car: CRX, b16, inlinepro, T3/T4 .63/.60 turbonetics, apsx D2 wideband (I believe lsu 4.2)

I finished welding up my downpipe last night... wow, there are some crazy little angles in there. I installed the sensor back before the CAT right under the firewall. I believe it's around 36" from turbo at about the 1:00-2:00 position. I'm not getting any accurate readings. It's showing very, very lean (stays on 19.0) Now, before I went turbo, I installed it on the stock 4-2-1 header to test it out. Calibrated it outside, set the fuel type to gasoline on the gauge and was getting good readings. Is it too far away or do I simply need to take the sensor back and and calibrate it again? The car runs good, idles great, doesn't ping, etc.. Turbo feels good (except above 5psi - I haven't gone there yet until tuned 100%). I'm using a base tune I made and it should be running a little rich if anything.

 

Are you still running the factory o2 sensor?

Also if you unplugged the battery for a long period of time when installing the turbo then the wideband's calibration reset and it likely out of cal.

 

No, factory o2 is not being used. I am currently running a tune that ignores the oem o2 sensor(s) - Car is running entirely on the fuel tables. As far as the wideband sensor, it's the one included in the kit which is a bosch lsu 4.2. I was under the impression that the gauge unit had a memory which didn't depend on power to retain the settings. Is this incorrect? Because when you turn the car off, it kills the power to unit altogether anyway. It only receives power from when the ignition wire is on. That's the way I was told to wire it.

 

right before the cat is pretty much stock location, that should be fine.

 

Guys can tune cars with tail pipe sniffers (sticking it in the muffler). The only difference here really is how long it takes the wide band to see changes. I would verify the calibration is correct first.

 

18" away from exit of turbo is 100% ideal, to avoid the delay of readings.

 

Did you change the MAP sensor, injectors, fuel pump or tune when the turbo went on?

If the car's battery was unplugged long enough for the wideband to lose its capacitive power then it lost its calibration and needs reset.

Also I don't understand why so many people remove the factory o2 sensor. At my shop I always reinstall them or wire the wideband to the Hondata S300 cars and use that for closed loop. It can only help the car run better.

 

Running walbro 255, using dsm 450's. I'm not using factory o2 because it gets in way of tuning. It's always trying to adjust so I have my chip set to currently ignore it and it's not in the car anway. I use BRE software and Eprom programmer to make and burn the chips I need. I had it tuned on N/A on the 450s's and the wideband was working great. Idk what's going on. I reset it, recalibrated it open air, still not reading right. It stays on 19.0. I've seen it dip a little under light boost down to 16, but something's off.

I don't know for sure, but I don't believe the wideband units rely on power to retain settings. They are wired in such a way to be totally off if wired to ignition wire, yet they retain the settings.

 

Car would run like complete *** if it really was that lean all the time. Soak a rag in brake cleaner and cover the wideband sensor with it. Should read rich.

 

They rely on capacitive power to store settings but that doesn't matter since you've recalibrated it. If you select the entire fuel table and increase it by 28.5% does that fix the ratios?

You could just go to AutoZone and get a Bosch 17014 wideband sensor for like $60. That's the sensor that comes with your gauge.

 

I appreciate the help. I'll attempt the rag in brake cleaner idea and if I have a problem, I'll get another sensor. Found them for $45 on ebay and 50 on amazon (Bosch 17014)

I'm running a different battery in the car, same grounds etc... I'm thinking about running the wideband ground back to battery. Also, I had to do some welding on the downpipe while it was on the car. The sensor wasn't connected to echaust, but was tucked in the engine bay with a cap on it. I wonder if welding could have done something.

 

I just heard back from a tech rep from apsx and he said that the unit stores the settings without power. Also, I wrapped sensor in fuel soaked rag and immediately got 9.0 (rich) on the gauge. Hmmm, maybe o2 is too far back

 

Ground the wideband to the thermostat housing. Recalibrate then try it again.

 

X2 Always a good idea to ground your wideband to the thermostat housing, as that's were the ecu is grounded. This is especially important if you are logging afr through your ems or using the wideband for closed loop operation.

 

Ok, I'm narrowing this down. I found about a 2" unwelded section on my downpipe where the little flex joint is. I did most of the welding outside in the dark after midnight so... whoops. I welded that up and welded another o2 bung where the downpipe goes down. It's probably 12" from turbo. I did a little researching and found several turbo cars from the factory have o2 sensors this close (VW) so it should be ok (aren't they using widebands?) Either way, if it's a problem, I can later put the factory o2 in this location and my wideband a little behind. But I'm getting good readings now. The tune I'm using is in fact a little lean, but mostly above 5psi. Shouldn't they be around 15 at idle? Mine are a touch higher (leaner). I'm just happy I can finally start tuning this baby!

Question: do the little stainless steel flex pipes leak air? They surely cannot be totally sealed so I wonder if this contributed to the o2 not reading right. IDK, it seems the turbo would have enough force to keep the exhaust somewhat pressurized if anything. (?)

Pic of my downpipe: (yes, the welds were done without gas so go easy on me). I'll get it grinded clean, painted, and wrapped

 

They do not leak until they fail, in my experience. Failure results in audible exhaust leaks on the cars I see for safety inspection.
Usually from impact damage, but I guess after alot of use and flexing it could fail.
The good ones have inner cores of different sorts of tubing, btw