Underground infrastructure of urban areas: selected and edited papers from the International Conference on Underground Infrastructure of Urban Areas, Wrocław, Poland, 22-24 October 2008

UNDERGROUND INFRASTRUCTURE OF URBAN AREAS

SELECTED AND EDITED PAPERS FROM THE INTERNATIONAL CONFERENCE ON UNDERGROUND INFRASTRUCTURE OF URBAN AREAS, WROCŁAW, POLAND, 22–24 OCTOBER 2008

Underground Infrastructure of Urban Areas Editors

Cezary Madryas, Bogdan Przybyła & Arkadiusz Szot Faculty of Civil Engineering, Wrocław University of Technology, Wrocław, Poland

Taylor & Francis is an imprint of the Taylor & Francis Group, an informa business © 2009 Taylor & Francis Group, London, UK Typeset by Charon Tec Ltd (A Macmillan Company), Chennai, India Printed and bound in Great Britain by Cromwell Press Ltd, Towbridge, Wiltshire All rights reserved. No part of this publication or the information contained herein may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, by photocopying, recording or otherwise, without written prior permission from the publishers. Although all care is taken to ensure integrity and the quality of this publication and the information herein, no responsibility is assumed by the publishers nor the author for any damage to the property or persons as a result of operation or use of this publication and/or the information contained herein. Published by:

CRC Press/Balkema P.O. Box 447, 2300 AK Leiden, The Netherlands e-mail: [email protected] www.crcpress.com – www.taylorandfrancis.co.uk – www.balkema.nl

ISBN: 978-0-415-48638-5 (Hbk) ISBN: 978-0-203-88229-0 (eBook)

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Table of Contents

Preface

VII

Scientific Committee/Reviewers

IX

Sponsors

XI

Problems of trenchless rehabilitation of pipelines situated under watercourses T. Abel

1

Building on underground space awareness J.B.M. Admiraal

9

New challenges in urban tunnelling: The case of Bologna Metro Line 1 G. Astore, S. Eandi & P. Grasso Numerical analysis of the effect of composite repair on composite pipe structural integrity A. Bełzowski & P. Stró˙zyk

15

27

Repair of RC oil contaminated elements in case of infrastructure T.Z. Błaszczy´nski

37

Modelling the behaviour of a micro-tunnelling machine due to steering corrections W. Broere, J. Dijkstra & G. Arends

45

Trenchless replacement of gas and potable water pipes with new PA 12 pipes applying the pipe bursting method R. Buessing, A. Dowe, C. Baron & M. Rameil

55

Experiences with Polyamide 12 gas pipes after 2 years in operation at 24 bar and new possibilities for HDD A. Dowe, C. Baron, W. Wessing, R. Buessing & M. Rameil

67

Simulation researches of pump-gravitational storage reservoir and its application in sewage systems J. Dziopak & D. Sły´s

75

New developments in liner design due to ATV-M 127-2 and case studies B. Falter

83

Concrete – durable composite in municipal engineering Z. Giergiczny, T. Pu˙zak & M. Sokołowski

97

Fly ash as a component of concrete containing slag cements Z. Giergiczny & T. Pu˙zak Rehabilitation of road culverts on the equator. Implementation of innovative open cut and jacking/relining trenchless solutions J.M. Joussin Urban technical infrastructure and city management W. Kaczkowski, K. Burska, H. Goławska & K. Kasprzak V

107

115 129

Maintenance of drainage system infrastructure in Butare Town, Rwanda A. Karangwa

141

Contact zone in micro tunneling pipelines A. Kmita & R. Wróblewski

149

Effect of variable environmental conditions on heavy metals leaching from concretes A. Król

155

Design of the pipelines considering exploitative parameters A. Kuliczkowski, E. Kuliczkowska & U. Kubicka

165

Management of sewer network rehabilitation using the mass service models C. Madryas & B. Przybyła

171

Utilizing the Impact-Echo method for nondestructive diagnostics of atypically located pipeline C. Madryas, A. Moczko & L. Wysocki

183

Selected problems of designing and constructing underground garages in intensively urbanised areas H. Michalak

193

Material structure of municipal wastewater networks in Poland in the period of 2000 to 2005 K. Miszta-Kruk, M. Kwietniewski, A. Osiecka & J. Parada

203

Two HDD crossings of the Harlem River in New York City J.P. Mooney Jr. & J.B. Stypulkowski Preliminary design for road tunnels on Trans-European Vc Corridor motorway, section Mostar North – South Border (Bosnia and Herzegovina) I. Mustapi´c, D. Šari´c & M. Stankovi´c Mapping the underworld to minimise street works C.D.F. Rogers Assumptions for optimization model of sewage system cooperating with storage reservoirs D. Sły´s & J. Dziopak

213

225 237

249

Curvature jacking of centrifugally cast GRP pipes U. Wallmann & D. Kosiorowski

257

Relining with large diameter GRP pipes U. Wallmann

269

Underground infrastructure of historical cities as exceptionally valuable cultural heritage M. Wardas, M. Pawlikowski, E. Zaitz & M. Zaitz

275

Author Index

287

VI

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Preface

In most cases, towns have grown up on the basis of industrial capital and relevant rules of industrialization. It has caused that such towns are malfunctioning, expensive, not ecological with all resulting consequences impeding everyday life of their inhabitants. Thus, public expectations are that portions of towns subjecting to modernization and also expansion of such towns would be progressed with utmost consideration for residential comfort by adapting newly grown town infrastructure to social, spiritual and cultural needs resulting from changed style of life and continuously changing scale of values. Creating urbanized space of such features is one among fundamental tasks that need to be undertaken to fulfil the expectations specified above. This task is also resulting from the necessity of unifying the towns and adapting them to standards becoming popular due to globalization process. New projects of modernization and expansion of towns having been now coming into being must be distinguished by better-than-before use of town space through stereoization of development, i.e. development of tower-block housing and underground structures. To meet this condition a higher level of integration of infrastructure systems among which the following equipment is distinguished: • equipment related to communication services for the town; • equipment connected with power management, water supply and sewage disposal system, waste disposal and management; • communication and information related systems which, assuming the need of control, also in respect to the remaining infrastructure systems, create the basis of urban management system. The most important is however to work out engineering solutions that would be the basis for creating integrated structures. The fundamental assumption for such studies must be creation of urbanized space enabling: • to exchange energy between systems/equipment and to use town heat from some structures, such as for example communication tunnels, sewage system or power systems, etc. • self-filling in of the water supply system (by waste water treatment), • to rise the safety of town inhabitants both in respect of natural threats (flood, seismic and paraseismic quakes, etc.) and external risks (terrorist or war actions), • to use the profits resulting from stereoization of town, i.e. temperature, humidity and acoustic conditions other than those existing over the ground, • to release the ground space from some functions (first and foremost the communication related functions) which shall be mainly used for residential and recreational purposes, • to renovate the historical, cultural and ecological environment of city centres. Thus, researches, planners and investors must focus their attention on making better use of underground space as the potential to improve town communication, on expanding centre capacity by moving many commercial and service function underground, and also on modernization and integration of underground system to improve their functionalities and to create conditions for construction and operation of other underground structures. Should the specified targets be reached, a package of administration regulations preferential for underground construction would be necessary, such which affects, but are not limited to the rules of crediting, subsidizing or commissioning the best solutions. According to experience gained to date in developed countries we can state that the underground space would be, and often already is, used without any limitations to generally all purposes (except residential function, which in this way could get more space on the ground). However this would VII

not mean that the topic has been exhausted and related problems resolve. Just the opposite. As it results from the materials included in this paper, this subject is still topical and many related issues need to be resolved. Hence, I hope that this paper would arise interest and inspiration for further examinations in persons engaged in widely understood shaping of underground infrastructure of urbanized areas. At the end of this preface, I would like to express special acknowledgements to institutions and companies which logos and names are included in this book as their financial support was a decisive factor allowing its publication. Main editor Cezary Madryas

VIII

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Scientific Committee/Reviewers

Han ADMIRAAL, President of Dutch Group ITA-AITES, The Netherlands Gerard ARENDS, Delft University of Technology, The Netherlands Rolf BIELECKI, President EFUC, Germany Bert BOSSELER, Wissenschaftlicher Leiter des IKT, Germany Józef DZIOPAK, Rzeszów University of Technology, Poland Bernhard FALTER, University of Applied Science-Münster, Germany Kazimierz FLAGA, Cracov University of Technology, Poland Piergiorgio GRASSO, Vice-President of ITA-AITES, Italy Wojciech GRODECKI, President of Polish Group ITA-AITES, Poland Eivind GRØV, Vice-President of ITA-AITES, Norway Alfred HAACK (D), STUVA, Germany Jens HÖLTERHOFF, President of GSTT, Germany Józef JASICZAK, Poznañ University of Technology, Poland Martin KNIGHTS, President of ITA-AITES, UK Andrzej KULICZKOWSKI, Kielce University of Technology, Poland ˙ Dariusz ŁYDZBA, Wrocław University of Technology, Poland Cezary MADRYAS, President of PSTB, Poland Herbert A. MANG, Technische Universität Wien, Austria Dietmar MÖLLER, Universität Hamburg, Germany Harvey PARKER, Past-President of ITA-AITES, USA Anna POLAK, University of Waterloo, Canada Chris ROGERS, University of Birmingham, UK ´ Anna SIEMINSKA – LEWANDOWSKA, Warsaw University of Technology, Poland Ray STERLING, Louisiana Tech. University, USA Markus THEWES, RUB – Bochum, Germany Roland W. WANIEK, President of IKT, Germany Andrzej WICHUR, University of Science and Technology Krakow, Poland

IX

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Sponsors

PLATINUM SPONSORS HERRENKNECHT AG

HOBAS System Polska Sp. z o.o

GOLD SPONSORS INFRA S.A

REHAU Sp. z o.o.

SILVER SPONSORS Amitech Poland Sp. z o.o.

Góra˙zd˙ze Cement S.A.

KWH Pipe (Poland) Sp. z o.o.

XI

OTHER SPONSORS ´ BEWA – Systemy Oczyszczania Scieków

Dolno´sl˛aska Okre˛gowa Izba In˙zynierów Budownictwa SIKA Poland Sp. z o.o.

Book supported by

XII

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Problems of trenchless rehabilitation of pipelines situated under watercourses T. Abel Wroclaw University of Technology, Poland, KAN-REM Sp. z o.o. Wroclaw, Poland

ABSTRACT: All types of collisions of pipelines with water race are frequently solved by elaborating sewer trap constructions. Such constructions are often encountered in sewerage systems when it is necessary to overcome an obstacle and for this reason by main collectors of large dimensions are used. Channels located in close neighborhood to surface water should be constantly monitored, since every damage or failure of the under-river pipeline or channel leading surface water may cause very serious consequences. Municipal and industrial wastes, when in contact with surface water may quickly result in contamination and ecological catastrophe. Surface water disturb water and sewerage balance and in extreme cases, with high pressures applied to the sewer trap construction, they may produce a very quick propagation of the damage and finally, construction disaster. Due to a very specific construction of a sewage trap (most frequently located under watercourses of Rother fix hindrances) it is impossible to repair or replace a pipeline network that is created by sewage trap in a traditional, dig technology. Owing to the development of civil engineering and the use of trenchless technologies, restoring the original condition of pipelines and assuring their safe exploitation are highly feasible. In the paper, examples of sewer traps will be provided and finished projects of sewerage systems rehabilitation, shown. The first structure to comment on will be a waste trap on the main drain of DN1200 located in the region of Pulawy (southeast of Poland) under Kurowka River. The sewer trap is composed of three steel pipeline networks of DN600. Rehabilitation works consisted in making short relining with PEHD modules. The second example will be a steel sewer trap of DN2200 located under discharge water of Thermal-electric power station in Konin (central Poland). Regeneration was made in Maxi-Trolining technology. The aim of this paper is to present trenchless methods of sewer trap structures rehabilitation and explain in detail all technological process as well as by-pass methods and materials solutions applied in pipelines and sewer traps entrance chambers.

1 INTRODUCTION As a result of intensive development, routes of sewage systems forming underground part of urban infrastructure are, in many places, situated in direct vicinity of other objects, particularly in urban areas. Engineering objects with which collision of sewage systems may occur are, first of all, road and rail tunnels, communication arteries and railway lines laid in towns in trenches, navigable waterways, water and heat supply mainlines and even buildings. The second group of obstacles that can be encountered while laying sewage systems are natural obstacles such as water courses and ravines. Depending on the depth of laying sewers as well as differences in grade lines, the ducting may pass above routes of engineering objects and other terrain obstacles or below them. 1

Laying a sewer above an obstacle can be performed on an aqueduct or in a vault passing over the engineering structure. When laying the sewer under an engineering structure, the following three cases can occur: • the sewer can run under the obstacle without change in shape or dimensions of the section. The bottom of the engineering structure intersects the ducting vault. • the section of the sewer enters the object structure only in its top part. In this case, it is necessary to change section of the sewer to a lowered one while maintaining drop of the channel bottom and speed in the channel possibly unchanged. • if the sewer enters into the structure of the colliding object with its whole section, a sewer trap should be planned under the obstacle.

2 ASSESSING THE TECHNICAL STATE OF DAMAGED SEWER TRAP PIPE The starting point of technology selection for rehabilitation of trap pipes, as structural element transferring definite load, is precise assessment of their technical state. In accordance with ATVDVMK M 127, three different technical states of damaged pipe can be distinguished. Depending on the kind of technical state, various loads act on the executed shell (grouting). Assessment of the technical state should be performed on the basis of TV camera inspection results. It should be realized that such mode of examining the structural state of sewer is not always sufficient. Especially in the case of trap construction of concrete and reinforced concrete pipes, very dangerous vitriol corrosion of concrete often occurs as a result of which it transforms into gypsum causing lowering in load capacity of the whole structure, even leading to its loss and occurrence of state of emergency. Level of corrosion cannot be assessed by optical examination but only by testing of samples taken. If justified, before rehabilitation, point repairs should be undertaken using robots. In such case, the recommended solution is to increase the strength of rehabilitating shell over the whole length.

3 TRAP CONSTRUCTIONS AS SPECIAL CASES DURING REHABILITATION OF SYSTEM Traps are constructions consisting of one or more pipes whose operation occurs under pressure. Trap constructions occur most frequently as objects made from pipes of cast iron, steel or steel in concrete or reinforced concrete casing. In case of repairing sewage systems, engineering objects as are trap constructions generate complications for application of trenchless technologies. Performing repairs of sewer traps using trenchless technologies requires conducting individual analysis of the case and drawing up a project, and particularly planning the technology for carrying out the works. Part of the trenchless methods of renovating the system does not find application in sewer traps. Characteristic features for structures surmounting terrain obstacles, causing narrowing of capabilities for application of solutions, are first of all: • very frequent changes in direction of laying trap lines in profile (figure 1). Location of trap lines at large angle disables application of most close-fitting technologies due to their limitations regarding susceptibility at arcs and occurrences of deformations that can be avoided. When using unconstrained linings in case of occurrence of even minimum change in direction of sewer route due to introduction into the pipeline of rigid pipe modules of lengths from 0.6 to 6.0 m, the trap arrangement completely disables execution of repairs in the abovementioned technologies. Technologies of unconstrained linings constitute a very good solution in case of occurrence of trap line arrangement as straight lengths (figure 2). • diameter of trap pipes. In case of passage pipelines, i.e. of diameters larger than 1000 mm, it is possible to utilize close-fitting technologies enabling formation of the insert directly in the 2

Figure 1.

Changes in direction of laying trap lines in profile.

sewer by the fitters thanks to which their precise execution is ensured. For non-passage pipes, due to absence of the possibility of direct intervention inside the sewer, the range of applications of close-fitting technologies narrows considerably. Lack of possibility of controlling execution of lining in every sensitive point (as in the case of passage pipes) forms a serious obstacle and aspect against the use of close-fitting linings. • trap construction in sewage system is most frequently applied in case of surmounting an obstacle in the form of water courses. Such a situation very often disables execution of by-pass type system enabling working on a section cut-off from utilization. Most trenchless techniques of repair are applicable only on inactive systems. Lack of facility for pumping over medium conducted through the pipe meant for rehabilitation defines to a certain extent the group of technologies that are possible for application. All the above-mentioned conditions constitute the group of factors which, occurring simultaneously, limit to a very small group of trenchless technologies that are possible for application.

4 REVIEW OF TRENCHLESS TECHNOLOGIES FINDING APPLICATION IN REHABILITATION OF SEWER TRAPS Linings used in trenchless methods of renovating sewage system trap constructions can be divided into close-fitting and unconstrained. The former (viz. in situ form) are methods consisting in making linings inside the existing pipe, whereas the latter consists in installation inside the section under repair, of pipes or modules of smaller dimensions than its inside diameter allows. Close-fitting methods are sleeves of technical fabric saturated with resins as well as polyethylene sleeves. An example of unconstrained linings, in case of sewer traps, is method of relining with short modules. 3

Figure 2. Trap line arrangement as straight lengths.

4.1 Sleeves of technical fabrics Technologies from group of close-fitting technical fabrics consist in inserting into the sewer a resinous shell – sleeve of technical fabric saturated with resin which, after filling with e.g. hot water or hot air, gets hardened and adheres closely to the old sewer structure. 4

At present, there are several variations of technical-fabric sleeve technologies, differing from the mode of introducing the shell, kind of medium used causing pressure in the shell, kind of agent hardening the shell. In certain situations, technical conditions make the CIPP sleeve to become the only rational solution. They could include, for example, deformation of the section. Considered here are such deformations as, for example, imperfections of the cross section. In other situations, application of CIPP sleeve is not possible, for example, in the case of loss of load capacity and break-down of structure of the trap construction utilized. At present, due to decrease in utilization of water and consequently in reduction in quantity of sewage, in many cases, reduction in cross section is a positive operation since, thanks to this, flow speed increases resulting in improved self-cleaning of the whole system and reduction in maintenance costs. 4.2 Polyethylene sleeves The basic example of this type of technology is Trolining system. It is a trenchless system for reconstructing combined sewage system and also sanitary, rainwater and industrial drainage systems as well as other pipelines both gravitational and also pressure, fabricated from various materials. This system belongs to the group of GIPP (grouting in place pipe) methods. Technologies of closefitting linings are characterized by eliminating to a minimum the need for carrying out earth works as are indispensable in case of classical technologies of underground infrastructure repairs; this facilitates reduction in communication difficulties. An important matter is also limitation of noise, dust and other nuisances. Deserving special attention however is the radical shortening of the period of work duration. In contrast with linings of technical fabrics saturated with resins (differing in fabric material and kind of resin and particularly the mode of hardening), TROLINING consists of polyethylene insert made from special kind of sheet/foil and filling layer. Large-size pipes are renovated using segments of PEHD panels stiffened with solid formwork, instead of “sleeves”. In case of infiltration of groundwater, panels are also used on the outside, adhering directly onto the renovated construction, protecting the new structure against external influences. The free space between the insert and the pipe walls is filled with concrete. Reinforcement can first be installed in it. The quantity of reinforcement, wall thickness and class of concrete are the deciding factors regarding rigidity of the repaired construction. These quantities are defined through precision calculations. The repair system of large-size pipes is also perfectly suitable for rehabilitation of inspection chambers. 4.3 Example of execution Discharge water trap in Konin Thermal-Electric Power Station.The object is located in Konin at Rybacka Street in northern part of the town. The trap construction has the task of carrying water from the discharge duct of Pa˛tnów Thermal-Electric Power Station under the duct leading water to Konin Power Station.This object belongs to the system of ducts and traps whose task is to carry cooling water of Pa˛tnów – Konin Power Plant Complex. The ducting is made from smooth St3S steel pipes with longitudinal seam and wall thickness of 20 mm together with external and internal anticorrosion protection. Length of each trap line is 32.5 m (figure 4). Assessment of technical state of the pipeline to be repaired showed very large corrosion cavities in places in pipe walls. Measurements taken with thickness gauge showed that wall thickness in certain places had reduced from 20 mm to 7.1 mm. From structural analysis conducted, it was concluded that in case of reduction of wall thickness to 5 mm, the trap construction would lose its strength. Technical state determined of the pipeline indicated the necessity for immediate reinforcement of the middle line of the trap. Deterioration of the technical state and the resulting loss of load capacity could threaten a construction catastrophe. 5

Figure 3.

GIPP system, TROLINING system.

Figure 4. Water trap in Konin.

Internal forces were calculated for two load situations: • case I – all loads occur, • case II – no water load in duct (pipeline is emptied of water). After dimensioning the construction, results were obtained ensuring adequate technical parameters for reinforced concrete layer of 150 mm doubly reinforced with ø14 rods in spacing of 100 mm. Structural and strength analysis for case II (trap construction emptied of water) indicated that for a pipe of such large loss in wall thickness (even up to 7.1 mm), its emptying of water is inadmissible since it could lead to breakdown. In view of the above, it was planned to provide a preliminary reinforcement by installing rings made of rolled 160 mm channel sections. The rings were installed by divers before pumping out water from the ducting (figure 5). 6

Figure 5.

Preliminary reinforcement.

4.4 Relining with short modules For rehabilitation of gravitational pipelines by means of short pipe modules, the modules utilized are of slightly less outside diameter than inside diameter of the renovated pipe, e.g. for renovation of DN 300 sewer, pipe modules of outside diameter 280 mm or 250 mm can be used. Rehabilitation consists in successive joining of consecutive pipe modules and simultaneous sliding the lining so assembled into the interior of the old pipeline. The modules have total length from about 0.6 m. This enables carrying out work inside the reinforced concrete chamber/pit and hence it is possible to rehabilitate the full length of trap construction without performing any earth work whatsoever. Available in the market are different methods of inserting renovation modules into the pipeline to be rehabilitated. Some firms propose pushing/jacking in the modules by means of hydraulic actuators/jacks; other propose pulling them in by means of winches. 4.5 Example of execution Sewer trap carrying communal sewage from the town of Puława to the wastewater treatment plant. The object is located on the route of the main drain of diameter 1200 mm. The trap consists of 3 lines of diameters 600 mm each, connected to the main drain through inlet and outlet chambers/pits. The trap construction lengths carry the sewage under the Kurówka River which constitutes a small tributary of the Vistula River. It was planned to perform rehabilitation using short relining technology. PEHD modules of length 1 m were used. The installation technology consisted in pulling in the modules by means of a hydraulic winch. The works were carried out under difficult winter conditions with sewer in operation. Sewage was transferred by means of a log stop provided through active part of the trap construction. This was possible thanks to division of the pit into cells. Before proceeding to execute the short relining, all the lines of the trap were subjected to hydrodynamic cleaning. After rehabilitation, the new pipeline obtained was of slightly less diameter, the modules installed being of outside diameter 580 mm.

5 SUMMING UP Due to the specifics of trap system operation which very often operates at 100% capacity, any method of rehabilitation used must not lead to deterioration of the hydraulic conditions. The repair must ensure improvement of hydraulic conditions, increase in load capacity and prolongation of 7

pipeline durability. Rehabilitation of the said objects must be carried out with maximum accuracy and precision due to inability of monitoring the lengths of trap construction at operational stage. Breakdown of sewer trap, in contrast with the whole sewage system, may carry more serious and dangerous consequences with it. As a result of sewer trap breakdown, damage can be caused to structures in its close proximity which very often include the objects for which provision of the sewer trap was necessary. Damage to a communication artery or water course may create a direct hazard to people. As can be seen, the consequences of no repairs or repairs carried out with errors may lead to several failures of structures. Sewer traps constitute a top-class challenge for any contractor undertaking their rehabilitation. Thanks to the possibility of applying the above-mentioned technologies for repairing such constructions, every designer has the capability of selecting the appropriate technology and planning out the rehabilitation procedure so as to acquire long-term effect and protection of the construction against breakdown. REFERENCES Sewage system. T. 1, Sewage system and pumping station. (In Polish). Kanalizacja. T. 1, Sieci i pompownie. Błaszczyk, Wacław, 1983. ATV-DVWK-A127P Instructions – Structural analysis of sewers and sewage system pipes. (In Polish). Wytyczne ATV-DVWK-A127P – Obliczenia statyczno-wytrzymało´sciowe kanałów i przewodów kanalizacyjnych. ATV-DVWK-M 127P Auxiliary Materials “Structural analysis for technical rehabilitation of sewage system pipes by introducing liners or by the assembly method”. (In Polish). Materiały Pomocnicze ATV-DVWK-M 127P “Obliczenia statyczno-wytrzymało´sciowe dla rehabilitacji technicznej przewodów kanalizacyjnych ˛ przez wprowadzenie linerów lub metoda˛ monta˙zowa”. PN-84/B-03264 “Concrete and reinforced concrete structures. Basic principles of designing”. (In Polish). PN-84/B-03264 “Konstrukcje betonowe i z˙ elbetowe. Podstawowe zasady projektowania”. Technical materials of M/s KAN-REM Sp. z o.o. (In Polish). Materiały techniczne firmy KAN-REM Sp. z o.o. Technical materials of M/s TROLINING GmbH. (In Polish). Materiały techniczne firmy TROLINING GmbH. Problems of trenchless rehabilitation of sewage system pipes. (In Polish). Problemy bezodkrywkowej odnowy przewodów kanalizacyjnych – Prof. Andrzej Kuliczkowski, 2004. Rehabilitation of Konin trap construction – Execution project. (In Polish). Naprawa syfonu koni´nskiego – Projekt Wykonawczy. Testing and acceptance of CIPP technical sleeves and their durability – Andrzej Kolonko, Trenchless Technology 3/2007. (In Polish). Badania i odbiory techniczne re˛kawów CIPP a ich trwało´sc´ – Andrzej Kolonko, In˙zynieria Bezwykopowa 3/2007.

8

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Building on underground space awareness J.B.M. Admiraal Centre Applied Research Underground Space – CARUS, Gouda, The Netherlands

ABSTRACT: Worldwide the demand on available space in urban areas is growing. The awareness that the use of underground space can offer a solution is often lacking. This paper will discuss the latest developments in the field of underground space planning. It will highlight why planning should be considered in order to avoid spatial conflicts which will be detrimental to the use of underground space. The paper will also discuss the sustainable use of the underground. Given the many benefits the use of the underground offers to life on the surface, it is often deemed to be sustainable. The author will argue that this is not necessarily always the case. A balanced decision is needed when considering the use of underground space in which the underground as a living organism should be considered.

1 INTRODUCTION With the worldwide search for more space in urban areas, the use of underground space is seen as a valid solution. The awareness that this is the case is in practice not widespread. In many cases the development of underground space is autonomous with far reaching effects. One of these being that further development is severely hindered and can only take place at great depth. Tunnelling is seen as one of the methods which can be applied to use the underground space. To prevent the autonomous development becoming common practice, a vision on the use of underground space needs to be developed at a local level. This in turn can facilitate planning the use of underground space, which should avoid conflicts between resource demands and also lead to multi-functional use of underground space. In many cases the underground space is a living organism. This will require balanced decisions on the use of underground space. Only in this way can the use of underground space be deemed to be part of sustainable development.

2 UNDERGROUND SPACE AWARENESS 2.1 The worldwide quest for urban space As the world population keeps growing, mega-cities are growing bigger and bigger. It is however not only the growth of the world population that leads to the development of mega-cities. The UNHABITAT programme has stated that as of mid-2007 more than 50% of the world population lives in cities. This means that the population shift from rural areas to urban areas is also contributing to this growth. One of the most common aspects of mega-cities is the struggle for space to accommodate all functions required to maintain liveability but also mobility. Cities can not survive without infrastructure. Infrastructure to allow its population to move, but also the infrastructure to provide the city with its power and water. Utilities also require space to be accommodated. With the adverse affects of climate change, sewer systems need to cope with 9

ever increasing amounts of rainwater. All this needs to be taken into account. The worldwide quest for urban space requires radical new insights into land use. Multiple land use is often seen as a new way to cope with the ever increasing demand. The use of underground space must be seen as a valid option within this context. 2.2 Sustainable development and climate change In The Netherlands a urgency agenda was published in 2007 by leading scientists and research programmes, calling on the nation to take to heart sustainable development and to climate proof the country. The so-called ‘Urgenda’ provides an action plan for the coming 40 years. The basis assumption is that The Netherlands will need to change more rapidly in the coming 50 years than in the past 500 years in order to cope with all the challenges the country is faced with. These challenges are both in the social cultural arena as in the civil engineering field. One of the statements in the Urgenda, is that within 15 years, intensive use of underground space will be common in The Netherlands. For the authors of the Urgenda it is a given fact that underground space use will play a vital role in the sustainable development of the country and its climate proofing, i.e. ensuring that the adverse effects of climate change are mitigated. 2.3 Underground space development Given, as show above, the role underground space use can play within the context of the worldwide quest for more urban space, there is a paradox which needs to be addressed. This paradox being that on the one hand many countries and cities already practice an intensive use of underground space whereas on the other hand there seems to be a worldwide ignorance to the fact that underground space can play a vital role in alleviating the spatial shortages at surface level. This paradox is one of the reasons for setting up the ITA Committee on Underground Space – ITACUS by the International Tunnelling and Underground Space Association, ITA-AITES. Worldwide many Western cities are finding the need to go deeper and deeper into the underground as the top layers are already congested with various functions. Users of transport systems need to be transported to great depth as the top layers are used for utility systems. This fact arises from the autonomous development of underground space without any form of coordination and no vision by city authorities on multiple use of underground space. The author has often publicly stated that the goal to achieve intensive underground space use in The Netherlands within the next 15 years is unreachable if the current practice of uncoordinated use of underground space remains. The autonomous development will lead to a chaos in the underground which will make future development impossible. Another problem which arises from the lack of awareness, is the simple fact that development of for example infrastructure will take place without even considering the possibilities of underground space use. This can lead to using contemporary methods which often give a suboptimal results as reported by the author (Admiraal, 2004). Awareness of the possibilities which underground space use has to offer is therefore needed on a large scale. Not only to ensure that this use is considered right from the start of development of cities, but also to ensure that once development takes place, it is done in a coordinated way. The role of ITACUS will be, to provide a platform for a worldwide dialogue on the use of underground space. A dialogue which will consider the use of underground space within the context of societal needs, environmental concerns, sustainable development and the climate challenge. 2.4 Underground space use There seems to be confusion in practice on what the use of underground space entails. Often tunnelling is seen to be the sole use of underground space. This is true in so far that tunnelling is a method which allows for various functional uses of underground space. Transport Use and 10

Production Use of the underground call for tunnels to make this possible (Admiraal, 2006). There are however many other uses which all compete for space in the underground. This fact in itself requires a balanced decision on how to develop underground space (Parriaux, Blunier, Maire & Tacher, 2008). A further complexity is added when we require this development to be sustainable as will be discussed later in this paper. As Parriaux and others point out, underground space can be modelled as consisting of four different resources: space, water, geo-material and geo-energy. All these resources can be used, but these uses can conflict with various results. These results can vary from pollution of drinking water to transportation projects not being carried out. There is need to consider the use of underground space in its entirety and not limit it to tunnelling. 3 PLANNING BASED ON VISION 3.1 Action without vision A Japanese proverb states that: ‘Vision without action is a daydream, action without vision a nightmare’. As stated above, with rapid autonomous development of underground space, a nightmare situation can arise as city planners discover the chaos which exists underground. In The Netherlands and in China the need for creating a vision on the use of underground space as the basis for a planned development is understood and put in practice. Although the scale is still limited, interesting results can be reported. The city of Zwolle in The Netherlands was the first to develop a vision on the use of the underground. One of the interesting results of this vision was the identification of polluted groundwater under a new urban development area. This has lead to the idea to combine the application of heat-cold storage with the cleaning-up of contaminated ground water over a period of 10 years. In this way the environment is served in two ways: the ground water is decontaminated and the carbon footprint for the development area is reduced as no gas or electricity is required to heat the houses in winter or cool them during summer. In the city of Shanghai in China, a pilot project is being carried out whereby for new developments of the city, underground space must be included in the planning of the development. It is evident that this situation will lead to a coordinated development in which an optimal use of underground space is ensured. China is a prime example of a country where the use of underground space is more and more seen as part of urban development. Vision and planned action is paramount for a controlled development of underground space. 3.2 Conflicts between resources The main problem which can arise from not planning the use of underground space can be best explained by two examples. As natural energy resources are seen to be limited, the search for alternatives is also a worldwide event. One of the most promising in this area is the application of geo-thermal energy systems. The application of these systems does however require vertical pipes to be inserted into underground space, often hundreds of meters deep. The use of these systems is very popular and a rapid autonomous deployment of these systems is observed both in Germany and The Netherlands. The downside to this is that these systems can become a serious obstacle for future development of underground space in the horizontal plane. It is not unthinkable that future alignments of underground mass rapid transport systems is severely hindered by the presence of these systems. Moreover, as observed by Parriaux, Tacher & Joliquin (2004), when a decision needs to be made for an underground mass rapid transport system or a geothermal application, the first is mostly chosen. In any case, no consideration is given to the possibility to combine these functions. Although not deemed to be feasible from an engineering perspective on the moment, there is no reason why in future this should not be the case. The point being made here is that without planning these situations can not be identified and therefore innovation as mentioned is not stimulated. 11

A second example stems from the conflict which arises from seeing underground space as a unlimited reservoir of space versus underground space as a nature reserve which needs to be preserved at all cost. This conflict arises from the fact that in more Deltaic regions, given the specific character of the subsoil, it is deemed to be the supporter of life on the surface, given the many natural processes and systems that exist below the surface. In one case this conflict has lead to a tunnel project being scrapped in The Netherlands for fear of changing the character of the Naarder Lake, a lake deemed to be part of an area of outstanding natural beauty. That the fear in itself is not unfounded can be demonstrated with the adverse effects of water inflow in tunnels as experienced in Norway (Grøv, 2008). Although the situation in Norway and The Netherlands are not comparable, the lack of understanding that the conflict between these two approaches exists, lies at the basis of the decision taken not to carry out the project. A balanced-decision making framework for the use of underground space can avoid these conflicts. Parriaux and others, are working on the development of such frameworks. The implementation of these frameworks can however only take place when there is a general awareness of the virtues of the use of underground space, this awareness is translated into a vision which in turn makes planning possible.

3.3 Planning methods in practice Various methods are in use regarding the planning of underground space. As reported by the author (Admiraal, 2006) research in The Netherlands has come up with practical methods which identify areas which are most likely to prove worthwhile for underground space development. These areas are identified by taking various aspects into account. Other methods try to approach the problem from a theoretical side and use a systems approach on which decisions can be based (Parriaux et al, 2008). The most common development is to focus on areas for development rather than on individual cases. The so-called area development approach often incorporates a dialogue based model in which all interested stakeholders are involved. Planners then use the outcome of this dialogue to develop a stakeholder based vision on which the planning can be based. In this approach the interests of all parties are taken into account and the conflicts as observed above can be avoided as all interests are weighed-up. An interesting tool for analysis is the so-called ‘layered approach’ to area development. In this approach an area is deemed to consist of three layers: habitation, networks and the underground. The method has been developed by the Ministry of Housing, Spatial Planning and the Environment in The Netherlands. The fact that the underground is recognised as an entity to be considered in future planning issues is a positive development. The city of Arnhem in The Netherlands is actively using this approach for the planning of developments with positive results in terms of underground space use. One of the results being the combination of functions underground. A prime example is the development of an underground car park in combination with an utility tunnel which incorporates an underground waste refuge collection system. In other cities these developments would eventually have taken place on an autonomous scale. In this example the combination of these functions through planning based on vision makes it an excelling best practice example.

4 THE UNDERGROUND AS LIVING ORGANISM The use of underground space takes place in different regions of the world with varying geological conditions. Mega-cities are most commonly found within 50 km of the sea. This makes that a lot of mega-cities and therefore underground space development takes place in Deltaic regions. These Deltaic regions have in common that the underground consists of soft soils being old river deposits. The regions also have in common that they are often very fertile, providing land for crop growth. 12

In Deltaic regions, the underground can be modelled as being the supporter for life on the surface as mentioned before. The time scale at which processes take place in the underground varies dramatically from life on the surface. Compare to appreciate this the time it took to produce coal, gas and oil as natural resources with the lifespan of an average building. The argument being made is that we often are not aware of the effects of manmade interventions in the subsoil. A prime example being the contamination of drinking water aquifers through storage of waste in the subsoil. The Netherlands is still facing a massive clean-up operation to this effect. The use of underground space as part of sustainable development must consider the above when decisions are taken. It clearly illustrates that autonomous development of the underground not only can lead to resource conflicts as mentioned in the case of transport versus geothermal energy. It can also lead to developments which may prove to be non-sustainable. In general it is felt by the author that resource conflicts can be avoided and a sustainable development of underground space can be achieved through planning. In the case of the city of Zwolle, the combination of a geothermal application with decontamination of groundwater, clearly shows what can be achieved through vision development.

5 CONCLUDING REMARKS The use of underground space is seen to be vital in a world where more than half of the population now lives in urban areas. Further concentration in mega-cities is a trend which can not be stopped. The use of underground space can contribute to sustainable development, maintaining liveability and preparing the world for the impact of climate change. Creating awareness on the use of underground space in this respect is vital. Furthering the development of visions on urban underground space use and rational use of the underground space through planning techniques is essential. The future of underground space use is furthermore governed by the ability to combine functions, e.g. combination of transport functions with water management. This requires a dialogue within city authorities across policy boundaries. The results can however be very positive as is demonstrated by the Stormwater Management and Road Tunnel project which is now operational in Kuala Lumpur, Malaysia. The ability to plan the use of underground space in combination with a multi-functional use will determine the future of underground space use as a valuable contributor to the worldwide quest for more urban space and a sustainable development of mega-cities. It will also show that society can really not afford not to use underground space. REFERENCES Admiraal, J.B.M. 2004. Developing a knowledge infrastructure for underground space in Indonesia. Proceeding 7th Joint Meeting JTA-COB, Stichting COB, Gouda, The Netherlands. Admiraal, J.B.M. 2006. A bottom-up approach to the planning of underground space. Tunnelling and Underground Space Technology, Volume 21, Issues 3–4, Pages 464–465. Grøv, E. 2008. Water control in Norwegian tunnelling. Proceeding South American Tunnelling 2008. Brazilian Tunnelling Committee – CBT, Sao Paulo. Brazil. Parriaux, A, Tacher, L. & Joliquin, P. 2004. The hidden side of cities – towards three-dimensional land planning. Energy & Buildings, Volume 36, Pages 335–341. Parriaux, A, Blunier, P, Maire, P. & Tacher, L. 2008. The urban underground in the deep city project: for construction but not only. Proceedings Underground space challenges in urban development, Stichting COB, Gouda, The Netherlands.

13

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

New challenges in urban tunnelling: The case of Bologna Metro Line 1 G. Astore, S. Eandi & P. Grasso Geodata SpA, Turin, Italy

ABSTRACT: The Line 1 of Bologna Metro is 7 km long with 12 stations which crosses the entire city from the Fiera District to the Maggiore Hospital. The system adopted is a light rail tramway operated with 34-m long, single-unit vehicles. The line is designed to be completely underground and its construction involves the use of all the available tunnelling technologies: cut and cover, TBM and NATM. The portion of the vertical alignment in the city centre is very deep to avoid damage to buildings and to allow the line to underpass the new High Speed Railway tunnel near the Central Railway Station, where it is foreseen an interchange with the metro. The Piazza Maggiore Station is the most complex and important in the entire line and represents a great challenge, for designers in particular, because at this station site the horizontal alignment has a turn of 90◦ which has to be built completely by conventional tunnelling techniques.

1 INTRODUCTION 1.1 The Bologna Case Located in northern part of Italy, Bologna, a medium-size city with about 400.000 residents, has an ancient historical centre and can be considered as the heart of Italian roads and railway network. In fact, its unique position, being in the centre of a crossroad linking north to south and west to east of Italy, creates a huge demand for public transportation systems. As a matter of fact, to improve and change the city layout, at least four infrastructural works will be built in 10 years time: a new high speed railway station; a people mover connecting railway station to airport, an important line for trolley bus called (Civis); and an enforceable Metro line. In the planned network, the metro Line represents one of the most complex works because of its length (7 km) and depth, alignment, and difficulties in constructing the civil works. The main constrains are the presence of old-built city centre with historical monuments and the possibility of the archeological findings, entailing the metro line to be completely underground. Geodata, leader of a team consisting of other design firms, has developed for the Commune of Bologna the Final Design of the whole metro line, which includes the civil works, the E&M installations and the tramway track system, besides the geological and environmental studies.

1.2 Description of the Metro System and the main project data Bologna will be equipped with a semi-automatic tramway system with drivers aided by ACC (Automatic & Centralized Control system), a system which allows for monitoring and tele-control of trains and subway traffic, and by ATP (Automatic Train Protection), a system for controlling the speed and distance between trains as well as for managing the mobilization of trains, once they have arrived at the station. Each station is to be equipped with platform screen doors that separate the platform from the train. These screen doors represent a relatively new technological addition to many metro systems 15

Figure 1.

Schematic plan of the entire line 1 from Fiera (right on top) to Maggiore Hospital (left).

around the world, with some platform doors added to the existing systems later. They are widely used in Asian and European metro systems. The modern low-floor rolling stock will run with 2’ headway during peak hours in the underground section, elsewhere with 4’ to 6’ maximum frequency. The principal characteristics of the line are: • • • • • • • •

Total length = 7800 m Semi-automatic tramway system (Driver aided by ACC /ATP technology) Underground stations = 12 Surface stations = 1 Ventilation shafts = 11 TBM(Tunnel Boring Machine) tunnel length = 5600 m Cut & Cover tunnel and U-shaped section length = 800 m Conventional tunnel length = 200 m.

2 GEOLOGICAL AND GEOTECHNICAL SETTING 2.1 Geology The geology of the top layers may be outlined through dividing the alignment into three sections, each of which has peculiar characteristics in terms of the depositional stratigraphy. The first section (from Michelino station to FS station) and the third one (from Saffi station to Maggiore Hospital station), corresponding respectively to the eastern and western side of the route, are characterized by alternate presence of gravel-sand sedimentary layers from river channel and variable bands of silt and clay from floodplain. The intermediate section concerning the historical town centre (from FS station to Saffi station), by contrast, consists almost entirely of fine soils. Coarse sedimentary bodies are almost absent and the main stratigraphic markers are represented by paleosols, passing through soil bands, rich of organic matter on the top, to over-consolidated bands with a lot of carbonate concretions. This stratigraphic layout is consistent with the well-known geomorphologic framework of the Bologna valley. In particular, in this case the extremes of the alignment cross alluvial cones, while the central section passes through predominantly fine inter-cone areas. 16

The soil-layers’ attitudes reflect approximately the complex geological reality of the project area, characterized almost exclusively by alluvial deposits which, by their lenticular geometry, show high vertical and lateral variations. 2.2 Geotechnical setting and hydrogeological regime The design geotechnical model of the project has been developed on the basis of historical data and new survey campaigns carried out in 2007 in the course of its design development. The depth of boreholes is directly linked to the depth of the stations and ventilation shafts, in order to provide a reliable definition of design parameters along the entire line. During the site investigation many in-situ geotechnical tests were carried out such as Standard Penetration Test (SPT) and also undisturbed soil samples have been obtained for laboratory tests. Other in situ tests carried out were: cone penetration tests with piezocone (CPTU) to measure porewater pressure; in situ dissipation tests for evaluating the coefficient of horizontal consolidation and horizontal hydraulic conductivity; Seismic Dilatometer Marchetti Test (SDMT) for defining the dynamic properties of the soils. Finally all boreholes were equipped with a piezometer in order to measure the hydraulic head in the aquifers along the entire route. In the first 40 m it can be found a multilevel groundwater aquifer. It consists in four different levels, partially saturated and locally under pressure, named respectively, from bottom to top: SUP1, SUP2, SUP3 and SUP4. Groundwater levels SUP1, 2 and 3 are located in sands and gravels, while SUP4 is included in sands, limes and silts, more superficial (−6 to −7 m from surface). In any case, SUP3 and SUP4 can not be easily distinguished, particularly where the soils change from one type to another. Tunnel and stations do not touch the deepest groundwater level (SUP1), but they are affected mainly by SUP3 and SUP4 The Geotechnical units are reported below. • Geotechnical unit A – Gravel layers This unit is formed by lens of coarse sand, gravely sand, gravel with sand, and sandy gravel, and clasts up to approximately 8 cm, as well as rare pebbles. The thickness of the soil lens are 6 to 8 m (NSPT = 15–50, in no case refusal). It is largely present in the Michelino-FS station section (Fig. 2) and partially in the section between Malvasia and Maggiore Hospital. The unit has good permeability and hosts major the aquifers of Bologna (SUP1, SUP2, SUP3) • Geotechnical unit B – Cohesionless sands This unit consists of uncemented sands, from coarse to fine, sometimes silty, predominantly saturated soils. SPT values do not show particular granulometric differences. The coarse sands are less compacted than the fine ones (NSPT < 10, until 2). • Geotechnical unit C – Fine cohesive soils This unit includes fine soils with a cohesive behaviour, mainly silty clay and clayey silt with peat trails. It is the unit that is most intercepted by the metro alignment, mainly in the section between FS Station and Saffi station. Table 1 shows the geotechnical properties of the three units defined for the design.

3 TUNNEL CONSTRUCTION TECHNIQUES 3.1 Main constraints and adopted solutions A very complex and time-consuming work in a historical, urbanized area is always a challenge for designers. In the case of Bologna Metro line 1, different and difficult items were considered in planning and designing the underground works such as tunnelling under water table in very poor ground conditions, underpassing of buildings particularly at historical places, protecting archeological findings related to the Roman era, siting of stations, exits and ventilation shafts 17

Figure 2. A part of the geological profile near Bolognina Station.

Table 1.

Geotechnical parameters

Unit

γn [kN/m3 ]

ϕp [◦ ]

c [kPa]

cu [kPa]

E [MPa]

A B C

18–20 18–20 19–20

30–36 26–30 22–28

0 0 0–20

– – 50–150

40–80 20–40 10–40

both in city centre and in commercial and congested areas (at north-east of the town), minimizing construction-site areas, and managing intensive surface traffic. In the light of the above constraints, the solutions adopted for the project are: • the entire line for its 7.8 km length, is conceived to be underground with a configuration of single tube, double track and the designed solution provides systematic recourse to the mechanized tunnelling; • external superficial tunnels are designed as cut and cover sections in order to realize connections between the depot and the Michelino station at the north side and between the Ospedale Maggiore station and the Malvasia station at the west side; and • for special situations, short conventional tunnels are foreseen, like, for example, in the Piazza Maggiore station. Specifically, the running tunnel will be realized using an EPB Shield with a diameter equal to 9.80 m, which can ensure (Fig. 3): • an internal tunnel diameter of 7.90 m (functional minimum = 7.80 m, plus 10 cm of tolerance); • tail void of 15 cm; 18

Figure 3. TBM tunnel typical section (left) and EPB operational scheme (right).

• final lining thickness of 35 cm with a nominal length of 1.4 m; • each ring of precast segments is tapered to negotiate the curved tunnel alignment; • the joints between segments are to be sealed using either neoprone or hydrotite gaskets. The key objectives contributing to the above choices: • altering as little as possible the original stress state of the soils; • avoiding unnecessary, extra-excavations in order to control the excavation-induced effects on the surface (subsidence). The EPB excavation mode can provide continuous support to the tunnel face, with the soils excavated by the cutting head accumulated under pressure in the excavation chamber and then extracted by a rotating conveyor. Geodata has successful experience in tunnelling with EPB-TBMs in Bologna because in 2006 Geodata was involved in the construction studies and technical assistance during the works of two parallel tunnels (9,4 m diameter and 6.112 m in length each) of the Bologna high speed railway line (especially Lot 5 of urban penetration in the quarter S. Ruffillo in the south of the city and the new central station). Both tunnels were realized by EPB-TBMs. The success was reflected by the tunnel daily production rate and the solutions to prevent damage of buildings. In designing the Metro Line 1 Geodata has used the technical and environmental known-how learned directly from the high-speed rail tunnel. In particular, back-analysis of the observed settlements have been made to determine the parameter values (Vp e k, O’Reilly and New,1992), which are necessary for input to the subsidence prediction and building risk assessment.

3.2 Risk analysis of buildings and soil improvement design Potentially, buildings affected by tunneling are as many as 400, of which 35 are underpassed directly by line or station tunnels. In such situations, a comprehensive, detailed survey in-situ was carried out to collect and organize critical building data. The crucial parameters have been managed using a GIS system in order to faciliate the assessment of potential building damages due to underground works: excavation of tunnels, stations and shafts. The evaluation of critical buildings revealed that where expected settlements are not compatible with prescribed safety limits, the design had to apply soil improvement: principally jet-grouting (Fig. 4) or compensation grouting in the very critical area like Bolognina. 19

Figure 4.

Soil improvements: tunnel crown completely grouted (left) or grouted wall to protect building edges (right).

Figure 5. Typical station – assonometric view (left) and internal rendering of stairs connecting atrium with mezzanine.

4 TYPICAL STATIONS 4.1 Functional and architectural layout The stations are the connections between the surface and the running tunnel and between the town and the line; with this point of view the functional layout of a station is defined in order to: • • • •

reduce the length of the paths to and exit from the platforms; create a wide, bright space where passengers can easily recognize the right directions; locate access where pedestrian flows are most significant; minimize the volume of the entire station.

In particular, the architectural image chosen for each station is built on the purity of the volumes, in which are avoided scarcement and blind spots; the clarity of functional space, where stairs and elevators are always visible to users; and the usage of finishing and furnishing elements are chosen for simplicity and elegance. With the above design principles in mind, 9 out of the 12 underground stations are designed according to a typological scheme, which is 18 m wide and 42 m long, and lies at an average depth varying from 15 m to 25 m. The platforms are separated from the tracks by platform screen doors. 20

Figure 6.

Piazza Maggiore Station plan view and detail of TBM passage.

4.2 Constructive method Typical stations are to be built with the top-down variant of the cut and cover method. This solution permits to reduce construction times and worksite areas, In fact, the surface areas above the cover slab can be returned to the city for realizing parking, viability or construction depot. The principal construction phases are: diaphragm walls built by hydromill; construction of concrete cover slab; excavation under the cover; construction of bottom slab; building other internal concrete works from bottom to the top. The stations of Piazza Maggiore, Riva Reno and FS differ from the typological scheme. In the next section a description of the Piazza Maggiore station will be given, which is no doubt the most significant piece of work of the whole Bologna metro project. 5 PIAZZA MAGGIORE STATION 5.1 General problem The Piazza Maggiore Station is located at a historical square of Bologna where two major roads cross the town centre: via Indipendenza and via Rizzoli, with a lot of critical and historical buildings and a lack of space to for construction sites. For all these constraints the Piazza Maggiore Station itself can be considered as a project in the metro project. The station is 150 m long and positioned on a curve (with 25 m radius), the station platforms on the two sides shall be staggered in order to optimize the unusual shape and permit an easier train stoppage. This station has 4 principal constituting elements (Figs 6, 7): • Platform tunnel on a curve to be built with conventional tunnelling method; • Large access shaft, used also as construction shaft; 21

Figure 7.

Piazza Maggiore Station longitudinal section (A-A section in Figure 6).

• Platform access tunnels to connect shafts with platforms; • Existing underground atrium. This structure will be upgraded to create a new atrium for passengers, but during construction it will be used as storage area to minimize demand for surface area. The 25 m curve is not a problem for the tramway alignment itself, because trains will approach the curve slowly, leaving after stopping at the platform. However, in the construction phase it has to permit the TBM to pass through that become a fundamental point in the entire work. Generally, a TBM of the required size can not excavate curves with a radius of curvature less than 200 m. Furthermore, in this particular case there is also not enough space to extract the TBM from a shaft located at one end of the station and lower it down at the other. For these reasons it is decided to create a conventionally-excavated platform tunnel with a wide cross-section shape to permit the shielded TBM to pass through the already-excavated station space. The TBM passing phase can be detailed as follow: shield machine enters from the north (via Indipendenza) in the already-excavated tunnel; backup are dismounted; the shield is moved through the curve with a special trolley system to the new start position aligned to via Rizzoli; shield starts excavation in the new E-W direction with back-up re-connected to the shield. 5.2 Tunnel and shaft calculation Such a complex work needs impressive studies and calculations to check the structural solidity and minimize risks in all construction phases, being in the town centre. Construction phases have been studied for a long time with experts in conventional and mechanized tunnelling, ground improvements, underground works, etc. in order to be sure that such a work can be affordable. Principal construction phases have been used to define: preliminary and definitive lining, extension and type of soil improvements, internal tunnel shape (minimum to move TBM in curve), connection tunnels principal parameters (Fig. 9). The circular shaft, from which curved platform tunnel and access tunnels should be constructed, will be constructed first. Being 20 m in diameter and 39 m deep with diaphragm walls of 45 m long, the circular shaft will be excavated using cut & cover top-down method. The shaft has intermediate slabs, every 4.8 m, to create support for stairs and to stabilize the diaphragm walls themselves. Bottom slab has an arched invert shape in order to reduce bending moment and to transfer bending forces stemming from compressive ground water forces on the diaphragms. 22

Figure 8.

Rendered assonometric view of the Piazza Maggiore Station.

Figure 9.

Major construction phases of the Piazza Maggiore Station.

23

Figure 10.

2D and 3D analysis performed in access shaft design.

Figure 11.

Conventional platform tunnel FEM analysis. Model and settlements result.

Both 2D Finite Difference Method (FDM) using FLAC and 3D Finite Element Method (FEM ) usingANSYS have been used to evaluate bending moments, shear and axial forces on the diaphragm walls, creating an axial-symmetric model with 3D analysis to simulate the creation of large openings in the structure (Fig. 10). For the curved platform tunnel, the top-heading and bench invert technique is considered to be more suitable given its big cross-section area of 170 m2 . This large section requires some special attentions in calculating the internal structures and in evaluating the settlement effects on surface buildings. This conventional tunnel is to be built in soils with very poor geotechnical characteristics and close to buildings and monuments. Thus, the construction is conditional to intense ground improvement. The techniques to be used are jet-grouting and soil freezing. Jet-grouting is used where there is enough space to work without large interference to surface activities, while soil freezing is used where surface works are not allowed. Compensation grouting is also foreseen in some particular situations like the under-passing of a critical historical building or to prevent settlements in the more critical sections of tunnel, where curve is at most. Like for the shaft, the tunnel is also designed using 2D (PHASE 2) and 3D numerical analyses (FLAC3D), modelling the excavation sequences. A series of PHASE 2 analyses was carried out to identify the more suitable solution in terms of the construction steps, and the type and extent of ground improvement (Fig. 11). 24

6 CONCLUSIONS The Bologna Metro line 1 represents a real challenge for designers. As a multi-constraint tunneling condition in urban area, the design of the Bologna metro project demanded special attentions to link properly the obstacle components of the projects, namely, conservation of a historical town, preservation of buildings and monuments, and minimization of the superficial impacts. All these constraints led to a very complex construction work. The job difficulty at Piazza Maggiore Station brought about also some challengeable design criteria, of which the most noticeable are: a staggered platform solution on curve, a new technical approach in using EPB TBM, a special solution to pass through the curve with a very small radius. The project also demanded for an extensive use of the available ground improvement techniques: freezing, jet-grouting, compensation grouting, etc. The solutions illustrated in this paper should be valuable for those who have to design similar underground works in analogous urban conditions. REFERENCES Amorosi A. & Farina M., 1994. Stratigrafia della successione quaternaria continentale della pianura bolognese mediante correlazione di dati di pozzo. 1st European Congress on Regional Geological Cartography and Information Systems, Bologna (Italy), June 13–16, 1994. Volume 5, 16–34. Amorosi A. & Farina M. 1995. Large-scale architecture of a thrust-related alluvial complex from subsurface data: the Quaternary succession of the Po Basin in the Bologna area (northern Italy). Giornale di Geologia, 57/1–2, 3–16. Geodata S.p.A., 2007. Metrotranvia di Bologna – Final design documentation. Guglielmetti et. al., 2007. Mechanized Tunnelling in Urban Areas. Taylor & Francis. Marchionni V. & Guglielmetti V. 2007 EPB-Tunnelling control and monitoring in a sensitive urban environment: the experience of the “Nodo di Bologna” construction (Italian High Speed Railway system), ITA-AITES World Tunnel Congress 2007 “Underground Space – the 4th Dimension of Metropolises” Prague, 5–10 May, 2007. Grasso P. & Guglielmetti V., 2008 High-speed Railway Underground-Crossing Bologna, Italy, Workshop on “Tunnels in densely populated urban areas” – Professional Association of Civil Engineers of Catalonia Barcelona, 7 April 2008.

25

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Numerical analysis of the effect of composite repair on composite pipe structural integrity A. Bełzowski & P. Stró˙zyk Wrocław University of Technology, Faculty of Mechanical Engineering, Wrocław, Poland

ABSTRACT: The paper deals with the problem of assessing the structural integrity of a composite reinforcement used to repair a nonpressure sewerage piping system with assembly damage. An analysis of the strain criteria applied to assess the structural integrity of polyester-glass composites was carried out. The technical aspects of the repair of the damage are discussed and a numerical model of the pipe section under repair is presented. The calculations made indicate the composite repair structural and material solutions’ potential for further improvements and measures which are less effective in this regard.

1 INTRODUCTION Polymer composites formed from glass reinforced plastics are characterized by relative lightness and stiffness and good resistance to environment action and caustic substances. Polymer composites are now commonly used to build all kinds of piping systems, such as: • urban infrastructure networks – pressure piping systems for water supply, sewers without pressure, sewage treatment plant fittings, etc.; • process plants, including storage tanks for petroleum derivatives and caustics, cooling circuit piping systems, flue gas pipelines in power plants, and so on. The minimum service life of urban infrastructure piping systems is 50 years (EN 1796, 2006), (EN 14364, 2006). The life of 20–30 years, 40 years and 10–30 years is assumed respectively in the petrochemical industry, nuclear power plants (Le Courtois, 1995) and chemical-resistant tanks. Polymer composites are susceptible to chemical and physical ageing and mechanical degradation (Bollaert & Lemasçon 1999), (Tuttle, 1996) which adversely affect the material properties. As a result, the composite’s elasticity modulus and strength during the anticipated service life may decrease by as much 10–60%. The changes in the composite properties are gradual but they can be predicted on the basis of accelerated ageing tests (Bełzowski, 2005). Long-term test procedures for composites are usually modelled on the tests described in (ASTM D2992, 1991). Tubular specimens (minimum 18 pieces) are subjected to pressures which should result in a failure within 10000 h (14 months). By extrapolating the simple regression determined from the coordinates of the failure points one can estimate the strength (ASTM D2992, 1996) or stiffness (PN-EN 1120, 2000) and (Farshad & Necola, 2004) of the material over its whole service life. The above procedure is considered to be a reliable but expensive way of predicting changes in the properties of composites. Irrespective of gradual degradation, most of the structures can be accidentally damaged (e.g. by impacts), which may additionally reduce their life. Discontinuities in the protective layers (PL) on the inner surface of pipes and tanks in structures exposed to the action of corrosion factors pose a threat to their durability. PL discontinuities can be technological defects or service damage. Protective layers should protect the glass fibres constituting the structural reinforcement against the corrosive effect of liquids filling the system. PL discontinuities may drastically reduce the durability of the piping system. 27

2 REPAIRS OF PIPING SYSTEMS Repairs on composite pipes, chemical resistant tanks and so on are carried out in order to remove: • technological defects, • local damage caused during transport or assembly and by incidental service overloads (e.g. accidental impacts), • deterioration in the properties as a result of long-lasting material degradation. The descriptions of repairs based on the composite reinforcement technology, found in the literature, focus on a few applications: a. Repairs of chemical-resistant, composite (often high-risk) process facilities for storing dangerous caustic substances, etc. The principles of carrying out such repairs are described in (ASME RTP-1, 2000). The criteria of qualifying a device for repair and the permissible repair range (limited to 3–10% of the inner surface) are quite stringent. b. The repair of the inner surface of whole sections of sewer pipelines by producing a new composite shell inside the old worn out conduit. Mainly preimpregnated sleeves hardened inside the pipeline by means of elevated temperature or UV radiation are used. The achievement of proper ring stiffness of the new shell can be the repair effectiveness criterion. c. Repairs of steel and composite piping with local corrosion damage causing leakage, by making external sealing composite rings. Chemical and petrochemical industry process pipelines are repaired in this way. The principles of designing, carrying out and evaluating a repair are described in (ASME PCC-2, 2006). Besides repairs to high-risk pipework, the codes also cover repairs to low-risk pipework but with its diameter limited to 1000 mm. Repairs on leaking places in steel pipelines are discussed in (AEA Tech, 2005). From an analysis of the standards and publications devoted to piping system repairs using composite-based technologies the following conclusions emerge: • Most attention is devoted to high-risk piping system repairs. • The criteria of qualifying damage for such repairs, defined in the standards, are quite stringent. • The considered techniques of repairing local damage usually do not take into account the specificity of pipelines laid directly in the ground. • The planning of repairs and their realization and technical acceptance include expert assessments of the damage extent and the effectiveness of the measures taken. The above mentioned publications on repairs aim to ensure high professionalism in this regard. This is understandable since most of the repairs are done on chemical resistant components of highrisk process plants. Professionalism is essential here as evidenced by descriptions of dangerous failures of unprofessionally repaired facilities (Bełzowski, 2004), (Myers et al. 2007). This paper presents a numerical analysis of the way in which assembly damage to a buried sewerage piping system without pressure was carried out, focusing on the comparison of the numerically calculated strains in selected points of the area subjected to repairs with the criteria for dimensioning composite components used in the construction of various piping systems. The possibility of improving the material-structural solution used in the repair is assessed. The pipeline section with damage was made from DN1400 mm polyester-glass pipes with ring stiffness SN = 10000 N/m2 . The wall thickness was about 34 mm. Because of the improper application of the force damage in the form of cracks, chips and spalls appeared in the pipes as they were being shifted. The damage was analyzed in (Bełzowski & Stró˙zyk, 2008). 3 STRAIN CRITERIA OF DIMENSIONING COMPOSITES The design of chemical resistant composite piping systems is based on the assumption that a 0.2– 1.0 mm long crack propagating across wall thickness represents dangerous damage. In the literature 28

Table 1. Allowable strain and safety factor values for chemical-resistant tanks and pipelines. Standard

Strain constraints

Other dimensioning criteria

BS 4994 NFT 57-900 ASTM D3299 AD2000 Merkblatt and WUDT-UC-UTS/01:10.03

εd ≤ min(0.2%, 0.1εB )∗ εd ≤ 0.2% and ε⊥ ≤ 0.1% ∗∗ ⊥ εd ≤ 0.1% for cylindrical part under test pressure: ε⊥ ≤ 0.2% in direction ⊥ to UD reinforcement and ε ≤ 0.35% for CSM reinforcement εd ≤ 0.1% in wound components εd ≤ 0.25% εd = 0.09%, 0.12%, 0.15%, 0.18% depending on temperature and environment

δ ≥ 8∗∗∗ δ≥6

ASME RTP-1 EN 13121 BS 7159

δ≥4 δ = 10 for hand laminating δ≥4 LCL/1.3∗∗∗∗

∗ε

– allowable strain for composite, εB – failure strain for resin – strain perpendicular to UD reinforcement fibres ∗∗∗ δ – load or stress safety factor ∗∗∗∗ LCL – long-term strength for failure probability of 2.5% d ∗∗ ε

⊥

it is referred to as the first ply failure (FPF) criterion. This crack size approximately corresponds to a single ply of typical chemical resistant (polyester-glass) laminates. Assuming a safety factor of 1.5 for the so defined damage threshold (Eckold, 1985), the criterion of allowable linear stress εd , which usually amounts to 0.1–0.25% (Table 1), was introduced into the design standards. This strain limitation protects the structure against the development of transverse cracks which would facilitate the infiltration of liquids from the inside of the piping to the structural layers containing glass fibre E which is susceptible to corrosion. Since the failure strain of polyester-glass composites predominating in such applications amounts to about 1.5–2.5%, relatively high safety factor values are obtained as a result. The actual strain safety factor values assumed for chemical-resistant process plants mostly amount to about 6–10. The load safety factor values are similar. Interesting conclusions emerge from (AEA Tech, 2005) devoted to repairs of leaking metal pipes. For one of the categories of composite repairs on high-risk equipment, defined there as Class 3, the allowable strain for a repair life of 2 years amounts to 0.30%. For a repair life of 10 years and 20 years εd = 0.27% and εd = 0.25% are respectively assumed. The allowable strain for composite repairs on low-risk equipment is respectively: 0.40% for a repair life of 2 years, 0.32% for a repair life of 10 years and 0.25 for repair life of 20 years. The above strain values show the conservatism of the authors of (AEA Tech, 2005) in the assessment of repair durability. The assumed allowable strains for composite repairs are not much different from the ones used in the design of high-risk chemical-resistant equipment (Table 1). The highest assumed allowable strain amounts to 0.40% and it applies to low-risk equipment repairs with a life of 2 years. The above analysis takes into account only the strain criteria used in the design of composite repairs, neglecting the calculations based on the long-term strength values of the laminates. The initial safety factor for pipes is defined as a ratio of the relative deflection of the pipe along its diameter (Figure 1), reached until the first symptoms of failure, to the allowable deflection. According to standard EN 14364, a composite pipe with ring stiffness SN10000 N/m2 should withstand deflection amounting to minimum 15% of its mean diameter. For an allowable deflection of 6% the value of this safety factor in new pipes amounts to at least 15/6 = 2.5. It was assessed that the pipes would meet this safety factor requirement also after a repair. It was also found (through numerical calculations) that linear strain ε in selected points of the reinforced area considerably exceeded the allowable values assumed in the various codes and studies concerning repairs: (AEA Tech, 2005), (ASME PCC-2, 2006). Moreover, one cannot ignore the fact that the required service life of the pipeline is 50 years and it much exceeds the repair durabilities adopted in (AEA Tech, 2005) and (ASME PCC-2, 2006), which amount to 2–20 years. 29

F y=0.03 dme

d

F

Figure 1.

(a)

(b)

Ring stiffness test. Deflection y = 0.03 × dme is used to determine stiffness SN.Allowable deflection amounts to 0.06 × dme . Symbol dme denotes mean diameter.

Similar disturbing conclusions emerge from the investigations described in (Farshad & Necola, 2004). On the basis of tests lasting up to 1000 h the regression line for pipes DN500/SN10000 loaded as in Figure 1, but in addition immersed in sulphuric acid with a concentration of 5%, was determined. The acid solution is used to model the effect of sewage on structural materials. It was found that hoop strain at the moment of failure decreases from 2.1% in the short-term test to 0.5% after 1000 h and to as little as 0.16% after 50 years. This is below the values assumed for the design of chemical-resistant piping systems (Table 1) in the standards which are highly conservative as regards the choice of allowable strain values. From the above investigations and analyses one can conclude that the maximum linear strains in the circumferential direction of the investigated pipes, loaded as in Figure 1, should not substantially differ from the values assumed in the design of chemical-resistant facilities. Although the corrosive effect of the water filling the analyzed pipeline is generally less harmful than that of sewage, studied by (Farshad & Necola, 2004), the required service life is very long (50 years). The technical aspects of repairs and the results of calculations for different material-structural designs of the composite repair are presented below.

4 DAMAGE AND REPAIR – TECHNICAL ASPECTS Depending on the extent of damage, several repair methods were proposed. The analyses presented here concern cases of quite extensive damage repaired through local composite reinforcements inside the pipe. The extent of damage in directions tangent to the inner surface mostly amounted to 100 mm (Figure 2), but in some cases it was larger (e.g. 100 × 250 mm in Figure 3). According to the ASME RTP-1 criteria for high-risk chemical-resistant facilities, the damage did not qualify for repair because of its extent. Also the criteria of acceptance for repairs of damage to low-risk facilities, described in ASME PCC-2 were exceeded. On the basis of its long operating experience the manufacturer of the pipes proposed its own procedures for assessing the damage and carrying out repairs. The damage affected the protective and structural plies of the pipes. The repair consisted in removing the damaged material (Figure 4), filling in the cavities with a polyester filler (UP) and making an inner reinforcing ring composed of five layers of polyester laminate with glass CSM with a mass of 600 g/m2 (Figure 4). The ring width towards the axis of the pipe was 400 mm. Finally, a protective layer made of resin was formed. 30

Figure 2. Damage at pipe’s end (37) and small impact damage distant from edge (17). Dark gap between pipes is about 15 mm wide.

Figure 3.

Cracks and chips in laminate.

Figure 4.

100 × 200 × 15 mm cut-out left after removal of cracked material.

Figure 4 shows that the void left after the removal of the material extends deeply into the structural layers, disturbing their integrity. An expert assessment showed that the pipe’s initial load capacity and stiffness in the circumferential direction could be recovered by means of the inner reinforcing ring. It was assumed that in a pipeline which does not carry axial pressure the axial load capacity and stiffness of the pipes are of secondary importance. 31

Figure 5.

Longitudinal section of pipe wall and denotations of individual layers.

Table 2.

Denotations and structure of pipe wall layers and reinforcement variants.

Layer denotation

Layer function

Structure and reinforcement

M1 M2 M3 M4 M5 M6 M7 M8 M9

outer protective outer structural pipe core inner structural inner protective resin filler repairing ring – variant A repairing ring – variant B repairing ring – variant C∗

UP resin with sand UP resin with staple fibre UP resin with sand as in M2 UP resin UP resin UP resin with CSM glass fabric WR + UP M7/M8/M9/M8/M7/M5 + UP

∗ In variant C the following system of reinforcement layers (moving from the contact with the pipe towards the ring inner surface): M7: 2.2 mm/M8: 0.84 mm/M9: 1.4 mm/M8: 0.84 mm/M7: 2.2 mm/M5: 0.5 mm was adopted.

5 MATERIAL PROPERTIES AND PIPE MODEL Figure 5 shows the structure of the wall of the repaired pipes. Thickness t was 34 mm. Layers M1, M2. . . are described in tables 1, 2. It follows from the structure of the reinforcement that most of the considered layers can be regarded as isotropic in volume (M1, M3, M5, M6) or in plane (M2, M4, M7). Only layer M8 and the laminate in variant C, representing improvements in the composite repair, are moderately anisotropic. The values of Poisson ratio ν and Young’s modulus (E) of the layers are give in Table 2. First the 300 mm long undamaged pipe section was modelled in ANSYS. Using SOLID 45 elements with 8 nodes the system of layers M1/M2/M3/M4/M5 was modelled starting from the outside and moving towards the inner surface (Figure 5). The layers are described in Tables 2 and 3. There were about 36000 elements. The aim of the calculations was to match the thickness and properties of the particular layers so as to obtain the actual pipe ring stiffness close to the nominal one (SN10000 N/m2 ). In the ring stiffness test, the ring cut off the pipe is loaded as in Figure 1 until deflection 0.03 × dme = 42 mm is reached in the case of the considered pipes. It was assumed that the allowable pipe deflection is 0.06 × dme = 84 mm (Madryas et al. 2002, EN 14364). When the properties of layers M1–M5 given in Table 3 and the load twice as heavy as the one corresponding to the rated ring stiffness were substituted the maximum deflection of 75 mm was obtained. The ring stiffness of the pipe model is about 11100 N/m2 and it is higher than the nominal one (10000 N/m2 ). The stiffness of new pipes is as a rule higher than the rated stiffness. 32

Table 3. Thickness of layers and adopted proper mechanical properties. Layer symbol

Layer thickness (mm)

Young’s modulus∗ (GPa)

Poisson ratio∗∗

M1 M2 M3 M4 M5 M6 M7 M8 M9*

1.9 2.4 19.6 7.7 2.4 10.1 7.25 7.25 0.84

12.8 8.8 10.5 8.8 3.0 3.0 8.0 18.0 35.0 and 10.4

0.23 0.3 0.23 0.3 0.3 0.4 0.2 0.13 0.281 and 0.083

∗

For the unidirectional layers (UD) axial and transverse Young’s moduli are given; the same applies to the Poisson coefficients. Longitudinal elasticity modulus G = 3.2 GPa was assumed for layers M9.

Figure 6. View of reinforcing ring model.

The extreme hoop strains for the deflection of 75 mm amounted to: εmax = 0.607 × 10−2 and εmin = −0.598 × 10−2 . Assuming failure strain under bending εB = 1.6 − 2.2 × 10−2 (according to the manufacturer specifications) one can find that the safety factor amounts to 0.606/1.6 = 2.64 and it is close to the one required by standard EN 14364. Numerically determined strain εmax = 0.607 × 10−2 exceeds the strain values assumed for high-risk chemical-resistant facilities (Table 1), but the analyzed pipeline does not belong to this class of equipment.

6 MODELLING OF COMPOSITE REPAIR 6.1 Inner ring reinforced with glass mat Then a model of a situation similar to the one shown in Figure 3 (200 × 200 mm cavity with its depth equal to the sum of layer M4 and M5 thicknesses, located in the pipe’s highest place) was created in ANSYS. Under the load as in Figure 1, the cavity is the area of the highest bending moment. The cavity was filled with a material having the properties of layer M3. Then a reinforcing ring with width b = 400 mm and thickness t1 = 7.25 mm, made of a material with the properties of layer M7 (Tables 2, 3, Figure 6) was introduced into the pipe. In Table 2 this is denoted as variant A (ring W7/filler W3). According to standard ASME RTP-1, five layers reinforced with CSM with a mass of 600 g/m2 have the above thickness, which corresponds to the structure of the rings formed during the repair. The protective layer formed from resin and surfacing veil, laid as the last one, was not taken into account in the calculations. 33

Figure 7.

Locations of selected analyzed points in repair area.

Certain simplifications were made in the analyzed model of the pipe under repair, i.e. • the influence of the end-of-pipe location of the repaired damage was neglected and • the loading scheme used for determining ring stiffness (Figure 1) was adopted. At a pipe deflection of 55 mm, the hoop strains in the most stressed points of the ring, denoted as 4 and 8 in Figure 7, amounted to: ε(4) = 0.494 × 10−2 and ε(8) = 0.564 × 10−2 . The highest hoop strains ε in material M3 filling the cut-out left after the damage had been removed amounted to 0.342 × 10−2 . When filling material M3 was replaced by the less stiff filler M6 (variant A1), the hoop strain in point 4 increased by 18.4% (up to 0.585 × 10−2 ) and in point 8 by 2.8% (up to 0.580 × 10−2 ). Hoop strain also increased in the filling – its maximum value in material W6 was 0.395 × 10−2 . Even though the stiffness of the pipe after the repair is considerably higher than the required one, the calculated hoop strain values in the ring are quite high – amounting to about 0.5–0.6 × 102 . 6.2 Inner ring reinforced with glass fabric M8 Similar calculations were done for variant B (ring M8/filling M3 as in table 2). The main difference with variant A is that the ring is reinforced with material M8 with a higher stiffness than that of the previously used M7. In comparison with the results obtained for variant A, hoop strain decreased in the ring’s points 4 and 8 down to respectively 0.356 × 10−2 and 0.416 × 10−2 . As a result of the replacement of M7 by M8, the value of εmax in the ring decreased by (0.564 − 0.416)/0.564 = 26%. Hoop strain εmax in the filling in variant B is lower than in variant A (previously it amounted 0.342 × 10−2 and now it is 0.247 × 10−2 ). Preserving ring reinforcement, M8 filling material M3 was replaced by M6 (variant B1). Similarly as in variants A and A1, this resulted in an increase of hoop strain εmax in the filling area from 0.231 × 10−2 to 0.247 × 10−2 (by 7%). Strains εmax in the ring also increased. 6.3 Inner ring reinforced with laminate containing UD layer Then reinforcing ring structure variant C described in Tables 2 and 3 was calculated. This ring structure takes into account the general principles of designing and repairing chemical-resistant facilities. The introduction of a UD layer with a circumferential orientation did not result in any 34

Table 4.

Strains and deflections before and after increase in ring thickness.

Reinforcement variant A A1 B B1 A × 1.5 A1 × 1.5 B × 1.5 B1 × 1.5

Ring thickness (mm)

Strain εmax in ring

Strain εmax in filling

Deflection (mm)

7.25 7.25 7.25 7.25

0.564 × 10−2 0.585 × 10−2 0.416 × 10−2 0.422 × 10−2

55 55 44 44

10.90 10.90 10.90 10.90

0.529 × 10−2 0.538 × 10−2 0.360 × 10−2 0.362 × 10−2

0.342 × 10−2 0.395 × 10−2 0.231 × 10−2 0.247 × 10−2 0.263 × 10–2 0.287 × 10–2 0.152 × 10–2 0.157×10–2

46 46 35 35

significant reduction of strain in the ring and in the filling in comparison with variants A and B. In point 4 strain amounts to ε = 0.381 × 10−2 while in point 8 to ε = 0.445 × 10−2 . In the area filled with material M3 strain amounts εmax = 0.248 × 10−2 . 6.4 Effect of increased ring thickness In the considered reinforcing ring structure variants (A, B, C), pipe stiffness in the reinforcement area was sufficient, but the maximum strain, amounting to 0.4−0.6 × 10−2 , was relatively high (section 3 in the paper). Therefore it was decided to determine what effect an increase in wall thickness by 50% will have on hoop strain. In discussed variants A, A1, B, B1 the ring thickness was 7.25 mm. Now it was assumed to be 1090 mm. The obtained calculation results are shown in Table 4. 7 CONCLUSIONS The analysis of the literature on piping system repairs by means of polyester-glass composites shows that strains in the repaired components of low-risk piping systems should not significantly diverge from the allowable strain values assumed for high-risk chemical-resistant equipment (table 1). Allowable values ε = 0.25–0.3 × 10−2 are proposed. The stiffness of glass CSM is insufficient for it to be used in the structural layer of such rings reinforcing the repair area. Glass fabric provides more effective reinforcement, significantly reducing hoop strains. The elastic properties of the material filling the cut-out left after damage removal affected only slightly the hoop stress and pipe deflection values. The introduction of a UD layer into the ring structure did not result in a substantial reduction in hoop stress. This may be ascribed to the fact that a very small quantity of the UD reinforcement was used. When the ring thickness was increased by 50%, the reduction in hoop stress was smaller than expected. It seems that hoop stress can be further reduced by increasing the ring’s thickness or the UD reinforcement fraction in its structure. The obtained results show that the optimization of such composite repairs has great potential. The paper does not cover problems relating to possible improvements in repair techniques, the effectiveness of the latter or durability assessments. REFERENCES AD 2000-Merkblatt. 2000. Pressure vessels in glass fibre reinforced thermosetting plastics. AEA Technology. 2005. Design of Composite Repairs for Pipework.

35

ASME PCC-2 Repair standard. 2006. Non-metallic composite repair systems for pipelines and pipework. ASME RTP-1. 2000. Reinforced thermoset plastic corrosion resistant equipment. ASTM D 2992. 1996. Standard Practice for Obtaining Hydrostatic or Pressure Design Basis for Fiberglass” (Glass-Fiber-Reinforced Thermosetting-Resin) Pipe and Fittings. ASTM D3299. 1995. Standard Specification for Filament-Wound Glass-Fiber-Reinforced Thermoset Resin Corrosion-Resistant Tanks. Bełzowski A. 2005. Koncepcje oceny wytrzymało´sci długotrwałej polimerowych kompozytów konstrukcyjnych. V Konferencja KOMPOZYTY POLIMEROWE.: 5–37. Politechnika Warszawska. Bełzowski A. 2004. Badanie uszkodze´n zbiornika z laminatu wzmocnionego włóknem szklanym. Raport SPR, nr 9. Politechnika Wrocławska. BełzowskiA. & Stró˙zyk P. 2008. Assessment of repair reinforcement of polyester-glass fibre pipe. Composites 8 nr 2: 179–184. Polish Society for Composite Materials. Bolleart F. & Lemasçon A. 1999. Analyse de défaillance pièces plastiques, élastomères ou composites. Guide Pratique. CETIM, France. BS 4994. 1987. Specification for design and construction of vessels and tanks in reinforced plastics. BS 7159. 1989. Code of practice for design and construction of glass-reinforced plastic piping systems. ECKOLD G. 1985. A design method for filament wound GRP pressure vessels and pipework. Composites, V. 16, Nr 1: 41–47. EN 1120. 2000. Plastics piping systems – Glass-reinforced thermosetting plastics (GRP) pipes and fittings – Determination of the resistance to chemical attack from the inside of a section in a deflected condition. EN 1796. 2006. Plastic piping systems for water supply with or without pressure. Glass-reinforced themosetting plastics (GRP) based on unsaturated polyester resin (UP). EN 13121. 2001. GRP tanks and vessels for use above ground. EN 14364. 2006. Plastic piping systems for drainage or sewerage with or without pressure. Glass-reinforced thermosetting plastics (GRP) based on unsaturated polyester resin (UP). Specification for pipes, fittings and joints. FARSHAD M., NECOLA A. 2004. Strain corrosion of glass fibre-reinforced plastics pipes. Polymer testing, 23: 517–521. Le Courtois T. 1995. PWR Composite Material Use: A Particular Case of Safety-Related Service Water Pipes. Proc. of Enercomp 95: 835–843. Technomic Pub., Montreal. MADRYAS C., KOLONKO A. WYSOCKI L. 2002. Konstrukcje przewodów kanalizacyjnych. Oficyna Wyd. Politechniki Wrocławskiej, Wrocław. MYERS T.J., KYTÖMAA H.K. SMITH T.R. 2007. Environmental stress-corrosion cracking of fiberglass: Lessons learned from failures in the chemical industry. J. of Hazardous Mater. 142: 695–704. NFT 57900. 1987. Réservoirs et appareils en matiéres plastiques renforcées. Tuttle M.T. 1996. A framework for long-term durability predictions of polymeric composites. Progress in Durability Analysis of Composite Systems: 169–176. Balkema, Rotterdam. ˛ URZAD ci´snieniowe. Stałe zbiorniki ci´snieniowe z tworzyw ˛ DOZORU TECHNICZNEGO. 2003. Urzadzenia sztucznych wzmocnionych włóknem szklanym. WUDT-UC-UTS/01.

36

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Repair of RC oil contaminated elements in case of infrastructure T.Z. Błaszczy´nski Civil and Environmental Engineering Faculty, Pozna´n University of Technology, Pozna´n, Poland

ABSTRACT: RC elements in case of underground and network infrastructure, are subjected to the activity of different agents, i.e. water, chemicals and oil-products. Because of it so important is their durability. Indispensable are then suitable repairs, because the possibility of the underground exchange of elements is comparatively difficult and expensive. In the relationship with the environment which can appear in objects of the infrastructure, repair methods must be resistant to the activity of different environments, and must also be prepared to the repair of concrete surfaces with the different state: moisture and oiling. Comparing the influence of various oil products on compressive strength of concrete, leads to the conclusion that there are large differences in effects. The repair of oiled structures is technologically difficult. The paper will present research works on some nano- and modern technologies in case of oiled RC structures.

1 INTRODUCTION In view of environmental influences and structural solutions the separate repair procedure often is also required. If one takes under consideration that concrete is a product of simple technology and complicated knowledge which begins to be only mastered, then right prognoses are, that XXI century for the building construction can be in large measure the age of repairs, rehabilitations and demolitions. The concrete is relatively cheap material in construction, however its repair during use of the property is very costly. Especially depressing and expensive, and sometimes even operable impossible, is the repair of already completed repair. Repairs became a civilisation-wide problem (Kucharska 2001). The acid content of crude oil products might be affected at the oil/water interface as a result of bacterial activity. Aerobic oil oxiding bacteria (OBB) produce acetic acid, which can reduce the pH level of water to 5. In addition, anaerobic sulphate reducing bacteria (SRB) also operate to produce H2 S, which can be converted to H2 S04 . This latter process has often been encountered in sewers. Research at Imperial College has indicated that the inorganic acid content (Onabolu at all 1985) of the crude oil products is unlikely to increase as a result of bacterial activity. That kind of environment create three different corrosion mechanism: biological, chemical and physico-chemical (Fig. 1). The less known is physico-chemical process, which take place in case of oil environment with low neutralisation number.

2 MINERAL OILS INFLUENCE ON CONCRETE Long term laboratory experiments have been conducted to assess the changes of physicomechanical characteristics of oil contaminated concrete. The compressive strength was determined from 100 mm cubes, according to EN 12390-1:2001, for concrete type C20/25 as most commonly used for industrial RC structures in Poland (5 specimens, sfc = 0.84–2.87, νfc = 2.25%−6.99%). The average 28 day compressive strength of concrete was fcm = 29.8 MPa. The water-cement ratio was 0.59 and aggregate-cement ratio was 6.70. 37

Figure 1. An example of oil/water environment in case of RC structure.

fc [MPa] 60 50 40 30 20 10 0 0 4 8 12 16 20 24 28 32 36 48 60 64 68 tz [months] 72

Samples M-40 TU-20 H-70

Figure 2. Variation of concrete C20/25 compressive strength during the period of exposure to H70, TU20 and M40 oils (Błaszczy´nski 2002).

Concrete was impregnated with the most commonly used industrial oils of different kinematic viscosities namely turbine oil TU20 (81 mm2 /s), machine oil M40 (211 mm2 /s) and hydraulic oil H70 (383 mm2 /s). These oils have low neutralisation numbers with values between 0.05 and 0.075 mgKOH/g. The oils was first applied to concrete 2 months after casting, subsequently the specimens were examined every 4 or 12 months during total period 72 months. The control specimens (samples) were additionally examined after 28 days and 2 months (Fig. 2). Time of oiling tz is started in age of two months with an average compressive strength of concrete fcm = 37.35 MPa. The results clearly show, that as a result of the influence of the oils used, the different degree of decrease of concrete compressive strength (comparing to control samples): from 55% for oil H70 to almost no influence for oil M40. Oils H70 and TU20 affected the analysed concrete compressive strength fcm the most. Contamination of concrete by hydrocarbons gives an almost new material, which behaves differently. The results of the stress (σc ) – strain (εc ) relation in non-oiled state and after 12-months of oiling by mineral oil TU20 for concrete C20/25 in function of the longitudinal strains are different. The non-linear behaviour of strength and strain variations depend on the contents of hydrocarbon and its type. It can be noticed that the strain εcl , corresponding to the maximum stress, is lower for oil saturated concrete than for non-oiled concrete (Fig. 3 – Błaszczy´nski 2006). 38

45 40 35

 c [MPa]

30 non-oiled

25 oiled

20 15 10 5 0 0,0

0,5

1,0

1,5
c [‰]

2,0

2,5

3,0

Figure 3. σc − εc diagram for non-oiled and oiled by oil TU20 concrete C20/25.

0.5 mm 1.0 mm Oiled section Fz [%]

Oiled section Fz [%]

0.05 mm 0.2 mm 100

50

100

a1 Crack width 0.3 mm

50

0

0 1

2 3 4 5

0

10 20 30 40

Figure 4.

0.05

0.10

0.15

0.20

0.25

Distance from crack a1 [m.]

Time of oiling tz [days]

Percentage of section oiling (TU20): a/ in case of different crack width, b/ in section grow away from crack (Błaszczy´nski 1983).

3 OIL PRODUCTS PENETRATION INTO RC ELEMENTS During the research on the influence of mineral oil penetration into crack’s zones, it was shown that oiling kinetics in case of smaller cracks is greater (Fig. 4.a). Besides of this also intensive oiling of all the crack region occurred (Fig. 4.b). This mechanism is a basic mechanism of the oil products penetration into RC elements. The permeability through porous material, like concrete, can be defined as the flow of liquid through capillary channels, pores aerial and others (about different sizes, often joint with itself by the net of microcracks) (Matti 1976). The knowledge of concrete permeability is important, because it exerts the influence on the resistance of aggressive liquid environment, and consequently decides – along with the reactivity parameter – about its durability. The concrete permeability in the greater degree is relative to pores structures than to the general porosity (Kagimoto 2000). 39

80

Depth of oiling [%]

Depth of oiling [%]

100

non oiled 60

partialy oiled

40 20

oiled

0 4.5

9.1

85.7

87

75

1198

Kinematic viscosity ηk [mm/s2]

Figure 5.

100

TU20

M40

30 60 90

30 60 90 30

H70 60 90

Time of oiling tz [dni]

Mineral oil kinematic viscosity influence on depth of section oiling: a/ concrete after 13 months of oiling (Manns 1977), b/ cement mortar after 3 months of oiling (Błaszczy´nski 1983).

The permeability of so heterogeneous material surely is relative to its internal structure, however insignificant is also viscosity of interfering liquid ηk , what is represented in Fig. 5 (Błaszczy´nski 1983, Manns 1977).

4 REPAIR TECHNOLOGY OF MINERAL OIL CONTAMINATED RC STRUCTURES There can be many causes of deterioration in a concrete structure. Concrete repair is a specialist activity requiring fully trained and competent personnel at all stages of the process. Up until now there has been no common European Standard in this field. Often simple “patch and paint” strategies have been employed as short term cosmetic repairs which have failed to address the root cause of the problem. This can, and has, lead to dissatisfaction from building and structure owners. The new European Norm EN 1504 will standardise repair activities and provide an improved framework for achieving successful, durable repairs and satisfied clients. Importantly, this Norm (expected date of full implementation: 31.12.2008) will deal with all aspects of the repair process including: • • • •

definitions and repair principles, the need for accurate diagnosis of causes before specification of the repair method, detailed understanding of the needs of the client, product performance requirements, test methods, material production control and evaluation of conformity, • site application methods and quality control of works. All of this especially difficult in case of oiled structures. Today we have got many modern technologies even nanotechnologies. Nanotechnology does not mean nano-sized particles. The better understanding of cement hydration has allowed us to improve the quality and density of the nanostructures in cement paste. This reduces micro-defects in the systems and improves bond between the cement matrix and the aggregate, and, the cement mortar with the substrate. Physical properties such as tensile strength are improved to reduce the possibility of cracking. This is the basis of applied nanotechnology in cement systems. The purpose of all range of superficial treatments is the assurance of suitable adhesion of repair material. This decides often about the efficiency of the repair. For structural repairs is recommended, and according to ENV 1504-9 even required, the usage of bond layers. For the purpose of detailed analysis of above problems two series of tests was done. In the first series of tests the efficiency of crack repair methods was taking into account. In the second series the efficiency of surface protection method was tested. Before using any repair treatment on oiled 40

(b)

(a)

Figure 6.

(c)

(b)

Oiled concrete surface in case of split samples treated with modern paste based on solvent with absorptive solid material: a/ splited sample, b/ surface after treatment, c/ surface after steel-brushing. (b)

(a)

Figure 8.

(d)

Oiled concrete surface treated with modern paste based on solvent with absorptive solid material: a/ oiled surface, b/ first treatment, c/ second treatment, d/ last treatment.

(a)

Figure 7.

(c)

Split test in case of concrete cylinders: a/ oiled concrete before bonding, b/ bonded with epoxy resin.

concrete surface the proper surface preparation is necessary. This preparation should give clean surface without any oil particles on it. 4.1 Surface preparation In case of oiled surface cleaning is necessary. There was used two types of cleaning one typical with the emulsifying medium and the second modern with the paste based on solvent with absorptive solid material, treated on the surface three times (Fig. 6). After each treatment concrete was cleaned with steel-brushing. The same treatment was done in case of split sample, necessary for crack repair testing (Fig. 7). 4.2 Cracks repairs Crack repair tests was done based on cylinders (160 × 160 mm) using split test. After 6 years of oil TU20 influence cylinders was splited and bonded by two kinds of resin: epoxy and polyurethane (both the most popular in crack repairing). After week time the bonded cylinders was also splited (Fig. 8). 41

Tensile strength [MPa]

2,5 2 Sample Epoxy resin Polyurethan resin

1,5 1 0,5 0

Figure 9.

Figure 10.

Split test results of oiled concrete cylinders bonded with two kinds of resins.

Examination of analysed layer adhesion to concrete surface.

Figure 9 is presenting the average split tensile strength of all kind of analysed elements (6 samples each): oiled cylinder, oiled cylinder bonded with epoxy resin and oiled cylinder bonded with polyurethane resin. The split test results clearly show, that in case of both resins crack in oiled concrete could be repaired with success. Better connection is of course in case of epoxy resin (20% of original split tensile strength loss), but the crack connection made by polyurethane resin gives only 38% of original split tensile strength loss. Comparing the cost of both these resins satisfactory effect for polyurethane resin is visible. 4.3 Surface protection Apart from crack repairing the surface protection is also necessary, both in case of new structures or cleaned and repaired. The variety of surface protection materials is high. For testing procedure paintwork based on copolymer vinyl-acetyl-ethylene, epoxy and polyurethane coat, was used. For the purpose of surface repairing two kind of cement repair material was checked. One ordinary and one nanomaterial. Both were single component. The nano repair layer was very high strength, high modulus, fibre reinforced, shrinkage compensated, expansive, structural repair mortar. On prepared samples with pull-off test adhesion of analysed layer to oiled (TU20) and clean concrete surface was examined. The method of research consisted in the measurement of minimal force applied perpendicularly to sample layer (Fig. 10). 42

clean concrete

oiled concrete

clean concrete

oiled concrete

3 Separating stress [MPa]

Separating stress [MPa]

2,5 2 1,5 1 0,5

2,5 2 1,5 1 0,5 0

0

a

b

a

a – nano system b – ordinary system

b

c

a – epoxy coat b – vinyl- acetyl-ethylene copolymer c – polyurethane coat

Figure 11. Adhesion (separation) stress results: a/ repair layer, b/ paint layer.

Obtained results testify hereof that values of bond force to oiled concrete surface are lower aside from all used surface materials (Fig. 11). The required adhesion, measured by pull-off test, for not structural repairs amounts above 0,5 MPa, and for structural repairs must be above 1,5 MPa. Comparing all results is clear, that in case of repair materials only single component nano-material has still adhesion above this value. In case of paintworks all of them have value of oiled surface adhesion still higher, then in above technical recommendations.

5 CONCLUSIONS The results clearly show the different degree of decrease of concrete compressive strength (comparing to control samples): from 55% for oil H70 to almost no influence for oil M40. Oils H70 and TU20 affected the analysed concrete compressive strength fcm the most. Contamination of concrete by hydrocarbons gives an almost new material, which behaves differently. The non-linear behaviour of strength and strain variations depend on the contents of hydrocarbon and its type. The split test results clearly show, that in case of both used resins crack in oiled concrete could be repaired with success. Better connection is of course in case of epoxy resin (20% of original split tensile strength loss), but the crack connection made by polyurethane resin gives only 38% of original split tensile strength loss. Comparing the cost of both these resins satisfactory effect for polyurethane resin is visible. Apart from crack repairing the surface protection is also necessary, both in case of new structures or cleaned and repaired. In case of repair materials only single component nano-material has still adhesion above the technical recommendations. All of tested paintworks have value of oiled surface adhesion still higher then necessary. REFERENCES Błaszczy´nski T., Kozaczewski J., Nowakowski B. 1983. About the mineral oils influence on physical and bearing concrete features, Proceedings of XXIX Polish Scientific Conference, Pozna´n – Krynica, 11–16, (in Polish). Błaszczy´nski T. 1994. Durability analysis of RC structures exposed to a physico-chemical environment, Proceedings of the Third Kerensky Conference, Singapore, 67–70. Błaszczy´nski T. 2002. Some effects of crude oil environment on RC structures, Foundation of Civil and Environmental Engineering, (2): 7–14.

43

´ Błaszczy´nski T. & Scigałło J. 2006. Ultimate bearing capacity assessment of RC sections under mineral oil exposure, Archives of Civil and Mechanical Engineering: 41–56. Kagimoto H., Sato M., Kawamura M. 2000. Evaluation of degree of ASR deterioration in concrete and analysis of pore solutions, Concrete Library International, (36): 480–493. Kucharska L. 2001. Failures and damages of concrete structures and its development, Proceedings of XX Structural Failure Conference, Szczecin-Mie˛dzyzdroje, 89–118, (in Polish). Manns W. & Hartmann E. 1977. Zum Einfluss von Mineralölen auf die Festigkeit von Beton, Schriften-reiche des DAfStb, Ernst, H. 289. Matti M.A. 1976. Some properties and permeability of concrete in direct contact with crude oil. Ph.D. Thesis, University of Sheffield. Onabolu O.A., Khoury G.A., Sullivan P.J.E., Sterritt R. 1985. Inorganic acid contents of NS oil: effect of anaerobic bacterial activity in tanks, Petroleum Review, (1): 42–45.

44

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Modelling the behaviour of a micro-tunnelling machine due to steering corrections W. Broere Delft University of Technology, Delft, The Netherlands A. Broere BV, Amsterdam, The Netherlands

J. Dijkstra & G. Arends Delft University of Technology, Delft, The Netherlands

ABSTRACT: Micro-tunnelling often encounters restrictions expanding into new areas. In the Netherlands, and more general in soft soils, one of the problems is the controlled boring of curves. To better understand the behaviour of the tunnel boring machine in such conditions, an analytical model has been developed, that takes translation and rotation of the TBM into account. The model takes the subgrade reaction and the stiffness of the soil into account as major parameters describing the soil behaviour. It is shown that the subgrade reaction in the inner and outer curve of the TBM differs substantially and a subgrade reduction factor is introduced in the model to deal with this effect. A first derivation of the subgrade reduction factor was made comparing field measurements with model parameter variations. In this paper a full three-dimensional finite element simulation is presented, using large deformation analysis to model the movement of the TBM through the soil, which has been used to study the behaviour of the micro-tunnel machine in more detail and derive the subgrade reduction factor.

1 INTRODUCTION In most of the Netherlands the upper layers consist of alluvial soil deposits, where the stiffness of the soil is low. The groundwater level is generally very high and locally reaches the ground surface. This combination has adverse implications for the drilling techniques used when construction tunnels using the micro-tunnelling technique. In such poor soil conditions, the control and steerability of the micro-tunnelling machine can become problematic (see e.g. Oreste et al., 2002). One aspect, the boring of curves, is described in more detail in this paper. In the past several borings were executed in soft soils without any significant problems, or only with limited problems (Broere et al., 2007). Recently however, problems occurred on a project where a curve was introduced in the trajectory coinciding with the transition from very soft to stiffer soils. At this location the concrete pipe snapped when the tunnel boring machine (TBM) had just entered the stiffer soils. A second boring, with a greater curvature, was successfully completed at the same location. This event raised questions concerning the actual behaviour of the TBM, the concrete pipes and the coupling forces between the pipes. To increase understanding of the behaviour of the TBM in soft soils, an analytical model was formulated that describes the behaviour of the TBM during the boring of curves in soft soils. In this model the behaviour of the soil is modelled as a subgrade reaction modulus. To correctly predict the TBM behaviour, a reduced subgrade reaction modulus is needed in the inner curve as compared to the outer curve. The necessary subgrade reduction factor has been derived first based on field measurements at the aforementioned project. 45



Figure 1.

 

Movement of the TBM is split into a rotational and translational mode. These are combined to calculate the reaction w of the TBM to a steering action a.

In order to study the soil behaviour around the TBM in more detail and derive the subgrade reduction factor, independently from the field measurements, a finite element analysis has been used that models the movement of the TBM through the soil during a curved boring.

2 ANALYTICAL MODEL OF A MICRO-TUNNELLING MACHINE IN CURVES In order to model the behaviour of a micro-tunnel machine during the boring of a curve in (very) soft soil, an analytical model has been developed by Broere et al. (2007). In this model the movement of the TBM through the soil is split in a translational mode and a rotational mode, as shown in Figure 1. For both modes the forces acting on the TBM are determined and the torque resulting from these forces is calculated. Supposing the soil is elastic, the forces can be superposed and the reaction of the machine can be determined from moment equilibrium. In soft soils often the reaction of the machine ω is less than the intended steering action α. Of the forces acting on the machine, the jack forces, the normal forces between soil and shield and the face support are the most important. Both the jack forces and the face support are considered as given, what is left is to determine the normal forces in the soil. The linear-elastic soil behaviour is characterized by a subgrade reaction modulus k. For a given steering angle α the normal force Fsoil and the reaction ω can then be calculated. See Broere et al. (2007) for full details of the model derivation. When the results of the model were compared to field measurements, it became clear that a different subgrade reaction modulus was needed at the inner and outer curves of the TBM. This is understandable as the TBM will excavate a slightly oval-shaped hole in order to make a curve. On one side the soil will not be excavated but rather displaced sideways, whereas on the other side an over-excavated zone exists where there is no direct contact between TBM and soil and some relaxation of the soil will occur. In this area the soil will have a lower subgrade reaction modulus, which is included in the model by introducing a subgrade reduction factor Cf for this area. Based on field observations in soft soils and realistic input parameters, a possible range of Cf = 0.15 to 0.35 was determined. Cf = 0.25 was selected for further parameter studies of the 46

3 2.5  [˚] 2

2

1.5 1

1

0.5

0.5

0

Figure 2.

2000

4000

6000

8000 k

10000

12000 14000 k [kN/m3]

Reaction of the TBM ω as a function of the subgrade reaction modulus k and steering action α.

subgrade reaction modulus k and the magnitude of the steering angle α. The results are given in Figure 2, which shows that in stiff sand layers (k = 10,000 kN/m3 or higher) the TBM reacts well to desired steering action. In softer (clay) layers (k = 2000 – 4000 kN/m3) the machine does not follow the desired steering angle completely, but can still be controlled. Only in even softer clay and peat layers the machine will hardly react to any steering actions taken. As the subgrade reduction factor Cf = 0.25 was determined for soft to very soft soils and has a significant influence on the results, a more detailed determination of this value is warranted. A numerical simulation of a curved boring is therefore made to derive the subgrade reduction factor independently.

3 NUMERICAL SIMULATION The curved boring of a micro-tunnel machine is simulated in the numerical framework FEAT, capable of a full Eulerian large strain analysis as well as traditional Lagrangian finite element calculations (TOCHNOG, 2007). In the Eulerian scheme the material flow and the mesh are decoupled. This allows for extremely large deformations to be modelled, whilst the calculations remain numerically stable. This approach has been used previously to model the jacked installation of piles (Dijkstra et al., 2007). The TBM, with a front and back part of the machine, and the first three tunnel segments are included in the simulation. In modelling the excavation and steering process, a two-step approach is taken. First, an Eulerian calculation is made in which a tunnel boring machine on a straight alignment is simulated. In this stage a forward movement of at least the length of the finite element mesh is simulated, to make sure that steady-state conditions are reached. This stage yields a proper stress and strain field around the TBM and tunnel and is used as the starting point for the next phase. In the second phase the front part of the TBM is rotated with respect to the back part until a 1 degree rotation is reached. This simulates the actual steering of the TBM. Additionally, a reference calculation is made in which the Eulerian inflow of the soil is skipped and only the initial conditions are set and the steering phase is simulated. 47

Figure 3.

Mesh and boundary conditions for the numerical simulation.

3.1 Details of the numerical schematisation In order to create a finite element mesh that includes the proper boundary conditions, a two-step approach is used. First, an axi-symmetric mesh is created, consisting of first order quadrilateral elements. In this mesh, the tunnel and TBM are situated on the axis of symmetry. In order to prevent mesh problems, a small gap is maintained around the axis. All elements, the soil as well as the TBM and tunnel, are represented by volume elements. No plate elements of any kind are used. In this mesh, the TBM is 6 meters long and 2.2 meters in diameter. In the middle, separating the front and back part of the machine, a slice of 10 cm thickness is included that will be used to model the steering jacks. Behind the machine 15 meters of tunnel with an outer diameter of 2.1 meters and inner diameter of 1.8 meters is modelled. This approach results in a tail void of 5 cm behind the TBM. The mesh is extended 10 m from the axis of symmetry and 5 m in front of the TBM. This ensures boundary conditions are sufficiently far removed from the tunnel. Subsequently, the mesh is rotated around the symmetry axis to create a fully three-dimensional mesh consisting of 36 10◦ wedges. The intentional gap on the axis is remeshed and merged with the full mesh at this stage. Following this approach, a numerically stable mesh as shown in Figure 3 is obtained. As shown in Figure 3, the outer boundaries of the mesh are considered fixed. The same hold for the tail end of the last tunnel segment. At the right, in front of the tunnel face, soil flows into the mesh at a steady velocity of 0.01 m/s. Soil leaves the mesh at the left boundary. In order to (somewhat crudely, but effectively) model the excavation process, soil is also removed from the mesh at the tunnel face, such that soil that would have flowed into the tunnel is removed from the calculation. Finally, in order to keep the calculation stable, a distributed force of 50 kN/m2 is used on the boundaries where soil flows out of the mesh. For the discretisation in time Euler backward time stepping is used, because of its high numerical stability. For the Eulerian formulation, where mesh and material state are decoupled, the convective terms (also known as state parameters) need to be transferred through the mesh. This is done by a Streamline Upwind Galerkin method. In this method material and state parameters are calculated in the nodes. 48

jacking ring 50 mm tail void

D_TBM = 2.2 m D_lining = 2.1 m

Figure 4.

Close-up of the boundary conditions around the TBM.

Table 1.

Material parameters for the soil.

Name

Symbol

Value

Unit

Dry volumetric weight Cohesion Friction angle Dilatation angle Young’s modulus Poisson’s ratio

γ c φ ψ E ν

17 1 35 0 30 0.3

kN/m3 kPa

◦ ◦

MN/m2 –

After 30 m of material flow (i.e. 30 meter of forward movement of the TBM and tunnel has been simulated), the steering action is simulated. In this stage the material flow is stopped. The steering action of the TBM is simulated by the thin slice separating the front and back of the machine, similar to the steering jacks in an articulated shield (see Figure 4). An asymmetric velocity field is applied to a single wedge of this slice, extending this wedge by 10 mm at a rate of 0.1 mm/s. Given the small deformations and in order to improve accuracy, the integration scheme is switched to a normal Lagrangian scheme. Therefore, the material and state parameters are calculated in the integration points and need to be interpolated from the nodes before the start of this calculation phase. Initial stress conditions are set to the horizontal effective stress representative for a 10 m deep tunnel, although these will be completely replaced at the end of the first calculation phase and before the steering action is modelled. The influence of pore water is not modelled in the current calculation. In the reference calculation, the Eulerian phase is skipped. The same boundary conditions are used, except for the equilibrium force on the outflow boundaries. These are replaced by a fixed displacement boundary condition, which are sufficiently far away from the area of interest as not to influence the result.

3.2 Constitutive model The soil is modelled using the Mohr-Coulomb model, with parameters listed in Table 1. A tension cut-off was used to prevent tensile stresses in the soil. Also, to prevent problems with large volumetric strains occuring during large deformations, dilatant behaviour of the soil was prevented, setting ψ = 0. 49

Figure 5. Vertical total stress at the end of the Eulerian inflow phase.

Both the TBM and the tunnel lining are modelled as a linear elastic material with Poisson’s ratio ν of 0.2 and Young’s moduli E of 23 GPa for the concrete lining and 200 GPa for the steel of the TBM. 4 NUMERICAL MODEL RESULTS 4.1 First (Eulerian) inflow phase The vertical total stress at the end of the first phase, i.e. after 30 meters of forward movement of the TBM is given in Figure 5. Figure 6 shows the plastic shear strains. In these figures, a cross-section through the soil is made, although the full tunnel is plotted, in order to highlight the results at the interface between the soil and the tunnel. A large stress increase is seen in the soil directly in front of the tunnel face, which peaks at the front thick edges of the TBM. This increase is due to the local boundary conditions and should not be considered a correct representation of the excavation process. However, compared to a closed face where all soil would flow around the TBM it is a significant improvement of the soil behaviour. It can also be observed in Figure 5 that stress increases as the soil flows around the TBM and decreases again at the tail end of the TBM. This is due to the distinct tail void modelled in the analysis, which causes a partial unload of the soil as it flows into the tail void. Similar behaviour can be observed in the shear strain plot (Figure 6). Large strains occur where the soil is forced around the TBM and the friction between the TBM and the soil causes further shear strain to develop. Once the soil is free to expand into the tail void and relax the magnitude of the strain decreases. Again, the tail void dominates the observed behaviour of the soil around the TBM. 4.2 Steering action phase In Figure 7 the normal component of the total stress acting on the tunnel, at the end of the steering action, is shown. At this point the front of the TBM is rotated by 1◦ compared to the back side, with 50

Figure 6.

Plastic shear strains at the end of the Eulerian inflow phase.

Figure 7.

Normal stress on the tunnel at the end of the steering phase (for the calculation with a first Eulerian inflow phase).

51

Figure 8.

Normal stress on the tunnel at the end of the steering phase (for the reference calculation without inflow phase).

Table 2.

Normal stresses acting on the TBM at the end of the steering action and derived local subgrade reduction factors. (Coordinates refer to distance from steering jack or TBM articulation point.) Normal stress (kPa) Outer curve

Normal stress (kPa) Inner curve

Local subgrade reduction factor

Coordinate (m)

Reference

Full Analysis

Reference

Full Analysis

Reference

Full Analysis

+1.6 +1.1 +0.6 −0.6 −1.1 −1.6

−91 −189 −311 −289 −54 −287

−790 −657 −930 −1940 −2789 −2080

−23 −21 −32 −135 −64 −16.5

−583 −779 −1100 −1068 −1097 −1061

0.26 0.11 0.10 0.48 1.19 0.06

0.75 1.19 1.18 0.55 0.39 0.51

the jack extension (or outer curve) modelled at the lower side of the TBM in this figure and the resulting curvature upwards. Figure 8 shows results for the reference calculation where phase one was skipped and no soil inflow was modelled. In both cases the tunnel and TBM are not shown for improved clarity of the stress scale. When the reference case (without inflow of soil) is compared to the complete calculation some differences can be observed. In the reference case two distinct zones with a stress increase below the tunnel can be discerned, which act as support zones to carry the reaction of the rotating TBM. In the full model only a single support zone has developed. Clearly, the proper initial stress state calculated from the first inflow phase has a significant influence on the failure mode. 52

In order to derive the impact on the subgrade reaction modulus for the analytical model, the normal stresses acting on the TBM on the outer curve are compared to those on the inner curve. Some key numbers are given in Table 2, where the coordinate refers to the distance from the jack position or articulation point. The subgrade reduction factor is clearly affected by the stress history, given the differences between the reference analysis and the full analysis. Strong local fluctuations in the local subgrade reduction factor occur, but overall the results from the reference analysis are lower than the range derived for the analytical model from field observations, whereas the results for the full analysis are somewhat higher.

5 CONCLUSIONS During the boring of curves in very soft soils the steerability of the TBM may become problematic or even completely impossible. An analytical model that can determine the reaction of the TBM based on a given steering action has been developed. This model needs a subgrade reaction modulus on both the inner and outer curve side of the TBM to yield proper results. The subgrade reaction modulus on the inner side is reduced with respect to the value on the outer curve. The local subgrade reduction factors derived from the full numerical analysis indicate that a single value for the subgrade reduction factor may not be appropriate. It would be better to use separate factors for the front and back part of the TBM. The conclusion from the earlier paper by Broere et al. (2007), that the analytical model needs improvements, still stands, but the presented numerical approach alone is not sufficient to reach the required level of accuracy and thereby verify the analytical model in a wide range of soil conditions. In order to reach that goal the numerical model needs several improvements. These include a further refinement of the mesh, a more detailed modelling of the soil excavation at the tunnel face and a continued inflow of soil during the steering action of the TBM. Although these are all possible, they would increase the computational effort beyond that of a common desktop PC. An alternative approach would be to slightly adapt the analytical model to allow for a variable reduction factor along the TBM, with different values for the front and back part of the TBM on both the inner and outer curve side. REFERENCES Broere, W., Faassen, T.F., Arends, G. & van Tol, A.F. 2007. Modelling the boring of curves in (very) soft soils during microtunnelling. Tunnelling and Underground Space Technology, 22(5–6), pp. 600–609. Dijkstra J., Broere, W. & van Tol., A.F. 2007. Numerical simulation of the installation of a displacement pile in sand. In Numerical Models in Geomechanics, Taylor & Francis, pp. 461–466. Finite Element Application Technology. TOCHNOG Professional User’s Manual Version 4.2. (http://www.feat.nl/) Oreste, P.P., Peila, D., Marchionni, V. & Sterling, R. 2002. Analysis of the Problems Connected to the Sinking of Micro-TBMs in Difficult Ground. Tunnelling and Underground Space Technology, 16 Suppl. 1, pp. 33–45.

53

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Trenchless replacement of gas and potable water pipes with new PA 12 pipes applying the pipe bursting method Reinhard Buessing SB Projektentwicklung, Herdecke, Germany

Andreas Dowe & Christian Baron Evonik Degussa GmbH, Marl, Germany

Meinolf Rameil Tracto-Technik GmbH & CoKG, Lennestadt, Germany

ABSTRACT: Pipe Bursting offers a real technical and cost-efficient alternative to the open trench method; it is also very effective when compared to other alternative pipe rehabilitation methods. Pipes made from Polyamide 12 (PA 12) extend the area of application for trenchless technologies, pipe bursting in particular. Polyamide 12 pipes showed that problems like point loads and crack growth resistance can be solved. A field test in Dortmund showed that the handling of PA 12 pipes is as easy as the pipe cracking with PE.

1 INTRODUCTION Pipe replacement by open trenching can involve traffic impairment, noise- and emission pollution when breaking open the surface. Also there is a risk of damaging existing underground pipe, soil and groundwater intervention and higher storage, transport and soil removal costs. These drawbacks can be almost completely avoided by replacing pipes using trenchless technologies. According to estimates of Advantica, the pipe bursting method has become the most widely applied trenchless pipe replacement method with more than 50,000 km worldwide. With the pipe bursting method, the defective or under dimensioned pipe is cracked with a burst head and then displaced into the surrounding soil. This creates space for the new pipe directly proceeding, of same or greater diameter. The pipe bursting technique allows the replacement of defective pipelines in the same path without any substantial influences on soil and groundwater. Open trenching which requires the breaking up and repairing of valuable surfaces is thereby almost entirely eliminated. In addition, this eco-friendly technique helps to cut down costs considerably. The pipe bursting distances usually have a length of approx. 100–200 m. In general, there are two methods to choose from: 1. hydraulically operated static pipe bursting with ladder shaped Quicklock rods 2. pneumatically operated dynamic pipe cracking with piercing tools or rammers The proportion of pipes that need to be rehabilitated or renewed will increase. The current situation in Germany: The length of the transportation network of water pipes is approximately 460.000 km (Wagner, V.: Sanierung von Wasserdruckleitungen – Eignungsprüfung für Druckliner, WWT Heft 9/2003) and 270.000 km for the gas supply lines (Bundesverband der Deutschen Gas- und Wasserwirtschaft e.V. (BGW): Jahrbuch Gas- und Wasser 2004, Oldenburg Industrieverlag GmbH, München, 2004). There are as well 160.000 km of house-connections for water and 80.000 km for gas. 55

In order to continuously ensure a fully functional mains and distribution network, an annual rehabilitation rate of at least 2% should be achieved. This rate of renewal would correspond to a service pipe life of 50 years. That is why the network-operators are looking for innovative methods which offer increased productivity combined with increased economical savings. The Pipe Bursting technology for trenchless pipe renewals is just such a method. 1.1 Pipe materials for transportation network of water and natural gas in Germany Grey cast iron was used with lead joint sockets until the end of the 1940’s and with screwed and bolted flange joints up until the mid-60’s. There are still many of these fragile grey cast iron pipes which are likely to leak and break and these now have to be renewed as a matter of priority. Ductile iron without sufficient coated protection was installed from the mid-60’s to the early-70’s. Steel pipes were installed with jute fibre/tar coating and joints until the mid 40’s, then with a more simple bitumen coating until the end of the 50’s, and after that with a double protective layer of bitumen but often the pipes were not laid on a granular bedding until the end of the 70’s and since then with PE-coatings and with/or without protective surface coatings. Asbestos cement- and lead water pipes need to be replaced just as soon as possible because they are considered harmful to peoples’ health. Plastic pipes (PVC) with bonded sockets are leaking and PE pipes are often not of sufficient carrying capacity and therefore need to be upsized and renewed simultaneously. 1.2 Choosing the proper method The most important question for the network operators is: do pipes have to be repaired, rehabilitated or renewed/replaced? To repair the pipes is feasible if the repair measure ensures a reliable and lasting result. Another option is pipe rehabilitation, but rehabilitation methods sometimes do not meet the required standards in terms of pipe carrying capacity, flow volumes and pipe life durability. Problems also occur if an annular gap is left between the new replacement and the host pipe. Also because of additional pipe cleaning and the need for encrustation removal, the overall costs may be comparatively high. The third opportunity is trenchless renewal. A pipe replacement is necessary: – when repair or renovation is technically or economically inappropriate – when the hydraulic capacity needs to be improved by a greater pipe diameter – when repair or renovation offers only a short term solution with a pipe replacement being inevitable – when there is request for a long lasting pipe durability and or a higher product life span – when the static loading capacity of the defective pipe would be otherwise negatively affected The renewal of pipes with traditional open trenches is consistent with traffic jams, noise, dust and not liked ecological effects as well as risks of damage to existing buried pipes cables and telecommunication lines, disruption of the groundwater flow and the need to remove previously undisturbed soil, storage and transport of the excavated soil and its associated high disposal and tipping costs. All this can be avoided to a large extent by using a trenchless method. With the Pipe Cracking method the effected or under capacity pipe is fragmented by the profile bursting blades and the broken pieces are displaced by an expander into the surrounding soil. Thus a sufficient channel space is preformed for the attached new pipe with the same or even bigger diameter that follows the blade/expander. The Pipe Cracking method permits renewals of damaged pipes by using the existing pipeline course without intrusion into the soil or disturbing the groundwater. The work method does not cause surface subsidence damages and by using a filler for the annular gap between the new and the burst pipe the bedding can be improved. As opposed to the open trench method where the open trenching and reinstatement of the road and pavement surfaces are largely unnecessary and thus pipe cracking is environmentally friendly and in most cases more economical. 56

1.3 The development of the Pipe Bursting technology to date Pipe Bursting has a long history and could not have developed as successfully as it did without soil displacement hammers and ramming technology. The idea of the Pipe Bursting technology, using the existing, damaged pipe in order to accommodate a new pipe has its origins at British Gas. As early as the 1980’s they commonly used adapted soil displacement hammers and pneumatic pipe ramming machines to trenchlessly lay pipes and cables. In Great Britain a contractor named D J Ryan & Sons had already taken out the first patents (Fraser: The Pipe Bursting Options, Conference documents “Asian Trenchless Tech 1994”) Even in the first years of the Pipe Bursting technology pneumatically and hydraulically propelled systems existed. The pneumatic machines were known as PIM-machines (Pipe Insertion Method). The first hydraulic machines were systems where the expander radially extended using a hydraulic force and thus shearing apart the pipe and the fragmented parts displaced into the soil around it. In Germany the machines were known, amongst others, by the name of “KM-Berstlining”. This expanding technique though did not become generally accepted in Germany. First users in Germany were companies like “Kanal-Müller” (KM-Berstlining), DIGA (PIM, Grundocrack) and Brochier (Grundocrack). In the USA the method known as “cracking” was introduced successfully more than 20 years ago. Ever since then it has been used very successfully on a daily basis. In the UK the first applications of the Pipe Bursting method were limited to grey cast iron gas pipes. The method rapidly spread all over Europe already including operations in the water and sewage sectors. In those times Polyethylene pipes were already used as new product carrying pipes. Mostly in the early years plastic protection tubes were installed first because PE pipes with protective coating have been not available. Nowadays the Pipe Bursting method is mainly used to replace pressure pipes (gas and water networks ND 80 to 1000) and for sewage pipes (ND 150 to 1000). In the USA even sewers up to 1200 mm in diameter have been replaced. In the mid-90s the Dynamic (Pneumatic) Pipe Cracking was complemented by a new method – the so-called Static (Hydraulic) Pipe Bursting. With static bursting ladder shaped rods without screwed threads were introduced. These new rods could easily be connected and disconnected and have had many practical advantages. The functionality of screw threads can easily be impaired by dirt and therefore have to be kept clean at all times. Often the screw threads are damaged on the jobsite and thus slow down the connection and thus the pushing and pulling process. In addition to that the tensile strength and pulling forces may be impaired. In contrast the QuickLock system provides an absolutely safe and reliable connection – for the tractive pushing and pulling forces. The new QuickLock rod joints are also flexible and can follow the radius of the existing curved offset pipe bends. Time efficiency is another key advantage – the QuickLock rods can be driven into the host pipe about 40% faster than conventional screwed rods because the ladder shaped design allows a fast and easy rod string assembly. In contrast to the screwed rods where friction drive clamp jaws transmit their force on the rods from outside, the QuickLock rods have positive locking brake and drive fingers that prevents any slippage of the rods when under traction load. Also the tiresome labour intensive screwing and unscrewing process is no longer necessary. A hydraulically operated rig pushes or pulls the rods into the host pipe. Once they appear in the arrival pit, an upsizing cone is connected and the new pipe then attached to it. By pulling back the rods the bursting and installation process begins. A traction bursting force of up to 2.500 kN (250 metric tons) is available. 1.4 Main options of the Pipe Bursting Pipe Bursting is the trenchless replacement of pipelines by using the host pipe’s line and level. With a dynamic or hydraulic bursting device the host pipe will be broken, fragmented and then displaced into the surrounding soil. Protection tubes can also be installed with this method. During this work process the new pipe with the same or even larger diameter is installed. For gas- and sewer pipes most of the time PE pipes are used. They are flexible and able to adjust well to the host pipeline course. Smooth outer but welded pipe joints guarantee an easy entry into the new burst line. Prior to the beginning of the burst process the host pipes have to be disconnected and the house services 57

Figure 1.

Dynamic pipe cracking method.

Figure 2.

Dynamic cracking hammer GRUNDOCRACK with bladed pipe cracking head and attachment for winch rope and expander at the rear end for pressure pipes.

have to be exposed at their connection points on the main. As Pipe Bursting is a really fast pipe renewal method some network operators do not install a temporary bypass fluid or gas supply. 1.5 The dynamic pipe cracking method After the constant tension winch is set up at the pit and the winch boom has been anchored, the cable is pulled through the host pipe until it arrives at the launch pit. At the same time, the cracking hammer is prepared and the new pipe, usually a PE or PA 12 pipe, is attached to it. This means, firstly single pipe sections are butt-welded together and attached to the rear end of the cracking hammer by welding to an adapter. The bladed pipe cracking head is attached to the front nose of the cracking hammer which itself is connected to the winch cable. The bursting machine together with the attached PA 12 or PE pipe is pulled into the pit by means of the cable winch until it comes up against the host pipe to be renewed. Then, by starting the compressor and the impact cracking machine via its control unit the pipe cracking process starts. The forward motion and the directional stability are maintained by the traction force of the cable winch. The pipe cracking hammer breaks the host pipe by means of its bladed head and radially displaces the fragments of the pipe into the surrounding soil. At the same time it expands the soil to accept the new pipe with an equal or bigger diameter. Once the bursting machine and the expander cone reach the arrival pit they can be removed. 1.6 The Static Pipe Bursting method The Static Pipe Bursting obtains its required force for the bursting, displacing and installation process hydraulically and transmits it via the ladder shaped QuickLock rods, which are absolutely safe and reliable for the thrust and traction forces. First the hydraulically powered rig has to be braced inside the arrival pit. After that the burst rods with their precursory guide-rod are pushed into the host pipe. In the starting pit the guide-rod has to be exchanged by the burst tools (burst head and expander with swivel, roller-blade, etc). The new PA 12 or PE pipe is affixed to a pipe pulling adapter and attached to the burst tools. By pulling back the burst rods towards the arriving pit the 58

Figure 3.

Static pipe bursting method.

host pipe will be broken and its fragments displaced by the expander into the soil right behind the burst head/expander. The bore for the new pipe can be the same or a larger diameter.

2 LIMITATIONS There are only few limitations (Rameil, M.: Handbook of Pipe-Bursting Practise, Vulkan Verlag Essen 2007) for the use of the Pipe Bursting method: – at this time, the application range is limited to circular existing pipes, – if necessary, the host pipe has to be taken out of service during the bursting process (not necessary e.g. for mains, drainage pipe replacement, etc.), – the course of the host pipe must be usable for the new pipeline (e.g. inclination). Heavily encrusted pipes must be cleaned so that bursting rods can be pushed through (static bursting) or respectively the winch rope can be pulled in when (dynamic/pneumatic pipe cracking), – the soil surrounding the host pipe must be displaceable, – house connections have to be installed using pits. This, however, guarantees a professional and safe integration, – sharp bends, flanged joints of steel and ductile iron pipes, etc., depending on selection of bursting tools require intermediate pits, – pipe slumps (sags) cannot be removed however may be reduced, – a minimum distance has to be kept away from existing parallel or crossing pipes as well as an adequate cover depth. However, the application of the pipe bursting method also offers all advantages of a modern, trenchless installation method: Pipe bursting is the installation of new, industrially produced pipes, which may be compromised when installed by open cut methods, – pipe bursting gives a considerable reduction in excavation and road works (almost no traffic disturbance, no annoyance of the public or noise and dust pollution, reduced construction time and the reduction of indirect costs), – high daily output up to 150 m gives a cost-effective replacement and considerable cost-savings compared to open cut, – almost any pipe material available for trenchless installation methods can be installed by pipe bursting, e.g. plastic (PE, PA 12), ductile iron (DIP), steel, glass reinforced plastic (GRP) and even vitrified clay (VCP) and polymer-concrete pipes (PCP), – pipe bursting allows the replacement of almost any host pipe material (some with limitations), – pipe bursting can be applied for any kind of pipe damage as long as the bursting rods can be pushed in (static bursting) or the winch rope can be pulled in (dynamic/pneumatic pipe cracking), – no reduction of pipe diameter, up-size of pipe diameter is also possible, 59

– preparation of the host pipe, like high-pressure cleaning, removal of debris and blockages are not necessary (but may possibly be necessary for specific reasons), – considerably less danger of unintended ground settlement compared to open cut, – applicable for pipes in sloped areas and areas with trees, shrubbery and the like, – pipe bursting can also be used for the replacement of laterals, – pipe bursting is controlled and described by worldwide standards, norms and regulations.

3 MATERIALS FOR THE NEW PIPES New pipes made of HD PE with a coating (e.g. SLM 2.0) are particularly suitable. They are more impact-resistant as the PE without coating, sufficiently flexible and align well to the course of the host pipeline. This material is mainly used for gas networks from 0.5 to 4 bars. Evonik Degussa GmbH knew from experiences in the gas pipe project about the excellent characteristics of Polyamide 12 (PA 12). After a lot of testing they decided to make a pipe cracking. After a lot of testing they decided to make a pipe bursting test with this material. All results turned out satisfactory. Now the first PA 12 pipes especially for the use in water networks are produced and will be installed soon. During this conference it may be possible to present the first results. Some details about the outstanding results of PA 12 material tests.

3.1 Crack growth resistance of PA 12 During the feasibility study of PA 12 for gas applications several tests were observed in acc with international standards. Considering the sensitivity of former PE materials to slow crack growth (SCG) and its influence on the installation technologies Evonik paid much attention on this issue. Because of the high resistance to stress cracking PA 12 is used for air brake tubing systems since decades. Therefore it could be assumed that SCG would not be an issue for PA 12. This is also indicated by the long-term hydrostatic pressure test in acc with ISO9080. Whereas PE is known to have a second branch at the time to failure curves this is not investigated by tests on PA 12 pipes. Even after 16,000 test hours at 80◦ C no stress cracking initiated failures are detected. Additionally the SCG behaviour of PA 12 was investigated with the Notch Pipe Test in acc with ISO 13479 and ISO 22621-1. Also the PENT test in acc with ASTM F1743 was investigated. The results are shown in Table 1. At 80◦ C and 20 bar no failures occur at the Pipe Notch Test after 2000 h. Even after an increase of the notch depth from 20 to 30% of the wall thickness no failures occur. The minimum requirement in acc. with ISO 22621-1 is 500 h without failure with a notch depth of 20%. The PENT test is very well established in the US gas society and PE materials are tested at 80◦ C and 2,4 MPa stress. Presently, the requirements within ASTM D2513 for PE materials require PENT time to failure of 100 hours. However, no such requirements are in place for Polyamide materials. Two replicates of the PA 12 material were tested in accordance to ASTM F1743 requirements at an increased stress of 4.8 MPa. The results of the testing indicated that there were no failures with any of the specimens after 1000 hours. The testing was discontinued after 1000 hours.

Table 1.

Results on slow crack growth resistance of VESTAMID LX9030.

Test

Standard

Pipe Notch Test 20% notch depth 30% notch depth PENT

ISO 13479 & 22621-2

ASTM F1743

Requirement

LX9030 pipe

>500 h

>2000 h >2000 h >1000 h

>100 h

60

3.2 Effects of secondary stresses on PA 12 pipes In addition to characterizing the SCG performance characteristics and influence of surface scratches, additional tests at GTI were performed to characterize the influence of secondary stresses. The main motivating factor for performing these tests was the fact that under actual field conditions, the piping systems are subjected to the combined effects of both internal pressure and other secondary stresses including rock impingement, earth loading and bending. Often, these secondary stresses, not internal pressure, are the root cause of many in-service field failures. Additionally the point load resistance is highly discussed relating to the various trenchless technologies and sand-bed free installation of thermoplastic pipes. Therefore, comprehensive long term sustained pressure tests were performed at elevated temperatures to characterize the effects of various types of secondary stresses. It is important to note that these tests are not a part of either the ASTM or ISO standard. The test methodology is an extension of previous research performed by Dr. Charles Bargraw– DuPont and further refined by Dr. Michael Mamoun – Gas Technology Institute to study the performance characteristics of older generation PE materials. For the case of the rock impingement, the intent is to evaluate the performance of pipe materials subjected to indentations by a 13 mm rock. For the case of the earth loading, the typical safe deflection limit that is specified is 5%. For the case of the bending strain, the typical bend radius limits for a pipe specimen without any joints or appurtenances is 20 times the outside diameter. Six (6) 2-inch SDR-11 LX9030 pipe specimens were placed in appropriate test rigs to simulate the effects of rock impingement, earth loading, and bending strain. The entire test assembly was placed under long term sustained pressure testing at 20 bar at 80◦ C. For all of the various types of secondary stresses which were evaluated, the results of the testing demonstrated that there were no failures after 1000 hours of testing, as presented in Table 2 below. In summary the test results show that PA 12 has a unique resistance to SCG and secondary stresses including point loading. Therefore it could be assumed that PA 12 pipes can be installed without sand-bedding and with trenchless technologies even without additional protection layers. At a usual gas pipe bursting site in Dortmund we made a test how the PA 12 material will resist compared to PE. This test showed that for welding the same equipment, with different welding parameters can be used. The employees have been able to weld PA 12 without any problems. The PA 12 pipe has been connected with the PE pipe through an inside connection because it is not possible to weld them together. The bursting process has been done like usual. Not a single problem has been recorded. PA 12 is as easy to use as PE. At the last pit the PA 12 pipe and a sample of PE has been taken out for the microscopic analysis in the laboratory. 3.3 Results of the microscopic analysis The pictures show very well the difference in the damage of the surface of both materials. Table 2. Test results of VESTAMID LX9030 pipes investigating secondary stresses Field test in Dortmund. Secondary Stress

Test Criterion

Results

Rock Impingement

13 mm Indentation Test Pressure = 20 bar Test Temperature = 80◦ C

Test Time >1000 hour with no failures

Earth Loading

5% Deflection of Outside Diameter Test Pressure = 20 bar Test Temperature = 80◦ C

Test Time >1000 hour with no failures

Bending Strain

20 times OD Test Pressure = 20 bar Test Temperature = 80◦ C

Test Time >1000 hour with no failures

61

Beside the above reported excellent resistance to SCG and point loads the PA 12 shows also a high scratch resistance. Summarizing these performances it can be assumed that Evonik Degussa found a promising solution for a pipe bursting material which can be used in pipes with high pressure. 3.4 Bursting tools for pressure pipes When static pipe bursting was introduced in the middle of the 1990s, it became possible to replace ductile host pipe materials like steel and ductile iron by splitting. Up until then, it had only been possible to burst brittle host pipe materials applying dynamic/pneumatic pipe bursting. Steel pipe bursting was limited and ductile iron pipes could not be burst at all. Meanwhile, different splitting and bursting tools were made available to select from, depending on the material of the host pipe to be replaced. Cast iron pipes, but also other brittle pipe materials (fibre or asbestos cement) can easily be burst into fragments by unbladed bursting heads and then displaced. Here, the tapered angle of the bursting head which induces powerful radial loads into the host pipe has a great influence on the destructibility of the host pipe. The fragments generated should be as small as possible to achieve an even load distribution covering the complete circumference of the new pipe, thus preventing twisting and misalignment of the old pipe. Differing from brittle host pipe materials are steel or plastic as well as ductile iron pipes which require splitting techniques that cut the host pipes and expand them open while also requiring

Figure 4.

PA 12 and PE pipes connected by an mechanical joint.

Figure 5. Typical pit on the pipe bursting site in Dortmund.

Figure 6. The PA 12 and PE pipes connected with the pipe bursting head.

Figure 7. The site is ready to start the pipe bursting process.

62

specialist tooling to break joints and pipe repair clamps. Particularly, within the area of joints and repair clamps (e.g. steel pipes with bell-and-spigot joints or not welded steel pipes or ductile iron flanges and fittings), the capacity limits of the pulling rig are often reached before they can be cut when using conventional splitting techniques. To split steel pipes, roller blade cutters (fig. 12) are used, with a splitting line of roller wheels that attack the host pipe and joints mainly in the lower third of their circumference. The roller blades are arranged to achieve optimal splitting geometry. This is the only way to minimise the required splitting and pulling forces. It is of the greatest importance that the split edges of the host pipe are bent outwards to keep them from damaging the new pipe. Meanwhile, in order to split ductile iron pipes, so-called “roller blade trailers” are used. The roller blades are not arranged in a single holder with minimal space between them, but instead every single roller blade is arranged in its separate roller blade holder. Usually, the space between the single roller blade holder has the length of a rod between. The diameter of the roller blades fitted is steadily increased. This arrangement leads to an even distribution of the splitting forces. When the blade holders run through the host pipe, the first profiled roller blade perforates the pipe so the following smooth roller blades can evenly split it. In order to cut PE pipes or defective inliners, special hook shape splitters with 2 or 4 blades are used. These special splitters prevent the host pipe from buckling and slide easily through the host pipe when splitting.

Figure 8. The pipe bursting head is cutting the existing cast iron pipe.

Figure 10.

Figure 9. The burst host pipe has very sharp edges.

Intermediate pit for the house connection. Also here the sharp edges of the host pipe can be seen.

Figure 11. The PA 12 and PE pipe pieces for the microscopic analysis.

63

Figure 12.

New Roller Blade to cut 1,000 mm steel pipes.

4 CONCLUSION Pipe bursting with its different variations and decades of applied experience is a comprehensive trenchless renewal option for a large proportion of the worlds damaged fluid, gas supply and sewer pipelines. This amounts to enormous potential for economical savings when carrying out the inevitable pipe rehabilitation programs. Especially, in times of critical financial situations, the cities and municipalities can hardly afford to simply give away this potential as a result of technological 64

conservatism. All in all, we can surely look to pipe bursting to be the preferred technique as more than 50,000 kilometres of supply and sewer pipelines worldwide have already been “cracked”, “burst”, “split” and replaced with new pipes, particularly, since with the new PA 12 material also gas and water pipes operated under higher pressure can now be installed. As a result in many countries a steadily increasing number of contractors are now successfully using this technology. In combination with the PA 12 pipes, which are resistant to issues like point loads and crack growth resistance gives pipe bursting the chance to enter the market in higher pressure pipes f.e. in water distribution networks.

65

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Experiences with Polyamide 12 gas pipes after 2 years in operation at 24 bar and new possibilities for HDD A. Dowe & C. Baron Evonik Degussa GmbH, Marl, Germany

W. Wessing E.ON Ruhrgas AG, Essen, Germany

R. Buessing SB Projektentwicklung, Herdecke, Germany

M. Rameil Tracto-Technik GmbH & CoKG, Lennestadt, Germany

ABSTRACT: For more than 20 years PA 12 is evaluated as a pipe material for gas distribution and for more than 10 years PA 12 pipes have been used for low pressure installations. In recent years PA 12 is being investigated for operation pressures above 10 bar. Since 2002 ISO/TC 138 is elaborating standards for PA 12 gas installation systems with operation pressures up to 20 bar. In 2007 the first three parts of the ISO 22621 were published. Evonik Degussa GmbH, Marl, Germany, as one of the four manufacturers of PA 12 and E.ON Ruhrgas are cooperating on a test installation of a 60 m pipe system (110 mm SDR-11) in the technical center of E.ON Ruhrgas in Dorsten, Germany. The system with various connections was installed in October 2005. After 2 years operating at a pressure of 24 bar first samples were taken and investigated. The authors will give a report of the test installation and the experiences after 2 years. The paper will include also an introduction of the installation technology “Horizontal Directional Drilling” (HDD). In combination with the exceptional resistance of Polyamide 12 pipes against point loading and slow crack growth HDD offers new possibilities.

1 PA 12 – A HIGH PERFORMANCE THERMOPLASTIC MATERIAL Due to the specific carbon-amide group in the polymer chain polyamides (PA) have strong intermolecular actions which induce high mechanical strength, high melting temperatures and chemical stability. The “long chain” PA 12 with 11 carbon atoms between carbon-amide groups has the lowest water absorption of all commercial available PA and represents the best compromise in thermal and mechanical properties. That is why PA 12 is the material of choice for some challenging applications relied on already for decades, e. g. in the automotive industry for fuel lines of passenger cars or for airbrake tubings of trucks. Compared with medium and high density polyethylene in use for low pressure gas supply, “long chain” polyamides like PA 12 provide “naturally” superior performance due to their described chemical structure (table 1). Besides PA 12 only PA 11 is commercially available with almost identical properties. PA 11 supplied by one manufacturer is based on a planted feed stock, castor bean, while PA 12 is synthesized from butadiene, a crude oil by-product, in a complicated multi-step process. Evonik Degussa is the only company of the four PA 12 suppliers who is fully backintegrated to butadiene. 67

Table 1.

Basic properties of PA 12 vs. PE.

Property

Unit

PA 12

PE 100

PE 80

Melting temperature

◦C

178

130

126

Tensile strength at yield Tensile elongation at break Flexural Modulus

MPa % MPa

45 200 >1200

20–23 >800 950

17–19 >800 700

Charpy impact strength

kJ/m2

No break

30

20

Hardness, Shore D

74

63

58

Permeability (23◦ C, mm3 /bar/day) Methane Hydrogen (data for PA 12 pending)

< 0.005 < 0.01

0.24 0.7

Table 2.

Maximal allowable operation pressures for PA 12 and PE based gas pipes with considering a safety factor of 2 and a SDR-11.

MOP in bar

20◦ C

60◦ C

PE 80 PE 100 PA 12

8 10 18

4 7 12

2 NEW VESTAMID GRADE FROM EVONIK DEGUSSA, PA 12 DESIGNED FOR HIGH PRESSURE GAS DISTRIBUTION Controlling molecular weight and intermolecular interacting forces is an Evonik Degussa core competence to design PA 12 extrusion grades with optimized processing and performance properties. Evonik PA 12 is sold under the registered trade mark VESTAMID™ L. When Evonik started the activities for large pipe applications new high molecular PA 12 grades for extruding pipes with bigger sizes was designed. For various tests and for test installations pipes with dimensions up to 300 mm have been manufactured so far using these new grades at several different pipe extrusion companies without any problems. 2.1 VESTAMID LX9030 – a material of choice for high pressure gas supply There are two major design criteria for thermoplastic materials used in gas supply: the Maximum Required Strength (MRS) and Rapid Crack Propagation (RCP). Due to the intermolecular forces in PA 12 the hydrostatic strength of PA 12 pipes is much higher than for PE-HD, even for the optimized grade PE 100. Long term hydrostatic strength investigations according to ISO or ASTM standards have proven that in a 50 years extrapolation PA 12 has got a MRS of 18 MPa. Therefore PA12 SDR-11 systems are able to operate at pressures up to 18 bar using a safety factor of 2 for gas (table 2). In contrast to PE PA 12 is naturally resistant against stress cracking in general and slow crack growth and passes easily all the relevant tests originally created for this weakness of PE. For the time being resistance to RCP at low temperatures is a matter of concern and discussion in the gas utilities. In the ASTM territory RCP is under consideration. RCP data have to be provided 68

but they are not limiting the listings for pipe materials from the Plastic Pipe Institute for use in the USA. In the European standards and also adopted in the ISO standard the maximum operation pressure (MOP) for a pipe material is limited from two sides, the hydrostatic strength of the material and the MOP derived from a RCP test at 0◦ C. Due to that a test institute performed a RCP full scale test on 110 mm and 6-inch SDR11 pipes of Vestamid LX9030 according to ISO 13478. The initiated crack arrested up to 30 bar respectively 25 bar internal pressure. 2.2 Technical positioning of VESTAMID LX9030 between PE and steel Steel has a long history in gas installation and it is still the only option for real high pressure gas transportation at 50 bar or higher. Reinforced plastic pipe constructions might be feasible to serve systems with pressures up to 45 bar (Wessing, Grass, 2006). Welding of steel is a well established technology and weld quality control procedures are in place. Also with respect to third party damages steel has naturally a high stability. On the other side steel pipes are rigid, heavy and corrosive, which makes them principally not attractive for installations. In Europe and USA the gas distribution networks are operating at pressures of up to 25 bar, with a big share around 16 bar. PE100 as the latest development of High Density Polyethylene received the approval for operating pressures up to 10 bar however this comprises only a small share of distribution networks. PA 12 pipe systems are technically able to carry gas at pressures up to 18 bar and withstand elevated temperatures up to 80◦ C, even in a longer term. This would fit quite well to serve distribution networks running at a pressure of 16 bar (Dowe, Baron, Buessing, 2007). 3 CO-OPERATION OF EVONIK AND E.ON-RUHRGAS ON EVALUATING PA 12 FOR HIGH PRESSURE GAS SUPPLY 3.1 First high pressure test installation in 2005 Evonik Degussa and E.ON Ruhrgas decided in 2005 to set up a high pressure test installation on the E.ON Ruhrgas Technical Center site in Dorsten. Evonik had to provide 60 m of 110 mm SDR-11 VESTAMID pipe on a coil and some straight pipes for assembling a system including butt fusion and electro fusion joints. Evonik also had to provide electro-fusion end caps. For the extrusion and coiling of the requested pipe Evonik choose the company Egeplast in Greven. Although having PA 12 the first time on their production line Egeplast was able to extrude the pipe within the tolerances and with excellent appearance without any problem. The online coiling on a 2.5 m diameter drum was running smooth without cranking the pipe on the coil at all. For the development of electro-fusion fittings and end-caps Evonik co-operated with company Friatec in Mannheim, one of world leading companies in electro-fusion fittings. Using the VESTAMID gas pipe material they manufactured the required components for the test installation in a perfect manner. Before the test installation on the E.ON Ruhrgas site a burst pressure test was carried out with a 3 m test pipe including a butt fusion and an electro-fusion joint and electro-fusion end-caps. The pipe sample burst at a pressure of 94 bar in a tough crack at the main tube, not at a fitting or joint. For the test installation the coiled tube was unrolled in the field without any mechanical stretching tool. Only at few points needed some heating assistance to be stretched for applying the connections. Butt fusion was carried out with standard equipment and slightly adjusted temperature profiles. For the electro-fusion fittings standard power generators from Friatec were used. The adjusted fusion conditions were read from a bar code adhered to the fittings. The system with butt fusion and electro fusion joints and two electro-fused end caps was installed and sealed in a 500 mm steel pipe. For 72 hours 36 bar natural gas was applied for checking the tightness of the system. Then the pressure was lowered to 24 bar. Additionally a similar pipe to the burst pressure sample was installed in parallel. This 3 m long pipe section was pressurized to 36 bar. 69

Table 3.

Properties after 2 years of test installation.

Properties

Standard

Tensile test at 23◦ C Strain at yield [MPa] Elongation at yield [%] Strain at break [MPa] Elongation at break [%]

ISO 527

Water content [%], across wall Outer Centre Inner

Virgin pipe from stock

After 2 years at 24 bar

After 2 years at 36 bar

38 13 49 315

39 12 46 287

40 14 49 293

0.69 0.46 0.56

0.63 0.27 0.28

0.52 0.34 0.31

3.2 Observations after 2 years of operation at 24 and 36 bar Both installations run very successful. Except an external leakage at a manometer after 8000 h caused by storm Kyrill no further distinctive features were observed. After 2 years (18000 h) of operation, the 36 bar pipe sample was depressurized and taken. The following burst pressure test did show a burst pressure of 86 bar. This is a reduction of 8 bar compared to the virgin pipe sample tested prior to the installation. The investigation of the water content in the pipe material did show a increase from 0,1% to about 0,4% as expected. PA 12 absorbs a maximum of 1,4% of water if fully immersed and 0,8% at 50% humidity at 23◦ C. This absorption causes a softening of the material. It must be emphasized that the investigation of the MRS value in acc. with ISO 9080 was done at saturated pipe samples. Therefore and because the test is done fully immersed in water this softening effect is already fully considered at the investigation of the MRS of the material. Also a pipe sample including a butt fusion joint and a electro fusion joint was cut after depressurization the 24 bar installation. Beside tensile tests on dog bones prepared from the pipe samples the water content was tested. At table 3 the results of the tests in comparison to virgin pipe samples are shown. A hydrostatic pressure test in acc. with ISO 22621-2 at 80◦ C and 20 bar was also set up. This test is still pending and results will be shown at the conference. 3.3 Expert report by TUVNord and SKZ Wuerzburg With the very good experiences of the test installations and the good test results in acc. with international standards Evonik assigned a review of all available data to the German TUVNord and SKZ Wuerzburg. As a result an expert report was set up in co-operation of TUVNord and SKZ which stated the general suitability of VESTAMID LX9030 for high pressure gas pipe applications up to 18 bar. It is also mentioned at the report that pipes made of VESTAMID LX9030 are as safe as steel pipes if installed following the recommendations of the report.

4 TRENCHLESS TECHNOLOGIES CONSIDERING PA12 PIPES 4.1 Crack growth resistance of PA 12 During the feasibility study of PA12 for gas applications several tests were observed in acc with international standards. Considering the sensitivity of former PE materials to slow crack growth (SCG) and its influence on the installation technologies Evonik paid much attention on this issue. Because of the high resistance to stress cracking PA 12 is used for air brake tubing systems since decades. Therefore it could be assumed that SCG would not be an issue for PA 12. This is also indicated by the long-term hydrostatic pressure test in acc with ISO9080. Whereas PE is known to 70

Table 4.

Results on slow crack growth resistance of VESTAMID LX9030.

Test

Standard

Pipe Notch Test 20% notch depth 30% notch depth PENT

ISO 13479 & 22621-2

ASTM F1743

Requirement

LX9030 pipe

>500 h

>2000 h >2000 h >1000 h

>100 h

have a second branch at the time to failure curves this is not investigated by tests on PA12 pipes. Even after 12,000 test hours at 80◦ C no stress cracking initiated failures are detected. Additionally the SCG behavior of PA12 was investigated with the Notch Pipe Test in acc with ISO 13479 and ISO 22621-1. Also the PENT test in acc with ASTM F1743 was investigated. The results are shown in table 4. At 80◦ C and 20 bar no failures occur at the Pipe Notch Test after 2000h. Even after an increase of the notch depth from 20 to 30% of the wall thickness no failures occur. The minimum requirement in acc. with ISO 22621-1 is 500 h without failure with a notch depth of 20%. The PENT test is very well established in the US gas society and PE materials are tested at 80◦ C and 2,4 MPa stress. Presently, the requirements within ASTM D2513 for PE materials require PENT time to failure of 100 hours. However, no such requirements are in place for Polyamide materials. Two replicates of the PA12 material were tested in accordance to ASTM F1743 requirements at an increased stress of 4.8 MPa. The results of the testing indicated that there were no failures with any of the specimens after 1000 hours. The testing was discontinued after 1000 hours. 4.2 Effects of secondary stresses on PA 12 pipes In addition to characterizing the SCG performance characteristics and influence of surface scratches, additional tests at GTI were performed to characterize the influence of secondary stresses. The main motivating factor for performing these tests was the fact that under actual field conditions, the piping systems are subjected to the combined effects of both internal pressure and other secondary stresses including rock impingement, earth loading and bending. Often, these secondary stresses, not internal pressure, are the root cause of many in-service field failures. Additionally the point load resistance is highly discussed relating to trenchless technologies and sand-bed free installation of thermoplastic pipes. Therefore, comprehensive long term sustained pressure tests were performed at elevated temperatures to characterize the effects of various types of secondary stresses. It is important to note that these tests are not a part of either the ASTM or ISO standard. The test methodology is an extension of previous research performed by Dr. Charles Bargraw – DuPont and further refined by Dr. Michael Mamoun – Gas Technology Institute to study the performance characteristics of older generation PE materials. For the case of the rock impingement, the intent is to evaluate the performance of pipe materials subjected to indentations by a 13 mm rock. For the case of the earth loading, the typical safe deflection limit that is specified is 5%. For the case of the bending strain, the typical bend radius limits for a pipe specimen without any joints or appurtenances is 20 times the outside diameter. Six (6) 2-inch SDR-11 LX9030 pipe specimens were placed in appropriate test rigs to simulate the effects of rock impingement, earth loading, and bending strain. The entire test assembly was placed under long term sustained pressure testing at 20 bar at 80◦ C. For all of the various types of secondary stresses which were evaluated, the results of the testing demonstrated that there were no failures after 1000 hours of testing, as presented in Table 5 below. In summary the test results show that PA12 has a unique resistance to SCG and secondary stresses including point loading. Therefore it could be assumed that PA 12 pipes might be suitable 71

Table 5. Test results of VESTAMID LX9030 pipes investigating secondary stresses. Secondary stress

Test criterion

Results

Rock Impingment

13 mm Indentation Test Pressure = 20 bar Test Temperature = 80◦ C 5% Deflection of Outside Diameter Test Pressure = 20 bar Test Temperature = 80◦ C 20 times OD Test Pressure = 20 bar Test Temperature = 80◦ C

Test Time > 1000 hour with no failures

Earth Loading

Bending Strain

Figure 1.

Test Time > 1000 hour with no failures Test Time > 1000 hour with no failures

Surface of the PA 12 pipe (left) vs. PE 100 pipe (right) after HDD trial.

for installations wihtout sand-bedding and with trenchless technologies even without additional protection layers. To solve remaining questions to this issue Evonik is running additional tests like field installations and laboratory tests at Hessel Ingenieurtechnik GmbH.

4.3 First directional drilling trial at Tracto-Technik Gmb&CoKG (TT) test centre To investigate the behavior of PA12 pipes during the directional drilling technology (HDD) of TT a 4 m test sample of a 110 mm SDR-11 VESTAMID LX9030 pipe was drilled at TT test field in Lennestadt, Germany. To enable a direct comparison to PE a standard PE100 pipe with the same dimension was drilled in parallel. Both pipes were drilled thru a depth of 4 m and a distance of 60 m. The drilling diameter was 135 mm and the soil consists of a pebbles – sand mixture. After pulling out the pipes a visual investigation of the outer surface showed that the PA12 pipe has only very tiny scratches compared to the PE pipe (Fig. 1). A closer look on the surface by microscopy and semi electron scanning (SEM) proved this result (Fig. 2). Additionally hydrostatic pressure tests were set up in acc. with ISO 13479 and ISO 22621-2 to show the excellent performance of the pipe material even after the drilling trial. Results will be presented at the conference. The excellent SCG and point loading behavior combined with the minimized notch sensitivity underlines that PA12 is a promising candidate for trenchless technologies like directional drilling and pipe bursting as well. 72

Figure 2.

Depth of the scratches after the trial, LX9030 (left) vs. PE100 (right).

5 CONCLUSION Summarizing the field test experiences and the laboratory tests in acc with international standards it can be concluded that PA12 is a suitable material for high pressure applications. Therefore PA12 extends the well known benefits of PE pipe systems to the pressure range of steel pipes up to 18 bar at gas applications. The test results also underline that along with the traditional way of installation also trenchless technologies are suitable. The outstanding SCG and point loading resistance of PA12 is one of the strength of this promising material for high pressure applications. REFERENCES Dowe A., Baron Ch., Buessing R. 2007. New VESTAMID® (Polyamide 12) grades to manufacture products with large geometrical dimensions for oil and gas applications. Conference book. Plock: Technical Conference Risk Management for Pipeline Operation 2007 Wessing W., Grass K., Kanet, J., Capdevielle, J-P. 2004. Novel PE gas supply system for a maximum operation pressure of 16 bar, Conference book. Vancouver: Gasresearch Conference 2004 Bayer, H.-J.2006. HDD-Practise Handbook. Vulkan Verlag Essen 2006 Rameil, M. 2007. Handbook of Pipe-Bursting Practise. Vulkan Verlag Essen 2007

73

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Simulation researches of pump-gravitational storage reservoir and its application in sewage systems J. Dziopak & D. Sły´s Department of Infrastructure and Sustainable Development, Rzeszów University of Technology, Poland

ABSTRACT: Low accessibility of municipal building sites and growing demands in the field of volume efficiency and reduction of net-work depth below storage reservoirs limits the range of economically reasonable applications of storage reservoirs. This fact brought to the beginning of the researches directed to elaboration of new constructions of storage reservoirs of limited building surface. The paper deals with the hydraulic systems’ solutions of pump-gravitational waste water retention. The range and results of simulation researches of such type of sewage systems’ objects are also presented.

1 INTRODUCTION The problem of waste water transport by sewage systems and waste water management in urban areas is actually very significant in the light of the increase of proofed surfaces in catchments and frequency of extreme atmospheric precipitation occurrence. The development of urban areas influences on the reduction of green and non-hardened surfaces, from which surface flows could feed ground waters. At the same time the demands and standards concerning the preservation and drainage of urban areas are becoming more rigorous. The effect of this phenomenon is the growth of storm water flows discharged by sewage systems. The growth of storm water and waste water quantity and their great variability in time negatively impacts the processes of waste water treatment and surface water quality and demands the expansion of existing sewage infrastructure. Modern sewage systems requires, then, the usage of efficient methods of waste water flows regulation at the stage of their transport by sewage systems and before waste water treatment plant. The problem of waste water flow control in sewage systems is the subject of many qualitative and quantitative researches. The researches are based on the usage of modern soft-ware instruments for simulation of sewage systems’ functioning. Interesting researches in the field of simulation of storm waters’ influence on sewage system and waste water treatment plant, acting jointly with storage reservoir, were carried out by Calabrò and Viviani (2006). Calabrò (2001) confirmed the significant influence of storage reservoir on contamination’s removal in the process of sedimentation, that is important in the case of objects located on canals, discharging storm water to recipients and before primary sedimentation tanks. The author also showed that first rain portion in different catchments has different content and influences the efficiency of storm water reservoirs, but under definite flow intensity the increase of storage reservoir’s volume do not impacts significantly on the suspended substances’ removal efficiency. Waste water flow control is very important in the case of sewage systems with significant contribution of combined sections, that brings to great variability of waste water flows and negative impacts on waste water treatment plants (Diaz-Fierros et al. 2002). In such cases significant improvement of storm water quality and flows stabilization are obtained by application of different hydraulic schemes, where the main role is played by storage reservoirs (Huebner & Geiger 1996). 75

2 HYDRAULIC SCHEMES’ CONCEPTION FOR PUMP-GRAVITATIONAL STORAGE RESERVOIRS The main problem in sewage systems’projecting is the necessity of obtaining of significant volumes of storage reservoirs in the areas of high level of urbanization and low availability of building surfaces. This problem concerns reservoirs located within waste water treatment plants and city areas. Taking into account the significance of this problem for the development of sewage infrastructure, the authors are carrying out the researches of storm water accumulation processes in storage reservoirs with pump-gravitational hydraulic scheme of waste water retention. The usage of pump methods of transport enables the constructing of reservoirs of high level of storm water accumulation without of necessity of excavation of chambers and canal, located below the reservoirs. In some solutions there is a possibility of shallow location of inflow canal of reservoir. The result of the researches is the elaboration of few reservoirs constructions which can be classified in following ways: – pump-gravitational reservoirs with upper pump accumulation chamber KAW, – pump-gravitational reservoirs with lower pump accumulation chamber KAD, – hybrid pump-gravitational reservoirs. All solutions of pump-gravitational reservoirs with upper pump chambers enable the usage of open constructions of pump chambers and their building in embankments. So, these solutions are especially useful as the reservoirs within waste water treatment plants, where they can play the role of waste water pump station of high accumulation capacity. Storage reservoirs with lower pump chamber enable the building of such objects on their surface as garages, storehouses, parking etc. Hybrid reservoirs, owing to storied scheme of accumulation chambers, have the most limited building surface, so their application field are mostly the areas of high investment level and low availability of building grounds.

3 HYDRAULIC AND MATHEMATICAL MODELING FOR CHOSEN SOLUTION OF PUMP-GRAVITATIONAL RESERVOIR Hydraulic modeling of storage reservoirs is the base of mathematical description of their functioning in sewage systems (Dziopak & Sły´s 2007, Sły´s & Dziopak 2006) and of simulation programs (Dziopak & Sły´s 2007, Sły´s 2006). The idea of hydraulic modeling of storage reservoir is based on separation of characteristic phases of its functioning in sewage system and definition of marginal conditions for reservoirs’ filling by waste waters in each chamber and the range of characteristic flows. Figure 1 illustrates the hydraulic scheme of pump-gravitational reservoir of GPWT type and characteristic parameters, used in elaborated model of reservoir functioning. Taking into account the complexity of hydraulic processes during waste water accumulation, synchronism of their occurrence in few chambers simultaneously and probable complexity of route of inflow function and the resulting curvilinear functions, describing the processes inside the reservoir, as well as the variability of outflow, the general mathematical model of reservoir functioning can be described by Equation 1.  dh   = QA · FT−1 · dt − Qt · FT−1 · dt − Qp · FT−1 · dt − Qr · FT−1 · dt     dt dHw = Qp · FW−1 · dt − Qw · FW−1 · dt − Qz · FW−1 · dt  dt      dhs = Qt · F −1 · dt + Qw · F −1 · dt − QO · F −1 · dt S S S dt 76

(1)

KS

KAW

KT

QO hsnor hsmax hs hz Qw Qt Hg

Figure 1.

Hc

Qr ha ho homin hiHw h

QA Qp hr hrmin

HPa Qz

Hydraulic scheme of GPWT type storage reservoir and characteristic parameters of hydraulic model (QO – waste water outflow from storage reservoir; Qw – waste water outflow from accumulation chamber KAW to steering chamber; Qt – reduced waste water flow from transport chamber KT to steering chamber KS; Qr – waste water outflow from transport chamber KT through emergency overflow; QA – waste water inflow from sewage system to storage reservoir; Qp – capacity of pump system transporting waste water from KT to accumulation chamber KAW; QZ – waste water outflow from accumulation chamber KAW through emergency overflow; KAW – gravitational accumulation chamber; KS – steering chamber; KT – transport chamber with pump system; hsnor – level of filling in steering chamber KS in the period of dry weather with reference to comparative level; hs – temporal level of filling in steering chamber KS with reference to comparative level; hsmax – maximum level of filling in steering chamber KS in the period of storm water inflow or in the period of accumulation chamber KAW emptying with reference to comparative level; hz – average bottom elevation of steering chamber KS with reference to comparative level; Hg – elevation of upper edge of outlet; Hc – average elevation of accumulation chamber’s bottom with reference to comparative level; ha – elevation of emergency overflow edge of transport chamber KT with reference to comparative level; ho – level of switching of waste water pump transport system from transport chamber KT to steering chamber KS and waste water treatment plant with reference to comparative level; homin – level of switching off the waste water pump transport system from transport chamber KT to steering chamber KS and waste water treatment plant with reference to comparative level; hi – average elevation of transport chamber KT bottom with reference to comparative level; Hw – temporary filling of accumulation chamber KAW with reference to comparative level; h – temporary filling of transport chamber KT with reference to comparative level; hr – level of switching of waste water pump transport system from transport chamber KT to accumulation chamber KAW with reference to comparative level; hrmin – level of switching off the waste water pump transport system from transport chamber KT to accumulation chamber KAW with reference to comparative level; Hpa – elevation of emergency overflow edge of accumulation chamber KAW with reference to comparative level).

Minimal marginal values for filling of storage reservoir’s chambers are: h ≥ hi, Hw ≥ Hc, hs ≥ Hz.

4 SIMULATION PROGRAM SIMTANK On the base of hydraulic and mathematical model of GPWT type storage reservoir, software instrument was elaborated that enables the simulation of hydraulic processes in storage reservoir of this type for any route of waste water inflow function. 77

Height of waste water filling in the accumulation chamber Hw, m3

8 Vzb max = 3829 m3, Hw = 7.66 m

7 Td = 50 min

6 Vzb max = 2824 m3, Hw = 5.65 m

5 Vzb max = 1709 m3, Hw = 3.42 m

Td = 35 min

4

3 Td = 20 min

2

1 0

0

20

40

60

80

100

120

140

160

180 200 time t, min

Figure 2. The level of filling Hw and waste water volume Vzb max, stored in accumulation chamber, for rain duration Td = 20, 35 and 50 min. The rest simulation parameters: average annual precipitation H = 720 mm; probability of calculation precipitation occurrence p = 50%; catchment’s area F = 125 ha; surface flow coefficient = 0,8; coefficient of domestic waste water dilution at overflow nrp = 6; coefficient of domestic waste water dilution in transport chamber nrz = 2.

Calculation program SIMTANK makes possible the dynamic simulation and analyzing of hydraulic processes in storage reservoir during waste water inflow. Simulation program SIMTANK considers the order and sequence of reservoir’s functioning stages in accordance with hydraulic model of its functioning. While operating, the program makes the analyses for demanded parameters of marginal conditions and calculated parameters. At the moment of achievement of marginal conditions for particular phases of reservoir’s functioning, the transition between consecutive stages of its functioning takes place. Such conditions are: characteristic waste water levels in reservoir; waste water inflow’s duration and intensity. Figure 2 presents the results of exemplary simulation of filling process in accumulation chamber KAW for the rains of different duration and different outflow functions in the shape of triangle. One of the most important stages of storage reservoirs’ calculation is the estimation of the route of function of waste water inflow to storage reservoir. The modern calculation instruments in the form of hydrodynamic catchments models enable the simulation of waste water flow functions’ routes for any precipitation intensity, that is very useful in the case of project parameters’ selection for storage reservoirs. In the case of the lack of information about precipitation and the lack of possibility of real waste water flow diagrams’ usage, the substitutive waste water flow diagrams can be used. In this situation the most significant calculation parameter is duration of calculation rain Tdm. The process of this parameter’s selection for substitutive diagrams were described by Dziopak in his publications (1997). 78

Duration of representative rain for projecting of storage reservoir Tdm, min

1200

1000

p = 50 % p = 20 % p = 10 %

800

600

400

200 500

600 700 800 900 1000 1100 Average annual precipitation H, mm

Figure 3. The influence of average annual precipitation H on duration of representative rain Tdm for projecting of storage reservoir under following calculation data: unit discharge of domestic waste water flow qs = 2 dm3 /s ha; catchments area F = 200 ha; surface flow coefficient = 0,6; coefficient of domestic waste water dilution at overflow nrp = 4.

Figure 3 presents the results of calculations of rain durationTdm, appropriate for storage reservoir projecting of GPWT type in dependence of the following hydrological parameters of catchments: average annual precipitation H and probability of the rain for the projecting of storage reservoir p. Rain duration Tdm depends on the number of hydrological and hydraulic parameters as well as on hydraulic capacity of waste water treatment plant, the way of catchment’s usage and water consumption. Figure 4 presents the results of simulation of rain duration Tdm appropriate for projecting of storage reservoir, located between storm overflow and waste water treatment plant in dependence on the level of surface sealing in catchment’s area. 5 CONCLUSIONS The exploited gravitational sewage systems are under constant expansion and modernization. The main problems for technical solution are: hydraulic overloading of sewage nets, surface water protection against contamination discharged by sewage systems and control of waste water flow to waste water treatment plants. These problems can be solved successfully with the help of modern efficient constructions of reservoirs for periodic retention of waste water. Taking into account the merits and imperfections of existing constructions of storage reservoirs within storm and combined sewage systems, the researchers of new group of solutions of pumpgravitational scheme of waste water retention have been carried out. Such type of storage reservoirs makes possible to store waste water in accumulation chambers of significant heights and limited grounds for building, that is especially important for urban areas. 79

800 750 Duration of representative rain for projecting of storage reservoir Tdm, min

700 650

H = 700 mm H = 800 mm H = 900 mm

600 550 500 450 400 350 300 250 200 150 100 50 0,3

0,4

0,5

0,6

0,7

0,8

Waste water flow coefficient for catchment's area 

Figure 4.

Influence of waste water flow coefficient for catchment’s area on rain duration Tdm, appropriate for storage reservoir projecting for the following calculation data: probability of calculation rain occurrence p = 50%; domestic waste water unit inflow qs = 1,5 dm3 /s ha; coefficient of domestic waste water dilution at overflow nrp = 5; catchment’s area F = 500 ha.

Simultaneously, owing to the idea of multi-chamber reservoirs’ use, which are equipped with transport chamber steering the operation of accumulation chambers, the high efficiency of waste water accumulation was achieved. Equipping of multi-chamber reservoirs by pump transport schemes and outlet schemes enables the options in chambers’ hydraulic forming, that influences on the broadening of the field of their application, economically and practically motivated. The elaborated calculation programs SIMTANK, serving for the purposes of simulation of GPWT type storage reservoir operation, are the modern researching and projecting instruments for detailed investigation of waste water accumulation processes in such types of reservoirs, located before waste water treatment plants. The use of SIMTANK programs in project processes for selection of main hydraulic parameters, chambers’ capacity and geometry, as well as installations acting jointly with them, enables the wide usage of pump-gravitational constructions of reservoirs, especially for the purpose of waste water treatment plants’ hydraulic unloading. REFERENCES Calabrò, P. S. & Viviani, G. 2006. Simulation of the operation detention tanks. Water Research 40: 83–90. Calabrò, P. S. 2001. Cosmoss: conceptual simplified model for sewer system simulation. Urban Water 3(1–2): 33–42. Diaz-Fierros, F., Puerta, J., Suarez J. & Diaz-Fierros, V. 2002. Contaminant loads of CSOs at the wastewater treatment plant of a city in NW Spain. Urban Water 4: 291–299.

80

Dziopak, J. 1997. Multi-chamber storage reservoirs in the sewerage system. Technical University of Czestochowa. Czestochowa. ˛ ˛ Dziopak, J. & Sły´s. 2007. Modelowanie zbiorników klasycznych i grawitacyjno-pompowych w kanalizacji. Rzeszów University of Technology. Rzeszów. Huebner, M. & Geiger W. 1996. Characterisation of the performance of an off line storage tank. Wat. Sci. Tech. 34(3–4): 25–32. Sły´s, D. 2006. Simulation model of gravitation-pump storage reservoir. Environment Protection Engineering 32(2): 139–146. Sły´s, D. & Dziopak, J. 2006. Simulation of Trough-Flow Chamber Operation in Storage Reservoirs. Ecological Chemistry and Engineering 13(10): 1143–1155.

81

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

New developments in liner design due to ATV-M 127-2 and case studies B. Falter University of Applied Sciences, Münster, Germany

ABSTRACT: The German Code ATV-M 127-2 published in 2000 has proved itself a helpful guideline to find the optimum wall thickness of any liner material, e.g. CIPP or stainless steel sleeves. Many rehabilitation projects in different European countries have been performed successfully using this code. The code differentiates between three host pipe states: State I for untight sewers without cracks, state II for sewers with longitudinal cracks but a stable soil pipe system and state III for cracked pipes with larger deformations and considerable risk to collapse in the near future. According to the code stress, deformation and stability tests are necessary. For many practical cases charts with stress factors and imperfection reductions allow to design without a computer. The paper reports about the progress in liner design since the 1st edition of the code. Additional clauses have to be introduced into the 2nd edition for non circular geometries (e.g. for horseshoe and rectangular profiles) and for new applications (e.g. railway crossings).

1 INTRODUCTION The critical water pressure equation is based on a Glock formula (Glock 1977) for elastic rings encased in a rigid boundary. This equation was used by many researchers and enhanced by reduction factors to describe the real situation of the host pipe (e.g. deformations and annular gap). In a few codes the critical pressure of an unsupported ring (Timoshenko 1961) is used and increased by a support factor K; for example K = 7 for good and 4 for poor installation conditions (WRc/WAA 2000). The critical soil pressure is treated less often in research papers. The reason might be that experiments on the broken pipe soil-system with an overburden (e.g. Watkins 1988) show a conservative behaviour without a collapse. On the other hand in such experiments uniformly distributed pressures are applied onto the sandbox surface more often than concentrated wheel forces. 2 DESIGN CONCEPT OF THE CODE M 127-2:2000, 1ST EDITION 2.1 Host pipe state The German liner design concept is based on the differentiation of three host pipe states. The state I and II (without and with longitudinal cracks) must be calculated only for groundwater acting as a pressure on the outside of the lining. In case of state III an additional calculation for soil and traffic loads is prescribed. 2.2 Buckling pressure, imperfection reductions and proof of stability Due to the Code ATV-M 127-2 the buckling load for the water pressure pa valid for all host pipe states is evaluated regarding three kinds of imperfections, cf. Figure 2: a) Local imperfection wv = 2% which must be chosen according to the relevant buckling mode b) Annular gap ws = 0.5% for CIPP caused by shrinkage of the liner material c) Global imperfection wGR,v ≥ 3% caused by the deformation of the cracked host pipe 83

Figure 1a-c.

Host pipe state I (a), II (b) and III (c) as defined in the German Design Code ATV-M 127-2.

(a)

(b)

(c)

wGR,v 3%

ws = 0.5% rL

rL

rL

cracked four times (hinges)

2 1

wv = 2%

Figure 2a-c.

Local imperfection wv (a), annular gap ws (b), global imperfection wGR,v (c), minimum values.

For the imperfections in Figures 2a-c reductions factors κ for the buckling load are given in the Design Code. The factors depend on the depth of the imperfection and the rL /sL ratio describing the slenderness of the liner construction and the character and the size of the host pipe damages. For the critical water pressure pa of a circular lining the following formula has been developed (Falter 1993). crit pa = κv,s · αD · SL

(1)

where κv,s ∼ = κv · κs · κGR,v is the common reduction factor for all imperfections due to Figures 3a-c; αD = 2.62 · (rL / sL )0.8 is the snap through factor (Glock 1977); and SL = EL /12 · (sL / rL )3 is the ring stiffness of the liner (EL = long-term Young modulus). The minimum values for wv , ws and wGR,v to be applied are given in Figures 2a-c and 3a-c. The main problem in the practice is the correct assumption of the global imperfection wGR,v . Usually the value of wGR,v has to be evaluated from a video screen which results in sometimes different opinions of the engineers about this issue. For non circular linings a computer evaluation of the critical water pressure is necessary or a substitute radius on the safe side has to be taken for rL in Equation 1. It is strongly recommended to calculate the substitute radii by separate non linear analyses; sometimes the amount and the location of the local imperfection must be varied to get proper results. The correct location of this kind of imperfection is usually the region with the smallest curvature, e.g. the invert for horseshoe profiles; the region beneath one springline for egg shaped profiles. 2.3 Proof of stresses and strains The section forces M and N caused by groundwater pressure pa are calculated by means of dimensionless factors mpa and npa given in appendix A4 of M 127-2 for the diameters ND 200 – ND 600, cf. Figure 4. Mpa = mpa · pa · rL2

with mpa due to Code M 127-2, appendix A4 84

(2a)

0,0

rL

rL/sL = 10

wv

15 20 25 35 50

s

v

1,0 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1

100

0

2

4

6

1,0 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0,0

8

rL/sL = 10

rL

ws 15 20 25 35 50 100

0

1

2

3

4

5

GK,v

ws/rL·100%

1,0 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0,0

20

15

rL/sL =

wGR,v 25 rL

35 50 100

0

2

Figure 3a-c.

4

6

8

10

Reduction factors κv , κs and κGR,v for the buckling pressure due to local imperfection wv , annular gap ws and global imperfection wGR,v , source: ATV-M 127-2:2000. 0,12

Host pipe state II (3% ovalisation) Host pipe state I

0,10

ND 600

mpa

0,08

sL[mm] = 10

10

12.5

0,06

15 12.5

0,04

15

0,02 0,00

Figure 4.

1

2

3 4 5 max hw above the invert [m]

6

7

Invert bending moment factors for liners ND 600, source: M 127-2, Appendix A4.

Npa = npa · pa · rL

with npa ≈ −0.8 to − 1.1

(2b)

The stresses are calculated using Equations 3a, b: N M (3a) + αki · with αki = 1 + sL /3rL , αka = 1 − sL /3rL A W N M σa = (3b) − αka · A = sL in mm2 /mm and W = sL2 /6 in mm3 /mm A W The resulting tensile stress is compared with the ultimate flexural strength of the lining material, reduced for long-term behaviour. As the pressure stresses due to Equation 3b are a bit larger an additional material pressure test leads in many cases to a more economical design. σi =

85

For host pipe state III the soil and the traffic loads qv are applied to the pipe’s crown and the structural safety of the total system of soil, cracked host pipe and lining is analysed. The axial force and bending moment factors nq and mq are given for ND 200 to ND 600 in appendix A5 of Code ATV-M 127-2. Elaborate experiments have been necessary to prove the design formula, referring to national and international research, cf. chapter 4.3. 2.4 Discussion The stress proof in chapter 2.3 is the most important test. The factors mpa calculated non linearly with appropriate imperfections contain the risk of stability failure. In many cases computer analysis is necessary as the validity of the cross section factors is restricted to a range of material properties. The stability proof in chapter 2.2 has the advantage of general validity for all diameters and all material properties. This proof can be used for approximate evaluations of the wall thickness without computer programs. Deformation analyses are less important and used sometimes for judgement of serviceability. 3 EXTENSIONS OF THE DESIGN CODE M 127-2, 2ND EDITION 3.1 Additional reduction factors for the buckling load The regular values for most rehabilitation situations with CIPP-liners are 2% local imperfection, 0.5% annular gap and an arbitrary global imperfection. For this constellation only one chart with reduction factors κv,s for the critical pressure is necessary, cf. Figure 5. 3.2 Simultaneous calculation of water and soil pressure In the Design Code ATV-M 127-2:2000 the load cases pa and qv are treated separately. Subsequently they must be combined by an interaction formula. New developments show that an enhanced numerical model is able to cover both load cases in one step. For host pipe state III without water 0,9 wv /rL = 2.0 % ws / rL = 0.5 %

0,8 rL / sL = 10

0,7

κv,s

0,6

15 20 25 35

0,5 0,4

50

0,3 100

0,2 0,1 0,0 0

1

2

3

4

5

6

7

8

9

10

wGR,v / rL·100%

Figure 5.

Reduction factors κv,s of the buckling pressure, 2% local imperfection, 0.5% annular gap and arbitrary global imperfection (Falter 2003).

86

table a double symmetry of the system could be assumed; in the case of a load combination this is however no longer possible, cf. Figure 6a,b. 3.3 Lining of flexible sewers Sometimes the future integrity of the old pipe is called into question. In the Design Code ATVM 127-2 it is assumed that the host pipe has enough strength to support the lining in the radial direction. If the absence of any host pipe structure is expected in the future new analysis problems arise: The liner without host pipe. The main stresses are now to be expected in the liner crown, cf. Figure 7a. 3.4 Soil bedded lining (host pipe fully detriorated) Sometimes the future integrity of the old pipe is called into question. In the Design Code ATVM 127-2 it is assumed that the host pipe has enough strength to support the lining in the radial direction. If the absence of any host pipe structure is expected in the future new analysis problems arise: The liner without host pipe. The main stresses are now to be expected in the liner crown, cf. Figure 7a. 3.5 Host pipe-soil system In order to prove the host pipe soil-system the equilibrium of a rigid circular ring with four excentric hinges is investigated, cf. Figure 8. For regular cases with elastic soil behaviour the load deflection curve is described by the following equation: ξ · ηS · (ρGy − ηS /3 − η) qν = (1 + ξ) · (2ρGx − 1 + ξ) − K2 · (1 − η) · (2ρGy − 1 − η) SBh

(4)

where ρGx and ρGy = horizontal and vertical distance of the hinges from the pipe wall centre; η = deflection of the pipe’s crown to the inside, ξ = outside deflection of the springline; (b)

(a)

Figure 6a,b.

(a)

Figure 7a-c.

Load cases groundwater pa and soil pressure qv – a system without horizontal symmetry, (a) bending moments of the liner, (b) contact forces of the liner versus the host pipe (Linerb 2008).

(b)

(c)

Bending moments (a), contact forces (b) and deflections (c) of a circular lining subjected to groundwater-host pipe neglected (Linerb 2008).

87

qv qh eG qh* rm 90°

s

Figure 8.

Pipe with longitudinal cracks and soil pressures qv , qh and qh∗ (support). 0,09 0,08

eG = 0,45s

0,07

eG = 0,25s

0,06

qv / SBh

0,05 0,04

wv = 0%

0,03

3%

eG = 0

0,02 6%

0,01 9%

0 0

5

10

15

20

v [%]

Figure 9.

Load deflection-curves of the pipe soil-system for soil group 1, varying imperfections and hinge excentricities eG.

ηS = vertical extension of the side bedding (all parameters related to the pipes radius); and K2 = factor of lateral soil pressure. Equation 4 yields the dimensionless ratio of the crown loading qv and the horizontal bedding stiffness SBh of the soil. The equation has been extended for initial deformations and plastic soil behaviour. The resulting load deflection curves show maximum values crit qv which are valid for cracked pipes surrounded by soil. In appendix A6 of the Design Code ATV-M 127-2 Equation 4 has been evaluated for varying soil groups and hinge excentricities, cf. Figure 9. 4 THEORETICAL BACKGROUND AND EXPERIMENTAL EVALUATION 4.1 Numerical models used for liner design Several models used in practice and research studies are shown in Table 1. The most usual one is the beam structure, cf. Figure 10a. 88

Table 1.

Numerical models in liner design.

Model

Application

Advantages

Disadvantages

Example

1. Beam model

Basis of the diagrams in Code M 127-2:2000, familiar in practice

Inexpensive, quick dimensioning, possible for all profiles

Lining material regularly linear

Chapters 5.2–5.4

2. Two dimensional

Special profiles

Non linear behaviour of soil and lining material

Relatively complex

–

3. Three dimensional

Special investigations, no constant situation in longitudinal direction

Research, spatial load distribution, anchored linings etc.

Complex and expensive

Chapter 5.1

Figure 10a.

Beam model for egg-shaped liners, loading.

3D-models are necessarily applied for non constant situations in the longitudinal direction e.g. anchored linings, non familiar geometrical or structural imperfection distributions or load cases from grouting the annular space between the host pipe and the liner. Stresses in lings for pressurized pipes must be calculated by such models as well. 4.2 Numerical model as a basis of the design code ATV-M 127-2:2000 The computer model is a structure of connected arcs described by beam elements and rigidly supported by the host pipe, cf. Figure 10a. All necessary imperfections are introduced in the model and can be chosen arbitrary. The calculation is performed non linearly in order to evaluate stress and stability limits as well. Figure 10b shows an appropriate solution for host pipe state II with a single lobe deflection. The deflections in Figure 10c have been calculated assuming a water pressure exceeding the buckling load. The beam model for an egg shaped liner in Figure 10a includes a local imperfection at the right side beneath the springline. 4.3 Experimental evaluation From numerous experimental research on lining stability two buckling test series are shown in the Figures 11 and 12. The correct way to perform such buckling tests regarding the material’s creep tendency is to estimate the time until buckling and to apply the pressure at the outside as a load constant with time. 89

Figure 10b. Admissible deflection with one lobe in the invert.

Figure 11.

Figure 10c.

Long-term buckling tests on CIPP linings, Louisiana Tech University, Ruston, USA (Guice et al. 1994).

Not acceptable multi-lobe figure for host pipe state I.

Figure 12.

Creep buckling tests on eggshaped PE linings, University of Applied Sciences Münster, Germany (Falter et al. 2008).

The results are in good agreement with the calculated critical pressures, using the buckling Equation 1 or the numerical model described in chapter 4.2.

5 CASE STUDIES 5.1 Hafenkanal Düsseldorf (2004) The Hafenkanal sewer ND 2500 mm in Düsseldorf, Germany made of reinforced concrete was heavily damaged by sulphuric corrosion. Nevertheless it could be classified as a host pipe state I sewer. One proposal for the renovation was to cover the sewer’s surface by anchored 8 mm thick polyethylene sheets with backside burls. Thus the sheets are drained by an annular gap and notches at both springlines. The distance of the bolts was originally planned as 0.5 m in axial direction but could be increased to 4 m by the calculations, cf. Figures 13–14. 90

Figure 13.

Beam model, bending moment. Figure 14.

3D simulation, imperfection, deformations.

Figure 15.

Erection device to press the 4 m long polyethylene sheets against the culvert.

Figure 16.

Sewer after renovation, Düsseldorf, Germany (Photos: H.I. Hammer).

Beam and three-dimensional models were examined. The 3D- model was necessary to analyse the shell bearing behaviour. The dead load of the liner and temperature changes of ± 7.5 K were applied; a maximum crown deformation of 15 mm was allowed for all load combinations. The installation of the sheets and the finished work are shown in the Figures 15 and 16. 5.2 Separation of the Münzbach River from waste water in Freiberg, Saxonia (2005) In the horseshoe-shaped concrete sewer in Freiberg, Saxonia in Germany, the water of a small river was transported simultaneously. Thus the aim was to separate the streams within an assembled lining and to improve the load-bearing capacity of the sewer. During the analysis of the sewer and the lining for host pipe states II and III some special problems occured: 1. Analysis of the existing sewer’s stability suffering from side cracks and infiltration. 2. Optimization of the sewer’s shape to avoid excessive arching forces from the flat invert caused by groundwater pressure, cf. Figures 17a, b. 3. Application of safe imperfections to the flat part of the cross section, cf. Figures 18a, b. In the present case a non symmetrical pre-buckle must be applied. 4. Quality control of the liner material and the bonding connections of the segments, cf. Figure 19. 91

(a)

(b)

reduced and distributed stress

concentrated force

Figure 17a,b.

Contact forces between polycrete liner and grout, (a) sharp edges and (b) rounded edges .

Figure 18a,b.

Model of the invert shallow arch subjected to water pressure pa , (a) symmetric and (b) non symmetric imperfection.

Figure 19.

Polymeric concrete invert element.

Figure 20. Assemly of the liner elements, bonding.

For the project dimensions l = 2.75 m and f = 0.25 m the following horizontal forces H caused by a water table of 1.5 m above the invert (pa = 15 kN/m2 ) result: H = pa · l 2 / 8f = 15 · 2.752 / (8 · 0.25) = 56.7 kN/m. As a consequence of the high forces H the liner shape must be a carefully chosen in this region and the edges should be rounded. Figures 19 and 20 show a bottom element with sharp edges as designed in the first contruction phase and the installation. For the second phase elements of the whole cross section with rounded edges were manufactured, cf. Figure 17b. 5.3 Masonry sewer crossing a railway in Krefeld (2007) As the egg shaped sewer B/H = 1200/1800 mm sewer crossed a railway in Krefeld, Germany it was necessary to rehabilitate the full length of 350 m between the two manholes in one step. Thus the 92

Figure 21.

Sewer with longitudinal cracks.

Figure 22.

Installation of the CIPP lining.

12 hW = 2.5 m 10

hW = 4.5 m

req sL

8 6 4 2 0 M 127-2

Figure 23a.

Danish St.

US Stand.

Comparison of the required wall thickness of linings, host pipe state I (water table hW ).

required wall thickness was a question if the rehabilitation could be done at all. The main design issue for this project was the crown deflection. The first approach for the crown deformation was 3%, but an inspection with different measurements (height, width, crown gap width and angle difference in the crown, cf. Fig. 21) yielded 5% deformation except a part of 15 m length with 10%. Thus two liner wall thicknesses were manufactured: 23 mm regularly and 30 mm in the largely deformed area. It was possible to restrict the whole weight of the wet lining to 180 tons, the ultimate weight for a street transportation, cf. Figure 22. 5.4 Railway crossing Banbury station sewer, UK (2007) The Banbury Station Sewer had a brick walled cross section changing over the length: rectangular, horse shoe, circular ND 800. Therefore and regarding the severe damages of the sewer a slip lining procedure was chosen with a grouted annular space. A part of the sewer had to be exchanged by a new GRP pipeline and open cut. Regarding the shallow cover above the sewer of only 0.69 m the main problem in design was the requirement of 93

req sL

18 16 14 12 10 8 6 4 2 0

host pipe support soil support

M 127-2

Figure 23b.

Danish St.

US Stand.

Comparison of the required wall thickness, host pipe state III (soil cover h and traffic load).

the railway authorities to ensure that the settlements under the sleepers should not exceed 3 mm. Due to the British Standards newly laid pipes should have a minimum cover of 1.5 m. Because of the vicinity to the wheel loads a fatigue analysis for the pipeline material was necessary. Many liner materials have not been tested under cyclic loading yet. 6 INTERNATIONAL COMPARISON OF DESIGN METHODS Several experts have compared the necessary liner wall thickness for similar loading situations – a recent paper was presented at the NoDig Roma (Kuliczkowski 2007). From Figure 23a it is obvious that the differences in wall thickness for state I (and state II with cracks) are quite small. Important for the dimensioning of the lining is that state III calculations show more significant differences. The reasons must be discussed for future development of design codes. 7 CONCLUSIONS The German Design Code ATV-M 127-2:2000 has proved itself to be applicable for designing liners in very different rehabilitation projects (e.g. host pipe geometry and material, state and damage case). The stress analysis is the relevant proof for the structural safety of the construction allowing the design of lining material with low flexural strength as well. Stability analysis is easy to perform without the aid of a computer as a first step. Appropriate imperfections have to be introduced into all analyses. The open discussion in an international group of experts is helpful to identify common problems and to improve some clauses of national design codes. Future research needs are seen for host pipe state III situations (US definition: fully deteriorated), for railway crossings (e.g. fatigue tests) and for liner quality assurance. REFERENCES ATV-M 127-2:2000. Static calculations for the rehabilitation of sewers with lining and assembly procedures. Hennef: German Water Ass. (available in English language). Danish Standards 2001. Static calculations of liners for the gravitational sewers, Taastrup. Falter, B. 1980. Grenzlasten von einseitig elastisch gebetteten kreiszylindrischen Konstruktionen. Bauingenieur 55: 381–390. http://hb051.fh-muenster.de/opus/fhms/volltexte/2005/165/ Falter, B. 1991. Standsicherheitsnachweise an Sanierungssystemen für Abwasserkanäle mit unveran-kerten Linern. 3R International 30: 50–55. http://hb051.fh-muenster.de/opus/fhms/volltexte/2005/142/

94

Falter, B., Hoch, A. & Wagner, V. 2003. Hinweise und Kommentare zur Anwendung des Merkblattes ATV-M 127-2 für die statische Berechnung von Linern. Korrespondenz Abwasser 50: 451–463. http://hb051.fhmuenster.de/opus/fhms/volltexte/2003/95/ Falter, B., Eilers, J., Müller-Rochholz, J. & Gutermann, M. 2008. Buckling experiments on polyethylene liners with egg-shaped cross-sections. Geosynthetics International. Vol 15(2): 152–164. Glock, D. 1977. Überkritisches Verhalten eines starr ummantelten Kreisrohres bei Wasserdruck von außen und Temperaturdehnung. Stahlbau 46: 212–217. Guice, L.K., Straughan, T., Norris, C.R. & Bennett, R.D. 1994. Long-Term Structural Behavior of Pipeline Rehabilitation Systems. TTC Technical Report #302, Louisiana Tech University Ruston, Louisiana, USA. Kuliczkowski, A., Kubicka, U. & Parka, A. 2007. The problems of design of resin liners. Proceedings of No Dig Conference Roma, September 10.-12. 2007. Linerb 2008. Structural analysis of linings – computer program. Interface in German and English language. Univ. of Appl. Sc. Münster. https://www.fh-muenster.de/fb6/personen/lehrende/falter/linerb.php Timoshenko, S.P. & Gere, J.M. 1961. Theory of Elastic Stability. New York: Mc Graw-Hill. Watkins, R.K., Schupe, O.K. & Osborn, L.E. 1988. Contribution of Insitupipe to the structural integrity of broken rigid buried pipes. Insituform of North America, Inc. WRc/WAA 2000. Sewerage rehabilitation manual. UK Water Research Center / Water Authorities Ass.

95

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Concrete – durable composite in municipal engineering Z. Giergiczny, T. Pu˙zak & M. Sokołowski Góra˙zd˙ze Cement S.A, Chorula, Poland

ABSTRACT: Cements and concretes with the addition of high quality mineral components are used in common and special engineering construction. Concrete based on these components have high quality parameters and resistance for external factors. Particular attention will be paid to the durability of concrete in different aggressive environments. As a sample of use of mineral additives (fly ash) authors present properties of concrete resistance to fuels and oil derivatives according to the requirements of PN-EN 858-1 and PN-EN 858-2. Concrete, the subject of the study, was made in technology of almost self-compactive concrete (ASCC). Achieved laboratory test results gave the ground for industrial applications. Composition and tests results of concrete and photos from mentioned realization will be also presented. The properties of cored specimens (visual inspection, resistance for chemical aggression) kept in different solutions will be also presented in paper.

1 INTRODUCTION Durability is the main issue of current standard PN-EN 206-1:2003 “Concrete. Requirements, properties, production and conformity”. Durability is defined as “the ability of construction to fulfil minimum of its function (property design) by forecasted lifecycle and in estimated conditions, without extra repair costs” [Brandt 2004, Fiertak 2004]. Durable concrete is the one made of components of proper quality, characterized by adequate density (w/c, cement amount, proper strength), structure (aeration) and resistance to chemical attack (proper cement and its proper amount, adequate selection of additive, proper strength). Mineral light liquids are specially hazardous for earth and underground waters. Petrol or oil once it has reached the soil decreases the absorption of oxygen and depress self-cleaning of process o water. Though, it is essential to separate oil-bearing liquids (petrol stations, logistic centres, garages) using sewage pre-treatment installations. Production of light liquid separators in concrete technology is one of the solutions. Concrete for the construction of such installations, due to repeated, extreme exploitation conditions, ought to fulfil special requirements, for example: resistance to aggressive environment, because of strongly diverse amount of pollutions in waters and sewage (oils, fuels, petrol, variations of oil-derived temperatures), high degree of water tightness, frost resistance and limited shrinkage (Czarnecki 2004, Madryas 2002, Giergiczny 2000, Neville 2000). Obtainment of concrete with desired properties requires appropriate selection of components and proper design process including density and reinforcement gauge, enough fluidity of concrete mixture, duration of transport and concreting, etc. Hereby paper presents concrete test results designed for the production of such installation type. Concrete was produced in almost self compacting concrete technology (ASCC) basing on Portland cement and the addition of siliceous fly ash. 97

Separator without protective coating

Figure 1. Table 1.

Separator with coating

Examples of light liquid separators. Requirements for concrete.

Property

Requirement

Compressive strength after 28 setting days Chemical resistance of concrete

≥45 N/mm2 (MPa) Concrete after exposure of samples for 1000 hours in following environments, must present compressive strength ≥45 N/mm2 (MPa): Deminaralized water with temperature 40◦ C ± 2◦ C, Fuel oil with temperature 23◦ C ± 2◦ C, Unleaded fuel with temperature 23◦ C ± 2◦ C, Mixture with content according to PN-EN 858-1 with temperature 40◦ C ± 2◦ C.

2 TESTS AND RESULTS 2.1 Project assumptions Plastic protective coatings with high resistance to chemical attack are commonly used solutions to protect reinforced containers against the influence of hazardous chemical substances (Fiertak 2004). Process of covering the containers with protective coatings requires the observation of hard discipline as well as leading the wide scope of tests during the exploitation time (thickness of dry plate, adhesion, resistance to impact, resistance to scratches, porosity). One of the disadvantages of the coatings is relatively quick process of ageing. Alternative solution, however, is the application of proper concrete quality for the production of containers and separators. Concrete resistant to light liquid performance must comply with the requirements of standard PN-EN 858-1 “Installations of light liquid separators. Part 1: design rules, useful properties and research, marking and quality control” and in standard PN-EN 858-2 “Installations of light liquid separators. Part 2: Selection of rated values, installing, usage and exploitation”. They are presented in table 1. 98

Table 2.

Design properties for ASCC concrete.

Requirements for concrete mixture

Requirements for hardened concrete

Consistency of concrete mixture: 60–65 cm measures as flow No component segregation Stable consistency in time min. 30 minutes Proper concrete vacuum process

Compressive strength for concrete after 1 setting day – minimum 20 MPa Concrete classes C 35/45 Water permeability ratio W 8 Depth of water penetration – maximum 50 mm Water absorbability – nor more than 5% Frost resistance ratio F 150 Chemical resistance according to PN-EN 858-1; resistance criteria are shown in table 1

No “surface water floating”

Table 3.

Requirements of Portland cement CEM I 42,5R.

Property Consistency change, Le Chatelier Beginning of setting time Compressive strength after 2 days Compressive strength after 28 days

Requirements acc. to PN-EN 197-1

Plant laboratory test results

≤10 mm ≥60 min ≥20 MPa ≥42,5 MPa ≤62,5 MPa

0,5 mm 153 min 26,4 MPa 50,4 MPa

Carried tests assume the production of concrete in almost self compacting technology (ASCC) with properties specified in table 2. They include the requirements related to production process of such product assortment in plant conditions. 2.2 Components of concrete mixture 2.2.1 Cement Baring in mind, that desired early strength level, allowing for unmoulding and transport of finished product at the area of precast plant, cement used in tests was of CEM I 42,5R strength class with the properties presented in table 3. Portland cement I 42,5R is commonly used in fine and large precast elements production, compressed elements, road and bridge construction as well as concretes BWW and SCC. 2.2.2 Aggregate To design concrete mixture recipe, sand and grit aggregate with maximum grain size of Dmax 16 mm were used, though designed grain distribution curve is presented on fig. 2. Special attention was given to proper selection of sand grain size. 2.2.3 Fly ash Fundamental for concrete mixture construction in ASCC technology is proper amount of dust fraction in concrete (grains with size under 0,125 mm). ASCC technology demands the implementation of adequate mineral fillers into concrete composition (Giergiczny 2002). It effects in low value of liquid limit and high flow of concrete mixture (Giergiczny 2002, Domone 2006). Cement can not play such role as it would lead to over-shrinkage; economical aspect is also here of great importance. Properly selected mineral additive, together with cement, guarantee correct flow of concrete mixture 99

100 Border curve (acc to PN-B-06250)

90 80

Mixture curve

Content, %

70 60 50 40 30 20 10 0 0

0,125

0,25

0,5

1

2

4

8

16

Sieve mesh dimension #, mm

Figure 2.

Table 4.

Grain distribution curve of aggregate mixture.

Physical properties of fly ash.

Loss of ignition [%]

SO3 [%]

CaOfree [%]

Cl− [%]

Pozzolana activity [%] After 28 days

After 90 days

Fineness [%]

Density [g/cm3 ]

2,24

0,67

0,07

0,007

78,4

89,7

34,0

2,13

Table 5.

Chemical composition of fly ash.

SiO2

Al2 O3

Fe2 O3

CaOfree

CaO

MgO

SO3

Na2 O

K2 O

Cl−

51,5

27,8

7,5

0,07

3,7

2,5

0,67

1,1

3,0

0,007

and provides for the extension of workability time (rework) of concrete. Siliceous fly ash, used in carried study, with its composition and properties is presented in tables 4 and 5. Used fly ash complied with the requirements of standard PN-EN 450-1:2007 “Fly ash for concrete” (category A and N). 2.2.4 Chemical admixture Early tests of concrete were carried with fluidisation admixture (superplasticizers) based on poly carboxyl ethers. It is the admixture designed for the production of precast elements, high early strength concrete and compressed concrete. The use of such superplasiticizer allows for the production of concrete with low w/c ratio, which results in concrete with high strengths, both early ones and in standard times (after 28 days of setting). Concrete with the admixture present the abilities to keep the correct consistency, even during higher external temperatures. The addition of such admixture accelerates the hydration of cement phases, and in consequence, more heat is emitted specially in early stage of setting process. It results in relatively high level of early concrete strength. 100

Table 6. ASCC concrete mixture composition. Component

Amount [kg/m3 ]

Cement CEM I 42,5R Sand 0–2 mm Gravel 2–8 mm Gravel 8–16 mm Fly ash Water Superplasticizer w/s (w/(c+0,4p))

350 611 462 633 120 167 3,85 0,41

Table 7.

Properties of concrete mixture.

Property

Achieved result

Consistency of concrete mixture: measured as flow Air content Speed of mixture flow to 50 cm diameter Segregation of components keeping the consistency in time De-aeration of concrete mixture Temperature of concrete mixture Occurrence of water “throw out”

63 cm 2,5% 7 sec. None 45 minutes Correct 18,8◦ C No

2.3 Composition and properties of concrete mixture The properties of concrete mixture are particularly important during the design of ASCC concrete composition. Concrete mixture should be characterized by: – flow at least 60 cm (invertedAbrahms cone method), which guarantees proper fluidity of mixture, allowing the mixture to flow independently in container or separators form. – keeping the consistency for at least 30 minutes. Essential time required to keep the consistency was estimated basing on the length of technological processes taking place in precast production plant, considering also some safety margin. – It is necessary to state, that the use of concrete with consistency different than estimated leads to improper aeration of mixture, which finally unfavourably influences on the properties of hardened concrete and the aesthetics of produces separator or container. – Flow speed of the cake to diameter 50 cm – 5–10 sec.; lets to keep the right motion of concreting. The composition of designed and analysed concrete mixture is presented in table 6, and its properties in table 7.

2.4 Properties of hardened concrete Once the concrete mixture composition is determined, it is placed in cubic forms with 15 cm leg and subjected to 2–3 seconds of vibrations, pretending lapse agitation of concrete – just as it happens in plant condition during mass production. 101

Table 8. Average compressive strength. Sample

Average strength fcm , [MPa], after

C 35/45 ASCC

1 day 25,3

2 days 40,3

7 days 54,2

28 days 63,1

90 days 74,9

180 days 83,6

Table 9. Water penetration depth under pressure. Sample indication

Water penetration depth [mm]

C 35/45 ASCC

11,0 11,0 9,0

Table 10. Water tightness test results.

Sample indication C 35/45 ASCC

Depth of water penetration after water tightness tests for W8 degree [mm] 32,0 36,0 35,0

Once the samples had matured in laboratory conditions (temperature 20◦ C ± 2◦ C, relative humidity 95% ± 5%) they were subjected to tests in order to confirm quality parameters achieved by concrete with the designed ones. The following scope of tests were carried with hardened concrete: – Compressive strength after 1, 2, 7, 28, 90 and 180 days defined by the procedures of standard PN-EN 12390-3 – Frost resistance F150 wg PN-B-06250 “Normal concrete”; determination of concrete frost resistance began for samples after 28 and 90 days of hardening – Depth of water penetration under pressure acc. to PN-EN PN-EN 12390-8 “Tests for concrete. Part 9. Depth pf water penetration under pressure” – Water tightness for W8 acc. to PN-B-06250 – Resistance of concrete to chemical attack of fuels and other light liquids (acc to the procedure given in table 1). Resistance tests results are presented in table 8. Achieved concrete characterized with high density, which was later confirmed by the results of water penetration depth tests (table 9), also by the test results of water tightness for W8 degree (table 10). Samples of hardened concrete (after 28 days of maturing) were subjected to tests determining the level of resistance to aggressive environments in standard PN-EN 858-1. Solutions for the tests were following: – demineralised water with temperature 40◦ C ± 2◦ C – petrol with temperature 23◦ C ± 2◦ C 102

Figure 3.

Concrete samples after 1000 hours of storing in petrol. 120

Strength, MPa

80 60

Figure 4.

83,8

84 77,6

demineralized water 63,6

standard solution

40

unleaded fuel

20

petrol

0

93,4

copy sample

100

28 days

1000 h

Strength of concrete samples hardening over 1000 hours in aggressive solutions.

– unleaded fuel with temperature 23◦ C ± 2◦ C – standard one, with temperature 40◦ C ± 2◦ C, consisting of 90% of water, 0.75% sodium hydroxide, 3.75% sodium orthophosphate, 0.5% sodium silicate, 3.25% sodium carbonate and 1.75% sodium methaphosphate. Samples were subjected to aggressive solutions for 1000 hours. After this period were observed and next put through strength tests. Received strength test results were compared with the strength of samples maturing in 18◦ C water. Surface of tested samples were free of any surface scaling and cracks. Their surface was smooth, without any efflorescence. Received compressive strength level is described on fig. 4. Analyzed concrete fulfilled the requirements of standard PN-EN 858-1 excessively (minimum 45 MPa after exposure in aggressive solutions). Frost resistance tests were carried for F150 degree acc. to PN-88/B-06250 Determination was launched both after 28 days of setting in laboratory conditions (this is the most common term proposed for the tests launch in project documentations) as well after 90 days of setting. Pictorial 103

120

Strength, MPa

100 80

samples after 150 cycles copy-samples 85,1

90,5

88,2

90,6

60 40 20 0 Samples after 28 days

Figure 5.

Samples after 90 days

Frost resistance tests of concrete after 28 and 90 days – decrease of strength. 4 3,5

Mass, kg

3

mass before freezing mass after test

2,5 2

2,376

2,368

2,358

2,355

1,5 1 0,5 0 Samples after 90 days

Samples after 28 days

Figure 6.

Frost resistance tests of concrete after 28 and 90 days – mass loss.

representation of received results is shown on figs. 5 and 6. Analysed concrete fulfilled the requirements for F150 frost resistance degree. The concrete characterized with slight mass loss (0,3% for samples after 28 days and 0,1% for samples after 90 days) and a little drop of compressive strength (6,0% for samples after 28 days of setting and 2,5% after 90 days).

3 CONCLUSION Tests were carried in order to prove the usefulness of concrete as fully valued material for the production of containers and separators resistant to the attack of organic light liquids, in compliance with requirements of PN-EN 858-1 “Installations of light liquid separators. Part 1: design rules, useful properties and research, marking and quality control”. Designed concrete was produced in almost self-compacting concrete (ASCC). Concrete mixture characterized with distinguished with proper fluidity, kept for adequately long time. “Surface water flow out” wasn’t observed on concrete surface. Hardened concrete characterized with high compressive strength (fcm = 63,1 MPa) greatly over the level of 45 MPa. Density of concrete was also considerably higher that specified in project assumptions. Basing on German concrete standard directions, tested concrete may be classified as dense one. Concrete achieved also estimated water tightness level W8 determined as per standard PN-B/8806250. After water tightness test, no stains giving the evidence of humidity in concrete were visible 104

on bottom surface of the concrete. Once the samples were broken, it turned out that the depth of water penetration was not deeper than 36 mm (Table 10). Concrete presented very good strength parameters after exposure in environments (liquids) determined by standard PN-EN 858-1. The increase of strength exceeding over 30% in comparison to the strength after 28 setting days had also been observed (Figure 4). Such a significant increase of strength is related to the increased temperature during tests operation (23◦ C and 40◦ C) and the presence of fly ash in concrete composition. It is commonly known, that temperature increases the process of pozzolana reaction, i.e. Ca(OH)2 by reactive components of fly ash [Giergiczny 2006, Siddique 2008]. In the effect, pores are being filled with the products of pozzolana reaction, which makes the structure of hardened paste more dense, which again on one hand increases compressive strength, however, on the other hand, makes the diffusion and interaction of aggressive ions more difficult (increased resistance to chemical attack [Giergiczny 2006, Siddique 2008]). Analysed concrete fulfilled the requirements for frost resistant concrete after 150 cycles of freezing and defrosting (frost resistance degree F150 acc. to PN-B-06250). The drop of compressive strength was no higher than 6,0%, that is considerably lower than determined by standard requirements (≤20%). It was stated, that this drop is lower for concrete after longer period of early setting (Figures 5–6). Basing on carried tests we may declare that the use of concrete in the production of separator for light liquids and containers for fuels in fully justified. Proper quality concrete is possible to be made in ASCC technology providing the concrete components are good quality and exact following technological discipline on ready-mixed plant and during the production of concrete goods. REFERENCES Beton według normy PN-EN 206-1 – komentarz. Praca zbiorowa pod kierunkiem prof. Lecha Czarneckiego. Polski Komitet Normalizacyjny – Polski Cement. Kraków 2004. s. 298. Brandt A.M. Uwagi o trwało´sci konstrukcji betonowych. Drogi i mosty, nr 3, 2004. s.5–14. DIN 1045-2: Tragwerke aus Beton, Stahlbeton und Spannbeteton. Teil 2: Beton – Festlegung, Eigenschaften, Herstellung und Konformität. Anwendungsregeln zu DIN EN 206-1 (Ausgabe: Juli 2001). Domone P.L. “Self-compacting concrete: An analysis of 11 years of case studier” Cement & Concrete Composites 28 (2006), pp. 197–208. ˛ wpływom czynników s´rodowiskowych. Fiertak M., Małolepszy J.: Beton materiał kompozytowy podlegajacy Sympozjum naukowo-techniczne “Trwało´sc betonu i jej uwarunkowania technologiczne, materiałowe i s´rodowiskowe” Kraków 2004, s.5–39. ´ Giergiczny Z., Małolepszy J., Szwabowski J., Sliwi´ nski J.: “Cementy z dodatkami mineralnymi w technologii ´ aski ˛ Sp. z o.o. w Opolu, Opole 2002. betonów nowej generacji”, Wydawnictwo Instytut Sl Giergiczny Z. “Rola popiołów lotnych wapniowych i krzemionkowych w kształtowaniu wła´sciwo´sci współczesnych spoiw budowlanych i tworzyw cementowych”. Monografia 325, Politechnika Krakowska, Kraków 2006. Giergiczny Z., Nocu´n-Wczelik W. “Rola cementu w kształtowaniu wła´sciwo´sci betonu”, materiały Konferencji “Beton cementowy w obiektach hydrotechnicznych”, Góra˙zd˙ze 2006. Madryas C., Kolonko A., Wysocki L. “Konstrukcje przewodów kanalizacyjnych”, Wrocław 2002. Neville A. M., “Wła´sciwo´sci betonu”, Polski Cement, Kraków 2000. Siddique R.: Waste Materials and By-Products in Concrete. Springer, 2008.

105

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Fly ash as a component of concrete containing slag cements Z. Giergiczny & T. Pu˙zak Góra˙zd˙ze Cement S.A, Chorula, Poland

ABSTRACT: Fly ash is common use (type II) mineral additive in concrete composition. According to European standards fly ash can be used in concrete produced with Portland cement CEM I or composite cement CEM II/A (excluding fly ash Portland cement CEM II/A-V). In the paper authors describe the use of fly ash for concrete with the following types of cements: Slag Portland cement CEM II/B-S 32,5R and Slag cement CEM III/A 32,5N – LH/HSR/NA. Ready mixed concrete as well as hardened concrete based on these cements and with addition of fly ash were characterized by positive properties – water absorption, consistency and compressive strength.

1 INTRODUCTION Modern concrete is composite material, where apart of traditional components as cement, aggregate and water, chemical admixtures and mineral additives are also present. There are fully valued concrete components, which in significant way influence the properties both, concrete mixture and hardened concrete (Lindon 2001, Siddique 2008, Giergiczny 2006). One of the most common additives to concrete composition is siliceous fly ash. They are by-products of dust coal combustion in electric power stations and in thermal electric power stations. They are valuable and desirable raw material for building industry, specially for cement and concrete producers. It has its reflection in current standards, which determine quality parameters for fly ash as main component of cement (PN-EN 197-1:2000) or additive to concrete (PN-EN 450-1:2007). The regulations of fly ash use in concrete composition are consisted in standard PN-EN 206-1 and national appendix to this standard (PN-B-06265:2004). Domestic resolutions (PN-B-06265:2004) admit the possibility of fly ash use as the addition of concrete containing Portland cement CEM I or composite Portland cement CEM II/A (excluding fly ash Portland cement CEM II/A-V). Some of European countries (CEN TC 104/SCI) allow for wider range of cement assortment, which fly ash can be produced with, it mainly concerns Portland slag cements CEM II/B-S and slag cements CEM III (Belgium, Czech Republic, Germany, Italy, Luxembourg, Holland, Slovakia). Alike the composition of composite cements CEV/A,B allows for the mixture of slag and fly ash, with slag contents higher than 20% (PN-EN 197-1:2002). Increased resistance of concrete to aggressive chemical attack is also an advantage of such solution (Łagosz 2008). Wider and proper use of fly ash in concrete technology inscribes in sustainable development technology, for it enables for the optimization of cement use (decrease of CO2 emission, decrease of natural raw materials use in the production of cement clinker), enables to save deposits of natural raw materials and limits hazardous influence on the environment by the limitation of surface for the deposits of by-products of industrial processes, which is fly ash. The subject of author’s study is the evaluation of the influence of siliceous fly ash addition on the formation of concrete mixture properties as well as hardened concrete. The concrete was produced basing on cements containing granulated blast furnace slag (slag portland cement CEM II/B_S 32,5R and slag cement CEM III/A 32,5N-LH/HSR/NA). 107

Table 1.

Chemical composition of fly ash. Content, % of mass

Ash type

SiO2

Al2 O3

Fe2 O3

CaO

MgO

Na2 O

K2 O

Siliceous

51,5

27,8

7,5

3,7

2,5

1,1

3,0

Table 2.

Properties of fly ash.

Density [g/cm3 ] 2,13

Ash type

Loss of ignition [%]

SO3 [%]

CaOfree [%]

Cl− [%]

After 28 days

After 90 days

Fineness, sieve 45 µm [%]

Siliceous

2,2

0,7

0,07

0,01

78,4

93,2

34,0

Figure 1.

Pozzolana activity [%]

Siliceous fly ash used in the tests – a) enlargement – times 3000; b) enlargement – times 8000.

2 CHARACTERISTIC OF MATERIALS Fly ash applied as mineral additive of II type must comply with the requirements of standard PN-EN 450-1:2007. Table 1 presents chemical composition of fly ash used in the study, whereas table 2 illustrates the fly ash properties essential for the evaluation according with the requirements of standard PN-EN 450-1:2007. Figure 1 presents the average structure of fly ash grains. Table 3 shows basic physical and mechanical properties of cements, which gave the grounds for concrete mixture preparation. The influence on water demand (determination of w/c ratio, qualitative and quantitative selection of chemical admixtures, the amount mineral additive in concrete composition), considering the formation of concrete mixture properties, is a relevant property of fly ash (Giergiczny & Pu˙zak 2008). Designation of water demand of fly ash used in the study was carried according to procedure given in the “Appendix B” to standard PN-EN 450-1:2007. Measure of water demand, as per applied procedure, is the flow of paste prepared on cement without and with the addition of fly ash. The designation was made for all cement types applied in the tests. The results of carried indications are listed in Table 4. Concrete mixtures were prepared with washed sand of 2mm granulation and gravel aggregates of fractions 2 ÷ 8 mm and 8 ÷ 16 mm. In order to achieve the consistency similar for all concrete mixtures, the addition of new generation superplasticizer, based on polycarboxylethers. 108

Table 3.

Physical and mechanical properties of cements.

Property

CEM I 42,5R

CEM II/B-S 32,5R

CEM III/A 32,5N

Blaine’s surface, cm2 /g Le Chatelier; mm Water demand, % Beginning of setting time; min. Compressive strength after 2 days; MPa Compressive strength after 28 days; MPa

3400 0,2 27,0 170 25,2 50,5

3400 0,6 28,7 216 18,1 49,8

3800 0,5 29,7 253 9,8 44,4

Table 4. Water demand of mortars with fly ash addition.

Table 5.

Binder type

Flow [mm]

Water amount [ml]

CEM I 42,5R 70 % CEM I 42,5R + 30 % siliceous fly ash CEM II/B-S 32,5R CEM II/B-S 32,5R + 30 % siliceous fly ash CEM III/A 32,5N CEM III/A 32,5N + 30 % siliceous fly ash

175 182 163 170 160 160

225 225 225 225 225 225

Concrete mixture composition. Amount of component, [kg/m3 ]

Mixture symbol

“k” value

Cement

Fly ash

Sand

Gravel 2–8

Gravel 8–16

SP

Water

CEM I CEM I/20-1 CEM I/20-2 CEM I/33-1 CEM I/33-2 C II CEM II/20-1 CEM II/20-2 CEM II/33-1 CEM II/33-2 CEM III CEM III/20-1 CEM III/20-2 CEM III/33-1 CEM III/33-2

– 0,2 0,4 0,2 0,4 – 0,2 0,4 0,2 0,4 – 0,2 0,4 0,2 0,4

320 308 296 300 283 320 308 296 300 283 320 308 296 300 283

– 62 59 99 93 – 62 59 99 93 – 62 59 99 93

699 676 681 663 671 699 676 681 663 671 699 676 681 663 671

612 592 596 580 587 612 592 596 580 587 612 592 596 580 587

641 620 625 608 615 641 620 625 608 615 641 620 625 608 615

3,6 2,7 2,6 2,6 2,6 3,3 3,2 3,3 3,1 3,2 4,0 3,5 3,5 3,0 3,1

160 160 160 160 160 160 160 160 160 160 160 160 160 160 160

SP-superplasticizer; CEM I- CEM I 42,5R;CEM II- CEM II/B-32,5R; CEM III-CEM III/A 32,5N-LH/ HSR/NA.

3 INFLUENCE OF FLY ASH ADDITION ON CONCRETE MIXTURE PROPERTIES The composition of tested concrete mixtures is shown in table 5. Fly ash was being added to concrete content in amounts of 20 and 33% in relation to cement mass. The amount of binder in individual mixtures was calculated in accordance with regulations specified in standard PN-EN 206-1 considering “k” value of 0,2 and 0,4 (s = c + k · p [kg]; where s – amount of binder in kg; c – amount of cement in kg; p – amount of fly ash in kg).Water-binder ratio (w/s) for all tested 109

Figure 2.

Table 6.

Concrete mixture properties.

Mixture symbol

Slump after pasting t0 [cm]

Slump after pasting 45 minutes t45 [cm]

Air content [%]

Mixture temperature [◦ C]

CEM I CEM I/20-1 CEM I/20-2 CEM I/33-1 CEM I/33-2 CEM II CEM II/20-1 CEM II/20-2 CEM II/33-1 CEM II/33-2 CEM III CEM III/20-1 CEM III/20-2 CEM III/33-1 CEM III/33-2

17 18 18 17 18 18 18 16 17 18 19 18 18 18 17

15 16 15 16 16 16 16 15 17 16 15 15 15 15 15

1,6 1,6 1,7 1,6 1,6 1,9 1,7 1,7 1,7 1,9 1,9 1,9 1,9 1,9 1,7

18,8 18,7 18,0 19,4 18,6 19,1 18,6 19,4 18,8 17,9 19,0 18,0 18,8 17,9 18,9

Slump of concrete mixture a) after pasting to b) after 45 minutes t45 .

concrete mixtures was 0,5. The amount of superplasticizer was assorted in such a way to reach the slump (PN-EN 12350-7) on the level of 15 ÷ 18 cm (pumped concrete). The following parameters were designated in concrete mixtures: • temperature of concrete mixture, • consistency by slump method as per standard PN-EN 12350-2 (after pasting- t0 and after 45 minutes- t45 ), • air content as per procedure contained in standard PN-EN 12350-7. The results of carried designations are presented in table 6. It is visible, that in all cases the consistency of concrete mixture was kept on estimated level (15–18 cm) for 45 minutes. Figure 2 shows the exemplary appearance of concrete mixture after consistency test. 110

Table 7. Average compressive strength fck,cube and concrete water absorbability.

Mixture symbol

Compressive strength after 2 days, [f ck,cube MPa]

Compressive strength after 28 days, [f ck,cube MPa]

Compressive strength after 180 days, [f ck,cube MPa]

Water absorbability [%]

CEM I CEM I/20-1 CEM I/20-2 CEM I/33-1 CEM I/33-2 CEM II CEM II/20-1 CEM II/20-2 CEM II/33-1 CEM II/33-2 CEM III CEM III/20-1 CEM III/20-2 CEM III/33-1 CEM III/33-2

30,4 21,8 28,2 21,8 26,3 21,6 19,0 16,1 18,9 19,1 9,6 8,5 7,8 6,6 6,3

58,7 56,8 57,1 52,7 56,3 54,2 56,3 49,3 54,7 53,6 54,9 53,0 51,7 52,6 47,5

70,1 74,7 75,9 76,2 78,8 66,1 73,2 70,5 70,1 70,4 70,0 76,8 72,1 72,8 70,1

4,4 3,9 3,9 3,6 3,8 4,2 3,3 4,0 3,3 3,4 3,2 3,3 3,8 3,6 3,9

4 THE INFLUENCE OF FLY ASH ADDITION ON THE PROPERTIES OF HARDENED CONCRETE Hardened concrete was subjected to the following scope of tests: • compressive strength after 2, 28 and 180 days according to the procedure contained in the standard PN-EN 12390-3, • water penetration depth under pressure according to the standard PN-EN 12390-8, • water absorbability according to the standard PN-B/88 – 06250, • frost resistance of concrete according to the standard PN-B/88–06250 for frost resistance degree F150. In compliance with the records of the standard PN-B/88–06250 concrete is defined as frost resistant providing, after 150 cycles of frosting/defreezing performed (−18◦ C/+18◦ C), the drops of strength are not higher than 20% and mass loss not higher than 5% in relation to unfrozen samples mass. The results of hardened concrete tests are presented in tables 7 and 8 as well as on fig. 3. 5 DISCUSSION OVER TEST RESULTS It is commonly known that fly ash as a concrete component influences both the formation of the properties of concrete mixture and hardened concrete (Lindon 2001, Siddique 2008, Giergiczny 2006, Neville 2000). Such influence on concrete properties depends on its chemical and phase composition, its amount in concrete content as well as the type of installation it derives from (acquire conditions) (Giergiczny & Pu˙zak 2007). Moreover, the type of cement used for the preparation of concrete with the addition of fly ash effects concrete properties, specially its durability (Łagosz 2008, Wawrze´nczak 2002). Considering analyzed, siliceous fly ash, its implementation into mortars content leads to the decrease of water demand of construction on Portland cement CEM I and slag Portland cement CEM II/B-S 32,5R (table 4; larger diameter of flow). It brings about the possibility of getting the estimated consistency with lower content of make-up water or lower amount of plasticizer. The consistency of tested concrete mixtures with siliceous fly ash was kept stable during test (for 45 minutes) on the level of 15–18 cm of slump. Air content in all mixtures didn’t exceed 2,0% (table 6) 111

Table 8.

Results of concrete frost resistance test results. Compressive strength after 150 cycles

Mixture symbol

Frozen

Copy samples

Drop of strength [%]

Loss of mass [%]

CI CEM I/20-1 CEM I/20-2 CEM I/33-1 CEM I/33-2 C II CEM II/20-1 CEM II/20-2 CEM II/33-1 CEM II/33-2 CEM III CEM III/20-1 CEM III/20-2 CEM III/33-1 CEM III/33-2

58,0 45,5 38,2 Destruction Destruction 55,6 62,7 57,8 58,4 53,8 68,4 68,2 63,0 61,8 51,4

63,6 67,8 67,7 76,5 78,6 62,5 70,4 65,3 68,6 66,9 70,5 72,3 66,8 68,4 67,3

8,8 32,9 43,5 Destruction Destruction 11,0 10,9 11,5 14,9 19,6 3,0 5,7 5,7 9,7 23,6

0,1 0,3 0,4 Destruction Destruction 0,2 0,3 0,4 0,4 0,1 0,1 0,1 0,3 0,4 0,4

Depth of penetration, mm

40

30

20

10

Figure 3.

CEM III/33-2

CEM III/33-1

CEM III/20-2

CEM III/20- 1

CEM III

CEM II/33-2

CEM II/33-1

CEM II/20-2

CEM II/20- 1

CEM II

CEM I/33-2

CEM I/33-1

CEM I/20-2

CEM I/20- 1

CEM I

0

Depth of water penetration under pressure.

and slightly increased in mixtures with slag Portland cement and the addition of fly ash. The analysis pf hardened concrete prove, that the implementation of fly ash decreases concrete early strength (after 2 days; table 7). It results form the pozzolana activity of siliceous fly ash, which is quite low in room temperature and its positive influence on concrete properties is visible only after longer setting time. In practice, the activity of binders containing siliceous fly ash is received by additional grinding, heat treatment and chemical activity (Giergiczny 2006). Analyzed concretes with slag Portland cements and fly ash have strengths in standard time similar to the strengths of concrete without any additives. However, in longer setting time (after 180 days) the strength of concretes containing slag Portland cements and fly ash is higher than concrete with Portland cement CEM I 42,5R (table 7). Specially high dynamics of growth performed concretes on slag cement with the addition of fly ash CEM III/A 32,5N-LH-HSR/NA, which is illustrated on fig. 4. All analyzed concretes, irrespective of the type of cement applied, characterized with high density. The depth of water penetration was under 40 mm, however, the least dense was concrete 112

70,1

72,8

47,5

40

52,6

51,7

50

53

54,9

60

72,1

76,8

70

70

28 days

6,3

6,6

7,8

10

8,5

20

2 days 180 days

30 9,6

Compressive strength, MPa

80

Figure 4.

CEM III/33-2

CEM III/33-1

CEM III/20-2

CEM III/20- 1

CEM III

0

Compressive strength of concrete made of slag cement CEM III/A 32,5N-LH-HSR/NA and the addition of fly ash.

containing Portland cement CEM I 42,5R (Fig. 3). The best result was achieved with the addition of 20 and 33% of fly ash to Portland slag cement CEM II/B-S 32,5R, granting that factor k = 0,2 (Fig. 3; table 5). Examined concretes were also identified by low water absorbability. Besides, concretes produced of slag Portland cement CEM II/B-S 32,5R, and the ones containing slag cement CEM III/A 32,5NLH-HSR/NA, reached lower water absorbability than concrete produced of Portland cement CEM I. The situation may proceeds from the positive impact of the additive, both granulated blast furnace slag and fly ash, on the formation of microstructure of hardened cement paste in concrete (Lindon 2001, Giergiczny 2006). Analyzing frost resistance test results of concrete for F 150 degree, it was stated, that concretes made with Portland cement CEM I and the addition of siliceous fly ash in the amount of 20 and 33%, as well as with slag cement CEM III/A 32,5N-LH-HSR/NA with 33% of siliceous fly ash with k = 0,4 (table 8) did not fulfill the requirements of standard PN-B/88–06250 (PN-88/B-06250). The author of hereby paper (Łagosz 2008) confirms, that concretes produced with slag Portland cements with the addition of fly ash perform lower durability considering frost resistance tests (freezing and defrosting) attending deicing salts. While evaluating the influence of fly ash on the formation of quantitative parameters of concretes with slag cements (CEM II, CEM III), specially those related to durability, it is necessary to consider the temperature affecting the processes of hardening and setting of those composites. In lower temperatures hardening and setting of concretes extends; the increase of compressive strength is slower, especially with larger cement reduction and application of bigger amounts of fly ash in concrete composition. Furthermore, in lower temperatures the influence of pozzolana activity of fly ash on the formation of concrete microstructure is considerably lower, which may effect in lower durability, either chemical and frost corrosion (Wawrzeñczyk 2002, Schneider & Puntke &Sylla & Lipus 2002). As the author’s state, the best way to improve frost resistance of concrete is proper aeration of the mixture (air content on the level 4–6%) and the production of concrete with lower water-binder ratio (plasticizing of concrete by the addition of chemical admixture reducing the water amount in concrete). 6 CONCLUSION Cements with mineral additives, especially the ones with the addition of granulated blast furnace slag (slag Portland cement CEM II/A,B-S, slag cement CEM III) are very popular hydraulical binders, commonly used in many building industry areas. 113

Tests carried by authors confirmed the experience of other countries (CEN TC 104/SCI 2006) concerning the use of fly ash as the additive of concrete made of slag Portland cement CEM II/B-S 32,5R and slag cement CEM III/A 32,5N-LH/HSR/NA. Siliceous fly ash of proper quality, complying with the requirements of standard PN-EN 450-1:2007 for category A, influences the rheological properties of concrete mixture in a positive way (consistency, workability, keeping the consistency in time). Concrete on slag cements with fly ash addition characterizes with high strength in standard time (28 days) and later one (180 days), moreover, resistance of concrete to chemical attack is also increased (Łagosz 2008) which leads the use of such concretes into direction of building object construction in exposure classes from XA1 to XA3 acc. to PN-EN 206-1. Resistance of concrete on slag cements with fly ash addition in lower temperatures is still a problem requiring further studies. Carried tests ought to contain various test methods of frost resistance, as well it is necessary to examine the influence of aeration on the formation of concretes frost resistance. REFERENCES CEN TC 104/SC1 (2006) Survey of national requirements used in construction with EN 2006-1:2000. Giergiczny Z.; Rola popiołów lotnych wapniowych i krzemionkowych w kształtowaniu wła´sciwo´sci współczes˛ nych spoiw budowlanych i tworzyw cementowych. Seria: In˙zynieria Ladowa, Monografia 325, Politechnika Krakowska, Kraków 2006. Giergiczny Z., Puzak T.: Properties of concrete with fluidal fly ash addition. Proceedings of the International Symposium “Non-Traditional Cement & Concrete III” organized by Brno University of Technology & ZPCV, a.s., Uhersky Ostroh, June 10–12, Brno, pp. 263–271. Giergiczny Z., Pu˙zakT; Wpływ rodzaju popiołu lotnego na wła´sciwo´sci mieszanki betonowej. IX Sympozjum “Reologia w technologii betonu”, Gliwice, 2007, s. 5–14. ´ Giergiczny Z., Małolepszy J., Scliwi´ nski J., Szwabowski J.: “Cementy z dodatkami mineralnymi w technologii ´ aski, ˛ betonów nowej generacji”, Instytut Sl Opole, 2002. Lindon K.A. Sear; Properties and use of coal fly ash. a valuable industrial by-product. London. Thomas Telford Ltd, 2001. Łagosz A.: Wpływ popiołu lotnego na trwało´sc´ betonu z cementami z˙ u˙zlowymi. Budownictwo Technologie Architektura, nr 1, 2008, s. 60–65. Neville A.M.: Wła´sciwo´sci betonu. Polski Cement, Kraków 2000. ˛ cementów PN-EN 197-1:2002 “Cement- Cze´ ˛sc´ 1. Skład, wymagania i kryteria zgodno´sci dotyczace powszechnego u˙zytku”. PN-EN 450-1:2007 “Popiół lotny do betonu. Cze´ ˛sc´ 1: Definicje, specyfikacje i kryteria zgodno´sci”. PN-EN 206-1:2003 “Beton. Cze´ ˛sc´ 1: Wymagania, wła´sciwo´sci, produkcja i zgodno´sc´ ”. PN-B-06265:2004 Krajowe uzupełnienia PN-EN 206-1 Beton – Cze´ ˛sc´ 1: Wymagania, wła´sciwo´sci, produkcja i zgodno´sc´ ”. PN-EN 12350-2 “Badania mieszanki betonowej – Badanie konsystencji metoda opadu sto˙zka”. PN-EN 12350- 7 “Badanie mieszanki betonowej – Badanie zawarto´sci powietrza”. PN-EN 12390-3 “Badania betonu. Wytrzymało´sc´ na s´ciskanie próbek do badania”. PN-B/88 – 06250 “Beton zwykły”. PN-EN 12390-8 “Badania betonu. Głeboko´ sc´ penetracji wody pod ci´snieniem”. ˛ Siddique R.; Waste Materials and By-Products in Concrete. Springer-Verlag Berlin Heidelberg, 2008. Schneider M., Puntke S., Sylla H.M., Lipus K.: The influence of cement on the sulphate resistance of mortar and concrete, Cement International, 2002, No. 1, pp. 130–148. Wawrze´nczyk J.: Wpływ dodatku popiołu lotnego na mrozoodporno´sc´ betonu. Konferencja “Dni Betonu”, Polski Cement, 2002, s. 479–488.

114

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Rehabilitation of road culverts on the equator. Implementation of innovative open cut and jacking/relining trenchless solutions Jean Marie Joussin HOBAS France SAS, Osny, France

ABSTRACT: Guyana, a French overseas department, is a vast region of some 83,500 km2 , located on the equator on the edge of the Latin American continent. The department has 466 kilometres of national routes, most of which are located along the coast, linking the country’s major towns – Cayenne, Kourou and Saint Laurent – with the strategic points constituted by the Guiana Space Centre (which will shortly be receiving the Russian Soyuz rocket and the Italian Vega rocket), the port of Dégrad des Cannes and the Rochambeau airport. Main passenger, goods and freight transport travels along the national routes. This road network comprises approximately 800 road-crossing water culverts of all sizes (width × height ranging from 700 × 500 mm to 4000 × 3000 mm). Of these, approximately 300 are metal, the remainder, of reduced section (diameter below 800 mm), being constructed of concrete. Following the collapse of structures in 2004, the Highways French Ministry (DDE) decided to secure the most critical structures. The Rouen CETE Laboratory was asked to produce a detailed list reporting on the state of these structures. This list identified approximately one hundred structures suffering from serious problems and distortion mainly caused by severe climate conditions with 4 m of rain per year, an aggressive environment and heavy road traffic. This led the DDE to consider replacement solutions or in-situ refurbishment and to issue the necessary invitations to tender, allowing bidders a free choice of techniques and products used. After a comparative analysis of bids, the following technical solutions were short-listed from among those submitted by bidding companies: • for deep structures (more than 5 m deep) or on sections with the greatest traffic: pipes replaced with GRP shells (mainly in the shape of tube-like arches) or by jacking GRP non circular panels (either into the ground close to the structures to be replaced, or in situ and in line with the structures dismantled during the works) • for other structures: trenched excavation of damage culverts, replacing them with GRP pipes having a diameter of between 600 and 1700 mm. This operation for which 9,2 M a have already been invested for 111 culverts should be completed by the end of 2008 and has clearly highlighted the benefits of trenchless solutions for safe and quick work which does not affect in particular highway usage. The purpose of this paper consists in describing: • the 3 different trenchless solutions, • the criteria in favour of each, • the design approach in accordance with the French National Project of Research and experimentation program RERAU 4 devoted to prefabricated channel lining systems and especially of structural GRP lining panels. For this project a new design method based on finite element method analysis (F.E.M.) was applied for the largest structures or with severe loads. • the site phases from preparation to completion. Keywords:

Lining, Jacking, GRP, Rehabilitation, Sewer, Design, Finite element analysis 115

1 BACKGROUND OF THE PROJECT In order to find the right solution for the project is essential to consider the local conditions. The geographical situation: French Guiana, a French overseas department with an area of around 83,500 km2 , located below the Equator, is the only part of the European Community edging the South American continent and remaining isolated from the European market. The National Road Network (NRN) covers the littoral from the Brazilian border in the east to that of Surinam in the west for 464 km providing service to over 95% of the population. It connects the big towns, Cayenne, Kourou and Saint-Laurent, and the economical strategic points, Guiana Space Centre, Degrad des Cannes Seaport and Rochambeau Airport. It constitutes the backbone of the road network in Guiana. The latest link that has been established is the bridge over the River Approuague towards Brazil, which was opened at the beginning of 2004. The NRN is currently the basis of the economic activities for Guiana. It assures the connection between the airport and the Guiana Space Centre mainly operated by the European Space Agency (satellite transports), the supply of comestible goods through the Degrad des Cannes Seaport and agricultural products (90% of Guiana’s needs are covered by two Mhong communities of Cacao and Jahouvey), the transportation of 80% of the wood coming from Guiana’s forests and the traffic generated by public construction works that is, a very active sector in Guiana). On RN1, the establishment of 2 new launching pads for Soyuz and the Italian rocket Vega will soon require a reinforcement of the connection between Cayenne and Kourou (high increase in special transports). Besides this, the mining at new quarries at Saint-Laurent leads to an increase in the resulting transportation and will require a reinforcement of existing road structures. On RN 2, the opening of new forest tracks between Regina and Saint-Georges and a giant gold mine together with the increase of traffic caused by tourism and the trading of manufactured products with Brazil after the completion of the bridge over the river Oyapock ask for a road improvement. About half of the population of about 170,000 is less than 20 years old. More generally, Guiana has a vigorous demography close to 3% and consequently an increase in travel frequency (to study, to work, for leisure).

Figures 1 & 2.

Geographical situation of French Guiana.

116

Additionally, the investors must consider that the construction cost in Guiana is 31% higher than in metropolitan France, mainly due to the lack of resources and the isolation of the territory.

2 THE HYDRAULIC HERITAGE IN GUIANA’S NATIONAL ROAD NETWORK The heritage of the culverts over the National Road Network consists of over 850 constructions in all sizes (width × height ranging from 700 × 500 mm to 4000 × 3000 mm) to canalize watercourses. About 380 of these are steel Arval, Armco or Tubosider constructions, or circular or arches mainly located in forest areas and in wet territory. They were installed during the 70s and 80s when paths were converted into roads and are today at the end of their life cycle. In fact, after heavy rainfalls during the rainy season in the years 2000/2004, several road ruptures were caused by the failure of these hydraulic constructions (see Photo 1). The embankment, pavement and culverts needed to be repaired immediately which required an expensive construction of temporary diversions both for water and road traffic which irreversibly destroyed the tropical forest (see Photo 2). The Direction Départementale de l’Equipement (DDE) de la Guyane (Guiana’s Management Department for Infrastructure from the French Ministry of Equipment & Transportation) decided to launch a census of these culverts since many of them were not even registered ([1]). A first census phase was conducted by work subdivisions in 2000 and 2001. Then a first diagnostic study performed in 2002 provided a better overview of this heritage. It seems that a third of the steel constructions (corrugated galvanized steel culverts) and a number of concrete culverts had visible defects (see Photos 3, 4, 5 & 6). A first rehabilitation program resulted from this first investigation and was approved in 2003. Two additional missions awarded to the Centre d’Etudes Techniques de l’Equipement (CETE) de Blois in 2004, the first one in April–May and the second in October, have set up a precise update of the existing census and a detailed assessment of the condition of the different constructions. Nearly half of the culverts suspected of being in poor condition are evaluated as “Grade 3” or “Grade 3U” according to the multi criteria analysis IQOA – Image Qualité des Ouvrages d’Art (Quality Image Works – [1]) – developed by the Ministry, and need to be replaced as soon as possible. This campaign involving 350 culverts has allowed the establishment of a status report on the inspection and situation for each of such. The main problems that appeared in the steel culverts are summarized in the following table 2 ([2]).

Photo 1. Collapse of a metal culvert. Photo 2. Emergency repair view of the temporary diversion.

117

Photos 3 & 4. Views of damaged steel culvert (internal corrosion, broken steel plates displaced, . . .). Photo 5. Damaged steel culvert (holes in the bottom. . .). Photo 6. View of damaged concrete culvert (mortar loss, holes, worn invert, infiltration through joints . . .).

Table 1.

IQOA state assessment coding.

Grade 1

Construction in apparently good order, regularly maintained according to Technical Instruction, October 19, 1979

Grade 2

Construction with structure in apparently good order or with possible minor defects, but needing non-urgent special maintenance

Grade 2E

Construction with structure in apparently good order or with possibly minor defects but urgently requiring a special maintenance in order to prevent any deterioration arising from the defect

Grade 3

Construction with an affected structure requiring non-urgent repair works

Grade 3U

Construction with a highly affected structure, requiring urgent repairs due to support problems or sudden problems may develop which will shortly cause a support problem

3 LOOKING FOR THE MOST COST EFFECTIVE SOLUTION The requirements concerning performance mainly cover the structural integrity and environmental and hydraulic performance aspects. To be more precise, the technical solutions for implementing the works must meet different criteria such as shown in Table 3. Different techniques were evaluated, such as reinforcement by high performance in-situ concrete lining technique or by continuous lining with cured in placed plastic pipe. These solutions were 118

Table 2.

Main problems in steel culverts.

Location of defect

Description of defect

Longitudinal profile

– General non-uniform settling of stagnant water – Localized settling leading to plate distortion and shearing behaviour in the assembly

Cross section (except ends)

– – – –

Structure

– Corrosion with steel loss (see Figure 3) – Abrasion – Steel perforation

Invert

– – – –

Figure 3.

Table 3.

Excessive vertical deformation (>10%) of circular culverts Vertical flattening for «arch» type culverts Horizontal joint breakage Curve inversion of radial invert

Gaps Abrasion Deposits, blockage Water flow passing under the culvert

Remaining steel thickness at the crown of an arch shaped metal culvert.

Considered criteria to find technical solutions.

Criteria

Description

Operation Weather Hydraulics Chemicals Mechanical Surrounding environmental/ geological conditions

Possibility of road diversion High risk of rain The hydraulic capacity needs to be maintained or increased Acidic stream water with a pH value within 3 and 6.5 (due to humic acid) Abrasion resistance to sand Heavy traffic loads (72 ton lorries) Embankment and backfill nature and behaviour Presence of tree stumps

not selected in the end due to the poor condition of the host pipes that could have affected the final construction quality, some restrictive and higher risk installation conditions and finally the expenses. The first tenders were issued to initiate works beginning of 2004. 119

Photos 7 & 8.

Full relining GRP arch shape panel and jacked pipe.

4 IMPLEMENTATION OF REHABILITATION The following aspects were considered by the Ministry’s technicians for the final choice: • • • • • • • •

Mechanical performance (structural integrity) with clear static calculations Compliance with referring standards and certificates Flow capacity Maintenance Jointing Field installation Level of corrosion resistance Abrasion resistance.

Following solutions were chosen from all bids after careful evaluation: • Constructions in over 4 m depth requiring no flow increase and with a circular diameter of DN 700 to 1200: relining with Centrifugally Cast Fiber Glass Reinforced Polyester (CC-GRP) Pipes with flush joints. • Culverts with non-circular arch type profiles and over 1.5 m2 requiring no flow increase: lining with GRP arch shape panels or CC-GRP Pipes (see Photo 7). • Constructions in over 6 m depth that do not allow sliplining: CC-GRP Pipes are jacked close to the existing defective pipe or in line with the old structure that is dismantled at the jacking head (see Photo 8) • Other culverts: damaged culverts are excavated and replaced with CC-GRP Pipes (see Photos 9 & 10). The CC-GRP Open Cut, Jacking and Relining Pipes and GRP panels are specially designed to suit the specific installation and difficult operating conditions of this project and perfectly meet all the previously mentioned requirements. They are designed for an operating life of 50 years and more. Their roughness coefficient is very low compared to existing even sound corrugated culverts, for which the Manning coefficient lies around 40/50 compared to a value of 100 of the proposed GRP products. Since 2004, 111 culverts have been rehabilitated, by both classical open cut replacement and trenchless technology, in order to recover their performance trying to minimize any inconvenience during the installation. The trenchless part accounts to 34.9% of the 9.2 M a launched between 2004 and 2006 (see Table 4). Open cut was applied for depths up to 4 m mainly for reasons of expense with standard 5,65 m CC-GRP Pipes PN 1, SN 10000, DN 800, 1000, 1200, 1400 & 1700 equipped with flush 120

Photos 9 & 10.

Replacement by open trench with standard CC-GRP pipes.

Table 4.

Jobs conducted between 2004 and 2007.

Implemented technique Replacement (open cut) Jacking Circular relining Non circular lining (cross section >1.5 m2 ) Total

Number 96 5 4 6

% of replaced works 86.5% 4.5% 3.6% 5.4%

111

100%

% compared with works cost 65.1% 9.6% 3.9% 21.4% 100%

leak-free FWC couplings. Furthermore, this is a very traditional design and a quite reliable installation method. However, this technique has shown some disadvantages: it disrupts the traffic (the contractor worked on one road lane); it can affect the quality of the laterite embankment particularly in case of high water content; the work can be stopped or even damaged in cases of heavy rain. . . 5 GRP PANELS AND PIPES – TRENCHLESS SOLUTIONS The proposed trenchless solutions with GRP are absolutely non-disruptive and allow an installation without having to divert the flow. CC-GRP Pipes and GRP panels were selected by both, contractor and designer, as the most reliable technical solution. 5.1 GRP jacking and relining pipes Depending upon the depth and the state of repair of existing pipes relining (see Photo 11) or jacking techniques were applied using both CC-GRP Pipes. The CC-GRP Pipes are jointed with flush belland-spigot joints (see Figure 4). They are made of unsaturated polyester resin, quartz sand and glass fibers and are manufactured by computer-controlled centrifugal casting. The following aspects are considered for the job preparation stage ([4]): • Final design. The pipe thickness and stiffness are calculated considering the construction specifications, the geotechnical data (mainly a mix of coarse gravel and tropical laterite which is 121

Photo 11.

Relining of a highly deflected metal culvert.

Figure 4.

GRP pipe for relining and jacking.

• • • •

highly sensible to water content), the operating conditions (traffic load, water table and earth loads) and the material characteristics. A structural analysis is made according to the French static calculation method – as described in the Projet National Microtunnel ([5]) for the jacking sections and RERAU ([4]) code for the relined ones. The surrounding soils with a high risk of presence of tree stumps (this is why the jacking pipe in line with the existing one was the preferred solution) The installation procedure. The grouting for relining and lubrication with bore fluids for jacking. Finishing of inlet and outlet

The jacking job involves CC-GRP Jacking Pipes – DN/OD 1100 and 1229 – installed for one work close to the structures to be replaced or directly in line with the structures that are dismantled during the procedure (see Figure 4). The maximal allowed jacking forces (196 tons for the DN/OD 1100, SN 40000 with 38 mm wall thickness; 229 tons for the DN/OD 1229, SN 32000 with a 40 mm thickness) for the pipes perfectly met the expected jacking stresses (jacking drive up to 38 m) and the jobs were completed in time without major difficulties. 122

Jacking for in-situ replacement of steel construction.

R9

P = 5076mm A = 1.915m2

D

1316

37

1816

R2358

s

60 10–12

25

100 10 40

1400

08

43,49

Figure 5.

58

R2

648,05 1900

Figure 6.

GRP shell cross section and joint.

5.2 The non-circular GRP panels The GRP panels from computer-controlled manufacture are tailor made and consist of unsaturated polyester resin, quartz sand and glass fibres. These special pipes with a leak tight spigot-and-socket jointing systems are designed to fit inside the old defective pipe. The following aspects are considered for the job preparation stage: • The final cross-sectional shape of the panel. The ideal shape is decided on after having checked the inside dimensions of the existing pipe considering the final level and alignment (see Photo 11). • Final design. The panel thickness is calculated taking the panel dimensions, the geotechnical data and the operating conditions (traffic loads, level of water table and earth loads) into account. A structural analysis considers the RERAU code and the material characteristics. • The installation procedure: every 2 meter long panel (to fit the sea container dimensions) is pulled on rails and connected to the next one. • The grouting and strutting. After installation (see Photo 12) the annulus is filled with a cement grout. A temporary wooden internal strutting is required at this stage. • Finishing of inlet and outlet. All the jobs – whatever the dimensions of the panel (from 1260 × 1660 up to 3200 × 2050) and the length (up to 45 m) of the line were – were completed within a few weeks. 123

Checking the internal dimensions of the existing culvert and installation of the panels.

R1 0

24

16,6431˚

R3567

96

R2

1557

Photo 12 & 13.

2469

Figure 7.

GRP lining.

6 RERAU DESIGN PROCEDURE FOR NON CIRCULAR SECTIONS The purpose of RERAU (Rehabilitation of Urban Network Sewers) was to establish a common rational design methodology for a wide range of lining systems. The methodology is a limit state design including partial safety factor on loads and on material properties. For circular linings there are many available design methods – such as the SRM design approach from WRc- but structural calculation of non circular lining is far more complex. That is why RERAU has focused on non-circular lining and also because man-entry sewers are generally non-circular. The design method, based on finite element method analysis (F.E.M.), has been developed specially for large (man-entry), non-circular GRP linings and for special non conventional loads. The lining system is designed to act as a flexible pipe with the old sewer, annulus grout (where appropriate) and soil providing the necessary support to maintain stability. The prime design requirements on the lining are therefore: • ability to sustain the grouting pressure during installation (where appropriate). • ability to sustain the external head of groundwater pressure that must be considered to arise once hydraulic integrity is restored. • and eventually ability to sustain soil loading transfer if the sewer loses its hoop compressive stiffness after lining. 124

Wedges

Props Metal Bar

Figure 8.

Internal vertical strutting.

Photo 14. View of the wooden construction.

Here one of the French Guiana corrugated steel pipe arch buried in a road’s embankment by a horseshoe GRP prefabricated lining. Because of the steel corrosion it was considered that the host steel pipe arch will loose its hoop compressive stiffness after lining. General description of the project: • • • • •

Height of soil above the crown; 12.0 m Height of the table water above the invert: 2,3 m Elastic modulus of soil Es = 10 MPa Specific weight of soil γ = 20 kN/m3 Embedment: mixed grained soils with low fine fraction, relative compaction Dr >95% OPN. 125

2

2

1

Figure 9.

1

Deformation of the panel calculated with F.E.A. (very exaggerated). Water table

2.1m

Figure 10.

Loads due to water table level.

Mechanical characteristics of the GRP lining: • • • • • • • •

Internal height: 1557 mm Internal width: 2469 mm Thickness: 33 mm Radius of the invert section: 3567 mm Mean Perimeter: 6500 mm Young modulus: 11000 MPa Long term modulus: 3700 MPa Bending strength: 60 MPa

6.1 The design for grouting The vertical height of the lining was restrained by an internal strut. The height of the first stage was limited to 0.5 meter. Finite element analysis was used to calculate the deformation of the panel and also the hoop force in the props. 6.2 The design for ground water table The design pressure is equal to the pressure of the groundwater at the invert: pw = 10 · (1.6 + 0.5) = 21 kPa 126

Figure 11.

F.E.M. mesh.

The Buckling pressure of the GRP lining is calculated with the Glock-Thepot’s analytical formula. The parameters used are the perimeter, the radius of the invert section (where the lobe can develop), the thickness and the long-term modulus: t2.2 P0.4 R 1.8 332.2 = 0.455 · 3700 · = 44.4 kPa 65000.4 35671.8

pcr = 0.455 · k 0.4 · EL · pcr

In this case, k = 1 because there is only one lobe (at the invert). The safety factor for the stability is: γF =

44.4 = 2.1 > 2 21

The safety factor for stability security must be greater than 2. 6.3 The design for ground and traffic loading A Finite element analysis was performed to model the transfer of the loading from the soil to the lining and also the traffic loading. The figure 11 shows the finite element mesh. The vertical deflection of the lining and the maximum stresses was calculated. Two conditions must be satisfied: • the vertical deflection must be less than 3% of the lining height; • the maximal bending stress must be less than the bending strength divided by a safety factor of 1.5. 7 CONCLUSION All involved companies and authorities are pleased with the result of the French Guiana National Road Network water culverts rehabilitation and renewal project. The complete project imposed a 127

series of challenges at different stages, all of which were mastered particularly with the different trenchless installation techniques and the GRP tailor made solutions proposed and implemented by the contractor. The Client is pleased with his choice, combining open cut and innovative trenchless techniques, and plans to include them in the upcoming tender to renew the last few culverts that are still evaluated as “Grade 3” and “Grade 3U”. REFERENCES [1] Guide de visite en subdivision, 1996, SETRA, F 9630PV, N◦ ISBN 2-11-085782-X. [2] Catalogue des désordres «buse métallique», 1996, SETRA, F 9641C, N◦ ISBN 2-11-085782-X. [3] Guide pour la surveillance spécialisée, l’entretien et la réparation, 1992, SETRA, N◦ ISBN 2-11-085717-X. [4] Projet National RERAU 4, REhabilitation de Réseaux d’Assainissement Urbain, Restructuration des collecteurs visitables, published in 2002–2004, Editions TEC & DOC [5] Projet National Microtunnel et Forages Dirigés, National Project Horizontal Directional Drilling and Microtunneling, published in 2004, Hermes Science Publishing [6] The Structural Design of Large Non-Circular GRP Prefabricated Linings – No Dig Rome 2008, JM Joussin & O Thépot.

128

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Urban technical infrastructure and city management W. Kaczkowski, K. Burska, H. Goławska & K. Kasprzak Municipality of Wrocław, Wrocław, Poland

ABSTRACT: Efficient city management is aimed at providing the Inhabitants with all conveniences necessary for comfortable life. Efficient functioning of a city is possible only thanks to an efficient urban tissue, which manifests itself in efficient transport, water supply, sewage collection and treatment as well as gas and energy supply. Ensuring efficient waste management and activities for the development of new technologies is crucial for good functioning of the City. It is essential to coordinate urban technical infrastructure development plans with the city’s development. Therefore, participation of the City in determining the directions of development and expansion of infrastructure, organising the underground space by constructing technological channels and zoning underground technical infrastructure is of the utmost importance. Municipal authorities must not forget about coordination of urban systems and cooperation with owners and administrators of networks in their activities. An efficiently functioning city must be ready for operational response to emergency situations and random incidents, and, at the same time, it also must not omit such an important variable as the aesthetic qualities of the City.

1 THE CITY – DEFINITION, FUNCTIONS AND URBAN INFRASTRUCTURE AREAS The term urban planning derives from the Latin word urbs, which stands for a city and is a field of knowledge dealing with planning and management of the development of human settlements. A city is to meet the basic needs in four key areas: • work, i.e. employment providing means of support and production of objects necessary for life, • protection of inhabitants both against external threats (war, crime), and climatic threats (construction of flats and municipal facilities), • recreation, i.e. provision of conditions for passive leisure (parks, pedestrian areas) and active leisure (participation in cultural and sports events), • transport, i.e. enabling inhabitants to use facilities in conditions of growing cities that cover bigger and bigger areas. Mutual interrelations and proportions of areas designed to fulfil these four basic city functions changed in each historical era and differed in form with respect to the level of dependence of social and economic development on natural conditions, regional traditions and other external factors. The definition, nature and functions of cities have changed from their very beginnings to their present-day image. The first human settlements, later on towns and cities, fulfilled defensive and administrative functions and were of a commercial, mining or industrial nature. Wrocław, just as most cities in this part of Europe, was established based on the German foundation charter in the middle of the 13th century. Just like other medieval, merchant cities, it expanded around a central square – the Marketplace, and the symbol of municipal authority – the town hall. Wrocław changed its national status several times in the past millennium, and these changes entailed modifications of a series of conditions related to the city’s existence, turns of directions of the most important forces, the need to adapt to another authority system, to different mechanisms 129

of cooperation and competition. Wrocław has always come out of historical tests unscathed. Even after the almost total destruction of the urban tissue during World War II, it managed to quite quickly build a considerable number of inhabitants and the position of the fourth largest city in the almost 40-million country. At present, Wrocław covers the area of 293 square kilometres inhabited by over 632 thousand people. It has 2,511 streets that are over 1,200 kilometres in length. The municipal water supply system is over 1,840 kilometres long, and the sewage system is 1,170 km long. Today, to ensure that the City fulfils its obligations well and to provide its inhabitants with the sense of living in a friendly city, its functioning and development should take place in all spheres of urban infrastructure including optimum coordination of all of its elements. The role of the manager is to simultaneously organise and adapt the urban infrastructure to the requirements of the city of the 21st century, both “on”, and “under the surface”. The establishment of an efficient communications and transit system of the city, modern public transport combined with urban railway system, a car park system for a very rapidly growing number of private cars is required. Systems of “the underground city,” imperceptible and hidden from the community, determine the condition of urban agglomerations. Water supply, sewage, energy, gas and heating systems, and finally teletechnology, complement the image of the underground urban bloodstream.

2 ENSURING EFFICIENT FUNCTIONING OF THE CITY 2.1 Efficient transport An efficiently functioning City must have a transport system that corresponds to the needs and the number of inhabitants and makes it possible to reduce the cost of living and save more and more precious time. Enabling rescue services to reach to individuals waiting for help quickly as well as efficient elimination of disruption in the functioning of urban infrastructure systems create the sense of security for inhabitants and its friendly image. One of the greatest transport problems concerns difficulties of inhabitants related to their everyday journeys to work, school, for shopping, to the cinema or theatre, etc. The transport system within the Wrocław agglomeration is characterised by a constant increase in individual mobility of its inhabitants, whose intensity is typical for large urban centres, which causes that its traffic capacity turns out to be too low in rush hours (Ernst & Young, 2007). The road system in Wrocław is consistently extended. The spatial policy in the field of the development of motor transport sets not only a general objective, namely improvement of external and internal road links, but also a series of objectives related to the city’s features and needs. The priorities include among other things: • transferring the supralocal transit traffic outside the central part of the city to the considerably developed City Ring Road. • establishing or improving connections between the city districts, relieving the density of traffic in the city centre and improving its transport services, separating economic and residential traffic, • improving connections with individual parts of the metropolitan area. However, investing only in the development of road infrastructure will not ensure full success in the field of improvement of the condition of transport in the city. It is practically impossible to keep modernisation of the road system up with the dynamics of the increase in the number of motor vehicles. In most European cities, it was recognised a long time ago that that the only way out of the situation that ensures efficient functioning of a metropolis is good quality of public transport. There are plans for the next years concerning a very significant expansion of the urban system, among other things, by means of constructing new tram connections, outlining new channels in the city centre area, constructing a dozen or so new or modernised and integrated interchanges, 130

introducing new means of transport with improved parameters, introducing a traffic control system at intersections with a preference for track traffic, introducing the system of dynamic passenger information at stops and in vehicles or implementing integrated electronic tickets in the whole public transport system. Wrocław places particular emphasis on the development of rail transport. The city perceives this form of transport as its chance to reduce road congestion related to individual transport. The planned track transport system will include: • agglomeration rail transport – Wrocławska Kolej Aglomeracyjna (Wrocław Agglomeration Railway), • the system of fast transport on routes linking the opposite ends of the city and crossing its centre called Tram Plus (Tramwaj Plus), • the present, traditional tram system. All track subsystems together with the bus system will constitute one, coherent integrated public transport system of Wrocław. The development of the track system parallels works on streamlining public transport by establishing the Intelligent Transport System. The system is to monitor and control traffic lights, the establishment of a road information system for passengers concerning, among other things, weather conditions, car parks, information boards, video supervision, ensuring priority for public transport, information on the location of public transport vehicles and an information system at bus stops. Wrocław, following the example of other European cities, also plans to expand the public transport system with a network of automatic municipal bicycle rentals. Bicycles are recognised as a means of transport that is an attractive alternative to other ways of moving within a city. The rentals will perfectly complement public transport. All these activities are aimed at improving the transport system, and consequently facilitating the functioning of the whole City. 2.2 Providing the city with basic utilities The basis for the functioning and development of cities is the provision of all necessary utilities. The local government act of March 8, 1990 obliges municipalities to meet collective needs of inhabitants. Own tasks of a municipality concern, among other things, water supply and sewage systems as well as water supply, sewage collection and treatment, maintaining cleanliness and order, sanitary facilities, refuse dumps and utilisation of municipal waste as well as electricity, heat and gas fuel supply. The basic documents for municipalities in the field of energy, heat and gas fuel supply are The Assumptions for electricity, heat and gas fuel supply plans for municipalities adopted under the Resolution of the Wrocław City Council number XXXI/2275/04 of December 9, 2004. A very important element of the implementation of the Assumptions is close cooperation of the municipality with energy companies and ensuring cohesion between enterprise development plans and their energy systems, and the strategy and development plans of the Wrocław agglomeration. 2.2.1 Water supply and sewage collection and treatment A substantial majority of the inhabitants of Wrocław (99.8%) are supplied with water by means of the municipal water supply system. The basic water intake is the Oława River and the Nysa Kłodzka River. The existing system ensures meeting the current and long-term demand for water of the Wrocław agglomeration. There are three water treatment plants in Wrocław. Mokry Dwór water treatment plant is supplied by surface water intakes, Na Grobli plant – by infiltration intakes, and the local water intake in Le´snica is based on underground water. One of the basic problems in water distribution is the poor technical condition of pipelines, among other things, high level of incrustation of the system, which secondarily pollutes treated water. Another problem is the poor technical condition of fittings and elements of pipes in the water 131

supply system, including poor condition of household connections. Some areas of the city lack the ring-shaped distribution system (Dutkiewicz, 2006). The length of the sewage system in Wrocław is 1,170 km. As the city is located on a flat area, the system includes 22 intermediate sewage-pumping stations. The Old Town in Wrocław is equipped with a combined sewage system, post-war housing estates use a mixed sewage system – partially separate, and partially combined. New housing estates located in the city centre already have a separate sewage system. Some housing estates located on the outskirts of Wrocław still do not have a sewage system. Sewage flow to a sewage treatment plant by means of gravity ´ ˛za, Bystrzyca, Kolektor Południowy, drainage systems, which include main collectors, e.g. Odra, Sle Kolektor Północny, as well as an extensive system of housing estate networks and sewage system connections. Currently, sewage flow to one of the three sewage treatment plants: • WOS´ Janówek – a modern, mechanical and biological sewage treatment plant with complete sedimentary management, which treats about 70 thousand m3 of sewage a day. • Pola Irygacyjne Osobowice (Osobowice irrigation fields) – a sewage treatment plant established in the 19th century, based on the process of neutral sewage treatment in the ground. About 45 thousand m3 of sewage a day flow into the fields. • Oczyszczalnia Raty´n – treats a small quantity of sewage from the southwestern part of the city (about 325 m3 a day). It is to be liquidated. Projects aimed at improving water and sewage management and the quality of water are currently being executed in Wrocław. The project entitled Improvement of water and sewage management consists in modernisation of 6 sections of water mains – in total almost 30 kilometres of the most important water supply systems in Wrocław, redevelopment of the Mokry Dwór water treatment plant, expansion of the sewage system, modernisation of the water supply system at Strachocin/Wojnów housing estate and expansion and modernisation of Janówek sewage treatment plant in Wrocław together with the sewage pumping system. All investments are to be completed until the end of 2010. As part of the second key project entitled Improvement of water quality in Wrocław, main ´ ˛za, the second Water Treatment Plant sewage collectors are being constructed: Bystrzyca, Sle Na Grobli is being modernised, the sewage and water supply systems are being extended and modernised, including construction of road surfaces at the following housing estates: Oporów, part of Muchobór Wielki, Stabłowice, Złotniki, Wojszyce, Partynice, Krzyki Południe, Klecina, Ołtaszyn, Brochów-Jagodno. ´ ˛za, has As part of the Project, the construction of main sewage collectors: Bystrzyca and Sle been completed so far. A sewage and water supply system was built together with reconstruction of road surfaces in a part of Muchobór Wielki and Stabłowice. Extension of the sewage system at the following housing estates: Złotniki, Wojszyce, Partynice, Krzyki Południe, Klecina and Oporów, Ołtaszyn, Brochów and Jagodno, is in progress.

2.2.2 Electric energy supply The basic source of electric energy supply for the city is the high-voltage system (110 kV), powered by: GPZ 400/110 kV Pasikurowice, GPZ 220/110 kV Klecina, the heat and power station at Łowiecka Street and the heat and power station in Czechnica. A fragment of the power high´ voltage line (220 kV) from Klecina to Swiebodzice, included in the national energy transmission system, runs through Wrocław. The medium-voltage power system (20 and 10 kV) is powered by the high-voltage systems (110 kV) through 21 main power supply points (110 kV/SN), including three stations located outside the city. 26 local generators are connected to the medium-voltage system, but their share in the electric energy turnover is very little. The medium-voltage system of 20 and 10 kV usually operates in the open-loop system ensuring duplex emergency power supply. 132

2.2.3 Heat supply The heat supply system is centrally based on the heat and power station located at Łowiecka Street, Czechnica heat and power station and the heating plant located at Bierutowska Street that constitute the main sources of heat. Industrial and local boiler houses of public utility institutions, commercial facilities, housing estate boiler rooms and boiler rooms of multi-family residential buildings, which generate heat for their own needs, also function in Wrocław. Heat and power stations supply the municipal heating system in about 98% and meet the current exhaust emission standards. 2.2.4 Gas supply Wrocław is supplied with high-methane natural gas, which is supplied to the City through the circumferential system of high-pressure as well as high- and medium-pressure gas pipelines. Nearly the whole gas pipeline ring network of Wrocław runs outside the city limits and is connected to the main gas pipelines, which radiate to other regions as well as to radial gas pipelines. The existing gas supply solution is recognised as very good. About 97% of the city inhabitants are within the scope of availability of the gas supply system. 2.3 Waste management Wrocław does not have a municipal solid waste management company. Furthermore, there are no municipal systems related to waste management, which causes that the key and practically the only operators of the system are entrepreneurs conducting business activity in this field. As part of this activity, contracts for the provision of adequate waste collection containers to properties, collection of waste and recycling or utilisation of waste are concluded. Some activities aimed at achieving adequate standards within the City area were initiated in 1998. Thanks to the efforts of the then Waste Management Board, the local system of selective glass and plastic packaging waste collection was organised and launched. Similar activities concerned the implementation of the campaign entitled Clean Housing Estates and the programme Expired medicine, which enabled the inhabitants to legally dispose of problematic large-size and dangerous waste. In 2008, the first municipal solid waste sorting plant started to operate. Its functioning will enable the City to achieve maximum levels of biodegradation of municipal waste to be disposed at landfill sites. The project entitled Solid waste management in Wrocław is currently being implemented. The development of a safe and modern biodegradable waste management system will contribute to the improvement in the environment. As a result, the project will prevent potential ecological damage in the environment related to waste management. 2.4 Teletechno and the Internet To enhance and modernise the image of Wrocław as the city supporting modern technologies, free access to wireless Internet was enabled at he Marketplace in Wrocław. The service is provided by the City and is addressed to tourists, inhabitants and businesspeople staying near the Marketplace. The transmission capability of access points was selected so that the signal covered the largest area possible. Access to the Internet was limited with respect to the possibility of using peer-to-peer programs and auction services as these forms of access were abused by anonymous fraudsters. Wrocław assumes gradual development of IT functionalities, technical infrastructure and systems as the means of providing an unlimited scope of services. The result of such an attitude is the development of the e–Wrocław strategy, in which informatisation was extended to ensure comprehensive and extensive operation and integration of all services and sources of information, which is to lead to the establishment of the so-called Smart Communities. The product of the e–Wrocław strategy is the development of an electronic office that will be not only transparent and efficient, but, above all, will serve citizens. This type of organisation may be established only thanks to full integration and cohesion of the basic links of internal IT systems in the local government, electronic circulation of documents, the geographic registry system and the Internet portal. Full integration with other public utility units in the field of combining the communications infrastructure and 133

the establishment of a broadband access infrastructure for the inhabitants is extremely important (Hanys, 2004). The above-mentioned activities are related to the need to adjust the management system to the rapidly developing IT systems. 3 URBAN TECHNICAL INFRASTRUCTURE AND CITY MANAGEMENT 3.1 Coordination of urban technical infrastructure development plans with the city’s development. Determination of development directions and extension of infrastructure enabling the City’s development The technical infrastructure of Wrocław leaves a lot to be desired. Part of the problems results from past negligence and some other reasons – the growth and transformation processes. It is impossible neither to improve the standard of life of inhabitants, nor the city’s attractiveness without extensive efforts in the field of modernisation and development of these systems. To do this, we need to ensure constant and satisfactory provision of all necessary utilities to the city and consistently develop their supply and distribution networks. The modernised city parts are expected to be characterised by high comfort of use, ecofriendliness, safety and flexibility. To meet these requirements and protect the inhabitants against nuisance related to excavated streets, it is necessary to coordinate development and renovation plans of individual sections of the underground infrastructure. In the previous century, when all networks supplying the inhabitants with utilities were state-owned and concentrated at one place, it was easier to manage development and operation in the command or task-oriented form. At present, in the age of transformation and competition, the establishment of directions of network development in accordance with the directions of the city’s development has become very difficult. Realising the importance of the problem, activities aimed at developing a permanent platform of cooperation with technical infrastructure administrators in the field of coordination of development and renovation plans of all sectors have already been taken. 3.2 Organising the underground city space by means of building technological channels and zoning of the underground technical infrastructure The element used to organise the location of networks in the road cross section is zoning of underground fittings consisting in observing the established distances between underground wires and pipes required by legal regulations. Under the Resolution of the Mayor of Wrocław no. 1749/07 of September 17, 2007, the rules and the method of distribution of the Catalogue of street cross sections together with zoning of the underground technical infrastructure and road concepts for streets outlined in local spatial development plans were introduced. Road concepts supplement the adopted local spatial development plan and are the basis for project studies related to investment processes, whereas the Catalogue of street cross sections together with zoning of the underground technical infrastructure forms the basis for designing streets that do not require concept development. Both the Catalogue, and the concepts are provided to designers by Network investors (MPWiK, EnergiaPro, Dolno´sla˛ski Operator Systemu Dystrybucyjnego, Fortum, Telekomunikacja, etc.) and at individual stages of assessing and developing design documentation (Department of Architecture and Construction of the Municipality, Design Documentation Coordination Team, the Road and City Management Authority (ZDiUM), Wrocław Development Office, Department of Urban Engineering of the Municipality etc.). The catalogue of typical street cross sections together with zoning of the underground technical infrastructure is also available on the Internet sites of the Municipality of Wrocław: www.um.wroc.pl A significant problem in the field of zoning of underground fittings is the constantly growing number of networks, which have to be located in the road cross section. Construction of multilayer technological channels may be a good solution in this situation. An experimental section of a multi-layer tunnel, which was to include water supply, gas, heat ducts as well as energy 134

and telecommunications lines, was built at the Gaj housing estate in Wrocław already in the 1970s. Constant monitoring of the operation of the tunnel was to enable the development of a technical and economic analysis of the functionality of the construction of these facilities as the method of land fitting at the national scale. Only the first stage of the investment was completed as part of the whole project, i.e. 370 m of tunnel, which did not provide an objective picture of profitability of the investment. Low level of precision of the tunnel (during the first year of operation, the tunnel was repeatedly flooded) and difficulties in operation (transport of pipes during renovation works, collisions of location of the gas pipeline with renovation works performed using electrical or gas welding) consequently resulted in resignation from the gas pipeline and later on removal of hot and cold water pipes from the tunnel. This was practically the end of operation of the multi-layer tunnel – at present, only low-voltage and internal lighting system wires run through the tunnel, and the basic problem of the housing estate administration is to ensure adequate protection of the tunnel against access of unauthorised people – mainly the homeless. The example above damages the image of multi-layer tunnels and does not encourage to their broader application. At present, the possibility of constructing technological channels for telecommunications lines is being analysed – the act on public roads of March 21, 1985, art. 39 sec. 6 provides that during construction of roads, a road administrator may locate a technological channel in a road lane to place technical infrastructure fittings not related to road management or traffic needs. Pursuant to art. 39 sec. 7 of the act, a road administrator provides interested entities appointed by way of a tender with access to such channels. Problems arise with respect to roads that are subject to redevelopment. During redevelopment, a road administrator may also locate technological channels in a road lane according to the same rules as it is in the case of newly built roads. However, in this case, a road administrator has to solve the problem related to fittings existing in road lanes whose vast majority was placed there before 2003. Such fittings are subject to art. 10 of the act amending the act on public roads, which obliges a road administrator to rearrange fittings and incur the costs related to such rearrangement. In this situation, a road administrator must be authorised to determine new terms and conditions for location of fittings in a road lane. Therefore, popularisation of the use of technological channels must be preceded by providing expert and legal opinions establishing the direction of further activities.

3.3 Current management and coordination of the functioning of urban Infrastructure systems. Permanent platform of cooperation with network owners and administrators. Assessment, transfer of information, identification of disruption and corrective action The basis of management is information on what is at one’s disposal. In the case of urban organisms, there is a great diversity of facilities. Individual elements are interrelated in such a way that any changes in one field may cause unexpected results in other areas. Most cities have extensive databases concerning individual facilities, which, however, as a rule, do not take into consideration mutual relations between different types of facilities. Nowadays, lack of information is not a problem. What city managing authorities really need is a clear method of presentation of information, which will enable simultaneous placing of information on various facilities. ICT infrastructure is indispensable both to ensure efficient city management, and a suitable level of standard of living for its inhabitants. The infrastructure of digital systems and tools supporting the city functioning overlaps the existing, traditional layers of the urban tissue, which enables their further, efficient operation and development. Remote traffic management systems, monitoring systems using camcorders – these are the most obvious, spectacular examples of the application of new technologies, which support current solutions and methods of work of municipal services. For instance, telemetric systems to measure important parameters of municipal water supply or gas systems or systems supporting the management and coordination of the activity of municipal units and institutions are less visible, but not less important. 135

Wrocław is becoming a more and more complex organism. If it is to develop, it has to be managed in an adaptive way responding to the constantly changing challenges of the global and local reality. This gives rise to many new, previously unknown problems in the field of management and planning. Integration of municipal information systems is becoming inevitable. Higher and higher complexity and dynamics of various structures and processes brings about the need to comprehensively embrace, consolidate and coordinate the growing number of spheres of urban life that have been functioning separately so far. Ensuring efficient functioning of the entire Wrocław requires access to a greater and greater amount of current, reliable, cross-sectional and interrelated information. Municipal administration and public services have to be able to constantly monitor their activities, constantly share information on the situation in areas of urban life handled by them. Only then will they be able to respond to different occurrences in a proper and efficient way. Today, the Geographical Informational System GIS seems to be the most natural integration outline for circulation and sharing of information on the situation in the city among various institutions and services operating in the city. Informatisation includes the basic functions of City management and all municipal organisational units. The most significant activity in the field of coordination of the functioning of urban infrastructure systems is integration of the GIS system with the principal map of the board of geodesy and cadastre in the online mode. Informatisation will also include the elements of infrastructure that are not owned by the city, but are the instrument of city management as a whole. For instance, as part of “The project of assumptions for electricity, heat and gas fuel supply plans for the Municipality of Wrocław,” a municipal database of the most important systems and fittings necessary to supply gas, electricity and heat is being developed. The system is to be integrated with the digital city map and maintained in cooperation with energy companies. Municipal companies Miejskie Przedsie˛biorstwo Wodocia˛gów i Kanalizacji (Wrocław Water Supply and Sewerage Company), Miejskie Przedsie˛biorstwo Komunikacyjne (Municipal Transport Company) and Zarza˛d Dróg i Utrzymania Miasta (the Road and City Maintenance Authority), Zarza˛d Zieleni Miejskiej (the Urban Greenery Administration), Zarza˛d Cmentarzy Komunalnych (Municipal Cemetery Administration) will develop specialised databases concerning a selected technical urban infrastructure within one IT platform. One of projects aimed at increasing the efficiency of activities of municipal services was the development of the Operation Plan. While Wrocław has developed other editions of the Long-Term Investment Plan for many years, containing hints for all entities operating in the city in the field of large investments, there was a lack of information on operational activities, which is of the utmost importance for efficient functioning of the City. The development of the Long-Term Investment Plan was aimed, among other things, at providing all interested parties with access to information on the scope of works planned for a given year and its coordination already at the stage of development of the document. It is planned to continue the development of the Operation Plan, and the possibility of developing a Long-term Operation Plan is also considered. These documents are to facilitate the planning process with respect to all entities operating within the area of Wrocław. Due to the need to improve the transport situation, the so-called Intelligent Transport Systems (ITS) will be implemented in Wrocław. The ITS will include tasks related to: • • • •

extension of the traffic management system, extension of the traffic control systems, extension of the system of detecting overspeeding for the police, monitoring of environmental conditions around roads, development of the warning and informing system in the form of indicator boards presenting data to drivers, • information radio and mobile transmission systems, development of a network of cameras monitoring: main junctions, routes for transport of dangerous materials etc., • development of an electronic public transport information system, • development of a digital plan of Wrocław to be used in GPS systems. 136

Other tasks will concern the integrated municipal card of Wrocław, whose functionalities will enable users to pay for car parks and season tickets in public transport. As far as public transport management is concerned, development of the local communication system (TETRA) and its integration with location systems will be necessary. As for public safety, the city’s activities are focused on the extension of the TETRA communications system, connecting intervention services to the municipal ICT network, development of a system monitoring: the water level, water and air pollution, development of a system of cameras to monitor schools, housing estates etc., connecting smoke detectors to the alarm of the fire brigade, implementation of an online system warning against breakdowns, various threats etc. (Wie˛ckowski, Najnigier, 2004). 3.4 Operational responding to emergency situations and random incidents in the field of infrastructure Two departments responsible for operational responding to emergency situations and random incidents function within the structures of the Municipality. The Security and Crisis Management Department performs tasks in the field of: • identification of natural and technical threats as well as development of programs preventing their appearance, • management of the Threat Identification and Alarming System, the Early Warning System in the field of monitoring and alarming, • coordination of rescue, maintenance and protective activities, execution of tasks in the field of civil defence and population protection, • management of crisis response and civil defence tasks executed by business entities, public institutions and other organisations. The Department of Urban Engineering performs activities aimed at systematising and developing a common code of conduct for organisations operating within the City in case of emergency situations and random incidents in the field of infrastructure. It is a very complicated process, as part of organisations responsible for the city’s infrastructure is not directly subordinate to the City. Accomplishment of objectives in the field of supplying the city with e.g. heat, electricity and gas fuel, requires searching for a compromise between the possibilities of the municipality and the local energy market with respect to the execution of the established objectives and obtaining approval for their execution from all entities operating on the local energy market. 3.5 Influence of technical infrastructure on the city’s aesthetic qualities The essential elements of technical infrastructure, which nowadays are at the disposal of all Cities, have a significant influence on their aesthetic qualities. In accordance with the act on spatial planning, the notion of technical infrastructure is to be understood as organisation or modernisation of roads and construction of water supply, sewer, heating, electric, gas and telecommunications systems or fittings under, on or above the ground. If elements of small-scale urban architecture such as: benches, bus and tram shelters, poster pillars, notice boards, telephone booths, fencing, power distribution unit etc., are designed well and made of high quality materials, they significantly influence the general appearance of a city. Uniformity and harmony of the adopted designs of municipal furniture is also very important. A catalogue of municipal furniture aimed at systemising its style and character depending on its purpose and location is being developed in Wrocław. While public transport stops should be identifiable in the area of the whole city and their style should be basically uniform, the appearance of, for instance, waste bins or bicycle racks, depends on the city area and the risk of possible damage. Elements of technical infrastructure are subject to industrial standards and their appearance sporadically has a positive influence on aesthetic qualities. It is of less importance in more modern city areas, where such elements as: power distribution units etc. are less visible among other fittings, e.g. modern-style municipal furniture, however, technical infrastructure in the historic part of the 137

City constitutes a challenge to people wanting to emphasise the historic character and aesthetic qualities of such places. Moreover, elements of this type are often used to put advertisements and announcements. One of activities taken to prevent such practices is the proposal to introduce poster pillars for inhabitants to place their small ads and information for free. From the point of view of aesthetic qualities, it is extremely important to plan necessary renovations and consult time limits of works that lie within the competence of the City with owners of networks. For instance, at the beginning of this year, a meeting of representatives of the City and owners of networks was held at Ostrow Tumski in order to coordinate renovation works. Owners of networks were obliged to check the technical condition of their property and provide information on time limits of possible renovations. Such coordination of works in the field of individual City areas will enable more effective utilisation of resources of all entities operating within the City area in the future. Another activity of the city aimed at improving the aesthetic qualities of its technical infrastructure is to develop a uniform system of road signs. This type of activity supports both the improvement of safety in the City, and improvement of the level of identification of municipal road signs, as well as of aesthetic qualities, e.g. by means of introducing a higher standard of temporary road signs. These assumptions determined the terms and conditions of a tender for the Municipal Information System. The Specification of the Essential Terms and Conditions of the Order precisely determines the selection of materials to be used to manufacture information boards and the supporting structures on which they are to be installed. The form, dimensions and colour scheme of road signs were specified. The Specification also included recommendations concerning the manufacturing technology and guidelines concerning transport and installation of information boards in order to ensure the highest quality and clarity of products to support, among other things, traffic safety. The whole plan was aimed at establishing a uniform system, which – apart from fulfilling its information function – will guarantee safety and aesthetic qualities. The Municipal Information System is ultimately also to include all tourist information signs currently existing in Wrocław. Clear and readable street signs and new designs of address plates at buildings will make it easier for both the inhabitants, and visitors to move around the city. It will also shorten the time necessary for rescue services and the police to reach a given address. Guiding boards for road traffic will supplement the existing road sign system. The Municipal Information System is to be a convenient source of information on public utility facilities, monuments, cultural facilities and access roads e.g. to offices. Lampposts are equally important elements of technical infrastructure. Most of them are managed by municipal units. In this case, attention to aesthetic qualities and the matter of selection of adequate “models” of lamps matching the style of a specific City area is much less complicated. The matter of aesthetic qualities of elements of infrastructure not managed by the city requires cooperation with numerous entities and, unfortunately, there still is much to be done in this area. The City provides illuminations of buildings attractive to tourists together with various entities. The first illuminations started to appear in Wrocław 14 years ago, when a fragment of the Marketplace and the University building was illuminated. Then, at the end of 1996 and at the beginning of 1997, light arrangements appeared at Solny Square and Ostrow Tumski. Since then, other sites gain new appearance each year. In 2004, Wrocław organised a competition for the concept of illumination of buildings located along the Odra River. The competition winner developed a concept of illuminations of selected facilities in Wrocław. The concept includes illuminations of 38 facilities, including bridges, islands, boulevards and selected buildings located by the Odra River.

4 DEPARTMENT OF URBAN ENGINEERING In 2007, as a result of reorganisation of the Municipality of Wrocław, the Investment and Technical Department (WIT) was transformed into the Department of Urban Engineering. The Department took over the duties of WIT and extended its duties with tasks of the City Engineer. The Department supervises the operation and proper functioning of Wrocław in its key areas and monitors the system 138

of collecting information concerning the functioning of the City. Another important task performed by the Department is coordination of the activity of administrators of all technical infrastructure sectors, including cooperation in assessing the reliability of services in the field of providing the city with electricity, heat, gas, water, sewage collection, waste management. The Department is also responsible for assessment of renovation plans in individual sectors, coordination of operating, investment and renovation works, improvement of safety of Wrocław with respect to flood hazard, supervision over operational responding in emergency situations and random incidents in the field of urban infrastructure. In 2007, the Mayor of Wrocław established the function of the City Engineer. A very significant task of the Engineer is coordination of development and infrastructure modernisation plans with respect to individual sectors and cooperation with all existing services responsible for emergency and crisis situations in the City. The establishment of this type of position was aimed at ensuring more efficient City management as part of its resources and possibilities. REFERENCES Dutkiewicz R. et al. 2006, Study of Local Factors and Directions of Spatial Development of Wrocław, Resolution No. LIV/3249/06 of the Wrocław City Council of July 6, 2006, Official Journal of the Wrocław City Council of July 24, 2006 No. 8, item 253. Ernst & Young, 2007, General track development plan in Wrocław, Part I: General concept of tract transport in Wrocław, Extended version, Warszawa. Hanys R., 2004, “E-Wrocław od czego zacza˛c´ ?” (“E-Wrocaw what to strat with?”), Wrocław 2000 Plus Studies on the city strategy, Book 4 (55) 2004, Wrocław, the Municipality of Wrocław, Wrocław Development Office, p. 128. Madryas C, Kolonko A., & Wysocki L., 2003, Poprawa efektywno´sci wykorzystania przestrzeni podziemnej miasta dla umieszczania infrastruktury sieciowej (Improvement of the efficiency of the use of the city underground space for network infrastructure), Wrocław 2000 Plus Studies on the city strategy, Book 7 (51) 2003, Wrocław, the Municipality of Wrocław, Wrocław Development Office, pp. 85–130. Wie˛ckowski T. & Najnigier S., 2004, Czy Wrocław potrzebuje miejskiej sieci teleinformatycznej? (Does Wrocław need a municipal ICT network?), Wrocław 2000 Plus Studies on the city strategy, Book 4 (55) 2004, the Municipality of Wrocław, Wrocław Development Office, pp. 5–22.

139

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Maintenance of drainage system infrastructure in Butare Town, Rwanda A. Karangwa National University of Rwanda, Butare, Rwanda

ABSTRACT: The applicable legislative framework as regards management of wastewater in Rwanda is not yet defined. The current sewerage system does not allow the connection of domestic wastewater from the buildings to the public storm drainage network. Any wastewater discharge other than rainwater, in the public sewers is prohibited. Vis-a-vis this situation, the town of Butare recommends that all the domestic wastewater is collected and discharged into septic tanks and soakaways and then joining the natural environment through groundwater infiltration. According to the importance of Butare it seems that the authorities have to take action in order to modernize its basic infrastructure and to meet acceptable technical standards. Upgrading of water and sanitation infrastructure has thus become one of the priorities for the Butare town, especially because the existing infrastructure is old and no longer sufficient. This paper looks at the sustainability of current drainage system infrastructure and maintenance practices in Butare. Because of lack of data in Butare, especially in terms of maintenance of infrastructure, a field survey was conducted in February and May 2008. The results showed that the water infrastructure suffered from the problems of corrosion, old age, cracks, breakages, and the water supplied is generally not adequate for the town. The open drainage system had problems of siltation, poor maintenance, and inadequate design capacity. The drainage network infrastructure had problems related to poor coverage, no institutional support for maintenance and control, no public sanitation facilities, no regulations and/or standards, etc. From this it was concluded that the drainage network in Butare is critical and not sustainable. It was therefore recommended that the authorities should make a condition inventory of the infrastructure and set up a rehabilitation programme.

1 INTRODUCTION Butare is having a 18% growth of the population (source: Huye District). This growth is not followed by the development of the basic infrastructures such as the network of water and sanitation. For the town of Butare buildings are not served by a wastewater collecting system and consequently they must treat onsite their wastewater before discharging them into the natural environment. Only the drainage network was planned and unfortunately the district does not have data on this network. However, it is noticed that most of the stormwater drains are not maintained. These issues are associated with an ineffective institutional framework for the proper management of the whole water and sanitation infrastructures which is not in place. This poor management is evidenced by the operation of the network which cannot adequately convey the expected loads. This observation gives the impression that the system is not operating to expected standards and therefore there is a need to establish a program of maintenance which will be able to determine the periods and type of intervention. This study aims at giving a contribution to the study of the maintenance strategies and management of the whole of the infrastructures of the drainage system network. This study could of course be used as tools of reference to the authorities of the Huye District and any person interested in the management and/or the maintenance of the drainage network. The results of this study moreover will make it possible to re-examine the means of management of the flow 141

Figure 1.

Satellite image of Butare (www.maplandia.com/rwanda/butare/).

according to the present and future needs. The extent to which these individual components could be influenced in order to arrive at a sustainable needs further study.

2 MATERIALS AND METHODS 2.1 The study area The main study area is the Town of Butare (Fig. 1). Butare was the largest and most important town in Rwanda prior to 1965, when it lost out to the more centrally located Kigali, 135 km to its north, as the capital of independent Rwanda. Today the site of several academic institutions, including the country’s largest university, Butare is a town with a population of about 77,000 and is still regarded to be the intellectual and cultural pulse of Rwanda. The drainage system in Butare is not yet well developed and is not available for all residents. In general, Butare is lacking adequate drainage infrastructure and a clear maintenance strategy. The lack of good management and the possibility to have effective cost recovery are usually blamed to lack of financial resources. Therefore, it is important to have insight in the budget allocation mechanisms, budget constraints, cost recovery mechanisms, subsidies and maintenance strategies. A model on how to proceed was given by Kok and can be apply in the context of drainage system investment. Special attention is to be paid to life cycle costing process so that the district authorities will be able to plan the maintenances activities in time. The following chart shows how the decision of maintenance activities could be taken (Kok et al, 2003). In practice, applying the life cycle approach means developing scenarios for each investment opportunity and forecasting all expenditures during its entire life cycle. When comparing a set of alternative investments, a District decision-maker intuitively is focus upon initial capital costs. The basic idea behind the life cycle costing technique is then to look at the balance between initial and future expenditures of an investment project. The presented here model will help the District decision makers to take an optimal action. 2.2 Approach, data collection and analysis For the purpose of analysis, three existing components of the water sanitation infrastructure were identified. These are: (1) unlined open channels, (2) lined open channels or riprap and (3) concrete pipeline. The study considered mainly structural and hydraulical parameters, focusing on technical 142

Generating investment alternatives

Determining the period of inspection

Forecasting cash flows

Applying financial appraisal techniques

Investment decision (maintenance)

Risk analysis

Figure 2. The life cycle costing process.

state and maintenance practices. A detailed assessment of damages for drainage infrastructures in general was dealt with in separate study (Karangwa, 1999 and Madryas et al, 2002). The current paper shows the state of existing drainage system in Butare and maintenance practices. Considering the fact that Butare does not have a sewerage system but stormwater drainage only, this study will concentrate mainly on the network of rainwater collection. The collection of rainwater by the existing stormwater drainage system is done by infiltration (unlined open channels) and/or by gravity towards a river.The question of sustainability of drainage system infrastructure is discussed in context of its adequacy of the infrastructure to meet current and future loads.

3 RESULTS In order to find out the real state of the drainage infrastructure in Butare, an inspection was made in February and May 2008. The specific objective was to evaluate the damages and propose the maintenance strategy. Because of the nature of Butare’s drainage infrastructure, the visual inspection was enough for determining the state of the above mentioned infrastructure. 3.1 Unlined open channel This category occupies approximately 75% of the whole of the network. This kind of infrastructure shows how poor is the investment in water sanitation infrastructures. In addition to that is lack of budget for maintenance of existing infrastructures. The inspection was conducted in February and May 2008 and the results concluded that the channels are in very bad technical state in terms structural design, strength and hydraulically discharge. The photographs taken at the time of the inspection show the actual state and confirm that the system does not function properly due to the lack of maintenance. According to the observation made in the period from January to May 2008, it seems that at least a monthly maintenance is needed if the channel has to be in acceptable technical state. The picture nr 1 shows the state after cleaning and grass planting on the side slopes. The picture nr 2 shows the channel with grasses. As can be seen on picture nr 2, the proposed channel protection measures disturb the flow velocity and contribute not only to the stopping of the waste but also to the deformation of channel’s geometrical form. According to the Butare town authorities, the mosaic vegetation pattern results from differences in hydro-morphology of the region and grass planting activities have to maintain mainly the edges of the unlined open channels. In this situation 143

Photo 1.

Photo 3.

Grazing of the cleaned storm drainage.

Lined open channel on unpaved road.

Photo 2.

Photo 4.

Drainage infrastructure with grasses.

Lined open channel on paved road.

the strength of the bed and edges protection against erosion is defined by the flow conditions and the strength and the type of soil. The sustainability of the unlined open channels is based on the rain season and the intensity of rain. This means that the maintenance strategy should consider rain season as an important parameter in maintenance planning.

3.2 Lined open channel (riprap) This category occupies about 20% of the Butare’s rainwater network. These channels are constructed in stones and/or bricks and are generally in good state in comparison with the first category. One of the main problems noted during the inspection is that the channels are not regularly cleaned. Amongst the observations were a mud deposit of soil, solid waste, etc. The following photographs show some of the Butare’s riprap channels. The above presented pictures show the cracks, siltation and corrosion of concrete and these damages are observed in many places of this drainage system. This kind of damage will slowly reduce the lined open channel construction to a total destruction if non maintenance is done in near future. 144

3.3 Concrete and reinforced concrete pipeline The concrete and/or reinforced concrete pipeline occupies about 5% of the total drainage system. The pipeline is used in the centre of the city and in culvert constructions. Also this category presents many damages. The photographs bellow show the decrease of strength (structural degradation) as a function of a long period without maintenance. The most impairment observed are cracks, partial and/or total destruction, corrosion, deposition of soil and solid waste, poor inlet construction/cover, exposed pipes, poor pipe joint construction, etc. The picture 5 shows how the current state of pipeline needs an urgent attention. The replacement of the pipeline is required to ensure a long and durable service life.

Photo 5.

Table 1.

Cracked pipe.

Photo 6.

Grated drainage inlet pit.

Results of storm drainage network inspection.

Category

Most observed damages

Picture illustration

Technical state

Grazing the unlined channel

Non regular dimensions, Erosion, Sediment, Tall grasses

bad

Unlined channel

Tall grasses, Stripped channel

very bad

(Continued)

145

Table 1.

(Continued)

Category

Most observed damages

Picture illustration

Technical state

Pipe under public light

Sediment deposit, Clogging of culvert pipes

Very bad

Lined open channel

Grasses in channel, wrench, cracks, corrosion of concrete

good

Junction open channel with a road

Structural degradation of pipe and edge, stripped pipe, stones or wall wrench

bad

Concrete pipe

cracks, corrosion, pipe without ground cover, partial destruction

Very bad

It was also noted that the pipeline is used especially alongside the road Butare Akanyaru and under the road in the places of junction of the canal with the road. The observed damages are presented in Table 1. Generally the rainwater network is characterized by a lack of maintenance, destruction of the pipe, lack of protection, etc. 4 DISCUSSION Deterioration of a drainage network infrastructure in Butare is a function of its physical use and the passage of time. It none that deterioration can be controlled to some extent by the provision of high 146

quality components and high levels of maintenance. According to the results of the inspection, the technical state of the drainage network infrastructure in Butare town is critical. A detailed analysis revealed the following defects (see the Table 1): partial or total destruction of the pipes, poor joint construction, clogged pipes, cracks, lack of regular cleaning (excessive deposition of waste), destroyed and/or missing covers. Other problems also include tall grasses in drains and poor sloping of unlined drains. The photographs in Table 1 illustrate the level of structural degradation. The technical state of the rainwater network infrastructure is therefore described as being very bad and interventions should be instituted as a matter of urgency. The delay in remedial actions will result in the accelerated degradation of the drainage structure and the road network in the district.

5 CONCLUSION 1. The deterioration of the drainage system infrastructure in Butare needs urgent attention. Most notably the ambitious targets set for drainage system upgrading are still not realized. The results of inspection indicate that the sewerage infrastructure needs a structural rehabilitation to meet the requirements of Butare’s drainage system network. 2. The good maintenance and management have an impact on durability of the drainage system, roads and safety of the town’s residents. 3. The standards for the management of all the elements that contribute to the drainage system maintenance have to be developed through risk assessment taking into consideration the real state of the infrastructure in order to meet the core requirements of waste and rainwater drainage.

6 RECOMMENDATIONS 1. The creation of an enabling environment with appropriate policy and legal frameworks is required; policy issues to be addressed include a focus on sustainable development, and drainage infrastructure and economic principles, 2. Institutional development, preferably building on existing institutions; includes national, local, governmental, public and private institutions, and community participation, 3. Human resources development; includes training, career development and performance incentives, and strengthening of managerial systems at all levels, 4. Creating and keeping up to date accurate inventories of sewerage system network, and 5. Creating a District’s infrastructure Maintenance Management System. REFERENCES Karangwa A. (1999) Optymalizacja doboru technologii rehabilitacji betonowych konstrucji przewodow kanalizacyjnych, Praca doktorska, Wroclaw, Polska Kok M. et al (2003) Asset Management, UNESCO-IHE, Delft, The Netherlands Madryas C., Kolonko A., Wysocki L., (2002) Konstrucje przewodow kanalizacyjnych, Oficyna Wydawnicza Politechniki Wroclawskiej, Wroclaw, Polska.

147

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Contact zone in micro tunneling pipelines A. Kmita & R. Wróblewski Wrocław University of Technology, Institute of Building Engineering, Wrocław, Poland

ABSTRACT: Contact zone of the pipes in a jacking pipeline is one of the critical points which decide on the distance which may be drilled form one well do another. The problem of a local crushing capacity in this zone of two jacking pipes is presented in the paper. Standard calculations indicate too optimistic values of the local crushing capacity for this particular case which was concluded from a numerical analysis. This analysis was preformed on three dimensional model with use of finite element method.

1 INTRODUCTION A review of products offered for the pipelines’ construction a very intensive increase of production of precast pipes is noticed in the recent period, particularly for microtunneling. This is a specific group of elements which design is very different than design of pipes used in the excavation method. A difficult task for designers of such elements are design principles introduced in new standards and very rapid technological progress in new generation materials. Since the subject covers a very wide field the paper presents selected issues of reinforced concrete pipes design with particular attention paid to the contact zone. This zone is one of the critical points which decide on the distance which may be drilled form one well do another. From the contractor point of view this distance is one of the most important parameters of the pipe which has an influence on the cost of the pipeline construction. And this is the genesis of the research on technological concepts which lead to resist an arbitrary jacking force in the contact zone of two pipes. The most popular method to increase the jacking force in the reinforced concrete pipes is application of high strength concrete.

2 CHARACTERISTICS OF THE PRECATS JACKING PIPES CONCRETE ELEMENTS Design and construction of the jacked pipelines requires consideration of specific load system both in construction and exploitation phase which is reflected for example in the reinforcement used. In Figure 1 an example of the leading pipe reinforcement is presented and a detail of the pipes’ connection is presented in Figure 2. The micro tunneling technique enables construction of the pipelines curved in horizontal and vertical planes. The curvature radius is determined with the pipe’s diameter, length and the divider thickness (Scherle 1977) which is illustrated in Figure 3, where: tan φ◦ ≈

f s l ≈ ≈ R Da l 149

(1)

160

60 0

φ1720 φ1400

φ8/600

60 0

7

.09

160

3000 28φ8

50

Figure 1.

5 28φ8

6

62 500 6 φ8 I  1062

500 5 φ8 I  1050

Reinforcement of the leading pipe DN1400.

140 230

Figure 2.

18

φ1720

3000

φ1696

φ1720 φ1400

160

10

130

3000

Detail of the pipes’ connection.

I

S

R

f

φ

Do

φ

Figure 3.

φ

Parameters for the curvature evaluation.

150

3 MAXIMUM JACKING FORCE Construction of the curved pipelines requires analysis of the pipe’s capacity In local compression and as a compressed eccentrically loaded element. In Figure 4 a schematic load system for the axially (straight-line pipeline) and eccentrically loaded (curved pipeline) separate element is presented. For eccentric compression two cases for the contact zone are distinguished: closed connection and opened connection. Depending on the resultant force position (PN-EN-1916. 2004) stress distribution presented in Figure 4 are achieved. According to the notation of PN-EN-1916. 2004 jacking forces are evaluated as follows: • Maximum theoretical allowable jacking force Fjmax (no deviations, jacking surfaces are perfectly parallel): (2) Fj max = 0.6fck × Ac with (Fj ≤ Fj max ) Partial safety factor for concrete is equal 1,67. • Maximum jacking force for closed connection Fcj Fcj = 0.5Fj max

(3)

Fcj = 0.3fck × Ac

(4)

• Maximum jacking force for opened connection Foj Foj = 0.3fck × e × Ac

with

(Foj = e × Fcj )

and e ≤ 1

(5)

where: e = a function of the resultant force eccentricity – z, internal and external diameter of the pipe – di and de . The value of e is calculated or may be obtained, to simplify the calculation, for z/de from graphs given for example in PN-EN-1916. 2004 or in PrEN 1916. 1999. If one consider for an arbitrary case of connection the pipe capacity calculated from the local crushing condition and capacity of compressed element it is evident that the local crushing decides. To determine a maximum length between the wells (jacking distance L) in a pipeline it is necessary to evaluate the maximum jacking force. The approximate jacking force according to Chapmann D. N. & Jchioka Y. 1999 has fulfill the following equation:

0.6fck

F = f0 + π · d e · P · L π f0 = (Do )2 Po 4

Figure 4.

1

(6) (7)

2

Stress distribution for: 1 – opened connection, 2 – closed connection.

151

with: F = total jacking force, fo = primary resistance, P = Frictional resistance along pipe, L = jacking distance, Do = outer diameter of excavation, Po = face resistance.

4 STRESS ANALYSIS IN THE CONTACT ZONE Proper design of the contact zone requires not only calculation of the maximum force for the given eccentricity but also stress distribution in the contact zone is necessary. To determine stress distribution in this zone numerical analysis was performed. The analysis was based on finite element method with use of Lusas 2008 software. Three dimensional model of the pipe connection was developed as presented in Figure 5. The top part was driven into the bottom one. Two approaches were made for the analysis. In both cases linear elastic material properties were assumed for the pipe’s concrete, steel ring, wooden divider, rubber divider and rubber gasket. In the simplified model there is no contact zone between the dividers and the pipes, elements are connected in nodes. For the advanced model the contact zone was assumed. In the first case both compression and tension are transmitted through the contact zone but in the second case only compression is transmitted so the open connection might be properly modeled. Unfortunately the second model failed to converge under eccentric loading and only results form the first model are presented. A series of calculations were made to obtain stress distribution in the contact zone for various jacking load eccentricities. The load level was assumed to achieve 0.8fck . Typical stress distribution in the section of the pipe is presented in Figure 6. Compressive stress concentrations are observed at the area close to the top surface of the wooden divider. This is the area were an extra reinforcement for local crushing might be necessary. The depth of this concatenation is approximately equal 5cm, which is measured from Figure 7, and is independent from the load eccentricity. Unlike assumed in PN-B-03264. 2002 stress distribution over the pipe perimeter is not linear, so the calculations may overestimate the local crushing capacity (this capacity is the largest one among all the forces presented in Table 1). As one can notice in Figures 6 and 7 large compressive stresses appear above the wooden divider. This is the area where in some cases an extra reinforcement for local crushing may by necessary.

Figure 5. Three dimensional model of the pipes connection.

152

Figure 6.

Stress distribution in a section of the pipes’ connection. Stress along the pipes’ length (blue – compression, red – tension). 0 0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0,45

0,5

Axial stress [kPa]

10000

20000

30000

e  0.78 m e  0.624 m

40000

e  0.468 m e  0.312 m

50000

e  0.156 m e0

60000

Figure 7.

Distance x [m]

Stress distribution along the pipe length for various load eccentricities e (x = 0.5 m – contact surface). Horizontal section above the wooden divider shown in Fig. 6.

Table 1. The pipes’ capacities (concrete C40/50). DN600/130

DN/thickness

DN1400/160

DN2200/200

Axial loading

Loading as for closed connection

Axial loading

Loading as for closed connection

Axial loading

Loading as for closed connection

Local crushing acc. to PN-B-03264. 2002

kN

6618

4412

17408

11605

33476

22317

Capacity of RC compressed element

kN

5024

3765

13320

9696

25388

18106

Fjmax | Fcj

kN

5141

2571

14679

7340

29880

14940

153

5 CONCLUSIONS The results presented in the paper indicate that standard calculations are too optimistic in estimation of the local crushing capacity for connection of two jacking pipes. This conclusion arises from the analysis of the numerical simulation because stress distribution is very different from the assumed in the standard calculations. Large compressive stress appear above the divider and this is the area were an extra reinforcement should be placed or particular attention should be pied in a design. The area in which this extra reinforcement should be placed is approximately equal to two wall thickness. REFERENCES Scherle M. 1977. Rohrvortrieb. Sattik, Plannung, Ausfuhrung. Band 2. Wiesbaden und Berlin: Bauverlag Gmbh. Kmita A. et al. 2001. Prefabrykowane rury z˙ elbetowe stosowane do mikrotunelingu. In Diagnostyka, utrzymanie, remonty, modernizacje oraz budowa obiektów budowlanych na terenie Lubi´nskiego Zagłe˛bia Miedziowego. Wrocław: Wydawnictwo Politechniki Wrocawskiej. Kmita A. 2005. Problemy projektowania rur i kształtek betonowych w s´wietle obowia˛zuja˛cych norm i przepisów budowlanych. In Infrastruktura podziemna miast. IX Konferencja naukowo-techniczna. Prace naukowe Instytutu In˙zynierii La˛dowej Politechniki Wrocławskiej Nr 53/20. Wrocław: Wydawnictwo Politechniki Wrocawskiej. PN-EN-1916. 2004. Rury i kształtki z betonu niezbrojonego, betonu zbrojonego włóknem stalowym i z˙elbetowe. Warszawa: PKN. PrEN 1916. 1999. Concrete pipes and fittings unreinforced steel fibre and reinforced. CEN. ATV_A 161 E. 1990. Structural Calculation of Driven Pipes. GFA. PN-B-03264. 2002. Konstrukcje betonowe, z˙elbetowe i spre˛˙zone. Obliczenia statyczne i projektowanie. Warszawa: PKN. Beckmann D. et al. 2007. Co Jack – A new statics method of computing and controlling pipe jacking. In Tunneling and Underground Space Technology 22. Elsevier. Chapmann D. N. & Jchioka Y. 1999. Prediction of jacking forces for microtunneling operations. In Trenchless Technology Research vol. 14, No. 1. Elsevier Science. Lusas 2008. Lusas Finite Element System. User Manual version 14. Kingston upon Thames: FEA Ltd.

154

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Effect of variable environmental conditions on heavy metals leaching from concretes A. Król Faculty of Environmental Engineering, Opole University of Technology, Opole, Poland

ABSTRACT: The more common use of alternative fuels and mineral additions in the cement industry and in the process of concrete manufacturing constitute the reasons for the increased focus on the impact of the environmental conditions on the level of heavy metals leaching from concretes. Such assessments should involve a variety of environmental factors which affect the concretes during their normal service life. This paper undertakes the subject of the level of heavy metals leaching from concretes based on the Portland cement and Portland fly ash cement. The factor associated with the effect of various environmental factors on heavy metals leaching has been particularly taken into consideration over extensive research periods. Hence, the effect of the reduced pH of the aquatic environment on the level of heavy metals leaching was recorded. The concretes were also subjected to the impact of sea water environment and heavy metals leachability levels were taken. The leaching was determined both from the monolithic and grounded concrete forms.

1 INTRODUCTION The major components of cements in commercial use include CaO, SiO2 , Al2 O3 , Fe2 O3 , SO3 , K2 O, Na2 O, TiO2 and P2 O. The admixture of trace elements in cements depends essentially on the incidence of natural raw materials (limestone, clay, sand, iron ore and gypsum ore as well as fossil fuels). The partial replacement of these natural resources by alternative fuels and raw materials, which are recovered from selected waste streams and industrial by-products, may result in variations depending on their trace element contents and the proportion used. The contact of the cement-based construction materials (concrete, mortar) with water during normal service life, recycling, and disposal may result in an increase of leaching of trace elements. Research undertaken on refuse-derived fuel in the co-incineration process examined the potential environmental impact of heavy metals produced in the co-incineration process (Gendebien et al. 2003). The more common use of alternative fuels and fuels from wastes streams in the cement industry and in the process of concrete manufacturing has encouraged interest into the subject of the effect of environment related conditions on the level of heavy metals leaching from concretes. While the impact and the properties of silica fly ash and granulated blast furnace slag does not on cement raise any doubt, the assessment of the environmental effects of the cement product based on them remains a long way from being comprehensive. The reason for that is associated with the long duration of the process of study necessary for the assessment of the heavy metals release into the natural environment. The amounts of hazardous substances released into the environment from concrete constructions is affected by a number of factors, which include the type of the concrete sample (monolithic, ground) and the environment related factors associated with the contact surface (ground, water, waste, chemically aggressive environment) (Dijkstra et al. 2005). The level of heavy metals leaching from the concretes is also determined from the structure of the hardened 155

concrete, in particular the compositions of the concrete mixture (i.e. mainly the content and type of cement). The huge role in the process of heavy metals binding is attributed to the C-S-H phase and the calcium aluminate and sulfoaluminate hydrates, the phases which originate from the process of cement hydratation. As a result, the porousness of the hardened paste is decreased together with the mobility of the heavy metals (Glasser. 1994). The entire system of research should also account for the various environment related characteristics which affect the material or the concrete construction. It is also indispensable to assess the release of the hazardous substances over the entire service life of the concrete materials (from the manufacturing, through exploitation until the “end of life” and the opportunity to recycle or reuse the materials). The effect on the level of heavy metals leaching from the material or concrete structure is subjected to the conditions of its exposition (i.e. application scenarios). The list includes six conditions of concrete exposition; however, the combinations are sometimes the case (Dijkstra et al. 2005): – granulated materials placed inside or on the ground surface, – monolithic materials placed inside or on the ground surface, – monolithic forms exposed to aquatic and dry environment (e.g. exposition to rainfall and sunlight), – loose granulated material e.g. crushed concrete with various particle size, – pipes (e.g. for supply of drinking water) placed in the ground (the leaching of heavy metals is possible both into the water stream and the surrounding ground), – monolithic forms exposed to aquatic environment (e.g. coastal constructions). The determination of the form in which the concrete will occur along with the environment related factors are relevant due to the complexity of the leaching processes, which can accompany each of the scenarios. The release of heavy metals from concrete materials may involve both the leaching from the external surface of the concrete as well as solution, perlocation and diffusion. Concurrently, one has to bear in mind that the level of the heavy metals leaching will be different for monolithic forms in the permanent contact with aquatic environment from the level for the identical concrete forms in contact with the soil. Hence, prior to the beginning of the assessment of the leaching level from concrete materials it is necessary to take into consideration the conditions of the exposition (van der Sloot. 2000). For the purposes of this research the material based on cement products and concretes will be considered a monolith. This means that the leaching is constrained due to conditions of the transportation. Since the diffusion of the ions from the interior of the matrix to the environment is a lengthy process. The test selected for the purposes of this study is called the tank leach test. The leachability can be modeled in the first approach by a one-dimensional diffusion model in order to assess release over longer time- scales than the duration of the leaching test (van der Sloot. 2000). In the second life of cement-based products mechanical demolition or by degradation will result in the reduction of particle size. The material can either be used as bound aggregate in new concrete or an unbound aggregate. When the material is applied as a bound aggregate in concrete, it is necessary to assess the leachability of the new product, which will be similar to the course of action on the specimens tested in their intact form (van der Sloot. 2000). The long-term properties of cement-based products in their second life in unbound form are assessed with the aid of the commonly available testing (e.g. method referred to in EN 12457-4:2002 standard).

2 MATERIALS AND METHODS The conducted research involved the design and development of concretes with the use of: Portland cement type CEM I 32,5R; Portland cement with fly ash CEM II/B-V 32,5R-HSR (content of fly ash amounted to ca. 30%). The chemical composition of the cements used in the research is summarized in Table 1. Table 2 contains the content of heavy metals in the Portland cement and Portland cement with fly ash. 156

Table 1.

Chemical composition of cements used in researches. Content, %

Chemical composition

CEM I 32,5R

CEM II/B-V 32,5R-HSR

Loss on ignition CaO SiO2 Al2 O3 Fe2 O3 MgO SO3 K2 O Na2 O Cl−

3.46 64.60 19.20 4.69 3.04 1.22 2.65 0.81 0.09 0.047

3.71 48.11 27.10 10.84 3.93 1.61 1.95 1.41 0.39 0.037

Table 2.

Content of heavy metals in cements used in researches. Content, mg/kg

Figure 1.

Heavy metals

CEM I 32,5R

CEM II/B-V 32,5R-HSR

Zn Cr Ni Pb Cu

316 54 18 24 60

262 52 26 31 54

Principles of placing concrete sample in a container in accordance with the EA NEN 7375:2004 standard.

The composition of the concreter mixture was as follows: cement – 300.0 kg/m3 ; sand – 685.2 kg/m3 ; gravel 2 ÷ 8 mm – 600.4 kg/m3 ; gravel 8 ÷ 16 mm – 628.6 kg/m3 ; water – 180,0 kg/m3 ; water/cement ratio (w/c) in the two mixtures was 0.6. The resulting mixture server as the material for formation of 10×10×10 cm cubes. After 24 hours the samples were removed from the mould and subsequently subjected to leaching tests. The aqueous extracts from concretes were taken by various methods, which comes as a result of simulating various life stages of concrete and the effect of environment related factors: method I – procedure follows on from the methodology in the standard EA NEN 7375:2004 standard. This test involves the preparation of aqueous extract from a monolithic sample. The concrete cubes are placed on pillars, hence enabling them to remain in contact with the liquid across the entire surface of the sample (Fig. 1). 157

The volume of the liquid which is used to immerse the sample must not be smaller than twice the volume of the examined sample and is to be determined in accordance with the formula (1). 2 × Vp ≤ V ≤ 5 × Vp

(1)

where: Vp – volume of sample, V – volume of liquid. In the conducted examinations the concrete sample was placed in a container holding 4 liters of distilled water. The total duration of the test amounted to 64 days and was divided into 8 research periods (0.25; 1; 2,25; 4; 9; 16; 36; 64 days). After each of the researched periods an aqueous extract was taken from the container and the liquid was replaced. After the period of the testing the samples were kept in distilled water for a period of 360 days. The content of heavy metals was subsequently taken in the resulting extracts. This test is a procedure to evaluate the release from monolithic material by predominantly diffusion control (e.g., exposure of structures to external influences). – method II – modified method I – the leaching liquid was acidified with HNO3 until pH = 4 was gained. This method illustrates the effect of an aggressive environment with a low pH (acid rain, reduced acidity of ground) on the level of heavy metal ion leaching. At the same time, pH = 4 must be noted to be the boundary level of exposition of concrete class XA3 (strong acid chemical environment) in accordance with PN-EN 206-1 standard, – method III – modified method I – as the leaching liquid was sea water prepared in accordance with ENV 196-X:1995. In order to prepare sea water the following admixture of substances is to be solved in each 1000 g of distilled water: • Sodium chloride (NaCl) 30,0 g • Magnesium chloride MgCl2 ·6H2 O) 6,0 g • Magnesium sulfate (MgSO4 ·7H2 O) 5,0 g 1,5 g • Calcium sulfate (CaSO4 ·2H2 O) 0,2 g • Potassium bicarbonate (KHCO3 ) – method IV – specified in standard EN 12457-4:2002 “Characterisation of waste – Leaching Compliance test for leaching of granular waste materials and sludges – Part 4: One stage batch at a liquid to solid ratio of 10 l/kg for materials with particle size below 10 mm (without or with size reduction)”. Water extracts were made on 100 g – samples of concrete, reduced to particle grain size under 10 mm. The material was flooded with leaching liquid (distilled water with pH = 7) and subjected to the process of leaching (agitation) for 24 hours. Weight ratio of water (L) to solid mass (S) was 10 (L/S = 10). This method was applied for the simulation of the application scenarios which concern the granulated forms of concretes (e.g. crushed concrete). The research was based on the grounded concretes based on the Portland cement and Portland fly ash cement. The concretes were left to mature in laboratory over 360 days. – method V – samples of concrete after 360 days of setting were subjected to leaching process by the application of modified version of method IV – since acidified liquid HNO3 to pH = 4 was used for the preparation of water extract. This method simulated the influence of aggressive medium with low pH (e.g. acid rain in environment, aggressive underground water) on the level of heavy metals ions leaching. This method illustrates the impact of an aggressive environment with reduced pH on the ground concrete. The content of heavy metals in water extracts received during carried tests was determined thanks to emission spectrometer with plasma inductively excited ICP-AES.

3 RESULTS AND DISCUSSION The total concentration of heavy metals in water extract from concretes taken after 64 days with method I are summarized in Table 3. Table 4 contains the leachability of heavy metals from liquid environment with pH = 4. 158

Table 3.

Heavy metals concentration in water extracts from concretes (monolithic concrete sample) – 64 days. Concentration of heavy metals in water extracts, mg/kg

Heavy metals

Concrete on CEM I 32,5R

Concrete on CEM II/B-V 32,5R-HSR

Permissible value*

Permissible value**

Zn Cr Ni Pb Cu

0.01169 0.0155 0.00333 0.00181 0.00781

0.01558 0.01235 0.00136 0.00145 0.00992

– 0.05 0.02 0.025 2.0

3.0 0.05 0.05 0.05 0.05

* According to Council Directive (98/83/EC) – the quality of water intented for human consumption ** According to Council Directive (75/440/EEC) – permissible heavy metals concentration in waters with A1category (surface waters requiring simple physical treatment, in particular filtering and disinfection)

Table 4.

Heavy metals concentration in water extracts from concretes (monolithic concrete sample) – 64 days/liquid environment with pH = 4. Concentration of heavy metals in water extracts, mg/kg

Heavy metals

Concrete on CEM I 32,5R

Concrete on CEM II/B-V 32,5R-HSR

Permissible value*

Permissible value**

Zn Cr Ni Pb Cu

0.01173 0.0158 0.00156 0.00198 0.00827

0.01913 0.01283 0.00138 0.00164 0.01197

– 0.05 0.02 0.025 2.0

3.0 0.05 0.05 0.05 0.05

* According to Council Directive (98/83/EC) – the quality of water intented for human consumption ** According to Council Directive (75/440/EEC) – permissible heavy metals concentration in waters with A1category (surface waters requiring simple physical treatment, in particular filtering and disinfection)

The analysis of the gained results indicates that the amounts of heavy metals released from concretes are small. The differences regarding the level of leachability for the samples placed in distilled water and pH = 4 aqueous environment are similar (Tables 3 and 4). The comparison between the recorded levels of heavy metals leaching and admissible values of heavy metals concentrations in waters group A1 in accordance with European Council Directive (75/440/EEC) regarding quality requirements which surface fresh water used or intended for use in the abstraction of drinking water after physical treatment through filtration and disinfection indicate that the levels do not contravene the regulations in force. Concretes in monolithic forms were left to be exposed to leaching liquid (distilled water and water acidified with HNO3 ) over 360 days. After that period the water extract have been taken. The concentration of heavy metals after 360 days of test duration is summarized in Table 5. The liquid with a reduced pH leads to an increased level of zinc leaching in concrete based on cement type CEM I; however, the leaching of chromium, nickel and lead is lower. For the case of 159

Table 5.

Heavy metals concentration in water extracts from concretes (monolithic concrete sample) – 360 days. Concentration of heavy metals in water extracts, mg/kg

Concentration of heavy metals in water extracts, mg/kg (leaching liquid of pH = 4)

Heavy metals

Concrete on CEM I 32,5R

Concrete on CEM II/B-V 32,5R-HSR

Concrete on CEM I 32,5R

Concrete on CEM II/B-V 32,5R-HSR

Zn Cr Ni Pb Cu

0.021 0.034 0.021 0.063 0.008

0.031 0.059 0.034 0.037 0.009

0.036 0.022 0.012 0.019 0.008

0.031 0.073 0.023 0.011 0.009

0,45 Concentration, mg/l

0,4 0,35 0,3 0,25

CEM I (s.w.) CEM I (d.w.)

0,2 0,15 0,1 0,05 0 Zn

Figure 2.

Cr

Ni

Pb

Cu

Comparison between heavy metal concentrations in water extract from concretes based on CEM I under the effect of distilled water (d.w.) and sea water (s.w.) after 360 days of test duration.

concretes based on the Portland fly ash cement under pH = 4 environment the release of chromium is higher, whereas the levels of nickel and lead are lower. For all concretes regardless of the used cement type and pH of the environment the leaching of copper remains constant after 360 days of test duration. After a longer period of liquid affecting concretes the levels of heavy metals leaching in the analyzed water extracts grow (see Tables 3–5). Chlorides tend to be corrosive factors affecting the strength and service life of concretes. The chloride based corrosion takes places mostly as a result of the impact of sea water on the concretes; such an effect occurs more rarely as a result of mine waters and de-icing agents used in the winter period on the roads (Giergiczny, 2008). In the conducted examinations the concretes based on Portland cement and Portland fly ash cement were subjected to sea water environment. The effect of the exposition of concrete samples based on Portland cement to sea water results in the leaching of bigger amounts of chromium (Fig. 2) in comparison to the concentration of this metal under the effect of distilled water. Similar results were recorded for the Portland fly ash cement (Fig. 3). The research shows that cements based on CEM I tend to indicate higher zinc leaching under the impact of sea water. The concretes under impact of chlorides due to sea water exposition are subjected to the reaction between chlorides and calcium hydroxide, which originated from the hydratation of the clinker phases and reduced pH of the solution. The diffusion of the chlorides which penetrate the concrete leads to the crystallization of salts in the capillary pores and to concrete failure as a result (Kurdowski, 1991). If the structure of the concrete develops a fault there is a hazard of leaching additional amounts of heavy metals from the newly exposed surfaces. Hence, 160

0,4 Concentration, mg/l

0,35 0,3 0,25

CEM I (s.w.) CEM I (d.w.)

0,2 0,15 0,1 0,05 0 Zn

Figure 3.

Cr

Ni

Pb

Cu

Comparison between heavy metal concentrations in aqueous extract from concretes based on CEM II/B-V under the effect of distilled water (d.w.) and sea water (s.w.) after 360 days of test duration. Table 6.

Heavy metals concentration in water extract from concretes (crushed concrete; 360 days of concrete maturation). Concentration of heavy metals in water extracts, mg/kg

Heavy metals

Concrete on CEM I 32,5R

Concrete on CEM II/B-V 32,5R-HSR

Permissible value*

Zn Cr Ni Pb Cu

0.0057 0.028 0.0012 0.00043 0.0011

0.0063 0.005 0.0005 0.00047 0.0019

– 0.05 0.02 0.025 2.0

* According to Council Directive (98/83/EC) – the quality of water intented for human consumption

this may have formed the reason for the elevated concentration of chromium and zinc. It is important to note here that the leachability of Cr and Zn is lower for the concretes based on Portland fly ash cement. This may be associated with the fact that the admixture of fly ash to the cement results in the reduction of the permeability of the concrete pastes. The pores are filled with the products of the pozzolanic reaction, which is reflected in the more compact microstructure of the hardened paste and results in a more complex diffusion of chlorite ions (distribution surface more developed), the higher concentration of ions in eluates (Giergiczny, 2008). The research also involved the assessment of how the grinding of concrete to grains sized below 10 mm affects the leaching of heavy metals from concretes based CEM I and CEM II/B-V. Grinding of matrices being subjected to tests has a considerable impact on the level of leaching of heavy metals into the water extracts – the lower grain-size (Hohberg et al., 1994). However, the resulting concentrations of heavy metals from aggregate concretes are well under the admissible levels of heavy metal concentration in surface fresh water used or intended for use in the abstraction of drinking (Table 6). The leachability testing of crushed concretes were conducted both in the environment of distilled water and liquid with pH = 4 (leaching methods: IV and V). The reduced pH of the leaching liquid leads to an increase in the concentration of heavy metals in water extracts (Fig. 4) regardless of the cement type. Concretes were also exposed to atmospheric conditions over the period of 360 days. After the time the crushed concrete was subjected to leaching in accordance with the procedure in method IV. It was assessed that the leaching of zinc is below the detection level (r µ(r − ρ)

(16)

– mean waiting time for starting of service: E(Tw ) =

– mean waiting time for service if all servicing sets are engaged: E(Tv ) =

1 rµ − λ

(17)

Solutions presented are valid on condition that ρ=

λ Kb2 1500 3100 12900 49900 199000

1000 1000 1000 1000 1000

1.5 3.1 12.9 49.9 199

4 5 6 7 8

where: Tbr = mean gross repair time, covering beside repair tasks proper also intervals in work (e.g. night breaks), but without intervals for time not utilized due to bad work organization, etc., Td = time not utilized, connected with unforeseen circumstances (e.g. inefficient equipment or means of transport, lack of materials, absence of workers, etc.). Time Tnbr is referenced to length of pipeline between wells. Taken in the example is µ = 0.05 1/d. ρ=

λs = 2.4 µ

Minimum number of teams: r = 3 – condition (18). Criterion function takes the form: f (r) = Kz1 E(U ) + Kb2 E(Or ) = Kz1 ·

ρr+1 + Kb2 · (r − ρ) (r − 1)!C(r − ρ)2

(28)

where: C according to formula (9). Number of teams calculated as assumed for requirement of example of costs Kz1 and Kb2 : Kz1 = 1000 zł/d, Kb2 = 1500 zł/d. (zł – currency unit). Calculation for establishing optimum number of teams r presented in Table 1. Function f (r) assumes minimum for number of repair teams r = 4. Studies were also conducted for function f (r) depending on changes of costs Kz1 and Kb2 . Gradually changing the costs, optimum number of teams were determined. The results are presented in Table 2. For determining the number of teams, relation of costs is important and not absolute values. Results of calculations show that for assumed intensity of damages and renovation of network, the number of teams r amounts to 3 or 4 for unit costs of unreliability lesser or equal to team employment costs. If team employment costs are less than unreliability costs, optimum number 178

Rλ E0

(R 1)λ E1

µ

(R 2)λ

2µ

(R i1)λ

... ...

E2 3µ

(R i)λ

iµ

(R r1)λ

... ...

Ei

rµ

(i1)µ

(R r)λ

rµ

States without queue

Figure 3.

(R j)λ

... ...

Er

... ...

Ej rµ

rµ

λ ER rµ

States with queue

Graph of states for system M(M)r:(R,L) (Filipowicz 1996).

of teams increases gradually from 4, whereby employment of consecutive team is justified with considerable growth of unreliability costs (ratio Kz1 /Kb2 ). 2.2.4 Example II The question arises whether the assumption treating lengths of pipeline between wells as elements of SMO (mass service system), is justified. Homogeneous lengths of pipelines (e.g. diameter, material) comprising a group can be taken as elements. Additionally, maximum length of an element can be limited. In this way, the number of elements can be significantly limited making it necessary to apply SMO type M(M)r:(R,L). Theoretically, it is possible to damage all the pipelines hence the queue capacity amounts to L = R − r. The graph of states of SMO type M(M)r:(R,L) is presented in Figure 3 (Filipowicz 1996). Formulae for probabilities of system states and corresponding values of waiting in stationary state can be found in subject literature. The criterion function here takes the form: R−r R!  n P0 ρr+n r! n=0 r n (R − r − n)!   r−1 r−1   + Kb2 · r − iPi − r 1 − Pi

f (r) = Kz1 E(U ) + Kb2 E(Or ) = Kz1 ·

i=0

(29)

i=0

where:  r  P0 =  i=0

−1 R  R! R! ρj  ρi + j−r i!(R − i)! r!(R − j)!r j=r+1

(30)

Pi =

R! ρi P0 for 1 ≤ i ≤ r i!(R − i)!

(31)

Pj =

R! ρ j P0 for r + 1 ≤ j ≤ R r!(R − j)!r j−r

(32)

In case of this SMO model, condition (18) need not be fulfilled. To illustrate the derivation, it was decided to calculate the optimum number of repair teams for group A pipelines in example I, distinguishing the following elements (maximum length of element taken as 500 m): brickwork pipelines of diameters ø 500 – 1 pipeline, brickwork pipelines of diameters ø 1000 – 3 pipelines, stoneware pipelines of diameters ø 200 – 4 pipelines, stoneware pipelines of diameters ø 400 – 5 pipelines, concrete pipelines of diameters ø 300 – 5 pipelines, concrete pipelines of diameters ø 500 – 6 pipelines, concrete pipelines of diameters ø 1000 – 15 pipelines. Total number of system elements R = 39. In SMO type M(M)r:(R,L), it is assumed that all elements undergo damage with the same intensity λ. For pipelines, mean intensity of damage λ should be assumed. λ=

λs R

179

Table 3.

Calculation for establishing optimum number of teams r SMO type M(M)r:(R,L), R = 39.

Kz1 [zł/d]

Kb2 [zł/d]

r

f (r)/103 [zł/d]

1000

1500

1 2 3 4 5 6

21.33 5.67 2.22 2.91 4.24 5.70

Table 4.

Calculation of function f (r) and ropt for changing costs Kz1 and Kz2 SMO type M(M)r:(R,L), R = 39.

Kz1 [zł/d]

Kb2 [zł/d]

Kz1 /Kb2

ropt

Kz1 < Kb2 100 200

1000 1000

0.1 0.2

2 3

Kz1 = Kb2 1000

1000

1

3

Kz1 > Kb2 1200 1300 6200 27000 115500 549500

1000 1000 1000 1000 1000 1000

1.2 1.3 6.2 27 115.5 549.5

3 4 5 6 7 8

Intensity of damage λs and intensity of renovation µ were assumed as in example I: λs = 0.12 1/d, µ = 0.05 1/d, λ =

0.12 λ = 0.003 1/d, ρ = = 0.06. 39 µ

Number of teams were calculated for Kz1 = 1000 zł/d, Kb2 = 1500 zł/d (Table 3). Function f (r) assumes minimum for number of repair teams r = 3. Similarly as is example I, studies were conducted of function f (r) depending on changes of costs Kz1 and Kb2 for defining the optimum number of teams. The results are presented in Table 4. In Table 5, the number of teams were defined for SMO consisting of 160 elements – as in example I, utilizing SMO type M(M)r:(R,L). For this system, the optimum number of teams is larger with the same relation Kz1 /Kb2 than in the case of a less numerous system. Comparison of Tables 5 and 2 shows the differences caused while SMOs of different types are taken for calculations. In the examples presented, the differences are not large, however the question arises as to what number of elements R justifies the application of the simpler M(M)r:(∞,∞) system.

3 SUMMING UP Maintaining the technical efficiency of large systems as are water supply and sewage systems is costly and comprises a significant position in budgets of enterprises managing such systems. Seeking optimum models for maintaining such systems is therefore fully justified and indispensable. It 180

Table 5.

Calculation of function f (r) and ropt for changing costs Kz1 and Kz2 SMO type M(M)r:(R,L), R = 160.

Kz1 [zł/d]

Kb2 [zł/d]

Kz1 /Kb2

ropt

Kz1 < Kb2 100 650

1000 1000

0.1 0.65

3 4

Kz1 = Kb2 1000

1000

1

4

Kz1 > Kb2 3500 3600 15000 59000 244000

1000 1000 1000 1000 1000

3.5 3.6 15 59 244

4 5 6 7 8

seems that mathematical basis created for models of mass service systems may find application in this case. Presented in the paper are analyses of the problem for which inspirations were put forth earlier, already in the nineties of the past century, with relevant studies conducted by Polish scientific environment. The material presented shows the full complexity of the problem which is most probably the reason that up to the present in Poland, these proposals have not found practical application in servicing water supply and sewage systems. Undoubtedly, the reason for this state of affairs is the fact that during the period when the said models were created, there were no computer databases about the systems, their states, procedures of rehabilitation processes, digital maps of the systems, etc., in the enterprises managing the systems. Presently, the situation has undergone complete change causing the subject to be worth a repeated, expanded scientific analysis.

ACKNOWLEDGEMENT Part of this research was made possible by funding awarded as part of professorial grants to Prof. Cezary Madryas by the foundation for Polish Science. REFERENCES Denczew S. 1998. Application of reliability principles in utilization process of water supply connections subsystem. Gas, Water and Sanitary Engineering (GWiTS) no. 9: 378–382. Filipowicz B. 1996. Stochastic models in operation studies, analysis and synthesis of servicing systems and railway networks. Warsaw: Scientific and Technical Publishers (WNT). ˛ Kłoss-Trebaczkiewicz H., Kwietniewski M., Roman M. 1993. Reliability of water supply and sewage systems. Warsaw: Arkady. Ko´zniewska I., Włodarczyk M. 1978. Reliability renovation and mass service models. Warsaw: State Scientific Publishers (PWN). Wieczysty A., Iwanejko R., Lubowiecka T., Rak J. 1990. Defining the number of repair teams in water distribution subsystem applying mass service models. Gas, Water and Sanitary Engineering (GWiTS ) no. 7: 136–138. Wieczysty A., Iwanejko R. 1996. Determining the required level of reliability of water supply system objects. Gas, Water and Sanitary Engineering (GWiTS) no. 2: 54–58.

181

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Utilizing the Impact-Echo method for nondestructive diagnostics of atypically located pipeline C. Madryas, A. Moczko & L. Wysocki Wrocław University of Technology, Wrocław, Poland

ABSTRACT: Presented in the paper is the specificity of nondestructive defectoscope measurements realized by means of the Impact-Echo method in diagnostics of underground infrastructure. Discussed herein are basic factors of the method with respect to testing this kind of building objects giving special consideration to the ring-section structure acting on which is the surrounding ground. Deliberations were conducted on the basis of results of applying this method for diagnostics of atypically located, reinforced concrete, underground pipeline, draining flotation tailings reservoir ˙ “Zelazny Most” (Iron Bridge) in Poland.

1 INTRODUCTION Progress in the field of diagnostic examination have resulted in elaboration of a number of new testing methods enabling relatively versatile identification of actual technical state of the examined object without the need for conducting expensive and time-consuming laboratory testing. For obvious reasons, the greatest interest is aroused by diagnostic methods which enable undertaking rapid and important decisions, directly on site, having not only technical significance but also substantial economical measure. Among the numerous technical solutions, such expectations are fulfilled to a considerable extent by the Impact-Echo method (ASTM C 1383–98, Carino et al., 1992, Sansalone et al., 1989) worked out in USA. It enables nondestructive measurement of thickness of all kinds of concrete elements having accessibility from one side as well as the widely understood defectoscope examination (flaw detection) of concrete structures. The range of examining possibilities using Impact-Echo make the only alternative method in practice to be laboratory testing of test pieces taken from the structure. Such testing, leading indeed to very precise results, however requires: • uncovering the structure which in some cases is not possible, • repairing the damage caused by taking test pieces, • more prolonged laboratory testing. In effect, in spite of the mentioned features of testing conducted on test pieces, even if such test pieces can be taken, such testing is prolonged and very expensive. Whereas the Impact-Echo method enables carrying out control tests in short period of time, in practically unlimited number of measurement points. As a definitive inconvenience of the method, one must acknowledge the fact that this type of testing requires the application of very costly and complicated equipment which can only be handled effectively by highly qualified engineering staff. This is probably the reason that in Poland, “ImpactEcho” is still insufficiently utilized particularly in case of underground infrastructure. Despite inconsiderable experience in the scope of examining underground infrastructure, the authors of this paper utilized the “Impact-Echo” method to determine the structural state of reinforced concrete ˙ pipeline transporting water in flotation process. The duct is located under the bottom of Zelazny Most – the largest flotation tailings reservoir in Europe. 183

2 TESTING DESCRIPTION 2.1 Description of the reservoir ˙ The reservoir “Zelazny Most” of area 13.94 km2 and embanked with a dam 14.3 km long, is located in the south-western part of Poland about 100 km from Wrocław. Construction of the storage reservoir was begun in 1974 and its utilization in 1997. Annual quantity of flotation tailings dumped into the reservoir ranges from 20 to 26 million tons. Only 25% of this mass is assigned for further processing and the remaining 75% mass of wastes constitutes the annual growth of the deposit. It is envisaged that the reservoir will ultimately accommodate 1.1×109 m3 of wastes which will cause the reservoir to be used until the copper deposits in Głogów Głe˛boki area are exhausted. A picture of the reservoir is presented in Figure 1 (www.kghm.pl). Apart from the basic function of the reservoir described above, i.e. utilization of wastes, it performs two additional functions: • settling reservoir clarifying the water above the sediment, utilized in the flotation process, • retention and dosing reservoir of excess technological water used in mining process circulation. 2.2 Description of the pipeline Taking into consideration the said technological processes, the basic accessory of the reservoir therefore is the pipeline carrying away excess accumulated water. This ducting is laid under the bottom of the reservoir under the constantly thickening layer of deposit whose depth reaches up to several dozen meters. It carries away water beyond the range of the structure through the overflow located on the reservoir. A picture of the overflow is presented in Figure 2 (www.kghm.pl). The subject of examination is the pipeline consisting of steel piping of diameter 1200 mm built up with a reinforced concrete structure whose main task is to carry the mechanical loading acting on the structure. Cross-section of the ducting so constructed is shown in Figure 3.

Figure 1.

˙ Picture of the reservoir “Zelazny Most”.

184

Figure 2.

˙ Picture of the overflow in the reservoir “Zelazny Most”.

1900 600

600

350

400

350

200 “P ” 2

500

600

“P ” 3

2100

400

“P ” 1

Existing reinforced-concrete housing Steel pipe 1200 mm diameter

Figure 3.

Cross-sectional view of the structure examined.

185

Figure 4.

Inside view of the pipeline.

Long-term utilization of the pipeline has caused substantial incrustation of its internal surface as illustrated in Figure 4, making the examination difficult. 2.3 Organization and procedure of testing The accepted testing program involved conducting nondestructive evaluation of structural continuity of selected concrete section of the pipeline construction housing (enclosure). The basic purpose of testing was to obtain an answer to the question about the scope of possible damages to the concrete housing adjacent to one of the compensators (expansion pipe joint) responsible for ensuring failure-free utilization/functioning of the duct. The task was exceptionally complicated because of the fact that access to the structure examined was only possible through the tilting valve leading into the pipeline through the technical chamber located outside the limiting outline of the reservoir. In the place of testing, the duct is laid at a depth of about 40 m below the water surface, under several dozen meters of deposit layer, which evidently precluded its detection. An additional difficulty in realization of testing was the necessity of ensuring adequate level of safety. Of special importance was the hazard of electric shock which could result from the necessity of measurements being conducted by testers kneeling in the wet interior of the duct with residual water at its bottom which oozed in after closing flow into the duct at the overflow. In order to eliminate this hazard, a special electric line was led into the pipeline through the ventilation channel, to provide safe supply for the measuring equipment. Testing was conducted at three selected measuring locations considered as crucial for safety of the tested structure, marking them on Figure 3 as: • P1 (top surface), • P2 (right side of pipeline), • P3 (left side of pipeline). 186

Figure 5. View of measuring location P2 .

Before proceeding with the measurements, sections of the pipeline assigned for testing, were thoroughly cleaned to remove sediments and dried. View of an example of testing location (P2 ) is presented in Figure 5. Utilized in the testing was the measuring unit DOC-ter consisting of two basic elements: • central processing unit, forming an integral whole with the high class computer adapted for field work, • combined sensor type Mark II facilitating, on the one hand, pulse excitation of elastic waves in the tested concrete element while simultaneously recording displacement of the concrete surface. It is evident that the whole procedure of getting the equipment inside the duct as well as its installation therein and conducting the measurements was very difficult because of the technically implicated conditions at the testing station. However, in spite of these difficulties, it was possible to carry out the testing with success. Before proceeding with the actual measurements, because of lack of possibility of directly measuring the propagation speed of “P” type waves in the tested medium, its value was initially taken at the level of 4000 m/s, and then after analyzing the results obtained, verification was conducted and its value finally assumed at VP = 4210 m/s. 2.4 Principle of the testing method Diagnostics of concrete structures by the “Impact-Echo” method is based on analyzing the effect accompanying propagation of elastic waves in a solid body. The principle of the method is to utilize the reflection of pulse-excited elastic waves from inside surfaces of separation of individual layers (delamination) of the medium occurring in damaged places of structural material, including also its outside surface. Diagram of this effect is presented in Figure 6 (Moczko, 2002). 187

∆t

Force

Time

Time

Time

Of the excitation

Excitation

Converter

T Fault

Figure 6.

Impact-Echo method – Principle of the operation.

From the practical point of view, of basic significance in this case is knowledge of propagation characteristics in concrete of “P” type elastic wave also called primary or longitudinal wave. Propagation of this type of wave in concrete causes slight displacement of its surface which can be recorded in the function of time by means of a sensor specially constructed for this purpose, usually placed in direct proximity of the place of pulse excitation. Suitability of this method for diagnostics of concrete structures results mainly from its “sensitivity” to occurrence of media of different acoustic impedance (product of medium density and propagation speed in it of “P” type elastic waves) in the tested element. Since, at the borders of media of clearly differing acoustic impedance, reflection occurs of propagating elastic waves. This type of situation occurs among others in the case of internal non-continuity of material filled with air or water. As an example, approximate value of acoustic impedance for selected media amounts to: • • • • •

concrete air water rock soil

→ → → → →

7.8–10.4 106 kg/m2 s 411 kg/m2 s 1.48 106 kg/m2 s 4.6–15.1 kg/m2 s 0.28–4.2 106 kg/m2 s.

A characteristic feature of most concrete structures constituting underground infrastructure elements and distinguishing them from other engineering structures is their linear course as well as thin-walled, ring section, characterized by fixed geometrical parameters along the length of the structure. Besides, these objects are usually subject to the action of considerable loading transferred by rock mass or ground surrounding it. For these reasons, the testing methods utilized for defectoscope examination of such structures must take into account their specifics and particularly the influence of reactions on the concrete-and-ground border. Considerable testing/research conducted in USA (Sansalone et al., 1997) showed that shortduration activation of mechanical pulse on the surface of concrete element of ring section, generates propagation of elastic waves in it, whose frequency spectrum approximately equals the distribution obtained while testing elements of slab character by means of Impact-Echo method. The differences observed in this case consist mainly in additional occurrence of frequency amplitude sequences characterizing individual forms of cross section vibrations (Figure 7A) caused by the fact that an 188

Wzbudzenie

Amplituda

fT f1 2 3

4

5 6

(a)

0

10

20

30

Czestotliwosc [kHz]

Osady

fT Wzbudzenie

Amplituda

f1 2

0 (b)

Figure 7.

10

20

30

Czestotliwosc [kHz]

Sample spectral resolutions of vibration amplitudes typical for ring section, expressed in frequency function. A/ – ring section in air, B/ – concrete section surrounded by ground.

element of ring section forms a limited medium for propagation of elastic waves in it. The amplitudes of these frequencies are however distinctly lower than the extreme values, viz. dominating frequency (fextr ). From the practical point of view, they can therefore be neglected and the thickness of undamaged ring section can be determined using the known formula, formulated for slab elements: T = VP / 2 fextr

(1)

In case the tested section is continuous (viz. no faults occurring in its outside to interrupt this continuity), value of frequency accompanying the extreme amplitude is defined as the expected frequency (fT = fextr ) which, for given speed of P (Vp ) type wave, corresponds to thickness of the tested element. Knowledge of the dominating frequency ( fextr ) enables locating relatively easily any possible places of non-continuity occurring in the tested concrete section and to estimate the depth of their locations. Interpretation of results obtained in this case consists in comparing the dominating value in the recorded frequency band with the expected value. Every absence of extreme amplitude for expected frequency ( fT ) in the frequency spectrum obtained signifies the occurrence of fault in the concrete section. We are concerned with a slightly different situation when testing a ring section on which the surrounding ground is acting (Figure 7B): • firstly, values of frequency amplitude are slightly lower because of the fact that part of the energy is absorbed by the ground, • secondly, higher forms of the cross section vibrations do not appear in the recorded frequency band since they are damped by the surrounding ground, and the frequency amplitudes visible 189

Figure 8.

Sample result obtained in the place where concrete section is continuous, extreme frequency (5.9 kHz) obtained corresponds to expected thickness of section equal to 357 mm.

in it, corresponding to two of the first forms of cross section vibrations, are not very clear. In comparison, value of the expected frequency (fT ) is explicitly dominating and its identification does not present any difficulties. Sample spectral resolutions of vibration amplitudes are presented in Figure 7 (Moczko, 2002). 3 TEST RESULTS Unique under Polish conditions, tests performed, limited to selected sections of the structure, showed full suitability of the Impact-Echo method for evaluating the continuity of difficult to access concrete sections typical for underground infrastructure. For illustration purposes, presented in Figures 8 and 9 are sample results obtained in places where: • integrity of tested concrete section was confirmed (Figure 8), • occurrence of its cracking was confirmed (Figure 9). As is concluded from Figure 8, the expected value of frequency conforming to extreme amplitude is 5.9 kHz which corresponds to the actual section thickness amounting to about 357 mm. Whereas the result in Figure 9 illustrates the case of measurement in which it can be accepted with considerable probability that the concrete section is cracked. This is signified by occurrence of individual extreme amplitude for relatively higher values of frequency than that resulting from expected thickness of the same element ( fextr = 8.1 kHz > fT = 5.9 kHz). Lack of dominating value of frequency characteristic for solid material ( fT = 5.9 kHz), with simultaneous absence of other extremes of amplitude in lower frequency band testifies to occurrence of damage that is large enough to practically disable penetration of the wave below the fault and is consequently almost 100% reflected from it. Such cracking therefore has the character of layer separation and should be considered as a significant hazard to safety of the tested section. 190

Figure 9.

Sample result obtained in the place where the concrete section is cracked, extreme frequency obtained (8.1 kHz) corresponds to layer separation type fault occurring to a depth of about 261 mm.

Detailed analysis of results obtained showed that the technical state of concrete housing of the tested pipeline is diversified. In measurement location denoted P3 (left side of pipeline), relatively good state of existing pipeline housing was confirmed. Only at a distance of about 10 cm from the compensator, the tests conducted showed non-continuity of concrete section (probably its cracking). A different situation existed in the measurement locations denoted P1 and P2 . In both these areas, considerable cracking of concrete was ascertained with high rate of probability. In case of measuring location P2 (right side of pipeline), the width of this crack can be estimated as not less than 20 cm and not more than 34 cm, and in case of the top section of pipeline (P1 ), not less than 15 cm and not more than 28 cm, respectively. It should here be stressed that in both cases the cracking most probably starts at the border of the compensator. 4 CONCLUSIONS • Testing the state of underground pipeline reinforced-concrete structure by the Impact-Echo method for the first time in Poland was successful and showed that it can be applied even under extremely difficult local conditions, which had raised the greatest anxiety in the authors. • On the basis of presented results and further testing, decision was undertaken to renovate the pipeline using trenchless technologies. The renovation works have been completed. • Since the time of conducting these tests, the authors have successfully examined underground and other structures by the Impact-Echo method several times, however never under such difficult local conditions. REFERENCES ASTM C 1383–98: Standard Test Method for Measuring the P-Wave Speed and Thickness of Concrete Plates Using the Impact-Echo Method. American Society for Testing and Materials, Philadelphia, PA. 19103, USA.

191

Carino, N.J. & Sansalone, M.J. 1992. Void detection in grouted ducts using the Impact-Echo method. ACI Materials Journal 89(3): 296–303. Moczko, A. 2002. Examination of structures – Impact-Echo (in Polish). Building Industry, Technologies, Architecture (Budownictwo Technologie Architektura) No 1: 48–50. Sansalone, M.J. & Carino, N.J. 1989. Detecting delaminations in concrete slabs with and without overlays using the Impact-Echo method. ACI Materials Journal 86(2): 175–184. Sansalone, M.J. & Streett W.B. 1997. Impact-Echo – Nondestructive Evaluation of Concrete and Masonry. Bullbrier Press, USA. www.kghm.pl

192

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Selected problems of designing and constructing underground garages in intensively urbanised areas H. Michalak Warsaw University of Technology, Poland

ABSTRACT: The paper describes the problems of designing and constructing garages erected in dense urban conditions, in deep trenches with shielding made of monolithic diaphragm walls, propped, according to the principles of the roof method or rigid bracing. Studies on the impact (effects) of deep-embedded new buildings upon ground deformation in the nearby are have been carried out.

1 INTRODUCTION Nowadays, ensuring the appropriate number of parking lots has become one of the most important interdisciplinary issues which require solving in order to ensure the normal operation of towns. Solving the car parking and garaging question in a correct way, especially in intensively urbanised areas, can help town development, while incorrect solutions in this respect may become an issue inhibiting this development. Buildings, including those with multi-storey underground parts destined for garages, should be designed and erected taking into consideration the requirements of currently binding legislation and adopted standards concerning the quality of these structures. At the same time it is necessary to ensure the safety of the existing neighbouring houses. This condition is mainly related to the necessity of limiting the deformations of the ground during the erection of a new, deeply founded, building.

2 EXTEND OF TRENCH IMPACT UPON VERTICAL DISPLACEMENTS OF SOIL MASS Excavation of deep trenches and the subsequent construction of objects cause displacements of the adjacent ground and movement of ground masses in the object’s vicinity. There are different research opinions and results proving a relationship between the extent of trench impact, the magnitude of the vertical displacement of the surface of the ground and lateral displacements of trench shielding, the type of shielding, method of propping, construction method of the underground part of the building, type of soil, lowering of ground water table, etc. (Burland, Simpson & St John 1979; Breymann, Freiseder & Schweiger 1997; Clough & O’Rourke 1998; Michalak, P˛eski, Pyrak & Szulborski 1998; Siemi´nska-Lewandowska 2001; Kłosi´nski 2002; Kotlicki & Wysoki´nski 2002; Michalak 2005; Michalak 2006). It is estimated that the extent of the trench impact upon the vertical displacement of the soil surface in the surrounding area depends, most of all, on the soil type; it is expressed as a factor of the depth of the trench h and equals: • 2 ÷ 4 h – according to Clough & O’Rourke, 1998 in the ground formed of London and boulder clays, 193

• 2 ÷ 2.5 h – according to Symons & Carder, 1992 in the ground formed of London and boulder clays, • 1.5 ÷ 2 h – according to Breyman, 1997 in non-cohesive soils (fine and medium grain size sand and gravel), • 2 ÷ 3 h (maximum 5 h) – according to Simpson, 1979 in strong cohesive soils, • in sand 2.0 h, in clays 2.5 h, and in silts 3 ÷ 4 h – according to Kotlicki & Wysoki´nski 2002; it has also been found that when the ground water table lowers, the extent of the impact is lower by approximately 20%. From the research works (Michalak, P˛eski, Pyrak & Szulborski 1998; Michalak, P˛eski, Pyrak & Szulborski 2001; Wysoki´nski 2002; Szulborski & Michalak 2003; Michalak 2005; Michalak 2006) it follows that the largest vertical movement of ground surface occurs within a zone of 0.5 to 0.75 h fart from the trench edge. The displacements disappear at a distance of 2 h, and when the ground water table is lowered (for depression wells located outside trench outline) it disappears within the distance 3 ÷ 4 h from trench edge. The published data presented here refer to an assessment of the effects of trench excavation, but do not include movements occurring during the erection of the underground part of the structure, followed by the erection of the above-ground part of the building.

3 THE EXTENT OF THE EFFECTS CAUSED BY THE CONSTRUCTION OF A NEW BUILDING WITH A MULTI-STOREY BASEMENT UPON THE VERTICAL DISPLACEMENTS OF THE GROUND SURFACE The excavation of deep trenches and the subsequent construction of objects cause deformations of the adjacent ground and movement of masses nearby. Among other things, they are triggered by: • modification of ground stress and strain condition, related to trench shielding displacement, • ground deformation due to strain relief by a trench (destressing), and load by a new building (Fig. 1), • settling of the ground surface due to a lowered ground water table. These deformations, although inevitably related to the execution of deep foundations, may occur more frequently when irregularities or errors happen to occur during execution. Vertical displacements of the ground surface near the excavations and the extent of the impact of a new building construction mainly depend on: the type of soil, applied trench shielding and the method of propping (spreaders, floors of basement storeys – roof method, injection anchors), the adopted static scheme, execution phases – de-stressing and loading the condition of ground. Vertical ground displacements within the zone adjacent to a newly constructed building are the result of the superposition of the subsequent phases of works covering: constructing a shielding (Berlin wall, pile wall, diaphragm wall etc.), deepening the trench and its successive lagging, constructing a building basement, and the whole structure and its usability (cf. Fig. 1). It is believed that the process of the movement of sandy soil in practice is finished just after the completion of the construction, while in cohesive grounds it lasts even 3 years longer. On average, it can be assessed that in non-uniform soil this process lasts approximately one year after the completion of a construction and the full loading of the new structure with its operational load. From my own surveys and calculation analyses of objects constructed in Warsaw in the last decade (Michalak 2005; Michalak 2006) it follows that the magnitude of the vertical displacements of the ground surface and the shape of the settlement basin mostly depend on the soil type, applied trench shielding and the static behaviour scheme, the construction method of the building basement and the construction phase. The present study covered shielding by diaphragm walls, top-propped and made by the roof method (Fig. 2). 194

0.0 10.5 21.0 31.5 42.0 52.5 63.0 73.5

200.00

190.00

180.00

170.00

160.00

150.00

140.00

130.00

120.00

110.00

100.00

90.00

80.00

70.00

60.00

50.00

40.00

30.00

20.00

10.00

0.00

77.2

(a)

200.00

190.00

180.00

170.00

160.00

150.00

140.00

130.00

120.00

110.00

100.00

90.00

80.00

70.00

60.00

50.00

40.00

30.00

10.00

20.00

0.00

0.6

(b)

Figure 1.

Numerical model of vertical displacements of the ground in the following phases of construction of a building with multi-storey basement: (a) de-stressing by trench, (b) loading by building.

It has been observed that in objects where the basement was constructed by the roof method (from “0” floor level upwards) or when the trench was propped by wedging the struts on the capping beam or pre-bracing from level “0”) lateral movements of trench shielding are insignificant compared to the vertical displacements of this shielding. An area outside the layout of the excavations for the planned building where vertical displacements exceeding ±0.6 mm occur, has been adopted as the zone of impact of a new building upon the displacement of the ground surface and the neighbouring houses located in this area. Such limited value results from the precision of conducting geodetic survey concerning vertical displacements, which can be achieved nowadays. 195

(a)

Figure 2.

(b)

Construction of building basement: (a) roof method, (b) propped diaphragm walls. 0 0,5

1

1,5

2

2,5

3

Y

0

X

Figure 3.

Extent of impact of a new building construction (phase III) founded in sands (Michalak, 2006): Y – factor of vertical displacement value v0 ; X – distance from trench edge expressed as a factor of trench depth h.

Within the extent of the impact of a new building, 4 zones have been set apart S0.75 , S0.50 , S0.25 and S0 with a reach dependent on the absolute value of vertical displacements on the trench edge v0 not exceeding 0.75v0 , 0.50v0 , 0.25v0 and 0, respectively (decline of displacements). The analyses which were carried out considered the following phases of construction: • phase II – corresponding to the execution of the basement structure, • phase III – corresponding to the erection of the building and loading with full operational load. It has been observed that the extent of the trench impact (of zone S0 ) depends on the ground type. The extent of the trench impact corresponding to buildings which are founded in non-uniform 196

1,2 1

Y

0,8 0,6 0,4 0,2 0 0

1

2

3

4

5

6

X

Figure 4.

Scope of impact of a new building construction (phase III) founded in clays (Michalak, 2006): Y – factor of vertical displacement value v0 ; X – distance from trench edge expressed as a factor of trench depth h.

Table 1.

Impact zones of the construction of new buildings upon the vertical displacements of the surrounding ground surface. Zone range

Construction phases

S0.75

S0.50

S0.25

S0

II

0.5 h

0.7 h

1.1 h

1.7 h – sands 5.4 h – silts

III

0.5 h

0.8 h

1.3 h

2.8 h – sands 5.4h – silts

soils with sandy deposits or silts deposited below the bottom slab level has been studied (Figs. 3, 4 and Table 1). From first-hand analyses carried out based on true (measured) lateral and vertical displacement of surveyed buildings, the following conclusions can be drawn as concerns the extent of the impact zones of new buildings on the movements of the ground surface: 1. the biggest vertical displacement of the ground surface of absolute value up to 0.75 v0 occurs within the distance up to 0.5 h in construction phases II and III, 2. the vertical displacement of the ground surface of absolute value 0.75 ÷ 0.50 v0 occurs at a distance of up to 0.7 h in phase II and 0.8 h in phase III, 3. the vertical displacement of the ground surface of absolute value 0.50 ÷ 0.25 v0 occurs at a distance of up to 1.1 h in phase II and 1.3 h in phase III, 4. the decline of the vertical displacement of the ground surface depends on soil type and occurs: • for sandy deposits: within the distance 1.7 h from the trench edge in phase II, and within a distance of 2.8 h from the trench edge in phase III, • for clays: within a distance of 5.4 h from trench edge in both phases, 5. the extent of zone S0.75 does not depend on the execution phase of the building construction, 6. an increase of extent of each zone in phase III compared to phase II has been observed outside zone S0.75 . 197

4 THE RELATIONSHIP DETERMINING THE VERTICAL DISPLACEMENTS OF THE GROUND SURFACE DUE TO THE CONSTRUCTION OF A NEW BUILDING A BASEMENT In order to determine the general relationship describing the vertical displacement of the ground surface in the vicinity of newly constructed buildings with an underground part, constructed in deep trenches, true (measured) values of vertical displacement obtained from a land survey and the vertical displacement value at the trench edge, calculated by the 3-axial strain method have been analysed (Michalak 2006). Calculating the analyses included buildings with a basement constructed by the roof method or by propped diaphragm walls during phase III of the construction. The uniform load of surface the of the elastic half-space of the analysed buildings has been adopted, diminished by the load value corresponding to de-stressing by the trench. In the cases analysed, a ratio linking the true and calculated values of the vertical displacement of the ground surface has been determined by the method of least squares. The empirical ratio obtained from the analyses and approximation of the curve of the vertical displacement of the ground surface applying the square function led to obtaining the sought after relationship for non-uniform soils formed of sandy deposits or clayey deposits below the level of the bottom slab . Based on my own analyses (Michalak 2006) it has been adopted that a function describing the vertical displacement of the ground surface due to the construction of buildings on non-uniform ground, formed within the zone laying below the sandy deposit slab has the following shape:  

 n q∗ ωi B 1 − υi2 x2 x (1) Vy = −0,00883 2 + 0,0482 − 0,0655 h h E0i i=1 where: h – trench depth; x – distance from trench edge; n – number of ground strata; q∗ – uniform load by building, considering de-stressing by trench (under the basement); ωi – coefficient dependent on shape of loaded area (foundation), its rigidity and location related to the loaded area; υi – Poisson’s ratio of ith stratum; B – width of foundation (trench); E0i – modulus of original deformation in ith stratum . In the case of buildings which are founded on clays below bottom slab, the relationship between vertical displacement of ground surface and the erected building can be formulated as follows:

 n x q∗ ωi B(1 − νi2 ) x2 V = 0,00614 2 − 0,0453 + 0,0652 h h E0i i=1

(2)

The specified functions enable the total vertical displacements of ground surface to be forecasted (including ground de-stressing, and later its loading with the new building weight) in the vicinity of buildings with basement parts embedded in deep trenches built by the roof method or by applying propped diaphragm walls. In practice, they can be applied to estimate the vertical movements of the ground in the case of strata similar to the one discussed in the present paper. 5 ANALYSIS OF THE EFFECTS OF THE GROUND DEFORMATION UPON THE TECHNICAL CONDITION OF HOUSING When a building is constructed in a dense urban area it is necessary to limit the lateral movements of the trench shielding for each phase of managing the investment: • investment preliminary phase: it is necessary to investigate the technical infrastructure, • design phase: in static calculations of trench shielding movements it is necessary to design the trench propping ahead of earthworks progress, • execution phase: it is necessary to obey technological regimes. 198

Based on studies (Michalak, P˛eski, Pyrak & Szulborski 1998; Szulborski 1999; Michalak 2005; Michalak 2006) it can be concluded that the most beneficial of the comments on the limiting of ground movement in the neighbourhood, is constructing multi-storey basements in trenches which are shielded by diaphragm walls propped by pre-stressed spreaders or the floors of underground storeys (roof method). Due to the required limitation of lateral displacements these walls should be adequately rigid, embedded below bottom slab level and cross-braced. Propping should be applied on several levels depending on the depth of the trench while limiting lateral displacements of this shielding is the most effective if it proceeds along with the earthworks progress starting from “0” level. It should however be noted that the circular plan of the basement of the building is optimal bearing in mind the limitation of ground movement and type of static interaction. This solution has been applied in France thanks to the development of a new method of cavity deepening and construction of diaphragm walls forming ring shielding, thus eliminating the need to prop the walls. This allows the amount of reinforcing elements of diaphragm walls to be reduced, provides water-tightness in wall contacts and is cheaper than traditional shielding methods in French conditions. Published data (Michalak, P˛eski, Pyrak & Szulborski 1998; Runkiewicz 2001; Kotlicki & Wysoki´nski 2002), engineering practice and results of my own studies (Michalak 2005; Michalak 2006) indicate that a relationship between the type and size of damages occurring in buildings (building facilities), and the type of building structure and ground displacements within a location area exists. Based on analyses and observations, many researchers have characterised empirical limits, usually measured by tilting θ of different types of building structures, the surpassing of which may cause various damages in the furnishings or the structural components of these buildings. From the aforementioned data and my own survey it follows that the damages of buildings (mainly their architectural components), including cracking, can occur when the tilting of a building equals or exceeds the value 1/600 ÷ 1/500, i.e. 1,66 ÷ 2,00 mm/m. In general, the probability of crack occurrence in building components appears when tilting amounts to 1/500 ÷ 1/300, i.e. 2,00 ÷ 3,33 mm/m, and cracking occurs when tilting is equal to 1/300 ÷ 1/150, i.e. 3,33 ÷ 6,67 mm/m. Small and medium structural damages occur usually when tilting equals to 1/200 ÷ 1/150 (i.e. 5,00 ÷ 6,67 mm/m), and heavy damages – when tilting is greater than 1/150 (greater than 6.67 mm/m). Evidently, the values related to specified building damages can differ slightly, depending on the true technical condition of building (its natural wear) during tilting, as well as to its susceptibility to the deformation of materials and structure applied in construction.

6 METHODS OF PROTECTING AND REINFORCING BUILDING STRUCTURE During preliminary construction works for a building with a multi-storey basement the existing housing should be scrutinized within the zone of the planned trench impact (Michalak, P˛eski, Pyrak & Szulborski 1998; Szulborski & Michalak 2003). Depending on the value and character of the ground deformation under the existing buildings (uniform or non-uniform displacements) and the technical condition of these buildings, sometimes it is necessary to protect or reinforce the structural components, namely: (a) reinforcement of the foundation zone, including.: – piling (including micropiling) in order to transfer the load of the building foundation to bearing ground strata laying below the wedge of the block formed in the ground during the deformation of trench shielding (when exceeding limiting strains), – circumferential reinforced concrete ties located on the ground floor level of the building, – high-pressure injection method consisting in increasing the load-bearing capacity of the ground under the foundations (especially, in the case of expected non-uniform settlement of substratum, optional loosening, etc.), 199

(b) reinforcement of the aboveground part of existing buildings, e.g. by steel braces along load bearing walls, usually on several levels, as well as the possible execution of additional elements cross-bracing the structure. Requirements concerning the maximum accepted dislocations and deformations of the existing buildings and ground deformation should be considered when designing a new building. In principle, in designing the trench shielding and conducting the works surpassing permissible wall deformations and ground settlement nearby should be avoided with reference to neighbouring housing.

7 SUMMARY The analysis of almost 50 buildings with multi-storey basements erected in Warsaw in the last decade, allows us to conclude that: • the dislocation of the ground depends on the depth of the trench, type of shielding, strutting, type of soil, phase of construction, • the value of the ground movement can be affected by selecting an appropriate type of trench shielding, and especially strutting, • in the case of constructing a building basement by the roof method or by means of propped diaphragm walls, vertical movements dominate among ground dislocations. The process of ground dislocation ends, depending on the soil type forming the ground; for sandy soil it ends directly after completing the construction and starting operation, while in cohesive grounds this process lasts for up to 3 years more. On average, it is estimated that in non-uniform strata this process continues the next year after completing the construction and after loading a new structure with its operational load. When the basement part is constructed by the roof method or by propped diaphragm walls, the total extent of the impact of a new building, also including constructing its aboveground part, is ca 2.8 h in non-uniform soil with sandy deposits in the foundation area, and ca 5.4 h in clays. REFERENCES Breymann H., Freiseder M. & Schweiger H. F. 1997. Deep excavations in soft ground, in-situ measurements and numerical predictions. Proceedings of the XIV International Conference of Soil Mechanics and Foundations Engineering. Hamburg. Burland J. B., Simpson B. & St John H. D. 1979. Movements around excavations in London Clay. Materiały Konferencji. “Design parameters in geotechnical engineering”. BGS, London. Clough G. W. & O’Rourke T. D. 1998. Construction induced movements of in-situ walls. Proceedings of Conference Design and Performance of Earth Retaining Structures. New York. ˙ nski R. 2005. Wybrane problemy budownGrodecki W., Madryas C., Tajdu´s A., Tokarz A., Wichur A. & Zyli´ ictwa podziemnego. “Górnictwo i Geoin˙zynieria”, Uczelniane Wydawnictwa Naukowo-Dydaktyczne AGH, vol. 3/1, Kraków. ˛ ˛ Kłosi´nski B. 2002. Projektowanie obudów głebokich wykopów. Materiały seminarium pt. “Głebokie wykopy na terenach wielkomiejskich”. IDiM PW oraz IBDiM. Warszawa, 19 November 2002. ˛ ˛ Kotlicki W. & Wysoki´nski L. 2002. Ochrona zabudowy w sasiedztwie głebokich wykopów. ITB, praca nr 376/2002. Warszawa. ˛ S., Pyrak S. & Szulborski K. 1998. O wpływie wykonywania wykopów głebokich ˛ Michalak H., Peski na ˛ zabudowe˛ sasiedni a.˛ “In˙zynieria i Budownictwo”, nr 1/1998. Michalak H. 2006. Kształtowanie konstrukcyjno-przestrzenne gara˙zy podziemnych na terenach silnie zurbanizowanych. Prace naukowe – seria architektura, vol. 2. Oficyna Wydawnicza Politechniki Warszawskiej, Warszawa. Michalak H. 2005. Wybrane problemy projektowania gara˙zy podziemnych na terenach zurbanizowanych. “In˙zynieria i Budownictwo”, nr 11/2005.

200

˛ ˛ Runkiewicz L. 2001. Wpływ głebokiego posadowienia budynków plombowych na destrukcje˛ istniejacych obiektów. Materiały XVI Ogólnopolskiej Konferencji “Warsztat Pracy Projektanta Konstrukcji”. PZITB, Oddział w Krakowie, Ustro´n, 21–24 February 2001. Siemi´nska-Lewandowska A. 2001. Przemieszczenia kotwionych s´cian szczelinowych. Oficyna Wydawnicza Politechniki Warszawskiej, Warszawa. Simpson B. & oth. 1979. Design parameters for stiff clays. Proceedings of the VII ECSMFE, Brighton. Symons I. F. & Carder D. R. 1992. Field measurements on embedded retaining walls. “Geotechnique”, nr 1/1992. ˛ Szulborski K., Michalak H., Peski S. & Pyrak S. 2001. Awarie i katastrofy s´cian szczelinowych. XVI Ogólnopolska Konferencja “Warsztat Pracy Projektanta Konstrukcji”. PZITB, Oddział w Krakowie, Ustro´n, 21–24 February 2001. ˛scia˛ Szulborski K. & Michalak H. 2003. Uwarunkowania realizacji budynków z kilkukondygnacyjna˛ cze´ podziemna˛ w strefach zabudowy zwartej. XLIX Konferencja Naukowa KILiW PAN i KN PZITB “Krynica 2003”. Warszawa-Krynica, 14–19 September 2003. Szulborski K. 1999. Problemy konstrukcyjne w realizacji inwestycji wznoszonych w zabudowie zwartej. V Konferencja Naukowo-Techniczna “Warsztat Pracy Rzeczoznawcy Budowlanego”. ITB, Kielce, 27–29 April 1999. ˛ Wysoki´nski L. 2002. Badania geotechniczne przed i w trakcie wykonywania głebokich wykopów budowlanych. ˛ Conference papers “Głebokie wykopy na terenach wielkomiejskich”. IDiM PW i IBDiM, Warszawa, 19 November 2002.

201

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Material structure of municipal wastewater networks in Poland in the period of 2000 to 2005 K. Miszta-Kruk, M. Kwietniewski, A. Osiecka & J. Parada Technology Institute of Warsaw, Faculty of Environmental Engineering, Water Supply and Wastewater Removal Department, Warsaw, Poland

ABSTRACT: The material structure of wastewater networks in Poland has been changing steadily since 1990. This paper presents actual results of changes in the period of 2000 to 2005. Effects show that the actual trend of systematic increase of PVC material applications. It pertains to both outside networks and sewer house connections. The Waterworks companies take into considerations life expectancy of pipes and sewer house connections as well as durability of joints during the process of matching the material to the sewer networks constructions.

1 INTRODUCTION The material structure of wastewater networks in Poland has been changing steadily since 1990. The trend shown by those changes points to a wide use of products made of thermoplastics, especially a system of pipes and pipe fittings made of PVC. It is so both in case of external networks and house sewers. Developments in application of various material solutions in construction of wastewater networks in Poland have been monitored by Water Supply and Wastewater Removal Department of Technological Institute of Warsaw for at least 15 years in five-year cycles (Kudra 2002, Kwietniewski et al. 1996, 1997, 1999, 2000, 2004). This paper discusses the research results from the most recent period (2000 to 2005). The research results have been combined with the results of the observations made in the preceding periods of time to create a general picture of developmental trends in the use of various material solutions for construction of wastewater networks in the entire 15-year period covered by the observation.

2 OBJECTIVE AND SCOPE OF THE RESEARCH The objective of the research was to assess the developmental trends shown by municipal wastewater networks in Poland in the period of 2000 to 2005. Emphasis was placed on identification of the material structure of the sewers under construction and the criteria applied in selection of materials the sewers used for the network construction purposes were made of. The sources of data included water supply and wastewater enterprises but some information was obtained from municipal offices, as well. In the data collection process it was found that in many enterprises the quality of the wastewater facility and network records continues to be far from the expected level. In many cases many important data required for a well-planned operation and maintenance of those facilities are missing. Nevertheless more and more frequently introduction of GIS-type computer databases is planned or in progress in many enterprises; the databases are an excellent tool which can be employed to organise the stocktaking of underground infrastructure facilities in a structured way. At present every second large enterprise is introducing those databases into their operations. 203

Table 1.

Examined wastewater networks∗ .

Specification

Length (km)

Share (%)

Wastewater networks in Poland Examined wastewater networks

80 130.8 13 216.5

100 16.5

∗ as

at 31 December 2005.

Table 2.

Home connections in the examined wastewater networks∗ .

Specification

Length (km)

Share (%)

Sewer house connections in Poland Examined wastewater networks

17 545.12 3 022.79

100 17.2

∗ as

at 31 December 2005.

Table 3.

Material structure of the wastewater networks examined in 2000 taking into consideration the range of sewer diameters. Length of sewers

Item

Material

Range of diameters mm

1 2 3 4 5 6 7

Stoneware Concrete PVC Reinforced concrete PE Cast iron Other

100 ÷ 1600 150 ÷ 2500 100 ÷ 1400 150 ÷ 2500 80 ÷ 1400 100 ÷ 1200 –

TOTAL

km 5 536.28 2 996.51 1 428.74 1 343.90 296.52 193.00 381.64 12 181.69

% 45.6 24.6 11.7 11.0 2.4 1.6 3.1 100

The research covered 13,200 km of wastewater networks, i.e. 16.5% of the municipal wastewater networks in Poland (Table 1) and over 3,000 km of house sewers, i.e. about 17.2% of all the sewer home connections in the country (Table 2). The examined networks were used by over 33% of inhabitants benefiting from collector-based wastewater networks. The scope of the research was as follows: • Structural analysis of the wastewater networks and home connections under construction in terms of the types of materials used; • Trends in the use of various materials for construction of wastewater networks in the period of 2000 to 2005; • Analysis of the criteria applied in order to select sewer materials to be used for construction of wastewater networks. 3 DEVELOPMENT OF WASTEWATER NETWORKS At the beginning of the period covered by the research stoneware pipes were the dominating material in the wastewater network structure (Table 3, Fig. 1). Sewers made of stoneware (most of them were old-type pipes) accounted for 45.6% of the length of the examined wastewater networks. 204

50% 45.5% 45%

40%

35%

30% 24.6%

25%

20%

15% 11.7%

11.0% 10%

5%

2.4%

3.1% 1.6%

0% Stoneware

Figure 1.

Reinforce concrete

Concrete

PVC

PE

Cast iron

Other

Material structure of the wastewater networks examined in 2000.

Table 4. Shares of accumulative gains in the length of wastewater networks made of various materials in the period of 2000 to 2005. Length of sewers Item

Material

Range of diameters mm

1 2 3 3 4 5 5 6

PVC Stoneware PE Composite materials PP Concrete Structural materials Other∗

90 ÷ 630 100 ÷ 1600 80 ÷ 1400 150 ÷ 1200 100 ÷ 1200 100 ÷ 1400 200 ÷ 1200 –

TOTAL ∗ other

km 1 618 832 172 323 51 685 37 647 21 170 20 034 11 436 48 176 1 993 826

% 81.2 8.6 2.6 1.9 1.1 1.0 0.6 2.4 100

materials: steel, east iron.

As you can see concrete was second most popular material used for construction of wastewater networks (accounted for 24.6% of the network length). The share of pipes made of thermoplastics, mainly PVC (11.7%), and reinforced concrete in the material structure of the network was considerable, too. In this context the following question can be asked: what materials were used for production of pipes that wastewater networks were made of in the period of 2000 to 2005 and what was their share in the total length of the networks built at that time? To answer this question all the materials used for construction of wastewater networks in that period of time have been specified in Table 4 below in an accumulative format, and the global percentage shares of sewers made of individual materials have been shown on Fig. 2. Moreover, Fig. 3 illustrates the development of wastewater networks built using various materials in individual years of the period covered by the research. 205

81.2% 80% 70% 60% 50% 40% 30% 20% 10% 0%

8.6%

Stoneware

PVC

0.6%

1.0%

2.6%

1.1%

Rainforce concrete

Concrete

PE

PP

1.9%

2.4%

0.6%

Composite Structural materials materials

Other

241435

288997

237361

250008

303877

297154

Figure 2. Percentage shares of accumulative gains in the length of wastewater networks made of various materials in the period of 2000 to 2005.

100000 90000 67525

80000 70000 60000

15970 18900

31723 2652 1880 4927 2050 0

10650 13041 0

2987 7477 1920 8161 6765

13983 7917 3336 10428

24429 160

10000

4961 816 8250 8010 500 1790 10578 12625

20000

3362 8527 890 1753 4316 8620 7183 11504

30000

7399 8116 780 7929 4700 3520 11709 19369

40000

29550

50000

0 2000

2001 Stoneware

2002 PVC

Composite materials

Figure 3.

2003

Reinforced concrete & concrete Structural materials

Other

2004 PE

2005

PP

Pressure conduit

Gain in the length of wastewater networks made of various materials in the period of 2000 to 2005.

The data shown on Figures 2 and 3 clearly show that in the period of 2000 to 2005 the wastewater networks were built mainly of PVC pipes. As much as 81.2% of the total network length was made of that material. Other materials were less important in construction of wastewater networks in that period of time. Only the new generation stoneware is worth mentioning in this context: about 8.6% of the examined wastewater networks was made of it. Moreover it is important to point out that PVC pipes continued to be quite a popular construction material for the entire period covered by the observations although a slight downward trend was recorded. The results produced by the most recent (available) research conducted in order to determine the material structure of the wastewater networks [6] are shown on Fig. 4. 206

50% 45%

44.0%

40% 35% 30% 26.6% 25% 20.9% 20% 15% 10% 3.8%

5% 1.1%

0.7%

PE

Asbestos-concrete

1.9% 0.1%

0% Stoneware

Figure 4.

Reinforced concrete & concrete

PVC

Iron

Composite materials

Other

Material structure of wastewater networks in 2003.

Figure 4 shows that the materials that the wastewater networks are made of can be classified in terms of their share in the networks in the following order: 1. stoneware −44.4%, 2. concrete/ reinforced concrete −26.6%, 3. PVC −20.97%, 4. PE −1.06%, 5. asbestos cement −0.71%, 5. cast iron −1.90%, 6. composite materials −0.1%. The share of other materials amounts to about 3.83%. Thus, it is quite clear that the material structure of the wastewater networks continues to be dominated by stoneware but, apart from traditional materials like concrete and reinforced concrete, the share of PVC is rather high, too. Generally, it can be concluded that over 90% of the wastewater networks is made of stoneware, concrete/reinforced concrete and PVC pipes. The results produced by the research point to a substantial diversification in the material structure of the wastewater networks that has taken place in the past few years (Fig. 1 and Fig. 4), namely: • The share of concrete and reinforced concrete pipes has decreased from 35.6% to 26.6% • The share of PVC pipes has increased from 11.7% to almost 21%. The share of the stoneware pipes has remained at the same level, however some traditional stoneware pipes have been replaced by new generation stoneware pipes. The share of the stoneware pipes has remained at the same level, however some traditional stoneware pipes have been replaced by new generation stoneware pipes.

4 DEVELOPMENT OF SEWER HOME CONNECTIONS Among other things, in the course of the research the development of sewer home connections regarded as separate network elements was analysed. Special attention was paid to the material structure and trends in the use of various materials for construction of the connections. The results of the analyses (Fig. 5 and 6) are very similar to those produced by the analysis of the whole network (Fig. 2 and 3). House sewers were built almost exclusively of PVC pipes (95.5% of all the house sewers). High share of the latter showed an upward trend in the examined period (2000 to 2005). 207

Structural materials 0.002%

Other 0.2%

Composite materials 0,002%

Stoneware 3,79%

PE 0,38%

PVC 95,63%

80000

65350

70258

90000

73266

100000

83065

91879

Figure 5. Percentage share of accumulative gains in the length of sewer home connections made of various materials in the period of 2000 to 2005.

55203

70000 60000 50000 40000 30000

5546

1110 0 0 108

302 0 0 184

2244

2826

137 0 10 220

3189

114 0 0 140

0 10 0 172

1962

1614

10000

60 0 0 80

20000

0 2000

2001 Stoneware

Figure 6.

2002 PVC

PE

2003

Composite materials

2004 Structural materials

2005 Other

Gains in the length of sewer home connections made of various materials in individual years.

5 ASSESSMENT OF CRITERIA APPLIED TO SELECTION OF MATERIALS TO BE USED FOR CONSTRUCTION OF WASTEWATER NETWORKS As a result of the research the following types of criteria have been identified. The criteria have been classified according to the weights attributed to them in the process of choosing a product made of a given material for construction of a wastewater network (Table 5). The analysis of the criteria applied to the selection of the materials used for construction of sewers indicates that in general their influence on the choice of the material is not excessively diversified and varies within quite a narrow range of 7.8 to 12.7%. Although no selection criterion 208

Table 5.

Classification of criteria according to their weight (importance in selection of the material).

Item

Criterion

1 2 3 4 5 6 7 8 9 10 11

Durability of pipes and joints Tightness of joints Construction costs Structural strength of the pipes Defectiveness of the sewers Negative influence of wastewater on the material Resistance to corrosion Pipe and pipe fitting purchase costs Easy assembly Availability of pipes and pipe fittings Other

Criterion assessment score∗

TOTAL

230 208 192 191 180 177 172 169 144 142 13 1 818

Share (%) 12.7 11.4 10.6 10.5 9.9 9.7 9.5 9.3 7.9 7.8 0.7 100

∗ Each criterion was assessed on the scale of 5 to 1. 5 – the most important criterion, 4 – very important criterion, 3 – important criterion, 2 – satisfactory criterion, 1 – the least important criterion.

clearly stands out it can be seen that the durability of pipes and joints was taken into consideration most frequently. One can notice that such criteria as: durability, tightness of joints, structural strength, defectiveness, negative influence of wastewater on the material and resistance to corrosion describe one common characteristic of the network well, namely its reliability. The criteria mentioned here were taken into consideration in 63.4% of decisions concerning the selection of the material. Thus, it can be noticed that how to ensure high operational reliability of the network is the biggest problem faced by the operators of wastewater networks. Therefore they assess the usability of the material and structural solutions available on the market for construction of the wastewater networks first and foremost in this context. The investment costs constitute another significant criterion. It is important to point out that such a crucial aspect as easier maintenance and operation of a network built of a given material or the operating costs connected with sewers made of that material has not been mentioned. 6 DEVELOPMENT OF WASTEWATER NETWORKS IN THE PERIOD OF 1992 TO 2005 The results of the ongoing research were supplemented with the results obtained in the preceding periods of time, i.e. 1990 to 1995 and 1995 to 2000, to generate information describing long-term trends shown by the wastewater networks built in Poland. This information is demonstrated in Table 6 and illustrated by the diagrams created on that basis and shown on Fig 7. Analysis of the obtained results indicates that there are two periods that can be distinguished in the development of wastewater networks, namely, Period I from 1992 to 1995 and Period II from 1995 to 2005. In Period I (1992 to 1995) a clear domination of stoneware pipes as a material used for construction of wastewater networks can be seen. At that time almost 50% of all the networks were built of pipes made of that material. But in the following period of time (starting from 1996) PVC pipes were used more and more widely and it is them that had a dominant share in the newly built wastewater networks. The share grew from 43.6% (in 1996) to 89.4% (in 2001). 209

Table 6. Percentage gains in the length of wastewater networks made of various materials in the period of 1992 to 2005. Material

1992

Stoneware Reinforced concrete Concrete PVC PE Composite materials∗ Other∗∗ Total

1994

1995

1996

1997

1998

1999

2000

50.06 55.93 54.94 33.33 27.20 23.06 19.26 10.51 11.80 13.51 14.33 14.46 10.60 4.37 3.20 2.43

2001

8.34 0.50

3.39 0.93

2002 4.28 1.05

2003

2004

2005

6.42 16.76 10.55 0.02 0.84 0.34

33.16 20.03 19.49 19.63 13.04 14.62 10.90 2.50 0.45 0.53 1.46 0.03 2.63 0.55 1.55 9.73 9.67 30.57 43.64 52.67 61.20 79.85 83.90 89.45 84.73 78.63 71.74 80.30 0.00 0.00 0.36 0.41 3.94 3.17 4.23 2.27 2.41 0.24 2.75 8.09 1.97 0.63 0.00 0.14 0.77 0.13 0.52 1.54 0.95 2.07 0.49 2.36 2.69 2.48 2.59 0.68 3.43 100

∗ Glass

1993

0.65 100

0.44 100

1.48 100

1.06 100

0.58 100

0.27 100

0.36 100

3.90 100

3.10 100

3.05 100

4.33 100

3.47 100

6.95 100

Reinforced Plastic – a resin strengthened (reinforced) with glass fibre; grey cast iron, ductile cast iron, structural pipes (of the Duo, Spiro etc. type)

∗∗ Other:

100,0 90,0

PVC

80,0

Increase rate [%]

70,0 60,0

stoneware reinforce concrete concrete PVC PE composite materials other

stoneware

50,0 40,0 concrete

30,0 20,0

reinforce concrete

10,0 PE

other composite materials

05

04

20

03

20

02

20

01

20

00

20

20

99

98

19

97

19

96

19

95

19

94

19

93

19

19

19

92

0,0

Figure 7. Trends in the construction of wastewater networks based on pipes made of various materials in the period of 1992 to 2005.

In this context it is important to point out that the upward trend in the use of the PVC pipes for construction of wastewater networks had been observed since 2001. Since the end of Period II a decreased interest in the use of this material has been recorded, resulting in a slight downward trend. Nevertheless, the share of PVC continues to exceed substantially the shares of other materials used for construction of new wastewater networks. Moreover, the analysis of data presented in Table 6 and on the diagrams (Fig. 7) indicates that the use of traditional materials for construction of wastewater networks, e.g. stoneware, concrete and reinforced concrete, shows a downward trend in the entire period (1992 to 2005) with a slight increase in the use of the new generation stoneware pipes in the period of 2004 to 2005. 210

7 CONCLUSIONS Recapitulation of the 2000 to 2005 research results leads to the following conclusions: 1. Wastewater networks and house sewers are built mainly of PVC pipes (81.2% of the network length); 95.5% of the house sewers (in terms of their length) was built of that material in the analysed period of time. 2. The effects of the wastewater network development in the analysed period of time are as follows: • the share of the stoneware pipes in the wastewater network structure remained at the level of about 44%; • the share of the PVC pipes increased by about 10%; • the share of the concrete and reinforced concrete pipes decreased by about 9%. Analysis of the wastewater network development in the period of 1992 to 2005 shows that the pipes made of PVC had been used quite extensively for construction of the networks since 1996 and their share changed from 43.6% (in 1996) to 89.4% (in 2001) of the newly built network length. REFERENCES Kudra, M. & Kwietniewski, M. & Le´sniewski, M. 2002. Wyniki bada´n zakresu wykorzystania ró˙znych materiałów w rozwoju sieci kanalizacyjnych w Polsce. In Materiały IV Ogólnopolskiej konferencji pt. Nowe technologie w sieciach i instalacjach wodocia˛gowo – kanalizacyjnych, 397 – 409, Ustro´n, Politechnika ´ ˛ska. Sa Kwietniewski, M. & Chudzicki, J. & Goła˛b, A. 1997. Stosowanie materiałów w wodocia˛gach i kanalizacji ze szczególnym uwzgle˛dnieniem tworzyw sztucznych w Polsce i na s´wiecie. In Rynek Instalacyjny nr. 1/1997, 7–16. Kwietniewski, M. & Goła˛b, A. 1996. Tendencje w zakresie stosowania materiałów do budowy sieci wodocia˛gowych i kanalizacyjnych w Polsce i na s´wiecie. In Materiały I Ogólnopolskiej Konferencji Naukowo-Technicznej, Nowe materiały i urza˛dzenia w wodocia˛gach i kanalizacji, 17–26, Kielce-Cedzyna Kwietniewski, M. 1999. Kierunki rozwoju rozwia˛za´n materiałowo – konstrukcyjnych w sieciach kanalizacyjnych. In Materiały VI Seminarium Instytutu Zaopatrzenia w Wode˛ i Budownictwa Wodnego Politechniki Warszawskiej, 204–212, Warszawa Kwietniewski, M. & Zawadzki, J. 2000. Ogólne tendencje w zakresie stosowania rur z tworzyw sztucznych do budowy przewodów wodocia˛gowych i kanalizacyjnych w Polsce ze szczególnym uwzgle˛dnieniem PE. In Materiały konferencji Nowe technologie w sieciach i instalacjach wodocia˛gowo – kanalizacyjnych, 18–28, ´ ˛ska. Ustro´n, Politechnika Sa Kwietniewski, M. 2004. Rurocia˛gi polietylenowe w wodociagach i kanalizacji – rozwój rynku w Polsce i niezawodno´sc´ funkcjonowania. Gaz, Woda i Technika Sanitarna, 70–82, nr. 3/2004.

211

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Two HDD crossings of the Harlem River in New York City James P. Mooney, Jr Consolidated Edison Company of New York, Inc., USA

Jacek B Stypulkowski MRCE, New York City, USA

ABSTRACT: This article discusses the challenges faced by the design team of two under river crossings of electric transmission feeders in an urban area. The under river crossing of a 138 kV transmission feeder was successfully completed through HDD. A second under river crossing of a 345 kV feeder, was initially designed as HDD, but was redesigned as a utility tunnel because of site constraints on both sides of the river.

1 INTRODUCTION 1.1 Site setting New York City is a deep water port connected to various estuarine courses to the east, west, and north. The city is comprised of five boroughs; Manhattan, Brooklyn, Queens, Staten Island, and the Bronx. Manhattan, an island, is the most densely populated part of the city, and is surrounded by the Hudson River to the west and the Harlem/East River to the north and east. New York harbor lies to the south of Manhattan. Utilities supplying Manhattan must cross one of these waterbodies. The electricity supply provided by Consolidated Edison Company of New York, Inc. (Con Edison) is only one of many utilities facing this challenge. Overhead crossings are not practical in this urban environment and have been out of favor with the public for quite some time. Space beneath bridges has been largely taken. Bridge crossings are also less desirable due to difficulty in installation, maintenance, and because of security issues. The only options left are water crossings. There are three types of crossings currently under consideration: direct burial in the river mud, horizontal directional drill (HDD) crossings, and utility tunnels. This article describes two electric transmission feeder crossings of the Harlem River with which the authors have been involved. Both crossings were originally intended to be installed using HDD technology. 1.2 Regulatory framework The potential permitting approvals necessary for the design and construction of any river crossing are numerous. Significant consideration must be given to the permits actually required. A list of potential permits and approvals for HDD crossings and utility tunnels is presented in Table 1. Additionally, direct burial in the river mud would require several years of environmental assessment and the time to perform actual construction is limited. A list of additional permits is presented on Table 2. 1.2.1 Article VII Recognizing the complexity inherent in the permitting process, Article VII of the New York State Public Service Law was established as a review process for the consideration of any application to construct and operate a major utility transmission facility. 213

Table 1.

Permit requirements. Regulation

Activity

Sect. 10 / Sect. 404

Notification of proposed activity Discharge of water generated by dewatering activities

NYSOGS

6 NYCRR Part 750–757 State Pollution Discharge Elimination System 6 NYCRR Part 750 State Pollution Discharge Elimination System; Article 17, Titles 78 and Article 70 of ECL 19 NYCRR Part 600 State Coastal Zone Management Program Article 6 Section 75 Public Lands Law

NYSOPRHP

Article 14 State Historic Preservation Act

Federal Agency USACE State agency NYSDEC

NYSDEC

NYSDOS

Local Agency NYCDCP NYCDCP NYCDBS NYCDBS

Title 62 CCNYRR Ch. 5 and E. O 91 of 1977 as amended in 1991 Section 197-c of the City Charter Section 1301 (2)(c) City Charter Section 1301 (2)(c) City Charter

NYCDPR NYC Landmarks NYC Planning NYCDEP

NYC Landmarks Law NYC Local Waterfront Revitalization Program NYCDEP Noise Control Code

Land disturbance greater than one acre. Stormwater management during construction Project located in New York State Coastal Zone Permission/license for use of lands underwater Potential impact on historic resources City Environmental Quality Review Uniform Land Use Review Procedure Construction on waterfront property Work Notice/Permit for pipeline construction Construction work in parkland tree protection Potential impact to historic resources Construction in NYC Coastal Zone Construction Noise Variance

USACE: United States Army Corps of Engineers NYSDEC: New York State Department of Environmental Conservation NYSOPRHP: New York State Office of Parks, Recreation and Historic Preservation NYSDOS: New York State Department of State NYSOGS: New York State Office of General Services NYCDEP: New York City Department of Environmental Protection NYCDBS: New York City Department of Business Services NYCFD: New York City Fire Department NYCDPR: New York City Department of Parks and Recreation

The NewYork State Legislature enacted Article VII in 1970 to establish a single forum for reviewing the need for, and environmental impact of, certain major electric and gas transmission facilities. Article VII was meant to be a “one-stop shopping” approval, intended to supersede local city, town, county, and other state agency permitting requirements. The law requires that an applicant must apply for a Certificate of Environmental Compatibility and Public Need (Certificate) and meet Article VII environmental and need requirements before constructing any such facility. As established by New York State law, the Public Service Commission (“PSC”), the five-member decisionmaking body that regulates investor-owned electric, natural gas, steam, telecommunications, and water utilities in New York State, is responsible for reviewing and issuing Article VII Certificates. Due to the extensive effort required to satisfy numerous regulatory requirements, the design process usually starts with preliminary concepts and, once the permits are granted, it is followed by 214

Table 2. Additional permit requirements for direct burial. Regulation Federal Agency USACE Department of Commerce Coast Guard State Agency NYSDEC NYSDEC NYSDEC NYSDEC NYSDEC NYSDOS NYSDOH

Dredge and Fill Permit (Clean Water Act, Section 404) Federal Coastal Zone Management Program Review (16 USC,_Chapter_33,_ Section_1451) Docking Approval State Pollution Discharge Elimination System (Environmental Conservation Law, Article 17, Title 8; 6 NYCRR Parts 750 through 757) Water Quality Certification (Clean Water Act, Section 401) Protection of Waters Permit (Environmental Conservation Law, Article 15, Title 15; 6 NYCRR Part 608) Tidal Wetlands Permit (Environmental Conservation Law, Article 25, 6 NYCRR 661) State Facility (Air) Permit (Environmental Conservation_Law,_Article_19;_6 NYCRR_200–317) Coastal Management Plans (Part 600 of Title 19 NYCRR) State Environmental Review Certification for New York Revolving Fund Program (Public Health Law, Sections_1161_and_1162;_21_NYCRR Part 2604)

a detailed design phase. Sometimes particular design details may change the preliminary concept entirely and the additional involvement of permitting agencies is inevitable.

2 SITE DESCRIPTION 2.1 General The Harlem River is a tidal strait in New York City that with Spuyten Duyvil flows eight miles between the East River and the Hudson River, separating the borough of Manhattan from the Bronx. At the time of European discovery, the Harlem River was approximately 900 to 1,000 feet wide in the area of the proposed crossings, as opposed to its current width of approximately 425 feet. A map of upper Manhattan at the time of discovery is shown in Figure 1. Filling on both the Manhattan and Bronx sides has taken place since then to create the current Harlem River shoreline. In heavily urbanized areas where river courses have been altered over the last several hundred years and numerous structures were built and demolished, discovering what might have been left in place is a challenge. Information on buried foundations and other manmade, steel-containing obstructions are very important for HDD technology. Initial planning usually relies on easily available information, such as known utilities and visible structures. 2.2 Willis Avenue Bridge setting The Willis Avenue site is located north of the Willis Avenue Bridge as shown in Figure 2. A search of historical records led to various interesting discoveries. For example, approximately 100 to 200 feet north of the proposed northern alignment, a steel railroad bridge crossed from Second Avenue in Manhattan to the Bronx carrying the elevated Second Avenue Transit subway line. This was demolished between 1951 and 1968. Between 1951 and 1968, the Harlem River Drive highway (“HRD”) was constructed in Manhattan, parallel to the shoreline, and the bulkhead was extended to its current location. At the project site, HRD is elevated and supported on deep foundations to bedrock. As a result, existing pile foundations are the potential obstruction in the path of the proposed crossing location. 215

Figure 1. The Island of Manhattan at the Time of European Discovery.

Figure 2. Willis Avenue Bridge Site.

There are also two existing live 13.8 kV feeder crossings installed by HDD in the 1990’s as well. A retired submarine crossing installed by cut-and-cover in the Harlem River exists south of the proposed crossing. All four proposed crossings are located in the proposed subway tunnel access paths (approximately 30 feet wide as shown in Figure 3) left between the HRD deep foundations and the bulkhead relieving platform deep foundations. These subway tunnel access paths were set aside in the 1950s for ten subway tunnels to carry an underground Second Avenue subway. Fortunately for the design team, these access paths are not proposed for use on any new subway construction based on known NYCT plans. A known waste transfer facility was identified on the Bronx side. The facility is owned by New York State Department of Transportation (NYSDOT), which leases it to Harlem River Yards (HRY). HRY, in turn, leases portions of the site to CSX (Waste Management and Inter-modal), which operates freight trains on the existing tracks. Since relocation of this facility was out of the question, a technology had to be found to satisfy railroad concerns. Fortunately, yard speeds are much lower than on mainline tracks and risks associated with potential track settlements are greatly reduced. On the Manhattan side, the site for the proposed northern alignment is a paved recreational park containing four basketball courts and one handball court along with trees and benches. The site for the proposed southern alignment is a grassy triangle of park land containing trees, bounded on two sides by the Harlem River Drive access ramps, and on the south by 127th Street. The recreational 216

Figure 3.

Proposed Second Avenue Subway Right-of-Way.

Figure 4.

Landmark “Crack is Wack”.

park has a landmark Keith Haring mural on the handball court wall called “Crack is Wack” as shown in Figure 4. 2.3 Willis avenue bridge challenges Drilling fluids were expected to impact the HRY railroad tracks. Downhole erosion caused by drilling fluid erodes soil beneath the railroad tracks, which results in potential settlement. Because the railroad agency required the use of surface casing beneath the tracks, the risk of impact to the tracks was expected to be low. However, during construction, the process of installing the surface casing caused settlement of the railroad track, and the track settled more than five inches. Since the rail cars move at speeds less than 20 mph, it has not caused any operational problems to the facility. As soon as the track instrumentation revealed settlement the track was releveled. Arborists voiced their concern that the existing vegetation in the park may be affected. The nature of the HDD operation combined with the large size of the necessary equipment essentially 217

Figure 5.

Pilot holes in the Park.

destroyed most surface plants, which will have to be replaced once the construction has been completed. Starting from the Bronx was somewhat easier since that site was a paved parking lot. The northern crossing successfully extended through soil below the Harlem River and the HRD (Figure 5). Minor misalignment of the southern HDD path through bedrock despite the very accurate technology that was used for the location of the pilot hole – Paratrack resulted in a problem. The actual drilled path crossed directly under the rock socket supporting HRD foundation. Numerical analysis indicated that the influence of the 30-inch opening on the foundation performance was negligible and the contractor was allowed to continue with reaming the hole to the final diameter. 2.4 Marble hill setting Another site is located in the Inwood section of northern Manhattan Island, and the Marble Hill section of the Bronx in the vicinity of the Broadway Bridge. This crossing is a part of the 9.5-mileslong 345 kV primarily underground transmission circuit, running north-south from Sprain Brook substation in Yonkers to upper Manhattan. Part of the current course of the Harlem River at this location is the manmade Harlem River Ship Canal. The canal runs somewhat south and west of the former course of the river in the general site area, isolating a small portion of New York County (Marble Hill) on the Bronx side of the river (Figure 6). In 1888, work began on the Harlem River Ship Canal to provide a navigable connection to the Hudson River via Spuyten Duyvil Creek. Construction of the channel involved dredging soil across Dyckman’s Meadow and excavating bedrock to provide a 15- to 18-feet-deep channel. At the site area, the channel was excavated through soft alluvial and sand deposits. The Army Corps of Engineers construction in the project area comprised stone-filled timber crib walls at the channel’s edge. Channel work by the Army Corps of Engineers was completed in 1895, and at some point between then and 1911, the railroad was realigned to follow the new channel. The banks are generally lined with rip-rap stone. Notes on the 1911 drawings state that the crib face moved toward the river several inches between 1908 and 1911 just southeast of the northeast pier of the Broadway Bridge and that 10 cantilever timber piles were driven at the face to halt the movements. In 1931, plans were made by the Railroad for the timber crib wall to be reconstructed. Subsidence of the track ballast and slope failures had occurred in the past, prompting concerns regarding the stability of the existing tracks due to proposed construction activities. 2.5 Marble hill challenges The first location was picked along the existing right-of-way and decommissioned low voltage crossing buried in the river mud. Since two large retail stores were built in the vicinity, there was 218

Figure 6.

Harlem River Crossing First Study Site Plan, Options 1 & 2.

no space to layout the HDD crossing. Therefore the design team suggested two technically feasible options (1 & 2) as shown in Figure 6, above. However, permission from all the interested parties still needed to be obtained. A number of considerations led to the location of proposed options, which are explained below. One constraint was the need to cross in bedrock to ensure the stability of the Metro North Railroad (MNR) commuter train tracks. These tracks receive heavy use bringing thousands of daily commuters from north of NYC to Grand Central Terminal. The preliminary geotechnical investigation concluded that Inwood Marble is shallow enough and HDD-friendly geometry was proposed. Other obstructions identified included the original Marble Hill Station. In 1975, the present elevated platform structures were constructed farther north and the original station abandoned. The original platforms were removed. However, foundations of the platforms and elevator pit remained. The platform foundation consisted of anchorage piles and steel cables or steel rods. The steel cables or steel rods span from the platform to the U.S. pier and bulkhead line. In addition, a 54-inch Intercepting Sewer has been identified in the project area. The sewer runs parallel to MNR tracks from Broadway to a pump station located south of the Applebee Restaurant. The invert is located approximately 25 feet below ground surface. Due to the presence of organic soils at this site, this sewer was founded on timber piles. A 48-inch Flat Top Reinforced Concrete Sewer on timber piles was identified under West 225th Street with an invert approximately 17 feet below ground surface. This sewer would interfere with option 1B. Numerous public and private shallow utilities exist in the vicinity of the site, primarily located below the streets and sidewalks. The existing utilities had been constructed over a period of time spanning from the late nineteenth century to the present, and they would most likely require relocation prior to any HDD installation. The HDD launching point was moved further north to allow a drill angle deep enough to bypass the numerous obstructions, settling in the parking lot of the Marble Hill Houses. The Marble Hill Houses are public housing owned by the New York City Housing Authority (NYCHA) and mortgaged by the federal government. Locating the launch point on NYCHA property resulted in additional permitting, environmental, and community related issues. 219

Figure 7.

Harlem River Crossing Second Study Site Plan, Options 1, 2 & 3.

At this stage, all parties were contacted and the first feedback was received from the Con Edison real estate group. Owners of the properties affected appeared very concerned with the proposed combination of the technology and the location of construction activities. This led to another study in the vicinity of the bridge as shown in Figure 7. Alignments 1 and 2 from Figure 7 would require lengthy permitting process associated with the NYC Parks requirements. Therefore Option 3 was chosen as the most suitable for further consideration. Options 1 and 2 would also require a full environmental assessment because of the wetlands. In addition, the HDD would have to miss deck foundations of the apartment building complex on the Bronx side and MNR tracks. With Option 3, the lay down area was limited to parking lots and city streets and was perceived to be a better option at the time. The lay down area on the Bronx side is located between high-rise housing. Any construction activities that cannot close after dark in the residential area create a problem due to NYC noise restrictions. Unfortunately, some of HDD operations once started cannot be stopped, which led to the evaluation of quiet, non-diesel-driven equipment and other noise reduction measures were identified, such as a noise-insulated tent. On the Manhattan side, which is the only side where pipe has to be spliced and laid down before the pull, an elementary school, and numerous businesses with drive-in access are located. Noise from the HDD activities, closing the street and limiting access was considered a difficult task and not practical. A temporary aboveground trestle to elevate the pipe above entrances was considered a feasible alternative. The presence of the elementary school complicated the traffic and access requirements during construction. A workable HDD alignment was found and the contract documents were prepared while waiting for the real estate department to finalize access and easement agreements. The geotechnical profile as well as the proposed HDD feeder routing is shown in Figure 8. The lay down area required access to the parking lot on the Manhattan side, which for years had been a coal and oil storage depot, was previously contaminated with hydrocarbons, and considered a “brown field” site. Environmental liability issues were very difficult to address, which led the design team to a new crossing concept: the utilidor. In addition to the constraints mentioned above, the mitigation costs of different aspects of HDD technology (risks of drilling fluid penetrating into the basements of nearby houses, settlement of underground utilities, etc.) led to the consideration of other alternatives. 220

Figure 8.

Geotechnical profile of the HDD Harlem River Crossing.

3 UTILITY TUNNEL The non-HDD solution required re-evaluation of the layout, since 345 kV pipes have a maximum bending radius of 25 feet. A traditional shaft tunnel layout had to be customized not only to facilitate sweeps but also to provide a space for splicing in the bottom of the shaft or in the tunnel. To determine which tunneling technology could be used, clearance requirements had to be established. A small tunnel shown in Figure 9 could be potentially excavated using micro tunneling technology; however this option would not allow for splicing and repairs. An internal diameter of 7 2 could be achieved inside of a large diameter concrete pipe jacked from the shaft using the micro tunneling boring machine in the front. However, access for inspection was not possible in this confined environment. To facilitate splicing of the 345 kV cable inside the tunnel, a medium-size was developed as shown in Figure 10. Tunnel internal diameters changed to 11 high × 10 wide and could be excavated using NATM technology or D&B in rock. The large tunnel option (19 × 17 ) was briefly contemplated if additional space in the tunnel could have been marketed to the other utilities. This option was not pursued any further beyond initial cost estimates. Utilidors are tunnels housing different utilities usually built in densely populated areas or in areas with catastrophic environmental hazards, like tsunamis. They provide access for maintenance and inspection and they do not require digging to look for problems. Their costs while high initially become competitive if numerous utilities are placed in them and long-term maintenance costs are considered. Looking at the geotechnical conditions presented in Figure 8, both soil and rock crossings should be considered. To check the feasibility of the small layout using micro tunneling technology, the design team looked at a number of configurations as shown in Figure 11. To satisfy the requirements of various facilities, multiple combinations were analyzed resulting in large rectangular shafts about 80 ft deep, 24 to 28 ft wide and 55 ft long. However, due to the presence of a manmade crib wall supporting MNR tracks, soft ground tunneling/micro tunneling was considered by Con Edison to be too risky, and would not have been acceptable to MNR. This study led to the conclusion that tunneling technologies needed to be considered in this case. The proposed Harlem River Tunnel crossing consisted of two shafts, with an inside diameter of 24 feet and approximately 160 feet deep and tunnel about 700 feet long. To take advantage of the economy of scale a tunnel provides, additional facilities were designed to be included in the HRT. The facilities include several additional 345 kV feeders, several 13.8 kV distribution feeders, and a high pressure gas header. The shafts and tunnel were sized to accommodate all the proposed additional Con Edison facilities. The Bronx Shaft will be located in the parking lot adjacent to 221

Figure 9.

Figure 10.

Small utilidor.

Medium utilidor.

Ground surface

Micro-tunnels

Figure 11.

Different launch wall of shafts considered in micro-tunneling.

222

Metro-North railroad tracks. The Manhattan Shaft will be located in the parking lot adjacent to Ninth Avenue. It is expected that the Manhattan Shaft will provide access for tunnel excavation and mucking. The tunnel to carry the new Con Edison facilities will be horseshoe-shaped, waterproofed, and concrete-lined. It will cross under the Harlem River, which has a water depth of between 17 and 25 feet and approximately 450 feet wide in the vicinity of the tunnel. At the time of the article submittal, Con Edison is in the process of negotiating the construction contract with notice to proceed expected in July 2008.

223

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Preliminary design for road tunnels on Trans-European Vc Corridor motorway, section Mostar North – South Border (Bosnia and Herzegovina) I. Mustapi´c, D. Šari´c & M. Stankovi´c Civil Engineering Institute of Croatia, Zagreb, Croatia

ABSTRACT: This paper presents preliminary design preparation for road tunnels on the TransEuropean Vc Corridor motorway, section Mostar North – South Border (LOT 4), which passes through Bosnia and Herzegovina. Total length of above mentioned section is 67,3 km, 14 km being in tunnels (which is about 21% of total length of the section). The challenges faced by the designers were obvious, especially due the fact that the Corridor route mostly passes through difficult, mountainous terrain. Due to this fact, 16 road tunnels of different lengths and cross-sections had to be designed, in a total length of tunnel tubes of approx. 29 kilometers. Therefore it can be concluded that the tunnels represent the key part of the section Mostar North – South Border (LOT 4) of the Trans-European Vc Corridor motorway which passes through Bosnia and Herzegovina. These tunnels were designed in accordance with the state-of-the-art guidelines, primarily the DIRECTIVE 2004/54/EC ON MINIMUM SAFETY REQUIREMENTS FOR TUNNELS IN THE TRANSEUROPEAN ROAD NETWORK. As a result of different soil characteristics, different construction methods were proposed, primarily the NATM (New Austrian tunnel method) and the “CUT & COVER” method.

1 INTRODUCTION The Vc Corridor is included in the TEN transport infrastructure network of South-Eastern Europe and passes in the direction Budapest (Hungary), via Osijek (Croatia), Sarajevo (B&H), to the Port of Ploˇce (Croatia). The length of the Vc corridor route through Bosnia and Herzegovina amounts to about 330 km, and runs in the direction North-South through the middle of the country, in most favorable natural conditions and through the valleys of the Bosna and Neretva rivers. It is expected that the construction of this highway shall be a prime mover of economic activities enabling B&H to be included in the main European traffic streams and global European economic system. Section Mostar North – South Border (LOT 4) of the Vc corridor is 67,3 km long and it is divided into five subsections, with different lengths, passing through various terrains. These sections are: – – – – –

Section 1: beginning of LOT 4 –Mostar North Interchange (9,800 km), Section 2: Mostar North Interchange –Mostar South Interchange (16,450 km), Section 3: Mostar South Interchange –Poˇcitelj Interchange (19,650 km), Section 4: Poˇcitelj Interchange –Med−ugorje (Zvirovi´ci) Interchange (12,100 km), and Section 5: Med−ugorje (Zvirovi´ci) Interchange – end of LOT 4 (9,329 km)

General layout of the section Mostar North – South Border (LOT 4) of the Vc corridor with marked tunnels is presented on the following page. 225

Figure 1.

General layout of the section Mostar North – South Border (LOT 4) with marked tunnels.

1.1 General information about the tunnels on LOT 4 of the Vc Corridor, passing through B&H Tunnels are designed with two tubes and the spacing between tunnel tube axes is min. 25 m. Anticipated tunnel execution technology is NATM (New Austrian Tunnel Method). Tunnels shorter than 500 m are designed with an emergency lane along its total length, and tunnels longer than 500 m are designed with lay-bys at a maximum distance of 1000 m instead of an emergency lane. The width of tunnel carriageway is selected on the basis of requirements for an equal width of traffic lanes as on the open part of the route. Tunnels shorter than 500 m are not mechanically ventilated and the tunnels longer than 500 m are ventilated with longitudinal reversible ventilation. On every lay-by, there is a transversal passage for vehicles. Transversal passages for evacuation are placed at maximum distance of 500 m. 226

Figure 2. Two-lane tunnel without invert.

Figure 3. Two-lane tunnel with invert.

Figures 4 and 5. Three-lane tunnel with emergency lane, without invert and with invert.

In tunnels with larger longitudinal carriageway slopes, transversal pedestrian passages are placed at maximum distance of 250 m. SOS recesses are spaced near the portals and at maximum distance of 150 m and are equipped with TPS and manual fire alarm with two fire apparatuses for initial extinguishing. Hydrant network is designed in tunnels longer than 500 m. All the tunnels are equipped with lighting system. Tunnels longer than 2 km are equipped with video and audio system and tunnels longer than 1 km are equipped only with radio system. 2 CROSS-SECTIONS OF TUNNEL TUBES 2.1 Types of cross-sections used There are three types of tunnel cross sections on the objective section, depending on their length, provided technology of execution, as well as on the number and width of traffic lanes on the open part of the route. Two- and three-lane tunnels are designed, taking into consideration that on the part of the route where there is an additional lane for vehicles, three-lane tunnels are somewhat larger in profile in relation to the designed three-lane tunnel applied in cases of tunnels with emergency lane (shorter than 500 meters). 227

Figures 6 and 7. Three-lane tunnel with additional lane, without invert and with invert.

According to the above mentioned terms, most frequently used are the three-lane tunnels with an emergency lane. There are 7 tunnel like this (Komi´c, Vijenac, Osoje, Gorica, Kiˇcin, Šunja Glava, Bijela Vlaka), total length of nearly 5 km, followed by two-lane tunnels, of which there are 5 (Orlov Kuk, Debelo Brdo, Oštri Rat, Samac, Kvanj), but their total length is approximately nearly 20 km. After this, we have 4 three-lane tunnels with an additional lane (Rudine, Koˇcine 1, gallery Koˇcine 2, Rožni Kuci), total length of nearly 4 km. All the above mentioned tunnels are designed with two tunnel tubes except Koˇcine 2 which has a single carriageway in a gallery and the other on an open route. Depending on the chosen cross-section type, we have different clearance surface areas. Profile clearance is 59,13 m2 for two-lane tunnels, 83,90 m2 for three-lane tunnels with emergency lane and 94,30 m2 for three-lane tunnels with additional lane. Such cross-sections completely satisfy demands for profile clearance as defined by the “Bylaw on basic requirements for public roads and their elements outside settlements regarding traffic safety”, as well as the requirements for profile clearance defined by the Austrian RVS guidelines and TEM standards. Moreover, the mentioned cross-section allows for the accommodation of all the necessary devices and equipment, and enables aeration by means of horizontal ventilation. The speed limit of 100 km/h must be provided , as required by the design. Width of tunnel carriageway is selected on the basis of requirements that the same width of traffic lanes and marginal strips should be maintained as on the open part of the route. Total width is 8 m for two-lane tunnels, 10,75 m for three-lane tunnels with emergency lane, and 11,75 m for three-lane tunnels with additional lane. Inspection lanes are elevated from the traffic surface of the tunnel by 15 cm. Total width of inspection lanes is 90 cm, for two-lane tunnels and 100 cm for three-lane tunnels. Under the inspection walkways there are channels for installation of the required tunnel facilities. The channel on the inner side is used for placing the hydrant network, while the channel on the outer side serves for supply and telecommunication installations. Power supply installation ducts should be executed with the cross fall of 2% towards the edge where the channel discharge is located. The tunnel cross section is rotated following the cross fall of the carriageway. All three types of cross-sections are shown below: 2.2 Lining Tunnel lining is made of concrete C25/30. Its minimum thickness is 30 cm for two-lane tunnels and 40 cm for three-lane tunnels. Generally, the lining is not reinforced, except near recesses, in the rock mass of V class zones (for three-lane tunnels lining is slightly reinforced in the base calotte 228

part in the rock mass of IV class zones), on lay-bys, transversal passages for vehicles which are equipped with TS substation recess, at points of connection with transversal pedestrian passages and passages for emergency vehicles and in portal zones. Also, the foundation threshold is to be reinforced along the entire tunnel length. Between the tunnel lining and primary tunnel support a permeability layer is provided by a PVC foil protected by geotextile. The PVC foil is provided with signaling layer, made in one piece, 2 mm thickness. The insulation is laid on the geotextile layer, minimum weight of 500 g. Tunnel insulation is provided along the entire length of the tunnel and on the entire calotte surface and tunnel flanks, and it shall be executed after the excavation and the execution of the primary tunnel security, and after the settling of possible displacements in the primary lining of the tunnel tube.

3 TUNNEL DRAINAGE The tunnel drainage consists of a centrally laid sewage pipe, 500 mm diameter, while carriageway drainage is done by means of a hollow curb system which collects water from the carriageway surface. Considering the characteristics of rock mass, the tunnel passes through a karts terrain, creating a possibility of significant water inflow, which needs to be collected and directed by a special pipeline to the discharge at a suitable place within the tunnel (cave) or out of the tunnel into the terrain. At the tunnel flanks, a RAUDRIL drainage pipe of 150 mm diameter is provided for collection of hill water and its drainage into the central sewer. In order to allow access and inspection of the side drainage pipe, recesses are designed where drainage manholes shall be situated. Discharge from these manholes to the sewer manhole is executed by PVC pipes of 150 mm diameter. Distance between manholes amounts to approximately 50 meters. Carriageway drainage in the tunnel refers to liquids deriving from liquid (inflammable) outflow during a traffic incident and requires a carriageway drainage system which ensures the drainage of incident liquids, with a intake capacity of 200 l/s at carriageway length of 200 m. In order to meet the above mentioned conditions, hollow curb is provided with a continued horizontal opening of reinforced concrete C30/37 of 30 cm diameter. The hollow curb is manufactured as a precast element which is integrated onto the prepared base on the foundation threshold. Siphon outlets are executed with a spillway dam in order to prevent spreading of fire in the sewer. The siphon outlet is executed of reinforced concrete C30/37. The liquid from the siphon outlet is taken by PVC pipes, 20 cm diameter, into the sewer manhole. The distance between siphon outlets is approximately 100 meters. The main sewer system is made of impermeable PVC pipes of 500 mm diameter. The sewer in the two lane tunnel is located in the carriageway center, in three-lane tunnels with an additional lane in the carriageway centerline and in three-lane tunnels with emergency lane, the sewer is situated in the emergency lane. The carriageway liquids, water collected through side drainage system and water from the drainage of pavement structure sub-base layer are taken into the sewer through manholes. Manholes are provided at mutual distance of approx. 50 m. Manholes are executed as monolithic constructions of reinforced concrete C25/30 or as precast, with precast reinforced plate and cast iron cover of 600 mm diameter, of 400 kN bearing capacity

4 TUNNEL VENTILATION Tunnels are designed with longitudinal ventilation. Tunnels with a total length less than 500 m are ventilated naturally. Tunnels from 500 m to approximately 800 m length are designed according to demand that ventilation system must be able to reach the air flow speed of min. 1,5 m/s. 229

Ventilation of tunnels longer than 800 m is designed according to the requirement that ventilation system must be able to reach critical speed of 3 m/s. Sensors for air flow velocity are designed, so that management of the ventilation system could function properly. Also there are CO concentration and rigid particles sensors. Minimum 2 pieces are required, also one CO sensor and one fire detection sensor in each fire zone.

5 FIRE ALARM SYSTEMS Fire alarm systems are designed in tunnels longer than 500 m. Its main purpose is fast and reliable fire detection. Except the above mentioned it has to satisfy the following requirements: – – – –

Selective alarm activation Communication with surveillance and management center Initialization of actions regarding to initial fire extinguishing Evacuation of people

At detection level, system is executed with use of two kinds of communicators. Sensor cable is placed through total length of main tunnel tubes and spot communicators are placed in SOS recesses, TS and UPS rooms.

6 ELECTRICAL INSTALLATIONS Lighting is designed in all of the tunnels as well as in portal zones. The designed lighting system is as follows: adjustable, basic tunnel lighting and anti-panic lighting. Following installations are connected on reserve power supply: – – – – – – – – –

recess lighting ventilation measuring devices safety tunnel lighting anti-panic lighting fluorescent SOS devices and SOS distribution facilities variable traffic signalization traffic signalization, lighting and remote guidance controllers and warning devices fire alarm system video system

Also, there is a low-voltage power supply unit for supplying and management of electro ventilation system.

7 PORTAL STRUCTURES Generally, portal structures are designed as structures extracted from the tunnel tube. The designed length of the portal construction is 12 m. Portal construction is provided from reinforced concrete C25/30. Lining thickness amounts to minimum 60 cm. A drainage system must be made near the tunnel flanks, with pipes RAUDRIL DN 150 type, wit a discharge into the manholes at tunnel portals. A channel must be executed at the approach cutting face. The portal construction waterproofing is provided in the part where portal construction is cut and covered with excavation material. Waterproofing is executed with PVC foil. 230

UPS

SOS

94 300 400 400 94

94

4000

H

400 400 94

170 350

R6

85

500

94 400 400 94

5

UPS

SOS

H

Figure 8.

4000

400 300 94

350 170

2500

4690

94 400

H

REGRADA

TRAFO

750

R68

385

Lay-bys and transversal passage for vehicles.

8 LAY-BYS AND TRANSVERSAL PASSAGES 8.1 Lay-bys Lay-bys are designed as surfaces 40 meters long and 3 meters wide. They are used for stopping broken-down vehicles, so they are equipped with SOS devices. Except that, in every lay-by, a surface for accommodation of UPS and hydrant is designed. 8.2 Transversal passages for vehicles Transversal passages for vehicles are used as an evacuation route in case of fire, and for redirecting traffic in case that one tunnel tube is impassable or closed. They are located opposite to the lay-bys and are equipped with substations. Transversal passages for vehicles are divided with fire walls. On every fire wall there is a slide door for vehicles with dimensions 4,5 m * 5 m and two glass doors for pedestrians with dimensions 100 * 220 cm. Doors and walls are fire resistant for 90 minutes. 8.3 Transversal passages for emergency vehicles Transversal passages for emergency vehicles are used for evacuation of people from one tube into another and for access of emergency vehicles at place of the incident. They are situated at distance of no more than 500 m. Transversal passage for emergency vehicles has profile clearance of 3,6 m * 3,5 m. They are divided with fire wall. On every fire wall there is a slide door for emergency vehicles with dimensions 3,6 m * 3,5 m and two glass doors for pedestrians with dimensions 100 * 220 cm. Doors and walls are fire resistant for 90 minutes. 8.4 Transversal passages for pedestrians At sections with larger longitudinal slopes, transversal passages for pedestrians were designed. They are placed halfway between passage for emergency vehicles and passage for vehicles. 231

220

220

350

300

360

100

100

Figure 9. Transversal passage for emergency vehicles.

100

100

Figure 10. Transversal passages for pedestrians.

They allow evacuation of tunnel users from one tube into another. Designed profile clearance is 250 * 225 cm, and at every end of the passage there is a fireproof bulkhead with built-in round door for pedestrians, dimensions 100 * 220 cm, with fire-proof glass dimensions 40 * 100 cm. Doors and bulkheads are fire resistant for 60 minutes. 8.5 SOS recesses SOS recesses are situated at maximum distance of 150 m. They are equipped with TPS, manual fire alarm and with two fire apparatuses for initial extinguishing of fire. SOS recesses have dimensions 240 * 225 * 130 cm. They are closed with fire-proof bulkhead. On bulkhead there are doors that have dimensions 210 * 70 cm, and they are glassed with glass dimensions 40 * 100 cm. 8.6 Hydrant recesses Hydrant recesses are designed with dimensions 240 * 225 * 100 cm, and they are situated at distance of approximately 125 m. By the design, they are placed near every transversal passage and halfway between two transversal passages. 9 EXECUTION METHODS Chosen method of execution is NATM (New Austrian Tunnel Method). CUT & COVER method is used at lower thickness of overburden. 9.1 Cut & cover method Following tunnels were designed to be executed using Cut & cover method: – – – –

right tube of tunnel Komi´c (0,457 km) part of right tube of tunnel Rudine (app. 0,132 km) part of right tube of tunnel Orlov kuk (app. 0,080 km) gallery Koˇcine 2 (0,120 km)

After the first temporary cutting is excavated, slopes must be protected. After that it is necessary to execute channels on the flanks of a working cut. 232

Cut & cover tunnels are designed to be executed from reinforced concrete C25/30. Thickness of lining is minimum 60 cm. It is necessary to execute drainage down the flanks. After the tunnel lining is concreted, in open cut, it is necessary to execute hydro isolation that has to be protected. Backfilling is executed in layers with maximum thickness of 50 cm, and compacts with appropriate compacting utilities.

9.2 NATM (New Austrian Tunnel Method) NewAustrianTunnel Method is an excavation method which is very flexible regarding often changes of geological and geotechnical conditions. Following tunnels were designed to be executed using NATM: – – – – – – – – – – – – – –

left tube of tunnel Komi´c (0,500 km) Tunnel Rudine (cca. 0,415 km) Tunnel Orlov kuk (cca. 4,797 km) Tunnel Gorica (0,804 km) Tunnel Kvanj (5,355 km) Tunnel Kiˇcin (0,748 km) Tunnel Bijela Vlaka (0,988 km) Tunnel Debelo brdo (2,096 km) Tunnel Vijenac (0,217 km) Tunnel Osoje (0,270 km) Tunnel Oštri rat (5,648 km) Tunnel Samac (1,267 km) Tunnel Koˇcine 1 (0,753 km) Tunnel Rožni kuci (2,404 km)

10 TUNNEL SUPPORT SETS Designed support sets as well as recommendations of measures on excavation and stabilization of tunnel cuttings, were chosen based on experience acquired from previous designs in carbonated rocks. It will be shown for following rock mass class zones: – – – –

rock mass of II class zone rock mass of III class zone rock mass of IV class zone rock mass of V class zone

10.1 Basic tunnel support set type II Support is used in basic rock mass of class II according to the geo-mechanical classification (RMR = 61–80), and contains the following support elements: – crown shotcrete of 5 cm thickness, systematic anchoring with adhesion bar anchors, corrugated steel Ø 25 mm, 3.0 m long, spaced at 2.5 m. – walls without support. 233

10.2 Basic tunnel support set type III Support is used in basic rock mass of the class III, according to the Geo-mechanical classification (RMR = 41–60), and contains the following support elements: – crown shotcrete of 10 cm thickness, steel welded mesh Q 131, systematic anchoring with adhesion bar anchors, corrugated steel Ø 25 mm, 3.0 m long, spaced at 2 m. – walls shotcrete 5 cm thick. 10.3 Basic tunnel support set type IV Support is used in basic rock mass of the class IV, according to the Geo-mechanical classification for RMR = 21–40, and contains the following support elements: – crown shotcrete of 15 cm thickness, steel welded mesh Q 131, systematic anchoring with adhesion bar anchors, corrugated steel Ø 25 mm, 3.0 m long, spaced at 1.7 m (if necessary with self-drilling injection anchors type IBO R25N Ø 25/14 mm). – walls shotcrete 10 cm thick steel welded mesh Q 131, systematic anchoring with adhesion bar anchors, corrugated steel Ø 25 mm, 3.0 m long, spaced at 2.0 m (if necessary with self-drilling injection anchors type IBO R25N Ø 25/14 mm). 10.4 Basic tunnel support set type V Has been provided fro integration in the fault and fracture zone sin the V class rock mass (RMR < 20), and it consists of the following support elements: – crown shotcrete of 20 cm thickness, two steel welded meshes Q 221, systematic anchoring with self-drilling injection anchors type IBO R32N Ø 32/18.5 mm, 4.0 m long, spaced at 1.4 m, truss girders, Pantex 95/20/30 spaced at 1.0 m. – walls shotcrete of 20 cm thickness, two steel welded meshes Q 221, systematic anchoring with self-drilling injection anchors type IBO R32N Ø 32/18.5 mm, 5.0 m long, spaced at 1.4 m, truss girders, Pantex 95/20/30 spaced at 1.0 m. – invert shotcrete of 20 cm thickness, two steel welded meshes Q 221 Considering the fact that there is no initial stability of ground excavations in the class V rock mass, the progress shall be made by fore poling of the steel ribs (corrugated steel Ø 25 mm, 4.0 m long) spaced at 30 cm, over the truss girders. Steel ribs need to be fore poled only in the zone of tunnel crown. 234

11 CONCLUSION Complexity of section Mostar North – South Border (LOT 4) of the Vc corridor is visible from a fact that from total length of 67,3 km, as much as 21% (14 km) is in tunnels. Because of this we can say that tunnels are one of the key parts of the objective section from aspect of financing, building, using and maintenance of the motorway. Apart from given TEM standards, these tunnels were designed in accordance with the state-ofthe-art guidelines, primarily the DIRECTIVE 2004/54/EC ON MINIMUM SAFETY REQUIREMENTS FOR TUNNELS IN THE TRANS-EUROPEAN ROAD NETWORK. According to given directives, special attention was given to safety standards. REFERENCES “Directive 2004/54/EC on minimum safety requirements for tunnels in the Trans-European Road Network”, European Parliament and Council, Bruxelles, 2004. Preliminary design corridor Vc motorway Mostar North – South border (LOT 4), designs of tunnels and structures, Civil Engineering Institute of Croatia, Zagreb, 2006. RVS 9.281, RVS 9.282, “Strasse und Verkehr” (FSV), Wien, 2002. “Regulations on basic conditions to be met with by public roads and their elements outside towns from the point of view of traffic safety”, Croatian Ministry of sea, transport and communication, Zagreb, 2001.

235

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Mapping the underworld to minimise street works C.D.F. Rogers Department of Civil Engineering, University of Birmingham, Edgbaston, Birmingham, UK

ABSTRACT: This paper seeks to describe the genesis, progress and outcomes so far from the Mapping the Underworld project, a government research council-funded project, jointly supported by UK Water Industry Research, that seeks to develop techniques to locate and map all of the buried pipes and cables that make up our utility service infrastructure without resorting to excavation. This is a ‘grand challenge’ that stems from a long-term vision, developed in 1996 between the UK water and gas industries, and has developed via a programme of drawing academia and the relevant stakeholders together to collaborate. This process of academe-industry cooperation is explained in the context of a 25-year vision of research and development that will ultimately benefit road users and society more generally.

1 INTRODUCTION 1.1 The problem and its ownership Most utility services, including electricity, water, gas and telecommunications, are distributed using buried pipelines or conduits, or via directly buried cables, and the majority of this buried utility infrastructure exists beneath roads. Trenching is usually required whenever they need maintenance, repair or extension and this often causes disturbance (and sometimes damage) to other utility services, delays to traffic and/or damage to the environment. Inaccurate location of buried pipes and cables results in far more excavations than would otherwise be necessary, thereby creating a nuisance and increasing the direct costs of maintenance to the service providers, yet greatly increasing the costs to others, the single most important being the enormous direct cost of traffic delays to business and direct and indirect costs to private motorists. These ‘social costs’ of congestion in the UK alone are estimated to be as high as £5.5 billion per annum (McMahon et al. 2005), 5% of which is attributed to utility works. There are also very considerable environmental ‘costs’ due to traffic congestion, a significant proportion of the damage to the planet deriving from vehicles that are delayed. Nevertheless utility service providers, who are under enormous pressure from the regulators to improve performance in all sorts of ways and minimise costs to customers, retain the open cut approach and accept the inconvenience of ‘dry holes’ as a marginal cost addition. There is an enormous benefit to be gained by accurate service location, but the impact is felt by business, society and the environment. Responsibility for funding the research necessary to bring this about rests with the organizations that work on behalf of society (i.e. governments), and in the UK this responsibility lies with the Engineering and Physical Sciences Research Council (EPSRC) since it acts as the conduit for technologically-based government-funded research; EPSRC is the UK body that needed to be convinced of the need and value of the enabling research. This government responsibility is now being accepted in several countries as the environmental pressures to reduce traffic congestion become ever stronger. 1.2 Aims of the paper This paper aims to describe the development of the Mapping the Underworld project in the UK, but it attempts to place the development in a global context of considerable complementary research 237

and development. While accepting that technologies are constantly improving, it aims to make clear the limitations of current location technologies when deployed sequentially, to outline best practice given these limitations and to explain a 25-year vision that will streamline these practices to minimise the disruption currently caused. More particularly, it will outline the genesis and progress so far of research that aims to address the key barriers preventing fast and effective location practices, and explains in detail proposed research to create and deploy a multi-sensor device that would facilitate remote location of all buried utility services without the need for proving excavations.

2 CURRENT PRACTICE 2.1 The need for trenches Utility services provide the basis for modern civilized living, clean water provision, waste water removal and energy (i.e. electricity, gas) supply being considered to be essential minimum requirements in many countries. Add to this the ever-growing demand for telecommunication cables and the need for service connections to each individual building along these supply networks, and the potential for congestion in the utility service ‘corridors’ becomes apparent. These utility service ‘corridors’ typically consist of the underground space beneath the streets between buildings, above which there is vehicular and pedestrian traffic. The underground space is restricted further by the need to service the streets, by street lighting cables, traffic light cabling, highway drainage, and so on. In many countries, the essential service systems have been in place for many decades, and even centuries, and the buried services have deteriorated such that they have been repaired or abandoned to be replaced by alternative service lines. Fitting this buried infrastructure together beneath the streetscape is problematic in itself, such that the ideal pattern of linear, parallel services with sets of parallel sub-perpendicular lateral connections at different depths does not exist. This is the legacy that current engineers face when dealing with buried utility services. When working in the ground beneath the streets, for whatever reason, it is clearly necessary to establish what is present before using mechanized or manual excavation. Shallow surface geophysical technologies provide the obvious solution, but they only provide an absolute solution if they are able to detect, and identify, everything that is present, and if the goal is to add to the congestion of the underground space thereby reveal the places where there is nothing present. A subsequent section of this paper will demonstrate that there is currently no absolute solution, so the best that can be done at present is to use shallow surface geophysics to provide as much information as possible and thereafter to proceed with caution while excavating locally to prove the results of the geophysics (whether this is the specific intention of the excavation or whether the excavation is for the primary purpose of working beneath the streets). Many new technologies are being developed to make working in the ground beneath the streets less disruptive, either by avoiding occupation of the surface or by making the operations more efficient to reduce the time required for occupation. However most of these technologies, such as the enormous number of processes that fall under the umbrella of trenchless technologies, rely upon an accurate knowledge of the sub-surface and often cannot be used with adequate certainty of avoiding damage to what is already buried. Quite apart from this, the utility service companies responsible for installing and maintaining the pipelines and cables buried beneath the streets often view direct excavation as a cheaper option than adopting less disruptive technologies, and fall back on the argument that without a sure knowledge of the make up of the underground space they have no option but to excavate. This tension is nicely expounded by Farrimond & Parker (2008). 2.2 The benefits of accurate location practices Procedures for working in the streets differ in different countries, partly as a result of historical legacies and the different owners of the buried utility services. The most coherent procedure is 238

perhaps the Subsurface Utility Engineering (or SUE) process adopted in the US (ASCE, 2002), if only because it makes it clear that the quality of information provided depends upon the level of investment made in the survey. SUE seeks to deliver one of four levels of service, only the highest level of which will guarantee to provide comprehensive information on the utilities buried beneath the survey area. This highest level of surveying service is only reached, however, following the use of local proving excavations to back up the prior desk studies, walk-over surface surveys and sequential deployment of shallow geophysical techniques. Without proving excavations, no guarantee can be given, and this accurately sums up the state of surveying technology development. If the SUE process is deployed fully and accurately, it has been shown markedly to reduce the costs of subsequent construction through accurate planning and mitigation of construction risks. For example, case studies at Purdue University (US DoT, 2000) and the University of Toronto (Osman & El-Diraby, 2005) showed that sums of $4.62 and $3.41, respectively, were saved in avoided costs for every $1.00 spent on SUE. These findings are consistent with the experience of Geotechnical Engineers, who consistently advocate an appropriate level of spending on and planning of site and ground investigation in order to avoid unnecessary construction costs arising from unforeseen circumstances. SUE is, in effect, an intelligent form of specific site investigation and long-term data capture. These figures make sense in direct economic terms as far as the contactors are concerned, but as our appreciation grows of the wider costs of the work that is carried out daily in the streets, it becomes even more cost-effective because of benefits to society and the environment. Governments are advocating a more sustainable approach to work carried out on their behalf and thus sustainability cost accounting, which uses a ‘triple bottom line’ approach in which the direct economic costs of work are balanced by social and environmental costs (see e.g. Hunt & Rogers, 2005; Hunt et al., 2008). Such accounting makes prior detailed surveying even more compelling, but user confidence in the results of the surveys is of paramount importance. 2.3 Capabilities of current utility location practices Current ‘locating’ techniques fall into two broad categories: the simple devices are strictly limited in their target detection capability and used immediately prior to excavation by site operatives, while the more sophisticated techniques are deployed, and the results require interpretation, by specialist contractors. Controlled trials carried out by UKWIR demonstrated that even with these sophisticated detection techniques, detection rates are often poor. These trials revealed that even in the best of scenarios (i.e. most helpful ground conditions), the detection rate in urban areas with many different types of utilities is rarely much better than 80% (Ashdown, 2001). Recent research effort to address this problem has focused on improving Ground Probing Radar (GPR) for utility detection. For example, a recently-completed EU project (GIGA), seeking to improve GPR performance, resulted only in an 81% success rate, and even then provided a caveat that such success could only be achieved in certain types of ground conditions, and a new EU project (ORFEUS, see www.orfeus-project.eu; Manacorda et al., 2007) is seeking to make further improvements. However, the problem is one of adequate signal penetration in wet soil and hence the application of GPR in ground conditions such as saturated clay soils militates against traditional surface GPR ever being adequate alone for buried utility detection. Although there have been some impressive recent GPR developments, for example those presented at the recent International Conference on Ground Probing Radar (GPR, 2008), they have been for different applications (such as land mine detection) and therefore have different targets; they avoid the essential problem of deep penetration into saturated ground. However, it should be noted that current GPR thinking solely adopts a ‘surface looking downwards’ approach. The only exception is an in-sewer GPR system, but this is for condition assessment of the sewer pipe itself and detection of voids immediately behind the sewer wall, and is restricted to man-entry sewers. It is clear that, even accounting for recent advances made with GPR systems, GPR will never completely solve the problem of utility location for all types of utility infrastructure and ground conditions, and hence other technologies must be used in parallel with GPR to detect every buried service. 239

Several other techniques are available (Costello et al., 2007; Deb et al., 2001; Metje et al., 2007), including acoustics and a variety of electromagnetic techniques. Recent research completed under the first phase of the EPSRC-funded ‘Mapping the Underworld’ location project (MTU Phase I), has demonstrated that all such techniques offer considerable potential to locate individual types of service in different ground conditions. However, they all suffer the same drawback in that, when deployed in isolation, even when deployed in sequence by expert operators without time constraints, they do not provide an adequate solution to identify all types of buried infrastructure and each type of system has its own specific limitations. This finding is confirmed by the fact that US Subsurface Utility Engineering (ASCE, 2002) states that the only method of guaranteeing the location of all buried utilities is by carrying out local proving excavations, in their case using vacuum excavation techniques. There is therefore a need to develop a multi-sensor device that combines the outputs of different complementary technologies in order to maximize the likelihood of detecting every buried utility service.

2.4 A 25-year vision This was recognized as the ideal scenario as far back as 1996, when the term ‘bodyscanner in the street’ was coined. It derived from a meeting between members of the UK water and gas industries, who recognized the problem and the potential impacts that it might have on their business, but equally recognized that the primary beneficiaries of the considerable research and development necessary to bring about such a device would not be the water and gas industries themselves. The adverse impacts to the utility industry stemmed from UK discussions of means of reducing traffic congestion and included such features as lane rentals for utility work, adding a substantial direct cost to utility work and penalizing inefficient practices arising from poor prior utility detection. The conundrum thus presented itself of an urgent need to avert a threat, but an inability to justify to the water and gas companies the expenditure on R&D that would do so. Thus the situation remained of an acknowledged 25-year vision, the overall framework for which is presented in Figure 1, but an uncertainty about how the vision could be realized. Telling the boards of the companies that there was a corporate social responsibility would raise the response that there should be some form of corporate responsibility for funding the solution.

Water / Gas Industry Vision: Bodyscanner in the Street

MTU Project - Location - Mapping - Data integration - Asset tags - Network

UJKWIR – AWWARF – KIWA 3-Day Workshop

1996

2000

UKWIR Commissions Location Trails

2004

MTU Location Project: MultiSensor Device Generation, Assessment, Protocols

2008

Ideas Factory: ‘Mapping The Underworld’ term coined

NETTWORK Workshop: Underground Mapping Pipeline Location and Condition Assessment

Figure 1.

Streetworks become more sustainable - road occupation minimised - night surveys - trenchless installtion / replacement / rehabilitation - congestion reduced

Location of Underground Plant and Equipment Initiative Minimising Streetworks Disruption

Project VISTA advances Mapping and Knowledge Management elements of MTU

2012

2016

Assessing the Underworld: Creating Multi-Sensor Device for Remote Assessment Monitoring of Asset Condition

Overview of the minimisation of street works disruption initiative.

240

2020 . . . and more sustainable forms of utility service provision researched

3 REALIZING THE VISION – THE MAPPING THE UNDERWORLD PROJECT 3.1 Genesis of mapping the underworld It has been established in Section 1.1 that it is the responsibility of governments, acting on behalf of the societies that they govern, to fund the necessary research to reduce traffic congestion, but equally it has been established in Section 2.4 that there should be cooperation from the utility industries in supporting such an initiative. However there are very many demands on a government’s purse and lobbying from all sides for funding; the government must be sure that it is investing wisely. Although the need for better means of utility detection and location was evident, quantification of the scale of the problem and the benefits that would derive from the considerable necessary investment, and indeed the size of the necessary investment itself, was not. Consequently UKWIR commissioned the trials reported in Section 2.3 (see Figure 1 andAshdown, 2001) to demonstrate the limitations of the then current locating technologies. A government-funded EPSRC Engineering Programme Network in Trenchless Technology (NETTWORK, see Rogers et al., 2004), which brought academic and industrial stakeholders together via a series of five workshops to establish the research needs in the broad topic area of trenchless technology, concluded at its first workshop that accurate utility detection was absolutely crucial to the growth of the trenchless technology industry and delivery of the enormous benefits that trenchless technology could bring (see Rogers et al., 2002). Subsequently UKWIR convened a three-day international workshop jointly with the American Water Works Association Research Foundation (AWWARF) and the Dutch equivalent (KIWA) of academics and (mostly) industrialists to scope out a set of projects that would deliver solutions to the problem now and in the future (see Figure 2 and Burtwell et al., 2003). This provided the final piece of the evidence base needed to convince the UK government, and EPSRC acting on its behalf, that funding should be devoted to the topic. At that time EPSRC was developing a novel means of awarding funding to the UK academic community to engage in large, multi-disciplinary challenges, such as that presented by the buried utilities problem, known as an IDEAS Factory, or sandpit (see www.epsrc.ac.uk). An IDEAS Factory involves a highly multidisciplinary mix of academic and industry participants, who are invited following a process

A

B

C

D

Making the best of what we have

Improved surface based survey techniques

Below ground survey techniques

Doing things differently

1

User need requirements

Development of a multi-system location tool

Mapping pig

Smart pipe technology

2

Data collection & Exchange standards

Improving GPR

Through wall in pipe visualisation

Asset tagging

3

Best practice for existing technologies

New technologies for buried assets location

Keyhole access

Novel appraches to underground infrastructure

4

Real costs of buried assets

5

Better regulation

Asset avoidance for Directional drilling

Figure 2. The minimising streetworks disruption research programme (after Farrimond and Parker, 2008).

241

of peer selection, are set the challenge and, over a period of several days, develop a series of research projects via facilitated workshop events. This process aims to drive lateral thinking and result in novel methods for the formulation of solutions to research problems. EPSRC chose the buried utility location challenge for its first IDEAS Factory and coined the term Mapping the Underworld (MTU) as its title. 3.2 The Mapping the underworld project and its developments The MTU sandpit identified four complementary research projects: – BuriedAsset Location, Identification and ConditionAssessment using a Multi-SensorApproach. This aimed to assess the feasibility of a range of potential technologies that can be combined in a single device to determine the location, and where possible identification, of buried pipes and cables (condition assessment was rapidly considered a step too far for now). Advances have been made in GPR technologies (Shan et al., 2006), acoustics (Muggleton & Brennan, 2006), low frequency electromagnetics (Lim & Atkins, 2006), the influence of soil and ground conditions on the resulting data (Thomas et al., 2007), and the potential for combination of these technologies into a coherent operating system (Rogers et al., 2008b). – Mapping and Positioning. This aims to develop a reliable positioning system, with an accuracy of approximately one centimeter, effective even in city streets containing high-rise buildings (so-called ‘urban canyons’, see Ogundipe et al. 2008; Taha et al., 2008; Roberts et al., 2007). – Knowledge and Data Integration. This project is investigating the construction of a unified database of all the location data stored by the various utility companies operating in the UK, together with new data generated daily from utility surveys, hence providing a common means for data sharing (see Beck et al., 2007). Constructing such a database is a particular challenge owing to the current incomplete, and partially inaccurate, state of records, some of which are not even in digital form, and inconsistent methods of data storage between companies. – Enhanced methods of detection of Buried Assets. This aims to develop methods of improving the visibility of underground pipes when surveyed from the ground surface using electromagnetic techniques. A series of ‘resonant labels’, or RFID tags, have been developed (Hao et al., 2007). These are relatively simple metallic structures that could be encapsulated within a new pipe prior to installation. They would provide an effective means of reflecting electromagnetic signals at predetermined frequencies and are expected to lead to cost effective methods of labeling new pipes, or repairs to existing pipes, so that they can be effectively located, when required, in the future. In addition, EPSRC also funded a MTU Engineering Programme Network that, like NETTWORK, aimed to bring together industrialists and academics to further develop knowledge and debate research needs in the broad field of buried utility location, mapping and condition assessment, as well as serve to co-ordinate the four MTU research projects. The mapping and data integration elements of MTU spawned a follow-on project entitled VISTA, which is worth £2.3 million and is funded jointly by the UK government (see www.vistadtiproject. org) and industry. VISTA aims to bring together existing paper and digital records with data from satellite- and ground-based positioning systems to formulate the means of creating a threedimensional electronic map of buried utilities. This is a pressing goal, given the requirement of the UK Traffic Management Act that all utility providers be able to exchange digital utility positions by June 2008. Most recently, a full research programme has been funded to build on the outcomes of the feasibility study of the multi-sensor location device and this is described hereafter. In a parallel development the Americans have recognized the compelling need for action in this area and have awarded two research contracts under their Strategic Highway Research Program (SHRP). The first is entitled Strategies for Integrating Utility and Transportation Agency Priorities in Highway Renewal Projects and is being led by IFC International. The second is entitled Encouraging Innovation in Locating and Characterizing Underground Utilities, is being led by Professor 242

Ray Sterling at Louisiana Tech University and is much more closely aligned to the Mapping the Underworld initiative (see Sterling et al., 2008). This international activity complements the EUfunded research referred to earlier, which like the MTU project (see Rogers et al., 2008a) was also presented at the recent GPR2008 conference. Specific interest in extending the scope of the work in India and Australia following the GPR2008 conference proves the assertion that the problem and its solution are truly of international importance.

4 MAPPING THE UNDERWORLD – MULTI-SENSOR DEVICE 4.1 Introduction Phase 2 of the MTU Location Project, worth ∼£3.5 million, has been funded by EPSRC to research in detail a multi-sensor device that can detect all buried pipes and cables, specifically its creation, assessment and protocols for its use (see www.mappingtheunderworld.ac.uk). The research seeks to utilize every possible advantage to see through the ground and focus on the targets, and is necessarily a multi-disciplinary initiative combining academics from the Universities of Birmingham, Bath, Southampton, Leeds and Sheffield and some 40 project partners and supporters. The project consists of eight work packages and these will be described below. 4.2 Ground penetrating radar The aim of this work package is to advance GPR technologies specifically in support of the multisensor device, for deployment both alone from the surface and when combined with in-pipe GPR transmitters/receivers. Novel arrays of GPR antennas and their integration with the multi-sensor arrays, will be researched to increase the probability of target detection and accuracy of location information (plan and depth). The anticipated deliverables of this work package include recommendations on antenna deployment strategies for surface and in-pipe deployment and creation of specifications and a prototype in-pipe device that meets current standards for safe, practical application in sewers. 4.3 Acoustic technologies The aim of this work package is to determine the effectiveness and limitations of acoustic technologies for buried utility service location, based on both pipe and ground excitation techniques, when operated alongside the complementary sensor technologies in the multi-sensor device. The anticipated deliverables include the design and construction of a prototype acoustic sensor for stand alone use and as part of the multi-sensor device; a comparative performance assessment between the outputs and usefulness of geophones and scanning laser technology; operational protocols when deployed in isolation and in the multi-sensor device; and methods to mitigate the potential adverse effects of sensor interference when deployed with other technologies. 4.4 Low frequency electromagnetic field technologies The aim of this work package is to conduct four inter-related studies based on low frequency electromagnetics: – To attempt to recreate the results achieved in the laboratory in the feasibility study under MTU Phase I (see Lim & Atkins, 2006) when testing in the field, and thereby make the transition from laboratory to field application. – To develop techniques that will distinguish discrete objects, including relatively short linear objects, from continuous (utility service) targets, and thereby minimise the number of ‘false alerts’ registered. 243

– To explore the efficacy of using streaming potentials to detect buried utility services in which flow occurs, and to examine whether the type of flow, and hence the type of utility service line, can be identified. – To explore the efficacy of utilising breaks in the integrity of services, such as insulation breakdown failures and perforations in the structure or coatings of pipelines, and also the leaks from the system thereby caused, as a means of detecting, and potentially identifying, buried utility services. The anticipated deliverables include an improved prototype low frequency electromagnetic detection device for mapping the underworld (particularly small, near-surface services), both as a stand alone device and when incorporated into the multi-sensor device; a new methodology to predict insulation breakdown failures, and therefore the potential to detect leaks; and a matrix of operational capability of low frequency electromagnetic field systems. 4.5 Magnetic field technologies The aim of this work package is to utilize a passive array of magnetic sensors together with advanced signal processing techniques to detect underground electricity cables and other metallic buried infrastructure, even when stacked or laid in close association, and to develop the technique so that it can be integrated in the multi-sensor device. The anticipated deliverables include a new finite element model to analyze magnetic fields and to locate ‘metallic’ buried infrastructure, particularly when in complex, close association; design and construction of a prototype sensor to be tested alone and in conjunction with the multi-sensor device; and a matrix of the operational characteristics of magnetic field technology for implementation within a multi-sensor head device. 4.6 Intelligent tuning of the device to the ground and targets This work package aims to develop equipment for use in field testing of geophysical soil properties and, in conjunction with the British Geological Survey (BGS), to create a Knowledge Based System (KBS) that will allow geophysical soil data to be predicted using geographically mapped geotechnical and geological data. Both are intended to inform, and significantly improve, utility location during the planning, site operations and data interpretation phases of site surveys. The deliverables from this work package include a KBS for application with the new multi-sensor device; a set of soil suitability maps for the four geophysical techniques, based on the KBS and associated geographical software model, for a number of case study sites; and a set of test methodologies, and associated apparatus, suitable for providing data on soil geophysical properties in the field, to further optimize the KBS and multi-sensor device, at survey locations where access to soil can be obtained. 4.7 Intelligent data fusion from the multi-sensor device and statutory records The aim of this work package is to fuse geo-referenced information from multiple sensors and to combine this with an integrated database of buried asset records to increase confidence in their presence and location, and to determine missing asset records. The anticipated deliverables include techniques to resolve differences in resolution, positioning and depth sensitivity for the sensor types and the available buried asset records; techniques to compute spatial correspondences between interpreted sensor readings and utility records; and techniques to determine whether a sensor has located an unrecorded asset or whether a recorded asset is undetected. 4.8 Creation of multi-sensor device, deployment strategies and signal processing The aim of this work package is to develop a multi-sensor array demonstration unit, with bespoke software to integrate data processing, fine-tuning to ground parameters and supply of data in an appropriate form for data visualization, by bringing together the outputs of all of the previous work packages, which are in each case specifically focused on supporting the device’s creation. 244

The anticipated deliverables include a surface multi-sensor array device that may be used to ‘map 100% of the underworld’ at shallow depths; an in-pipe device for use in tandem with the surface device; protocols for the use of the devices; and a means of data fusion that will result in a cross-sectional probability map of the likelihood of an asset being present. 4.9 Proving trials and specification of national MTU test facility The aim of this work package is to conduct a comprehensive and rigorous programme of proof tests of the above technologies both when applied in isolation and when combined in the multi-sensor device, and prove the efficacy of accurately tuning the devices to different soils and groundwater conditions. A further objective is to refine the set of criteria compiled under MTU Phase I for a UK National Test Facility. The deliverables will include objective reports on the performance assessment of both existing technologies and the prototype technologies being developed by MTU; and development of a specification and construction guidelines for a new UK National Test Facility including layout and best practice for operation. 5 CONCLUSIONS There are several essential conclusions that can be drawn from the experience of bringing about an ambitious research programme that addresses the complex problems of how to locate and record the position of buried utility services buried beneath the streets of our urban areas. Accurate detection, location and recording of the position of buried utility services delivers enormous benefits to those working in the streets and, importantly, to society in general since the consequence of inadequate utility location is traffic congestion and its manifest adverse consequences. Those representing societies, i.e. governments, should fund the necessary research to facilitate accurate recording of utility locations, albeit with the full cooperation of utility service providers. This is what is happening in the UK. The case for the research, though obvious in a qualitative sense, needed making quantitatively and this was done by independent trials of equipment to prove current capabilities and via academicindustry workshops to define precisely the research needs and the outline projects that would address these needs (these were valued at more than £10 million in 2003; research funding approaching this level has now been secured – see Farrimond and Parker, 2008). A multi-sensor location device is an essential component of the solution to the problem, but a feasibility study was necessary to prove the case for the undoubtedly ambitious research that the creation of such a device requires; funding of a full research programme in the absence of the feasibility study would almost certainly not have succeeded via the peer review processes that, rightly, guide the award of government-funded research. The Mapping the Underworld project would welcome the collaboration of any person or organization that is interested in advancing its aims. Contact details are available on the mapping the Underworld web site (www.mappingtheunderworld.ac.uk). ACKNOWLEDGEMENTS The authors gratefully acknowledge the financial and other support provided by the UK’s Engineering and Physical Sciences Research Council (EPSRC) and UK Water Industry Research (UKWIR). REFERENCES American Society of Civil Engineers. 2002. Standard Guideline for the Collection and Depiction of Existing Subsurface Utility Data. ASCE 38-02, American Society of Civil Engineers, USA.

245

Ashdown. C. 2001. Mains Location Equipment – A State of the Art Review and Future Research Needs. UKWIR Report 01/WM/06/1, UK Water Industry Research, London, UK. Beck, A.R., Fu, G., Cohn, A.G., Bennett, B. & Stell, J.G. 2007. A framework for utility data integration in the UK. In Coors, V., Rumor, M., Fendel, E.M. & Zlatanova, S. (Eds) Urban and Regional Data Management – Proceedings of the Urban Data Management Society Symposium (Stuttgart, Germany, October 10–12, 2007), Taylor and Francis, 261–276. Burtwell, M., Faragher, E., Neville, D., Overton, C., Rogers, C.D.F. & Woodward, T. 2003. Locating Underground Plant and Equipment – Proposals for a Research Programme, UKWIR Report 03/WM/12/4, UKWIR, UK Water Industry Research, London, UK. Costello, S.B., Chapman, D.N., Rogers, C.D.F. & Metje, N. 2007. Underground Asset Location and Condition Assessment. Tunnelling and Underground Space Technology 22 (5–6): 524–542. Deb, A.K., Hasit, Y.J., Williams, J.A. & Jacob, R. 2001. New Techniques for Precisely Locating Buried Infrastructure. AWWA Research Foundation, Denver, Colorado, USA, 157 pp. Farrimond, M.S. & Parker, J.M. 2008. The Importance of Seeing Through the Ground – A Utility Perspective. Proc. of 12th International Conference on Ground Penetrating Radar (GPR2008), Birmingham, UK, 16–19 June. GPR 2008. Proceedings of 12th International Conference on Ground Penetrating Radar, June 16–19, 2008, Birmingham, UK. Hao, T., Burd, H. J., Edwards, D. J. & Stevens, C. J. 2008. Enhanced Detection of Buried Assets. Proc. of Loughborough Antenna and Propagation Conference (LAPC 2008), March. Hunt, D.V.L., Lombardi, D.R., Rogers C.D.F. & Jefferson I. 2008. Application of Sustainability Indicators in Decision-Making Processes for Urban Regeneration Projects. Engineering Sustainability, Proceedings of the Institution of Civil Engineers 161(ES1): 77–91. Hunt, D.V.L. & Rogers, C.D.F. 2005. Barriers to Sustainable Infrastructure in Urban Regeneration. Engineering Sustainability, Proceedings of the Institution of Civil Engineers 158(ES2): 67–81. Lim, H.M. & Atkins, P.R. 2006. A Proposal for Pipe Detection Using Low Frequency Electric Field. Proc. of 1st International Conference on Railway Foundations (Railfound 06), Birmingham, UK, 11th–13th September, 84–93. Manacorda, G., Scott. H., Rameil, G. & Pinchbeck, D. 2007. The ORFEUS Project: a step change in Ground Penetrating Radar technology to locate buried utilities. European Forum Gas (EFG) 2007, Paris, 12–13 September. McMahon, W., Burtwell, M.H. & Evans, M. 2005. Minimising Street Works Disruption: The Real Costs of Street Works to the Utility Industry and Society. UKWIR Report 05/WM/12/8, UK Water Industry Research, London, UK. Metje, N., Atkins, P.R., Brennan, M.J., Chapman, D.N., Lim, H.M., Machell, J., Muggleton, J.M., Pennock, S.R., Ratcliffe, J., Redfern, M.A., Rogers, C.D.F., Saul, A.J., Shan, Q., Swingler, S.G. & Thomas, A.M. 2007. Mapping the Underworld – State-of-the-Art Review. Tunneling and Underground Space Technology 22(5–6): 568–586. Muggleton, J.M. & Brennan, M.J. 2006. The use of acoustic methods to detect and locate underground piping systems. Proc. of the IX International Conference on Recent Advances in Structural Dynamics, 17–19 July, Southampton, UK. Ogundipe, O., Hancock, C., Taha, A. & Roberts, G.W. 2008. The Use of High Sensitivity GPS for Mapping Sub-surface Utilities. In Proceedings of European Navigation Conference 2008, Toulouse, France. Osman H & El-Diraby TE. 2005. Subsurface Utility Engineering in Ontario: Challenges & Opportunities. Report to the Ontario Sewer & Watermain Contractors Association. Centre for Information Systems in Infrastructure & Construction, Dept. of Civil Engineering, University of Toronto, Canada. Roberts, G.W., Hancock, C., Ogundipe, O., Meng, X., Taha, A. & Montillet, J-P. 2007. Positioning Buried Utilities Using an Integrated GNSS Approach. Proc. of the International Global Navigation Satellite Systems Society (IGNSS) Symposium 2007, University of New South Wales, Sydney, Australia, 4th–6th December. Rogers, C.D.F., Chapman, D.N. & Costello, S.B. 2002. Report on Asset Location and Condition Assessment. UKWIR Report 02/WM/12/1, UK Water Industry Research, London, UK. Rogers, C.D.F., Chapman, D.N. & Karri, R.S. 2004. UK Engineering Network in Trenchless Technology (NETTWORK). Proc. of 12th Int. Conf. on Plastic Pipes, Milan, Italy, April. (CD ROM). Rogers, C.D.F., Zembillas, N., Metje, N., Chapman, D.N. & Thomas, A.M. 2008a. Extending GPR Utility Location Performance – The Mapping the Underworld Project. Proc. of 12th International Conference on Ground Penetrating Radar (GPR2008), Birmingham, UK, 16–19 June.

246

Rogers, C.D.F., Zembillas, N., Thomas, A.M., Metje, N. & Chapman, D.N. 2008b. Mapping the Underworld – Enhancing Subsurface Utility Engineering Performance. Proc. of Transportation Research Board 87th Annual Meeting, Washington D.C., USA, January 13th–17th. Shan, Q., Pennock, S.R. & Redfern, M.A. 2006. GPR for Mapping the Underground. Proc. of 11th International Conference on Ground Penetrating Radar (GPR2006), Ohio, USA, June. Sterling, R.L., Anspach, J., Allouche, E., Simicevic, J. & Rogers, C.D.F. 2008. Encouraging Innovation in Locating and Characterizing Buried Utilities for U.S. Transportation Projects. Proc. of 12th International Conference on Ground Penetrating Radar (GPR2008), Birmingham, UK, 16–19 June. Taha, A., Kokkas, N., Hancock, C., Roberts, G.W., Meng, X. & Uff, J. 2008. A GIS Approach to GNSS Simulation in Urban Canyons. In Proceedings of European Navigation Conference 2008, Toulouse, France. Thomas, A.M., Metje, N., Rogers, C.D.F. & Chapman, D.N. 2007. Soil Electromagnetic Mapping for Enhanced GPR Utility Location. Proc. of 25th International No-Dig Conference and Exhibition, Rome, Italy, 9th–12th September. (CD ROM). US Department of Transportation (DoT). 2000. Cost Savings on Highway Projects Utilizing Subsurface Utility Engineering. Federal Highway Administration and Purdue University, Publication No. FHWA-IF-00-014.

247

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Assumptions for optimization model of sewage system cooperating with storage reservoirs D. Sły´s & J. Dziopak Department of Infrastructure and Sustainable Development, Rzeszów University of Technology, Poland

ABSTRACT: The paper presents assumptions and theoretical fundamentals of an optimization model, called SEW, which shall then be developed further to a software tool for conducting a multi-criterion analysis of available investment options of sewage system cooperating with storage reservoirs. Objective function for the optimization task, as well as decision variables of the developed model are presented.

1 INTRODUCTION Storage reservoirs are the elements of the majority of modern sewage systems. They are used for hydraulic relief of storm water sewage systems, waste water treatment plants within combine sewage systems and waste water recipients. Storage reservoirs, located before outlets to surface water recipients, are often used in combined and separate sewage systems for control of waste water outflow to recipients and treatment processes (Huebner & Geiger 1996, Jacopin et al. 1999, Michelbach & Weiß 1996). Storage reservoirs located before waste water treatment plants regulates waste water flow and equalizes its composition, enabling the reduction of quantity and frequency of storm water discharges from sewage systems and stabilization of functioning parameters of waste water treatment plants. In case of reservoirs located within storm water and combine systems of canalization, which have to lighten the burden on sewage system and reduce its diameters by transformation of short storm inflows of high intensity into inflows of greater longevity and lower intensity, the treatment process of waste water not always is favorable. So in many cases this process is renounced and waste water is treated directly before its discharge to surface waters. It is dictated by economic and exploitation reasons and the necessity of simplifying of sewage sludge management. From the point of view of reservoirs projecting within sewage system for its relief, the establishment of number and localization of these objects is very important task. These two factors are closely connected with following project parameters of storage reservoirs and sewage system: storage reservoir’s accumulation capacity, waste water flow reduction, maximum filling of reservoir, canals, slope, its underground level and diameters of canals. These parameters are the deciding factors of economical effects of storage reservoirs’usage within sewage systems. Taking into account the significance of problem and possible investment’s savings during the building of storm water transport system, the researches were carried out in order to elaborate the project methodology for sewerage relieving storage reservoirs comprehensively considering the choice of their optimal number and localization within sewage system. The investment options analysis, which is the subject of the researches, is based on the elaborated soft-ware instrument SEW and simulation programs of storage reservoirs’ functioning within sewage system (Sły´s & Dziopak 2006, Sły´s 2006). 249

2 GETTING STARTED PURPOSE FUNCTION OF OPTIMIZATION MODEL SEW The task of optimization is to find the set of decisive variables x = {x1 , x2 , x3 ….xp }, which maximizes or minimizes the purpose function. In case of sewage systems’ optimization from economical point of view, the purpose function is under minimization and can be described by Equation (1): min F(x) = Kc x

(1)

where Kc = total investment cost; x = decisive variables. Researches of sewage systems’ optimization, co-acting with storage reservoirs, were began in Poland by Dziopak (1997, 2006), who in his optimization model considered the following decisive variables: filling level, sewage system length and waste water flow reduction. Optimization model of sewage system SEW, elaborated by the author, aimed to the establishment of optimal number and localization of storage reservoirs in complex sewage systems, considers additionally the influence of canals’ slopes before and behind the reservoirs, as well as the influence of average diameters on economical efficiency. The purpose function for this model id described by Equation (2): F (idz , iwr , hm , Ln , βi ) = Kc (2) min (idz , iwr , hm , Ln , βi )

where idz = canal slope behind the reservoir; iwr = canal slope before the reservoir; hm = maximal filling by waste water in gravitational part of reservoir; Ln = distance from the beginning of sewage system section to reservoir; βi = coefficient of flow intensity reduction for storm water in reservoir. The decisive variables in optimization model SEW are the following: – coefficient of flow intensity reduction for storm water in reservoir βi which varies: β1 , β2 , . . ., βn in the limits of 0–1; – the distance from the beginning of sewage system sector to storage reservoir Ln defined by the values of L1 , L2 , . . ., Lk ; – maximal filling by waste water in gravitational part of reservoir hm , with following values h1 , h2 , . . ., ho ; – canal slope behind the reservoir idz which has the following sequence of values: id1 , id2 , . . ., idy ; – canal slope before the reservoir iwr with following values: iw1 , iw2 , . . ., iwt . The number of optimization task Z in accordance with the number of combination possibilities for variables can be determined from Equation (3): Z =n·k ·o·y·t

(3)

where n = accepted number of possible values of flow reduction coefficient β; k = accepted number of possible localizations of storage reservoir within sewage system section, o = accepted number of possible maximum fillings of reservoir by waste water; t = accepted number of possible slopes of canal before the reservoir; y = accepted number of possible slopes of canal behind the reservoir.

3 ALGORITHM OF SEW OPTIMIZATION MODEL CALCULATION FOR SINGLE SEWAGE SYSTEM SECTION In case of single sewage system section’s analysis the calculation model SEW provides the usage of elaborated method for establishment of optimal number of storage reservoirs located within single sewage system section. First step of optimization calculations for single sewage system section is the determination of optimal canal’s slope iopt and its diameter D without storage reservoir. Total investment cost is Kc1 considering the cost of exploitation or not. 250

Next step is the optimization in order to determine the most economically favorable storage reservoir localization Lopt . To achieve this aim it was necessary to calculate the following hydraulic parameters and total investment costs connected with them for individual storage reservoirs’ localization L1 , L2 , . . ., Lk within analyzed system section. The calculations have to be provided for particular values of maximum filling level in reservoir: h1 , h2 , . . ., ho . It is also necessary to consider the optimization of canal slope which influences waste water flow rate and capacity of canals of different diameters. So, for storage reservoir localization L1 and values of maximum filling level in gravitational part of it h the following equations, describing project parameters were received: V11 (β1 ), Dd1111 (β1 , id1 , L1 ), Dw111 (iw1 , L1 )forh1 , V21 (β1 ), Dd2111 (β1 , id1, L1 ), Dw211 (iw1 , L1 )forh2 , up to Vk1 (β1 ), Ddk1 (β1 , id1 , L1 ), Dw21 (id1 , L1 ) for hk . where V = demanded capacity of storage reservoir; Dd = pipe diameter below storage reservoir; Dw = pipe diameter above storage reservoir. The same calculations must be made for the following values of flow reduction coefficient β2 , β3 , . . ., βn and appropriative localizations of storage reservoir within sewage system L2 , L3 , . . ., Lk and the values of canals’ slopes id2 , id3 , . . ., idy and iw2 , iw3 , . . ., iwt . On the base of estimated project parameters’calculations total investment costs can be determined for the following reservoir’s localizations L1 , L2 , . . ., Lk considering different flow reduction coefficients β1 , β2 , . . ., βn and appropriative diameters behind storage reservoirs Dd11 , Dd12 , . . ., Ddnyk , particular values of maximum filing level in reservoir’ chambers h1 , h2 , . . ., ho , canals’ slopes id1 , id2 , . . ., idy and iw1 , iw2 , . . ., iwt . Determination of optimal reservoir’s localization within sewage system demands the establishment of minimal value of investment costs among investment costs’ range for particular project parameters. For the following values of flow reduction coefficient β and constant values of other decisive variables investment cost’s rates are the following: Kc11111 = V11 (β1 ) · Kinzj + L · Kinsdj (β1 , id1 , Lc − L1 ) + (Lc − L1 ) · Kinswj (iw1 , Lc − L1 ), Kc12111 = V12 (β2 ) · Kinzj + L · Kinsdj (β2 , id1 , Lc − L1 ) + (Lc − L1 ) · Kinswj (iw1 , Lc − L1 ), up to Kconkyt = Von (βn ) · Kinzj + L · Kinsdj (βn , idv , Lc − Lk ) + (Lc − Lk ) · Kinswj (iwt , Lc − Lk ). A priori accepted assumption, that one ore less storage reservoirs can be located within sewage system section, is improper. So it is necessary to make an optimization of sewage system section under assumption of storage reservoirs’ number more than one. Waste water section optimization with one or more storage reservoirs requires the acceptation of definite methodology of optimal solutions’ selection. Optimization model SEW assumes the selection of number and localization of these objects based on principles of dynamic programming. The precursor of the methodology of optimal solutions’ search is Bellman (Bellman 1952) who formulated the following optimization principle used in dynamic programming: “every section of optimal track is also the optimal track”, that means, that optimal steering depends only on actual state, but not on previous states of the system. The system at every stage can be in one of definite states that are the results of definite decisions. The search of optimal solution for greater number of reservoirs within analyzed sewage system section is based on the method of optimal solution’s choice under assumption of one reservoir. This method is determined as the result of total investment costs’ analysis for all possible investment variants. 251

(a)

Kc(0), L, iopt , D A

(b)

B

Kc(1), Lopt, opt, hopt, idopt, iwopt , Dd, Dw

A

B Kc(2)  Kc(1)  Kc(1)

(c)

Kc(1), Lopt’, opt’, hopt’, id opt’, iw opt’, Dd’, Dw’ A

Figure 1.

A'

Kc(1), Lopt”, opt ”, hopt”, id opt”, iw opt”, Dd”, Dw” B

Draft of calculation stages’ sequence for single sewage section: a) stage 1 – sewage system without storage reservoir; b) stage 2 – sewage system with one storage reservoir; c) stage 3 – sewage system with two storage reservoirs.

Methodology of selection of optimal storage reservoirs’ number and localization, suggested by optimization model SEW, are based on hypothetical division of section into two parts and search of optimal solution for particular parts of the section according to dynamic programming principles. Then it is necessary to make optimization calculations of sewage system and reservoirs’ parameters and investment costs within particular sections according to previous scheme. If total investment cost Kc(2) in case of two storage reservoirs within sewage system section exceeds the total cost of the investment with one reservoir Kc(1) it must be recognized that the increase of reservoirs’ number is pointless. Therefore, optimal solution for this system must be selected among optimal solutions for sewage system without storage reservoir or sewage system with one reservoir. In case when total investment cost of sewage system section with two reservoirs Kc(2) is lower than total investment cost of system with one reservoir Kc(1) and of the system without reservoirs Kc(0) it is necessary to carry out the calculations for greater number of reservoirs with following divisions of analyzed section. Calculations for optimal solutions for particular subsections must be made according to waste water flow direction and dynamic programming principles, i.e. optimal selection of parameters for particular sections brings to optimal solution for total system. The draft of calculation stages’ sequence for single sewage system section is presented by Figure 1.

4 CALCULATION ALGORITHM OF SEW OPTIMIZATION MODEL FOR COMPLEX WASTE WATER SYSTEMS In optimization model SEW the principle of dynamic programming is used for analysis and choice of optimal solution for multisection sewage system. Optimization of consecutive sewage system’s sections are carried out according to waste water flow. In conformity with optimal track principle the choice of optimal solutions is made for particular stages that are the consecutive sections of projected sewage system. It means that optimal solutions for particular sections are simultaneously the optimal solutions for total sewage system. The idea of optimization model SEW, the aim of which is the choice of optimal number and localization of storage reservoirs within two sewage system’s sections arranged in rows, is shown 252

Kc(00)  Kc(0)  Kc(0), (a)

Kc(0), L, iopt, D

Kc(0), L, iopt, D A

B

C

Kc(01)  Kc(0)  Kc(1) (b)

Kc(0), L, iopt, D A

Kc(1), Lopt, opt, hopt, idopt, iwopt, Dd, Dw B

C

Kc(01)  Kc(0)  Kc(2) Kc(0), L, iopt, D

(c)

A

Figure 2.

Kc(1), Lopt’, opt’, hopt’, Kc(1), Lopt”, opt”, hopt”, id opt’, iw opt’, Dd’, Dw’ id opt”, iw opt”, Dd”, Dw” B

B’

C

Draft of calculation studies sequence for two sections of sewage system arranged in rows, assuming that optimal solution for AB section excludes the construction of storage reservoir: a) stage 1 – sewage system without storage reservoir within sections AB and BC; b) stage 2 – sewage system without storage reservoir within section AB and one reservoir within section BC; c) stage 3 – sewage system without storage reservoir within section AB and two reservoirs within section BC.

Kc(1), Lopt, opt, hopt, idopt, iwopt, Dd, Dw

(a)

A

Kc(0), L, iopt, D B

Kc(1), Lopt, opt, hopt, idopt, iwopt, Dd, Dw

(b)

Kc(10)  Kc(1)  Kc(0),

A

C

Kc(11)  Kc(1)  Kc(1), Kc(1), Lopt, opt, hopt, idopt, iwopt, Dd, Dw B

C

Kc(12)  Kc(1)  Kc(2), Kc(1), Lopt, opt, hopt, idopt, iwopt, Dd, Dw

(c)

A

Figure 3.

Kc(1), Lopt’, opt’, hopt, id opt’, iw opt’, Dd’, Dw’ B

B'

Kc(1), Lopt”, opt”, hopt”, id opt”, iw opt”, Dd”, Dw” C

Draft of calculation studies sequence for two sections of sewage system arranged in rows under assumption of AB section’s optimal solutions with storage reservoir: a) stage 1 – sewage system with storage reservoir within section AB and the lack of reservoir within section BC; b) stage 2 – sewage system with storage reservoir within section AB and one reservoir within section BC; c) stage 3 – sewage system with storage reservoir within section AB and two reservoirs within section BC.

on Figure 2 and Figure 3 for possible calculation cases. In case of situation presented by Figure 2, when optimal solution for initial section AB excludes storage reservoir’s localization, then for next section BC the following calculations are possible. Analogical situation is in the case of one or more storage reservoirs within initial section AB. In this situation it is assumed that choice of optimal solution was provided and optimization is been 253

carrying out for the next sewage system section BC, arranged in this case parallel to AB. The draft of optimal solution search for two-section sewage system, which optimal solution for first section considers the localization of one storage reservoir, is illustrated by Figure 3.

5 STORAGE RESERVOIRS’ FUNCTIONING SIMULATION During long time in order to establish the required capacity of storage reservoirs the simplified methods, assuming the definite waste water flow type though sewage systems in the form of triangle or trapezium, were used. Actually owing to the development of soft-ware instruments and significant resource of precipitation measurement data it is possible to carry out the dynamic simulations of storage water reservoirs’ functioning for any precipitation situations and storm water flow rates transported by sewage systems (Dziopak & Sły´s 2007). Equally intensively the researches of storage reservoirs are developed in order to analyze not only hydraulic parameters but also qualitative parameters concerning waste water treatment in processes of sedimentation in concrete and ground constructions (Hubner & Geiger 1996; Jacopin et al. 1999; Michelbach & Weiß 1996; Calabrò & Viviani 2006). In presented optimization model SEW the elaborated programs of one- and multi-chamber storage reservoirs’ functioning simulation are used (Dziopak & Sły´s 2007; Sły´s 2006; Sły´s & Dziopak 2006), which allows to investigate any waste water flow function in sewage systems and to select the hydraulic parameters of such type of objects.

6 CONCLUSIONS Storm water retention in separate and combined sewage systems has a great significance not only for the reasons of water ecosystems’ safety but for economical and exploitation reasons. The key issue of storage reservoirs’ usage for the purpose of sewage system hydraulic relief is their number and localization within sewage net, that determines the project parameters of sewage system and reservoirs. The feedback of parameters, which are closely connected and mutually influenced, takes place. This demands very complicated calculations carried out in traditional way. The paper deals with the matter of modeling of storage reservoirs’ number and their localization within sewage system on the base of elaborated optimization model SEW. It considers the influence of all significant project parameters of sewage system cooperating with storage reservoirs: canals’ slops above and below the reservoir, maximum filling of storage reservoir, coefficient of waste water flow reduction, applied hydraulic scheme of waste water accumulation. The elaborated soft-ware instrument is based on previously elaborated programs for hydraulic processes’ simulation in multi-chamber reservoirs and their project parameters’ investigation for any type of waste water flow functions. REFERENCES Bellman, R. 1952. On the Theory of Dynamic Programming. Proceeding of the National Academy of Sciences USA. Biedugnis, S. 1993. Metody optymalizacyjne w wodocia˛gach i kanalizacji. Wydawnictwo Naukowe PWN. Warszawa. Calabro, P. & Viviani, G. 2006. Simulation of the operation detention tanks. Water Research 40: 83–90. Dziopak, J. & Sły´s D. 2007. Modelowanie zbiorników klasycznych I grawitacyjno-pompowych w kanalizacji. Rzeszów University of Technology. Rzeszów. Dziopak, J. 1997. Multi-chamber storage reservoirs in the sewerage system. Technical University of Cze˛stochowa. Cze˛stochowa.

254

Dziopak, J. 2006. Selection of Optimal Variant of Storage Reservoir within Sewage System. Ecological Chemistry and Engineering. 13(11): 1269–1286. Huebner, M. & Geiger W. 1996. Characterisation of the performance of an off line storage tank. Wat. Sci. Tech. 34(3–4): 25–32. Jacopin, Ch. et al. 1999. Characterisation and settling of solids in an open, grassed, stormwater sewer network detention basin. Wat. Sci. Tech. 39(2): 135–144. Krajewski, K. 1993. Metody optymalizacji w in˙zynierii s´rodowiska. Oficyna Wydawnicza Politechniki Warszawskiej. Warszawa. Michelbach, S. & Weiß, G. 1996. Settleable sewer solids at stormwater tanks with clarifier for combined sewage. Wat. Sci. Tech. 33(9): 261–267. Sły´s, D. & Dziopak, J. 2006. Simulation of Trough-Flow Chamber Operation in Storage Reservoirs. Ecological Chemistry and Engineering 13(10): 1143–1155. Sły´s, D. 2006. Simulation model of gravitation-pump storage reservoir. Environment Protection Engineering 32(2): 139–146.

255

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Curvature jacking of centrifugally cast GRP pipes Ulrich Wallmann HOBAS Rohre GmbH, Bottrop, Germany

Dariusz Kosiorowski HOBAS System Polska, Be˛dzin, Poland

ABSTRACT: In recent projects for example Krefeld Germany or Warsaw E1 (3,5 km DN2000 min. Radius 300 m) centrifugally casted GRP jacking pipes have proven their favourable properties in curvature jacking installations in Europe. With the use of the curvature jacking technology GRP pipes have been installed safely in a very economical way. The smallest radius was realised in 2004 by driving centrifugally casted GRP jacking pipes DN1000 in Zielona Gora Poland in a radius of 90 m. Particular in curvature jacking the high compressive strength and the elasticity of GRP are favourable features. The centrifugally cast GRP jacking pipe is because of the elasticity of the material able to react to eccentric loads by means of spigot deformations. Therefore the contact between the centrifugally cast GRP jacking pipes for the transmission of jacking force remains totally until the deformation limits are no exceeded. So there is no need of wood packers. Pipe tests in Germany have shown, that below the allowable compressive stress of 90 N/mm2 the GRP material reacts linear elastic without been influenced by wetness. Therefore the calculation of the deformation and the allowable jacking force for the curvature installation of centrifugally cast GRP jacking pipes is very accurate and reliable.

1 INTRODUCTION For environmental and economical reasons curvature jacking is more and more required by designers and clients. In a lot of cases the separation of the project into straight sections with bends only inside of the jacking pits requires more and deeper jacking pits. With the help of curvature drives the quantity and the depth of the jacking pits can be optimized. On the other hand curved drives require a special jacking equipment and a special design of the allowable jacking forces. A curvature drive results always in an angular deflection of the joints and an eccentric longitudinal loading of the pipes. The eccentric longitudinal loading has to be considered in particular by the calculation of the maximum allowable angular deflection of the joint between the pipes and the maximum allowable jacking force of the pipes. 2 THE PIPE SYSTEM Centrifugally cast GRP jacking pipes are produced in diameters from DN200 up to DN2900 in unit length of 1 m, 1.5, 2 m, 3 m and 6 m. The pipes are designed either for gravity or for pressure applications up to PN10. The coupling of the pipe is available in GRP (figures 3 and 4) and stainless steel (figure 5 and 6). The allowable jacking force of centrifugally cast GRP jacking pipes is depending of the wall thickness. Centrifugally cast GRP jacking pipes have been supplying 257

Figure 1. Transport of a centrifugally cast GRP jacking pipe DN1400 in Zielona Gora, Poland.

Figure 2. Curved installed centrifugally cast GRP jacking pipes DN2000 in Warsaw.

jacking projects since more than 20 years. The pipes are inherently corrosion resistant, rugged, solid-walled and provide a long maintenance free life. Because of the high compressive strength these pipes have a smaller wall thickness and therefore a smaller external outside diameter than most of the other conventional pipes for jacking. The smaller external diameter results in a reduction of excavation , lubrication and equipment costs. 258

Figure 3.

Pipe spigot for the stainless steel coupling.

Figure 5.

Pipe spigot for the GRP coupling.

Figure 4. The stainless steel coupling.

Figure 6. The GRP coupling.

Centrifugally cast GRP jacking pipes have a long record in Europe, Japan and the US with more than 500 km of pipes installed by the jacking method.

3 THE JACKING FORCE CALCULATION Centrifugally cast GRP jacking pipes have the following properties in the axial direction: Ultimate compressive stress in axial direction Elongation at break in axial direction 259

σa = 90 N/mm2 εa = 0.7%

D0 D1

D0 D1

A

r

A

x

r r

P

r

P Gmax = 2G0

volting

0

Figure 7.

0

0volting

0min = 0

Centric loading.

Figure 8.

0volting

Eccentric loading.

The allowable jacking force under centric loading can be calculated with the equation (1): P=

σa · AR S

(1)

AR = Cross-sectional area of the pipe wall at the thinnest point S = Material safety (1.75) P=

π σa · (Da2 − Di2 ) · 1.75 4

(2)

The jacking force P is a purely theoretical figure that can never be applied in practice. As some steering always has to be anticipated, even for planned straight installations. For the calculation of allowable jacking force allP at least the following eccentricity has to be considered. allP =

1 ·P 2

(3)

P – from equation (2)

π σa · (Da2 − Di2 ) · (4) 2 · 1.75 4 If the pipes are subjected to the eccentric allowable jacking force allP the spigot is deflected as shown in figure 4, which can be calculated by equations 5 and 6. allP =

allε =

εa 1.75

1 = allε · L L = Applicable length (usually half of the pipe length) 260

(5) (6)

P

α

∆1

L

Figure 9.

Spigot deformation under eccentric loading.

Table 1. Allowable angular deflection of the joint under the assumption of a full contact between the pipes. Max. permissible angular deflection (Pipe length) Outside Diameter OD

1m

2m

3m

272–376 401–550 616–752 818–860 924–1099 1229–1348 1434–1720 1842–2047 2252–2740

0.48 0.33 0.24 0.21 0.16 0.13 0.10 – –

1.0 0.6 0.5 0.4 0.3 0.25 0.2 0.15 0.10

1.5 1.0 0.7 0.6 0.5 0.4 0.3 0.25 0.2

For the calculation of 1 half the pipe length can generally be taken as the applicable length. Due to the elastic spigot deformation 1 the allowable angular deflection of the joint shown in Table 1 occurs under the assumption of a full contact between the pipes. V =

1 max σ σo

·P

(7)

P from equation (1) Z = Width of the contact area If the angular deflection shown in Table 1 is exceeded during pipe jacking, the joint between the pipes will gape and the full contact between the pipes is lost, which in turn reduces the contact area. 261

z

d1 d2

dmax d0

10 9

di d2 0.9 di 0.8 d2

8 7 6

di 0.7 d2

dmax 5 d0 4

d1

d1 L

3 2

Z

1 0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 z d2

Figure 10.

Relation between the stress ratio max σ/σo and the ratio Z/D.

Figure 11.

Spigot deflection with a gaping joint.

To prevent the permissible compressive stress σa from being exceeded, the allowable jacking force has to be reduced. The reduced permissible jacking force V in a gaping joint can be calculaed by taking figures 10 and 11 into consideration. The following considerations and equations only apply provided that no wooden packing rings are inserted between the pipes. In this case the allowable angular defection between the pipes is only the result of their elastic deformability and the accepted gap in the joint. The partial deflection caused by the eccentric load V can be calculated as follows: tan α =

L εa · 1.75 Z

(8)

The partial gap width 2 in the area of the gaping joint is calculated as follows: 2 = tan α · (Da − Z)

(9)

The total deflection β of the pipe joint is then β = α r + αl

(10)

β =2·α

(11)

or by approximation 262

Figure 12. The test rig of IKT Gelsenkirchen.

Figure 13. The bended GRP pipe.

The total gap width  in the pipe joint is then  = 2r + 2l

(12)

 = 2 · 2

(13)

or by approximation

4 PIPE RESEARCH TESTS Over the past 25 years dozens of independent verifications of the properties of centrifugally cast GRP pipes have been performed. Tests have been carried out by testing samples of strips and full pipes under centric and eccentric loads. At University of Illinois, at University of Bochum, at University of Dortmund and at the test laboratory of IKT Gelsenkirchen full length pipes in diameters between DN500 and 700 were tested. In 2005 at the University of Aachen a test of centrifugally cast GRP pipes under a cyclic angular deflected joint has been carried out with the result that these pipe react linear elastic under eccentric longitudinal loading. While the pipe was loaded by a longitudinal force and the pipe ends were moved ±20 cm rectangular to the pipe axis the longitudinal stress and strain were measured at the spigots and over the pipe length. The material properties of: ultimate compressive stress in axial direction elongation at break in axial direction

σa = 90 N/mm2 εa = 0.7%

were even under this test circumstances verified. 263

Figure 14. The test rig of Aachen University.

Figure 15. The test rig of Aachen University.

Figure 16. The route of the pipeline of the E1 project in Warsaw.

5 PROJECTS 5.1 The Warsaw Project E1 For the optimization of the down town sewer system Warsaw Waterworks designed a 3,5 km gravity sewer pipeline in a diameter of 2000 mm for the connection of a down town living quarter with the new sewer treatment plant “Czajka”. For the realization of a slope of 0,063% the pipeline had to 264

Figure 17. The location of the pipeline of the E1 project in Warsaw.

Figure 18. The location of the jacking pit of the first drive of the E1 project in Warsaw.

be installed in a depth between 4.7 m and 10.6 m. The maximum expected groundwater level was 2 m below the surface. Particular problems of the route of the pipeline are the crossing of the Warsaw Underground in a distance of only 0.6 m, the crossing of a main railway to Warsaw Main Station and the crossing of several sewers. Because of this problems and the installation under partly very narrow roads Warsaw Waterworks decided for jacking installation with curved drives in a minimum radius of 300 m. For the pipe material Warsaw Water works finally choose centrifugally cast GRP pipes for several reasons. First of all Warsaw Waterworks made a good experience during the installation of more than 10 km of centrifugally cast GRP pipes in diameters between DN1000 and DN2400 within the past 10 years. The high corrosion resistance would guarantee a long lifetime and the superior hydraulics would provide a high flow capacity. The project was contracted in summer 2006 and started in October 2006. For the realization the pipeline was partitioned in 15 sections. The longest drive had a length of 543 m with a curve in a radius of 400 m. In order to satisfy the various requirements of all sections the contractor finally decided for 3 different pipes. For the straight parts a pipe with a nominal stiffness of 32000 N/m2 and an allowable jacking force was selected. For the curved parts pipes with a nominal stiffness of 50000 N/m2 were selected in a unit length of 1.5 m and an allowable jacking force of 3600 kN 265

Figure 19. Jacking of centrifugally cast GRP pipes DN2000 at the E1 project in Warsaw.

for the 300 m radius and in a unit length of 1 m and an allowable jacking force of 3000 kN for the 200 m radius. 5.2 The Zielona Gora Project in Poland In the year 2003 a pipeline in a diameter of 1000 mm in a length of had to be build in Zielona Gora, Poland. The location of the pipeline was down town under a road between an eggshape sewer DN700/1000 and the foundation of a multistory building. Under consideration of the difficult traffic situation the City of Zielona Gora for a curved jacking installation. The route with the least disturbance of traffic was a curved drive with a radius of 90 m. The client was fully aware of the risk of a drive with a radius of 90 m. Because of good experiences with several curved drives with centrifugally cast GRP pipes in diameters 1200 and 1400 mm the City of Zielona Gora finally approved the installation of centrifugally cast GRP pipes DN1000 with an outside diameter of 1099 mm and a nominal stiffness of 160000 N/m2 . These pipes were designed for an allowable jacking force of 4000 KN for the straight parts and for 1400 KN for the curved drive. All the pipes which had to pass the curve had a unit length of 1 m, while the rear straight part was made by 3 m long pipes. Despite of the technical challenge without any problems up to 24 m per day were successfully installed. 5.3 The Krefeld Project In summer 2000 the city of Krefeld had to build a sewer with an overflow to increase the existing capacity of the “Rundweg combined sewer”. Because of the aggressive industrial effluent, the existing DN 2000 interceptor, which was in operation, could only be accessed from poorly ventilated manholes with breathing apparatus. To provide the greatest possible corrosion resistance for the new sewer, the client demanded centrifugally cast GRP pipes. The tender specified pipe installation by jacking.The project totaling 318 m in length was divided into two sections. The first push measuring 282 m was carried out at a depth of 6.5 to 7.5 m, partly parallel to a private railway line under an inner-city side road. At a length of 36 m and cover of 4 to 7 m the second push had to cross the busy B 288 federal highway. GRP 266

Figure 20. The location of the pipeline in Zielona Gora, Poland.

jacking pipes with an outside diameter of 2400 mm and a wall thickness of 76 mm were selected for jacking. These GRP jacking pipes have a nominal stiffness of 32,000 N/m2 and an allowable jacking force of 9460 kN (angular deflection per non-gaping pipe joint 0.2◦ ). A civil engineering contractor was awarded the contract for manned pipe jacking. An auger protected by a shield was used to remove the native soil at the face, which was then loaded onto trolleys by a conveyor. A winch pulled the trolleys, which ran on rails, to the jacking pit where they were lifted by a crane. The pipe jacking route was measured with a gyrocompass and the elevation with inclinometers. To check, the jacked pipeline was also measured electro-optically at certain intervals.The 2,9 m long GRP jacking pipes were installed according to the following procedure. Behind the shield, which was 2840 mm in length, there were two 2900 mm long working pipes made of centrifugally cast glass reinforced polyester. These working pipes housed all the auxiliary equipment required for jacking, such as conveyor, winch etc. Next to them the first intermediate jacking station was set up. After installing another 120 m of GRP pipes, which were jacked in a straight line with minimal steering adjustments, came the second intermediate jacking station, although the maximum jacking forces measured up to this point were less than 45% of the permissible limit. Once the intermediate jacking station had been installed the pipes were jacked at a curve radius of 1500 m until after another 160 m and a total of 32 working days, the reception shaft was reached. On the pipe manufacturer’s recommendation, no wooden packing rings were used to transfer the pressure at the joints. The elastic deflection in centrifugally cast GRP jacking pipes of 8 mm per pipe was sufficient to install them round the curve at a radius of 1500 m without damage and still have enough reserves for steering. Only in exceptional cases, e.g. when starting jacking again after a weekend, did jacking forces of just under 7000 kN have to be applied to get the pipeline moving again. As the required jacking forces were only 70% of the allowable limit for GRP jacking pipes, the second intermediate jacking station was not used at all during the jacking process.The first intermediate jacking station was operated at a jacking force of 500 to 700 kN and served only to overcome the peak resistance of the shield. This means that the entire push of 288 m including the curve was jacked with average of 6500 kN and a maximum jacking force of 8000 kN. Over the entire drive, the native soil mainly consisted of medium-dense to dense sandy gravel. At the beginning of the first push, clay had to be removed 267

Figure 21. The jacking pit in Krefeld.

Figure 22. The pipe installation in Krefeld Germany.

from the pipe crown area. The groundwater table, which was at a height of 1.0 to 1.5 m above the pipe bottom, was lowered below it with wells during the construction work. During normal operation, jacking was carried out at a rate of approx. 9 m per shift and a maximum of 12 m per shift. This was also achieved because of the comparatively small volume of material transported thanks to the thin walls in GRP jacking pipes. Compared to concrete pipe of the same inside diameter, the volume of excavated material here was approx. 30% less. When applied to the total installation length of 318 m, the total saving in soil transport is around 650 cu m. In addition, the contractor confirmed that the favorable jacking force development resulting from the smoothness and close tolerances of the pipes was an added bonus during installation.

268

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Relining with large diameter GRP pipes Ulrich Wallmann HOBAS Rohre GmbH, Bottrop, Germany

ABSTRACT: The relining of large diameter pipelines with GRP pipes is worldwide continuously growing over the past 20 years. For the structural design of liner pipes with significant deviations to circular shapes are no standards or guidelines available. The ATV-M127 Part2 and the WRC SRM (sewer rehabilitation manual) provide only hints and notes for a coarse calculation of non circular liner pipes, but no comprehensive structural design. In particular arch-profiles, kite-profiles or profiles with inverts can merely designed by the previous mentioned guidelines. A reliable approach for the structural design of these kinds of profiles is the finite element method.

1 INTRODUCTION After the flood disaster in the eighties and two floods in 1993 and 1995 the city of Cologne decided in 1996 for a flood control concept for the protection of the threatened city areas. An integral part of the flood control concept was the adjustment of the storm water outfalls into the river Rhein to the new height of the dykes. This led to higher internal and external loads on the pipelines. The outfalls are 90 year old masonry constructions with circular and non circular profiles in diameters up to 3,5 m. Previous inspections revealed a high number of leaks. Because of the higher requirements of the new flood control concept, in particular the internal pressure of 0,75 bar and the external pressure of 1 bar, the structural load capacity of the masonry could no longer be guaranteed. So an extensive rehabilitation of the outfalls was imperative.

Figure 1. The GRP pipe: DN3190/2585 L = 2 m.

269

Figure 2.

Drawing of the pipeline.

Figure 3. The relined pipeline.

2 PROJECT DESCRIPTION The city of Cologne contracted various consulting companies to carry out complex analyses of rehabilitation strategies and their costs. The prime target of the rehabilitation was to size the renovated outfalls as tight as possible to the old. Besides the size of up to 3,5 m, misalignments and curves of the masonry outfalls were further problems for rehabilitation. Because of the high structural load requirements, the issues of leak tightness and the short possible construction time the client finally decided to rehabilitate the outfalls with prefabricated GRP pipes. Before the pipes were ordered the contractor carried out a detailed 3 dimentional survey of the masonry in order to have a clear picture of the old pipeline. After the evaluation of the survey data the GRP pipes were selected as following: for the masonry arch profile of DN3500/2900 a GRP pipe with an outside diameter of 3190/2584 mm, for the masonry arch profile of DN2700/1750 a GRP pipe with an outside diameter of 2368/1454 mm and for the masonry circular profile of DN2400 a circular GRP pipe with an outside diameter of 2520 mm were selected.

3 THE STRUCTURAL DESIGN 3.1 Introduction For the determination of the wall thickness of the GRP pipes the structural design of the relined non circular masonry was carried out by finite element calculations, while the circular parts were calculated by ATV-M127 Part 2. For all calculations the following pipe properties were used: Flexural modulus (short term): Ek = 9000 N/mm2 Flexural modulus (long term): EL = 5625 N/mm2 Allowable bending stress (short term): bk = 120 N/mm2 Allowable bending stress (long term): bL = 75 N/mm2 Poisson ratio υ = 0,35 The finite element calculation was done by LGA Bautechnik GmbH Tillystraße 2 90431 Nürnberg Germany with the help of the finite element program NISA. For 5 different load cases 3 different finite element models were formulated. 3.2 Load case1 and 2 (LC1 and LC2) Load case 1 and 2 are considering the minimum and the maximum depth of cover plus traffic load (SLW60). The calculation was carried out for 50 years design life. For the GRP pipe, the masonry and the soil “PLANE elements” were used. For the connection between the pipe elements and the 270

Figure 4. The finite element model for LC1 and LC2.

Figure 5. The deformation due to LC1.

Figure 6. The finite element model for LC3 and LC4.

Figure 7. The deformation due to LC3.

Figure 8. The finite element model for LC5.

Figure 9. The deformation due to LC5.

masonry elements and for the connection between the masonry elements and the soil elements “GAP elements” (pure compression transmition) were used. The GRP pipe was represented by its actual profile, while for the masonry 4 joints, 1 in the crown, 1 in the bottom and 1 in each springline, were considered. The soil elements on top and besides the pipe were given an E-modulus of 5 N/mm2 , while the soil elements underneath 20 N/mm2 . 3.3 Load case 3 and 4 (LC3 and LC4) For both load cases the masonry carries the imposed soil and traffic loads. Load case 3 considers only an external water level of 7,5 m above the invert. Load case 4 is considering an external water level of 1,5 m. Load case 3 is for short term, while load case 4 is for long term design. For the 271

Figure 10. The trolley for the transport of the pipes.

Figure 11. The trolley pulled by a forklift.

Figure 12. The steel cradles for fixing the pipes.

Figure 13. The misalignment.

GRP pipe “PLANE elements” were used. The stiff embedment was represented by “GAP elements. For the bottom of the GRP pipe a predeflection of 20 mm and between the GRP pipe and the stiff embedment (the grout) a gap of 1 mm was assumpt. 3.4 Load case 5 (LC5) In load case 5 the liner pipe is calculated for short term external pressure of grout 1 m above the bottom of the pipe. For the GRP pipe “PLANE elements” were used. The grout pressure was considered as an equal circumfentural load. 4 THE PIPE INSTALLATION The evaluation of the finite element calculation together with the result of the evaluation of the survey data gave the following result for the prefabricated GRP pipes: Outside diameter: 3190/2584, Outside diameter: 3190/2584, Outside diameter: 3190/2584, Outside diameter: 2368/1454, Outside diameter: 2490,

wall thickness: 60 mm, length: 2,0 m, weight: 2300 kg wall thickness: 60 mm, length: 1,0 m, weight: 1150 kg wall thickness: 60 mm, length: 0,5 m, weight: 575 kg wall thickness: 50 mm, length: 2 m, weight: 1300 kg wall thickness: 45 mm, length: 2 m, weight: 1800 kg 272

Figure 14. The lamination of the steps.

Figure 15. The laminated steps.

Figure 16. The pressure test of the joints.

Figure 17. The installation of the pipe DN2400.

The prefabricated GRP pipes were inserted into the old pipeline through installation pits. From the pits the GRP pipe were transported to their destination by a forklift pulled trolley. After jointing the GRP pipes were fixed by steel cradles with a spacing of the pipe length. The cradles are preventing the GRP pipes from floating during the grouting of the annular space between the old masonry and the liner pipes. For the relining of the curved sections GRP pipes in length of 0,5 m and 1 m were used. To avoid wide gaps due to the angular deflection after jointing the spigots of the GRP pipes for the curves were 1◦ oblique cut. At the misalignments of the masonry the pipes were connected by hand laminates. After an installation time of 4 months the total pipeline was checked by a pressure test of all joints of 1 bar. The tests were carried out with the help of a special joint test equipment for circular and non circular profiles. Because of the high finished quality of this project the city of Cologne in 2008 to do another 300 m of relining with the same GRP pipes DN3190/2584. REFERENCES Doll, Dr.Ing . 2006. Gutachterliche Stellungnahme für Köln, Theodor-Heuss-Ring. Doll, Dr.Ing. & Schimmel. 2007. Refurbishing of large caliber drain and sewer sectios.

273

Underground Infrastructure of Urban Areas – Madryas, Przybyła & Szot (eds) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Underground infrastructure of historical cities as exceptionally valuable cultural heritage M. Wardas & M. Pawlikowski University of Science and Technology, Kraków, Poland

E. Zaitz & M. Zaitz The Historical Museum of the City of Kraków, Kraków, Poland

ABSTRACT: Historical sequence layers and underground infrastructure of Kraków, Kazimierz and their suburbs have been explored, for many years, by archaeologists and historians. Groundexplorers, (hydro)geologists and geomorphologist, as well as mineralogists and geochemists, in the course of interdisciplinary study, have completed the knowledge of the origin and extent of pollution of the, formed in the past, deposited matter, together with its cultural levels. The data, achieved as a result of multidisciplinary approach to the problem, particularly, information about heavy metal content, make probably possible the determination of the contaminant migration routs, as well as a reconstruction of the environment state in the past. A mineralogical-chemical study of historical sequence layers of underground Kraków showed a strong geochemical Pb and Cu anomaly, which revealed a considerable activity of inhabitants, concerning manufacturing and use of metal objects. The localities of higher anomaly have been analyzed, as far as the presence and origin of metallogenic phases, as well as the presence and age of, found by archaeologists, artifact are concerned. Identification of places characterized by nonstandard metal content may also throw the light to the problem of ecology and a threat to health and environment of a man living in Kraków in the past. Historical sequence layers, due to the content of artifacts but also due to the possible identification of ecofacts, as a result of the determination of elements, organic matter, pollens and seeds, constitute, same times the last, “disappearing” source of knowledge of the history and life of people of that time. That is why, the deposited matter, even if devoid of artifacts, should be recognized as a particularly valuable cultural heritage, unfortunately, extinguishing in the course of road and building investments.

1 INTRODUCTION 1.1 Historical stratification as a record of the city’s history There are many forms of recording history and events. The most classical is the written word in the form of manuscripts on paper, or before that on papyrus made of plants, or parchment made of animal skin, or on the bark of various trees. Another onetime form of making records was rock drawings or paintings, refined and transformed into an alphabet (hieroglyphics, for instance). A contemporary form of record is photography or film images (whether static or moving) on conventional or electronic media. Technology is still incapable of recording taste, smell or touch impressions. There is, however, another form of “recording” events – geological or archaeological strata, known as stratification. Knowing a specific (geological or archaeological) language, one can “read” a great deal of interesting information from stratification. This is objective information that has not been altered by an artist – a painter, poet or writer, all of whom add interpretation to the facts they render. 275

The language researchers use to “read” information from sediment and rocks of various sorts is specified and often difficult. In many cases, even someone who knows the layers well can discern or understand almost nothing. Often, however, there are many new and interesting facts to be learned. For example, was it wet or dry during the sedimentation period of the layers? Warm or cold? Was it created by wind, water, or man? How did it come about? What are its component parts? If the layer has anthropogenic components, this may be seen as evidence of human activity, and it might tell us what people did here. This is most often unrecorded information. Did a potter or a glazier work here, or perhaps there was a smithy where bronze or lead casts were made? Perhaps iron was smelted and forged in the research site, or grain was ground for flour, leaving behind bran. The stratification can contain pollen and seeds from plants that allow us to reconstruct the plant life in a given place – in spite of the fact that it grew here many millennia ago, and that there seem to be no practical remains of what once was. Isotope dating allows us to establish how long ago the stratum was formed (Kluj et al. 2006, Sokołowski et al. 2006). All these questions posed and answers achieved broaden our knowledge about the past and are often the basis for current activities and decisions. One example might be the present research under newly-built architectural sites or future highways. Sometimes precious historical structures are unearthed through this exploration, which results in necessary changes to previous intentions and protection of the given site. This stratification and knowledge, both what has been researched and what lies in its potential, is our cultural heritage, vital for understanding the here and now. This is why it should be both researched and protected. When it is excavated, irreplaceable historical material is lost. This is why historical research of stratification is also so vital, particularly in the region of old, historical cities. If we accurately decipher and draw conclusions from the historical records of our ancestors, we can sometimes avoid unfavourable decisions, and make only those essential for the further development of the people and the nation. 1.2 Historical stratification as cultural heritage The historical stratification and underground infrastructure of Krakow, Kazimierz and their suburbs have been researched by archaeologists and historians for many years. For some time now, land specialists, (hydro)geologists and geo-morphologists, as well as mineralogists and geo-chemists have been carrying out interdisciplinary research to supplement their knowledge on the genesis and level of the pollution, the embankments formed in the past, and the state of their culture. Data acquired through a multi-tiered approach to this question – and information about heavy metal content in particular – might allow us to establish the path of pollution migration and to reconstruct the state of the environment in the past (Kluj et al. 2006, Sokołowski et al. 2006, Wardas et al. 2006a, b, 2007, Wardas et al. in press., Wardas & Such in press.). Archaeologists in particular consider the underground infrastructure of historical cities to be exceptionally valuable cultural heritage. These researchers, knowing the “language” of archaeological stratification, are able to very accurately estimate what historical period it should be affiliated with, and at a glance (Zaitz & Zaitz, in press.). 2 THE ATTRIBUTES OF KRAKOW’S HISTORICAL STRATIFICATION 2.1 The early mediaeval period Krakow early-mediaeval settlements are dominated by sandy-clay strata of relatively low pulpiness (up to 60–80 cm), which have accumulated material traces of various farming activity. They most often contain shards of clay dishes, splintered animal bones, crumbled charcoal and bits of pugging; much more infrequently there are metal objects (from iron, bronze and copper), glass products, animal bones, horns, semi-precious stones etc. There are also organic and non-organic micro-traces of tree branches, stems and leaves of plants, seeds, fruit and pollen, or even slag affiliated with the waste, production and manufacture of metal objects (Radwa´nski 1975, Zaitz 1998). 276

On the top of this stratification were wooden buildings, often log-style and with no foundations. Various earth-formed sites dug into the sandy, natural subsoil were also produced from the surface of this stratification. These included the lower parts of residential half-dug-out buildings, various caverns affiliated with farming and production, and waste ditches for organising the settlement terrain, liquidating buildings, and removing the wreckage following various cataclysms. Other earth-formed sites included wells and various water reservoirs, both for consumption and production purposes. The early mediaeval settlements were almost totally devoid of plumbing facilities. Excess water and sewage was drained across the surface of the land to flowing water, or it gradually soaked into the cultural strata, penetrating to the natural subsoil. Among the highest humus strata, the remains of levels that once functioned as roads, streets and squares hard-paved with pebbles or gravel stand out, as do the levels shaped by the interiors of buildings. Among these was a dirt floor, traces of wood floors, and in more representative sites, flooring made of various stones, plaster, mortar, or fired clay (Zaitz 2006, Firlet & Zaitz 2007, Zaitz & Zaitz 2007a, b). 2.2 The late mediaeval period A new type of settlement appears in Krakow in the late mediaeval period. These are urban settlements (the newly-founded “nova civitas in Okol” cities, Kazimierz, Florencja – known as “alta civitas”, and later Kleparz), which were placed on the fringes of the early-mediaeval settlement concentration, just outside the Royal City’s domain. These settlements were densely populated, bred livestock, and farmed various crops (Radwa´nski 1975, Zaitz 1998). Humus and sandy stratification still formed, not unlike the early-mediaeval formations. But it started to be replaced by other stratification in the areas of most concentrated farming activity. Apart from the humus formations with dominant sand and clay components, there also appeared a series of strata directly affiliated with farming and livestock. It reached a pulpiness of 300–400 cm, while its chief component was a large quantity of various plant materials (from leaves, branches and splinters, to seeds, pollen, fruit, remains of wooden constructions and everyday objects) and animal waste (animal dung and manure), as well as brick and stone rubble, charcoal, glass, metal, bone and stone objects, mortar and other refuse associated with the farming activities of the city and outlying residents. Among these humus and organic formations there appear the cobbled surfaces of roads and streets, in Krakow most often made of calcium stone laid on a sand base, flagstones (made of bricks and flagstones), clay threshing floors, wooden floors and grassy surfaces which were not paved. The late-mediaeval development of stone construction also bore fruit in the appearance of construction layers of brick fragments and tiny calcium (or sand) stones and strata of wreckage. They were also very often accompanied by levelling embankments, whose task was to achieve a correct configuration of the terrain situated inside the constructions, on the streets, or in their nearby vicinity. The new urban agglomerations were distinguished by condensed buildings and intensified economic activity (trade, various handcrafts, farming), as well as attempts to satisfy people’s basic needs. Larger and larger wells for collecting water appeared within the city (there were at least four on the Main Square alone, in its four corners), a wooden water pipe was built to draw water from the Rudawa to wells and barrels scattered around squares and in the vicinity of bourgeois flats, and reservoirs were made to collect water for agriculture and fire-fighting. The paved surfaces of roads and squares, on the other hand, had gutters to lead the excess water and sewage outside of the range of the buildings, while septic pits of over ten metres deep were dug in the backs of some properties, thus collecting a significant portion of human and animal waste materials from the property (Zaitz 2006, Firlet & Zaitz 2007, Zaitz & Zaitz 2007a, b). 2.3 Modern Times In modern times – after the 17th century – organic strata to a large degree vanish, such as that affiliated with livestock breeding within the bounds of mediaeval cities. Their place is occupied 277

by compensatory layers and levelling embankments, as well as practical levels with a soft surface (earth) or reinforced with stones (calcium stones), breakstone, gravel, rubble, as well as grounds made of bricks, flagstones and poured calcium. The process of excessive collecting and piling of strata gradually comes to an end. The rebuilding of a surface now more frequently taking apart the old level and replacing it with a new surface. At the same time, old cobblestones (or the less frequently used flagstones or bricks) are often used repeatedly for a new surface and supplemented with new materials. Because of the various ground and construction jobs on embankments, pedestrian levels and other modern strata, there occur formations composed of destroyed older cultural strata (mainly early and late mediaeval) (Zaitz 2006, Firlet & Zaitz 2007, Zaitz & Zaitz 2007a, b).

3 RESEARCH SUBJECT 3.1 Archaeological characteristics Archaeological researches of Kraków City, conducted during recent years in the areas of the Main Market Square, Reformers’ Monastery and other architectural constructions, registered series of undisturbed settlement sequence layers, of early and late Middle Ages. In the western part of the Main Market Square the before town location settlement (Early Middle Ages) was represented by a humus layer, of relatively small thickness. It was covered by the late mediaeval strata, the oldest of the second half of 13th century, the youngest associated with the beginning of 15th century. Within them remained fragments of trade-, representative- and residential constructions, built of both stone and wood. The layers were covered by leveling deposited matter of 19th and 20th century, with preserved remains of paving of that time. In the ceiling of the late mediaeval and modern sequence layers the various water and sewer installations have been found. The oldest of them (stone block canal) have been built in the beginning of the 19th century and rebuilt then, severed times, in the 20th century. Above those installations, a concrete construction of the paving foundations, originated from 1960s, have been found (Zaitz 2006, Kluj et al. 2006, Wardas et al. 2006a, b). Over them, in turn, there are the new constructive layers of the recently paved surface of the Market. From the sequence layers, explored within the Main Market Square, several dozen of thousands of various artifact have been gained. From them, the samples for the paleobotanical, palinological, geological and geochemical investigations have been taken.

3.2 Geochemical characteristics The subject of detailed study are layers of deposited matter, exposed in the course of rescue excavations, conducted in the western part of the Main Market Square, and regard the variability of Cu and Pb accumulation (Table 1, Figs 1–2). The results are compared to those of the strata excavated in the area owned by the Monastery of Reformers in Kraków. The aim of the comparison was to present the extent of the manufacture activity, associated with trade and metal melting. The possible centre of that activity was the SE part of the today Main Market Square. Particularly interesting place, as far as the migration routes of pollutions are considered, is the Reformers’ Monastery. It is placed not only far from the sources of Cu and Pb emission, but according to the town topography, higher than historical manufacture centers, like the Small- and the Great Balance, as well. The previous investigations of historical strata of grounds and sediments showed considerable variations and a strong Cu and Pb anomaly, exactly in the area of mentioned abovemanufacturetrading objects. On the map of Kraków, limited to the area within the Planty Garden and the Wawel Hill (Fig. 3), the diagrams of metal accumulation variability, and morphology of the then town are presented. 278

Table 1.

170 cm thick profile in the western part of the Main Market Square.

Layers Depth in cm

Microscopic description of layers and proportion of characteristic constituents

0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80 80–90 90–100 100–110 110–120 120–130 130–140 140–150 150–160 160–170

Mixed material, ignored in research, Yellow-gray, sandy, aggregates of organic matter, fragments of debris and bricks, Elements of building infrastructure, samples not taken, Light, sandy, badly sorted, fragments of limestones and debris, Grey-brown, loose, Brown, compact aggregates, fragments of organic matter, Black, compact aggregates, fragments of organic matter, bricks, limestones, Grey-brown, compact aggregates, single short sticks, Light-grey, mixed with black, compact, fragments of ceramics (4 cm), Brown, black, pretty fragments of ceramics and organic matter, Brown-black, rich in organic matter, compact aggregates, fragments of bones, Brown-black, pretty fragments of organic matter, ceramics, bricks, limestones, Black, fine grained, compact, fragments of plants and stems, Brown-black, fragments of ceramics, clay, organic matter, shells of nuts, Grey and brown, silt, clay, fragments of ceramics (5 cm), organic matter, Brown, compact with aggregates, pretty fragments of ceramics (6 cm), Black, compact, big amount of organic matter, grain of quartz.

Figure 1.

Sampling site (arrow) of historical sequence layers from the western part of the Main Market Square. The profile was metrically divided and the samples taken from each 10 cm thick layer, along groove profile.

4 METHODOLOGY OF GEOCHEMICAL RESEARCH The way of geochemical treatment of the samples, coming from the explored by archaeologists areas, is always the same (Wardas et al 2006a, b, 2007). On the preliminary stage, while the grainsize and the phase composition are determined, the measurement of physic-chemical factors like pH and Eh takes place. For that purpose the water suspensions with proportion of sample to water 1:10 and 1:3, depending on the expected concentrations of contaminants, are prepared. In water − extracts, using Ionic Chromatography (IC) method, the concentrations of F− , Cl− , SO2− 4 , NO3 and 3− PO4 are measured. 279

Figure 2.

General view of the remains within sequence layers of underground infrastructure of Kraków.

The samples are usually collected directly on the expositions of archaeological strata or as a result of drilling-probe. The samples of ca. 500 g are air dried, diminished to the laboratory size (100 g) by quartering, and separated to fractions using plastics sieves of 2 mm, 1 mm, 0,5 mm and 0,18 mm. Sometimes, in order to assess the proportion of silt-clay and clay fractions, the sieves of 0,063 mm and 0,020 mm, respectively are used. The samples are treated 2 h with concentrated HNO3 at 130◦ C, to obtain extracts of heavy metals: Cd, Co, Cr, Ni, Cu, Pb, Mn and Fe. Their concentration is determined by AAS method. Analytical quality control is realized by analyses of double samples, repeated analyses, and reference analyses of certified samples. Additionally, the inter-laboratory calibration using AAS, ICP-AES and ICP-MS methods are performed. Especially carefully are treated the samples coming from the, so called, undisturbed layers, as a rule, list polluted.

5 RESULTS OF GEOCHEMICAL INVESTIGATIONS 5.1 Concentration of heavy metals Investigations of sequence layers, in the western part of the Main Market Square, showed a high variability of metals, particularly, Pb and Zn (Fig. 4). The levels of metal concentrations, especially Pb, revealed in numerous analyzed samples the many times higher values than observed in natural grounds, and even anthropogenic ones. Most distinct contamination is observed on the depth of 70–80 and 120–130 cm below the surface level. The higher content of metals in the surface layer is probably a result of its contamination by the earth taken up while excavation works. On the diagram, the two distinctly less polluted layer may be observed, on the depth of 90 and 150 cm. 5.2 Salinity The next diagram (Fig. 5) shows variable salinity of the grounds. The highest concentrations of water soluble Cl− anions represent the layers on the depth of 120 and 130 cm. Lying slightly higher, 110 cm deep, so younger layer, shows the highest concentration of SO2− 4 . Elevated concentrations of F− and NO− 3 were found in the highest layers. In others ones, their concentration were immeasurable by IC method. It was a consequence of the too high dilution of anions in the suspension used (1:10). That is why, the future investigations of that type will be continued with sample to water proportion 1 : 3. 280

Figure 3.

Sampling sites and areas of geochemical research of historical layers – Kraków – Centre, morphological reconstruction in the Early Middle Ages, the level of the primary soil – undisturbed layer (according to Radwa´nski 1974, Wardas & Such, in press., Wardas et al., in press).

are observed in the layer on Summarizing, the highest concentrations of Cl− and SO2− 4 the depth of 110 cm. On the depth of 150 cm, the highest conductivity of the samples has been registered. That suggests that some other anions, unmeasured by the method used are present there. 281

(mg / kg) 0

100

(mg / kg)

100000

0

0

200

400

600

800

0

Fe Mn

Ni 100 Cr

Pb

50

Cd depth (cm)

depth (cm)

50

100

Cu

Co 150

150 Zn

Figure 4.

Concentrations of heavy metals of the ground samples (fraction < 0,18 mm) of sequence layers in western part of the Main Market Square in Kraków.

20

40

40

50

50

60

60

70

70

cm depth

cm depth

cm depth

cm depth

cm depth

cm depth 20

20

20

20

40 50

40

40

50

50

60

60

60

70

70

70

60 70 80

20 40 50

80

80

80

90

90

90

100

80 90

80

90

90

100

100

100

100

110

110

110

110

110

100 110

120

120

120

120

120

120

130

130

130

130

130

130

140

140

140

140

140

150

150

150

150

150

140 150

160

160

160

160

160

160

170

170

170

170

170

0

1 2 mS/cm EC

Figure 5.

3

0

10 20 mg/dm3

30

Cl-

0

50 100 150 mg/dm3

0,0

0,3 0,5 mg/dm3 NO3-

S042-

0,8

0

170 0,1 mg/dm3 F-

0,2

0

50 100 150 mg/dm3 anions

Salinity of the ground samples (proportion of raw sample to water 1:10) of sequence layers in western part of the Main Market Square in Kraków.

5.3 Grain-size distribution and the part of organic matter In the figure 6, the grain-size distribution, and the part of organic matter are presented. It is visible that the percentage of organic matter and that of the finest fraction (