(3337 irro, * Industrieanlagen Betriebsgesellschaft mbh, Ottobrunn, Germany SHOT PEENING - A CHOICE TO IMPROVE FATIGUE LIFE OF TITANIUM COMPONENTS

Transcription

1 SHOT PEENING A CHOICE TO IMPROVE FATIGUE LIFE OF TITANIUM COMPONENTS W. Oberparl ei ter, W. Schutz* Shot peening is one of several possibilities to improve fatigue life of titanium components by surface treatment. The improvement is different for the two Titanium alloys TiA16V4 and TiA16V6Sn2, it depends on the type of loading, the specimen geometry (notched or unnotched) and the optimized peening parameters. Shotpeening may improve fatigue life by a factor of about 2 up to 40. INTRODUCTION Among other surface treatment techniques shot peening is a wellknown method to improve the fatigue life of critical parts of a structure. The impact of the different shot (steel, ceramic or glass beads) to the surface increases the fatigue resistance of a mechanical part under cyclic loading. This increase is particularly due to the compressive stresses induced in the surface layers of the material (I). On the other hand, however, there is a deleterious effect of a higher roughness of the surface after shot peening. The positive effect of shot peening has been verified for several steel, aluminium and also titanium alloys in the past, but mainly under constant amplitude loading (2), (3), (4), (5) using flat specimens. But most components especially in aircrafts or vetiicles are subjected to stochastically or deterministjcally variable loads in service. It is also generally assumed that residual compressive stresses affect fatigue strength only as long as they are not relaxed by the service stresses. In order to account * Industrieanlagen Betriebsgesellschaft mbh, Ottobrunn, Germany (3337 irro,

2 for these residual stresses it is necessary to have an idea of their relaxation during the service life of the peened component. The residual compressive stresses are assumed to relax when the material yields in the highest loaded zone, e. g. in a notch. Therefore the effect of shot peening on the fatigue life of notched specimen is of special interest. In the authors' company results are available from two larger programmes dealing with the effect of shot peening on the fatigue life of the two titanium alloys TiA16V4 and TiA16V6Sn2. From these results some conclusion are drawn on the effect of shot peening to the fatigue life of Tialloys of different chemical composition and static strength. MATERIALS AND SPECIMEN In the following tables the chemical composition and the static properties of both titanium alloys are given. TABLE 1 Chemical Composition Material A1 V Sn Fe O2 N2 Cu C Ti The tests were done using 4point bend specimen, unnotched flat specimen of TiA16V6Sn2 and TiA16V and notched specimen with a stress intensity factor Kt = 1.6 only of TiA16V4, see Figure 1. Most of thespecimenswere manufactured in the authors' workshop because already machining may induce significant compressive stresses in the surface of Tispecimen (6). TEST PROGRAMME SNtests have been carried out with flat unnotched specimen of both titanium alloys with and without shot peening. For TiA16V6Sn2 tests were done under a flightbyflight sequence using peened and not peened specimens. The effect of shot peening on the fatigue life of notched specimen under constant amplitude and variable amplitude loading was only determined for TiA16V4. In this case the standardized load sequence "FALSTAFF" was used in variable amplitude tests. All tests were done under 4pointbending. The test equipment is schematically shown in Figure 2. The stress ratio under constant amplitude loading was R = +0.1 for TiA16V4 and R = 0.25 for TiA16V6Sn2,Under variable amplitude loading the stress ratio generally was R = The maximum nominal stresses in all tests were kept below 0.2 percent offset yield strength in bending. SHOT PEENING PARAMETERS TiA16V4 TiA16V6Sn balance balance The shot peening conditions were controlled using the Almen test (7) which involves the exposure of one side of clamped, fixed standardized steel stripes to the shot. The optimized parameters for shot peening were found from SNtests using unnotched specimen: TABLE 2 Static Properties TiA16V4 TiA16V6Sn2 0.4 A2 Almen 0.3 A N steel shot S 170 steel shot + glass beads 200 percent coverage > 100 percent coverage (additional wet peening with glass beads) A couple of notched specimen from TiA16V4 also were shot peened with a somewhat lower intensity of 0.25 to 0.3 A2 at a second workshop using steel shot S 170 hard (Rc = 62).

3 SUPPLEMENTARY MEASUREMENTS The residual stresses after machining, i.e. without shot peening, as well as after shot peening were measured by Xray on the surface of several notched und unnotched specimens. In a few shot peened specimens the residual stress distribution across the thickness was also measured, as well as the surface roughness before and after shot peening. Finally the decay (or otherwise) of residual stresses due to the applied stresses was determined in unnotched and notched specimens under constant and variable amplitude loading. All Xray measurements were carried out by Dr. Franz at MBB, wh~ch the author gratefully acknowledges. RESULTS Unnotched flat specimen made from TiA16V4 or TiA16V6Sn2 peened with similar parameters and tested under constant amplitudes have quite the same fatigue life, Figure 3. The improvement of fatigue life by shot peening, however, is very different for both materials because of the significantly different SNcurves for the unnotched specimen without peening. Shot peening improves fatigue life of unnotched flat specimen of TiA16V6Sn2 by a factor of 40, while at the same stress amplitude + Ca = 400 MPa the improvement for the TiA16V4specimen is only about factor 5. The higher improvement for TiA16V6Sn2 may not only be caused by the higher compressive stresses induced in the flat specimen, Figure 4 and the different value of R = 0.25: The SNcurve for unnotched specimen from TiA16V4 without shot peening is very steep (k = 3.3) compared to the SNcurve of TiA16V6St12 (k = 4.9). Also the relatively high compressive stresses already after milling (up t o 6 = ~ 500MPa) may have shifted the SNcurve for TiA16V4 to higher cycles to failure. The difference between compressive stresses already from fabrication and the maximum compressive stresses after shot peening became relatively small and therefore the improvement of fatigue life is not very large for TiA16V4. For flat specimen under a flightbyflight sequence (only determined for TiA16V6Sn2) there is an improvement of fatigue life by shot peening of about factor 10, already for a maximum stress near Cy. Figure 5 shows the stress distribution in thickness of flat specimen (TiA16V6Sn2) under a fl ightbyfl ight sequence after 0, 800 and fliqhts. There is only a small relaxation of the compressive residual stresses by theservice stresses for TiA16V6Sn2. There is a smaller improvement of fatigue life of notched specimen by shot peening, about factor 2,under constant amplitude loading (Kt = 1.6 only tested for TiA16V4), Figure 6. It can be seen from this figure too, that shot peening done with similar intensity in two different workshops leads to about the same fatigue life in the finite life region while the fatigue limits are slightly different. Figure 7 shows similar results for notched specimen of TiA16V4 under a flightbyflight sequence. The improvement of the fatigue life by shot peening is about factor 2 to 3 even at maximum stresses well above 6 for axial loading. Y CONCLUSIONS In general there is a favourable effect of shot peeningon the fatigue life of TiA16V4 and TiA16V6Sn2 using optimized peening parameters. The factor of improvement ratio of fatigue life with and without shot ~eenin~ mav be influenced bv the manufacturing quality of the specimen iested. There seems to be a larger improvement o f fatigue life by shot peening for TiA16V6Sn2 (factor 10 to 40) than for TiA16V4 (factor 5) but only part of it may be explained by the d ifferent compressive stresses induced by shot peening. REFERENCES Ni kulari, A., March 1983, Experimental Techniques, 21 Schutz, W., 1981, International Conference on Shot Peening, Proceedings ICSP I. Wagner, L., and Luetjering, G., Influence of Shot Peening on the Fatigue Behaviour of Titanium Alloys Schutz, W., and Schrader, P., 1976, Fatigue Testing and Design, Vol. 1 Macherauch, E., and Starker, P., 1983, Z. f. Werkstofftechnik, 14 Franz, E., 1978, VDIBericht No. 313 Almen, I.O., 1943, Iron Age

4 6a Stress Amplitude ( MPa 1 Z!? gr, c : 1 t Distance from the Surface

5 Asn 1325'32 \ THE EFFECT OF DYNAMIC STRAIN AGEING DURING HOLD TIME ON FATIGUE BEHAVIOUR S.W. NAM *, I. S. CHOI *, and K. T. RIE ** not shotpeened IABG workshop/,25 0,30A2, S170hard Ti Al6VG Kt= 1.6 RE0,25 ( FALSTAFF ' ' L 6 I L 16 No. of Flights to Failure Figure Fatigue life lines under a flightbyflight loading not shotpeened k=6,7 7 Effect of hold time on low cycle fatigue behaviour of AISI 4140 steel at K was investigated. Experimental results show that dynamic strain aging is occurring during hold time period and the dislocations are locked by the Cottrell atmosphere. Formation of the Cottrell atmosphere results redistribution of carbon or carbides. These randomly dispersed carbides behave as the increased number of dislocation pileup sites from which fatigue cracks may initiate. Therefore, the dislocation density at each site will be lowered to retard the fatigue damage or improve fatigue life. INTRODUCTION According to Coffin and Manson, under lowcycle fatigue condition, the relation between the number of cycles to fracture (NF) and the plastic strain range (hep) is + (NF)" = constant which is the CoffinManson rela Rie (5) and Lachman (6) have shown that, instead of using the data of NF in Coffinhlanson relationship, the value of Ncr (the number of cycles leading to the damage state which noticeably weakens the specimen) can be used for the better fitting of the Coffinhlanson relationship with the real data.?i?rl,. I a ' r T " r T m 3.. Io~ lo5 To6' ' ' ' """. N NO. of Cycles to Failure Figure 7 SNCurves for notched specimens It has been suggested, in AS%E Code Case N47, that the life of the material under the situation of combined * Dent of Materials Science and Engineering_ Korea Advanced Institute of Science and Technology 3.0. Box 131 Cheongryang, Se ** Inst. fur Schweisstechnik u Technische UniversitYt Braunschwein..OLD 4 Langer Kamp 8, 3300 Braunschweig, West~errnan~