jpurban

Hi,

Long post, but this is a complex issue and I've found very little on the subject after hours of searching...

I have a 96 LT-4 Vette with a modified powertrain (e.g. 383, ported heads/intake, etc.). I'm in the process of fine tuning the VE and Spark tables. So, I removed my factory narrowband O2 sensors from my headers and installed a couple Innovate MTX-L wideband sensors in their place.

PCM appears to be receiving the simulated signal properly -- verified via datalogging with ScanXL, PCM voltage varies between 100 and 900 mv on each side. MTX-L "brown" analog2 wires are splice connected to the purple & purple/white wires, the "high" side PCM circuit for the pre-cat sensors on each side. "Low" side of PCM circuit, tan and tan/white wires for the pre-cat sensors, are grounded to the chassis near the battery to complete the circuit (MTX-L doesn't have an analog ground like the 7 wire LC-1).

I have set the Analog2 voltage to correlate with the following AFR:
14.3 = 900mv
14.7 = 450mv
15.0 = 100mv

I'm continuing to set DTCs P1133 and P1153 -- Insufficient switching. I've seen these codes almost every time I retuned with the factory sensors, the tune isn't quite perfect and the LTFTs haven't had a chance to adjust. However, I'm now setting these codes more often.

I understand wideband sensors are not "switching" sensors, like narrowband, but I've read others on the forum have founds settings that work for simulating narrowband in a way acceptable to the PCM.

The car operates incredibly well in closed loop with the simulated narrowband signal -- probably because the signal is calibrated and more accurate. So, I don't want to go back to the factory sensors.

Any advice for making the PCM happier with this signal? Or do I stop worrying about making the PCM happy and disable these codes (e.g. no error reporting)? If I disable the DTCs, will that keep the car from entering "open loop" and force the PCM to continue using the signal in closed loop?

Thanks for the help.

jpurban

Update...

I seem to be having better luck with a tighter, more extreme AFR vs. voltage curve. The tighter the curve, the longer between setting codes. Have yet to set a code with the following settings:
14.58 = 900mv
14.70 = 450mv
14.80 = 100mv

The steeper slope seems to induce more "artificial" switching as the PCM chases 14.7 with different fuel trims.

TastyBacon

Hey jpurban, thanks for posting this. I've been throwing P1133 occasionally on my C5 with a wideband in one bank, and simulated narrowband output from the MTX-L fed into the computer for that bank. The car runs great in closed loop, but I can see on my logger that the simulated narrowband output doesn't switch as fast as a real narrowband sensor. I'm waiting for a USB->RS232 adapter to show up so I can program these settings into the MTX-L. Hopefully that will make the PCM happy by increasing the number of transitions it sees.

jpurban

Congratulations on incorporating wideband -- It really is a critical component to proper tuning.

The switching DTCs are purely emissions related and can be disabled without adversely impacting the closed loop control. This is the method I eventually used to solve the problem permanently.

I used tuning software to disable the switching checks. Once disabled, the PCM would continue to operate normally without regard to switching frequency (limp mode avoided).

C5 is a little different from C4, but I assume the same approach would work.

TastyBacon

Thanks, I can just disable the P1133 DTC too. I'm not too worried about it, since the car runs very well in closed-loop as-is. Do you think it's still worth tightening the AFR vs. voltage thresholds, just so that the MTX-L output looks a little more like a real narrowband to the computer? Not sure if it would help, hurt, or make any difference at all.