I'm currently trying to bench test an ECM and apply the O2 sensor voltage with a power supply. Using TunerPro RT to datalog at the same time just to verify readings I can get the various temperatures I need with a voltage divider circuit, but the O2 sensor readings are stuck at 450 mV. Does the ECM need to have the "O2 sensor ready" setting correct before it will check the voltage? Is there a way to trick the ECM so it will read this voltage? Many thanks in advance!

 

A voltage divider that is between 1V and ground should do the trick. On The Jimstim I use for bench testing there is a 10K pot connected to ground on one side, the other is connected to %V from the ECM through a 39K ohm resistor and the wiper is connected to the O2 input of the ECM.

It may just be the values you are using for the voltage divider are right inline with the base voltage that is set by the ECM, right at 450mV. With nothing connected to that pin the ECM will read 450mV.

It's been a while since I bench tested an ECM, but I'm sure that the O2 voltage changes will be displayed on through the ALDL connection before closed loop is enabled.

 

Using the power supply directly, I was putting in 0.49 to 0.51 V and saw 450 mV. After that I adjusted the voltage from 0 to around 800 mV just to see if the value would change and didn't see anything different. When datalogging with my car, IIRC, I can see the voltage swings even without the O2 sensor being ready.

Is that 39kohm a pull-up resistor for the ECM or is that for the jimstim? I was thinking maybe the internal resistance of the power supply wasn't high enough, but I'm just not sure if that would effect anything. As long as the voltage applied was within range of a narrowband O2 I thought I would see a readout on the screen.

 

Be sure to apply ground to the O2 sensors other input of the ECM. Pin E15.

RBob.

 

Yeah, that won't work, at least not if I'm understanding what you did there.

Look up JimStim schematics, or maybe even the Megastim schematics to see how it is set-up. I know that Jimstim circuit works perfectly.

 

Make sure that the engine is running, some masks won't read/report O2 sensor value unless it is.

Of course the ECM could be bad...

RBob.

 

Are you trying to see if the o2 is good or bad?

 

So the overall goal is to get an Arduino to interface with the ECM for our wireless sensor network. Realistically we could have made our own ECM and minimized the interfacing needs but its a senior design project and has a limited time frame. (You might remember a post back in August of last year where I took a survey of people willing to use wireless sensors in new or used cars)

Right now I'm testing the ECM in the lab and just providing the same inputs to the ECM that the Arduino is capable of outputting. I looked at the Jimstim schematic and attached it below. It looks as if the 5V drops across the 39k to give it roughly 1V across the 10k pot. I see the wiper position goes to pin 15 on whatever header they're using, which I assume is the ECM input. I'm doing virtually the same thing but I can't get anything on TunerPro other than the 450 mV. The "engine running" could be my problem, though I don't have the capabilities to input a signal the ECM could use to believe the engine is running.

Unless I could put in a square wave or sine wave input into the distributor ref connection? Will this be enough to tell it the engine is running?

Thanks for the input, everyone. It is definitely appreciated!

 

A 0 to 5 volt square wave on REF+ at about 100 Hz will have the engine running. IIRC, that is 1500 RPM. This will cause as SES light with code 42, but the ECM will still see the engine as 'running.'

Be sure to ground the REF- pin of the ECM, same as the O2 sensor ground input to the ECM. These are differential inputs, with the O2 sensor one being isolated.

RBob.

 

After having no luck, I just tried a sine wave input and increased the voltage to about 2.2V. The engine speed jumped up in tunerpro and sure enough O2 voltage started changing. Then I saw Rbob's comment and realized a was close. Next time I'll use a square wave and a little higher voltage if need be. I appreciate everyone's help!