Cold/cool start idling problems

[Guest G Froelich]

Just a week old new owner of a 90 coupe and loving every moment of it! Unfortunately, it came with a few kinks. One of them is that although it easily starts up, after 1.5 minutes the rpms drop from 10 to below 5 on the gauge and, if I don't rev the engine several times for several seconds or gun it if I'm in gear, it will die. Other than that the engine runs smooth. It was at the mechanics last week (before I narrowed down the problem) and they didn't detect anything obvious. It has passed CA smog test. Appreciate any ideas!

[Guest CL_Reatta]

Im thinking that its your idle air control, they can be cleaned. But I prefer just to buy a new one from autozone for $60 or whatever they go for now.

[jonlabree]

If you prefer to save that $60.00 or so, Take it out and clean it. It's a simple fix. Two screws and it's out. although you need to put a rag under it so as not to drop the screws. They are NOT magnetic.

[Guest G Froelich]

Thanks for the responses. I took the IAC out and it looked pretty well carbonized (if that's the term). Tried my best at cleaning it (including unscrewing the mini piston and WD40ing the whole thing). Unfortunately, the cleaning didn't do the job. But then neither did the brand new IAC. It continues to idle rough after about a minute of warming up (not absolutely always, but 9/10 times). Other than that, the engine runs really smooth.

I noticed two of the plugs into the coil were looking rusty. Scraped away as much as I could in hopes this may solve the problem. Couldn't get most, though. Could faulty connections cause it to idle rough without running rough?

Thanks for any other ideas! (BTW the tip about the screws was really, really helpful.)

[D-a-n-i-e-l]

When was the last complete tune up? (plugs, wires, O2 sensor, fuel filter, ect.)

[jonlabree]

Have you replaced the plugs and wires? Which coil pack do you have? Square (Magnvox) or round? (Delco) If you have the square, you should replace them with the round ones. (Delco) If you do this replace the unit under them at the same time. Junk yard is the best and cheapest source.

[Guest G Froelich]

Thanks for the replies. The last complete tune-up was last week, but I'm wondering how complete it was now. I know they changed the air filter, but leaving wires with apparently rust-like corrosion on the connectors (I misspoke when I said plugs--sorry!) makes me doubt they changed much else. I'll bring it back.

The plugs on the squarish coil look good to me (drab gray). I'm sure it's the original and may be due for a replacement, if rough idling is a sign.

Thanks again. This is a great forum!

By the way, got new BF Goodrich touring tires on the car and it rides quieter than most cars I've been in--with the windows down!

[DAVES89]

Do the seafoam thru the plenum vaccum hose trick.

[Guest weewilly]

What is the "Do the seafoam thru the plenum vaccum hose trick" and specifically how do you do that? should and can this be done for maintanence type action?

[DAVES89]

Disconnect the hose that is just to the left of the Mass Airflow Sensor. Install a hose long enough to go into your Seafoam can. Start car,insert tubing into your Seafoam can, keeping a kink in the hose. Allow the vaccum to draw out the Seafoam. Directly above that hose is the throttle linkage. If you choose to, you can increase the RPM's of the motor.

Do not do this in an enclosed area as all kinds of carbon smoke will exit thru your exhaust system. This the easiest and best way to clean the plenum short of removing it.

[Roadster90]

Dave,

I think I may do the Seafoam thing just as preventative maintenance. I imagine the "kink in the hose" is to control the amount of Seafoam entering the engine (the Seafaom must bog down the engine, and try to cut it off....regulate with the "kink" and the acc cable???). How much Seafoam or how much of the can is necessary to clean appropirately please? Any other procedure tips please?

[DAVES89]

I believe the can is 12 ounces. Use it all.

[Guest G Froelich]

Just an update on the original problem of the thread: Took the car back to the shop and had them replace the plug wires, but alas to no effect. Cleaned out the MAP (?) which regulates airflow and the IAC intake--no effect. Tested the fuel pressure and noticed decrease in pressure when revving the throttle during the rough idle. The mechanic is thinking faulty fuel pump. Fuel filter is brand new. So, I'll probably fork over some more Ben Franklins in the next couple days. Haven't tried the plenum cleaning, but intend to.

[Ronnie]

Did you clean the passageway behind the IAC where idle air flows when you cleaned the IAC? It still sounds like a IAC problem to me. If It runs good with the current fuel pump I should idle as well. Don't spend your money unless you know for sure the pump is bad.

If you had an 88 or 89 model you could open and close the IAC with the touch screen in diagnostics mode to trouble shoot it. I'm not sure about the 90 model.

It would really help us if you would edit your profile and put the model of your car in your signature line. That way everyone knows what model of car you are talking about at a glance.

[Guest G Froelich]

Yes, the passageway (intake) got a good cleaning. There was the usual residue. The IAC is new.

The mechanic said that a faulty fuel pump may take a while to build up sufficient pressure (thus the rough idling initially), and once pressurized will run smoothly. I do have the slightest of lags at around 70mph and asked him if that might be related to the pump. He said when there's high fuel demand the pump may not be able to sustain pressure perfectly. Sounds reasonable to me, except that I think the lagging should occur at every acceleration from stop (greatest fuel flow demand, right?). It does lag initially, but after I've gunned it a couple times it smooths out. What do you think?

[jonlabree]

Check you fuel pressure before you buy a new pump.

[Ronnie]

Clarify a few things. Does the car idle rough all the time or just the first time it is started after sitting for a while? Does it idle good after the first start?

If it idles good after the first time it starts you could possibly have and injector that is leaking down while the engine is off (there is still some pressure in the system) and flooding one or more cylinders. Look for some black smoke when the car is first started and idling rough.

Just another idea.

[Guest G Froelich]

Another update. I read in the service manual that one way to test whether the problem is the IAC is to unplug the electrical connection to it when it's idling rough. If it doesn't change the situation, then it's not the problem. Which is exactly what happened. No change. I did use Seafoam through the manifold intake--but not as much as I would've like for fear of the neighbors and since I kept cutting the engine out--and it SEEMS not to be as stuck in the rough idling mode. In fact, if I gun it in gear from a dead stop once the ABS light goes out (about 10-15 seconds), you know, a jack rabbit start, then I don't have the problem at all. If I drive off a little too late or don't have the chance to gun it and it goes into the rough idling as I'm driving, then I'll gun until it clears its throat (about 2-3 seconds). Then it's smooth sailing.

The car starts up just fine. No problem whatever. Just after about a minute it will go into the rough idling, until a rev it a lot or gun it. No black smoke.

Thanks for the ideas. The service manual has a few more leads for me to follow.

[Guest Mike_s]

Just a thought but you might consider checking the vacuume lines.

[Guest G Froelich]

Final update (I think!). Cleared the problem (hunting idle at cold start) with a new O2 sensor and PCV valve. Don't know for sure which was the culprit, but metering pointed to the oxygen sensor. Now on to ironing out other little wrinkles in this gem of a car. Thanks again for all the help!

[Guest G Froelich]

Unfortunately this problem of a hunting/rich idle at cold start crept back over the months, but fortunately I just noticed the car starts acting up at the exact time it goes into closed loop.

I've read some of the threads about this, but my problem is different, most especially because I have no problem in closed loop once the car "clears its throat", for lack of a better way to put it (as a matter of fact, if I gun it in gear before it goes into closed loop then I don't encounter the problem at all). Once it gets past this rough phase it idles smoothly at 6Krpm and accelerates fine.

So that probably eliminates the crank sensor as cause, right? I haven't switch out the old coil pack for a Delco. Could that be the culprit? The IAC is new and clean, and besides the car still hunts with it unplugged. The O2 was replaced last year, as was the MAF. I've cleaned the EGR. Vacuum hoses are all in topnotch shape.

Thanks for any suggestions!

[Guest Bobby Valines]

Harbor freight has a fuel presser gage for $16.00. Buy it hook it up start car and monitor the pressure. Report back here with the numbers. I had my fuel pump changed for $300.00 sure made my car smother and stronger.

[Ronnie]

I've read some of the threads about this, but my problem is different, most especially because I have no problem in closed loop once the car "clears its throat", for lack of a better way to put it (as a matter of fact, if I gun it in gear before it goes into closed loop then I don't encounter the problem at all). Once it gets past this rough phase it idles smoothly at 6Krpm and accelerates fine.

Thanks for any suggestions!

The only thing I can think of that gunning the engine does is burn out raw gasoline quickly. One possibility is that you have one or more injectors leaking when the engine is shut off allowing gasoline to puddle in the intake manifold. The engine will run rough until that gasoline is burned.

Install a pressure gauge on the fuel rail and see if you quickly lose pressure when the engine is shut off.

[Guest G Froelich]

Thanks! I'll check the fuel pressure, but I have two concerns about fuel delivery being the issue: 1) Does the closed loop problem relate to fuel delivery? 2) Why would the problem clear when revved only in gear? I can rev it all I want in park and it won't clear any faster than it will if I let it hunt and run rich in idle (though it does help with it not stalling before it clears).

[Ronnie]

Thanks! I'll check the fuel pressure, but I have two concerns about fuel delivery being the issue: 1) Does the closed loop problem relate to fuel delivery? 2) Why would the problem clear when revved only in gear? I can rev it all I want in park and it won't clear any faster than it will if I let it hunt and run rich in idle (though it does help with it not stalling before it clears).

I didn't realize it was only when you rev it up in gear that the problem would clear up. In that case you can probably ignore what I said previously. Will the problem go away if you just put it in park and hold the RPM at about 2500 rpm for a while?

[Guest G Froelich]

Yes, the idle will smooth out in park, but it takes the same amount of time to clear as it would if I left it alone (assuming it doesn't cut out). In both cases, right before the problem clears the engine will rev up about 1K all by itself and then settle down into a smooth, low idle (and unlike with the closed loop starting the problem, I haven't been able to identify what the car is doing when the problem clears). But you know, let me check on this so I can be as exact as possible. Your idea about fuel pooling up is one I definitely want to pursue. Thanks.

[wws944]

You might also check for vacuum leaks.

I had similar symptoms with mine. Noticed that the little Z-shaped vacuum line (that has the 90° bend in it) that goes into the downstream side of the MAF kept popping out. Like 5 or 6 times in a couple of weeks. I suspect that even when the hose was connected it was loose enough to let some unmetered air in behind the MAF - which will totally confuse the computer. I put a 3/8" quick disconnect hose clip on it, and now it runs like new.

Note that since the computer uses 'adaptive' algorithms, it might take a few drives for things to completely smoothen out.

[padgett]

BTW the 1.5 minute delay (when cold, bet more like 25 seconds hot) matches the changeover from "open loop" to "closed loop" when it starts to trust the O2 reading.

[Guest G Froelich]

Any suggestions how to test the O2 sensor, short of buying another one?

[D-a-n-i-e-l]

They are only$20 and need to be changed every 30K or so.

[Ronnie]

Any suggestions how to test the O2 sensor, short of buying another one?

When in doubt... change it out.

[Guest Bobby Valines]

Auto zone had a O2 sensor made in CA. USA. G for $16.00 I think you should check your fuel pressure. Not much time not much money.

[wws944]

Given that things seem to be working ok in closed loop mode, the O2 sensor is probably good. (And you mentioned it is only a year old.)

Look carefully for a vacuum leak. Those 20 year old hoses can be deceiving.

[Guest G Froelich]

As suggested, I checked the fuel pressure at the rail. It hit 45 bar when pressurized w/o starting the engine and then in second went down to a steady 42. It stayed there for several minutes w/o budging. When we cranked the engine it went to 45 again but back down to 42 like before.

I did notice at cold start this time that it acted up exactly when it went to closed loop (as I've mentioned) but then after about 45 seconds it went out and back to open loop with a surge to over 1000rpm. When it started settling back down to regular idle rpm it went back into closed loop and started struggling again. I think it cycled between closed/open loop about three times. The rich/lean indicator ("LOW") flashes very slowly at this point, mostly off.

Is there a way to diagnose the O2 sensor from on board?

[Guest Kingglory]

I had the same issue and it was loaded with 20 years of carbon. I used Sea Foam and it smoked more crap out the tail pipe then a mosquito fog machine.

For a minute I thought I blew out the head gasket, LOL.

It idles smooth now and the choke works again with cold starts. I also added 1/3 to the gas tank and ran it through.

I am not a huge believer in "snake oil" products, but this worked very well for 9 bucks.

[Guest G Froelich]

Thanks, but I went through a can several months ago. Put it through one of the vacuum intakes and in the fuel tank. Certainly cleared out a lot of gunk!

[padgett]

You diagnose the O2 sensor by watching the flashes, and watching the cross counts, the instantaneous fuel and the BLM. AT 35-45 both should be close to 128-132 and counts should be above 10.

I treat the O2 sensor as a tune up idem - if the plugs get changed, so does the sensor.

[Guest G Froelich]

Thanks, Padgett. Here's what I came up with:

Between 35-45mph:

• O2 Sensor Cross Counts=5-9, but mostly stayed at 6
  
• Fuel Integrator=122-135
  
• BLM=126
  

I noticed that my "Knock Signal" (ED17) was stuck at 379 from start to finish--which is well over the 255 counts the fsm lists as max. Is that something to worry about?

I also noticed that the PCV wasn't seated into the intake grommet very well, so fixed that, and at least it seems the manifold isn't getting quite as hot as before, but it seems to also have helped with the spot of rough idling in closed loop. We'll see.

But, do the low O2 counts indicate the need for a new sensor?

Thanks!

[padgett]

Those do not sound bad, mixture is right on. Knock counts are cumulative from startup (and there are usually a bunch at cold start. What you do not want to see is them incrementing except possibly on a sudden acceleration.

[Guest G Froelich]

Finally decided to replace year-old+ O2 sensor because of decreasing fuel economy and lo and behold the new one also fixed the closed loop rough idling! After I got the old one out, I found that the garage I took it to for the replacement in '08 had cross-threaded it and may have damaged it in the process, too. It's taken me a while but now I know always to do the simple jobs myself. Thanks to all of you for the advice and encouragement.