Hi guys! After some investigation I have figured out why the shaft seal failed on me, I had to assemble the rotary parts and measure the throw in the lathe before I understood what really happened during the run.
The main problem is that the shaft tunnel and the compressor housing aren´t perfectly concentric, I didn´t use dowels to fix the engine core together but relied on the four M10 insex bolts to keep the parts alligned. That turned out to be a mistake since the two could move over 0.5mm with the bolts loosely torqued.
When I assembled the engine before this run I had the misfortune to get very bad concentricy for the engine core so the shaft piston ring seal
rested lightly against its seat. When the engine started this made the seal heat up to the point that the ring grooves collapsed from the pressure from the highly torqued compressor nut.
This was what I figured out when I fitted the assembled shaft in the lathe, properly torqued it was bent to the degree that I didn´t even bother to measure it. I removed the parts from the turbine shaft, measured them and found the shaft seal to be bent by 0.1mm. This in turn bent the entire turbine shaft and produced the imbalance clearly heard during the second start in the video, it bent so much that the compressor wings touched its housing just before the engine came to a halt but luckily without any damage to the compressor wheel.
Right now I am lining up the shaft tunnel and compressor housing in the lathe and preparing to fit dowels so they will fit together exactly the same every time I assemble the engine, a friend will drop by this afternoon with a reamer and dowels so hopefully I will have it done before the end of next week.
It is great to have found the problem, hopefully I will have most design flaws removed from the engine by the time I start racing the bike.
Cheers!
/Anders