Originally Posted by
n305fa
From the REDL report the epicyclic planets are made of the following steels: Gear/outer raceway 16NCD13, rollers M50, inner race M50. The REDL gives the reasons for the different materials and details manufacture. From experience a normal spalling failure would start on the inner race if all of the materials were the same due to the increased loading of the inner race. The gear/outer race on the 225 planet is a "softer" material due to the load cycle of the gear teeth, therefore there is a tendency of the 225/332 planets to start with outer race spalling
Good post.
16NCD13 is equivalent to AMS 6263 (9315), and the material is likely supplied in the form of a roll forged ring. This material is carburized, quenched, and tempered at a temperature that gives optimum core strength for the intended operating conditions. There is large difference between the case and core hardness.
You are also correct about the loaded sector of the inner race surface being the most likely area to experience a spalling failure under normal conditions. The inner race is fixed with respect to the applied radial load, the roller/race profile contact is convex/convex, so there is one small area of the race surface in-line with the load vector that is subject to a large number of load cycles at high stress every time a roller passes by.
The bearing roller/race contact is hertzian, and as noted, the most common failure is a race surface spall initiated by a sub-surface shear fracture. One very important consideration with carburized rolling element bearing race surfaces is ensuring the case depth after finish grinding is adequate for the hertzian contact conditions. The depth of max shear stress from hertzian contact must lie well within the very high strength carburized case. If the depth of max shear stress lies at the case/core transition or even within the core, a sub-surface shear fracture initiated spall will occur fairly quickly.
There are also a couple issues that can produce a local area of reduced strength in the gear/race material. One is excessive heating of the material during grinding operations that causes local de-tempering. This is fairly easy to do if the grinding is not performed carefully, since the gear material (9315) is tempered at a fairly low temperature (typically <400degF). Of course, there are normally NDI procedures used after finish grinding to check for this problem. Another potential cause of de-temper is excessive local heating of the gear/race material during operation, due to lack of cooling oil flow, or abnormal function of the bearing such as skidding/sliding.
One reason the inner race and rollers of this bearing were made from M50 is that M50 can operate at much higher temperatures (>500degF) without loss of strength/hardness.. There are some new carburizing gear alloys that have recently become available, such as C64, that can operate at very high temperatures (~900degF) without significant loss of strength/hardness. So we'll likely see greater use of these alloys in new gearbox designs.