Stalling when going to idle

[Swordsmith]

Earlier I had an odd problem; my Sonic would stall every time after I filled up the tank, and occasionally at other times. Unable to find the problem, I took it to my local mechanic, who found two problems. One, there was a wire from the bypass valve to the ECU which was unplugged. Two, the purge valve was stuck in the open position. Both these conditions had check engine codes associated with them, and those codes have gone away.

They also found a leak in the water pump, and replaced it.

However, when I got the car back, I immediately noticed a new, but probably related problem: the car now stalls MORE often. Instead of occaisionally stalling, in addition to the constant stall after fillup, it now stalls all the time, but under new specific circumstances, namely, the AC (or defrost) has to be on. AC/Defrost off, it doesn't stall out at all. But with them on, it stalls about half the time, but only when it tries to drop down to idle.

I push the clutch pedal down, foot off the gas, and the RPM gauge drops down to either 700 or 500. If it's around 700 it almost always swings back up to 800 and sits there (a little higher, actually, if I am still moving, 800 if I come to a stop). If it's around 500, though, it almost always rallies slightly, maybe to 550, and then falls off to zero.

If I am still rolling I can bump start it, or I can use the key to start it rolling or not.

I can avoid this by turning the AC off just before pushing the clutch in, and can turn it back on right away once the RPM's bump back up to 800.

It also seems to happen less often when it is under 80, maybe 1/3rd the time, versus about 2/3rds the time if it is 80+ out.

I have seen the engine code p0172 on and off since, but I also have seen that on and off for many months before. The code is for fuel rich, and I assume it is something to do with the flexfuel conversion or intake or exhaust... but maybe it is a clue of some kind after all.

Our current theory is that the tune may have been set up while the purge valve was stuck open, and that therefore perhaps a retune is needed. Otherwise, we're going to have to open it up and start searching for something else wrong, something that isn't showing up on the check engine system. Fun.

Any ideas?

 

[sonic_braaappp]

had this problem in my cobalt ss. didnt start until after i installed bigger fuel injectors and got tuned. fuel injectors i had were big enough to run e47 but i was tuned for/running 93 octane due to availability. whenever a/c was on it would bog down and die when i came to a stop. still does it if i use a/c, so i just quit using the a/c lol

 

[Swordsmith]

Hmm, I ran for over a year on the current injectors and never had this problem in any weather... but maybe I had some other problem which was preventing this problem from showing up until repaired?

 

[Natesriv]

all vacuum lines hooked back up?
fuel filter (my Riviera would die at idle after a bad tank of gas)....

 

[Swordsmith]

Something to look at to be sure. That wouldn't throw a code?

 

[Natesriv]

I wouldn't think the filter would...but a vacuum could...

Heck the last engine I had to worry about that was the 96 Riviera with a 3.8 S/C... it put up the CEL.. but seemed to have far more vacuum lines.

If you've not opened the hood I'd sure take a look around that bypass valve area, there's quite a bit of tubing there...rips or tears or just plain disconnected.

 

[ss2000slp]

Also seen a similar problem caused by faulty alcohol sensor. The computer was "seeing" 80% ethanol content, but customer was running regular 87. Do you have a way of monitoring your sensor, either scan tool or maybe ultragauge?

 

[Scot]

Our current theory is that the tune may have been set up while the purge valve was stuck open, and that therefore perhaps a retune is needed. Otherwise, we're going to have to open it up and start searching for something else wrong, something that isn't showing up on the check engine system. Fun.

Any ideas?

You reasoning is sound - look at your injector pulsewidths to see if they're hitting the low clip (1.5ms ? I don't know what our injectors' flow curve is) and still injecting too much fuel. If the purge valve was stuck open, you were constantly scouring your canister - so the purge flow would be very similar to an air leak.

Have you tried a battery cut / NVM reset to see if things change?

 

[Swordsmith]

Here is the response from Brian at VT:

Your fuel trims are very negative, over -30% long term fuel trims. This could be due to a leaking injector, a bad MAF sensor, or a boost pressure leak between the turbo and the head. A bad purge valve could also cause negative trims. I would bring it back to the shop and tell them the fuel trims are very negative.

I fiddled around with the various hoses and don't see any that aren't clamped down or have any visible leaks, but will advise my mechanic that the community thinks that's a possible cause.

You reasoning is sound - look at your injector pulsewidths to see if they're hitting the low clip (1.5ms ? I don't know what our injectors' flow curve is) and still injecting too much fuel. If the purge valve was stuck open, you were constantly scouring your canister - so the purge flow would be very similar to an air leak.

Have you tried a battery cut / NVM reset to see if things change?

I don't even know what that means... non volatile memory? Wouldn't that remain with a battery cut? At any rate no I have not, and if all it takes is removing a battery terminal and putting it back on, that's well within my capabilities

 

[cndctrdj]

Vac leak is my main idea but also a typical Gm thing is a dirty throttle body. I would check that first because it's easy and free.
The cobalts were notorious for that. It would stall all the time while the car was in neutral, or the Rpms would bounce all around

 

[Natesriv]

I had a bad MAF on my Riviera and it wouldn't idle for sh)t...

Anyone near you with a Sonic you could do a quick MAF swap, see if that affects it?

 

[cndctrdj]

Was it figured out

 

[rsSteve]

I had a bad MAF on my Riviera and it wouldn't idle for sh)t...

Anyone near you with a Sonic you could do a quick MAF swap, see if that affects it?

Swordsmith, if this is still an issue it looks like you are about 1.5 hours away from me. I would be willing to meet up and let you try my MAF to see if it works.

 

[Swordsmith]

So there was a problem with the engine coolant sensor, which was causing the deeply negative trims. It was odd, the shop's OBDII display device could see the coolant temp, and when unplugged the temp reading went to -40F, but no effect at all on the engine (while doing the same to a Cavalier made the car go crazy). At any rate, after replacement the fuel trims are now +/-5 or less... but sadly the stalling problem remains.

I am going to drive it to Pittsburgh and back over the weekend, and see if it "learns" and fixes the problem, but if not I will be experimenting. Would love to come up and borrow that MAF for a few miles, see if that does the trick. Personally I am still thinking it is the 02 sensor in the cat.

 

[rsSteve]

So there was a problem with the engine coolant sensor, which was causing the deeply negative trims. It was odd, the shop's OBDII display device could see the coolant temp, and when unplugged the temp reading went to -40F, but no effect at all on the engine (while doing the same to a Cavalier made the car go crazy). At any rate, after replacement the fuel trims are now +/-5 or less... but sadly the stalling problem remains.

I am going to drive it to Pittsburgh and back over the weekend, and see if it "learns" and fixes the problem, but if not I will be experimenting. Would love to come up and borrow that MAF for a few miles, see if that does the trick. Personally I am still thinking it is the 02 sensor in the cat.

Keep us posted, hope you get it figured out soon

 

[Natesriv]

Option 2 ... borrow his O2 sensor. .. if it's easy to get to

 

[Preloader]

I have had the same stalling issue before. And also had to turn off the A/c at stops.

Turn down for what? To avoid stalling.

I tried all kinds of tune updates and even had my idle rpm bumped up, which helped a little, but was only a band-aid.

It turned out to be the injectors. BNR had sold me chinese knock offs at an authentic price.

You may not have the knock off injectors, but I bring it up because the symptom was identical, and didn't appear for almost half a year.

The day I replaced the bad injector/s, the stalling went away.

You may also want to just inspect your injectors. Make sure they're sealing properly. If you have done any work in that region, like IM stuff or purge stuff, the seals may have been disturbed. Worth a look.

 

[Swordsmith]

Yes, I remember having a similar problem with my original injectors, and replacing them with these. I hadn't considered them as the problem since the problem wasn't the same, and since these are the real deal... but that doesn't mean they aren't also at fault. Seating and seals have also been fooled with and could be at fault, but I wouldn't be surprised at bad injectors.

 

[Preloader]

I just remember how frustrating the stalling at every light was, and all the time spent in trying to solve it. And the instant that the new injectors went in, it was cured and a wonderful feeling.

Also, relevant info..... My new working injectors have been slowly developing an occasional stall out. Maybe once/twice a week. I do a lot of gridlock commuting. So once or twice is not bad considering I spend the majority of my commute at an idling stop. But if it gets to be more frequent, I'll start with new injectors again, before looking at anything else.

The Green 42#s are common in other mfg'ers as well (BMW, Ford, etc.) So it's not hard to find them online for $40/ea. Just make sure you verify the Bosch authenticity before purchasing.