Picked this up off the DSM tuners forum in the States! John A will appreciate this one! ( This guy has years of experience in turbo tuning Jap engines and some whipper snappers were trying to tell him his business, the Tefal head puts em right )
I'm going to go a little more in depth, since some people are still having a hard time grasping this concept, despite the fact that it has been discussed to death on various forums for at least the last 3 years that I can recall.
People think that high EGTs equal high combustion temps, which would indicate a lean mixture. So when they see high temps, the typically add fuel. And when it is low they lean out. This method of tuning is flawed for various reasons, and I will touch on all of them.
Myth number one: The internal engine parts will melt at over 900 degrees C or 1650 F. Here is some big news. Combustion temps are over 2000 degrees and can reach 3 to 4 thousand. So why dont the parts melt? Just a few quick reasons. One is heat sink capacity. Two is heat transfer from the piston to the cylinder wall through the rings. Three is heat transfer to the oil that is splashed up or squirted on the piston and other parts. Four is heat transfer into the cooling system through the head or block. In fact, so much heat is taken out through these methods that by the time the combustion gasses get to the manifold and EGT probe, temp is down to 900 c or 1650 f ;o) Some heat is also lost in the conversion of heat energy to mechanical energy in the form of gas expansion. So that myth goes down the terlet.
Myth number two. If EGT is high the mixture is too lean. Well what happens when the mixture is (way) too rich? It is still burning as the exhaust valve opens. Fire in the manifold WILL heat up the probe. Not rocket science. I have seen a dozen people add fuel and add fuel and EGT temps dont go down. Lean out to where they are supposed to be though and its all normal again. I know it goes against the "Internet Laws of EGT Guages," but its the truth. Also, what happens when the mixture is WAY too lean? Temps go down. Why? Less fuel to support combustion. It is entirely possible to be so lean that combustion temps drop so low that there isnt enough heat energy to increase pressure enough to cause knock. I have seen this happen on my car after a baro wire mod gone wrong.
Myth number three. If EGT is too low, the mixture is too rich. Not necessarily, though this is way more possible than vice versa. It depends entirely on the setup. The more knock prone you are the more fuel you need. That slight difference in EGT temp will show itself if and only if you are getting maximum timing advance for your setup.
So what does the EGT tell you? I've said it a thousand times before, as have hundreds of other DSMers, but ponder this. You are knocking. For ANY reason, not just lean mixture. The knock sensor picks this up and reports to the ECU. Now this may be hard for some of you to believe apparently, but when the ECU senses knock, it pulls timing. The amount of timing it pulls will vary based on the "normal" knock sensor noise threshold, how strong the signal is, and how long it has been going on, among other factors. The ECU comes up with a kncok retard value, and that amount of timing is pulled back from where the full timing would be based on the timing map you are currently on. Here is what I mean by this. The timing map you are currently on is based on airflow mostly, but also coolant temp and intake temp and other factors. This map or table includes timing values for each rpm point. So lets say for where your car curently is, you are on a map that says at this rpm maximum advance would be 17 degrees. But knock retard is 3 degrees. What will your timing be? Get out the calculators. 14 degrees is what you will get.
Now what does retarded timing have to do with this... If the timing is retarded enough, as in LESS ADVANCE, the ignition event occurs LATER in the combustion cycle, in degrees of crank rotation. If the mixture is ignited late enough, it is still burning as the exhaust valve opens. Refer to "Myth number two." Fire in the hole will heat up the probe. THIS IS WHY high EGT temps mean that you are getting knock and reduced timing advance. Read it a few times. Let it soak in.
Furthermore, the amount of change in EGT temps per change in air fuel ratio is MINOR, while conversely, the change in exhaust gas temps is SUBSTANTIAL when the timing is retarded and there is fire on the probe. Why? The increased heat caused by a leaner mixture is mostly absorbed by the motor or converted to additional mechanical engery. The latter is the cause for the increase in power when you can lean out without knocking (<- key phrase, pay attention). By the time the combustion gasses get to the probe, the difference is now slight. In contrast, the difference in temps between normal exhaust gasses and fire is very significant and it show up easily on the guage. Now it is important to remember that the higher combustion temperatures caused by a lean mixture not only increase mechanical output, but increase a motor's (or cylinder's) tendency to knock. And again, this knock brings on timing retard, which brings on fire in the hole, which brings on a high EGT temp reported on the guage. THIS IS WHY poeple confuse high EGT temps caused by mixture with high EGT temps caused by knock. Because both usually happen at the same time when fuel tuning. You can also have one without the other, as is the case with mechanical knock or phantom knock.
So now that we got that straight...
Other reasons why EGT is not a good way to tune, without a logger. The ideal temperature will vary with EVERY car and EVERY setup and EVERY set of conditions. I do upwards of 300 drag passes a year and have clearly observed and datalogged this. With that established, the only way to find the ideal EGT temp for your setup/car/conditions/gear is to datalog to see what your timing is. At the point you develop knock you can get an idea of where you want EGT. The problem is, as I stated above, when you have full timing the difference in temperature effected by changes in air fuel ratio is very small. This is complicated even more by my next point:
EGT guages are not very accurate, nor are they precise. Yet another reason to need a datalogger to verify and validate what you are seeing on the EGT guage... The probes are slow to respond, and I will explain why. The rate at which an object changes temperature varies directly as the the difference between the temperature of the object and the temperature of its surroundings. In this case the object is the probe, and the surroundings are the exhaust gasses. What this tells us is that the greater the difference in the temp of the probe and the temp of the gasses the greater the RATE OF CHANGE of temperature. So as the temp of the probe rises (or falls) to match the temp of the exhaust gasses, the difference between the two temps is decreasing. This slows the rate of change. And the cycle continues... So as you can see this is not a linear function, its a curve. This is why poeple often say that the guage reacts slow, or to estimate how fast the needle is moving to estimate where it would really be if it was faster, etc. This is a fundamental problem for both accuracy (especially) and precision (not so much). Another problem is that in general they tend to fail on the low side, meaning they will read colder than what the case really is. So you'll lean out more and more. What happens then is you seem knock "easier" and the tuning window (based on EGT temps) becomes very small and touchy. I have been there, and done that. This is not the way I would choose to tune my car But if so many people insist this is the way to go, what can I say.
This is not rocket science. This is not new information. It has been documented, measured, qualified, quantified, and posted about for as long as I have been into DSMs. But not enough apparently. I am going to copy this into a file on my PC, so I can paste it into the next 500 posts I read about this. If I have to type all of this again I'm gonna snap...
The link to the whole thread can be supplied!