04-18-2010, 01:10 PM
Join Date: Jul 2005
Total Cats: 68
Hot restart hi-vacuum misfire
I'm running an AEM and recently installed a VVT motor in my 2000, and switched from 500cc RC injectors to 600cc Deatschewerks injectors. It is in parallel mode so factory ECU is controlling the idle air solenoid.
After some tuning the car runs well and fueling is pretty accurate (O2 corrections are small) when driving.
However I have a persistent problem which is that when the engine is hot I have a misfire at high vacuum (e.g. revving in neutral) which also seems to be the cause of idle surging. It shows up when the engine is good and warm, and is much worse after a hot restart - it seems to improve somewhat several minutes after a hot restart. The fueling around idle was tuned after an initial warmup when the problem doesn't show, because it's impossible to tune when the misfire shows up.
What seems to happen is that while idling, or while holding the throttle steady at say, 2500 RPM, the motor will periodically go lean and thus misfire even though injector PW has not changed. The RPM will drop and MAP will go up, which causes the inj PW to increase.
Attached is a datalog of the idle surging. O2 feedback is off so as to eliminate it as a cause of the oscillation. During this datalog my foot was off the gas.
Note that in the log of idle surging I show in the bottom window:
1) injector PW
2) injector batt offset (in case battery voltage was drooping and causing the lean spot)
3) the calculated value (inj_PW - batt_offset) / MAP. This value represents fuel delivered divided by MAP. It should remain fairly constant assuming the engine VE remains constant, and you can see in the logs that it is pretty flat thru the surging.
I *did* check if VVT was causing it:
- the problem is there whether or not the cam is locked at full retard, or if I have it set to start advancing at 1250 RPM, advancing linearly to 24* at 3200 RPM.
- I checked the VVT cam phase and it doesn't move around abnormally during the surging or misfire.
Later I'll get some logs of when the misfire happens while revving in neutral. When it does this after a warm restart, the car drives like *** from like 20" vac to 10" vac, until after a few minutes. It feels like the injectors are completely off sometimes in that vac range.
I keep thinking my batt offset table is wrong for the Deatschewerks injectors, but it runs great and AFR hits the target until after the engine is good and hot. I also was suspicious that my GM 2bar MAP sensor is flakey but the its output seems normal (see how it varies inversely with RPM)
It almost seems like, the fuel pressure is dropping causing the lean condition, but the car runs great under load.