I am kind of suspecting that my single channel AEM WB is on the fritz. It just doesn't seem to behave in a rational manner. I'm about to drive it around for a bit and see.

I have an obiwan link ecu.

Questions:
*How can I tell whether it is working properly? It would help if I had multiple wideband sensors so that I could check them against each other, but I don't.
*if it isn't working, will an Innovative LC-1 work? I don't see why it wouldn't so long as I give LINK the correct target lambda values. I have all the conversion tables from when I wrote my datalogger, with full interpolation out to like a zillion digits.
*is this a good deal? 139 bucks for the controller with no sensor (I now have 2 bosch sensors lying around, I figure at least one should be good). http://www.mphparts.com/prod/103-3744.html

 

No one has any advice?

 

Did you just hook it up? Or was this an old installation that is just starting to malfunction?

When I first installed mine. I didn't hook it up to the ecu ground on the head. It was all over the place. Once I hooked the ground to the head. I was good.

Otherwise I would call AEM.

Good Luck.

 

It's a 3-4 year old install. I bought it in early 2004 I think. It's been fine.

I let the car sit for about a year while I did various mods (cooling and ignition related)

The sensor seemed to work OK when i ran the car but it soon started to behave weird. I thought it was a bad sensor, but it apparently isn't. I think it's a bad controller, but I want to be sure before I spend 130 bucks on an LC-1.

Has anyone else used an LC-1 here?

 

Can I reuse the sensors I have from my AEM on the LC-1?

 

Just thought I would post the resolution of this problem.

Turns out the WB is fine. I guess I must have knocked something loose, but after checking all my connections with a multimeter, it started working fine again. One interesting thing I noticed is that the narrowband simulator is not actually a 0-1v signal. It uses .5v for lean instead of 0v.

The really cool thing is that westach offered to recalibrate my gauge if I would send it back with a map of voltage to AFR that the AEM puts out. So this way I will actually get a functioning gauge that reads from the narrowband signal of the AEM. My testing with the voltmeter shows that it is actually an accurate signal that matches the wideband output. Not particularly high resolution, but high enough to warn of anything catastrophic happening.

 

And in the future, I will go with an LC-1. Configurable outputs and auto-calibration of sensors is worth it's weight in gold.