From what I can tell from the pictures this is a pretty odd failure.....
From the lubrication standpoint: Consider that the gear teeth faces and the splines are lubed by the same lubricant (or supposed lack there-of) that the splines would be lubed by. Since the gear teeth faces take FAR more lubricant related grief than the splines do (sliding contact versus no relative motion) and the gear teeth faces look absolutely perfect I think you can safely rule out any sort of lube failure. In fact, I would be pretty certain it is not lube related.
From the condition of the splines it would certainly look like poor metalurgy...i.e..lack of or poor heat treat... IF only one set of the splines had fatigued away. Since both are fretted heavily to the point of dissappearing it would seem that the distress was equal on both sets of splines thus implying that both sets of splines had poor metalurgy. Not likely. If metalurgy was the top culprit I would suspect (from previous failures of gears and such I have seen) that only the part with the poor metalurgy would have the problem or be much more damaged than the other. So...for the moment I would rule out metalurgy.
I am not sure of how the unit bolts together but I can suppose that the gear is clamped quite firmly to the inner support bearing race. That type of gear requires proper geometry during install such that the "depth" of the gear is critical to properly mesh with the longitudinal shaft gear. So it would not be just "floating" on the splines. The depth of the gear relative to the case would have to be fixed by clamping to the inner race of the bearing that seats in the case. So...there shouldn't be any relative motion of the gear on the splines per se. It should be firmly clamped up in an actual press situation so the motion on the splines would be virtually zero.....unless....the nut were not tightened sufficiently. If that were the case then the gear could fret around on the splines and wear and fatigue them over time and lots of miles.
More evidence of this is on the inner face of the gear. Look at where the anti-rotation washer rests against the inner face. See any shiny polishing? I don't. If that washer were firmly clamped against the gear by the nut (properly tensioned) then I would expect to see some pretty shiny material from the washer moving around on the gear....or the gear moving around relative to the washer per se.... as the splines fretted and there was more and more free movement of the gear relative to the shaft. More possible evidence that the nut was not tight enough.
Since this seems to be a total one off failure you might suspect something like a loose nut (or just inadequately torqued/tensioned nut) rather than a metalurgical problem. When things like poor heat treats happen it NEVER EVER happens to just one part. Never. LOL. Loose nuts can happen on a one-off basis.
My vote based purely on speculation and absolutely no knowlege is that........................ the nut on the drive gear that failed was not properly tensioned at the factory thus allowing the gear to float around on the splines that were not designed for the type of drive system and that wore the splines and caused the shaft and gear splines to wear and fret evenly and fail at the same time.
A right angle drive gear like that has considerable thrust from the gear teeth face which would try to constantly "rock" the gear on the splines if it were floating and "loose" on the splines thus causing the type of wear and fretting shown. A gear with that type of side loading would never be designed to float on the splines simply from a spline longevity standpoint (see what happens when it does). That type of gear would always be "nailed" to the shaft in a tightly compressed joint via a nut with a lot of compression on the joint.
Probably the gear/splines would have failed much sooner but for the excellent lubrication it was receiving....LOL
I cannot imagine that the motoman breakin procedure (or any break in procedure) would cause or influence this failure at 70K miles. If the gear teeth faces had scuffed and failed at 5000 miles then the breakin might be suspected from overloading "green" gears but the splines are not really influenced by the lube and not a "moving" joint that would be affect by load like that for breakin. Splines that do not move do not "break in" per se so the heavy loading while green wouldn't hurt them. Might hurt the gears, though, but since they are fine I would eliminate the breakin as a suspect.
Understand, though, that harsh breakin does have an effect on hypoid gears (such as the rear end gears in cars) and the load bearing or torque loading capability of the axle goes up at least double after breakin. MANY MANY motor homes have been bought in places like Phoenix or Denver, loaded to GVW while green and then driven up a steep grade at full load....only to fail the ring and pinion due to excess heat from lack of breakin and subsequent lack of load bearing capability while green. Excess heat always shows up when this happens.
My 78 XS1100 has over 108,000 on it and the middle gear set has never uttered a peep. It is lubricated separately with gear oil like the final drive as mentioned and the oil is always clean and fine. So, I expect the XS1100 failure is a lot like this one on the FJR.....a one off. That is what I am trying to make myself believe, anyway.....LOL.
As long as we are on the subject:
The item in the picture that looks like a gear was a gear until 90% of the teeth fretted away. The remaining "fence" of nubs along one side is what the gear teeth looked like. That area was not in engagement with the mating gear (by design) so they were not fretted away. The mating portion of the gear fretted completely away. This was top gear in a Hewland gear box from a March Indy car circa 1985 or so. Poor metallurgy on the failed gear was the culprit...both metallurgy and heat treat, in fact. Shouldn't have happened. The mating gear looked perfectly fine despite the total destruction of the gear in the picture. The mating gear had the correct material and heat treat.
Bonus points....identify the source of the "prop" in the background....
