Many times not all graphics are stored by the WaybackMachine, so if there are graphics missing from the following VFAQ, please don't email me about them (unless you are emailing me to tell me you have a copy of the missing graphic you can send me).

If you know the original author of this VFAQ, and have their contact info, please have them email me if they have any missing info to add, have the VFAQ at another site I could not find, or wish to have it removed from my site.

The following post to the Talon Digest is in my opinion, a GREAT summary of much of the info listed in the other Theory links on this site. It is re-posted here with Dennis' permission.

Date: Wed, 21 May 1997 12:04:45 -0400
From: dg50@chrysler.com
Subject: Detonation
Message-ID: <#20>

(I've seen a number of posts in the archives on this subject, but I haven't seen a concise summary yet - so here's my attempt at one)

Detonation when the air/fuel mixture in the cylinder explodes, instead of burning rapidly. (To be really technical, detonation is when the flame front moving through the cylinder exceeds the speed of sound.)

This is bad for two main reasons:

1. The sudden shock load can break parts - the repeated shock loads of constant detonation is like running a jackhammer in your engine.
2. The temperature inside a detonating cylinder is much higher than normal, and can get hot enough to melt aluminum.

Symptoms of Detonation

1. A loud pinging or rattle/knocking noise. It sounds like a bunch of marbles inside a tin can. (This is why detonation is commonly called "knock")
2. Higher than normal exhaust gas temperatures
3. Little "flyspecks" of aluminum on spark plug center electrodes and the center insulator. This is the dead giveaway indicator for detonation. Examine your plugs with a magnifying glass. Little balls of aluminum clinging to the plugs means that you are detonating. (It's aluminum from the pistons being vaporized and condensing on the plugs)
4. Engine damage (holes in pistons, melted spark plug ground electrodes, etc.)

Causes of Detonation:

1. Low octane fuel
2. High static compression ratio
3. High boost levels
4. Excessive spark advance
5. Lean fuel mixture (may be insufficient fuel pump capacity/pressure, insufficient injector flow, air flow sensor out of range)
6. High inlet air temperature / low air density
7. Combustion chamber hotspots (Dirty combustion chamber or poor coolant circulation)

For a given fuel octane rating, an engine can only stand so much boost, spark advance, compression ratio, and mixture leanness before it begins to detonate. Therefore, as increasing boost, spark advance, compression, and leaning the mixture tend to increase power:

The onset of detonation is normally the limiting factor on the amount of power a given engine configuration can produce

Collorary: To produce maximum power, an engine should be running right on the limit of detonation

In a non-computer-controlled car, this is a very difficult thing to do. (Do a pass, pull the plugs, check for the flyspecks, change something, go faster, pull the plugs, check for the flyspecks etc. ad nauseum) However, the DSM CPU has a knock sensor that "listens" to the engine to detect knock. If the sensor detects knock, the ECU first cuts back the ignition advance, and if it still knocks, cuts the boost levels. Result: a drastic loss of power, and a safe engine. If you have somehow limited the ECU's ability to retard the spark (unlikely) or cut the boost (easy) then it is still possible to get into a destructive detonation scenario. It is still a good idea to pull the plugs every couple of runs and inspect them for signs of detonation. Keep an eye on the boost and EGT gauges!

Fixing a Detonation Problem:

1. Use higher octane fuel
2. Reduce the level of whatever it is you have increased (boost, ignition advance, etc.)
3. Reduce inlet air temp with a bigger intercooler
4. Confirm proper ECU/knock sensor operation.
5. Clean all carbon deposits from the combustion chamber
6. Ensure proper cooling system operation

Making more Power

OK, if I can easily adjust boost, spark advance, and mixture ratios, (and to a certain extent inlet air temp) what should I have more of to make the best power?

Excellent question. I don't know. I suspect that boost is worth more than spark advance, with mixture ratio somewhere in the middle. If I had a PMS, a datalogger, and a dyno, I think I'd set spark advance to about 2/3 of the possible range the computer provides, set the mixture a little on the lean side, and then run the boost up to the detonation point, and back off a little. Then I'd play with the mixture to see if I can get more power, or get the motor to stand more boost. Once I've got a good WOT boost level and mixture ratio, then we see how much more advance it can take. Maybe.

Note that we haven't taken inlet air temp or density into account here, nor have we allowed for real-world factors like: how fast can the wastegate react to boost spikes? This is just an educated guess. Bottom line? Lots of testing. Lots. And you can't get the inlet air temp too cold.

Hope this helps someone,
Dennis Grant
dg50@chrysler.com

Please send your suggestions and comments to Damian_Sigman@homemail.com
My Home Page

Page created June 10th, 1997 - Damian Sigman