What Every Engineer Should Know About Threaded Fasteners Materials and Design Alexander Blake Lawrence Livermore National Laboratory University of California Livermore, California ii. Mancet Dexxer, Inc. New York « BasetLibrary of Congress Cataloging-in-Publication Data Blake, Alexander. Threaded fasteners. (What every engineer should know ; vol. 18) Includes index. 1, Fasteners. 2. Screw-threads, I. Title. IL Series: What every engineer should know ;¥. 18. TJ1338.B57 1986 621.8°82 86-13527 ISBN 0-8247-7554-6 COPYRIGHT © 1986 by MARCEL DEKKER, INC. ALL RIGHTS RESERVED Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without per- ‘mission in writing from the publisher. MARCEL DEKKER, INC. 270 Madison Avenue, New York, New York 10016 Current printing (last digit): 10987 PRINTED IN THE UNITED STATES OF AMERICA1 | Introduction Volumes can be written about the development of mechanical fasteners, which date back to Herculaneum in 79 A.D. The use of a bolted joint, however, became truly practical during the 18th century. Since then much valuable information on fastener design, testing, manufacture, and service has been developed in such countries as the United States, United Kingdom, Germany, Holland, Austria, Japan, France, Belgium, Switzerland, and Italy. ‘The time interval between the Second World War and the close of the 1960s ‘was most productive and it might be termed the “golden era” of fastener development. About half a century of progress in developing fastener standards in the United States has centered around the procedures of the American National ‘Standards Institute, Inc. (ANSI) and the Industrial Fasteners Institute (IFI). The design engineers and managers of manufacturing and construction indus- tries can consult ANSI and IFI standards covering dimensions, geometry, and practice for a great number of mechanical fasteners. Further detailed infor- ‘mation is also available from engineering and production handbooks. This volume provides a guide to the dimensional, geometrical, and materials as- pects of the fastener, together with fundamental design formulas, as a ser-2 Chapter 1 vice to practical engineering designers faced with technical and economical decisions in relation to the performance of mechanical connections. The following additional acronyms are used throughout this book: AISI American Iron and Steel Institute ASTM American Society for Testing and Materials CDA Copper Development Association ISO _ International Organization for Standardization SAE Society of Automotive Engineers UNS — Unified Numbering System (for Metals and Alloys) Although the mechanics of a bolted joint, for instance, is not a mundane ‘engineering topic there are several essential features of geometry, materials, and design of fasteners which the generalist can appreciate without being an experienced metallurgist, designer, or stress analyst. This book assumes little specialized knowledge in the field and it is intended to serve the practical needs of a design office, test laboratory, and production department. The material covers the areas of definitions, dimensions, thread forms, strength properties, fastener grades, materials, and elementary design formulas sufii- cient to make a procurement request or a preliminary design estimate. The area of materials selection is particularly important for design purposes under various environmental conditions. For this reason the book includes many practical details of material properties, composition, fabrication, and classi fication. The attempt is made to offer a complete package of information that an engineer should have for his every day encounters with fasteners. ‘The name bolts is often extended to all threaded fasteners such as studs and machine screws, and the number of different fasteners defined by the various standards in the world is truly enormous. This situation, combined with the long and tedious process of conversion of all specifications to one international standard, creates additional difficulties in correlating informa- tion on the theory and practice of fastener behavior. This problem is complicated further by the fact that the methods of deal- ing with fastener behavior are of a statistical nature. The larger the total Population of fasteners and joints we are trying to understand the more data appears to be needed to get a reasonable prediction of fastener performance. In the real world, however, we can only afford to test a few samples. We are, of course, interested in the scatter of test results and the more variation we find, the less certainty we have about our estimates. The grouping of test results may or may not follow the so-called normal distribution diagram. Such diagrams can give us the overall picture of fastener behavior in whichIntroduction 3 it may be difficult to zero in on a particular variable. If we can have a relatively homogeneous population of the properties tested, then 25 test samples should provide a reliable result. Although many tests have been documented for individual threaded fast- ‘eners, and considerable progress has been made in numerical analysis, our knowledge of fastener behavior continues to be incomplete. Reliance, there- fore, has to be based on our own experience, engineering handbooks, and regulatory documents. The areas of new material properties and the quality of fabrication are particularly difficult to define. The specifications and standards developed in many countries are intended to reduce the cost of this process, and to promote reliability. ‘Several typical examples of bolt configurations are shown in Fig. 1. The design of these fasteners should follow a number of standard procedures such as, for instance, the following: Determine the external force for the most highly loaded fastener Select a factor of safety based on the type of application and best en- gineering judgment Select a fastener material for given environmental and economic con- straints Select fastener dimensions Calculate the required tightening torque Figure 1. Typical bolt configuration.4 Chapter 1 Figure 2. Example of a complex shape fastener. The fasteners shown in Fig. 1 have uniform cross-sections. The problem of design and application however can be somewhat different when the geom- etry of the shank varies as indicated in Fig. 2. It is clear from Fig. 2 that in addition to the changes in fastener diameter, the transition radii may become ‘of importance in selecting the maximum allowable axial load on the fast- ener. ‘The condition of shear loading on a particular fastener depends on the application and the number of shear planes involved. The cross-section in the case of a threaded fastener can be either through the main body of the Figure 3. Example of fastener in double shear.Introduction shank or the threaded region, The application shown in Fig. 3 involves both types of cross-sections. While our primary interest may be tensile, rather than shear strength of the fastener it is important to recognize that the tensile strength may be reduced when torsional or shear loads are present. Such effects, however, are best determined experimentally.2 Definitions of Screw Threads Screw threads are remarkably complex machine elements; for that is exactly what they are. When properly assembled they have a capability to develop and transfer tremendous forces. There are over 125 separate geometrical features and dimensional characteristics in the design and construction of screw threads, each with its own term and definition. However, with a familiarity of only about 30 the engineer can be comfortably conversant in the language of screw threads and, more importantly, can have a good working understanding of their function and performance capabilities A screw thread is a ridge of uniform section in the form of a helix on the external or intemal surface of a cylinder. External threads are screw threads on bolts, screws and studs; internal threads are in nuts and tapped holes. The configuration of the thread in an axial plane is its profile and the three parts making up the profile are the crest, root, and flanks (see Fig. 4), The crests of threads are at the top, the roots at the bottom and the flanks join them. A thread having full form at both crests and roots is a complete (ot full form) thread. When either crest or root is not fully formed it is an incomplete thread. Incomplete threads occur at the ends of extern- ally threaded products which are pointed, at the runout where the threads8 Chapter 2 Mininum major diem. ¥ 7 Ls Maximum minor diam at intersection v Minor diam. | of root and flank contour ‘oot ootona 90°) Axis of external thread Figure 4 Unified thread form. blend into an unthreaded shank, or at the countersinks in the faces of nuts and tapped holes. On an external thread, the diameter at the thread crest is the major diam- eter, and at the thread roots is the minor diameter. On an internal thread it is just the opposite, the diameter at the thread crest is the minor diameter and that at the thread root is the major diameter. The angle between a flank and « perpendicular to the thread axis is the flank angle. When both flanks haveDefinitions of Screw Threads 9 the same angle we have a symmetrical thread with the flank angle termed ‘half-angle of thread. Unified and ISO metric threads have a 30° flank angle and are commonly known as 60° threads because this is the included angle between their thread flanks. The fundamental triangle height 1 is the height of the thread when the profile is extended to a sharp-V form. For a 60° thread H equals 0.866025 times the thread pitch. Thread height is the distance measured radially be- tween the major and minor diameters. In American practice thread height is usually expressed as a percentage of 0.75H. Several design details of unified screw threads are given in Fig. 4. Thread pitch (p)s the distance measured parallel to the thread axis be- tween corresponding points on adjacent threads. Unified threads are desig- nated in threads per inch, consequently their thread pitch is the reciprocal of the number of threads per inch. Metric threads, however, are designated using actual thread pitch. Thread series are groups of diameter-pitch com- binations distinguishing one from the other by the number of threads per inch or thread pitch applied to a series of specific diameters. For fasteners {n general the most popular inch thread series are coarse and fine. Often used metric series involves coarse thread. Pitch diameter is the diameter of a theoretical cylinder that passes through the threads in such a position that the widths of the thread ridges and thread grooves are equal. On a perfect thread these widths should each equal one- half of the thread pitch. The thread axis is the axis of the pitch cylinder. With anything less than a perfect thread, the actual pitch diameter, as meas- ured at any position throughout the length or circumference of the thread, will vary depending on the variations in manufactured thread form within the permitted limits of size. Consequently, the definitions of measurement and significance of pitch diameter are quite controversial among screw thread experts. At the same time these values are important for computation and thread design purposes, production of manufacturing tools and dies as well as thread acceptance gages and gaging. The pitch line is the generator of the pitch cylinder. ‘The basic thread profile establishes an absolute boundary between the product external and internal threads. If either trespasses beyond this boundary then potential interference exists and product threads may not, assemble. It is from this basic thread profile that the actual limits of size are derived by applying allowances and tolerances. An allowance is an in- tentional clearance between the mating threads. In other words, when both the external and internal threads are manufactured to their absolute10 Chapter 2 ‘maximum material condition there will be a finite space between them. For fasteners the allowance is generally applied to the external thread which ‘means that its maximum diameters are less than basic by the amount of the allowance. The minimum diameters of internal threads (its maximum ma- terial condition) are basic. Tolerances are specified amounts by which di- mensions are permitted to vary for manufacturing convenience. The toler- ance is the difference between the maximum and minimum permitted limits. Thus, for external threads, the maximum material condition less the toler- ances (moving toward the thread axis) defines the minimum material condi- tion. For internal threads, the maximum material condition plus the relevant tolerances (moving away from the thread axis) define the minimum material condition, ‘The combination of allowances and tolerances in mating threads is fit which is a measure of tightness or looseness between them. A clearance fit is one that always provides a free running assembly. On the other hand an interference fit is characterized by specified limits of thread size which en- sure a positive interference between the threads when assembled. When assembling externally threaded fasteners into internally threaded nuts or tapped holes the axial distance through which the full threads of each are in contact is length of thread engagement. The distance the mating threads overlap in a radial direction is depth of thread engagement. Itis determined by the major diameter of the external thread minus the minor diameter of the internal thread, ‘Strengths of screw threads, that is their ability to support and transfer loads, are dependent on four stress areas, However there are five possible thread failure modes, three of which are critical and two that are not too common. The externally threaded part can fracture in tension through its threads, the external thread can strip off, and the internal thread can strip ‘out. The other two modes are shear through the threaded section normal to the thread axis or torsion, and twist-off due to overtightening. Tensile stress area is the assumed area through the thread used to compute the tensile load carrying capacity of fasteners, The tensile stress area is equivalent to the cross-sectional area of a theoretical cylinder of the same material and mechanical properties capable of supporting the same ultimate tensile load. Thread shear area is the effective area through the thread ridges, parallel to the thread axis and spread over the full length of the thread engagement supporting the applied load in shear. The shear plane for internal threads occurs at the major diameter of the external thread, and for extemal threadsDefinitions of Screw Threads " at the minor diameter of the internal thread, Thread root area is the cross- sectional area through the external thread of its minor diameter. This area is used to evaluate the susceptibility of the thread to either a cross-sectional or torsional shear failure.Thread Forms ‘There are literally dozens of different screw thread forms. However, for con- ventional mechanical fasteners only the following five forms have signifi- cance: UN, UNR, UNS, M, and MJ. All are defined as 60° threads of essen- tially the same basic profile. The principal differences between the various forms can be traced to the contour of the root of the external thread. Prior to 1948 the American National thread form (Fig. 5) was the stan- dard in North America, In that year, the United States, Canada, and Great Britain agreed to adopt a single screw thread system to supplant the American ‘National in the United States and Canada, and the Whitworth in Great Brit- ‘ain, ‘The new system was called Unified and it is currently recognized as the standard throughout the world for all inch series fasteners. In a metric type screw thread (Fig. 6) the radius is expressed in terms of the triangle height H. ‘The Unified thread form—designated UN-is practically identical with the ‘now obsolete American National. In fact, threads manufactured to either of the standards are functionally interchangeable. The UN thread form, as ori- sinally designed, provided for either flat or rounded roots in the external thread at the option of each country’s national standards. The United States chose to permit flat roots, even though there was a complete understanding that stress concentrations in threads could be lowered by rounding their 134 Chapter 3 Figure 5 American standard screw thread. roots. However, economics dictated. Thread-roll dies and thread cutting tools at that time were expensive, particularly in fabrication of the radiused crests, Also influencing this decision was the production fact-of-life that new tools wear so that within the manufacture of few hundred parts the thread crests of the tool becomes rounded off. Consequently the roots in the pro- duct thread no longer remain sharp, but take on a contoured form. Figure 6 International screw thread.Thread Forms 15 AAs performance demands increased, particularly for safety-critical fast- ‘eners subjected to fatigue-inducing loads, it became imperative to search out ‘opportunities for fatigue behavior improvement. An obvious answer was to no longer depend on tool wear but to actually require controlled radiusing of the external thread root. This led to the design and introduction of a modified thread form, designated UNR, with the single difference from UN being a mandatory root radius having the limits of 0.108 to 0.144 times the thread pitch. The minimum radius of 0.108p is the largest radius that can be fitted into the UN profile without violating the minimum material condition of the external thread, and 0.144p is the largest radius that can be accomo- dated without causing theoretical interference with an internal thread at its ‘maximum material condition. The UN thread completely encompasses the UNR form. Again, the only difference is that UNR external threads must have a radiused root. ‘When first introduced in the late 1950s it became necessary to specify the UNR form to assure delivery of fasteners with radiused roots. However, today virtually 100% of all inch series fasteners in nominal sizes 1 in. and smaller are manufactured with radiused roots whether UNR is specified or not. This is because threads in this size range are today normally produced by thread rolling utilizing thread roll dies with standard rounded crest For larger size fasteners, which are beyond the capability of thread-rolling ‘equipment and whose threads are cut, UNR should be specified whenever the radiused form is needed. Otherwise the manufacturer will probably supply the UN form of the fastener. Even before UNR threads were introduced, extensive research was in progress to develop an optimum thread form, one that would have superior fatigue resistance without sacrificing the static strength. In other words the ‘question was just how generous could the root radius be to satisfy the above criteria. The answer was a new thread form, designated as UNS. UNJ threads are basically UNR thread forms with a larger mandatory root radius having the limits of 0.150 to 0.180 times the thread pitch. Because of the enlarged radiusing it became necessary to modify the internal thread by increasing its minor diameter in order to prevent any possibility of interference with the external thread root during assembly. UNJ threads are now standard for all fasteners used in aerospace and many other critical applications. UN internal threads assemble correctly with UN and UNR external threads. Theoretically, UN internal threads will not assemble with UNJ ex- ternal threads. However, this combination has been used by major fastener users for several years now claiming trouble-free experience. Computer16 Chapter 3 studies also substantiate that the statistical risk of actual interference between the manufactured parts is negligible. Even so, this practice is not recom- mended especially if the fasteners are plated. UNJ internal threads assemble correctly with UNJ external threads and also fit both UN and UNR. How- ever, this latter mating should be used with a degree of caution because the increased minor diameter of the UNJ internal thread reduces the stripping area of the extemal thread. In 1969, ISO 68 was published establishing for the first time the basic thread profile of an international screw thread system. This profile, which is now commonly designated M, has exactly the same geometry as the UN thread profile, and like the UN form it permits a flat root in the external thread with a special recommendation of a minimum radius of 0.1 times thread pitch, During the follow-up ISO Meetings North American spokes- men argued in favor of adopting as the world standard for metric commercial fastener threads a metric converted UNI thread form, Delegates from other countries, while agreeing with the potential of beneficial effects of root radiusing, were reluctant to accept the enlarged UNJ roots because of the ‘accompanying need to alter the internal threads as originally defined in 1SO 68. The eventual compromise was that M profile threads would have ‘a mandatory rounded root in their external thread with a minimum radius of 0.125 times thread pitch for all fasteners having specified tensile strengths of 800 MPa (116 ksi) and greater. For lower strength fasteners the root would preferably be rounded but it could also remain flat. In the United States virtually all metric fasteners in nominal thread diameters M24 and smaller are roll-threaded and consequently have radiused roots regardless of their strength properties. Unlike UN however, for sizes larger than M24 all higher strength fasteners must have radiused roots. ‘As metric interest intensified in the United States it became apparent that there would be a need for a metric counterpart of the UNJ thread form. ‘Such a thread, designated MJ, was introduced in North America in the early 1970s and has since been accepted in principle for adoption internationally as an ISO standard. MJ threads are identical in their geometry to UNJ form. ‘Thus there are just two metric screw thread forms, the M profile which is the standard for commercial fasteners and the MJ which is the standard for aerospace quality fasteners. Its interesting to reflect on the five thread forms for fasteners just dis- cussed and to observe that they are essentially the same with the one excep- tion involving root contour in the external thread. In summary then, while‘Thread Forms 7 in the case of UN threads there is no mandatory minimum root radius, UNR requires 0.108p, M profile has 0.125p, and UNJ and MJ forms recommend 0.150p. It is difficult to believe that such modest differences could be im- portant, but they are. Radiusing the root of external threads adds only slightly to the tensile strength of a fastener subjected to static tensile loading. And the reason is cone of simple geometry. As the root radius increases minor diameter in- creases and the cross-sectional area through the thread becomes greater. However, the amount of area growth is so small that the same tensile stress area formula is used in stress calculations for all thread forms. Itis rare to find an assembly joined by mechanical fasteners which is not, to some degree, exposed to dynamic loading during its service life. Extreme- ly few mechanical joints remain static and totally insulated from some form of fluctuating stress, vibration, stress reversal or impact. Fortunately, in only 1 small percentage of all joints are the fatigue resistant properties of the fast- ener itself the primary design consideration. When they are, no opportunity for improvement can be ignored. And itis here that root radiusing really counts. The larger the root radius, the better the fatigue resistance of the fastener. Fatigue failures of stressed parts generally occur at locations of high stress concentration, such as notches or abrupt changes in cross-sectional configuration. Screw threads with their large variations in cross-section and their thread roots which act as notches are particularly susceptible to fatigue. The highest stress concentrations in threads occur at the roots and the magnitude of the stress concentration factor relates directly to whether the root is rounded and to what degree. ‘Computation of thread stress concentration factors is an exceptionally ‘complex exercise and the answers are not always reliable. Consequently, physical research programs, including photoelastic studies, have been con- ducted to investigate the influence of root radiusing on fatigue properties of threaded sections. A generalized conclusion is that when all other vari- ables, such as fastener size, thread pitch, material, manufacturing method, ctc., are uniform with the only variable being root radius, stress concentra- tion factors can be reduced from about 6 for sharp or flat roots to less than 3 for UNJ and MJ threads. Translated this means a possible doubling of fatigue life due to root radiusing. Research has additionally demonstrated that this potential applies to all fastener strength levels.18 Chapter 3 Roots of internal threads are not normally rounded. Mandatory root radiusing would mean use of taps with rounded crests and this would add unnecessarily to their cost. In a properly designed bolt/nut combination the rut is the stronger member of the team with the intent that should failure ‘occur it will invariably be the externally threaded member.A Comparison of Thread Series ‘Thread series are groups of diameter/pitch combinations differing from each other by the number of threads per inch (or thread pitch) applied to a series of diameters. ‘The current Unified coarse thread series (UNC) is patterned on the thread series introduced by Whitworth in the mid-19th century. The relatively coarse pitches he selected were probably chosen as much for being those which lent themselves to the manufacturing skills of the day as for any other single reason. Over the years, as production capabilites improved, it became ‘economically feasible to produce threads to greater degrees of accuracy and with finer pitches. Many special purpose thread series were developed and the one now known as the Unified fine thread series (UNF) enjoys consider- able popularity. As screw thread technology advanced it became evident that just two thread series, coarse and fine, would be inadequate to efficiently satisfy all engineering applications. Consequently, a number of constant pitch thread series were added to the system. Such series have a single thread pitch which is common for all diameters in the series. Metric thread systems evolved in a similar pattern and the current ISO system features a coarse and fine thread series supported by a number of constant pitch series. 1920 Chapter 4 Today there are 11 standard Unified thread series for inch products and 13 for metric. Fortunately, for fasteners only four have genuine impor- ‘tance—Unified coarse (UNC), fine (UNF) and 8-thread (SUN) in inch and metric coarse (M) in metric. Few aspects of fastener engineering have been debated more intensely, and frequently with vehemence, than the pro and con merits of coarse versus fine threads. For a given diameter, fine threads are stronger in tension because of their larger tensile stress area. For the same conditions coarse threads have ‘greater stripping strength over a given length of thread engagement. Because of their larger minor diameter fine threads develop higher torsional and trans- ‘verse shear strengths. Stress concentration factor at thread roots decreases as. the thread pitch increases. ‘This suggests that, al else being equal, coarse threads should exhibit a superior fatigue resistance. Fine threads are better for tapping thin walled members while coarse threads protect against deleterious loss of thread overlap due to nut dilation. Also fine threads are easier to tap in hard materials since less metal is being removed. On the other hand coarse threads are better for tapping brittle materials which tend to crumble or spall. Strength loss due to corrosion attack is not as harmful to coarse threads. Additionally coarse threads have slightly larger allowances in clearance fits (2A/2B) which means that thicker coatings and platings can be accomodated before thread adjustments have to be made, Coarse threads can tolerate more abuse during handling and can be easier and quicker to assemble and disassemble. Fine threads provide better adjust- ment accuracy because of their smaller helix angles and require less torque to develop a given tension. However the actual difference is modest and usually ccan be ignored. Theoretically, fine threads have greater resistance to self- loosening under vibration. However, the tendency is so small that no cre- dence can be given to fine versus coarse as a mechanism for maintaining bolt/ ‘nut tightness integrity. Over the years the general debate has continued with no overwhelming support being generated for either series providing a reasonable indication that the technical merits and deficiencies are equally shared. The aerospace ‘and automotive industries were originally strong proponents of fine threads. However in recent years there has been a visible shift toward increased usage of coarse thread fasteners. The motivation here is primarily to enjoy the econo- mics of simplification rather than to advance any technical justification. In the United States, fine thread fasteners smaller than No. 10 (0.190 in.) and larger than 1 in, are practically non-existent. Also for the sizes larger‘Comparison of Thread Series a than 1 in,, the 8-thread series (8UN) commands an equal popularity with coarse threads. ‘The ISO metric coarse thread series is uniquely positioned, in terms of its thread pitches, between the Unified coarse and fine inch threads. In other words, for comparable diameters, metric coarse thread pitches are finer than Unified coarse but coarser than Unified fine. And this makes the metric coarse series an ideal compromise with arguable technical advantages over ‘either of the inch thread series. ISO metric fine threads have pitches con- siderably finer than the Unifed fine series and use of this thread series for metric fasteners is not recommended.