Eratic AFR's all of a sudden

[mk3forme]

First of all I have changed nothing to the car. Last weekend all of a sudden the AFR on my LC1 are jumping around more than usual. And going more into the lean area. Normal previously was jumping between mid 14"s to low 15's at idle. Same at cruising speeds. Now all of a sudden they are anywhere from low 14's to mid 16's at idle. Still going mid to high 15 at cruising. But always dipping back to high 14's. Like I said I have changed nothing, so im looking for suggestions on what it could be. I used a stethscope to check the injectors and they are all ticking like they are suppose to. I have been having problems with the car starting right up when warm. It will fire up on the first crank when cold, but have to crank it alot when warm to get it to start. Then there is a little puff of black smoke and smells like gas. Im thinking maybe an injector doesnt have a good spray pattern or something.

This is a pretty fresh rebuilt. Probably have 2-3000 miles on it. The injectors are stock and were cleaned and flow tested before installation. I have a walbro FP and AFPR. The pressure is set to around 28 PSi. Not lex afm. A 57 trim ct26 and 3" DP and exhaust. That is about the just of my mods.

Maybe someone else has had this problem in the past. Thanks in advance for any help you can offer.

 

[figgie]

wait sorry

that seems like normal operation...

The stock ECU bouces from lean to stioch and back x numbers of time.

http://www.cygnusx1.net/Supra/Library/TSRM/MK3/manual.aspx?S=FI&P=119

 

[Supracentral]

Sounds like normal operation. If you're seeing a variance in behavior, it's likely the o2 sensor aging/getting dirty.

 

[grimreaper]

time to dig up the voltimeter and play

 

[mk3forme]

Innovate thought the o2 sensor might be going too. Im sure thats a possibility, but its new as well. ONly has a couple of thousand miles on it. They said its not the wideband sensor. And im not arguing with the fact that its normal to bounce around like it does, cause it has done a lil of that from the get go. But its doing alot more jumping than before. It never went above 15.2 or so. Its jumping all the way into the 16 range at times. Something has changed but I dont know what.

I also asked innovate about cleaning the sensor. They said dont. I did wipe if off with a dry paper towel cause it was covered in soot. I took it out to recalibrate it to air thinking that might change things. But it didnt. I would imagine that if the wideband sensor was covered in soot that maybe the narrow band is as well. Maybe thats affecting its performance? Shouldnt I have other issues though if the narrow band is acting up?

 

[zurud]

boost or vauum leak?

 

[mk3forme]

I've had that mentioned too. Im thinking no since I dont have any issues with the way the car runs. I think I will look into the narrowband and see If I find anything there. Any other ideas?? Any ideas on why the car may be hard to start when its warm?

 

[mk3forme]

small update. I ruled out the narrowband. I still need to check for vacuum leaks. Now the wideband is buried over to lean all the time. But without an error code innovate denies it being the wideband as a problem. I need to check my connections. The guage is now jumping back and forth between afr and percentage. Once it goes through the warm up process then it used to stay on AFR. So Im not sure whats going on without it now. All this and the car doesnt seem to have changed at all. Without the wideband readings I would not be missing a beat driving this thing. And wouldnt know any different.

 

[grimreaper]

when i cracked my dp at the turbo flange, the car compensated for the "lean" reading the oem o2 was getting. Wideband showed the compensation to an extent until is slowly worked its way to max lean, and the car still ran great in closed loop.

 

[mk3forme]

when i cracked my dp at the turbo flange, the car compensated for the "lean" reading the oem o2 was getting. Wideband showed the compensation to an extent until is slowly worked its way to max lean, and the car still ran great in closed loop.

Thanks for mentioning that grim. Its funny you mention that cause I have a crack or gap in a weld on top of my turbo flange too. I wasnt sure if it went all the way through and it has been there since I installed this motor last year. Maybe with the stress of heat cycles and movement and all is just started working its way through. That could very well be my problem too. I wonder if there is someway to check that w/o getting it welded right away. I dont know of anything that would withstand the heat, that I could temporarily plug it with.

 

[grimreaper]

nothing that i know of, gasket maker melted off on warm up (orange exhuast stuff)

get your hand down there cold and feal for the warm air. Mine started out as a full inch crack but only had one tiny hole that you could see (carbon build up) and feel as leaking, after a few heat cycles and 130+ mph runs, it opened up a bit.

 

[mk3forme]

Well mine is one the turbo flange. So I have heat coming from everywhere pretty much right away.

 

[grimreaper]

Turbo to manifold flange? Imho your one step ahead of the curve man, you already know where the leak is . Heat yes, pulsing hot air that can be felt a few inches away no, block the rad fan or disable it helps.

Get it welded up and see what that changes.

 

[mk3forme]

Well I think that could be a leak anyway. I always thought that could be an issue but never acted on it. You made me take a second look. So thank you sir. It makes perfect sense too. If it is sucking in unmetered air then I can see how it could make my wideband read lean. I figured it would make my car run worse though since it has to be affecting my 02 sensor too. Kinda wierd that it just started happening now. I hope that is the problem.