Rough Idle - Hesitation - near stalling - mostly when cold

[Zander916]

Now, before the flamers come marching in.

I've been researching this topic like a madman and am well aware of its popularity.

 

I changed the OCV's.

 

Recently, I got hold of a CAPv2. After loading in the tune it seems to be a bit worse. I tried a couple of variations. It identified my car as a 2005.5 so I went with that one. Then I used the ACN tune and it was SLIGHTLY better. v1.18 btw.

 

I've been monitoring different things trying to pin this down. (NO CEL)

AFR is normal 14-14.7 putzing around town. 12's on throttle tip in... so normal.

However, I watched the short term fuel trim and even after the engine had reached 190* and just sitting idle, it was still pulling a lot of fuel. -8% to -11%.

 

I'm kind of leaning towards the O2 going bad although I could see a MAF screwing around with it - I did clean that a few weeks back.

I bought it two months ago 104K miles. I have some history on it, but both spark plugs and O2 I have no idea if they've ever been done - I don't think they have.

I forgot important info. It is completely stock down to the paper air filter.

Thoughts?

[Rush600]

Not trying to be a flamer but you checked the blue t connection from the bpv under the intercooler? that was the problem with mine!

[Zander916]

I did check that. Well.... looked at it and it looked pretty snug. Plus sprayed it with carb cleaner and nothing happened.

 

I just thought of something. It would throw a code for sure but could I disconnect the O2 sensor - therefore forcing the car to ignore that sensor. Then, if it runs better due to the ECU defaulting to a baseline, or using an algorithm based on other parameters (I don't know which it does) I've discovered my issue?

 

Sound valid?

[Rush600]

I did check that. Well.... looked at it and it looked pretty snug. Plus sprayed it with carb cleaner and nothing happened.

 

I just thought of something. It would throw a code for sure but could I disconnect the O2 sensor - therefore forcing the car to ignore that sensor. Then, if it runs better I've discovered my issue?

 

Sound valid?

 

Sounds valid. But I'm not the person to ask. I'm just starting to get into this stuff!

[Zander916]

Sounds valid. But I'm not the person to ask. I'm just starting to get into this stuff!

 

Yeah... I just came from an 08 Cobalt SS Turbo. (Intake, tune, downpipe, bored 20 over and forged pistons). That damn thing kept me on my toes so I had to learn a bit. It was certainly fast though!

 

I think that's it.

What do you guys think?

[Zander916]

Are these a pain to replace?

 

It looks like a pain....

[12sechatch]

Sound valid?

 

I doubt it would go into a 'baseline' mode, but I'm not sure.

[Strizzy]

Intake manifold to TGV gaskets.

[Zander916]

I got it... 99.9% sure.

 

I unplugged the upstream O2 and needless to say the light came on instantly.

However, it ran flawless since not monitoring that sensor. So I have a new one coming tomorrow and will just take it easy until I can get it in.

[P-UNIT]

Please post an update, would be great if this were a possible fix!

[Strizzy]

http://legacygt.com/forums/showthread.php/diy-rough-idle-cold-start-cold-temps-lgt-07-09-155325.html

[Zander916]

http://legacygt.com/forums/showthread.php/diy-rough-idle-cold-start-cold-temps-lgt-07-09-155325.html

 

 

I'm not completely dismissing that as a possibility but I'm highly doubtful for three reasons. I do very much appreciate the input as well, I don't want to come off wrong. Forums are funny like that.

 

In that DIY post, it states that 2007-2008 are the most prone. Mine is 05. Not a strong point, but that's what it said.

 

It also stated that it's most likely to show up when temps are below freezing. It's nice and warm here with lows not dropping below 70.

 

But the most important point, the most telling I believe, is that it said the car will run lean. Which is true, but mine is running extremely rich. Once in a while it's difficult to start just as if flooded like old carb cars would do.

 

I'll give the O2 sensor a shot and if that doesn't do it, I don't have much option but to give this a shot next.

[4CEF3D]

Actually that write up is sort of wrong. In 07-08 i believe Subaru made that o-ring slightly thicker for that exact reason. The issue with the o-ring mostly happens with 05-06 as did mine. And it doesn't just happen when that cold on the 05-06s it would happen more often then the newer years because our o-ring is smaller. Mine would happen on Southern California Mornings.

[Zander916]

Actually that write up is sort of wrong. In 07-08 i believe Subaru made that o-ring slightly thicker for that exact reason. The issue with the o-ring mostly happens with 05-06 as did mine. And it doesn't just happen when that cold on the 05-06s it would happen more often then the newer years because our o-ring is smaller. Mine would happen on Southern California Mornings.

 

 

That makes absolute sense. I thought that was a bit weird, with the years anyway.

 

But, running rich is the opposite of what would happen I'm pretty sure.

It did have a vacuum leak earlier after I bought it but I fixed that. I used the same method to find leaks that I did everywhere else. (spray with carb or brake cleaner) That's not the ideal method, but if it were THAT bad it should react I would think. Also, when it is fully up to temp it still pulls tons of fuel constantly.

I think the O2 is reading incorrectly, probably never has been changed after 104K miles. The reason it acts up in that warming up zone is I believe, because when first started it's not using the O2, at least that's how every car I've known has been. That's why it idles high, and dumps fuel, to get a quick warm up of the O2 sensor and the CAT. But once it's done with that, I think my O2 still isn't as hot as it expects and reading incorrectly. But once fully hot, good enough to run a little better, although still pulling lots of fuel back because it's reading wrong.

 

I don't know... we can go back and forth for weeks.

Hopefully I can get a chance to dig into it before work tomorrow and I'll let you know.

[4CEF3D]

Ya i had an O2 go bad was dumping fuel like crazy and was still reading lean swapped it out was all good. 10 minute fix.

[Zander916]

Ya i had an O2 go bad was dumping fuel like crazy and was still reading lean swapped it out was all good. 10 minute fix.

 

 

I don't want to keep beating a dead horse over, and over, and over and annoy people. So just once more.

 

At this moment I am 100% completely, absolutely certain it's the O2.

 

I fired it up a little after 11 tonight after sitting for 9 hours. I waited about 30 seconds before driving off (with the a/c on).

Nothing... no issues whatsoever. Idled fine, pretty responsive throttle... it drove great!

I'm sure it'll be better with a properly working O2 but at the moment it's far better just unhooked. It's a good idea to stay out of boost though.

 

Oh, I never did mention this, sorry. I'm getting just over 11mpg and that's grandpa driving.

[Strizzy]

Plug the O2 back in, and unplug the MAF sensor. See if that makes it run fine, too.

[Zander916]

Ok, if anyone wants a further update I will, but this should be it.

 

I picked up the O2 and put it in.

I drove it (With a/c on) just long enough for it to reach operating temp and stay there for 5-10 minutes.

Initial thoughts are mixed, but promising. It drove fine, no stumbling, hesitation, or funny idling. The STFT was still just crazy but the promising thing was that it was EXTREMELY stable at idle. It didn't move around like it did before. Also, it steadily got better and I'm told it takes a couple hundred miles for these ECU's to properly relearn. So I'm hopeful.

Before the change, at idle it was bouncing around -8 to -11 or so. Today, it started out holding steady at -12.9 at idle, which is nuts I know. But by the time I shut it off it had slowly worked its way down to -3.

So I feel as though it was the correct part (luck on my part or otherwise). I think it should just get better from here as it relearns.

 

It could prove me wrong and be the MAF or something else but it looks good so far. I took a shot with the O2 because, I don't know the statistics but I'm pretty sure at least twice as many O2's go bad than do MAFs just because of the job duties. The MAF sits around just measuring clean, filtered air and I'm willing to bet quite a few that go bad are due to aftermarket oiled filters. (That happened to me on my last car) Not to say they don't go bad sometimes but I just thought.... if I were a betting man... I'd go with the O2 before the MAF. Which, it kind of was a bet since I was spending money.

[Max Capacity]

FWIW, you may want to think about replacing the front O2 sensor too. I have replaced both mine in the 05, car ran smoother afterwards.

[Zander916]

FWIW, you may want to think about replacing the front O2 sensor too. I have replaced both mine in the 05, car ran smoother afterwards.

 

The front is the one that I replaced.

 

EDIT:

 

I did some tests that I saw on some auto tech sites.

I only did first gear so far to test but 2nd or 3rd would be better. I didn't leave far from home, in the middle of town and I have to be careful. The whole reason I sold my 08 Cobalt SS is because it was too fast for my own good. I'll just leave it like that. LOL

WOT - peak g/s was 223. They say that it should equal roughly +/- 80% of expected hp. That for me would be around 278 (flywheel), which is about right. Peak boost was 14.1. Cobb claims 15 peak, but again this was first gear. Knock was 0. Peak MAF volts was 4.4, not bad.

[Zander916]

Back from the dead.

 

It was HOT as hell for quite a while and finally we've had some doable days and cool nights. I just discovered that the car runs NEARLY flawless with the a/c off. (it still has some bucking around 2500 rpm cruising)

It only acts up a lot with a/c on. The a/c works but I saw another forum where someone's a/c was overcharged and once that was corrected all was good. I'm just going to double check that tomorrow.

 

I've spent a lot of time looking for vacuum leaks, engine first started cold and starter fluid. I've looked a LOT over the last few weeks and it just doesn't reacted to any spray anywhere.

 

Tonight I hit a two lane highway - 55 mph - put it in sport mode, and selected 4th gear which put the rpm at 2500. Sure enough it kept bucking sometimes but it was very incline/decline dependent. I decided to watch the STFT and I finally caught something. Shortly after a brief episode of bucking the STFT would go EXTREMELY rich (-20 or more) for just a fraction of a second and then return to near zero. So I'm pretty certain it's missing. I have spark plugs on the way. I miss my Aeroforce Interceptor guages in my old car. You could monitor tons of stuff and watch the miss count - whether or not you felt a miss you could see it being recorded.

 

I've never had to deal with plugs going bad so I honestly don't know. I've always stayed on top of them as maintenance. Does it make sense that they would only act up under certain load / frequency conditions?

Maybe 2500 rpm is just the right frequency of spark and lighter load that it misses (spark escaping or being blown out?) WOT doesn't seem to act up but maybe the spark is stronger then. I don't know if that makes sense.

Either way, plugs will be done. I'm nervous about breaking one off in there though. LOL They don't look fun but I've checked out the DIY thread and I'll just take my time.

[Max Capacity]

If the plugs are old, under load they could cause a problem.

 

Like you said, just take your time and "think it thru" before you put a wrench in the plug.

[Zander916]

Interesting, as it turns out the a/c WAS overcharged by quite a bit. I got that resolved and that took care of most of the idle issue. It still gets a LITTLE funky from time to time but I think the plugs should take care of that.

 

I'll let you know. Plugs are not here as of yet. I think there was a vacuum leak that I fixed long ago but with the continuing rough idle I kept searching anyway.