AGMA 16FTM14 PDF

The actual power loss due to friction in a gear pair is of significant interest to the transmission community. This loss impacts fuel efficiency on almost every form of transportation. In aircraft applications, this loss converts to heat, which must be removed using cooling equipment, inherently detracting from the payload of the aircraft. Consequently, the impact of gear tooth surface characteristics and lubricant on friction and frictional losses at the gear mesh is of significant interest to designers of high performance gear boxes. While a prior paper [1] compared the change in losses due to these factors, this effort is intended to actually evaluate the effective coefficient of friction at the gear tooth flank under the conditions described below.

By instrumenting the input and output ends of the test box of a four-square, power re-circulating gear test rig with high-accuracy torque transducers, losses at the meshing gear pair in the test box are measured. The difference between the readings of the two torque transducers are the losses in the test gear box. By subtracting bearing and seal losses and neglecting windage losses, the friction losses at the test gear mesh are obtained. The nominal torque in the four-square loop is known, therefore the “effective coefficient of friction” at the tooth flank is calculated.

The gears evaluated in this study are 6.25-mm face width, 21-tooth, 4.233-module, 20-degree pressure angle spur gears. Gears were fabricated from AMS 6308 steel, gas carburized, and oil quenched, resulting in a surface hardness of 60–64 on the Rockwell C scale. This experimental procedure was carried out for three operating speeds (5000, 8100 and 10 000 rpm) and three different gear tooth flank conditions: ground, Isotropic Superfinished (ISF®), and Tungsten Diamond Like Carbon (W-DLC) coated. Surface roughness (Ra) on the tooth flank ranged from 0.06 to 0.29 μm. Two different gear lubricants (Mobil Jet Oil II/MIL-PRF-23699/ISO VG22 and Mobil SHC 626/ISO 12925-1 CKD/ISO VG68) and two different torque levels (96 and 192 N-m) are also considered in this study. This paper summarizes the results obtained and the effective coefficient of friction at the gear tooth flank under the conditions described above.