OK, I need some help... I am sort of lost here.

The engine is the one in the sig (383, ZZX cam, Miniram, Brodix heads, yada yada). I built this engine last summer and installed it just before winter. I really didn't have time to do much tuning until recently. I also just got my hands on an LM-1 wideband....

OK, what I am running into is BLM readings (using Datamaster) that are either 128 or right close to 128 but the LM-1 is telling me I am around 13.2-13.6... I recalibrated the LM-1 but it is working just fine. I figured this was the case as I can smell the richness and it even smokes a bit on cold starts.

On the cold starts.... I figured I just needed to lower the values in the startup enrichment table (TunerCat) to help with that and it did help but it is still too rich on startup (depends on coolant temp but it is high 12s to 13s on the LM-1.

Even if the BLMs aren't right on 128, if I sustain the rpm/load it will go to 13.5 AFR even if the BLMs are trimmed and sustain a value within the range of adjustments (this should put it pretty close to 14.7 after it is trimmed and if the rpm/load is sustained, right?????)

So I just don't really understand what could be going on here. I appreciate any ideas on this. Follow the links below for a data log from today and the latest .bin file (a Super AUJP based one):

Data log (datamaster)

bin file

Feel free to make any comments on it - I won't be offended!

TIA

 

Are you tuning closed loop with the stock sensor, or from a converter from the WB?.

A slight puff of black at startup seems about normal for some cars.

 

Well, I am trying to look at everything and figure out what is going on. I see that the LM-1 wideband says I am rich. I can also smell that I am rich and observe some smokage on cold starts. But the BLMs read 128s or close at times when the wideband says mid to low 13s for AFR..... So I am confused....

But... I got an idea after talking it over with some friends. Lets say that I adjust the VE and timing tables to get more consistent 128s all over but I am still seeing this 13.5 AFR.... What if I adjusted the stoichiometric constant in the .bin file the opposite direction the same amount it is off from 14.7. So in this case I move it to 15.9 to make up for the 1.2 points rich it is and move to 1.2 points lean....

What I am thinking is this will compensate for whatever is causing my ECM to be thinking that 13.5 is actually where it should be for stoich and move it to an actual 14.7 and get the BLMs in line. The theory we came up with is that the Bosch 3-wire O2 (p/n 13077) I have is off a bit or the manner in which it is mounted or wired is messing it up. I have this sensor in the driver side collector of Hooker coated long tubes with another bung welded onto the passenger collector for using the wide band.....

Does this make sense? Does altering this stoich setting alter it for all resulting calculations like PE and AE fueling?

TIA again.

 

Well, I thought I might try a few things to verify my theory before I make any changes....

So, I got a datalog from my buddy that I borrowed the LM-1 from that came from his 730-run TPI Suburban. I averaged the O2 values for both vehicles just to see if there would be a correlation but it seems there is not. His Suburban will plug along at about 15.1 AFR when the BLMs say it is fine (around 128) but mine does the same thing but it is 13.5 AFR when the BLMs say it is OK (well, not solid 128s or anything.....).

Anyways, I guess the theory on the O2 being the cause is incorrect.... What we were thinking is that the O2 is reading 450mv when it actually is not at 14.7 (as in it is not calibrated correctly) which, in turn was causing the ECM to think everything was fine when it was actually too rich.... But the averaged O2 voltage values for both my datalog and that Suburban one were almost identical (~570mv) and the vehicles read differently on the LM-1 for AFR by quite a bit..... And it is the same exact LM-1....

 

This might sound weird, but try running a ground from the ecm case to the ground on the LM-1.

I've run my LM on a couple cars, and against a couple other WBs and all the numbers have jived.

I've always thought there had to be a reason for all the O2 sensor parts numbers, and maybe there is a difference, and maybe you've found the two extremes of the O2 calibrations.

 

Try running your ECM in closed loop with the Analog Out 1 signal of the LM-1 simulating a NB O2 in place of your factory O2. If your BLM's show lean but the LM-1 shows ~14.7, then you know it is your O2 sensor that is causing the problem. I don't have an extra bung in my exhaust so I tune closed loop with the NB O2 then in stall the LM-1 to simulate a NB O2 and tune WOT. When I do this the BLM's stay right where I tuned them. It has worked well on 2 cars so far.

 

Thanks guys, very good ideas. I will get on them ASAP......

Using the LM-1 as the NB and the WB should remove the problem if it is related to the NB O2s signal.

Grumpy, do you mean ground the LM-1s sensor harness or the control unit to the ECM case? If so, I don't know if I can as the LM-1 is on loan.

Also, I don't kow if I explained this properly or not, but the same LM-1, sensor and all, was used on both vehicles. The NB on the Suburban is also a Bosch but he has to take a look at it and find the part number (or the box, etc.). Both my GTA and his Suburban have long tubes on them and as such both have three wire O2s in them. Mine is the 13077 Bosch but I am not sure what his is at this time.

I guess I had figured that with averaged O2 values from cars reading quite different AFRs in closed loop (idle and cruise I am talking so there should be no AE and PE affecting it) that it wouldn't be a calibration problem with the O2s..... But using the LM1's NB output should answer that question. Buying another NB O2 to run the car would probably help that but I can't see spending $$$ on a new sensor if I can adjust a setting in the .bin to tailor it to my O2 and get my desired AFRs......

 

O2 sensor placement (i.e. exhaust temperature) can have a big effect on where the stioch voltage range is. Check out this article . It shows that the voltage threshold for stoich can vary by as much as 175mV depending on EGT. This could explain why averaging the O2 volts from the 2 different vehicles did not provide the correlation you were looking for. This could also explain why the rich/lean voltage threshold tables are included in the calibration. Modifying O2 sensor placement from stock could definitely cause your problem. This also proves why a WB O2 is such a valuable tool when tuning modified setups.

 

.

 

Sorry bout the lack of replys.... Twas finals week and I didn't touch it at all last week.

I got an idea.... I am going to get a 4-wire O2 from a friend and wire that in and see if possibly I am having an issue with the grounding of the sensor through the coated header or just the sensor body itself. I hesitate to do anything with the LM1 I have since it isn't mine and I don't want to mess anything up with it.

I did run the car yesterday for a little while with a few changes to the PROM but it is no use.... The thing wants 13.5-13.8 to 1 regardless of what I do. I also noted that my free air value on the narrow band is 451mv.... Not sure if that completely ruins the grounding theory or not.....

Crossfire, I'll shoot you an Email with his Email address.

 

OK, I finally got a 4-wire installed a few days ago..... Took a while to get a connector and wire it up somewhat nicely as I decided to try (and have the ability to switch back and forth easily) moving the NB to the other header (passenger side) and switch the NB and WBs around to see if things changed.... Well they definately did. But things are still not making all that much sense.

Now I am getting a lean condition when the BLMs are close to 128s. Not real bad, but in and around the low to mid 15s at idle and part throttle cruise. Another thing I noted is that it seems to act much different based on engine temp (coolant) regarding this issue. When the engine is colder the discrepencies in the readings off the NB (in other words the BLMs) and the actual AFR as read by the WB are greater - it will idle into the 16s for AFR when cold. But once it is warmed up and I have driven it around it will come down to the low 15s to high 14s for idle. Light throttle cruise is much the same except it does not go quite as far into the lean.

I haven't messed with WOT much but it is still pretty far off. It was mega lean but I have worked it into the 13s and will fine tune it at the dyno, along with WOT timing.

I am starting to lean towards ignoring BLMs and tuning for the AFR I need/want. I mean not trying to get perfect 128s across the board that is.

I dunno if anyone has any ideas or anything, but I figured I would try to keep this thread updated on my "progress"....

There does seem to be one thing that I am verifying and that is how big of POSs the NB O2 sensors are and how much they can vary and throw a tune off.....

 

There's a new SAE paper out based on some Ford research showing how *waxy* fuel is at lower temps., and how the spray pattern is really lousy. There's a real change in injector spray patterns at ~130dF. The opening and closing dibbles also change alot. And at low DCs that can make alot of difference.

The cold engine commanded AFRs will rarely match since the engine is in a warm up mode. ie thermostat closed, alt in heavy charge, poor ring seal, oil contamination burning off, high engine oil viscosity, and the list goes on and on, a close engine has nothing good relaly going on, and the intial hot restart isn't much better. All sorts of subtle things that one at a time don't matter too much when ganged up will effect things, and in some cases dramatically.

Glad you got to see what junk a stock O2 really is.