Adding WBO2 to 1227747

[RobertISaar]

the outputs of wideband controllers are 0-5V, the O2 sensor amplifier inside the ECM/PCM is what takes a 0-1.x range signal and converts it into a 0-5V signal, which is then read that way(since the A/D converter works on a 5V range).

it is kind of over-complex upon first glance, but it is done to increase sensor resolution. with a narrowband's 0-1.1 volt signal being converted to a 0-5V internally to the ECM, narrowband O2 sensors can have a resolution of about ~4.4mV. if read without the amplification, then resolution would be ~4.5 times that amount(or ~19.8mV).

documents from GM specific to the C3 ECMs show the narrowband as being 4.34mV resolution. a full range(0-255) would be 0 to 1.1067 volts.

[mmigacz]

Alright, this is starting to make more sense. Attached is a screen shot of an example where the wideband gauge would be reading 12.49 AFR (.853 lambda).

Wideband conversion.jpg

Does this look right? Would the ECM be reading 1.276 volts (which is approximately 1.25 volts) after amplification?

Here is my next question. I noticed in the acquistion file that it multiplies the value from the ECM by 4.35. See below:

ADX screenshot o2.JPG

There has to be some other conversion going on between the voltage amplifier and what is being sent through the data stream?

[RobertISaar]

the reasoning behind the conversion in the ADX is because of the range being dealt with. since we're dealing with an 8-bit 5V system, a normal 0-5V signal gives .0196 volts per A/D count(5/255). now, since the O2 amp does a multiplication of 4.5, the .0196 volts per A/D count becomes .0196/4.5 = .00435555. there are 1000mV per volt, so .00435555 X 1000 = 4.35555mV per A/D count.

your math looks correct. .026 volts of error is pretty insignificant and something that could be taken care of in the ADX conversion if it were necessary.

[mmigacz]

Thanks Robert! I really appreciate your patience in showing me how this works. I think I understand exactly whats going on. I revised my spreadsheet to show more of the conversions in what is happening prior to being datalogged. I also changed ADX conversion factor formula, so Lambda would be displayed instead of (0-1.1V) O2 millivolts.

wideband lambda.jpg

Originally the ADX conversion was:

ADX1.jpg

I would change it to:

ADX2.jpg
ADX3.jpg
So Lambda would be displayed when datalogging. Is there anything I missed?

[RobertISaar]

other than calling an item "digital volts"(A/D counts would be a proper term), i think all of that works, but i'm doing it mentally since i don't have time to sit down and run some examples through a calculator myself.