Oil Pump Breaks on Dyno-Torsional Harmonics?

[raptor04]

When dyno testing SBC sprint car motors on a Superflow X test dyno we are breaking the cam driven Dailey Engineering engine oil pump. It can happen on the first or second pull. The allen head screws break that hold the drive gear on the lobe and locks up the pump. Sprint car motors do not run a harmonic balancer. Do you think torsional vibration is causing this issue? Any ideas how to cure this issue. I have attach the link to vibration spectral data.

 

[Teacups]

Mike H - Maybe I shouldn't have used the word "sheared". But there still is no answer to the original question of why the oil pumps repeatedly failed. The main question wasn't - "Where is the vibration coming from"? It was - "Why is the oil pump breaking"? Or maybe I have totally misunderstood (- always possible).

 

[raptor04]

Teacups, I can only guess at what was braking the pump lobe set screws. Maybe the cracks in the impeller caused some sort of flexing/loading/unloading that generated torsional energy which transferred through the crank-cam gears-oil pump gears/lobes and stress/shock loaded the oil pump lobe screws. The overall G's went from 24.83 G's to 16.96 G's and floor energy dropped 8.4 to 3.8. I have attached the before and after spectral data. If anybody else has a guess at the WHY please chime in.

 

[Teacups]

I don't have enough knowledge of vibratory-type matters to really comment. However my intuition would make me inclined to think that a pump etc. meant to take racing stresses would not be prone to break so easily from vibration.