- Email: [email protected]
Measurement and testing during strain hardening through shot-peening
beneficial residual compressive surface stresses. Quality depends on conscientious control of the process parameters. A listing of the parameters that influence the residual stress profile, which is the key factor for improvement of fatigue strength through shot-peening, is given. Direct measurement of the depth and character of the shot-peening effect on production parts is impossible without destructive tasting. An internationally recognized alternative approach i0 the use of Almen test stripe to measure blast intensity, taking advantage of the functional relation between it, the strength of the material shot-peened and the depth of the layer in which the residual compressive stress is present, Measurement procedures and equipment used for Almen strip testing are explained in detail. Graphs.
M a t e d e l p r o p e r t y evaluation o f t h i c k t h e r m a l barrier coating systems.
Brink, R.C. J. Eng. Gas Turbines Power (Trans. ASME) J u l y 1989 111, (3), 5 7 0 - 5 7 7 Coating system optimization is a critical step in the design and development of plasma-sprayed thermal barrier coatings (TBCs) for diesel engines (made of 17-4PH steel and grey iron). Physical and mechanical property measurements 0nd bench screening tests were performed to evaluate the candidate TBC systems. Additional understanding of the material beheviour and failure mechanisms of thick TBCs gained from the property measurements and bench testing was identified. A specific graded zirconia (ZrO2)/bond (NiCrAIY) coating material system with good strain tolerance and fatigue resistance was selected for follow-on diesel engine durability testing, Graphs, photomicrographs. 16 refs.
Joining Weld repair w o r k . van den Brink, S.H. Lastechniek Feb. 1989 65, (2), 4 0 - 4 7 (in Dutch) Submerged arc welded 25 and 50 mm T-pieces of steel Fe510 were examined in a project supported by Nederlands Instituut veer Lastechniek on the repair of welded structures showing penetration defects. The properties of repaired fatigue cracks and zones subject to dynamic loading, and fatigue testing of repairs carried out with restraint ore also reported. The most critical repair conditions arise in the first four runs, where the weld stresses can attain 150 N / m s 2. The restraint factor K is quantified, and a graph relating this to the C~ for preheating at 20 and 100 °C is compared with standard NEN 2063.
initiation end coalescence o f f a t i g u e cracks in w e l d e d j o i n t s in steel. (Dissertation). Otegui, J.L, Diss. Abstr. Int. Apr. 1989 40, (10) Non-load-carrying bead-on-plate and T-plate manual and automatic welds were tasted in three-point bending in air. A strain gauge method to measure crack depth and closure levels of small surface cracks was developed, and the effect of crock geometry on the proposed calibration methods was analysed. Three-stage link staining, replica techniques and SEM microscopy were also used to monitor multiple cracking at the weld toes. Minor undercuts, slag inclusions and surface spotter at the weld toes were detected as fatigue initiation sites. A first microcrack coalescence stage at a depth of < 0.3 m s , and a second stage at crack depths epprox 1-3 m s , determined crack aspect ratios. Depths to coalescence were proportional to the mismatch between the planes of neighbouring surface cracks. Surface growth of fatigue cracks was related to the effect of previous damage at the toe, and crock-microcrack interactions. The much faster surface growth observed in automatic welds, that led to lower aspect ratios and propagation lives, resulted from the straighter crack paths due to a smaller, more continuous toe radius. Small crack effects were analysed in terms of the variation of plasticity induced crack closure levels. Growth rates were related to AK~ff, through the Paris-Erdogan relationship. It was found that LEFM techniques could be confidently used for cracks deeper than ao = 0.3 ram, which was therefore proposed as an initial crack depth ~ to be used in models of fatigue propagation. Several models of the evolution of crack aspect ratios with crack depth were compared. In the small crack (a < 3 mm) regime, most models overestimated aspect ratios and propagation lives. Corrosion f a t i g u e o f hot-galvanized c o m p o n e n t - l i k e w e l d e d j o i n t s in a chloride medium under changing corrosion conditions. Sonsino, C.M.
and Bottcher, H.-J. Werkst. Korros. J u n e 1989 40, (6), 351-363 (in German)
The fatigue beheviour was investigated of welded specimens of hot-galvanized plate material of steels Rst 37-2, St 52-3 and St 70-2 as frequently used in steel construCtions which were mended by thermal Zn-spraying in the fillet area and tested under changing corrosion conditions. The fatigue beheviour of welded specimens with additional PVC coating of both non-galvanized and galvanized types with changing corrosion reaction was also investigated. The established findings can be summarized as follows. (1) Under constant amplitude loading, hot-galvanizing largely prevents the drop in endurance which would occur under corrosion. An increase of fatigue strength in the corrosive medium due to galvanizing is only found conditionally under variable amplitude loading. Welding of hot-galvanized parts brings improved protection against corrosion at variable amplitude loading only if 0 given thickness of the Zn coating, that depends on the material, is not exceeded. A coating which is too thick results in the formation of pores on subsequent welding and reduces the fatigue strength in spite of later Zn-spraying. (2) The structural steels RSt 37-2 and St 52-3 in the hot-galvanized state are more suitable for welding than the more notch- and corrosion-sensitive St 70-2. (3) Soldering brittleness was not found to be a cause of the failure, (4) The application of PVC protective layers causes high rates of increase of fatigue, for the nongalvanized specimens. 26 refs.
Frestographic s t u d y o f f a t i g u e crack p r o p a g a t i o n in b u t t - w e l d e d joints.
Kitsunai, Y. and Yoshihisa, E. J. Sac. Mater, Sci., Jpn. M a y 1989 38, (428), 527-532 (in Japanese)
Fatigue crack growth tests in butt-welded joints of HTS0 and STS42 steels were carried out using compact type (CT), centre-crscked tension (CCT) and single edgecracked tension (SECT) specimens at room temperature and 173 K, and their fracture surfaces were examined using a scanning and a transmission electron microscope
Int J Fatigue March 1990
to evaluate the influence of residual stress on the formation of fracturemorphologies. As the fatigue crack propagated through compression residual stress fields, the fracture surface was covered with a thin oxide film which was produced by mutual contact of crack surfaces. At AK > 20 MPa m t/2 or the crack growth rate > 2 x 10-s m/c, the fracture morphologies in the base metal and welded joints were dominated by striations independent of the test temperature. In the low &K region or where the crack growth rate was below 1 x 10-" m/c, fracture morphologies showed that the fatigue crack propagated along the microstructurec with fevoudta orientation to the crack growth. The intergranular like facets were formed in the weld metal of HTeO steel at AK ranging from 4-12 MPa m 1/2. The maximum area percentage of the intergranular like facets, however, was considerably lower than that of intergranular facets formed in the base metal. The striation spacings in the welded joints were correlated well with the effective stress intensity factor range, AKeffR, which was estimated by superpooition of the respective stress intensity factors for the residual stress fields and for the applied stresses. 7 refs.
A statistical s t u d y on f a t i g u e fracture o f s p o t w e l d e d j o i n t s in cross tension t y p e . Sakai, T., Kikuchi, T. and Fujisawa, Y. J. Sac. Mater. Sci., Jpn. Mar. 1989 38, (426), 287-293 (in Japanese) A fatigue test to obtain the L-N characteristics was performed on spot welded joints of cross tension type. Then the fatigue life distribution was experimentally examined by assigning a number of specimens at two different levels of applied load. Moreover, SEM observations were made on the fracture surfaces by paying attention to the crack initiation site and the crack propagation path. The main results obtained on this type of welded joint are summarized as follows. (1) Both distributions of the static strength and the fatigue life were well represented by the three-parameter WeibuII distribution. (2) The fatigue fracture was classified into several types depending on the combination of crack initiation sites and crack initiation lives at the respective sites. (3) The fatigue fracture was caused by the propagation of the second crack, which was stimulated by the first crack. (4) A correlation was found between the fatigue life and the crack reflected angle, and this was qualitatively explained by the variation of the stress intensity factor in the mixed mode. 20 refs.
Fatigue strength at t o e o f w e l d e d joint. Inoue, A,, Yoshioka, S., Hoshinouchi, S. and Sawaki, Y. J, Sac. Mater. ScL, Jpn. Mar. 1989 38, (426), 2 4 9 - 2 5 4 (in Japanese) The high cycle fatigue strengths at the toe of cruciform welded joints of SM41, and HT60 were investigated experimentally under tensile and bending fatigue loading. The effects of welding process (toe treatment), mean stress and welding residual stress on fatigue strength were studied. The main results obtained are as follows. (1) The fatigue limit of SM41A welded joints under tensile loading was increased up to 64% by the toe treatment of CO2 soft plasma welding. Similarly, in the case of HT60 welded joint the fatigue limits under bending loading were increased to 77, 43 and 25% by the toe treatment of CO2 soft plasma welding, TIG welding and grinding, respectively. These improvements can be attributed to the change of the toe part geometry; lowering the elastic stress concentration factor at the toe. (2) The relation between the fatigue limit of welded joints and mean stress can be expressed by the straight line intercepted at the true rupture stress on the axis of mean stress in the fatigue limit diagram. (3) The fatigue limit of welded joints was decreased by tensile welding residual stress. This decrease can be estimated well by regarding the welding residual stress as the static stress. 7 refs.
C u m u l a t i v e d a m a g e o f w e l d e d joints: U. Test results. Gurney, T.R. Joining Mater. Aug. 1989 2, (8), 3 9 0 - 3 9 5 The results of tests over the 10st six or seven years are reviewed. All tests were done in air and the data are reviewed in light of the implications of fatigue design rules. A total of 1292 individual test results are used for the study and the data are summarized as ";(n/N) for each test series. The results are based upon mean experimental S-N curves extrapolated linearly downward. Several of the problems associated with interpreting such large bodies of data are discussed. For example, the reduction in life between the mean and mean minus 2SD design curve takes into account only scarer found in constant amplitude tests as differences in local joint geometry. Other variables are not considered. In spite of this, the detailed analysis has shown that Miner's rule, which is based upon the mean minus 2SD design S-N curves, gives setisfactory design criteria. Finally, it is stressed that variable amplitude fatigue tests are useless unless they are properly planned and reported, in terms that designers can use such as peak stress in the range spectrum. Graphs.
Fatigue a n d ten=die properties o f EB w e l d e d 17-4 PH steel. Islam, M.V.,
Campbell, G, and Hsu. R. Weld. J. Sept. 1989 68, (9), 4 5 - 5 0
Fatigue properties of electron beam (EB) welded 17-4 PH stainless steel were studied. Chemistry data of the material, welding parameters, and photomicrographs of the resultant structures are presented. Results indicate that welding in the solution treated condition and aging at 482 °C do not affect the tensile strength of the material. Failures in fatigue specimens did tend to occur in the heat affected zone which was not in the minimum section of the fatigue specimen, indicating the fatigue strength of the weld was lower than that of the base metal. Additional tanoion-compression tests ere suggested as these tests were run under the rotating and bending mode. Photomicrographs. 8 refs.
Effect o f w e l d surfacing on t h e f a t i g u e strength o f c o m p o n e n t s m a d e f r o m C45 steel. Robakowski, 7, Schweisaen Schneiden M a y 1989 41, (5), 229-231 (in German) The fatigue strength of weld surfaced specimens (manual arc welding, semiautomatic active-gas metal-arc welding using carbon dioxide or a mixture of 80% Ar + 20% CO2 fully automatic plasma welding with a powdery filler metal) was determined under pulsating tensile stress and a stress ratio of R = +0.2. The results were statistically evaluated on the basis of 10s or 2 x 10s cycles and a 50=/0 probability of survival. It was found that the fatigue strength falls off noticeably as compared with the base metal (heat treatable steel C45), irrespective of the welding process used. Graphs, photomicrographs. (In English p. E76-E77).
141
M a t e d e l p r o p e r t y evaluation o f t h i c k t h e r m a l barrier coating systems.
Brink, R.C. J. Eng. Gas Turbines Power (Trans. ASME) J u l y 1989 111, (3), 5 7 0 - 5 7 7 Coating system optimization is a critical step in the design and development of plasma-sprayed thermal barrier coatings (TBCs) for diesel engines (made of 17-4PH steel and grey iron). Physical and mechanical property measurements 0nd bench screening tests were performed to evaluate the candidate TBC systems. Additional understanding of the material beheviour and failure mechanisms of thick TBCs gained from the property measurements and bench testing was identified. A specific graded zirconia (ZrO2)/bond (NiCrAIY) coating material system with good strain tolerance and fatigue resistance was selected for follow-on diesel engine durability testing, Graphs, photomicrographs. 16 refs.
Joining Weld repair w o r k . van den Brink, S.H. Lastechniek Feb. 1989 65, (2), 4 0 - 4 7 (in Dutch) Submerged arc welded 25 and 50 mm T-pieces of steel Fe510 were examined in a project supported by Nederlands Instituut veer Lastechniek on the repair of welded structures showing penetration defects. The properties of repaired fatigue cracks and zones subject to dynamic loading, and fatigue testing of repairs carried out with restraint ore also reported. The most critical repair conditions arise in the first four runs, where the weld stresses can attain 150 N / m s 2. The restraint factor K is quantified, and a graph relating this to the C~ for preheating at 20 and 100 °C is compared with standard NEN 2063.
initiation end coalescence o f f a t i g u e cracks in w e l d e d j o i n t s in steel. (Dissertation). Otegui, J.L, Diss. Abstr. Int. Apr. 1989 40, (10) Non-load-carrying bead-on-plate and T-plate manual and automatic welds were tasted in three-point bending in air. A strain gauge method to measure crack depth and closure levels of small surface cracks was developed, and the effect of crock geometry on the proposed calibration methods was analysed. Three-stage link staining, replica techniques and SEM microscopy were also used to monitor multiple cracking at the weld toes. Minor undercuts, slag inclusions and surface spotter at the weld toes were detected as fatigue initiation sites. A first microcrack coalescence stage at a depth of < 0.3 m s , and a second stage at crack depths epprox 1-3 m s , determined crack aspect ratios. Depths to coalescence were proportional to the mismatch between the planes of neighbouring surface cracks. Surface growth of fatigue cracks was related to the effect of previous damage at the toe, and crock-microcrack interactions. The much faster surface growth observed in automatic welds, that led to lower aspect ratios and propagation lives, resulted from the straighter crack paths due to a smaller, more continuous toe radius. Small crack effects were analysed in terms of the variation of plasticity induced crack closure levels. Growth rates were related to AK~ff, through the Paris-Erdogan relationship. It was found that LEFM techniques could be confidently used for cracks deeper than ao = 0.3 ram, which was therefore proposed as an initial crack depth ~ to be used in models of fatigue propagation. Several models of the evolution of crack aspect ratios with crack depth were compared. In the small crack (a < 3 mm) regime, most models overestimated aspect ratios and propagation lives. Corrosion f a t i g u e o f hot-galvanized c o m p o n e n t - l i k e w e l d e d j o i n t s in a chloride medium under changing corrosion conditions. Sonsino, C.M.
and Bottcher, H.-J. Werkst. Korros. J u n e 1989 40, (6), 351-363 (in German)
The fatigue beheviour was investigated of welded specimens of hot-galvanized plate material of steels Rst 37-2, St 52-3 and St 70-2 as frequently used in steel construCtions which were mended by thermal Zn-spraying in the fillet area and tested under changing corrosion conditions. The fatigue beheviour of welded specimens with additional PVC coating of both non-galvanized and galvanized types with changing corrosion reaction was also investigated. The established findings can be summarized as follows. (1) Under constant amplitude loading, hot-galvanizing largely prevents the drop in endurance which would occur under corrosion. An increase of fatigue strength in the corrosive medium due to galvanizing is only found conditionally under variable amplitude loading. Welding of hot-galvanized parts brings improved protection against corrosion at variable amplitude loading only if 0 given thickness of the Zn coating, that depends on the material, is not exceeded. A coating which is too thick results in the formation of pores on subsequent welding and reduces the fatigue strength in spite of later Zn-spraying. (2) The structural steels RSt 37-2 and St 52-3 in the hot-galvanized state are more suitable for welding than the more notch- and corrosion-sensitive St 70-2. (3) Soldering brittleness was not found to be a cause of the failure, (4) The application of PVC protective layers causes high rates of increase of fatigue, for the nongalvanized specimens. 26 refs.
Frestographic s t u d y o f f a t i g u e crack p r o p a g a t i o n in b u t t - w e l d e d joints.
Kitsunai, Y. and Yoshihisa, E. J. Sac. Mater, Sci., Jpn. M a y 1989 38, (428), 527-532 (in Japanese)
Fatigue crack growth tests in butt-welded joints of HTS0 and STS42 steels were carried out using compact type (CT), centre-crscked tension (CCT) and single edgecracked tension (SECT) specimens at room temperature and 173 K, and their fracture surfaces were examined using a scanning and a transmission electron microscope
Int J Fatigue March 1990
to evaluate the influence of residual stress on the formation of fracturemorphologies. As the fatigue crack propagated through compression residual stress fields, the fracture surface was covered with a thin oxide film which was produced by mutual contact of crack surfaces. At AK > 20 MPa m t/2 or the crack growth rate > 2 x 10-s m/c, the fracture morphologies in the base metal and welded joints were dominated by striations independent of the test temperature. In the low &K region or where the crack growth rate was below 1 x 10-" m/c, fracture morphologies showed that the fatigue crack propagated along the microstructurec with fevoudta orientation to the crack growth. The intergranular like facets were formed in the weld metal of HTeO steel at AK ranging from 4-12 MPa m 1/2. The maximum area percentage of the intergranular like facets, however, was considerably lower than that of intergranular facets formed in the base metal. The striation spacings in the welded joints were correlated well with the effective stress intensity factor range, AKeffR, which was estimated by superpooition of the respective stress intensity factors for the residual stress fields and for the applied stresses. 7 refs.
A statistical s t u d y on f a t i g u e fracture o f s p o t w e l d e d j o i n t s in cross tension t y p e . Sakai, T., Kikuchi, T. and Fujisawa, Y. J. Sac. Mater. Sci., Jpn. Mar. 1989 38, (426), 287-293 (in Japanese) A fatigue test to obtain the L-N characteristics was performed on spot welded joints of cross tension type. Then the fatigue life distribution was experimentally examined by assigning a number of specimens at two different levels of applied load. Moreover, SEM observations were made on the fracture surfaces by paying attention to the crack initiation site and the crack propagation path. The main results obtained on this type of welded joint are summarized as follows. (1) Both distributions of the static strength and the fatigue life were well represented by the three-parameter WeibuII distribution. (2) The fatigue fracture was classified into several types depending on the combination of crack initiation sites and crack initiation lives at the respective sites. (3) The fatigue fracture was caused by the propagation of the second crack, which was stimulated by the first crack. (4) A correlation was found between the fatigue life and the crack reflected angle, and this was qualitatively explained by the variation of the stress intensity factor in the mixed mode. 20 refs.
Fatigue strength at t o e o f w e l d e d joint. Inoue, A,, Yoshioka, S., Hoshinouchi, S. and Sawaki, Y. J, Sac. Mater. ScL, Jpn. Mar. 1989 38, (426), 2 4 9 - 2 5 4 (in Japanese) The high cycle fatigue strengths at the toe of cruciform welded joints of SM41, and HT60 were investigated experimentally under tensile and bending fatigue loading. The effects of welding process (toe treatment), mean stress and welding residual stress on fatigue strength were studied. The main results obtained are as follows. (1) The fatigue limit of SM41A welded joints under tensile loading was increased up to 64% by the toe treatment of CO2 soft plasma welding. Similarly, in the case of HT60 welded joint the fatigue limits under bending loading were increased to 77, 43 and 25% by the toe treatment of CO2 soft plasma welding, TIG welding and grinding, respectively. These improvements can be attributed to the change of the toe part geometry; lowering the elastic stress concentration factor at the toe. (2) The relation between the fatigue limit of welded joints and mean stress can be expressed by the straight line intercepted at the true rupture stress on the axis of mean stress in the fatigue limit diagram. (3) The fatigue limit of welded joints was decreased by tensile welding residual stress. This decrease can be estimated well by regarding the welding residual stress as the static stress. 7 refs.
C u m u l a t i v e d a m a g e o f w e l d e d joints: U. Test results. Gurney, T.R. Joining Mater. Aug. 1989 2, (8), 3 9 0 - 3 9 5 The results of tests over the 10st six or seven years are reviewed. All tests were done in air and the data are reviewed in light of the implications of fatigue design rules. A total of 1292 individual test results are used for the study and the data are summarized as ";(n/N) for each test series. The results are based upon mean experimental S-N curves extrapolated linearly downward. Several of the problems associated with interpreting such large bodies of data are discussed. For example, the reduction in life between the mean and mean minus 2SD design curve takes into account only scarer found in constant amplitude tests as differences in local joint geometry. Other variables are not considered. In spite of this, the detailed analysis has shown that Miner's rule, which is based upon the mean minus 2SD design S-N curves, gives setisfactory design criteria. Finally, it is stressed that variable amplitude fatigue tests are useless unless they are properly planned and reported, in terms that designers can use such as peak stress in the range spectrum. Graphs.
Fatigue a n d ten=die properties o f EB w e l d e d 17-4 PH steel. Islam, M.V.,
Campbell, G, and Hsu. R. Weld. J. Sept. 1989 68, (9), 4 5 - 5 0
Fatigue properties of electron beam (EB) welded 17-4 PH stainless steel were studied. Chemistry data of the material, welding parameters, and photomicrographs of the resultant structures are presented. Results indicate that welding in the solution treated condition and aging at 482 °C do not affect the tensile strength of the material. Failures in fatigue specimens did tend to occur in the heat affected zone which was not in the minimum section of the fatigue specimen, indicating the fatigue strength of the weld was lower than that of the base metal. Additional tanoion-compression tests ere suggested as these tests were run under the rotating and bending mode. Photomicrographs. 8 refs.
Effect o f w e l d surfacing on t h e f a t i g u e strength o f c o m p o n e n t s m a d e f r o m C45 steel. Robakowski, 7, Schweisaen Schneiden M a y 1989 41, (5), 229-231 (in German) The fatigue strength of weld surfaced specimens (manual arc welding, semiautomatic active-gas metal-arc welding using carbon dioxide or a mixture of 80% Ar + 20% CO2 fully automatic plasma welding with a powdery filler metal) was determined under pulsating tensile stress and a stress ratio of R = +0.2. The results were statistically evaluated on the basis of 10s or 2 x 10s cycles and a 50=/0 probability of survival. It was found that the fatigue strength falls off noticeably as compared with the base metal (heat treatable steel C45), irrespective of the welding process used. Graphs, photomicrographs. (In English p. E76-E77).
141