ASM Metals HandBook Vol. 8 - Mechanical Testing and Evaluation
Подождите немного. Документ загружается.
Weldability Testing
Weldability, while sometimes defined in the general sense as the measure of the compatibility of base metal and
any added filler metal with the heating cycles used for welding, is more commonly defined as the measure of
the resistance of the materials to the formation of cracks during welding. Methods are available for assessing
both hot cracking, at or near the solidification temperature, and cold cracking, at or near room temperature. Test
methods do not check for all types of cracking at once, but, instead, each test checks for the susceptibility to a
certain type of crack.
Weldability tests have been designed to allow small-scale specimens to mimic the cracking behavior of large,
rigid welded structures. These tests either use specimen geometry to force the shrinkage of the weldment as it
cools to be counteracted by plastic extension of the weld and HAZ, or they use additional loading to achieve
plastic extension of weld and HAZ in addition to that caused by weld shrinkage. The bulk of this section will
discuss the types of tests where additional loadings are required, because these are properly mechanical tests. A
limited discussion of some test methods without additional loading is also provided for comparison.
The techniques that include welding and loading to measure weldability have been designed primarily for hot
cracking at temperatures at or just below that for the last solidification of weld metal. The loading augments the
shrinkage strains. Cracking when the weld has cooled, such as cracks due to hydrogen, is usually tested using
welding without additional loading.
Varestraint Testing
Varestraint testing uses a cantilever beam specimen that is bent downward by a rapid application of force while
a weld is being made on the top surface of the beam (Ref 21, 22). The weld is made parallel to the direction of
tension once the force is applied with the welding torch moving toward the support point, as shown in Fig. 7. A
die block 51 mm (4 in.) long is placed beneath the specimen to force it to a limiting value of augmented strain.
The radius of the die block can be varied to examine the dependence of cracking on augmented strain, or a
single radius can be used to examine the effect of other welding variables such as base metal chemistry or weld
heat input. Common values of the augmented strain are between 0 and 4%. Standard varestraint testing methods
are discussed in AWS B4.0.
Fig. 7 Varestraint test fixture and specimen. Source: Ref 9
The specimen size is usually 305 mm (12 in.) long and 51 mm (2 in.) wide with a thickness of either 6 or 12.5
mm ( or in.), although minivarestraint specimens are sometimes tested, which reduce all dimensions by a
factor of 2. To prevent local kinking of the specimen, auxiliary plates are bent along with the test plate. The
auxiliary plates, rolled steel 305 mm by 51 mm by 12.5 or 6 mm (12 in. by 2 in. by or in.), are clamped to the
edges of the test plate. A minimum of three specimens are tested for any experimental condition. The weld on
the top surface is produced by gas tungsten arc welding (GTAW).
Measurements are based on crack length of cracks found in the HAZ or fusion zone. Hot cracks typically form
radially along the trailing edge of the weld pool and in the HAZ. Either the length of the longest crack or the
total combined crack length of all cracks is the measured parameter. Cracks are counted and sized on the
surface under a low power microscope (40 to 80×). One parameter of interest is the maximum augmented strain
without cracks. However, because only one augmented strain is measured per specimen, the value is most
closely estimated by extrapolating to zero the crack length measured at higher augmented strains.
Transvarestraint testing also uses welding on the top surface on a piece that is bent over a die block (Ref 23).
However, the force is applied so that tension is rapidly applied across the direction of welding rather than along
it. This method detects cracking sensitivity in weld metal more effectively than the original varestraint method.
Cracks in weld metal are more likely to be subsurface, so weld cross sections or nondestructive methods, such
as radiography, are required to observe the subsurface cracks.
One problem with transvarestraint tests is that cracks formed during the rapid loading may continue to
propagate after the specimen has been bent.
Spot Varestraint Testing
The spot varestraint test, or TIG-A-MA-JIG test, is a modification of the varestraint test. This test uses a
stationary welding torch (Ref 24). The torch is shut off at the instant of bending of the specimen or a specified
short time before.
The size of the weld pool should remain relatively constant between tests that are to be compared. This may
require changes of welding current or arc time when materials of differing thicknesses are compared.
The die block for the spot varestraint test is curved in only one direction, that is, it is shaped like the surface of
a cylinder. Cracking, thus, will be primarily found transverse to the tension above the highest part of the surface
of the die block. Radii for the die block have ranged from 44.5 to 1270 mm (1.75 to 50 in.).
Like the varestraint test, cracks are measured visually on the surface using a low-power microscope. Because
this test is primarily used to test for HAZ liquation cracks, crater cracks in the center of the weld metal are
ignored.
A measure of the relative susceptibility to HAZ liquation cracks that is not available in the varestraint test is the
shortest time from torch shut-off to bending that produces no cracks.
Weldability Testing without Augmented Strain
Weldability testing without augmented strain requires that the weld be made in a highly restrained specimen.
Highly restrained geometries prevent weld shrinkage in more than one direction. Cracking is typically from the
root of the weld, either at a fillet weld or a partial penetration butt weld. Specimen shapes sometimes surround
the weld metal with base metal, as in the Lehigh restraint test. Other test geometries, such as the oblique Y-
groove test, join two plates with welds at the ends and then put the test weld in the center.
References cited in this section
9. “Standard Methods for Mechanical Testing of Welds,” ANSI/AWS B4.0-98, American National
Standards Institute/American Welding Society, Miami, 1998
21. W.F. Savage and C.D. Lundin, The Varestraint Test, Weld. J., October 1965, p 433s–442s
22. C.D. Lundin, A.C. Lingenfelter, G.E. Grotke, G.G. Lessmann, and S.J. Matthews, The Varestraint Test,
Bulletin 280, Welding Research Council, August 1982
23. B.F. Dixon, R.H. Phillips, and J.C. Ritter, Cracking in the Transvarestraint Test, Part 1: A New
Procedure for Assessment of Cracking, Met. Constr., Feb 1984 and Aust. Weld. J., Vol 28 (No. 4),
summer 1983
24. W. Lin, J.C. Lippold, and W.A. Baeslack III, An Evaluation of Heat-Affected Zone Liquation Cracking
Susceptibility, Part I: Development of a Method for Quantification, Weld. J., Vol 72 (No. 4), 1993, p
135s-153s
Mechanical Testing of Welded Joints
William Mohr, Edison Welding Institute
Weldability Testing
Weldability, while sometimes defined in the general sense as the measure of the compatibility of base metal and
any added filler metal with the heating cycles used for welding, is more commonly defined as the measure of
the resistance of the materials to the formation of cracks during welding. Methods are available for assessing
both hot cracking, at or near the solidification temperature, and cold cracking, at or near room temperature. Test
methods do not check for all types of cracking at once, but, instead, each test checks for the susceptibility to a
certain type of crack.
Weldability tests have been designed to allow small-scale specimens to mimic the cracking behavior of large,
rigid welded structures. These tests either use specimen geometry to force the shrinkage of the weldment as it
cools to be counteracted by plastic extension of the weld and HAZ, or they use additional loading to achieve
plastic extension of weld and HAZ in addition to that caused by weld shrinkage. The bulk of this section will
discuss the types of tests where additional loadings are required, because these are properly mechanical tests. A
limited discussion of some test methods without additional loading is also provided for comparison.
The techniques that include welding and loading to measure weldability have been designed primarily for hot
cracking at temperatures at or just below that for the last solidification of weld metal. The loading augments the
shrinkage strains. Cracking when the weld has cooled, such as cracks due to hydrogen, is usually tested using
welding without additional loading.
Varestraint Testing
Varestraint testing uses a cantilever beam specimen that is bent downward by a rapid application of force while
a weld is being made on the top surface of the beam (Ref 21, 22). The weld is made parallel to the direction of
tension once the force is applied with the welding torch moving toward the support point, as shown in Fig. 7. A
die block 51 mm (4 in.) long is placed beneath the specimen to force it to a limiting value of augmented strain.
The radius of the die block can be varied to examine the dependence of cracking on augmented strain, or a
single radius can be used to examine the effect of other welding variables such as base metal chemistry or weld
heat input. Common values of the augmented strain are between 0 and 4%. Standard varestraint testing methods
are discussed in AWS B4.0.
Fig. 7 Varestraint test fixture and specimen. Source: Ref 9
The specimen size is usually 305 mm (12 in.) long and 51 mm (2 in.) wide with a thickness of either 6 or 12.5
mm (¼ or ½in.), although minivarestraint specimens are sometimes tested, which reduce all dimensions by a
factor of 2. To prevent local kinking of the specimen, auxiliary plates are bent along with the test plate. The
auxiliary plates, rolled steel 305 mm by 51 mm by 12.5 or 6 mm (12 in. by 2 in. by ½or ¼in.), are clamped to
the edges of the test plate. A minimum of three specimens are tested for any experimental condition. The weld
on the top surface is produced by gas tungsten arc welding (GTAW).
Measurements are based on crack length of cracks found in the HAZ or fusion zone. Hot cracks typically form
radially along the trailing edge of the weld pool and in the HAZ. Either the length of the longest crack or the
total combined crack length of all cracks is the measured parameter. Cracks are counted and sized on the
surface under a low power microscope (40 to 80×). One parameter of interest is the maximum augmented strain
without cracks. However, because only one augmented strain is measured per specimen, the value is most
closely estimated by extrapolating to zero the crack length measured at higher augmented strains.
Transvarestraint testing also uses welding on the top surface on a piece that is bent over a die block (Ref 23).
However, the force is applied so that tension is rapidly applied across the direction of welding rather than along
it. This method detects cracking sensitivity in weld metal more effectively than the original varestraint method.
Cracks in weld metal are more likely to be subsurface, so weld cross sections or nondestructive methods, such
as radiography, are required to observe the subsurface cracks.
One problem with transvarestraint tests is that cracks formed during the rapid loading may continue to
propagate after the specimen has been bent.
Spot Varestraint Testing
The spot varestraint test, or TIG-A-MA-JIG test, is a modification of the varestraint test. This test uses a
stationary welding torch (Ref 24). The torch is shut off at the instant of bending of the specimen or a specified
short time before.
The size of the weld pool should remain relatively constant between tests that are to be compared. This may
require changes of welding current or arc time when materials of differing thicknesses are compared.
The die block for the spot varestraint test is curved in only one direction, that is, it is shaped like the surface of
a cylinder. Cracking, thus, will be primarily found transverse to the tension above the highest part of the surface
of the die block. Radii for the die block have ranged from 44.5 to 1270 mm (1.75 to 50 in.).
Like the varestraint test, cracks are measured visually on the surface using a low-power microscope. Because
this test is primarily used to test for HAZ liquation cracks, crater cracks in the center of the weld metal are
ignored.
A measure of the relative susceptibility to HAZ liquation cracks that is not available in the varestraint test is the
shortest time from torch shut-off to bending that produces no cracks.
Weldability Testing without Augmented Strain
Weldability testing without augmented strain requires that the weld be made in a highly restrained specimen.
Highly restrained geometries prevent weld shrinkage in more than one direction. Cracking is typically from the
root of the weld, either at a fillet weld or a partial penetration butt weld. Specimen shapes sometimes surround
the weld metal with base metal, as in the Lehigh restraint test. Other test geometries, such as the oblique Y-
groove test, join two plates with welds at the ends and then put the test weld in the center.
References cited in this section
9. “Standard Methods for Mechanical Testing of Welds,” ANSI/AWS B4.0-98, American National
Standards Institute/American Welding Society, Miami, 1998
21. W.F. Savage and C.D. Lundin, The Varestraint Test, Weld. J., October 1965, p 433s–442s
22. C.D. Lundin, A.C. Lingenfelter, G.E. Grotke, G.G. Lessmann, and S.J. Matthews, The Varestraint Test,
Bulletin 280, Welding Research Council, August 1982
23. B.F. Dixon, R.H. Phillips, and J.C. Ritter, Cracking in the Transvarestraint Test, Part 1: A New
Procedure for Assessment of Cracking, Met. Constr., Feb 1984 and Aust. Weld. J., Vol 28 (No. 4),
summer 1983
24. W. Lin, J.C. Lippold, and W.A. Baeslack III, An Evaluation of Heat-Affected Zone Liquation Cracking
Susceptibility, Part I: Development of a Method for Quantification, Weld. J., Vol 72 (No. 4), 1993, p
135s-153s
Mechanical Testing of Welded Joints
William Mohr, Edison Welding Institute
References
1. “Standard Welding Terms and Definitions,” AWS 3.0, American Welding Society
2. L.P. Connor, Ed., Welding Handbook, 8th ed., Vol 1, American Welding Society, 1991
3. “Specification for Carbon Steel Electrodes for Shielded Metal Arc Welding,” ANSI/AWS A5.1-91,
American National Standards Institute/American Welding Society, Miami, 1991
4. “Recommended Practice for Preproduction Qualification for Steel Plates for Offshore Structures,” API
2Z, 2nd ed. (includes November 1998 addenda), American Petroleum Institute, Washington DC, 1998
5. “Standard Test Method for Young's Modulus, Tangent Modulus, and Chord Modulus,” E 111, Annual
Book of ASTM Standards, Vol 3.01, ASTM, 1999
6. “Standard Methods of Tension Testing of Metallic Materials,” E 8, Annual Book of ASTM Standards,
Vol 3.01, ASTM, 1999
7. “Standard Methods of Tension Testing Wrought and Cast Aluminum, and Magnesium Alloy Products,”
B 557, Annual Book of ASTM Standards, Vol 3.01, ASTM, 1999
8. B4 Committee on Mechanical Testing of Welds, Mechanical Testing of Welds, Part 1: Summary of
Tension Testing of Welds, Weld. J., Jan 1981, p 33–37
9. “Standard Methods for Mechanical Testing of Welds,” ANSI/AWS B4.0-98, American National
Standards Institute/American Welding Society, Miami, 1998
10. B4 Committee on Mechanical Testing of Welds, Mechanical Testing of Welds, Part 2: Bend Testing of
Welds. A Summary, Weld. J., Feb 1981, p 34–37
11. ASME Boiler and Pressure Vessel Code, Section IX, American Society of Mechanical Engineers
12. Aluminum Design Manual, The Aluminum Association, 1994
13. “Standard Test Methods for Notched Bar Impact Testing of Metallic Materials,” E 23, Annual Book of
ASTM Standards, Vol 3.01, ASTM
14. “Standard Method for Conducting Drop-Weight Test to Determine Nil-Ductility Transition Temperature
of Ferritic Steels,” E 208, Annual Book of ASTM Standards, Vol 3.01, ASTM
15. M.G. Dawes, H.G. Pisarski, and S. J. Squirrell, Fracture Mechanics Tests on Welded Joints, Nonlinear
Fracture Mechanics Vol II: Elastic-Plastic Fracture, STP 995, J.D. Landes, A. Saxena, and J.G.
Merkle, Ed., ASTM, 1989, p 191–213
16. “Standard Test Method for Crack-Tip Opening Displacement (CTOD) Fracture Toughness
Measurement,” E 1290, Annual Book of ASTM Standards, Vol 3.01, ASTM
17. “Standard Test Method for J-Integral Characterization of Fracture Toughness,” E 1737, Annual Book of
ASTM Standards, Vol 3.01, ASTM
18. N.J. Rendler and I. Vigness, Hole-Drilling Strain-Gage Method of Measuring Residual Stresses,
Experimental Mechanics, Vol 6, 1966, p 577–586
19. W. Cheng and I. Finnie, A Method for Measurement of Axisymmetric Axial Residual Stresses in
Circumferentially Welded Thin-Walled Cylinders, J. Eng. Mater. Technol. (Trans. ASME), Vol 107,
July 1985, p 181–185
20. D.S. Kim and J.D. Smith, Residual Stress Measurements of Tension Leg Platform Tendon Welds, Proc.
of the Offshore Mechanics and Arctic Engineering Conference 1994, Vol 3, American Society of
Mechanical Engineers, 1994
21. W.F. Savage and C.D. Lundin, The Varestraint Test, Weld. J., October 1965, p 433s–442s
22. C.D. Lundin, A.C. Lingenfelter, G.E. Grotke, G.G. Lessmann, and S.J. Matthews, The Varestraint Test,
Bulletin 280, Welding Research Council, August 1982
23. B.F. Dixon, R.H. Phillips, and J.C. Ritter, Cracking in the Transvarestraint Test, Part 1: A New
Procedure for Assessment of Cracking, Met. Constr., Feb 1984 and Aust. Weld. J., Vol 28 (No. 4),
summer 1983
24. W. Lin, J.C. Lippold, and W.A. Baeslack III, An Evaluation of Heat-Affected Zone Liquation Cracking
Susceptibility, Part I: Development of a Method for Quantification, Weld. J., Vol 72 (No. 4), 1993, p
135s-153s
Testing of Bearings
Charles A. Moyer, The Timken Company (Retired)
Introduction
BEARINGS can be divided into two major classes: rolling bearings, which include rolling elements (balls,
rollers, or needle rollers) between the inner and outer raceways, and sliding, or plain, bearings, which have
motion from one surface directly imposed on a stationary support.
Rolling bearings include radial, thrust, and angular contact designs. A review of the many versions of these
bearings can be found in the article“Friction and Wear of Rolling-Element Bearings” in Friction, Lubrication,
and Wear Technology, Volume 18 of the ASM Handbook (Ref 1). The primary requirement of rolling bearings
is proper and adequate lubricant that provides separation of the moving surfaces under all conditions, maintains
appropriate temperature, and provides an operating environment so that bearings will achieve their expected
lives.
Sliding bearings include sleeve and thrust bearings of various designs. Based on designs and materials
selection, plain bearings operate under dry or boundary lubrication conditions, partial film or mixed lubricant
film conditions, or a full film, which means the “rubbing” surfaces are essentially separated. More details of
sliding bearings can be found in the article “Friction and Wear of Sliding Bearings” in Friction, Lubrication,
and Wear Technology, Volume 18 of the ASM Handbook (Ref 2) and in Ref 3.
References cited in this section
1. T.A. Harris, Friction and Wear of Rolling-Element Bearings, Friction, Lubrication, and Wear
Technology, Vol 18, ASM Handbook, ASM International, 1992, p 499
2. R. Pike and J.M. Conway-Jones, Friction and Wear of Sliding Bearings, Friction, Lubrication, and
Wear Technology, Vol 18, ASM Handbook, ASM International, 1992, p 515–521
3. M.M. Khonsari and J.Y. Jang, Chapters 61, 62, and 63, Tribology Data Handbook, E.R. Booser, Ed.,
CRC Press, LLC, 1997, p 669–707
Testing of Bearings
Charles A. Moyer, The Timken Company (Retired)
Rolling Element Bearings
Rolling bearings date back to the Neolithic period, or the new Stone Age (Ref 4). When heavy materials needed
to be moved, or when primitive vehicles had wheels, someone devised bronze or wood “bearings,” with or
without oils or fats. Archaeologists have found bronze balls, cylindrical rollers, and wooden tapered roller
bearings dating back to 300 to 500 B.C.
Considering the different materials used, it is clear that finding the proper material, shapes, and sizes was then,
and continues to be, important and is the core of testing to determine what performance can be expected from
rolling element bearings. Advancements continued over the years, but bearings that resemble those in use today
did not appear until the late 19th century. Modern rolling bearings developed gradually from 1850 to 1925 (Ref
4). One driving force was the need for bearings in bicycles.
While working in Berlin, Germany, Professor Richard Stribeck undertook early bearing tests, the results of
which were published in 1901 and 1902 (Ref 5, 6). His goal was to determine safe ball loads statically and in
complete bearings over different speed ranges. He made use of Hertz's work covering elastic bodies in contact
(Ref 7) and started with a press arrangement with three hardened steel balls in contact, two steel balls with a
steel plate between them, and finally, a single ball between two short cylinders set on end with cup shapes to fit
the ball. He found the relation of load P on the ball diameter d to be:
P = Kd
2
(Eq 1)
K was a constant based on steel type and contact geometry. Based on the materials he used, Stribeck established
that Eq 1 held to the elastic limit and somewhat beyond. He also ran tests on complete bearings and established
the way balls shared loads under radial load conditions. Using bearings with 10 to 20 balls and no clearance
between balls and rings, he determined that the most heavily loaded ball (P
0
) was related to the total load P by:
P
o
= 4.37/Z·P
(Eq 2)
where Z is the number of balls in the bearing. To be conservative, Stribeck changed 4.37 to 5.0.
In 1912, Professor John Goodman published in England studies on rolling element bearings (Ref 8), covering
life, friction, and wear. He knew of Stribeck's work and determined reductions in bearing load capacity based
on bearing speed (N). Goodman's equation for this reduction was:
P = Kd
2
/ND + Ad
(Eq 3)
where D is the ball-race path diameter; A is a constant; and P,K, and d are the same as in Eq 1.
Studies of bearings and bearing materials (primarily steel) were followed by the emergence of the first
manufacturing companies to patent and produce rolling element bearings. Bearing lubrication at this time was
greatly influenced by the Reynold's equation (Ref 9) and the successes experienced in the hydrodynamic films
generated in conformal bearings. It was assumed that the nonconformal contacts in “antifriction” bearings
generated much thinner and, thus, less protective films for the contact region. The actual lubricating means for
rolling bearings was thus a puzzle.
However, in 1949, Grubin (Ref 10) considered the elastic behavior of the contact materials and the pressure
characteristics of the viscosity under the high stresses in the contact region in an equation that gave lubricant
films one or two orders thicker than with the usual rigid cylinders assumption used for determining lubricant
films as in hydrodynamic bearings. For rolling bearings, elastohydrodynamic lubrication came to be recognized
as an important factor in determining life expectancy under a wide range of operating conditions. Tallian (Ref
11) described the stages of rolling element bearings as an empirical state until late 1920. A classical period then
emerged that saw the engineering inclusion of Hertz's elasticity theory, the Weibull function, and statistics
leading to worldwide bearing standards. Finally, the modern period began in the 1950s. The advancements in
bearing technology allowed the new era to develop meaningful testing for rolling bearings.
References cited in this section
4. D. Dowson, History of Tribology, 2nd ed., Professional Eng. Pub., Ltd., London, Edmunds, U.K., 1998
5. R. Stribeck, Kugellager für beliebige Belastungen, Z. Ver. dt. Ing., Vol 45 (No. 3), 1901, p 73–125
6. R. Stribeck, Die Wesentlichen Eigen-schaften der Gleit und Rollenlager, Z. Ver. dt. Ing., Vol 46 (No.
38), 1902, p 1341–8, 1432–8; (No. 39), 1463–70
7. H. Hertz, On the Contact of Elastic Solids, J. reine und angew, Math., Vol 92, 1881, p 156–71
8. J. Goodman, (1) Roller and Ball Bearings, (2) The Testing of Antifriction Bearing Materials, Proc. Inst.
Civ. Engrs., clxxxix, Session 1911-2, Pt. 111, 1912, p 4–88
9. O. Reynolds, On the Theory of Lubrication and Its Application to Mr. Beauchamp Tower's
Experiments, Including an Experimental Determination of the Viscosity of Olive Oil, Philos. Trans. R.
Soc., 177, 1886, p 157–234
10. A.N. Grubin and I.E. Vinogradova, Investigation of the Contact of Machine Components, Book No. 30
(DSIR Trans. No. 337), Kh. F. Ketova, Ed., Central Scientific Research Institute for Technology and
Mechanical Engineering, Moscow, 1949
11. T.E. Tallian, “Progress in Rolling Contact Technology,” Report AL 690007, (SKF) Industries, King of
Prussia, PA, 1969
Testing of Bearings
Charles A. Moyer, The Timken Company (Retired)
Sliding Bearings
There is evidence from potters' wheels and ancient vehicles that plain bearings date back more than 5,000 years
(Ref 1). Materials were stone or primarily wood for wheels of vehicles. It is assumed that the axles were
stationary and the wheels rotated. Changes of material and “design” brought plain bearings to the point that
scientific studies were underway in the mid 1800s by people such as Gustov Hirn (Ref 12), who, by empirical
means, provided information about fluid film lubrication.
Others, such as Robert Thurston (Ref 13), who studied journal bearing friction and lubrication; Nikolai Petrov,
who explained the hydrodynamic characteristics of journal bearing friction (Ref 14); and Beauchamp Tower
(Ref 15), who discovered the pressure that developed within an operating journal bearing and that a bearing
with sufficient lubrication floated on a film of oil, followed Hirn's studies. Although the studies first