breaking one thing fixes another...

[MyBeatGSX]

He's the short story to explain the background...

Ever since I installed the 550's the engine has misfired at idle like crazy. No amount of compensation on the SAFC in either direction did even the slightest thing to fix it or even change it. I've since checked, replaced, and fixed everything idle related except the head internals, the MAF, and the ECU. But to no avail. It misfires like mad, it fluctuates anywhere from 16-19in/hg, its rough when you hit the throttle, it occationally stalls when putting the clutch in, and the idle fluctuates +/- 70rpm of where its set at.

Now some of you may have seen my topic about blowing my J-Pipe gasket out. This had an odd side effect...

The engine now pulls a dead steady 20in/hg. It idles within 10rpm of where its set. It no longer misfires. Its not rough when reving. And it doesn't drop way low or stall when the clutch is depressed. I find it quite amazing. The car just drives better and feels smoother overall (of course its not safe to boost, but off boost its beautiful). I almost don't want to fix it and just drive the car NA for the winter.


Can some one please explain this to me? I'm pulling unmetered air right? So leaning it out must fix the idle right? Wrong. Even -20% on the SAFC didn't change the idle one bit before this (its currently at -10%). Please keep in mind, that anything you're thinking of suggesting for check or replacement has already been done. Don't bother. I just want to know how having an unmetered air leak fixed my idle and even off boost driveability in general.WTF

 

[Dream On]

That's a ton of vacuum for having a boost leak like that. How much pressure do you hold on a boost leak test?

 

[MyBeatGSX]

What do you mean that's a ton of vacuum for having a boost leak like that? The leak is at the J-Pipe, not post TB. It would have absolutely no bearing on manifold vacuum readings. I don't care what I hold on a boost leak test now, probably not much. That's not the point.

Before it blew out it dropped like 1psi every 5 seconds at 20psi, then took like 5+ minutes to go from 10psi to 0. It was way beyond acceptable. But this is all irrelivant.

Please try to focus on what I'm getting at here. I'm well aware that I have a boost leak and I'm well aware of how to fix it. That's not really what I care about.

 

[talon96tsi]

When you get on it does it misfire?

 

[donmagicjuan]

One question I have is did you ever swap out the 550s for another set? It's a long shot, but what if the flowrates on the ones you have are mismatched enough to cause problems? I know some brands have looser tolerances than others. I'm still struggling to come up with an explanation for your current state of affairs, however, short of a clogged air filter. Take it for what it's worth.

 

[Dream On]

I was just trying to help. Don't be an asshole though. If nothing else, your thread got bumped.

 

[BreathinHeavy]

have you logged any of this with a data logger of sometype? a bad o2 sensor will cause a car to run rich all the time. it is a safety thing in the computer. did you try a leak test of any kind? possible a pcv valve did you try that.

 

[MyBeatGSX]

I was just trying to help. Don't be an asshole though. If nothing else, your thread got bumped.

Please don't take it as me being an asshole, its just how I come across. I'm very sarcastic and was kinda hyper at the moment. Sorry if I offended you, I tend to do that... nothing personal.


As already stated, EVERYTHING possible was already done. Yes the O2 is brand new, the air filter is brand new. I don't even use a PCV valve, its all vented to a catch can, this isn't the issue either, don't bring it up.

For awhile it though it could be the injectors. Andy and I discussed this in another topic and came to the conclusion that I was probably sent 650's instead. I think this is completely irrelivant however because its running just fine now. Are they too big? Maybe. Are they mismatched enough for it to not run well? Doubtful because its running just fine now. However I like your idea and the #3 cylinder's spark plug always looks extra rich. I'll swap them around when I get a chance and see if the rich plug changes cylinders with that injector. Buying another set just to see what it would do isn't really an option, I'm not that rich. BTW, they're PTE's, any known issues with those?

No it doesn't misfire any time but idle. It runs perfectly at any other throttle position.
No I haven't logged it, I'm in the process of trying to buy a logger off my friend.

But here's the point from my "asshole" post. I've long ago decided that diagnosing the idle issue is a lost cause, so I don't want this topic's focus to turn to that. When I say I've replaced EVERYTHING except internals, the MAF and the ECU I'm not exaggerating or joking.
I want to try to figure out what the boost leak changed, why it fixed the idle, and how I can keep it that way after fixing the leak.

 

[zippyshoe]

This is a shot in the dark, especially since it is unrelated to fuel injectors, but I had a rough idling issue some time ago that resulted in occasional misfires as well as relatively low vacuum (18 in Hg in park/neutral, 15 in Hg idling in drive.) In my case, it turned out that the intake camshaft was off by one tooth. The car appeared to run fine except for the crappy idle. Now that this has been corrected, my vacuum is at 22 in Hg in park/neutral and 20 in Hg while idling in drive, and the idle is fine. If you have not already done so, you may want to double-check the timing marks on the camshaft sprockets at top dead center to ensure that nothing has jumped a tooth. Hope this helps!

 

[selmerguy]

The hard part about EFI cars is that without a wideband or logger it is hard to tell where your problem is at. With the info in your post I would suspect that you are running rich and the j pipe leak allows more air that in not seen by the maf that kinda evens things out. The ruel of thumb here is that you can run up 650cc injectors on a safc with a walbro 255 and afpr. Now every car is different so if they infact did send you 650s you current fuel contorller might not be able to handle them. But again without a wideband or logger tuning your idle is very hard. This is the only thing I can think of that can be your problem. Get a monitoring device and tell us what your o2 voltage or afr is. Good luck

 

[donmagicjuan]

Actually, PTE's have some of the tightest tolerances as far as I know. I think you could rule out mismatched injectors based on that alone. If you really wanted to know, and you could afford the down time, I'm sure you could work out something with PTE to have them test your particular set. However, on the notion of getting 650s vs. 550s, you would have to lean out to about -30% to get close to the same amount of fuel being injected by the stockers. Based on your level of frustration, I'm assuming you already tried this as well. But if not, perhaps enough unmetered air from the massive leak is getting in and making up for the defecit.

Have you tried putting the stock injectors back in since the problem started? Maybe that would at least narrow the scope some if the problem goes away. Worst case is it doesn't and we can move on to something else.

I'm sorry for dwelling on the same issue. I'd really like to help you out, but I don't think I'll be able to come up with anymore theories tonight.

 

[MyBeatGSX]

The engine is in time both mechanically and ignition. The belt was just done and I checked the timing myself right after.

No actually I haven't tried going as high as -30%. I tried to -20% and this didn't even make the slightest difference or change so I didn't bother going higher. Maybe I'll have to try -30% or lower just to see what it does. My stock injectors were sold as soon as they came out, so those can't go back in. I guess I'll try contacting PTE as a last resort if we can't get anything else figured out.

I do have an narrowband A/F gauge (watch, you're about to see how useful they are). When I come to an RPM and stay in a steady curise at that point, the gauge shows steady a couple bars rich. But after a few seconds this starts to lean down and I'm back to the normal stoich back/forth bounce. So this would indicate to me, that my current settings are slightly rich (which they should be, you always want the ECU to pull fuel not add it) but the ECU is easily able to compensate and bring it back to a stoich mixture. So IMO this would seem to rule out running overly rich from too large injectors.

Thanks for the help everyone, keep the ideas coming if you have any.

 

[selmerguy]

I have done alot of messing around with my cars after I bought a wideband and you would be supprised that the nbo2 sensor can show lean when Infact you are running 13.5 AFR. I tried to duplicate your problem last night cause I had nothing better to do. I found that when I get into closed loop my car runs ok untill I get to about 11.6 afr at idle. Then it starts to miss alittle and run poorly. So I still think you are rich. You wont hurt your car at idle so lean it out on the safc untill it idles better. Try this and get bak to us.

 

[MyBeatGSX]

Sounds like an idea, I'm going to fix that leak when I go home today and we'll see how some super lean idle settings on the AFC do. I'll post back the results.


Thanks for going to that trouble for me.

 

[honduraneclipse]

Normal idle should be 18 - 20hg vacuum.
When you have unmetered air leaking out , like a boost leak even at idle. The ecu overtime adds fuel to the LTFT to compensate for the unmetered air leak, The IAC motor does not have enough time to catch up to the ecu commands so causing your idle problems" fuel cut mode".So since your car is already running rich,, adding or subtracting fuel at idle is not going to do a damn thing remember your injectors open and close time stays the same.

 

[spaceball]

owning a 5 gas analyzer would make life soo much easier!

 

[spaceball]

if you feel like driving to norwood, mass(uti) i'll be happy to 5 gas it for you and tellyou exactly whats happening. i work at forte's parts connection but i havent taked mike forte into buying a 5 gas yet. keyword ....yet.

 

[MyBeatGSX]

Normal idle should be 18 - 20hg vacuum.
When you have unmetered air leaking out , like a boost leak even at idle. The ecu overtime adds fuel to the LTFT to compensate for the unmetered air leak, The IAC motor does not have enough time to catch up to the ecu commands so causing your idle problems" fuel cut mode".So since your car is already running rich,, adding or subtracting fuel at idle is not going to do a damn thing remember your injectors open and close time stays the same.

Please read the entire post before responding. My car now idles perfect WITH the leak, the leak is not causing the idle problem. I don't know how many times I need to repeat that.



New Haven to Norwood Mass is a long ride. If I get desparate enough I'll stop by.

 

[MyBeatGSX]

UPDATE:

Fixed the leak and got the logger. The first thing I did was get my fuel trims zero'd out. The results were strange.

Idle fuel trims are now 0% STFT and -1% LTFT, that sounds good to me. However the SAFC is now adjusted to -21% at idle... but wait that's not the weird part. It still idles as bad or maybe even worse than before.

I tired playing around with the adjustment for awhile. And at +23% the engine idles perfectly.WTF I don't even know what to think. Not only is that rich way beyond the ECU's compensation range, it has the O2 pegged riched and it goes against everything we just came up with. Anything within the ECU's open loop adjustment range causes misfire and rough idle. If I do say, 10% rich. It will run well for a couple seconds until the ECU brings it back down to stoich and then it will start to run like crap again.

The O2 cycles perfect and often so I don't see that as the problem. Is it possible the O2 isn't calibrated correctly and what it sees as 14.7:1 could actually be more like 16:1 or higher? This the second O2 that I've used so its hard for me to believe I got 2 bad ones in a row.

 

[donmagicjuan]

Did you get a chance to swap the injectors around to see if the plug fouling follows the injector for #3 cylinder?

 

[MyBeatGSX]

No, unfortunately I didn't and I probably won't be able to until next weekend. This is all I have for now...