Code 35 on Grand Am

The fan coming on is a sign that the computer may have problems. Check the ecu ground and plugs before replacment.

The fan comes on automatically with most fault codes.

I didn't know that. Thanks.

Well, I've now flushed $100 down the toilet in trying to get rid of trouble code 35. No luck at all.

Latest moves:

Replaced the computer with one from a nearby salvage yard. Exactly the same results.

Replaced the IAC with a new one from Advance. Exactly the same results.

It seems to me that this code might be setting as soon as the ECM goes into closed loop control mode. That's a guess on my part. When it does its thing after warming up in PARK for a short time the engine speed does suddenly drop off a little bit in conjunction with the check engine light illuminating.

While driving the machine there is no check engine light at stops. Pull into the garage and put it in park or neutral and the light quickly shows up again.

There are no vacuum leaks anywhere that I can detect, but that doesn't mean there isn't one lurking somewhere.

I'm at my wit's end at the moment. Could the oxygen sensor be causing this behavior? It's less than a year old.

You have cleared the codes and made sure that 35 is still there correct? That code could be set by other things than just sensors or the ECM. Check the ground going to the IAC and the connector on the coil for crud. If it was receiving a low signal from either area it could set a code. O2 wouldn't set it.

Why not take it to the dealer or to a garage? Throwing parts at a car is costly. The computers seldom fail and should always be a last resort, but you're stuck with it now because it's an electronic part.

Yes. Cleared the codes for certain because the computer was completely disconnected for a couple of hours while I waited on the replacement to arrive.

Wiring to the IAC is intact per my measurements from the disconnected computer harness out to the IAC.

Perhaps if I knew how the ECM went about its business in closed loop maybe it would help me figure out what's going on. However, those things seem to be proprietary and little information exists (as far as I know) on their inner workings. Probably don't have the proper tools to probe deeper anyway.

My thinking on the O2 sensor was that if it happened to be sending bum scoop to the ECM the computer, not knowing any better, would try to use all its abilities to control fuel and air to compensate for an error but in doing so the engine idle would be adversely affected and knocked out of expected bounds. Does that make any sense?

Let me admit that the thought of taking this machine to a pro has been creeping around the darkest corners of my mind. However, doing so would be a terrible blow to my self-esteem. My mindset is that I should be able to eventually fix almost anything even though "professional automotive technician" doesn't show up anywhere on my resume.

Clint Eastwood's Dirty Harry movie character once said on screen that, "A man's got to know his limitations," or words to that effect. Perhaps I've reached my limits on this thing, but I'm not giving up yet.

Well, I sure as hell understand that. But... I have to admit, it doesn't take me too long before I swallow my pride and take a walk over to see my neighbor who is a Chevy mechanic. Or sometimes pull up to the back door of the dealership for a little free look and advice from him or his associates at work. I'm proud, but I'm also lazy. Don't like to spend any more time on these things than I have to.

I didn't see your original post, just these follow-ups. Is the engine idle speed way too high when it's cold? I seem to get that impression from these follow-ups. I had a '92 Olds Cutlass with the dreaded 3.4L DOHC V6 that did this and continually set the 35 code. No vacuum leaks that I could find, tried a new IAC motor, etc. Once when I had the rubber duct off the air plenum inlet, it was sitting there running about 2200 rpm with the engine cold. I took a cardboard drink coaster and completely plugged the intake, and the engine still ran at 1500 rpm. Says I to myself, "There is most assuredly air getting in this engine somewhere, self." Turned out the intake gasket between the upper plenum and the lower intake was cracked in three places. When the engine warmed up, (and always just about the same time that it went into closed loop on my scanner), metal expansion closed up the leak. Hope this helps in some way. Good luck.

Garrett Fulton

Actually the idle sounds like the engine is running just about the right speed before it goes into closed loop. It drops down noticeably though as soon as closed loop initiates. But thanks for the input. Your experience and findings make perfect sense, and I may yet find that air leakage is somehow plaguing my machine.

IIRC, 200 rpm above or below the commanded rpm for X seconds will set a code 35. As another poster suggested, look for vac leaks. Does the engine have an egr system? Check the PCV operation, also. The ecu uses the gear selector indicator light pcb to determine gear selection. If (when) the solder joints go bad, the car will exhibit undulating idle, stalls at idle, and sometimes lock the doors.

Now that's an interesting idea, but it will be about a week before I can check it out. I'm here in North Carolina visiting my daughter. Will put your theory to the test next week. Thanks for giving my problem some thought.

"Here we go Steelers, here we go."