Please help me understand narrowband output

[daniel76309]

My car is a 1970 Camaro with LS376-480 crate engine, along with GMPP controller (E67 ECM). The only modification is long tube headers. I am using an NGK-AFX wideband O2 sensor.

I have calibrated the MAF such that AFR runs pretty close to Lambda when in open loop.

However, in CLOSED loop it tends to run a bit (3-4%?) rich (and if anything I would like it to be a little lean in closed loop).

Lowering narrow band O2 "switch points" didn't seem to have any effect on AFR until I went all the way down to 80 mv (factory setting was 450mv), and even then was still on the rich side.

I set up a histogram to plot narrowband O2 sensor output voltage against Lambda (i.e. Lambda, as reported by the wideband sensor), and found that when Lambda = 1 on the wideband, O2 voltage is NOT 450 mv as I would have expected, but rather, something like 60mv (and it varies some depending on the log and on which cylinder bank).

Can someone explain to me why this is the case?

Is it perhaps due to O2 sensor placement? (the wideband is in the passenger side collector and the narrow bands are in the header reducers directly behind the collectors)

Or faulty O2 sensors?

Or could it be that my wideband is in error? (I HAVE calibrated it several times to be sure of that part of it.)

Other ideas?

[daniel76309]

OK, follow-up question...

The narrowband O2 sensors are those that came with the GMPP controller package for the LS376-480. If I were to replace the O2 sensors to see if they are the problem, which sensors would I use?

Are these sensors somewhat "universal", i.e. if I got sensors for a newer Corvette or Camaro, would they work with the GMPP system?

[Brandoneaves]

I think a constant heated o2 sensor would do as a wideband would do when heated, produce a lean voltage might be one place to look.

[daniel76309]

Thank you for replying, but could you please elaborate? (I don't understand what your point is.)

[Brandoneaves]

like when a wideband heats up it goes lean until normal working temp then comes back down once there I believe the o2 sensors do the same so if a o2 heater stays on (unsure if they always stay on or not but my thought's would be the heater turns off) it would constantly produce a lean voltage.

[daniel76309]

Thank you, that is an interesting theory, but not sure how I would test that or what to do about it. In any case, you could be on to something.

I think my O2 sensors (both narrow band and wide band) read "lean" at startup, and then function properly after heating up. I suppose it could be that the narrow bands are running a bit cooler than they should be (since they are in the header reducers rather than exhaust manifolds) and therefore are putting out a slightly "lean voltage". But then I would think everyone with long tube headers would have the same problem...

[5_Liter_Eater]

Unfortunately the switch point tables don't seem to function in the 4th gen PCM's. I get what you're saying but 3-4% doesn't seem worth worrying about.

[Brandoneaves]

it would be the opposite over heating a narrow/wideband would cause a lean condition

[Brandoneaves]

Unfortunately the switch point tables don't seem to function in the 4th gen PCM's. I get what you're saying but 3-4% doesn't seem worth worrying about.

I agree with him tho 3-4% isn't bad

[daniel76309]

Unfortunately the switch point tables don't seem to function in the 4th gen PCM's. I get what you're saying but 3-4% doesn't seem worth worrying about.

I didn't realize that. It SEEMED to make it leaner when I lowered switch points but I guess I will need to recheck that.

I am not worrying about it exactly, and the car runs great. I mostly wanted HP tuners to solve the stalling problem (which I did), and at this point am just messing with other things for my entertainment, i.e. tweak something, datalog to see what effect it has, etc. I don't necessarily need to have it run leaner under cruise conditions (in closed loop), but it would be nice if I could, and it seems as if there ought to be a way of doing this.

[daniel76309]

it would be the opposite over heating a narrow/wideband would cause a lean condition

You may be correct--I will have to recheck that also. I THOUGHT that the voltages were low at startup and then they crept up as the engine warmed up, but I could have that backwards.

[Brandoneaves]

no you are correct, and are me rethinking my understanding for that, narrowbands do start at 450mV ish then go up until warm or maybe lower I haven't looked in a while since ive been running with them shut off and using a wideband.

[5_Liter_Eater]

In the third gens the switchpoint tables work as expected. So you can set up a histogram to show what the narrowband output is when the wide band is close to stoich and then use that value. But I haven't seen it work on mine anyway (E40).

[daniel76309]

OK thanks. I am going to do some more experimenting with it and see what I can do. I tried changing switching voltage to 60 mv in just one of the cells leaving the others at 450, and with just one short log there did not seem to be a "zone" of leaness anywhere, so it could be that it doesn't have any effect on mine (E67) either.

[daniel76309]

BTW, my narrow bands seem to have an upper plateau around 850 mv and a low plateau around 30-40. Is that normal?

[ingsteve]

My car ls2 e40 is tuned with wideband 12.5afr...my narrowbands are in 890mv range and sometimes in 830mv range. What doesi it mean? Don't trust to narrowbands...

[5_Liter_Eater]

Doesn't mean anything. Can't trust narrow bands outside of stoich.

[daniel76309]

OK, thanks!

I AM making some progress with this but I still have some experimenting to do. I have gotten it to run lean in closed loop by setting ALL switch points (including max and min) to 50. I would like to have it slightly lean at cruise, but then be near stoich at idle and under load.

PS: I think fixing exhaust leaks helped with this also.