996, 997, etc. cyl thickness dimension

Looking for published cylinder head thickness dimension/dimensions for all 996, 997 and other water cooled engines in the Porsche line . Specifically, I am in need of the minimum thickness (deck to measuring reference point) on all of the water cooled heads. We do all Porsche heads and want to fill in the blanks in our head spec book. There must be a minimum thickness dimension, or the heads are not to be resurfaced and scrapped. Shop manuals are not useful here in so far as the later engine generations are concerned. Any help would be appreciated. Jack

I think you would just have to make experimental 'skims' on junk/cracked heads. Certainly no shortage of junked M96 engines !

Like all other M96/M97 internal specifications, these values must be gained emperically... Pretty much all the heads are safe to cut .015", after that you'll be losing structural integrity on some heads. When this happens the heads will twist easily. The biggest issue on most heads other than 3.4 996 is cutting the valve deck far enough that the valves crash into the side of the cylinder due to valve diameter, and angle. I have cut some heads .040", I have won, and lost many battles... What will bite you is core shift in some castings.

@Flat6 Thank you for the reply. In measuring heads for being planer and we have found almost 40% or so show approximately .002"-.003" high in the middle (bowed up). Which on being bolted back to a block return to perfect cam bore alignment. We have straightened a few 996/997 heads that were bowed more than this. What would you say is the maximum misalignment allowable for cam bores on a 996/997 head as installed? Example scenario: Suppose one were to take a clean up cut on a head that was bowed as described above, at say .003" high in the middle of the head compared to each end. This would result in the cam bores being out of alignment the .003" after the head was re-assembled to the cylinder block. Your comment would be appreciated. Jack

The variable you haven't considered yet is the biggest contributor here... Yiu are assuming that the cylinder deck of the bloc is 100% flat, has zero twist, and isn't also bowed.. the head conforms to whatever the cylinder deck has for a surface in most cases. The head will often bow, and twist because of the block.. so trying to correct the head issue without correcting the block will often CAUSE issues with other things. Youll have to play with these surfaces and variables to see what works for you. That's what we did over a decade ago, and we still learn today..

From notes on file: The block (referenced above in test back about twelve years ago) was near zero in flatness check on CMM. I took a skim cut on the block (both halves torqued together) just to make the test above accurate. Skim (.0010") cut visually mimicked results of CMM work. My original question is still valid. There is a minimum deck thickness for the heads (deck to reference plane/pad or whatever). That is what we need for records here. We are not hogging material off for compression increase or any other "performance" searching. The quest is not to test for limits. We want to know how much the decks (head & block) can be resurfaced before the geometry of the cam drive, etc. becomes a factor. If the p-car company refuses to publish technical information, they are quite foolish and the failure rate of their current crop of engines proves this. These engines (996 with iron crank shaft carrier) are of poor design throughout. We know this. We have been doing race engines since this company was founded in the sixties. This engine is not in any way a performance engine as per any 911 based design --- excellent. Thank you for the time. Jack