BGA RESEARCH PROGRAMMES

HMS InvincibleEquipped with two forward and reverse gear units, each weighs 180 tonnes and incorporate 19 gears.

Courtesy of David Brown plc.

DTI - MoD (N) funded research

The Fatigue Strength of Gear Materials

• 655 H13 (En 36) carburising steel
• 20 MnCr 5 carburising steel
• SAE 8620 H - carburising steel
• 722 M24 (En 40B) - nitriding steel.

Details of the Project work were presented at two seminars, in June 1996 and July 1997 in Sheffield for GRF Members. The final report was available to Members in the form of a full report and a summary of design data. Individual seminars and workshops will be held according to demand. The programme determined the surface and root bending fatigue strength characteristics for gears manufactured from the following gear steels:

• 655 H13 (En 36) carburising steel
• 20 MnCr 5 carburising steel
• SAE 8620 H - carburising steel
• 722 M24 (En 40B) - nitriding steel.

All test gears were manufactured from clean steel produced by British Steel Engineering Steels and heat treated to best industrial practice, before finish grinding to DIN quality 4/5. The fatigue characteristics (SN curves) were determined by testing gears to failure in four purpose built back-to-back rigs, using gears of 2mm, 4.2mm and 8mm module, running at 75mm and 160mm centres. To validate bending fatigue strength correlations with less expensive pulsator testing, a full programme of single tooth pulsator testing for bending fatigue strength was also undertaken.

In 1996 the following work was carried out:

• The test gears in 722 M 24 (En 40 B) nitriding steel were machined, hobbed, ground and nitrided, the heat treatment being carried out by Flame Hardeners and Allen Gears.
• The programme of back-to-back testing of 20 Mn Cr 5 gears was completed, with very satisfactory
• performance recorded for this low alloy steel.
• The pulsator testing of the carburised 20 Mn Cr 5 test gears, and of the 722 M 24 (En 40 B) nitrided gears was completed which showed that fatigue strength varied signifcantly with tooth size.
• The surface fatigue (Hertzian) strength of the nitrided gears in 722 M 24 (En 40 B) has been established by back-to-back testing on both the 75mm and the 160mm test rigs. Excellent surface fatigue strengths were achieved, but these are again significantly size dependant, with different modes of surface fatigue initiation and propagation.
• Further metallurgical investigations were carried out to seek to understand the cause of the different fracture initiation observed in nitrided gears, and the relative importance of case thickness and residual stress distribution.

The subsidiary programme of research into gear cleaning and the effect of different cleaning media and shot peening on root bending fatigue strength was extended and the following cleaning techniques rigorously evaluated:

• Vapour blasting with wet and dry aluminium oxide
• Blasting with chilled iron grit
• Blasting with steel shot
• Blasting with glass beads

It was found that there were significant differences in the performance of these different techniques, and that this also depended on the size of the gears.

A second subsiduary programme of research into the measurement of and the effect of grinding burn was also extended, with the evaluation of the Barkhausen Noise technique for surface residual stress and hence grinding burn measurement.

The work was completed in early 1997 and has generated reliable design data for the four steels being considered, providing considerable insight into the effect of process variables on the performance of gears. In supplying base data for current gear steels and processing and improving our manufacture and test procedures, this work has laid a firm foundation for BGA Research Project 5 to "Improve Gear Performance by Enhancing the Fatigue Strength of Gear Material".

The final report is available to Members in the form of a full report and a summary of design data. Individual seminars and workshops will be held according to demand.

index