Grettings,

Thanks for looking at this.

Motor: 10.4:1 2.3L
Injectors: FIC2150
Pump: 255
ROM: 9417
Method: SD w/forced open loop

Issue is after idling for 15 seconds or so AFR's go crazy lean. It takes driving for a few seconds for them to come back. However, characteristics of the car don't seem to change like they should with a instant lean condition as the wideband shows.

My question for you guys who know these ecu's so well is does the ecu make any sort of transition at this point? Perhaps its transitioning to a table that I've overlooked.

Video:


& yes i know its very rich - this was me attempting to compensate for the lean issue with scaling.

 

I have to have mine idle in open loop to keep control of the AFR. 2150's are well known for heavy AFR swings in closed loop at idle.

 

With the way you were able to take off while the AFR was (supposedly) full lean, seems to me like something's just wrong with the wideband.

 

its just so weird that it idles fine for some time then goes crazy lean. I'm almost thinking that maybe the sensors dirty and needs the added exhaust flow to clean off.

Is anyone else on 9417 open loop, sd & 2150's?

 

Where is the wb sensor? Exhaust leaks do funny things to these readings. I forced open loop, and dialed afrs to 13.5 idle, later fixed my exhaust unintentionally and saw high 11 at idle. A leak at idle usly draws in fresh air.

May or may not help you out but that was my issue.

 

Its in the downpipe, car doesn't have any exhaust leaks. I've had a number of pre wb02 exhaust leaks and none behaved like this they were consistently off.

 

My vote is wideband is going bad. The car should start to stumble/lope when it leans out like that and you are getting absolutely no change in idle at all from richest reading to leanest reading.

 

I had issues with my aem serial guage a few times on the af. It's probably just you sensor going bad. Replace the sensor and see what happens.

I've had to replace my sensor twice in 2-3 years. One time it would just read 14.8 at all times and never change. Another time it would go full lean after idling for a bit, once I started driving it would be normal but would say i was going full lean on pulls. I replaced the bosch wb sensor and it was good as new.

You can usually pick them up in stock at advanced auto parts. The bosch replacment wb sensor I get is part # 17018. It should be the same for the wideband as well

 

From my experience, AEM Wideband gauges are not every accurate. Regardless of the vehicle or EMS (Hondata S300 & original version AEM EMS for Hondas/Acuras or Evoscan for Mitsubishis). I've typically seen the gauge verse EMS, read a 1.2 to 1.4 difference. I've also seen on 3 occasions, the sensors got bad in a fairly quick time, after use of race gas or E85. Now those could have been coincidence, but it seemed strange that they all occurred within a short period of time.

I'd absolutely recommend at least buying a new sensor, and seeing if that fixes the issue.

 

Ok so it wasn't my wb.
Boostleak tested the car and I've got a massive boost/vac leak. I'm thinking this may be it.

 

even with a boost leak i have never seen a wideband respond like that unless it had a weak ground... on my own car in fact. the ground came loose and i thought my sensor was bad. it would read erratic and then give an error of overheating... replaced the sensor,then said **** it checked the wiring and found to loose ground. sadly i cut out the first sensor so i was stuck... autozone wont return electronics that have been opened normally anyway so whatever.

 

Currently, I have 2 gauges / sensors and an extra bung.. i'll wire them both in and compare.

 

 

Bosch Replacement part Number 17018... Incase you feel like replacing the o2 sensor itself... Its like 65 bucks