ECU Issues (Started as a high idle issue)

[Muskrat]

So I finally got the car back together and running again, now I just need to figure out it's original issue /ubbthreads/images/graemlins/smile.gif

I have different issues depending on which ECU I'm using:

1) Stock ECU:

Caps were replaced by Terry Posten a few years ago. He also had to repair a couple traces.

I started the car and it was idling really high, about 2000 rpm. I hooked up the datalogger (MMCd) and saw the TPS was at about 5%. I went to adjust it and found out it was already fully adjusted. Luckily I had another laying around, I swapped this in and saw an immediate improvement in the idle, around 12-1300 RPM.

After adjusting the closed throttle position switch, and playing with base idle here's where I've ended up:

Idle RPM: 1250 - fairly constant
TPS: 8.5% - constant
TIMA: 8* - fluctuates +/- 2
ISC: 10 - fluctuates +/- 2

When setting the base idle, and grounding the ignition timing plug the ISC jumps to about 60 and RPM's jump to about 2000. Adjusting the BISS all the way in results in the idle dropping to ~1160 but the ISC remains at 60-70 once the pin in ungrounded.

Also, with the key in the run position (engine not running) the ISC is clicking away like mad. Not sure if this is normal, but I've seen it mentioned in a few posts I've read.

2) MD 166260 ECU - Non EPROM - From a FWD turbo 1G DSM (Federal MT, Premium Line). 6/90 - 6/92 according to VFAQ

Rebuilt in 2002 I believe.

Ok. So as I was typing the above, I thought to try this other ECU I had laying around. Low and behold, after tweaking the base idle, it run's steady at about 800 RPM! However, running with the theme of this car, there are still some issues. Here's what I pulled from MMCd:

RPM: 812 - Fluctuates slightly
TPS: 0.0% - not reading
TIMA: -1* - Fluctuates slightly
ISC: 12 - Fluctuates slightly

CEL comes on, codes 14 (TPS) and 25 (barometric sensor)


The Questions:

Now I'm wondering, could this ECU be from a '90, with pins 14 and 6 switched and causing the CEL's (and no TPS reading)? Number printed on the inside is: 6582M

The second question is: Any idea what's wrong with the stock ECU? And is it fixable? I'd like to have a working EPROM in the future.

Thanks in advance for your replies and for reading this. I know it was kinda long.

 

[toybreaker]

One trick to see if the iac and ecu are playing nice with each other is to load the load the system at idle and watch the iac count.

Turning the steering back and forth gently will cause the ecu to request about five to nine extra iac counts while the pump is under load.

Cycling the a/c will also cause an increase in the iac count.

The ecu number you have listed as being a 90 is actually the right one for your car, as far as the pin swaps go. Even though it's production date may be from 1990, it was actually used in the 91 to 92 model year fwd turbo dsm's. Other than the fuel trim being a little off, it should run your car just fine.

Sometimes extra parts can be a blessing and a curse... Do you have access to another car that you can put your ecu's in to verify they are operating properly?

I would test the iac coils as per the vfaq, before going much further. The original, factory installed units are beginning to reach the end of their service lives. All kinds of problems can begin to manifest themselves with high mileage units.

It would be a shame to blow one of the drivers off the board with an iac coil that was out of spec.

Good luck troubleshooting this issue, and please report back on your progress.

 

[Muskrat]

Unfortunately I don't have a spare car to test things on.

So that rules out the first theory for the differences.

So any idea's why the TPS reads on the Stock, but not the 1G? and Why the 1G is throwing codes? (well, code 14 is obvious...)

I'll check the ISC and report back.

Thanks.

 

[toybreaker]

Quote:
So any idea's why the TPS reads on the Stock, but not the 1G? and Why the 1G is throwing codes? (well, code 14 is obvious...)



It's possible one of the traces that runs under the caps was damaged in it's previous life, prior to the caps being replaced.

Can you take a close up picture of the board?

Terry, Steve, or Jeff might be able to tell the tale if they can get a pic of the board in that area.

They deal with ecu's all day long, and they are pretty sharp when it comes to these kinds of issues.

 

[Muskrat]

Ok, I checked the ISC following the VFAQ instructions. The resistance between the coils was about 32 Ohms, and using the 6V battery the motor had a audible "tick". So I think it's ok.

Click on the pictures for full resolution:

Stock GVR4 ECU


1G Turbo FWD ECU

 

[Terry Posten]

There is a "feed-through" hole directly under the 47uf cap (closest to the heat sink).

Remove the board from the case, use a ohm meter (set to continuity) and place the probes at both ends of the runs that the feed through "feeds".

It is a little hard to find the run to the SIP so take a pic of the underside of the board (just like the one you just took) and post it up, Iwill circle the two spots and resubmit.

 

[Muskrat]

Stock GVR4


1G FWD Turbo


I think I see where you mean (The repaired trace on the galant board). And if that's the case both boards have continuity between those pins.

Thanks

 

[Terry Posten]

Yes, I did that wire. And that was what i was talking about.

About 75% of the boards have that feed-through eaten.

 

[Muskrat]

Update:

I plugged the stock ECU back in the car and ran it. Still doing the same thing, but now it's also throwing codes 14 and 25.

I ran the harness diagnostic for both the barometric sensor and the TPS, using the 1G FWD Turbo ECU. Both failed for power supply voltage (registering about 30 mV on the baro and 400 mV on the TPS). Both are supplied by pin 23 on the ECU. I checked continuity for the wires leading from pin 23 to the plugs and both were good.

I also tested the ground circuit for continuity for both plugs. Both failed (no buzzer) registering about 200 Ohms. I checked continuity along the wires to the sensor ground pins (17 and 24) and it was good. I also checked continuity across pins 17 and 24 on the ECU, and it was good. (Strike this, ignition was in the ON position). With ignition OFF (as test should be conducted, I believe) there is continuity on the ground circuit.

So it seems the issue is with supply voltage from pin 23.

I check the voltage from pin 23 to ground (nothing else connected) and it IS putting out 5 V. So my guess is I have a short to ground somewhere along that wire.

Well, I've solved my sensor issues. Turns out the pin wasn't making contact. Now I just need to readjust the TPS and set the idle and hopefully the car will be streetable again!

Thanks.

 

[Terry Posten]

Good news. Great job tracing that down.

 

[Muskrat]

Yeah, that was a fun one. Still don't know whats wrong with the original ECU, but at least it's running again.