More AFR idle/part throttle cruise issues HELP!

[gixxer750]

k. My car runs right sometimes, but other times is runs like shit.

No codes now.

Heres what happens. Car is idling great between 14.5/15.0, but then it slowly drifts leaner until about 17.9 or so, maybe leaner. This is completly random, and has no set amount of time that it lasts. Does this part throttle cruise and idle. It's ok on over 25% accel...

Watching on my emanage with the real time display, I get a constat signal from the AFM around 28-30 HZ at idle. Idle remains the same, except for a lean miss that starts up when it does this. It takes 40-50 % correction factors in the emanage just to get it back to the 15's.

Then, all the sudden it's fixed.

Only thing I've noted is that when this happens, inj duty cycle goes to zero....

It stays about 1% duty cycle when it runs fine.

I dunno... I'm lost. Tried AFM's and 02 sensors.

 

[Nick M]

I have no experience with an E-man, or other piggy back. But if you are going lean like that, often the ECU is trying to correct what it sees as a rich condition.

Duty cycle of zero is fuel cut. Does the E-man read vehicle speed?

 

[gixxer750]

no speed... just RPm. But how can the car run with 0 duty cycle @ idle?

 

[gixxer750]

and what causes it?

 

[Nick M]

Fuel cut happens when you are decelerating and the TPS is at idle. When you are in fuel cut, the engine isn't running. But when you open the throttle or reach a certain rpm, the fuel is turned back on.

It sets a 31 when the voltage is under 100 Mv. I do'nt know if a Hz reading triggers it against IDL opening. Perhaps you have a bad connection since you have traded out meters already.

And just to make sure, see if you have continuity from E2 and IDL when your foot is on the pedal.

edit:
http://www.cygnusx1.net/Supra/Library/TSRM/MK3/manual.aspx?Section=FI&Page=51

The IDL is yellow with a blue tracer, and the E2 is always brown. I didn't mention when you do a continuity check, or anything on the Ohm scale, the power needs to be off. The voltage check is better than Ohms.

 

[gixxer750]

Well, I did a bunch of checking. I monitered Vf while the car was going from weird to normal, and got 2.25 volts around where the car was running normal, 0 volts when the car ran lean, and around 4 volts on the occasional rich bounce...

But all in all, I found the problem. Pulled the cover off the ECU just for the hell of it, and found a smoked resistor and a transistor right close to the plug end of the ECU....

I tried to pull apart an old N/A ecu, desolder the needed components out, and solder them to my board... but being what seemed to be a multilayer board- it didn't work.

Now the car won't run, no tach signal and no fuel pump kick on...lol

Guess I'm in the market for a yellow plug MT ECU