Lean condition fixes itself?? (help me before it unfixes)

[williillii]

Sorry for the novel, , but if you have any input on this it would help me out alot.

I was picking up some tires today at a friends shop, and when I was leavin there were some dude outside houndin me to do a burnout when I was leaving. Normally I wouldnt indulge but Im changing my tires tomorrow anyway so I thought what the hell. Light em up a bit, roll off.

Now when the revs came back down and the car starts to idle after I pulled up to a light, the car is struggling to idle and runs real lean. Like 18.5:1. Then after like 20 seconds it will trim itself back to 14.7.

So I was like, damn...thats what I get for being irresponsible, I ####ing broke something or sprung a boost leak. So I drive real slow to a friends house about 20 miles out. Car is actually running real good, except at idle. So on the way back I decide to get on it merging onto i275 and DAMMNN the car hauls ass now. I look at the AFR...and its pegged at 10:1 right where it should be with no tune.

For the past year Ive been having problems with fuel delivery, and have tried everything I could think of to fix it. At WOT, it used to read about 12:1AFR. After Id been driving on the highway for about a half hour, some how it would richen up to about 11.3-11.5:1 (keep in mind this is a stockish setup at stock boost with no tune so it should be like 9.8:1) I ended up just assuming it was my 5 year old wideband controller malfunctioning. But after feeling the difference today compared to the readings it really does seem to be working correctly.

So anyway I've got to assume now that this idle problem is related to my fuel delivery issue. What could cause the car to run real lean at WOT, and run great everywhere else, and then all of a sudden cause it to run SUPER lean at idle, and run great at mid throttle and WOT??

First thought is boost leaks. So I will test that tomorrow but Im pretty sure its sealed up tight...it was a week ago and its only been on 10psi..with very little hard driving. Plus it runs fine after you get off idle.

Again sorry for the novel but Ive been dealing with this shit for over a year and really would like to put an end to worrying about popping my motor. Any help is appreciated

 

[gofer]

A lean idle on a stock fuel setup is when you'd typically find a vacuum leak around the throttle body and intake manifold area. Also, depending on where your wideband is located in the exhaust system you might be getting false AFR readings. For whatever reason most put their WB o2 sensor after the flex section in the downpipe and after years of driving you'll get an exhaust leak at the flex section which will throw off your AFR's.

 

[williillii]

Well thats what I was thinking. So I checked for leaks and its just a tiny one by the throttlebody shaft seal like usual. Ill get around to that soon. Also usually the idle will rise a bit with a leak. My idle dropped from about 1000rpm to ~700. Then when its able to richen back up and stop sputtering it was idling at 750rpm.

Of course when I got in the car this morning it was back to doing the same old crap. It holds an idle fine now but wants to run real lean at WOT again.

As far as the wideband, I tested the sensor and it seems to be working fine. I did used to have it behind the flex pipe in the rear o2 spot. But I had an exhaust shop move it up infront of the flex section for me, and check for leaks. Also, when the read out changes, I can definitely feel the change in how the car is acting. So Im inclined to believe what it says.

Here is a list of what I've done to attempt to fix this:

Fuel pump rewire
Evo9 fuel pump
New fuel filter
New FPR
Changed fuel pump oring
replaced line from fuel pump assembly to the rubber line
Tried spare MAF (could be broken too I guess..)
Different injectors
Even put 550cc injectors on it with an SAFC and couldnt tune the problem away (not how you fix things anyway)

 

[gofer]

Your profile needed updated 2 years ago when you joined this site and I nor anyone else won't be able to help you until it gets done. I can't even tell what year your DSM is and there are drastic differences between the 1g and 2g...

 

[williillii]

^Very sorry about that, I meant to do it when I graduated from the newbie section but I forgot. Its all in there now.

Something else Im not sure I made clear before...after driving on the highway for at least 5-10 minutes, the AFR at WOT will be as much as .7-.8 richer than flooring it after cruising at 45.

 

[gofer]

Theres a plastic screw on the bottom of your MAF, it might be covered in RTV to access. Try screwing it in to richen things up a bit, then see what happens at WOT. With your throttle body leaking like that you can't expect your idle to be good, once thats fixed then see if it continues.

---EDIT---
Its also possible you've got a blown sensor ground inside the ECU as well. If you install your WB o2 sensor and power it off sensor ground or somehow you shorted an o2 sensor, you can blow this ground inside the ECU. If the track is burned, you need to fix whatever caused it to burn or else after repairing the track, you'll just blow it again or, worse, blow something else in the ECU if your repaired track is stronger than the original.

Sensor ground is pin 92 on a 2G DSM ECU. 2g ECU Pinout
Once you pull out your ECU pull out a multimeter and check continuity like the picture shows...

You must be logged in to view this image or video.

 

[williillii]

I already tried the MAFS calibration thingy. I didnt observe any change really. Also I bought a fuel pressure test kit without realizing there is no test port on the rail. So I was thinking about just tossing some evo 560cc's I have on the car and see if it does anything. If not that should point me towards fuel pressure, no?

As for the sensor ground, that thought had crossed my mind, but then I thought it ran too good for it to be that trace. I will take the ECU out tomorrow and look. The wideband was wired separate from the ECU but the car has been through so much I suppose I could've damaged the trace another way. And I know I cant expect a perfect idle with a leaky system. Its just that it started idling funny then running correctly at WOT at the same exact time so its got me thinking the problems are related.

Thanks for your help gofer...I think everyone sees these threads I make about this and are just like "uuhhh...*backbutton*" . Ill try that stuff tomorrow and see what happens.

*edit*

As I suspected the ECU looks fine...didnt get around to changing the injectors. Really Id rather find someone with a fuel pressure gauge for a DSM.

 

[williillii]

You must be logged in to view this image or video.

Ok well I scored a fuel line off a 3000gt that was just long enough to work...

Pressure is about 45psi at idle with the regulator disconnected. With it connected it drops a bit (dont remember the number).

Took the car out for a spin...pressure goes up to about 55psi (8psi boost creeps to 11psi sometimes by redline) but then around 6000rpm sometimes it will all-of-a-sudden drop to about 51psi. On one occasion I saw it rise to about 51 then spike to 55 real quick. Never lower than about 51psi at 9-10psi boost.

How does that sound? I was thinking of replacing the o-ring in the fuel sending unit with 2 new ones to get it to seal tighter. Maybe that will get me my 3psi back?

Does this seem like it could be the problem though? Cause the car runs lean throughout the whole rshared_pm range, not just above 6k. During these pulls it was about 11.7-11.8. Thats pretty scary cause the car likes about 10.8-10.9:1. Due to the 9.5:1 compression..

Also I picked up a TPS and an alternator plug today..I would put those on but my multimeter broke so I cant adjust the TPS anyway .

 

[williillii]

Well I just tested the TPS and it was only putting out 3.9V. So I adjusted it so I have .63V at idle and I now get 4.96V at wot. Ill take the car out in a bit and test it out...

I was going to just change the TPS but the one I got is off the wrong throttle body apparently and is about 1/2'' too small across..oops.

Also found out my alternator was separating (bolts backed all the way out of the casing) so I put that back together and the car runs a bit smoother now, but it didnt really fix anything. Ill be wiring in a new alternator plug tomorrow because mine is cracked. Probably not really an issue but it needs to be done.

 

[williillii]

If anyone finds this by searching, just wanted to let you know Ive resolved the issue.

It was a combination of a bad TPS, broken alternator plug (causing voltage and thusly fuel pressure to fluctuate) and a wideband sensor that was dying causing inconsistent readings.