I'm STUMPED!!

[GN_Guy]

If you have any thoughts or suggestions - I'd really appreciate hearing them.
Here's the deal:

I have a 1984 GN totally stock execpt for the Hypertech Street chip.
It has been stored for about 15 years (58K orig miles).
Before I laid it up in storage, I was having this intermittent, random problem with (what I believe is) the electonic ignition/ECM/computer system.
I could be driving down the road happy as a clam and the car will suddenly quit. Dead as a doornail!!! In fact, it dies so hard and fast that the ECM will not register any error codes. Nothing!!.
It will eventually and ALWAYS restart, but it can take from 20 seconds up to 20 minutes.

About 6 months before I stored the car I took it one last time to the Buick dealer in town and explained everything to the mechanic that was going to work on it. I told him to drive it like it was his own. Take it to work, to the store, to the track - wherever. He did this and called me about 2 weeks later and confirmed what I posted above. His recommendation was to replace the ECM box and chip (which had a recall). I told him to replace the computer and that I would replace the chip with the aftermarket Hypertech street chip. We did and the car ran great for months. Never missed a beat. I drove it 1200 miles to my folks to put it in storage and it never quit once. Life was taking me in another direction, but the car was running great.

Last summer I broke the car out of storage and got it running. I did all the right things, fuel change, filter changes, oil change, etc. etc. The car was running good - not great, but good - for a while. I took it to a body shop to have some work done and when I called them to see if it was ready they said "Yeah, but it's stuck in the paint booth - it won't start". CRAP!!!!

So I went and pulled the ECM, took out and reseated the chip, checked the board to make sure that there wasn't any massive corrosion or problems. Checked all the grounds and put it all back together and guess what! Yep - the %$#*&!@(&% thing started right up. Two days later I had the car idling in the driveway and it quit again. 2 minutes later it started right up.

So, the problem is apparently back.

Has anyone heard of a similar problem with a stock '84 GN????

Because it won't HARD FAIL, I can never really get to troubleshoot the the stupid thing!?!? Totally frustrating problem.
That leaves just blindly replacing parts and hoping that I eventually find it - awefully expensive.
I personally believe that the problem is in the computer connector/harness. It seems to be the only thing that would explain it running flawlessly after the computer change for 6 months or more and then coming back to haunt me. Possibly a cracked wire, or loose pin or seat???? Seems to me that when the mechanic changed the ECM box and I changed the chip that we somehow made this intermittent connection more positive for a while. But after a long storage it must have slipped back or something.

Or, I have considered just yanking the whole stock electonic iginition system and installing an aftermarket system. I could use the knowledge gained from this endeavor on another project that I have coming up soon (build a scratch LS-1 for installation in an aircraft). I just want the thing to run reliably - I actually want to sell it when the economy improves again. Which aftermarket electronic engine control systems do most folks put in these cars when they mod them for the track??? Which system would be the least complex to install that would provide me with a reliable, close to stock electonic engine control system?

My preference would be to keep the stock system - as the car would eventually be more valuable that way to a stock purist. But if it won't run reliably, the car's value is diminished anyway.

Anyone got any ideas or suggestions???
Sure would appreciate your insight.
Thanks.

 

[TurboTerry87gn]

Trouble Shooting

So you say it takes from 20sec. to 20 min to restart? Are you trying the whole time for the 20 min or just waiting 20 min?

Here are my thoughts. I have had a simular issue with a diffrent car, I found it to be the ignition switch. As the keys swung around the contacts would intermitanly break killing the car.
If the ecu looses power it will not record any error codes...


http://www.gnttype.org/techarea/images/wiring_images/ecm_turbo.jpg

You could put a meter on I blieve pin A6 AT the ecu Pink with black stripe and check the voltage, this electrical print is hard to read poor quality scan. But it looks like pin A6. and when the car dies don't touch any thing check the voltage here. There are some other points I marked on the print that might be worth checking WHEN the car dies. Giggle the keys around and slightly move the ignition switch back and forth while in park and see if the car dies. Also it helps the use a small diameter needle to poke through the middle of the wire to gain acces to that wire.

Because it only takes somtimes 20 seconds to restart it sure sounds like a bad connection or intermittant contact like the ignition switch and not like an electronic componet. Another area where I have seen my gn have electrical issues, was at the bulkhead connector that plugs into the back side of the fuse block. The terminal for my fuel pump hot wire had a bad connection here.

Connector Pinout Diagrams

 

[bobc455]

FWIW, when I hear about problems with a large time effect (several seconds to several minutes), I usually think about a thermal issue of some sort. For example something gets hot, and when it gets hot enough it breaks a connection until it cools down again.

-BC

 

[GN_Guy]

Terry, Thanks for the thorough reply.
Yes, I would try the car on and off with some long waits in between, to prevent running the battery down.
It would eventually, always, restart.
But you are right on with this being a way to interrupt power to the ECM and that the ECM would not record any error codes.

Here's what has gone on since my last post:

Initally the car would not start.
Found the fuel pump frozen up.
We got the fuel tank cleaned out and re-installed.
Pump works great, FP was good, but it still wouldn't start.
So, using the GM service manual we started troubleshooting I determined that:
1. The ECM was alive and kickin'.
2. No error codes were available
3. There was no spark.
4. The 1-4 and 5-2 coils showed ~12Kohms, but the 3-6 was open.
5. Further digging showed that the ingnition control module was getting power, but no signals were being sent to any of the 3 coils.
6. As a side note, I noticed last summer that the tachometer signal in the instrument panel was gone (this signal comes directly from the ignition control module).
Everything seems to point to a bad ICM.

So, I called Casper's and ordered the 86-87 ICM, Coil pack, adapter plate and wiring harness adapter. I'm now $400 poorer - but it should all be delivered on Thurs. I'll try to get this thing back together before I have to head back home. There really wasn't much of a choice but mod it to the 86-87 ICM setup. Ultimately, I would have to replace the 84-85 coil pack and they are scarce and expensive. It was much easier to just order all the parts from Casper's and do the mod.

Oh- I also found a place were one of the wires going to the injector harness had been pinched somewere below where the ICM mounts on the 84-85 (front - slightly right of center) engine.

Also, it seems to me, as I remember, that there were times when it would "cut out" and there would not be spark. I remember one time that I had a plug wire off trying to see if there was spark and the damn thing fired up and scared the %*#$& out of me. But, the best that my aging brain can remember is that somtimes it did not have spark. It would always start again though after some time.

I don't know if there will ever be a way to tell if any (or all) of this will fix the problem that has been frustrating me all along. But if I put it all back together and the thing runs reliably - it will all be worth it. If the problem comes back - I'll just keep troubleshooting.

At this point all I can do is wait for the parts, put it all together and try it. I'll drive it as much as I can and see if it quits again (fingers crossed).

I will study the info that you posted more and let ya know what happens when I try the car again. Probably be late Thrus or early Fri (5/8-5/9).

Thank you very much for your interest and assistance.
R/ Todd

 

[GN_Guy]

Bob,
I agree with you totally.
It may be that this thing has been the ICM all along (see my post above).
I sure does sound like a bad solid state device that was temporarily going ka-put.
Maybe the new ICM setup will fix all this.
Thanks for your input - I really appreciate it.

R/ Todd

 

[GN_Guy]

Oh BTW - that wire that was pinched was the power wire (pink/black) that provides power to all the injectors! It was pinched before both C111 and S111 (this is the splice that sends +12V to all the injectors. Each of the grounds for the injectors goes to the ECM individually and the ECM grounds each of them individually at the appropriate time to provide the current path that fires the injector.

It was partially bare and could have easily been making intermitent contact with ground. That would definitely cause the car to quit!

The question is - would the ECM recognize this and provide some kind of error code???

R/Todd

 

[TurboTerry87gn]

If the power wire to the injectors grounded out the motor would die too fast for the ecm to log an error I would think. From my understandig the emc looks for parameters out of range for a specific time then it sets a code. Once the ecm detected the engine had stopped the error logging would stop.

Also a note on the Pink/black power wire that is the common hot wire for the injectors:
I'm guessing it's simular the 86/87 where the same fuse powers the fuel pump and injectors... If this wire shorts out and it dosent blow the fuse then there is a high resistance (poor connection) in the line some where. Most likely at a connector.

An example of this is a bad connection at the battery. When you would try to start the car the load through the bad conncetion would cause a voltage drop across the bad conncetion and the soleniod would click on/off. Each time the starter contacts made the voltage would de-energize the solenoid and the solenoid would drop out, then pull back in as the starter contacts open back up and the voltage rises back to the supply voltage.

 

[GN_Guy]

Thanks for the ECM info Terry. I'm good with electricity, but just not too familiar with how the ECM programming is set up.

Up until this last time that I broke the car out of storage, it's always started and I never found any blown fuses. When I was still driving it frequently and it was doing this "cutting out" thing, it also never blew any fuses.

This time, however, I did find the FP/INJ fuse blown! Which, I can't quite figure out, because the car was running when I put it the garage the last time 10 monts ago!?!?!? I suspect that it was the frozen up fuel pump that caused the fuse to snap while I was cranking the car trying to get it started.

Good point about the wire grounding out and not snapping the fuse. I'll check the resistance from the fuse block to the injector harness.

Thanks for the insight and assistance.
R/ Todd

 

[GN_Guy]

UPDATE: SO FAR SO GOOD!!!

Got the package from Casper's today. I ordered all four parts to do the modification, but ended up only getting the adapter plate and harness adapter. Called Casper's and discovered there was some kind of miscommunication on what I really wanted so they only sent me the adapters.
I ordered the other two and they will be delivered tomorrow (Fri 5/9).

In the mean time I hit the bone yards and found a place that had the ICM and Coil out of an 1986 Olds 88 for $50. Sold.

Brought it home, hooked it up and the car fired up on the first try!!! Seems to run great. I let it warm up for a while and then started driving it. Put 30 miles on the car today and it idled for probably and hour or more (total combined time in between drives). So far no problems at all. Didn't quit once. It seems to run at least as good as it ever did. Maybe better. It's almost been too long ago for me to remember.

Checked the FP and it is between 35-40 psi. Should be good.

Tachometer still doesn't work!?!? That's the white wire that comes directly out of the ICM so I thought that the new ICM might fix this. I'll have to troubleshoot that under a different post. Maybe someone has been down that path already. The GM service manual pretty much just glosses over the wiring paths for this system by including it in the instrument panel block in the diagrams.

So at this point:
Is the cutting out problem over?????
Was it the pinched injector supply wire that I found????
Was it the ICM???
Was it the Coil???
I guess time will tell.
Thanks for all your insight, help, inputs and suggestions.
If the problem with it cutting out and quitting comes back - I'll be back.
Thank you all.
R/ Todd