Riching out, intermittent, maddening

Hi...

'91 B2600i, previously quite reliable, seems to be in good mech shape, not burning oil, used oil looks good.

Symptom: randomly goes super-rich and starts stumbling, loses power, stalls, blows black smoke, stinks of unburnt fuel. Then the next day, may start and run perfectly -- for many miles. Then on another occasion, suddenly goes nuts and blows black smoke, stumbles, loses power, barely drivable.

Things I've tried: disconnected O2 sensor which seemed to fix problem, but that was apparently an illusion; it has now failed with it disconnected. Deinstalled, reinstalled FPR, thumped on it. Wiggled every connector in sight. Took it (when it was running OK) to a mechanic who, of course, couldn't diagnose anything 'cos it was running fine. Also replaced air filter, checked plugs (OK, all look the same, gaps are good, not too carbony).

Feeling kind of desperate. I have the shop manual (pdf) but I don't have any of the special tools (the SST box, vac gauge etc) for diagnosis. And the problem is so intermittent it's crazy-making.

Help...?

Sounds like a computer. Remove the passenger side kick panel and pull the computer. Open the top case and look at the area surrounding the two capacitors. There will be only two that look about the size of a AA battery but only 1/4 the length. They will leak eletrolytic acid on to the board and compromise one or more of the circuits. Sometimes it will be obvious and you will see green corrosion on the board. Other times it will be so subtle that it's easy to miss. You might only see a slight discoloration on the board around them. Basically, those areas of the board should be just as clean and shiny as the rest of the board. If not, your caps are leaking and that is probably your problem.

Another thing to check is the resistance of the coolant temperature sensor. It's the one in the thermostat housing that has a two terminal connecter (the one with only one goes to the temp gauge in the cluster.) The book calls for 2,200 - 2,700 ohms at 68 degrees F. If it's a little cooler or warmer, then your numbers will deviate slightly. Of course this probably won't tell you much if it's running smooth at the time but it's still worth checking. The ideal time to test it is when the engine is acting up. At 176 F it should read 200-400 ohms. So get it up to operating temp and check it there too - hopefully while the engine is acting up.

Same goes for the Air Charge Temp Sensor (intake air temp). At 77F, the resistance should read 29,700 - 36,300 ohms. You will really need to check this with the engine cold because after you warm the engine up, the air inside the intake manifold would probably be much warmer.

There are also some basic checks you can run on the Mass Air Flow, Throttle position Sensor, Idle Switch. Also, ignition components can sometimes get hot and become intermittent. If they start delivering a weak spark, then that could give you the rich and stumbling running. You might check the voltage on the "Engine" fuse in the cabin when it's acting up to see if you are getting low voltage. A bad ignition switch can cause the ignition system to cut out. If you don't have a good voltage there, then that could be a cause. But I would check the computer and temp sensors first and see what you have since the computer is the most likely culprit.

I hadn't thought of ignition switch (gee, you just expect those to work). I don't think it's hot ignition components because when the trouble starts, a cold engine runs just as badly as a warm or a hot one. In fact it gets a *little* better as it warms up, because I guess a hot engine burns a fraction more of the excess fuel that for some reason is getting dumped into the FI intake.

I prefer the computer as a culprit because it looks easy to remove and inspect and would explain everything but I'll do these other V and R tests that you suggest as well. Many thanks!

Looks like I have a smoking gun!



I pulled the computer and sho'nuff, cap 103 has been leaking and has compromised the board both uphill and down. There is corrosion damage all the way to Q112 downhill and Q9, C3 uphill.

Just pulling the board and wiggling a few things has changed the behaviour slightly, which seems like more evidence.

I answered my own question: without a modern circuit board fab bench, I don't think I can repair this. Looks like a multilayer board and there are several microcomponents which would be impossible to replace w/o specialty tools.

So now here comes the $64K question: can I still get this board from anywhere? If I get a used one, will it have the same darned problem? can I get a new one? I'll start my hunt now, but any clues would be greatly, greatly appreciated.

Many thanks to scotch and eddie who pointed me in the right direction.

Yep, that looks like pretty moderate damage. Very typical. I can set you up with one.

PM Sent. (I see you're new here. If you're new to forums, PM is private message. Check you PM "InBox" here: " target="_blank)

Just wanted to report back: PROBLEM SOLVED.

I sent the damaged board to Nick at Avpro in Florida. Expensive shipping from BC, and it took a while to turn around, but his repair cost was quite reasonable. The board came back to me this last Weds; installed it Thurs AM, and WOW.

The truck runs better than it ever has. No backfiring, smooth acceleration, quieter all around. Plenty of power. No more black smoke and stuttering. Idle speed self-regulates as the engine warms up. All works just like it's supposed to. It's like having a new truck!

Anyway I'm very pleased with Nick's service, and I want to thank everyone again for helping me to identify the problem. Much appreciated.