Indy RHS head failure

nm9stheham said
The .032" number comes from working out the geometry. I actually worked it out last night for a 1.6 PRW rocker and for the most ideal rocker shaft height setting, it worked out to .034" of roller contact movement across the valve tip.

The PRW has a shorter fulcrum length than the one I used for my calculation. Shorter fulcrum=more sweep.

nm9stheham said
The scrub pattern above does not look all that wide; it scales to around .080" wide. But don't look at that pattern in the context of a standard stamped or 273 rocker, or even for other roller rocker brands. They will be considerably different, 'specially the stock types versus a roller type.

Sorry, but I think a sweep pattern that is 200% wider than it should be is problematic.

nm9stheham said
Having said that, I am not yet convinced that engines will always explode with something different than the 'most ideal' rocker shaft height setting for many or most engines. This whole theory has to do with the valve velocities varying too much and thus causing valve float. My initial results last night were that with .600" valve lift, and the shaft centerline too low by about .300" (a VERY extreme situation); the scrub width would be .130" wide the valve velocity varies no more than around 10%. (These number are for the PRW 1.6 rocker only so don't take it as a general result that applies to all rockers.)

First, .300" is not unheard of. I've seen and corrected plenty that were off that much and more. Also, I believe you might want to recheck your work on the velocities. There should be no velocity at full lift when the geometry is correct because the valve stops and dwells, just like the piston does at TDC. Try to visualize what would happen if it was possible to have max piston speed at TDC.

nm9stheham said
That max velocity is at peak valve lift in that really bad case scenario, but all that gets tempered by the fact that the velocity is reaching zero at that point due the cam profile anyway. My bet is that this may be an issue when you get to extreme cases and highest lifts, up in the high .500's range, but not a worry at lifts at .500" and lower.

Really, I had a .490" lift small block pick up 600 rpm on the dyno from correcting the geometry, and it was unstable 1000 rpm below peak. Also, the cam lobe dwells very little at TDC on a flat tappet, and the lobe velocities are multiplied by the rocker ratio. Try running full speed at a brick wall, and then just stop right at the wall, never mind trying to turn around and run back to where you started.