Hi, first time long time...
I've got a 94 Firehawk (hope that isn't a dirty word on a Z28 board) with 27K miles. I purchased it with 15K miles and it has been doing this since purchase.

It doesn't seem to remember idle. When it is first started cold it idles around 1000 rpm and then climbs to about 1400 as it warms up. If I shut it off and restart it will idle lower. If I let it warm up completely, or drive to warm it up completely and shut it off and restart it will idle properly (about 650 rpm in drive, 700-750 rpm in park). Once it has relearned the proper idle it runs fine and can be restarted with the proper idle until it completely cools off again and then the cycle restarts.

Things I've done to try to fix this: replaced IAC, replaced intake manifold gasket, traced and replaced any questionable vacuum lines, OPTI was replaced at time of purchase because it failed, checked all the sensors using scan94/95 and everything seems to read ok.

The only thing I can think of at this point is that maybe it needs a new PCM? When on scan94/95 the IAC counts are mid 30's when it has relearned the idle. Prior to that the IAC counts are in the 100's, the RPM is higher than commanded but the counts don't change to try to command the IAC to close and change the idle.

Any input is appreciated. I hate to throw a computer at it without knowing if it is actually bad or not. It has never thrown any codes of any kind.
Thanks!

 

It’s probably not “relearning” when it warms up. There is a table in the PCM that elevates the target idle speed based on coolant temp. At very low temps cold start idle can be as high as 1,200 RPM. Fully warmed up idle for the A4 is 550 RPM in drive and 650 RPM in neutral. Not unusual for the factory tach to read 100 to 200 RPM higher than actual.

If you have or can run a Scan9495 log from cold start, I can take a look at it for you.

 

If you have or can run a Scan9495 log from cold start, I can take a look at it for you.[/QUOTE]

Thank you for the reply!

I will log it an post what I have.

I agree the factory tach is less than accurate. This morning's example of what happens; I drove 50 miles to work this morning and put it in park. The idle was still 1200-1400ish rpm (appears to be 900ish in drive). I shut the car off and restarted it and the idle went down to 650ish on the tach. When I go out to go home the idle will be back up at the 1200-1400 range and stay there no matter how warm the car gets until I shut it off and restart.

 

I would bet once the car goes into closed loop it's the 02 sensors not reading correctly or dead ... I am having the same problem with my 94 ... the only problem with 94's is that I use a lab top plus data cable and eehack from fbodytech to read whats going on with car because you can't use a regular scan tool for this year. once you shut car off it goes into open loop till the pcm read proper temp before switching to closed loop etc

 

Finally got a chance to do a couple of logs.
The first log is a cold start. The second log is a restart after its been idling for a few minute (right after the first log). The third log is actually one I did a couple of years ago that I forgot I had.
Thanks guys

 

Thanks for the tip siguy.

I think the o2's are reading properly though, at least according to the scan I did yesterday.

I can't actually point to anything except the computer not commanding the IAC to move (until after a restart) closed enough to lower the rpm. I'm an extreme novice though when it comes to these scans.

Edit: I did notice that at idle hot it seems to be bouncing between cel 16 and 18 in all the scans even in closed loop mode.

 

A quick look at the first log shows a problem with the TPS sensor, which may be the cause of idle problems. The TPS voltage is jumping up and down over a range of 0.49 volts to 0.59 volts. The PCM has baselined closed throttle at about 0.54 volts. When the voltage is below that, it shows the TP % as 0, and treats the idle control as if the engine was actually idling. But when the TPS volts are above that, it switches away from the "idle" fuel trim cell, and shows the throttle % as 1% or 2%. To the PCM, the engine is no longer "idling" because it appears you are moving the throttle. So it is no longer trying to pull the idle speed down to the target idle speed.

Just a quick first impression, but something you have to investigate. Why is the throttle position voltage unstable?

There are also repeated lines of corrupt data. That can often be caused by high voltage interference on the DLC data wire. But his corrupt data seems less than random. For example,. the left bank injector base pulse width error is almost always 500.00 milliseconds, which is totally bogus/impossible. May be an issue with the PCM.

Again - this is simply a reaction to a quick first glance, setting the file up so I can review it using Excel tools. But I would recommend you look for the cause of the erratic TPS voltage.

Actually, several of the sensor readings are jumping around over a small range, including coolant temp and inlet air temp. Could be a problem with PCM ground or voltage supply, preventing the PCM from regulating the reference voltage to the sensors.

 

Thanks for taking a look at it. I appreciate it.

I'll take a look at the tps and see what I find.

I also plan to finding what grounds I can and seeing if I can clean them up.

I have a string suspicion that the car was parked at a dealership with the front end over a water sprinkler for an extended period of time, or something like that. The 99% of the car is as clean as you'd expect a TX/CA car to be, but the harmonic balancer and front sway bar have rust like they were sprayed with water for extended periods of time (and the Opti died at 16K miles).

 

Couple other quick comments:

- when you first start the engine, the TPS volts are variable, but the PCM interprets all readings as 0% throttle. When the engine starts, the IAC is at 129. The target idle RPM is 900. The PCM immediately starts pulling the IAC counts down to control the engine at 900 RPM. When the RPM reaches 900 (=/- 25 RPM) the IAC is at 99. That indicates that without the confusion of the varying TPS volts, the PCM is fully able to control the idle RPM using the IAC valve, and the valve is responding to the PCM command. It loses that control when the TP % starts rising up above 0% in response to the varying TPS volts.

- your battery or charging system seems weak/problematic. The system voltage at the PCM drops to 5.1 volts at cranking.... that is low enough to start affecting the operation of the PCM. When the engine starts the battery eventually recovers to ~13 volts. But the voltage at the PCM is jumping up and down a few tenths of a volt. Makes me wonder if the voltage regulator in the alternator is leaking some AC voltage into the electrical system. This gets a bit outside my sphere of knowledge at this point.

- as far as checking 12 volt power supply and grounds, focus on the PCM. The pins for your 94 are the same as shown in Shoebox's PCM wiring diagrams:

12 volt feed to PCM:

http://shbox.com/1/1995_pcm3.jpg

grounds to PCM:

http://shbox.com/1/1995_pcm1.jpg

The two PCM grounds come from a wire attached to the engine block, passenger side, right above front of starter, and the stud that holds the ICM/coil bracket to the driver side head (with about 5 wires and a ground strap attached).

Still looking at the log. Unfortunately I have a problem staring at the screen for extended periods of time, because of an eye problem and repairs that I put off 15 months ago because of COVID.

 

Thanks for the links and the insight.

The battery is weak, but the problem existed even when the battery was new. I will have to replace it this weekend as part of the search for the repair just to make sure it doesn't add a wrench to any issue I may correct.

I'll work on the PCM grounds and see if there is any improvement. I see what you say about the numbers moving around quickly. Maybe my gut instinct was correct and there is an internal PCM issue.

Not sure if you noticed on the scan, but I did shut the car off and restart it once it went into closed loop mode and the idle came down quite a bit after. As you mentioned, the tps is still flashing 0 to 2% so it still moves between cel 16 and 18 even after the idle is "better" but still not as commanded.

 

Just an update for anyone who is following this.

Put a new PCM in it yesterday and knew the second it fired up it was fixed.

It now runs just like it is supposed to

Thanks for the assistance Injuneer!

 

All the data/sensor readings stopped jumping around? Must have been the reference voltage regulation in the PCM.

Thanks for the feedback. I wasn’t able to go deep into the multiple logs, because of some eye repairs I've put off for too long. Staring at a bright screen for an extended period of time is difficult. Appointment with the surgeon tomorrow, hopefully to get it scheduled.

 

All the data stopped jumping around, the exhaust doesn't smell rich anymore, the idle comes down to 550 in drive on it's own and it runs so much smother now. It's a completely different car now. I have owned a 02 LS1 since new, this is my first LT1 and the only one I have ever driven so I didn't have anything else to compare it to. The change in response and driveability is dramatic. It now acts as close to the LS1 does as I'd have expected it to.

Good luck with the surgery! Hope you have a quick recovery with excellent results.

 

Wow! Glad to hear that a different PCM fixed it. I have been battling the same range of issues on my 94 Z-28.

Throwing a PCM at it is on my list of things to try. Might have to move it to the top of my list. I had beat my head against the wall and got aggravated with it and stepped away from it.

 

Pioneer
I was really glad that fixed it also. I started checking grounds and each one I looked at looked as perfect as can be. After Injuneer said that one bank of fuel injectors was reading the same time for communication at all times I figured I'd bite the bullet and throw a computer at it. I wish I had done it 4 years ago when I bought the car! I suspect that is why someone got rid of it in the first place because it looked like all of the typical vacuum leak things had been touched as well as some sensors since the symptoms mirrored issues with those. I purchased the computer and got it in two days. It took all of 10 minutes to swap and was an instant night and day difference.