Interesting ECU experiment

[1qckser]

This is from one of the other forums im on , Q45TECH I hope you have some feedback on this dangerous experiment.

http://www.sr20forum.com/showt...ystal

[Q45tech]

Experimenting is good as long as you have a vague idea what you are doing and don't moan and whine about the potential to destroy something. You can always buy replacement ecu.

The problem with changing the 8 MHz clock on the microprocessor is every cycle gets faster/slower by the amount of the change. The ecu thinks the engine is turning 10% slower than it really is.

The CAS provides 360 [on a V8] pulses per cam revolution times two is a rpm.........468,000 pulses per minute is 650 rpm...... the ecu clock creates periods of action - data samples at 0.00001 seconds.............100,000 adjustments per second to arrive at a 100th of a millisecond injection resolution............the Consult on shows the AVERAGE [10.1 or 10.2 milliseconds] yet it is calculated to 10.xx milliseconds...........greater than a 1% accuracy.

10% changes are pretty big, he should have acquired crystals in 1% increments and zeroed in on an optimum for his individual engine.

Injector timing is so many milliseconds but when you speed the clock up........spark and injection duration decrease by the amount of the speed up...........10% leaner might be ok at WOT but not at cruise.