CarTalk.com Best of Deals Car Reviews Repair Shops Cars A-Z Radio Show

Rough idle

2002 Chevy S-10. I recently had repair work to replace intake manifold gaskets. After the repair, the mechanics had trouble starting the truck, and finally found they they had not placed the distributor back in correctly. (they also let gasket material get into my oil, but that is another story!) That was three weeks ago, and since then I have had a rough idle and every so often, the engine tries to die, but I can tell the computer is working to compensate by watching my tach (the rpm drops below 400, then rebounds to 500-550). I have changed the oil, put in new air filter, and ran a few bottles of fuel injector cleaner though the system. The spark plugs, plug wires and distributor are all less than 8 months old. I had the throttle control valve replaced 6 months ago. The computer is throwing no codes, at least none my OBD tester will register. The repair shop told me they tested fuel pressure and put it through a full diagnostic before returning the vehicle to me. But something is different after they worked on my vehicle. Anyone have any ideas here?? Any help is greatly appreciated!

Well, everyone makes mistakes. So if, for instance, they installed the distributor incorrectly OR let gasket material into the oil you just chalk it up to being human. However, I see that and the fact that the truck is now running poorly and I just have to wonder about their general level of competence and care taken in the work (instead of just mistakes).

One common cause of a rough idle - vacuum leaks. One place you can get a vacuum leak - a bad (or improperly installed?) intake manifold gasket. Of course, there could also be a more benign thing like a vacuum leak from elsewhere - something they missed hooking back up; something that cracked or split during the work, or just out of pure chance. But I think that this is likely their problem.

If you do want to check out the new manifold gasket for air leaks, take a torch (butane, propane or whatever), don’t light it - just turn on the gas and feed it out all around the intake. If you hit a spot that suddenly changes the idle then you are looking at an intake leak.

I am wondering what the 02 sensor waveforms look like,possibly damaged by the coolant leak. How was it leaking,into the cylinder,into the oil,onto the ground?

Much better chance there is a vacuum leak.

What about EGR that not setting a code?

When I had the coolant leak, I never saw any signs of it in the oil. There was never enough to pool on the ground either, although there were signs that it had run down the front and back of the engine. Much of it may have gone onto a cylinder, I don’t know.

I am baffled that it is not setting a code. Wouldn’t a bad O2 sensor code out?

Also, I failed to mention earlier that the severity of the problem seems to go up and down. Sometimes it does pretty good sitting at a traffic light (still a bit rough, but not trying to die) and at times it is worse. The weird thing is that the first two or three days after the repair on the manifold gaskets, I didn’t notice the problem at all. Then the check engine light came on, and I found a vacuum leak and fixed that (no more check engine light). That repair did not change the idle, one way or another. It seems to be getting worse over time.

Thanks for your help.

Additional info—When I do a cold start and just let the engine idle and come up to temperature with the transmission in “Park”,it runs well. When inside the truck with the doors closed, you can barely hear the engine. There is a slight vibration at low rpm (you can feel it when you are in the vehicle), but I have had that for quite a while. The vibration seems to disappear at about 600-650 rpm (normal idle is about 500-550).

In the old days, you could adjust up the idle a bit by turning a screw. Is something like that possible with modern, computer controlled vehicles?