Jump to content

More 88 strangeness...


KDirk

Recommended Posts

Thought I would post this here for input. As a follow up to my previous threads about ECM/MEM-CAL problems with my 88, I now have a new problem that has developed.

For some background, I had the ECM hanging out of the dash on the passenger floor pan while I was testing my new PROM, and during this time (a few days) everything worked great.

Since I tucked it back up behind the glove box Friday, I am getting intermittent E013 (open 02 sensor circuit) and E047 (loss of serial data comm) from the ECM. In the case of the latter, it only shows as a history code, as the ECM is off line when this occurs, so no ECM codes are shown while the problem is present. What is strange is that the engine continues to run well while the ECM is not on the network. So, I suspect this is not another mem-cal problem and rather is related to CKT 800 (serial data line) wiring. This seems to happen mostly when the car is hot (from running or just sitting in the sun). May also be a faulty ECM with a thermally induced open circuit, but right now I suspect a wiring problem.

What I would like to know (having read the FSM and the related fault checking chart) is that if this is just a serial comm problem, but the ECM is still powered up and talking to the engine sensors, will the engine run normally? Because if that is the case, then my troubleshooting is a bit less involved. If it were a mem-cal (PROM) issue then I have to think the engine would go into limp-home mode (reduced performance/driveability).

While this fault is present, A/C is locked out, SES light is on, I loose the tach and temp gauges on the CRT, and calculation of avg mpg and fuel usage is suspended, as would be expected due to loss of ECM communications. The voltage and oil pressure gauges do still function. Nothing else seems affected.

To add yet another layer to this screwiness, today (for the first time) I started the car, everything was fine for about 30 seconds, then I got the "electrical problem" telltale, lost the IPC readouts (00's and ERROR displayed) and the CRT would only show the radio screen, as in retained accessory power [blacked out after 10 seconds]. Now I wonder if this is all related to a bad ignition switch (intermittent power loss to some modules) or maybe a central power supply issue. Shutting off the car and restarting cleared this condition immediately.

I will be working on this, but would welcome any thoughts or suggestions regarding the above symptoms in case anyone has gone through the same exact issue and resolved it.

KDirk

Edited by KDirk (see edit history)
Link to comment
Share on other sites

Is the ALDL cover (the one that says "Do Not Remove" by the parking brake) in place ? The ALDL is the serial link and it makes a loop around the car. Part of the loop routes through the cover.

Link to comment
Share on other sites

Padgett: Yes. First thing I confirmed was that the plug was seated. Have also manipulated the 3 harnesses/plugs going to the ECM while the problem was present, and could not induce a restoration of communications. Headed back outside to try a few things.

KDirk

Link to comment
Share on other sites

If this did not begin until you tucked the ECM back in place I suggest looking there first. As you rotate the ECM in the wired are tight against the carpet for a bit. I suspect somthing may have loosened. Think you can push on/wiggle the plugs without removing the ECM

Engine will have enough info to run well enough but anything relating to the BCM or speedo or CRT won't, the ECM communicates with the BCM mostly over the data link.

Link to comment
Share on other sites

Guest Mc_Reatta

Kevin, you sure can come up with some strange issues to make us respond to.

The intermittent E013 code proved to be a bad ECM in my case and is easy to confirm by looking at the .45 reference output that the ECM needs to supply to the circuit. If it's not there consistently, the ECM is bad.

Turns out the replacement ECM I put in to fix the O2 sensor issue gave me intermittent serial data issue same as you have. I dropped the ECM down, cleaned and reseated the connectors and wiggled all the wires trying to induce or correct the problem without much success.

This problem has no effect on how the engine runs so I kept on driving the car with the ECM laying in the foot well so I could play with it.

I carry an umbrella in behind the seats this being sunny FL and all, and when I was running down the road and the data loss occurred, I picked up the umbrella and poked at the wires to see if anything changed.

Moving things around didn't have any effect, so I poked the case a nice rap with the pointed end of the umbrella, and lo and behold the problem went away.

I repeated this experience several times as the problem would return and each time the problem went away. Then feeling a little braver I poked the ECM when it was working and instant serial data loss.

Knowing now the I had a mechanical problem in the ECM I put in my last spare ECM and had the exact same issue with that one.

Pulled one of the ECMs and opened up looking for the issue, resoldered some joints, cleaned and reseated the PROMs etc never finding a smoking gun or fixing the problem.

Broke down and took one ECM in to Advance as a core and picked up a replacement, and issue solved.

So I'd suggest you replace you ECM and you will probably solve your problems.

Link to comment
Share on other sites

MC-

Thanks for the tips. One thing I failed to mention in my first post was that this is a rebuilt ECM I picked up (claims on the sticker to have been "fully computer tested") and I did not have a problem with this unit until I put it back up inside the dash. Now, if it is a thermal problem I figure it would've shown up even when it was hanging out.

Probably even more likely to show up then as when inside the dash it is against the air ducts, and I was running the A/C on or near max cool each times this occurred. The ambient temperature behind the glove box is probably less than on the floor pan due to the proximity of the air ducting.

So, this all leads me to a short list:

1. Wiring/connector issue at the ECM harness.

2. Poor solder joint on socket(s) inside ECM, so related to #1 above.

3. Other cold solder joints on ECM circuit board (yet it passed QC test when rebuilt)

4. Intermittent loss of contact with the PROM pins inside the MEM-CAL (will be pulling the ECM and popping the cover to test this theory shortly) but it seems like this would cause major engine performance issues, which I am not having.

If nothing else pans out, I will be exchanging this ECM under warranty.

KDirk

Link to comment
Share on other sites

Guest Mc_Reatta

I had both original ECMs and rebuilt ECMs but all were used / untested when I inherited them, but I'm pretty sure your problem is internal to the ECM.

Even if your ECM is a rebuilt, I still suspect the 20 year old circuitry and components inside can go at any time now. Points out the difference between rebuilt and refurbished parts.

You would think the seria data failure failure would cause the engine to run poorly, but when you look at what the data line is used for, you can see it really doesn't.

The ECM uses dedicated lines to get information to and from the sensors, and supply signals to the ICM and injectors etc.

The data lines are used to talk to the BCM, the IPC, and the CRT via the CRTC.

So the failure of this output of the ECM causes the IPC and CRT data display dropouts you see, but do not affect the signals that control the engine functions.

Link to comment
Share on other sites

Guest Mc_Reatta

Believe the data comm lines are akin to the original computer LAN Token Ring setup.

It can withstand a break in one location, but two breaks or more and data doesn't flow to each component anymore.

Link to comment
Share on other sites

MC & Padgett,

Correct on both counts. The ALDL line is a ring topology (in network/IT speak) although it differs from token ring in that there are no "tokens" used to transport packets of data. There is a busmaster (in the case of the 86-93 E platform vehicles the BCM) that manages the network and all other modules are considered slave nodes. GM designed it in such a way that almost all nodes (modules) had two drops, one from each direction.

This insured communications were maintained to all modules if there was one break in the ring, or if one module died; assuming it didn't pull the data line to voltage or ground in the process. On the Reatta (for those who have fitted the compass option) the compass module has only one connection to the data line (as shown in the FSM). I suppose it was considered non-critical, and so the redundancy of two connections was not utilized. All other network attached modules have redundant connections however.

To get back to the original point of this thread, I have not yet found the culprit for this latest episode. Will work on this more in the next couple of days and will post back when I have identified the cause.

KDirk

Link to comment
Share on other sites

An update:

Pulled the ECM down from the dash again. Manipulation of wires, connectors, etc. made no change in symptoms, nor did "percussive maintenance" [whacking] of the ECM. Removed the MEM-CAL cover and pressed and prodded on the MEM-CAL, engine would get rough and sometimes die while doing this. I've decided it is a bad ECM (gotta love reman'd stuff) given this latest finding. Have a replacement coming tomorrow to my FLAPS to exchange under warranty.

Car is running ok right now, but there is clearly a heat induced failure of the ECM, so either a failing electronic device within the ECM or bad solder joints on the ECM circuit board are apparently the root cause.

As of today, I was getting codes E013, E024, E026 and E027 (all history) when the ECM would come back online. Code 13 is 02 sensor circuitfailure, 24 is VSS circuit failure, 26 is a quad driver failure and 47 is loss of serial communications fault. Combined, all are further evidence of a flaky ECM.

Should be solved by this time tomorrow.

KDirk

Link to comment
Share on other sites

Yes, it popped a b336. Now that it's dark and cooled off somewhat, the ECM is playing nice again. Cleared all codes, and everything is as it should be; until the daytime heat tomorrow. Another reman ECM will be picked up Wed PM, will be swapped in. I fully expect this to fix the remaining issues. Unless the next ECM is also bad.

Edited by KDirk
added some detail (see edit history)
Link to comment
Share on other sites

Another update:

After swapping in another rebuilt ECM Wed PM, two days of testing have shown the previous ECM to be the cause of all ills. All is now operating normally, though my block learn value is high at 155. Need to figure out what is screwing up now, and get the BLM down closer to the median value of 128. O2 sensor readings seem low (barely over .45 which is the reference voltage) so I wonder if my new O2 sensor is also bogus.

KDirk

Link to comment
Share on other sites

Padgett,

True that. I probably need to wait a week or two for a break-in with the new ECM to see if it normalizes or not. If it is still high after sufficient number of miles driven, then I will suspect a further problem. Not many highway miles on it as yet since the ECM change, so that may also explain it.

KDirk

Link to comment
Share on other sites

Yes,

I changed the O2 sensor, plugs, wires, ICM/coils (to Delco) PCV valve, stat and cleaned the EGR and MAF as a regular tune-up, before I figured out the ECM had the wrong MEM-CAL in it. So, all the major stuff is new or in good condition otherwise. The O2 sensor seems active in diagnostics. The OXY indicator cycles regularly and the cross counts are active. The O2 readout moves around but stays pretty close to .45 at idle.

Coolant sensor reads good, and temp gauge is always well within normal operating spec, even with the near 100 degree days we've had this week.

KDirk

Link to comment
Share on other sites

Got a reading of 155 while parked (idling) to do diagnostics. Int value was staying around 128 (120-136 or so, moving around but this is normal per the FSM). I thought the BLM was the long term figure, and the Int was the short term adjustment. Do I have that reversed, or is that correct? If correct, seems the BLM should be lower unless something is wrong. Again though, I figure it may take some time to settle in, since the ECM being replaced reset those values.

KDirk

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...