Lean running at low load B18CR
I have a recent problem that I can't figure out, hopefully someone here has run into this.
Recent B18C swap into EF running on a virgin JDM P30 small case ECM, CAI, header, exhaust, adjustable FPR.
When off idle but at low load(approx. 10-30% throttle) my wideband O2 shows extreme lean (>18:1 air/fuel ratio), if left lean for a while it will start to misfire a bit. If you give it more than about 50% throttle it goes back to normal 14.5-15.2 A/F ratio. Wide open throttle is just fine at 13:1 A/F
It previously ran perfectly with a near stoic mixture until you really got into the throttle.
I have checked fuel pressure, smoke tested for vacuum leaks, swapped TPS and MAP.
I jumpered the service connector and grounded the yellow wire on my ECM adapter harness but only get a solid "check engine" light.
My narrow band O2 sensor is only as old as the swap(about 4 months old) and I don't have a spare to try...yet.
Any suggestions?
Recent B18C swap into EF running on a virgin JDM P30 small case ECM, CAI, header, exhaust, adjustable FPR.
When off idle but at low load(approx. 10-30% throttle) my wideband O2 shows extreme lean (>18:1 air/fuel ratio), if left lean for a while it will start to misfire a bit. If you give it more than about 50% throttle it goes back to normal 14.5-15.2 A/F ratio. Wide open throttle is just fine at 13:1 A/F
It previously ran perfectly with a near stoic mixture until you really got into the throttle.
I have checked fuel pressure, smoke tested for vacuum leaks, swapped TPS and MAP.
I jumpered the service connector and grounded the yellow wire on my ECM adapter harness but only get a solid "check engine" light.
My narrow band O2 sensor is only as old as the swap(about 4 months old) and I don't have a spare to try...yet.
Any suggestions?
Instead of using a narrow band and your wideband I would probably run the analog out from the wideband to the PCM to act as your narrowband. You say the sensor is only four months old but is it a Honda Genuine O2 sensor? I've seen aftermarket sensors cause similar issues.
-Amp
-Amp
Instead of using a narrow band and your wideband I would probably run the analog out from the wideband to the PCM to act as your narrowband. You say the sensor is only four months old but is it a Honda Genuine O2 sensor? I've seen aftermarket sensors cause similar issues.
-Amp
-Amp
I'll get onto that to see if that makes any difference.
If it's a stock narrowband on a stock ecu it should be fine. Check and make sure the sensor is good (if you can't remember it's easiest to replace it with a new oem one). Check for exhaust leaks upstream of the sensor.
Entirely replacing the stock sensor with the simulated narrowband on your wideband will generate a CEL since there is noheater circuit. You might be able to get around it by only using the signal pin and leaving the heater controls hooked up... I've never tried this though. .
Note that misfire manifests itself to a lambda reading as a false lean condition.
Change your fuel filter for good measure.
Entirely replacing the stock sensor with the simulated narrowband on your wideband will generate a CEL since there is noheater circuit. You might be able to get around it by only using the signal pin and leaving the heater controls hooked up... I've never tried this though. .
Note that misfire manifests itself to a lambda reading as a false lean condition.
Change your fuel filter for good measure.
If it's a stock narrowband on a stock ecu it should be fine. Check and make sure the sensor is good (if you can't remember it's easiest to replace it with a new oem one). Check for exhaust leaks upstream of the sensor.
Entirely replacing the stock sensor with the simulated narrowband on your wideband will generate a CEL since there is noheater circuit. You might be able to get around it by only using the signal pin and leaving the heater controls hooked up... I've never tried this though. .
Note that misfire manifests itself to a lambda reading as a false lean condition.
Change your fuel filter for good measure.
Entirely replacing the stock sensor with the simulated narrowband on your wideband will generate a CEL since there is noheater circuit. You might be able to get around it by only using the signal pin and leaving the heater controls hooked up... I've never tried this though. .
Note that misfire manifests itself to a lambda reading as a false lean condition.
Change your fuel filter for good measure.

To the OP, best thing to do is source OEM parts.
-Amp
Trending Topics
OP saying a "recent" issue meant to me it was previously running okay and this began to crop up with no other changes. I missed the part where this is a new swap.
Honestly, at the throttle angles/speed ranges he's talking about a P30 should run pretty acceptably, and the P30 is about the closest stock program to a B18CR. Obviously and preferably it would be better off with a tune.
Honestly, at the throttle angles/speed ranges he's talking about a P30 should run pretty acceptably, and the P30 is about the closest stock program to a B18CR. Obviously and preferably it would be better off with a tune.
The car did run perfectly previously. I just used the adjustable FPR to tune the A/F at WOT to 13:1 initially. The rest of the time it ran very close to stoic.
It just suddenly started to run lean at light load.
I was looking up which wire to use as output from my AEM wideband and realized the analog output is 0-5 volts and stoic is 2.38V. Narrowband O2s are usually 0-1V. so I don't think I can substitute the Wideband analog for the narrowband output :-(
I'll try to measure my narrowband output when lean with a scope at lunch today.
It just suddenly started to run lean at light load.
I was looking up which wire to use as output from my AEM wideband and realized the analog output is 0-5 volts and stoic is 2.38V. Narrowband O2s are usually 0-1V. so I don't think I can substitute the Wideband analog for the narrowband output :-(
I'll try to measure my narrowband output when lean with a scope at lunch today.
I was looking up which wire to use as output from my AEM wideband and realized the analog output is 0-5 volts and stoic is 2.38V. Narrowband O2s are usually 0-1V. so I don't think I can substitute the Wideband analog for the narrowband output :-(
I'll try to measure my narrowband output when lean with a scope at lunch today.
I'll try to measure my narrowband output when lean with a scope at lunch today.
A narrowband sensor still ha a 0-5V signal, it's just that it doesn't have a nice linear slope to it. It only works right around lambda (in a narrow band!) and is really just a rich/lean switch as such. Further more, a wideband sensor is scaled oppositely from a narrowband. The wideband's voltage output normally increases with increasing lambda (more lean) while a standard HO2S is high at rich and low at lean. That being said, the PHO2S input on your stock ecu is scaled to 0-3.7V, so if later on you did want to use that for your wideband input you would need to clip a component on the board to make it into a 0-5V scaled input.
Wideband vs narrowband outputs:

Not quite how it works but no, you can't just substitute your wideband's output for the narrowband. Some wideband controllers have a simulated narrowband output that you can use (it's a separate wire) but that won't help you here and they're only really useful if you don't have a narrowband at all and are able to shut off the heater error in your cal. As long as you have a stock ecu you should use a stock sensor.
A narrowband sensor still ha a 0-5V signal, it's just that it doesn't have a nice linear slope to it. It only works right around lambda (in a narrow band!) and is really just a rich/lean switch as such. Further more, a wideband sensor is scaled oppositely from a narrowband. The wideband's voltage output normally increases with increasing lambda (more lean) while a standard HO2S is high at rich and low at lean. That being said, the PHO2S input on your stock ecu is scaled to 0-3.7V, so if later on you did want to use that for your wideband input you would need to clip a component on the board to make it into a 0-5V scaled input.
A narrowband sensor still ha a 0-5V signal, it's just that it doesn't have a nice linear slope to it. It only works right around lambda (in a narrow band!) and is really just a rich/lean switch as such. Further more, a wideband sensor is scaled oppositely from a narrowband. The wideband's voltage output normally increases with increasing lambda (more lean) while a standard HO2S is high at rich and low at lean. That being said, the PHO2S input on your stock ecu is scaled to 0-3.7V, so if later on you did want to use that for your wideband input you would need to clip a component on the board to make it into a 0-5V scaled input.
I am doing exactly what I've highlighted in bold though and it works flawlessly. I have only a wideband sensor with the analog out to the PCM and my vehicle runs open loop/closed loop like factory with no other sensor. Granted, my PCM is chipped and I do have the heater circuit turned off.
-Amp
Thread
Thread Starter
Forum
Replies
Last Post
01 LudeDude
Honda Prelude
3
Feb 9, 2008 01:34 PM
LOGONZOJLS
Forced Induction
3
Jan 29, 2007 02:26 PM








