I have determined that my TPI motor's ECM Isn't resetting to zero after I turn the car off, causing a stall and searching idle on cold start. Does anyone have an idea what can lead to this other than a broken ECM? The only thing I can think of is that my ignition lock cylinder is missing the pin that causes the car to recognize when the key is in the ignition while the cr is shut off, but that's a pain to get to and replace.

 

Not cold start enrichment to lean ?

 

Unfortunately the tuning is present for it but it is inactive because the ECM isn't returning to zero engine runtime.

If I disconnect the battery and drain the power to manually reset it to zero, it works fine and does cold start without issue.

 

Is this the '86 in your signature? 1227165 ECM or has it been changed to a later S/D for the 357?

 

If it is the 1227165, connector terminal A6 on the ECM (pink/black wire) is the signal for ignition power. If that 12V+ is interrupted when the ignition switch is turned to the "OFF" position, it should work.

You can also check the voltage at terminal C in the ALDL connector to be sure it is only on with the ignition switch in the "RUN" position.

 

It is the factory ECM for the 86. Are there wiring diagrams for this that I can use to test it with a DMM? And what process would I use to do so?

 

 

I just did the test and it seems as though the ignition signal is oscillating with the key on and getting constant zero with the key off.

 

Just thought I throw a few more details.
After reviewing the @themagicman data log, the reason it's causing the engine to stall is because, the ecm skips the Start-up Open Loop Commanded AFR and goes directly to OL Commanded AFR (14.03 instead of 10.9). That makes the real AFR too lean when engine is cold. It skips because engine runtime does not get reset to zero like he explained.

 

I asked about checking voltage at the "C" terminal of the ALDL since it would be an easy way to test for power at the ECM through the GAGE fuse circuit. That should also shut off when the ignition switch is in the "OFF" position.

 

Measuring across the C terminal of the ALDL port, the measurement drop to 0.1mV when the key is out and up to 12V when the ignition is on. I assume I need to replace the ECM now.

The question is, where do I go from here? With a larger engine than factory with a cam, injectors, and heads, is there a better option out there that I can do in a few hours? I daily drive the car so I couldn't do a full rewire, but a more modern ECM would be nice. I see that Haltech makes one that's pretty much plug and play, but it's also extremely expensive compared to a factory type replacement. Is there a middle ground from a company not so well-known that may work better for me? For a daily driven car I'd like it to be as smooth and easy of a transition as possible, as I don't have access another vehicle if this one is down for some time.

 

As you probably recall from seeing these (a long time ago) the '86 had the stand-alone time/thermal control in the intake water jacket to operate the cold start injector, and that was dropped in about 1988. The enrichment tables on the old CSI version were not quite as skewed as some of the later versions for cold-start and warmup enrichment using just the service injectors. We probably should verify whether the CSI system is still in place and functioning on the vehicle @TheMagikMan is working on.

Ironically, some of the last PROMs I burned for my '86 had the internal cold start tables active as well as the CSI still being in place. It's a bit of a "belt and suspenders" approach, but it actually works surprisingly well in my example. As you also probably remember, the added start fuel from the stand-alone cold start control goes away after a couple of seconds, so there still needs to be an injector pulse adder skewed by IAT and CTS.

 

It is in place, but not programmed or plugged in as I'm using the $6E definition that gets rid of it.

 

0.1 mV (0.0001V) is effectively nothing. Based upon that, I would assume that there is no back-feeding of any of the ECM electronics from external sources. Most of the TTL logic circuits in these devices have a minimum "on" signal threshold of over 2.0V, and a maximum "off" signal of 650mV (0.65V) or below. Any of the CMOS gates in the ECM would have a maximum "off" level of 1.5V or below. Your measured 0.0001V is safely below either of those. At least you are narrowing down the possibilities.

The 1227165 ECM originally supplied with the vehicle is basically the "DOS" version of ECMs compared to the more current "UNIX" version PCMs used on vehicles. That doesn't mean that they can't be made to work well, but do have limitations. They are just about fast enough to manage what you have successfully compared to the even earlier version ECMs. With a modified installation it will be very critical to have an accurate and functioning MAF sensor. It also means that many of the original constants, adders, and lookup tables from an original tune are not going to be optimal for your modifications. But you probably already understood that.

The issue that you have identified seems to point to a hardware problem within the ECM. A stock replacement unit with your MEMCAL installed might solve it - Maybe.

It is also NOT against the rules to pull out the existing ECM, split the aluminum covers, and inspect/clean the board for obvious signs of damage, contamination, failed components, and solder failures. Contamination is actually a potential problem, so cleaning can actually make a difference.




One of my hobbies is vintage electronics, as a diversion from my normal "day job" and to keep the brain cells rubbing together. I encounter this frequently, and many people are aghast when they find out a $3,000 vintage amplifier can be dumped into the utility tub and cleaned with soap and hot water to "fix" lots of random problems.




If you're not into taking a shower with your favorite ECM, it's at least worth a look for anything obvious. If that bears no results, a replacement ECM might be a consideration, even if you could just "borrow" a known working one for testing.

 

I've attached images of the board, I was trained as a technician at the vocational high school I went to, and I can't find anything immediately wrong. Maybe you'll be able to take a better look at it.

Backside

Frontside

Closeup of a cap that I think could be bad, but just appears to have a glob of coating on it.

 

It looks fairly clean. The conformal coating applied at the factory to prevent corrosion and dust accumulation seems to be intact, even if it's a little heavy in areas. If you can't see any cracked solder joints, signs of heat damage, and shorting, I certainly won't be able to see it with a photo. The surface mount components can be about impossible to check without really good magnification, and even then poor solder connections can be impossible to find.

It may be worthwhile to find a "loaner" to try out.

 

Yeah at this juncture, the car still starts and runs fine with a little help, so I think I'll save up for a MAF conversion and a Haltech Sport Platinum GM ECM so I have more resolution and ability to run a wideband and flex fuel in the future. If it's a daily, might as well modernize it when a part has to be replaced, right? 😂