I will have to qualify and say I did not see the stellite we lost, only a verbal report from Herb Becker. I don't know if it crumbled as you show and describe or if it "came detached". It was likely either Tim Joyce or Rob McClendon who was racing it at the time.
The point is you describe what has happened (what you see and hear) and I am trying to gain some insight into why it is happening. Most all of your spintrons show valve bounce at the seat but we don't know if the valve is lofting off the nose or not. Do you suppose the damage to the stellite is while the valve is bouncing on the seat - but how can that be when there should be valve lash and perhaps an adequate hydrodynamic wedge of lubricant.
Furthermore, my 750 USS ran a N480 with periodic excursions to 9,000 rpm and regular excursions to 8,000 rpm with only one failure. The 500 USS even lived with a rev limiter set to 9,500 rpm and we used every bit of that with no loss of stellite.
Both engines used lighter (probably stiffer) cam followers and lighter and enhanced valve train components though with the 500 we resorted to a gear driven valve train and purpose built cam (with center bearing of Herb Becker's design) designed by Professor Gordon Blair.
What I am saying is that with all we did, I would consider myself very lucky if either of my valve trains did not hammer yet we only lost one stellite. I am looking at the differences between your and my valve train components and applications here and trying to make some sense out of it.
I am just trying to get my head around the whole thing and the stellite loss is just one bit of the whole thing.
Thanks again for all your effort