AF stumper

[TurboDog's Dad]

Hi Guys (and gals),
New here, but not new to tuning in general. Running an '02 LS1 in an '02 Miata with the GM computer to match the engine. Have been studying up on all the FAQs, tuning tips, etc, but haven't seen this issue. The car has two stock narrowbands in the downpipes, no cats and a single AEM wideband behind where the two banks join in a 2.5>3" collector. Car has ported heads, fairly mild cam, big throttle body, stock ratio roller rockers and GMPP good lifters. Problem is this: I'm trying to tune the VE table using the narrow band for now, MAF is unplugged. I have PE pretty much disabled by setting enable map to over 80 kpa. Our ambient pressure is only 86 kpa usually. Problem is with LT disabled and STT trims enabled, I'm seeing good STT numbers of +- 2-3, but wideband is consistently about 15.3-15.5 AF, although NBs are supposed to be aiming for stoich. NBs both seem to read well and I swapped between four different widebands tonight, including two brands. Seems like there is an offset somewhere making the car run lean. Any help appreciated. Let me know if I should post the current map. Thanks!

[dimented24x7]

Is there any bias between each cylinder bank? Or are they in good agreement with each other? Are the NBs new stock ACDelcos, original, or aftermarkets?

The slight lean operation may also be due to the calibration itself. The O2 PID is driven by the rich/lean O2 voltage thresholds, which define what voltage the PCM will be targeting. These are not necessarily stoich at all engine loads and airflows. On stock engines its sometimes advantageous to target a slightly leaner AFR to increase fuel economy. The only way to know for sure would be to check the rich and lean threshold tables in the O2 calibration within the PCM. If you have access to these, you can also adjust them to make the PCM target a richer AFR if the engine doesn't like the slightly lean operation.

[pdsq98gt]

Why are you even bothering with the narrowbands if you have a wideband? Put the wideband in place of the bank 1 narrowband and tune away. If you then have fuel trim problems, it's probably due to either the o2 locations or something in the tune as described above.

[TurboDog's Dad]

Thanks, guys. In answer to the second question, I'm still using the narrow bands because I haven't figured out how to get a histogram working showing wideband % correction. I have my wideband added to the table, but it won't let me go to (IIRC) histogram 6 or 7. Ideas?
On first idea- Are you talking about the oxygen sensor R/L vs airflow tables?
Here's what's in there:Labels 0 2 4 6 8 10 12 14 16
Rich/Lean Voltage 366.01439 435.79289 435.79289 435.79289 450.84249 500.91723 500.91723 551.02594 551.02594
I'm starting from a tune somebody else did, so there may be something completely whacked. I've attached current file if the attachment worked.

[TurboDog's Dad]

Anybody? Ferris?

[Eagle]

You really want to spend the time to get the wideband working properly. The narrowband tuning is a PITA.

What wideband is it? Does it have its own logging software you can check the #s from to verify the sensor is working? Have you calibrated it yet?

[TurboDog's Dad]

It's an AEM gauge type. Cal is good, I tried five different ones between two brands. I have it wired into the HP Tuners interface so I can log it. Problem is I don't understand how to get a histogram of WB % against the VE table. I do have the wideband recognized and installed in the HP Tuners, it does show in the table and does read.

[Eagle]

You should be able to just setup a gauge in the scanner then to verify that its collecting data. As for the histogram, you need to create a custom histogram, then select the PID for "Air to Fuel Ratio Error" and set the table to "Primary VE vs. RPM vs. MAP". Set the required cell hit count to the desired level, usually at least 15, some even say higher like 50... then save it.

That should get you in business.

[TurboDog's Dad]

Now I'm feeling dumb. er. I had setup the wideband, but when I went in to histograms, it would let me open 1-6 histograms, but not 7, which is VE and wideband %error. I have the feeling I just had to close out or reboot after installing the wideband in the scanner, as the #7 histogram is now available. Duh. Guess I just need to turn off the NBs now. Thanks for your help!

[TurboDog's Dad]

Would it make sense to build in an offset to the stoich target so NB and WB agree? I would still like to run closed loop.