good point. its probably the denonation in a cylinder or two that you cant hear where are running either near stoich or just happen to have the temps and pressures to ignite the intake/fuel charge. we all are so trusting of the dyno sensors for A/F and, even more importantly, trusting of the averaging of the sensor for all 8 cylinders out the tail pipe or header! I dont think having individual EGT sensors is that bad of an idea for the boosted crowd.

Mk

 

Boosted cars eat head gaskets on a much more regular basis than large displacement engines. The edge between max power and knock is much finer. When knock occurs, in cylinder pressure goes through the roof, that along with the accompanying temperature melts whatever is the weakest link.

Most of the boosted 928s probably aren't running stand alone controllers, so the tendancy towards knock under load is increased due to not fully optimized control. Thus, more wasted head gaskets. This is mostly related to running too much (or sometimes zero) ignition advance rather than not enough fuel.

 

Because I have built about 10 different engine configurations on my 944, I have gone back and forth on this.
My final decision is that BMEP (WITH PROPER COMBUSTION), will not lift/push headgaskets.
Knock does it.
My problem was that there are so many potential causes of knock, you have to prevent ALL of them.
The mls gasket on my car is viton coated and has been installed 3 times.
It is 100% leak free.
I run 20 psi on 93 pump (even in 5th gear to 7200 rpm)
This is with a turbo larger than probably ANYONE runs on a 944.
My car is extremely modified, but the block/head union is stock, so its relevant here.
On c16 I will run over 30 psi at Lebanon Valley dragway (7/11) and on the dyno it will be between 600 and 650 at the wheels.

I will say that the sneakiest thing I had to track down was reversion - exhaust gas will light you off early, I dont care if you have 1000000 octane fuel. Obviously this is turbo only.
Those wth stock cams dont have to worry about this one - there is no overlap.
2 valve heads are particularly prone.

 

+111

 

What company made it, who sells them, what bore sizes and thicknesses are available?

 

Mark brings up a good point regarding the averaging of the 8 cylinders into 1 A/F reading, because that's exactly what it is, an average reading of 8 cylinders.

Because of the differing runner lengths of the stock S4 intake manifold which was done to broaden the torque curve for a street car, at any given RPM, every cylinder will have a different A/F ratio than any other cylinder; there will be varying degrees of leanness and richness.

Looking at high RPM, all else being equal, the cylinder with the shortest intake runner will fill it's cylinder with the most air. The problem is, since the stock fuel computer doles out the same amount of fuel to each cylinder, that short-runner cylinder at high RPM will be the leanest. Likewise, the longest runner at high RPM will be the richest, etc.

For a stock car this is not a bad thing since one assumes that the fuel maps were created by the factory to compensate/compromise. However, for a modified engine, whether stroked or blown, this issue is exacerbated when using the stock fuel maps since the volumetric efficiency of the engine has radically changed.

How does one get around this conundrum using the stock intake?
This is a good idea, but the problem with EGT sensors is that their response time is very slow. A better solution would be wideband O2 sensors on every cylinder whose reponse is nearly instantaneous. Better yet, one can have the EGT's in conjuction with the WB O2's.

But that alone does nothing but allow one to simply observe what's going on with the engine. Using those individual WB O2 sensors with an aftermarket fuel computer would allow precise fueling for each cylinder individually, regardless of what RPM each cylinder reaches is peak volumetric efficiency.

This would be more practical for the centrifugal crowd since the stock intake manifold scales well with more boost. On the other hand, the stock intake does not scale well at all for the stroker motors, and independent throttle bodies appear the hot ticket.

Just like anything, all one needs is time and money!

 

You can use the SharkTuner on >MY87 cars to find which cylinders start to knock first..... no additional hardware required.

You can at least then remap the EZK to run safe.