Hello friends, longtime listener first time caller. I've recently supercharged my Miata, and am experiencing AFR oscillations in tunerstudios. I am using an AEM 30-0300 controller with a Bosch sensor; along with an MSPNP-Pro. I have searched the forums, and this is what I've tried to remedy the situation:

checked wideband calibration in TS
rewired the sensor output directly to the optional connector on my ecu, using sensor ground and analog in 2.
given controller its own dedicated power and ground, both good and clean connections.
confirmed I am using the analog out wire on the sensor controller, and not the other data out wire.

I understand that the ECU almost certainly refreshes faster and has less smoothing than the gauge, but I see much more drastic swings in TS than I do on the gauge, and it "feels" like there's more at hand here. Hoping someone could shed some light on what I'm experiencing, and what my next steps might be. Attached is two logs, one showing oscillations and one from before, not oscillating. This issue was not present prior to supercharging the car, but I didn't touch anything sensor side when supercharging, so nothing has actually changed.

Thank you for your time.

 

I doubt it's the case but you might double check that the gauge didn't somehow get into narrowband mode.

It'd be a bandaid, but if you just want to get the car out for a drive you might be able to get away with some filtering on the gauge in your tune.

 

Good point, on the controller settings. I'll check that. It's definitely drivable luckily, the controller reads a safe AFR still, and the average of the ecu looks reasonable.

thank you!

 

I have wired the gauge via canbus now, just to test whether it was a voltage offset or calibration issue. The problem persists. I believe this points to a sensor or tuning issue, and not a wiring problem. If anyone could speak to the sensor or my logs it would be greatly appreciated.

 

I just finished up dealing with the same problem, I had to replace the sensor with a new one(Napa p/n 17025) and if you have a ms3 or mspnp pro, ground the sensor through one of the high current outputs on the ecu. The documentation for the wiring can be found on the diyautotune wiring diagram page for wide bands

 

Awesome, thank you. I did see the suggested wiring configuration, but thought I knew better than the engineers for some reason. Was it a sudden issue for you? I can't figure out what would cause this, and it seemingly just happened one day after a month or two of not having driven the car. the sensor is very new though, I'm sure it has less than 1000 miles on it.

 

Solved: I spoke with AEM. They seemed immediately aware of the issue, and somehow knew I had received one of the newer AEM FAE sensors instead of the Bosch LSU. They kindly offered to send me a replacement, but I didn't want to wait so I bought a Bosch 17025 from the local parts store. Problem solved. If anyone reading this in the future is suffering from an oscillating o2 reading in tunerstudios but not on the gauge; it might be worth replacing the sensor.

 

I used to have an AEM wideband setup a while ago. I couldn’t get it to work properly for more than 20 minutes, despite replacing the gauge/controller 2x and the sensor 3x.

AEM has a bit of quality control problem. They can still design perfectly good stuff, but their qc sucks. So a lot of defective parts get sent to customers. I’m not saying too ditch it and switch to a different companies wideband. If you can get it to work then use it, don’t get rid of it unless you can’t get it to work properly at all.