- Issues with the Bearings on the Secondary Shaft -
Overload, Spalling Failures, Optimistic Life Claims
The table below is a duplicate of a portion of the table from Loads on the Secondary Shaft which shows the comparative loads applied to the bearings on the secondary shaft in all four operating conditions considered in this study.
|Description||Chain (max)||Chain (cruise)||Belt (max)||Belt (cruise)|
|Upper Bearing Load (lbs.)||3209||2591||4095||4060|
|Lower Bearing Load (lbs.)||165||24||575||561|
As you can see, the load on the upper bearing imposed by the belt system at cruise 1.57 times larger than the load imposed by the chain drive at cruise. Perhaps more interesting is the fact that the load on the lower bearing imposed by the belt system at cruise is over 23 times larger than the load imposed by the chain drive at cruise. These increased loads have an important effect on calculated bearing life.
The prediction of the life of a rolling-element bearing (ball, roller, needle) is a statistical calculation, in which life is stated as the number of hours that a specified percentage of an apparently identical population of bearings will survive under a specified load in a specified set of operating conditions. (See Determining Bearing Life for a complete explanation of how bearing life is determined.)
The usual life rating for industrial applications is called "L-10" life. The L-10 life is the number of hours in service that 90% of an apparently identical population of bearings will survive when subjected to the values of load, speed, lubrication and cleanliness that are specific to the application.
The SKF website provides a life-calculation program which produces THREE different values for an L-10 (B-90) life rating:
- the original Palmgren calculations,
- the proven "A23" method, and
- the new, “marketing-friendly” rating.
The one you pick depends on how realistic you want to be.
Our bearing life calculation program matches the SKF-website values calculated for an L-10 life under the "A23" method for the same input values. However, our program also includes the ability to calculate life ratings for complete load models, with combinations of radial and axial loads, and for survival probabilities more appropriate to aircraft applications (95, 96, 97, 98 and 99%).
Using our program, the expected L-5 life (the life of 95% of a population under this load) of the lower bearing (Fafnir RA100RRB) is essentially infinite at 2168 RPM with a 24 lb. radial load and a 20 lb. axial load and with grease lubrication (as in the chain drive).
However, the life of the same lower bearing using the same boundary conditions (survival %, lubricant viscosity, and cleanliness) with a 561 lb. radial load and a 20 lb. axial load (as in the toothbelt drive) is 1302 hours. (The lower bearing from the failed 30mm system we examined was near failure at 153 hours.)
The upper bearing is in a worse situation. The expected L-5 life of the 22207-E (Factory 35mm) bearing at the chain drive cruise load (2591 lbs radial and 100 lbs. axial) is 6211 hours. When subjected to the 4060 lbs. radial and 100 lbs. axial load in the belt drive system, the life drops to 941 hours.
We have not examined any 35mm spherical bearings which had failed. However, in the failed 30-mm secondary we examined, the top bearing in the upper-bearing pair had failed from contact fatigue (overloading) BEFORE the shaft broke, so there was a second major failure about to take place. Figures 1 and 2 show the condition of one of the ball elements from that bearing and he spalling which had begun in the upper race.
Because this bearing was part of the drive system which failed, it could be argued that there is "the chicken or the egg" scenario. However, our opinion, based on many failure analysis projects, is that the condition of these components indicates that the bearing failure began well in advance of the shaft failure.
This is yet another verification of the fact that the bearings are being overloaded by the ProDrive belt system. Using our bearing-life calculation program, the "B-95" life for a side-by-side pair of 3206 bearings, subjected to a realistic load model based on the steady-state (non-vibrating) load in the tooth-belt system (4363 lbs) is 48 hours. The "B-90" life (L-10) in the same load model is 78 hours. In the chain drive system, at cruise power, the B-95 life of that same bearing pair is 227 hours.
The detectable onset of bearing failure probably occurred somewhere around 10 hours prior to the shaft failure, so the life of this bearing was approximately 143 hours.
Another Amazing Discovery
We were surprised to discover an interesting procedure used in the RotorWay manufacturing process for the Two-Bearing, 30mm configuration. The two upper 30mm secondary bearings have the part number 3206B.2RSR.TVH and the manufacturer's identity (FAG) stamped into the edge of the outer races. That number identifies a bearing delivered with two seals installed.
RotorWay apparently tears one of the seals out of each new bearing, and by hand-holding the outer race of the bearings to a bench grinder, they cut a crude notch in the outer race so grease can be pumped into the back-to-back pair. (This might not be unusual in the manufacture of hay balers and the like, but certainly is unusual as an aircraft practice.)
We can't help but wonder how the stone and metal particles from the grinding process escape contaminating the grease which FAG delivers already installed in their sealed bearings. A little contamination like that can cut a bearing's life significantly. Wonder if that's a factor here.