Can anyone guess how this happened?
 

This is from a 930 engine that has less than 2000 miles on a complete rebuild. Pistons are custom JE with 8.0:1CR, rings are custom moly stuffed in steel lined cylinders.

The car is an EFI conversion using electromotive. At the time I noticed this happened, I was sitting at a stop light in heavey traffic. I hooked up the laptop and found all parameters within limits.

This is cylinder #4. When I performed the leakdown, #4 cylinder held 35% and #5 held 70% with hissing through the crankcase. Upon inspection of #5 on teardown, I found nothing remarkable. Both cylinders show no wear.

Ideas?

Thanks

http://forums.pelicanparts.com/uploa...1144631260.jpg

Are you sure you're only running @ 8.0:1?? Looks like low octane fuel in a high compression condition. Not absolutely sure, just my 2 cents.

Well, I am as sure as I can be about the CR. I've had a top notch Porsche machine shop do all the measuring from the spigots up to the heads, including head volume. The deck height is .069 because the heads had work by the PO. As for the fuel, who knows. I always put 91 octane pump gas in it.

Clear case of pre-ignition/detenation. All the standard possible reasons: Compression ratio, spark advance, lean condition, low octane fuel, too much boost.

If it's not a daily driver you might just want to mix race gas in.

-Andy

The only time I go lean is on decel using the TEC3r. I have the decel turned on where the inj shut off below a certain kPa, then add 6% as they come on at 2000rpm.

What the max boost pressure, and the injector duty cycle at max hp? Looks familiar, kinda like this post from awhile back, though not nearly as bad.

http://forums.pelicanparts.com/uploa...1144638511.jpg

From a subaru. Wastegate malfunctioned, I think.

OK - I'm convinced. Thx

Don:

Andy is spot-on.

Further,.....069 deck height is too much and makes any 911 engine much more detonation prone due to residual end-gasses left at the squish band. In short, not enough squish to push the A/F mix toward the plugs. Your piston crown looks VERY lean.

You need:

More octane than 91 (thats just not enough for these engines and your 8:1 CR at .8 to 1 bar really pushes the envelope))

.040 Deck height (understanding that this might be easy to do)

Richer AFR's for your running conditions

No more than 24-25 deg total timing from 4K on up.

Don, looks to be probably a tuning issue rather than something mechanical. You said that you only ran lean on decel. But how rich did you run under boost? This is more important. Also, how much ignition timing.

Plugs that are too hot are could also be a culprit but I would suspect tuning as the highest probability.

Hi Steve

I've had time to review this with JB Racing as well as the BIN file in the TEC3. For some reason, this cylinder produced a hot spot, as the AFR's below 80 kPa are 14 at cruise, 80- 110 kpa is 13 and above that is low 12's. In addition, if you stick the throttle, you will immed get 11 for 1 second, then low 12's from there to red line.

Timing above 100 kpa is 24, tapering off to 15 at 208 kpa.

The deck hieght is an area of heated (no pun intended) discussion. JB tells me that, since the heads had work prior to me owning the car (heads were shaved to produce 7.5CR), we need a deck height of 069 to make the math work out for proper deck height and CR. Since they have had the engine and build great endurance cars, I always take their side. However, since its my motor, I can do whatever....

As you know, I tune my own car. I've used a dyno-dynamics dyno that loads the engine and I can dial in any AFR under controlled conditions. The only time AFR's get above 14.5 is on decel. On hard decel, I have the TEC3 shut off the injectors until 2000 rpm, then I hit the cylinders with 6% additional fuel for a good transition and wash the cylinder walls.

Comments?

What spark plugs did you use? Did you ever hear it knock?

I use NGK R5672A9 plugs. There was no warning - it happened as I pulled away from a stop light in traffic.

How do the other pistons and spark plugs look?
If they do not show any indications of going lean then this could possibly be the result of a single bad injector causing only one cylinder to go lean.
Either way I would suggest having all of them tested prior to re-installation.

The damage was definitely done under a high load situation, so it was well before you pulled away from the stop light. Really looks like a classic detonation failure. Can you provide pics of the other cylinders?

Your plug, with a NGK heat range of '9', is proably hotter than a Bosch '3', so that could have also contributed.

Don

Sent you an email.

Every piston has been "frosted" - what high temps do to aluminum - exactly where the failed piston had damage. So it seems that the entire engine ran lean at some time rather than just one cylinder.

I am considering a different CR - 7.5 rather than 8.0.

What wideband did you use? Did you monitor EGT? It seems like more of a tuning problem than anything. Going from 8.0 to 7.5 shouldn't make a huge difference--the engine will fail in the same manner if you don't change anything else.

I use the UEGO wideband system, coupled to the TEC3r. I do not monitor EGT. I agree on the tuning problem - I tuned it.

Sorry to hear about that.

What PCs are you considering? Are you going to install EGT sensors?

Good luck.

Regards,