crazy boost spike after fiddling with the timing belt
 

So here's what my MP62 supercharger with 120mm crank pulley and 65mm nose is supposed to look like:

http://pic100.picturetrail.com:80/VO.../411602369.jpg

boost is nicely in the 12psi range, which is fine for my stock motor. I replace the cam seals put the timing belt back on and all hell broke loose (not yet, thankfully, just a figure of speech.)

Here's what it looks like now:

http://pic100.picturetrail.com:80/VO.../413940215.jpg

To help diagnose the FM Link ECU graph, I offer the comment of my most excellent tuner, Ken Hill.

"Through the RPM range, the manifold pressure builds from 170kpa (9psi) to 213kpa (16.5psi) at 6300RPM. After 6300RPM the boost shoots up almost vertically to 230kpa (19.5psi) before falling off. I find it hard to believe the spike at the end of the WOT run is real, but it’s in every run when the engine hits 6300RPM. What you are feeling is the ECU’s maximum boost cut." Ken notes that the fuel is way fat, around 10:1 AFR and that there is insignificant knock so it's unlikely that I'll damage anything if I stay under 6500 rpm but that requires staying under 6500 and the boost spike is a bit of a mystery.

Here's the story of the T belt. When I first put it back on, the scribe I'd put on the CAS was a 1/4 inch to the left with timing set to 10 degrees as required by the Link ECU.

http://pic100.picturetrail.com:80/VO.../413940214.jpg

The car ran like poo and sure enough, I was one tooth off between the cams. I figured no problem, I'll just re-do the belt (I was just taking the valve cover off and the top cam shield plate.) With everything apparently aligned, the freaking CAS scribe is now a 1/4 off from the initial setting in the other direction! I was using the TDC mark on the crank pulley and the 4:00 and 8:00 O'Clock marks on the cam gears to align with the marks on the backing plate. Four tries later I have them all lined up but something is clearly wrong.

What could be causing the boost spike on a pulley driven system? And why only after 6500 RPM? Thoughts and experiences much appreciated.

Fix the timing. Then fix the boost spike if it still occurs. Dont pull a thumpetto and waste time with conjecture. Take action! Mush!

I hear you and think you are right (sounds like the voice of experience) but I'm still curious about why, what must be a slight variance in timing, would create additional boost. From a pulley system designed to generate 12-13 psi, I'm getting 19. And why only over 6500 rpm? I just finished a track day at Lime Rock, 4 25 minute sessions, and the car pulled strong all day, only all under 6500 rpm. My off the cuff theory is that if the exhaust valves aren't opening enough or in time to vent the cylinders then it's creating a back pressure restriction that increases the manifold pressure. Just weird that it happens only at the top.

Conjecture aside, I've got two requests into local shops to see if they'll set the timing for me. :)

Where are you at? If you feel like coming up to northwest NJ I will show you how to set the timing properly.

The boost with a supercharger is directly related to the VE of the engine. The supercharger moves X amount of air per rotation (give or take depending on RPM abd back pressure, but that is a smallish variance.) If the timing is off and the engine flows less air, the pressure goes up as the supercharger tries to force more through.

yep, second this. This would probably be more accurately described as a VE drop, not a boost spike.

Got it. Thanks for the input. Going to local shop tomorrow to have the timing set. I'm not really sure how I screwed it up, since my points seem to match, 19 teeth and all that, but I have. I'm sure the VE has an impact on the fatness of the AF too. Not enough air. Looking forward to having it straightened out.

Meh, Thumpetto007 was right. With the cam marks aligned, the crank mark was 5 degrees advanced. Apparently I hadn't tension belt from the crank to exhaust cam enough when setting. Fixed, now behaving as desired.
http://pic100.picturetrail.com/VOL48.../413948054.jpg