Parametric Analyses of High-Temperature Data for Aluminum Alloys

5202 PA FM i-vi.qxd

10/21/08

9:41 PM

Page i

Copyright © 2008 ASM International®. All rights reserved. Parametric Analyses of High-Temperature Data for Aluminum Alloys (#05202G)

www.asminternational.org

PARAMETRIC ANALYSES OF HIGH-TEMPERATURE DATA FOR ALUMINUM ALLOYS

J. GILBERT KAUFMAN

ASM International® Materials Park, Ohio 44073-0002 www.asminternational.org

5202 PA FM i-vi.qxd

10/21/08

9:41 PM

Page ii

Copyright © 2008 by ASM International® All rights reserved No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the written permission of the copyright owner. First printing, December 2008

Great care is taken in the compilation and production of this book, but it should be made clear that NO WARRANTIES, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE GIVEN IN CONNECTION WITH THIS PUBLICATION. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorable results will be obtained from the use of this publication alone. This publication is intended for use by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM’s control, ASM assumes no liability or obligation in connection with any use of this information. No claim of any kind, whether as to products or information in this publication, and whether or not based on negligence, shall be greater in amount than the purchase price of this product or publication in respect of which damages are claimed. THE REMEDY HEREBY PROVIDED SHALL BE THE EXCLUSIVE AND SOLE REMEDY OF BUYER, AND IN NO EVENT SHALL EITHER PARTY BE LIABLE FOR SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES WHETHER OR NOT CAUSED BY OR RESULTING FROM THE NEGLIGENCE OF SUCH PARTY. As with any material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific testing under actual conditions is recommended. Nothing contained in this book shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters patent, copyright, or trademark, and nothing contained in this book shall be construed as a defense against any alleged infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement. Comments, criticisms, and suggestions are invited, and should be forwarded to ASM International. Prepared under the direction of the ASM International Technical Book Committee (2007–2008), Lichun L. Chen, Chair. ASM International staff who worked on this project include Scott Henry, Senior Manager of Product and Service Development; Charles Moosbrugger, Technical Editor; Ann Britton, Editorial Assistant; Bonnie Sanders, Manager of Production; Madrid Tramble, Senior Production Coordinator; Diane Grubbs, Production Coordinator; Patty Conti, Production Coordinator; and Kathryn Muldoon, Production Assistant

Library of Congress Control Number: 2008934668 ISBN-13: 978-0-87170-715-4 ISBN-10: 0-87170-715-2 SAN: 204-7586

ASM International® Materials Park, OH 44073-0002 www.asminternational.org

Printed in the United States of America

5202 PA FM i-vi.qxd

10/21/08

9:41 PM

Page iii

Contents Data Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Wrought Alloys 1100-O, H14, H18—Stress Rupture Strength and, for the 0 Temper, Creep Strengths . . . . . . . . . . . . . . . . . . . . . . . . 23 2024-T851—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . . . . 41 2219-T6, T851—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . 50 3003-O, H12, H14, H18—Stress Rupture Strength . . . . . . . . . . . . 56 3004-O, H32, H34, H38—Stress Rupture Strength . . . . . . . . . . . . 64 5050-O—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . . . . . . . 68 5052-O, H32, H34, H38, and H112—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 5083-H321 As-Welded with 5083 Filler Alloy—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 5154-O—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . . . . . . . 82 5454-O, H32, H34, As-Welded H34—Stress Rupture Strength and, for the O Temper, Strength at Minimum Creep Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 5456-H321 As-Welded with 5556 Filler Alloy—Stress Rupture Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 6061-T6, T651—Stress Rupture Strength, Creep Strength, and Strength at Minimum Creep Rate . . . . . . . 112 6063-T5, T6—Strength at Minimum Creep Rate . . . . . . . . . . . . 142 Cast Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 A201.0-T7—Stress Rupture Strengths. . . . . . . . . . . . . . . . . . . . . 144 224.0-T63—Stress Rupture Strengths . . . . . . . . . . . . . . . . . . . . . 145 249.0-T63—Stress Rupture Strengths . . . . . . . . . . . . . . . . . . . . . 147 270.T7—0.2% Creep Strengths . . . . . . . . . . . . . . . . . . . . . . . . . . 148 354.0-T61—Stress Rupture Strengths . . . . . . . . . . . . . . . . . . . . . 148 C355.0-T6—Stress Rupture Strengths. . . . . . . . . . . . . . . . . . . . . 149

Foreword and Acknowledgments..........................................................iv About the Author ....................................................................................v Introduction and Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Theory and Application of Time-Temperature Parameters . . . . . . . . 3 Rate Process Theory and the Development of Parametric Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Larson-Miller Parameter (LMP) . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Manson-Haferd Parameter (MHP) . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Dorn-Sherby Parameter (DSP). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Observations on the LMP, MHP, and DSP. . . . . . . . . . . . . . . . . . . . 4 Illustrative Applications of LMP, MHP, and DSP . . . . . . . . . . . . . . . . 4 Notes about Presentation Format . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Alloys 1100-O and H14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Alloy 2024-T851 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Alloy 3003-O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Alloy 5454-O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Summary of Parametric Comparisons . . . . . . . . . . . . . . . . . . . . . . . 7 Factors Affecting Usefulness of LMP . . . . . . . . . . . . . . . . . . . . . . . . . 7 Normal Rupture Test Reproducibility . . . . . . . . . . . . . . . . . . . . . . . 7 Testing Laboratory Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Lot-to-Lot Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Effect of LMP Constant (CLMP) Selection . . . . . . . . . . . . . . . . . . . . 8 Choice of Cartesian versus Semi-log Plotting . . . . . . . . . . . . . . . . . 9 Choice of Scales and Precision of Plotting . . . . . . . . . . . . . . . . . . 10 Effect of How the LMP Master Curves are Fitted to the Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Applications When Microstructural Changes are Involved . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Illustrations of Verification and Limitations of LMP . . . . . . . . . . . . . 11 Alloys 1100-O and H14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Alloy 5454-O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Alloy 6061-T651 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Limitations of Parametric Analyses . . . . . . . . . . . . . . . . . . . . . . . . 12 Presentation of Archival Master LMP Curves . . . . . . . . . . . . . . . . . . 13 Software Programs for Parametric Analyses of Creep Rupture Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Application of LMP to Comparisons of Stress Rupture Strengths of Alloys, Tempers, and Products. . . . . . . . . . . . . . . . . . . 15 Comparisons of Stress Rupture Strengths of Different Tempers of an Alloy . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Comparisons of Stress Rupture Strengths of Different Products of an Alloy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Comparisons of Stress Rupture Strengths of Welds with Parent Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Comparisons of Different Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Application of LMP to High-Temperature Tensile Data for Aluminum Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Application of LMP to Microstructural Changes and Corrosion Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Appendix 1: Aluminum Alloy and Temper Designation Systems ......................................................151 Appendix 2: Terminology and Nomenclature ..................................153 Appendix 3: Nominal Compositions and Typical Mechanical Properties of Some Aluminum Alloys ..........................155 Appendix 4: SI/Metric Unit Conversions ..........................................159 Index ....................................................................................................161

iii

5202 PA FM i-vi.qxd

10/21/08

9:41 PM

Page iv

Foreword and Acknowledgment It is the objective of this book to describe the potential usefulness of parametric analyses in analyzing and extrapolating the properties of aluminum alloys at high temperatures. It is also the intent to illustrate the use of such methods by presenting a broad spectrum of high-temperature creep data for aluminum alloys generated from a single source and developed using consistent testing procedures and practices. The author gratefully acknowledges the support of Alcoa, Inc., and in particular the efforts of Dr. Gwendolyn Dixon and her management in arranging and approving the release of the information contained herein. Alcoa, Inc. enabled the author to include many previously unpublished data and related information from Alcoa’s archives that add immeasurably to the depth and breadth of coverage. The archival parametric analyses presented herein are illustrative examples typical and representative of the respective alloys and tempers, but have no statistical basis and therefore are not to be considered as the basis for design.

iv

5202 PA FM i-vi.qxd

10/21/08

9:41 PM

Page v

About the Author J.G. (Gil) Kaufman has a background of more than 50 years in the aluminum and materials information industries, and remains an active consultant in both areas. In 1997, he retired as Vice President, Technology for the Aluminum Association, Inc., headquartered in Washington, D.C., and is currently president of his consulting company, Kaufman Associates. Earlier in his career, he spent 26 years with the Aluminum Company of America, where he managed engineering properties and fabricating metallurgical research at Alcoa Laboratories. Many of the data presented in this volume were generated over the period when the author was active in and/or managing Alcoa Laboratories engineering properties research. Kaufman spent 5 years with ARCO Metals, where he was Director of R&D and, later, Vice President, Research & Engineering. Kaufman also served for 9 years as President and CEO of the National Materials Property Data Network where, working with STN International and Chemical Abstracts Service, he established a worldwide online network of more than 25 numeric materials databases. Gil is a Fellow and Honorary Member of ASTM and a Fellow and Life Member of ASM International. He has published more than 130 articles, including five books, on aluminum alloys and materials data systems.

v

Parametric Analyses of High-Temperature Data for Aluminum Alloys J. Gilbert Kaufman, p 1-2 DOI: 10.1361/paht2008p001

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

Introduction and Background The properties of aluminum alloys are dependent on both the temperature to which they are exposed and also, for temperatures above room temperature, to the length of time of exposure at temperature. One consequence of this is a need for designers of structures intended for very long-life service at high temperatures to be able to anticipate the combined effect of temperature and time at temperature on the properties for the entire service life based on data from relatively shorter time experimental testing. For relatively short-life structures, the need is addressed simply by planning ahead and carrying out a test plan that replicates the intended service conditions. This is quite practical for structures whose life may be as much as a year, or perhaps even 5 years, but does not generally cover rather typical design lives of 10, 20, or 30 years, or longer. The need for the ability to judge performance for relatively long service lives has been addressed for more than 50 years (Ref 1–3) through the use of time-temperature parametric equations that permit the folding of data obtained over a variety of temperatures and exposure times into a single relationship. Once that relationship is established with adequate consistency and reliability, it is possible to extrapolate the available data to anticipate service lives that substantially exceed the range of test data. This must always be done cautiously and with awareness of the extent of the extrapolation, but it provides a better perspective than simply extrapolating individual strength life curves. The need for some sort of parametric relationship involving stress, time, and temperature may be visualized readily by observing a representative set of stress rupture strength data for 5454-O in Fig. 5454-1 (Ref 4) plotted as rupture stress as a function of time under load at temperature. The data for each temperature appear as discrete lines of decreasing rupture stress with increasing time at temperature. Typically, such curves extend out to between 1000 and 10,000 hours because that represents the practical limits of testing time in advance of designing some commercial structure. Despite the individual lines for each test temperature in Fig. 5454-1, it appears intuitively that there is some relationship among these curves. It would be highly desirable and helpful in extrapolating to longer service lives if these curves could be combined and consolidated into a single relationship representing all of the data for all temperatures, as for example in Fig. 5454-2. It is precisely such consolidation that parametric analyses try to accomplish, and it is the background and application of such analyses that we discuss in this book. The theoretical background and development of the time-temperature parametric relationships are covered in greater depth, but

it is appropriate to introduce those that are the focus of this volume at this point. A number of fairly commonly used parameters have been developed over the years, and most are based on what is termed “rate process theory.” Three versions of such timetemperature parameters are dealt with herein, notably: The Larson-Miller parameter (Ref 1): LMP = T (C + log t)

The Manson-Haferd parameter (Ref 5, 6): MHP =

Log t − log t a T − Ta

The Dorn-Sherby parameter (Ref 7): DSP = t e⫺A/T

where T is the temperature, °R; t is the time at temperature; and ta, Ta, A, and C are constants defined by the respective experimenters. These parameters have been applied with considerable success over the years, especially to stress rupture data for a variety of metals and, to a lesser extent, to creep rates and total accumulated creep of various amounts. While not necessarily showing a technical advantage over the other two parameters, the Larson-Miller Parameter (LMP) has become the most widely used, for aluminum alloys at least, primarily in the author’s judgment because of its ease and simplicity of application. As a result, the bulk of the information presented herein focuses on the LMP. It is timely and useful to consider the value of parametric relationships not only for creep and stress rupture data but also for elevated temperature tensile property data and even for resistance to stress-corrosion cracking. For this purpose we focus on data for aluminum alloys, notably those used for ASME Boiler & Pressure Vessel Codes and for other high-temperature applications and also for alloys subjected through service exposure to long-term hightemperature service. Briefly, then, the scope of this book includes:

• • • • • • •

Review of the theoretical basis for the parametric relationships Some illustrations and comparisons of the application of three parametric relationships for several aluminum alloys Factors affecting the usefulness of time-temperature parameters Illustrations of the verification and limitations of time-temperature-parameters Presentation of archival Larson-Miller parametric analyses to stress-rupture data for a variety of aluminum alloys Presentation of some new analyses of archival data Application of LMP to comparisons of the stress rupture strengths of aluminum alloys, tempers, and products

2 / Parametric Analyses of High-Temperature Data for Aluminum Alloys

• •

Application of the parametric relationships to high-temperature tensile properties of aluminum alloys Applications of the parametric relationships for anticipating microstructural changes in aluminum alloys

It is appropriate to note that throughout this volume, the Aluminum Association alloy and temper designation systems are used and that the compositions and tensile properties of the materials for which data are presented herein met all aluminum alloys specifications of the Aluminum Association. Both the alloy and temper designation systems and the composition and tensile property specifications for all aluminum alloys are presented in Aluminum Standards and Data published by the Aluminum Association and updated on a regular basis (the current issue at this writing was published in 2006) and also in the American National Standards Institute publications H35.1 and H35.2, published for ANSI by the Aluminum Association. A brief summary of the Aluminum Association Alloy and Temper Designation Systems is presented in Appendix 1. A list of the aluminum industry terminology and nomenclature used throughout the volume is presented in Appendix 2. Most of the industry terms are those from Aluminum Standards and Data. There are a few abbreviations used regularly in the text, tables, and figures:

• • • • • • • •

LMP, Larson-Miller parameter MHP, Manson-Haferd parameter DSP, Dorn-Sherby parameter CLMP, the constant C in the Larson-Miller parameter T, test temperature t, time at test temperature AW, as welded HTAW, heat treated and aged after welding

Appendix 3 provides for background information the nominal compositions and typical mechanical properties of all of the

aluminum alloys and tempers for which creep rupture data are presented or referenced in this volume. It is also appropriate to note that throughout most of this volume, principal focus is placed on the calculation and presentation of mechanical properties in the English or engineering system, rather than the International Standard System of Units (SI) or metric units. This was done because all of the tabular and graphical data presented herein were generated using the English/engineering system, calculated conversions other than when convenient would have added the potential for distortion of the presentations. For those interested in a more in-depth discussion of SI/metric units and their use in parametric analysis, see Appendix 4. REFERENCES 1. F.R. Larson and J. Miller, A Time-Temperature Relationship for Rupture and Creep Stresses, Trans. ASME, Vol 74, July 1952, p 765–771 2. F.C. Monkman and N.J. Grant, An Empirical Relationship between Rupture Life and Minimum Creep Rate in Creep Rupture Tests, Transactions of 59th Annual Meeting of ASTM, ASTM, Philadelphia, PA, 1956, p 593–605 3. J.G. Kaufman, Discussion Ref 2 in Transactions of 59th Annual Meeting of ASTM, Philadelphia, PA, 1956, p 606–612 4. K.O. Bogardus, R.C. Malcolm, and M. Holt, “Extrapolation of Creep-Rupture Data for Aluminum Alloys,” presented at the 1968 ASM Materials Engineering Congress (Detroit, MI), D8-100, American Society for Metals, 1968, p 361–390 5. S.S. Manson, “Design Considerations for Long Life at Elevated Temperatures,” Technical Report TP-1-63, NASA, 1963 6. S.S. Manson and A.M. Haferd, “A Linear Time-Temperature Relation for Extrapolation of Creep and Stress-Rupture Data,” Technical Note 2890, NACA, March, 1953 7. O.D. Sherby and J.E. Dorn, Creep Correlations in Alpha Solid Solutions of Aluminum, Transactions of AIME, Vol 194, 1952

Parametric Analyses of High-Temperature Data for Aluminum Alloys J. Gilbert Kaufman, p 3-21 DOI: 10.1361/paht2008p003

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

Theory and Application of Time-Temperature Parameters Rate Process Theory and the Development of Parametric Relationships Much of the early application and evolution of the high-temperature parametric relationships to data for aluminum alloys were carried out during the 1950s and 1960s under the auspices of the MPC, then known as the Metals Properties Council (now the Materials Properties Council). However, the real origins of the relationships go back considerably further. The “rate process theory” was first proposed by Eyring in 1936 (Ref 1) and was first applied to metals by Kauzmann (Ref 2) and Dushman et al. (Ref 3). It may be expressed mathematically as: r AeQ(S)/RT

where r is the rate for the process in question, A is a constant, Q(S) is the activation energy for the process in question, R is the gas constant, and T is absolute temperature. Over the years from 1945 to 1950, several investigators, including Fisher and McGregor (Ref 4, 5), Holloman (Ref 6–8), Zener (Ref 7), and Jaffe (Ref 8) were credited with recognizing that for metals high-temperature processes such as creep rupture performance, tempering, and diffusion appear to obey rate process theories expressible by the above equation. In 1963, Manson and Haferd (Ref 9) were credited with showing that all three of the parametric relationships introduced in the section “Introduction and Background” derive from: P=

(log t ) σ Q − log t A (T − TA ) R

where P is a parameter combining the effects of time, temperature, and stress; s is stress, ksi; T is absolute temperature; and TA, log tA, Q, and R are constants dependent on the material.

Larson-Miller Parameter (LMP) For the LMP, Larson and Miller (Ref 10) elected to use the following values of the four constants in the rate process equation: Q0

R 1.0 TA 460 °F or 0 °R tA  the constant C in the LMP

Thus, the general equation reduces to: P  (log t + C) (T) or LMP  T(C + log t)

This analysis has the advantage that log tA or C is the only constant that must be defined by analysis of the data in question, and it is in effect equal to the following at isostress values: C  (LMP/T)  log t

In such a relationship, isostress data (i.e., data for the same stress but derived from different time-temperature exposure) plotted as the reciprocal of T versus log t should define straight lines, and the lines for the various stress values should intersect at a point where 1/T  0 and log t  the value of the unknown constant C. Larson and Miller took one step further in their original proposal, suggesting that the value of constant C (referred to as CLMP hereinafter) could be taken as 20 for many metallic materials. Other authors have suggested that the value of the constant varies from alloy to alloy and also with such factors as cold work, thermomechanical processing, and phase transitions or other structural modifications. From a practical standpoint, most applications of the LMP are made by first calculating the value of CLMP that provides the best fit in the parametric plotting of the raw data, and values for aluminum alloys, for example, have been shown to range from about 13 to 27.

Manson-Haferd Parameter (MHP) For the MHP, Manson and Haferd (Ref 9, 11) chose the following values for the constants in the rate process equation: Q0 R  1.0

4 / Parametric Analyses of High-Temperature Data for Aluminum Alloys Under these assumptions, the general equation reduces to: P=

log t − log t A T − TA

In this case, there are two constants to be evaluated, log tA and TA. Manson and Haferd proposed that isostress data be plotted as T versus log t and the coordinates of the point of convergence be taken as the values for log tA and TA. It may be noted that the key difference between the LMP and MHP approaches is the selection of TA  absolute zero as the temperature where the isostress lines will converge in the LMP while in the MHP TA is determined empirically, or in effect allowed to “float.”

Dorn-Sherby Parameter (DSP) Dorn and Sherby (Ref 12) based their relationship more directly on the Eyring rate-process equation: DSP  teA/T

where t is time, A is a constant based on activation energy, and T is absolute temperature. This relationship, like the others, implies that isostress tests results at various temperatures should define straight lines when log t is plotted against the reciprocal of temperature. However, it differs from the other approaches in that these straight-line plots are indicated to be parallel rather than converging at values of log t and 1/T.

Observations on the LMP, MHP, and DSP The essential significance of the differences in the three parameters described previously and applied herein may be illustrated by the schematic representations in Fig. 1 based on the

relationship assumed of the relationships between log t and 1/T (Ref 6). As noted in the previous discussions, the LMP assumes that the isostress lines converge on the ordinate of a log time versus inverse temperature plot, while the MHP assumes convergence at some specific value of both log t and 1/T. The DSP assumes the isostress lines are parallel rather than radiating from a specific value of coordinates log t and 1/T. As representative data illustrated in this book show, the impact of the differences on the results of analyses with the three different parameters is not very great. It is appropriate to note that a number of variations on the three parameters described previously have been proposed, primarily including such things as letting the values of the various constants, such as the C in the LMP and the activations energy A in the DSP, “float.” None of these have seemed a useful extension of the originals. It is common practice to use the available raw data to calculate or determine graphically the values of the needed constants, but then once established to hold them constant. Allowing the constants in any of the relationships to float, for example, the activation energy in the DSP, results in a different type of analysis in which the isostress lines are curves, not straight lines, and considerably complicates its routine use.

Illustrative Applications of LMP, MHP, and DSP Several interesting facets of the value and limitations of the parametric relationships may be seen from looking at representative illustrations for the following four alloys and tempers where all three parameters are applied to the same sets of data.

• •

σ1 < σ2 < σ3 < σ4 σ4

σ4

ta, Ta

σ3

σ3 σ2

σ4

σ1

0

1/T

σ2

σ3

σ1

0

σ1

σ2

T

0

1/T

Comparison of assumed constant stress versus temperature relationships for Larson-Miller (left), Manson-Haferd (center), and Dorn-Sherby (right). T, exposure temperature, absolute; t, exposure time, h; σ, test/exposure stress.

Fig. 1

•

•

1100-O, commercially pure aluminum, annealed (O) 2024-T851, a solution heat treated aluminum-copper (Al-Cu) alloy, the series most widely used for high-temperature aerospace applications. The T851 temper is aged to peak strength, so subsequent exposure at elevated temperatures results in overaging, and some microstructural changes may be expected. 3003-O, a lightly alloyed non heat treatable aluminum-manganese (Al-Mn) alloy, widely used for heat exchanger applications. It is annealed so no further transitions in structures are anticipated as it is further exposed to high temperatures. 5454-O, the highest strength aluminum-magnesium (Al-Mg) alloy recommended for applications involving high temperatures. Because of the higher alloying, there may be diffusion of constituent with high-temperature exposure even in the annealed temper.

Many other alloys and tempers are included in the group for which master parametric relationships are presented in the section “Presentation of Archival Master LMP Curves.” It is appropriate to note that some components of the following presentations are based on the efforts of Bogardus, Malcolm, and Holt of Alcoa Laboratories, who first published their preliminary assessment of these parametric relationships in 1968 (Ref 13).

Theory and Application of Time-Temperature Parameters / 5

Notes about Presentation Format Generally, plots of stress rupture strength or any other property are presented with the property on the ordinate scale and the parameter on the abscissa, as in Fig. 1100-8. From the descriptions in Chapter 2, all three of the parameters discussed herein include both time and temperature, so it is useful to note that the parametric plots can also be presented as in Fig. 2043-3, 2024-6, or 2024-7, examples of the three parameters in which at the bottom, abscissa scales showing how the combination of temperature and time are represented. This type of presentation is often useful for individuals using the parameters for extrapolations, but it is not a necessary part of the presentation. Therefore, the multiple abscissa axes showing time and temperature are not included as a general rule through this volume unless the archival version included them. It is also appropriate to clarify at this stage that the values shown for the Larson-Miller parameter on the abscissas are in thousands and are presented as LMP/103; thus for example, in Fig. 1100-8, the numbers from 13 to 21 on the abscissa are actually 13,000 to 21,000. For the Manson-Haferd and Dorn-Sherby parameters, the values are as shown.

Alloys 1100-O and H14 Table 1100-1 presents a summary of the stress rupture strength data for 1100-O and 1100-H14; the discussion immediately following focuses on the O temper data. This summary is for rather extensive tests of single lots of material. Other lots of 1100 were also tested, as is illustrated later, but this material was the basis of the best documented master curves for 1100-O and H14. The data are plotted in the format of stress rupture strength as a function of rupture life in Fig. 1100-1 and Fig. 1100-2 for the O and H14 tempers, respectively. LMP for 1100-O. Figure. 1100-3 shows the archival master LMP curve developed for 1100-O derived with a value of the Larson-Miller parameter constant CLMP of 25.3. The isostress calculations leading to the selection of this value of CLMP no longer exist. Scatter and deviations are small, and the curve appears to represent the data reasonably well. MHP for 1100-O. The isostress plot of log t and temperature is shown in Fig. 1100-4. The isostress lines are not straight nor do they seem to converge as projected by Manson and Haferd, but values of the constants may be judged from projections of the straight portions of the fitted lines as: log tA = 21.66 and TA = –500. The resultant master MHP curve is illustrated in Fig. 1100-5. With the exception of several points obtained in tests at 250 oF, the fit is reasonably good. DSP for 1100-O. Calculations of the activation energy constant for the DSP resulted in a value of 44,100, and the resultant master curve is illustrated in Fig. 1100-6. With the exception of the data for the lower temperatures, the fit is reasonably good. Comparisons of the Parameters. All three parametric relationships represent data for 1100-O reasonably well. An additional useful comparison test is the degree of agreement in extrapolated values for predicted rupture life after 10,000 and 100,000 h:

Temperature, °F

212 300 400 500

Desired service rupture life, h

10,000 100,000 10,000 100,000 10,000 100,000 10,000 100,000

LMP rupture strength, ksi

6.0 5.3 3.7 3.0 2.5 2.1 1.1