So I'm driving home from work and the SES comes on, the engine starts running like crap. I pull up to my first stoplight and the rpms are surging up and down. A couple of weeks ago I had the same sort of problem and it gave me a code 42(problem with the EST circuit). I replaced the ICM and everything has been fine up to today. I was thinking that it was another ICM gone bad, but when I scanned for codes I got 7 of them! wtf ? Codes that the ECM gave me are: 13, 14, 22, 24, 33, 34 and 42.

So I disconnect the battery to clear the codes and re-scan. No error codes this time. Start the engine and it runs fine. I'm fairly certain I know what the codes are, but what are the odds of that many codes coming up at one time? My logic is leaning towards a bad ECM. I mean for the 02(code 13), CTS(code 14), TPS(code 22), VSS(code 24), MAP(code 33,34) and EST(code 42) to go wrong all at the same time is very strange.

Another thing that keeps running across my mind is what if its just one bad sensor that throwing off the other sensors? What are the chances of that happening?

Grr.. why couldnt this happen on a weekend and not a workday!

Any and all help is greatly appreciated. Thank you.

 

not that odd to have several or a lot of codes stored if you have a problem.

 

But all at one time? I cleared the codes earlier in the week when I disconnected my battery to install my stereo. So ll of these codes came up as I was driving home yesterday...

 

Debris,

One bad sensor can induce several DTCs at once, since the ECM relies on the data from some critical sensors to determine the acceptable input signal ranges of other sensors.

However, you may also have a wiring issue. One bad connection can affect the reference voltage to several soensors at once, creating unacceptable input signals from all of them. Grounds are critically important at the engine and ECM. 5VDC reference voltage circuits are often shared by several sensors, and a poor connection there can affect all the sensor on that circuit as well.

 

Ahh a person who knows what flotsam is hehe.

Thanks for the reply Vader, I always look forward to your knowledgable input.

So lets assume its not wiring and say its a sensor. Which ones would you replace first? Also do you have any good methods of testing the ground wires out?

Much appreciated, thank you.

 

What kind of car do you have?

 

I've seen several cars with LOTS of random codes before. 99.9% of the time the ECM was bad. I think GM says to try replacing the PROM first, but I never do (simply because PCM's are usually in stock and PROM's are NEVER in stock).

Try a PCM and if nothing changes, just return it.

 

That may not be a bad idea, but before you do it, spend a few minutes checking out your old ECM so you don't waste your time and money.

Presuming you have a TBI engine (by virtue of the MAP sensor), it appears that there are several common links in your system:

1. The CTS, MAT, and TPS share a common ground connection wire/circuit (A11). The VSS input coincidedntally happens to be on pin A10.

2. The MAP and TPS share a common 5VDC+ reference circuit (C14);

3. And while many of the passive sensors don't have a 5VDC+ reference feed to them, the ECM still uses a voltage divider and sinking input to develop an appropriate signal from them.

The common thread here may be some of the connections at the ECM, but more likely the 5VDC supply from the ECM, or the ECM grounding, since ALL signals are referenced to ECM ground.

Grab your DMM and test the reference voltage on the gray wire at either the TPS or MAP sensor. Use black wire at the TPS or the purple wire at the MAP as your ground reference. If you don't get 5.0 VDC there, you have a problem with the wiring or ECM, not a bunch of sensors. If you discover that the voltage is low, meter again but use the battery negative terminal as your ground reference. If you get a different voltage, , you have an ECM or chassis ground problem. If the voltage is still low, you need to clean and re-seat the ECM connectors and try it again.

 

Carbed cars have MAP sensors too...but you're right because they don't have a code 33. Or maybe that's what you meant?

 

Make sure the connectors at the ECM are fully seated....you wouldn't believe the problems this can cause.

 

Thanks everybody, today I'm going to troubleshoot more per Vaders instructions first. Thank you Vader.

BTW, is a PCM the same thing as a ECM? The General wants 220 bucks for a ECM while Kragens Auto Parts wants 105 bucks for a part refurbed by A1-Cardone.

Yesterday I removed my ECM and the connectors and wiring looked good. No crusty stuff and clean. I'm the original owner and this would be the first time Ive taken it out. So I put it back in and I'm going to replace the MAP, TPS and 02 sensors since theyre 12 years old. Thanks again guys, more later.

 

Just an update..

I replaced the above sensors plus the CTS. I checked per Vaders instructions and got +5 volts at the TPS and MAP. So I think my wiring is okay, unless its some sort of intermittent wire.

I took it for a drive and seems okay. But I still dont have alot of faith in it just yet. It drove fine for 2 weeks after I replaced the ICM then threw the codes above out. So I think the next thing I'll replace is the ECM, EGR, ESC and perhaps the coil. Sound logical? Thanks again, appreciate any feedback.

 

Jetsam,

Citing several of your key points, I'm going to guess that the ECM (which is the same as a PCM in out cars) is working correctly:

1. You have the required 5.0 VDC at the appropriate places, which means the power supply in the ECM is working correctly. The 5VDC supply is used for both the logic and signal derivation;

2. The ECM is detecting and storing DTCs (error codes) which are appropriate to your car and engine;

3. The codes are cleared when the power to the ECM is interrupted, which means they are not being randomly generated, but are being detected only after some inputs are going awry;

4. The system worked fine until the ignition control went bad on you. There certainly isn't any infant mortality issue with the ECM.

And you just performed several steps at once. You replaced the MAP, TPS, and O² sensors (if I understood your text correctly). The MAP is not typically a "wear" item, and can last the life of the car. The TPS is similar, but can experience wear due to the constant physical movement of the potentiometer wiper across the carbon block. I trust you adjusted the TPS correctly when you installed it (pre-1990 cars). The O² is definitely a wear item, whose signal degrades slowly over time or with contamination, and should be replaced every 30-40,000 miles. If your's was 12 years old (1990 car?) it probably should have been replaced some time ago for the most accurate signal.

Now your ECM is going to take some time to re-learn the input ranges from the new sensors. It may take a week or more of normal driving to get the BLM tables completed enough so that the ECM can decide between and acceptable signal and a bad one. Don't be too surprised if you get a DTC after several days.

Additionally, you disconnected and reconnected the ECM edge connectors. This act alone can sometimes solve this kind of rogue problem, since the wiping action of the connectors during the course of disconnection and reconnection can be enough to clean the contacts and provide good connections. Re-seating control cards in NC and CNC controls has been an acceptable maintenance practice for 20+ years for the same reasons.

The only other questionable item would be system grounds. You should use a high current ohmmeter (like an analog Simpson, Triplett, etc. or high-current digital) to test the continuity and resistance of the ground paths from the ECM and sensors to the battery negative terminal. Any resistance over 50 milliohms (0.050) can cause signal level problems with some of the sensors. A common place for ground resistance is at the engine ground strap on the rear of the right cylinder head, but any ground connection is suspect.

 

Vader,

Man you rawk! Very logical and
makes complete sense. Thank you.

Like you mention, it seems as though my ECM is okay and not the culprit. It runs the code 12 diag check and does what its suppose to. My wiring and harnesses all look good, no krusty terminals or freyed wires. But like you also mention, my ground strap could be an issue.

Yesterday, while doing my visual check of the wires, I noticed that the ground strap behind the passenger side engine block is rusty. This is the strap that connects to the cylinder head, firewall and frame. When taking a closer look, my frame was damp from a tiny drip coming from the heater hoses which is right above the ground strap and ground strap bolt. The strap itself looks okay though rusty.

Could this be a possible cause of an intermittent ECM or sensor DTC? Also, while doing my check, I noticed that my positive battery terminal has a fusible link that goes to the terminal block on the firewall. This fusible link has some of its insulation missing in the first couple of inches after the battery connector. This makes me wonder if this fusible link is the one that the ECM gets its main power from.

About the testing you mention, I'll have to look at getting a high current ohmmeter. I think if I cant find one I'm just going to look at replacing the ground strap thats mentioned above.

Thanks again Vader, I truly appreciate your help.

 

The grounds are very important, loose grounds at the back of the heads can throw a bunch of random codes like that. Maybe try cleaning that connection, and make sure all the wires are hooked up back there. There's usually 2-3 terminals on the passenger side and one on the driver side.

 

Thanks madmax, the straps arent actually loose, but rusty throughout. It looks as though it would still provide adequate continuity as it isnt freyed, but I'm not certain. I didnt know of the other ground strap on the drivers side. I'll definately take a look at that one. Thanks madmax.