Properties of Densification by Surface Rolling and Load Bearing Capacity of 1.5Cr-0.2Mo High Density Sintered Steel Rollers and Gears* Teruie TAKEMASU**, Takao KOIDE***, Yoshinobu TAKEDA****, Daisuke KAMIMURA** and Masato NAKAMOTO** ** Department of Mechanics and Systems, Tokyo University of Science, Suwa, 5000-1 Toyohira, Chino, Nagano 391-0292, Japan E-mail: t-tkms@rs.suwa.tus.ac.jp *** Department of Mechanical and Aerospace Engineering, Tottori University, 4-101 Minami, Koyama-cho, Tottori 680-8552, Japan **** Hgans Japan K.K., Akasaka Shasta East 2-19, Akasaka 4-chome, Minatoku, Tokyo 107-0052, Japan
Abstract Surface rolling experiments and surface durability tests were carried out using powder metallurgy (P/M) gears made of 1.5Cr-0.2Mo single-press single-sinter (1P1S) high density (7.5510 3 kg/m 3 ) sintered steel. The fundamental densification properties of this P/M material were first examined using P/M rollers. A high precision form rolling machine of two roller-dies transverse type was employed. The porosity became nearly zero from the surface down to a depth of at least 0.5 mm when the amount of decrease in the roller radius was more than 0.15 mm. Next, the gear tooth profile analysis was done by finite element simulation (FE-simulation) to optimize some shape parameters of a modified convex tooth profile of P/M spur gears before rolling. Simulation results agree well with the experimental data and gears with good tooth profile accuracy and a fully densified surface layer of about 0.5 mm in depth on the gear flank could be obtained. The gear running tests were finally conducted using a power re-circulating type gear testing rig with a counter pinion gear made of Ni-Cr-Mo wrought steel. Both P/M gears and counter pinions were case-carburized and finished by grinding under the same conditions. The test results confirmed that high density 1.5Cr-0.2Mo P/M gears had sufficiently high load bearing capacity as an automotive power transmission gear and those surface durability tended to increase by surface rolling. Key words: Powder Metallurgy, Sintered Steel, Roller, Gear, Surface Rolling, Densification, Load Bearing Capacity
1. Introduction The gears used in automotive transmissions are made of low alloy wrought steels for case hardening. They are manufactured through many processing steps such as soft machining, heat treatment, and hard finishing. Thus, it is very difficult to reduce the total production costs for such high precision, high performance gears. In the automotive industry, there is strong demand for net shape forming with low environmental impact, and many kinds of automotive parts are currently produced by chipless processes such as plastic forming or powder metallurgy (P/M). P/M technologies are especially suitable for *Received 28 June, 2011 (No. 11-0354) [DOI: 10.1299/jmmp.5.825] Copyright 2011 by JSME
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 826 producing complex-shaped parts such as gears. However, P/M gear preforms lack the extremely tight accuracy often required by the automotive industry. Furthermore, the fatigue strength of P/M gears is also limited because of the porosity of the P/M texture. Therefore, P/M gears have been associated with relatively low-cost and low-performance gear applications, such as oil pumps, silent chain sprockets in automotive engines, and gears for two-wheeled vehicles. It should be noted that previous studies have demonstrated that P/M gears can be manufactured with fairly narrow tolerances on the tooth profile, very good surface finish, and highly improved load bearing capacity due to the surface densification provided by finish gear rolling (1-3) . The development of high performance sintering steel powder and the introduction of high density technologies remarkably improved the density levels of P/M parts and their mechanical strengths. Warm compaction or high velocity compaction offered density levels of P/M parts up to 7.210 3 -7.310 3 kg/m 3 . High pressure compaction produced P/M parts with a higher density of up to 7.510 3 -7.610 3 kg/m 3 by the single-press single-sinter (1P1S) route, and offered a lower cost alternative to the double-press double-sinter (2P2S) route. Recently, high performance chromium alloyed sintered steels have been developed for automotive transmission gears. It has already been demonstrated that gears of these P/M materials with a density of 7.210 3 -7.610 3 kg/m 3 can achieve gear tooth bending strengths and gear pitting resistances higher than 1000 and 1500 MPa, respectively (4,5) . At this time, we have obtained high density 1.5Cr-0.2Mo sintered steel packs. In this article, surface rolling experiments were first carried out using rollers and gears made of this newly developed P/M material. Surface densification was examined and the rolling process for a P/M spur gear was optimized. Subsequently, the load bearing capacity of the P/M gear was investigated through gear running tests in a force square rig. Failure modes, fatigue strengths and the effects of surface rolling on endurance strengths were determined. The purpose of this study is to optimize a modified tooth profile of the P/M preform gear by FE-simulation to obtain surface rolled gears with true involute tooth profile, and to assess its applicability as a power transmission gear material by measuring fatigue strength. 2. Experimental and Simulation Methods 2.1 P/M Material Table 1 shows the chemical composition of the tested 1.5Cr-0.2Mo P/M material, which is manufactured by 1P1S route. This material (hereafter called FL520X) is made from completely pre-alloyed sintering steel powder to bring the composition close to that of the wrought alloyed steel of SCM415 in JIS, which is the most commonly used material for transmission gears. The carbon content is suppressed to a relatively low 0.23 % to control work hardening and to improve surface densification by surface rolling. Compared to conventional 1P1S P/M materials, FL520X has a very high density of 7.5510 3 kg/m 3 and a low porosity of 3.2 % due to the special process requirements of worm die compaction at 70 C, high compaction pressure of 950 MPa and high temperature sintering at 1,280 C.
Table 1 Chemical composition of P/M material (wt %) C Cr Mo Fe 0.23 1.50 0.20 Bal.
2.2 Roller and Gear Specimens Fundamental surface rolling experiments were carried out using P/M rollers shown in Figure 1, which were machined from P/M preform disks (80 mm 20 mm). Gear surface rolling experiments and fatigue strength tests were carried out using spur
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 827 gears and pinions shown in Figure 2. Two kinds of P/M gears of as-sintered GH and surface rolled GHR machined from P/M preform disks (110 mm 30 mm), and a wrought steel counter pinion were prepared for the gear running tests. Since it was found that the case-carburized wrought steel pinions of SCM415 did not have enough load carrying capacity in the pre-running tests, SNCM420 was employed as a reference material of counter pinions in this study. The dimensions of these test gears are summarized in Table 2. The surface rolled P/M gears of GHR were cut by the modified hob optimized by FE-simulation. Other test gears of as-sintered GH and wrought steel were cut by the standard hob. All gears and pinions used in the fatigue test were case-carburized under the same conditions as shown in Figure 3 and finished by grinding to have a maximum surface roughness of Rz < 2 m. Figure 4 shows the micro-hardness profiles around the pitch points of GH and GHR, respectively. The effective carburizing depths are about 0.6 mm. The effect of surface rolling on the surface hardness is found to be very small.
Fig. 1 Shape and dimensions of P/M roller specimen
(a) P/M gear (b) SNCM420 pinion Fig. 2 Samples of test spur gears
Table 2 Dimensions of test spur gears Gear (GH, GHR) Pinion (SNCM420) Module 3 mm Pressure angle 20 Number of teeth 26 13 Face width 6 mm 17 mm Outside diameter 84 mm 46.41 mm Pitch diameter 78 mm 39 mm Root diameter 70.5 mm 32.91 mm Addendum modification 0 0.24
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 828
Fig. 3 Case-carburizing conditions
Fig. 4 Micro hardness profile of test gears
2.3 Surface Rolling Apparatus and Methods Rolling experiments were conducted using a form rolling machine FA-20/CNC of two roller dies transverse-type developed by Nissei Co. Ltd. Figure 5 (a) and (b) show the configurations of the two rolling tools and a P/M roller or a P/M gear between the tools. In
(a) Surface rolling of P/M roller
(b) Surface rolling of P/M gear Fig. 5 Configuration of P/M test specimens and rolling die tools
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 829 these processes, the two tools are rotated synchronously, and only the tool on the right side is pushed radially into a P/M workpiece until a pre-set center-to-center distance between the right and left tool spindles is reached. The CNC system can effectively achieve the exact center-to-center distance with a resolution of 1 m. After the pre-set center-to-center distance between the tool spindles is reached, the rolling state is maintained for about 1 s (i.e., dwell operation), after which the processing ends. In the surface densification tests of the P/M rollers, the tool is a roller die of SKD11, whose outside diameter is 160 mm. A tool rotation speed of 60 min -1 and a radial displacement speed of 0.05 mm/rev were employed. The amount of radial tool displacement, d 0 , was varied in the range of 0.0-0.8 mm. In the surface rolling experiments with gears, the rolling die tool is a spur gear of SKH51 with a standard tooth profile meeting an accuracy of JIS 2 grade (Figure 6). The tool dimensions are 3.0 module, 20 pressure angle, 57 teeth, and 30 mm face width. A tool rotation speed of 60 min -1 and a radial displacement speed of 0.167 mm/rev were employed. As d 0 was set to less than 1.50 mm, the net processing time was less than 10 s. The commercial press oil LF-7 (Idemitsu Kosan Co. Ltd.) was used as a lubricant.
Fig. 6 Tooth profile of the rolling die tool
2.4 Finite Element Simulation Model The gear surface rolling process was modeled and simulated using the commercial finite element code, ABAQUS/Standard. Figure 7 shows the initial state of FE-model in two-dimensional plane-strain condition proposed in this study. The configuration of the two rolling tools and the gear specimen is the same as in the experiments. The tool is an elastic body, whose Young modulus and Poissons ratio are 200 GPa and 0.3, respectively. The gear is an elastic-plastic porous material according to Gursons constitutive equation (6) , whose initial porosity, Youngs modulus, and Poissons ratio are 3.2 %, 190 GPa, and 0.29, respectively. Flow stress-strain relationship of P/M material obtained from the compression test is given by
261 . 0 924 = (1)
where, is the equivalent stress [MPa] and is the equivalent strain (7) . The manner in which deformation occurs is as follows. The internal surface of the spindle hole of the gear or the tools sticks to a rigid beam element that rotates freely around its rotation center. The two tools rotating synchronously in anti-clockwise direction are pushed radially into the gear at the same constant speed of 0.0835 mm/rev depending on the experimental condition. The total amount of radial tool displacement is set to 1.20 mm to obtain a surface densification layer of about 0.5 mm depth. Coulombs friction law is assumed in the contact area between the gear and the tools, and the coefficient of friction is set to = 0.05. The
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 830 modified tooth profile of the P/M preform gear of GHR, which was assumed to be markedly convex around the pitch point, was designed to yield the true involute tooth profile after surface rolling. Figure 8 shows the tooth profile of the modified hob to cut GHR. Four kinds of shape parameters 1 , 2 , 1 and 2 were optimized in this tooth-profile analysis.
Fig. 7 FEM simulation model
Fig. 8 Tooth profile of the modified hob
2.5 Durability Test Methods Gear running tests were conducted using a power re-circulating type test rig as shown in Figure 9. A 3.7 kW variable speed motor was used as the power source. The maximum re-circulating power was 10 kW. The rotation speed of the wrought steel pinion was set to n = 1800 /s (the gear speed = 900 /s). The mineral lubricating oil RO150 (Nippon Oil Corporation) , whose kinematic viscosity is 150 mm 2 /s at 313 K and 14.6 mm 2 /s at 373 K, was supplied at a rate of 0.4 L/min at 313 K. The running-in operation before an actual running test was conducted under a light loading condition of 50% fatigue strength at 1.510 7 cycles for 10 5 loading cycles. The contact fatigue life was defined as the number of load cycles at which 2% of the tooth surface of the test gear was pitted, and the surface fatigue strength was assessed at 1.510 7 cycles (8) . In these tests, the pitting failures of wrought steel pinions always occurred earlier than those of gears. Subsequently, a failed pinion was changed to a new one, and the test was continued until a test gear was pitted. The surface contact stresses of gears at the pitch point were deduced from the Hertz stresses that arise from the contact pressure of two contacting cylinders with the same curvature
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 831 radius for the gear tooth involute surface (9) . The maximum surface compressive stress H is given by
| | . |
\ |
+
+ = 2 2 2 1 2 1 2 1 2 1 H 1 1 ) ( E E R R R R w
(2)
where, w is the load per unit length [N/mm], R is the equivalent curvature radius of each cylinder [mm], E is Youngs modulus [GPa], and is Poissons ratio. Figures 10 and 11 show the tooth profile and lead curves of a P/M test gear and those of a wrought steel counter pinion, respectively, used in the gear running tests. Errors for both the tooth profile and lead are very small. These correspond to JIS 3 grade accuracies.
Fig. 9 Power re-circulating type gear running test rig
Fig. 10 Tooth profile and lead curves of P/M gear
Fig. 11 Tooth profile and lead curves of wrought steel pinion
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 832 3. Results and Discussion 3.1 Surface Rolling of Rollers Surface densification was examined by observing the pore structure in the surface densified layer of rolled P/M rollers. A sample specimen was cut out parallel to the roller axis as shown in Figure 12. In this figure, the observation area is the hatched portion, which is 3 mm in width and 2.5 mm in depth. Figure 13 shows sample metallographs of each surface densified P/M roller when the amount of decrease in roller radius r is 0.05, 0.10, and 0.15 mm. These r values correspond to the target amounts of stock rolled normal to the gear flank surface. Figure 14 shows the porosity distribution in this area as a function of r. These distribution curves were obtained using the binary image processing, in which an area ratio of pore judged black in the bitmap image of the observation area was calculated at an interval of 0.1 mm depth. These figures indicate that the surface densification of FL520X sintered steel is intensively promoted at the surface up to a depth of about 0.5 mm regardless of r. Subsequently, the porosity gradually increases to a depth of about 2.0 mm, and becomes equal to the initial value. Specifically, the porosity in the surface layer of about 0.5 mm in depth becomes less than 0.5 % when r is 0.10 mm, and reaches almost zero when r is 0.15 mm. Accordingly, this high density Cr-Mo sintered steel could open up new possibilities to make the surface layer fully dense up to a certain depth by adjusting the amount of surface stock rolled.
Fig. 12 Sample specimen and observation area
r : 0.05 mm 0.10 mm 0.15 mm Fig. 13 Metallographs of surface rolled P/M rollers
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 833
Fig. 14 Porosity distribution of surface rolled P/M rollers
3.2 Surface Rolling of Gears According to the experimental results using rollers, the amount of stock rolled normal to the gear flank surface necessitates an adjustment of 0.15 mm or more to secure a fully densified area after surface rolling. Hence, the shape parameters 1 and 2 in the modified tooth profile of P/M gear preforms (Fig. 8) were set to 0.15 mm. The other parameters 1
and 2 were optimized using FE-simulation and determined as 2.0 and 1.75 mm, respectively. Figure 15 shows a comparison of changes in gear tooth profile curves with the amount of radial tool displacement d 0 . The tooth profile curves calculated by FEM agree well with those obtained in experiments throughout stages of deformation. Specifically, as d 0 increases, the tooth profile curves gradually align with the true involute, and P/M gears with good involute tooth profile curves on both the driven side and the follower side (5) can be obtained when d 0 reaches 1.20 mm. On the other hand, the tooth lead profiles in experiments remained almost flat regardless of d 0 .
Fig. 15 Comparison of rolled gear tooth profile between FE-simulation and experiment
Figure 16 shows a metallograph of a surface rolled gear tooth on the follower side when d 0 = 1.20 mm. A uniformly densified layer can be observed on the entire gear flank surface from the tip to the base circle. The densification level of FL520X gear is extremely high especially at the surface down to a depth of approximately 0.5 mm. The porosity
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 834 distribution around the pitch point of this surface rolled gear is shown in Figure 17. Note that the material in the gear surface layer at a depth of about 0.5 mm on both sides is fully dense. These densification properties are similar to those of FL520X rollers. They would appear to be ideal for improving surface fatigue strength, since the maximum shear stress in a sliding rolling contact typically acts beneath the surface at a depth of 0.2 mm depending on the contact width and magnitude of friction.
Fig. 16 Metallographs of surface rolled P/M gear tooth on the follower side
Fig. 17 Porosity distribution around the pitch point of surface rolled gear tooth
3.3 Durability Test Results Figure 18 summarizes the gear running test results. The ordinates indicate the Hertz stress H , whose values are calculated using Youngs modulus E = 206 GPa and Poissons ratio = 0.3 for wrought steels. The failures of all FL520X test gears involved pitting. Thus, the load bearing capacity of this P/M gear is dominated by the surface fatigue strength and not the bending fatigue strength. Figure 19 shows comparisons of the surface fatigue strengths for the test gears after 1.510 7 cycles. These results confirm that surface densification by finish rolling can substantially enhance the surface fatigue strength of FL520X gears. Specifically, when a P/M gear is set as the driver, the surface strength of the as-sintered GH is about 10 % smaller than that of the case-carburized wrought steel pinion of SNCM420. However, when it is used as the follower, its strength level almost matches that of SNCM420 pinion. On the other hand, the surface fatigue strength of the rolled GHR after 1.510 7 cycles increases by about 5-10 % compared to that of the unrolled GH. Thus, the load bearing capacities of GHR become higher than those of SNCM420 pinion, both in the number of load cycles and in the surface fatigue strength at 1.510 7 cycles.
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 835
(a) S-N curves as driver
(b) S-N curves as follower Fig. 18 Gear running test results
Fig. 19 Comparison of surface fatigue strength
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 836 In this study, we also conducted the gear running tests using P/M pinions, and confirmed that there was little difference in the surface durability between P/M pinions and P/M gears. Therefore, it seems to be no problem to compare the strength level of P/M gears (GH and GHR) with that of the wrought steel pinions of SNCM420. Figure 20 shows failures on the gear tooth surfaces. Pitting failures occur near the pitch point of the FL520X gears GH and GHR, as is the case with the wrought steel gears.
Fig. 20 Surface failures of P/M test gear
4. Conclusions The main results obtained from this study are summarized as follows. 1) A fully densified surface layer of at least 0.5 mm in depth can be obtained when P/M rollers made of high density 1.5Cr-0.2Mo sintered steel were surface rolled by setting the amount of decrease in the radius to 0.15 mm or more. 2) A modified convex tooth profile of the P/M test spur gear is proposed. The shape parameters are optimized using FE-simulation, whose results agree well the experimental data. 3) A high precision form rolling machine enables sufficient control to make surface rolled spur gears with tooth profiles of JIS 3-4 grade accuracy and a fully densified surface layer of about 0.5 mm on the gear flank. 4) The surface fatigue strength of high density Cr-Mo P/M gears without surface densification after 1.510 7 cycles almost matches that of case-carburized gears made of the typical Ni-Cr-Mo alloyed steel SNCM420. The surface durability of these gears tends to increase by surface rolling. 5) The surface rolled P/M gears made of high density 1.5Cr-0.2Mo sintered steel have sufficient load bearing capacity as automotive transmission gear material. References (1) Hayasaka, H. and M. Suzuki, M., Roll Finishing of the Sintered Compact, J. of the Japan Society for Technology of Plasticity (in Japanese), Vol.18, No.195(1977), pp.307312.
Journal of Solid Mechanics and Materials Engineering
Vol. 5, No. 12, 2011 837 (2) Roger, L., Rolling-Contact Fatigue of Surface-Densified Gears, Int. J. Powder Metallurgy, Vol.42, No.1(2006), pp.1729. (3) Koide, T., Ishizuka, I.,Takemasu, T., Miyachika, K. and Oda, S., Load Bearing Capacity of Surface-Rolled Sintered Metal Gears., Int. J. of Automation Technology, Vol.2, No.5(2008), pp.334340. (4) Kotthoff, G., Leupold, B. and Geiman, T., Predictive Model Development for PM Gear Rolling, Int. J. Powder Mettallurgy, Vol. 42, No.1(2006), pp.3145. (5) Takemasu, T., Miura, H. and Komoto, K., Development of High Precision and Strength Sintered Ferrous Alloy Gear by Rolling -1st Report- Numerical Tooth Profile Analysis in Finish P/M Gear Rolling Using Screw-Shaped Tool, J. Jpn. Soc. Powder Powder Metallurgy (in Japanese), Vol.56, No.6(2009), pp.330-335. (6) Gruson, A. L., Continuum Theory of Ductile Rupture by Void Nucleation and Growth: Part I Yield Criteria and Flow Rules for Porous Ductile Media, Trans. ASME, J. Engng. Materials and Technology, Vol.99(1977), pp.2-15. (7) Goto, M., Theory of Plasticity or Mechanics of Plastic Materials, (1982), pp.37-68, Corona Published Co., LTD. (8) Sonsino, C. M., Course of SN-curves Especially in the High-cycle Fatigue Regime with Regard to Component Design and Safety, Int. J. Fatigue, Vol.29, No.12(2007), pp.22462258. (9) Timoshenko, S. P. and Goodier, J. N., Theory of Elasticity -Third Edition-, (1951), pp.409-420, McGraw-Hill Book Company.
