divil

I want to log the output of my wideband sensor, but I noticed that the analog output is very noisy. I had a look at all the wires on a scope and it seems that the noise is present on the power and ground wires too. It seems to be caused by a PWM signal that's part of the sensor harness. I think this is the heater circuit. Whenever it switches, I see spikes of up to 50v. If I unplug the sensor harness from the controller, then the power supply traces are nice and clean. As soon as I reattach it, the noise is back.

I contacted AEM and they said I could send the sensor and controller back to them, but there would be a $50 inspection charge, plus shipping. I presume that won't cover the cost of fixing anything either. The thing is, a new sensor is only $50 - but how likely is it that this issue is a bad sensor vs a bad controller? I'm tempted to just buy a new sensor and see if that fixes it.

Has anyone else experienced anything like this? Could the controller be causing an issue like this?

Adker

Sounds like a ground problem to me, you should not see 50 v spikes on the ground lead. Power/ground leads should be short in length, good gauge of wire and have a proper connection to the chassis.

divil

Thanks. The 50v spikes are only on the sensor PWM wire, which I think is a heater. The PWM switching seems to be causing the noise on the power, ground and output of the controller. At least the noise im seeing on those traces lines up with the PWM. But its its a few volts, not 50.

Ill post a screenshot later.

Adker

Ground noise should be in the millvolts.

divil

OK so I tried running a wire directly from the battery ground to the AEM controller ground, and that does eliminate most of the noise on the ground wire. It improves the signal a bit, but it still doesn't get rid of most of the noise.

Here's a screenshot of the traces capturing the moment I plugged the sensor into the controller.

Blue is battery gnd/controller gnd
Red is controller 12v
Green is the 5v analog output of the controller
Yellow is the PWM signal from the sensor harness

Since this is with the controller grounded directly to the battery, I don't know what else I can do to improve it. Any suggestions?

Adker

Try connect the controller 12V to the battery, if it helps, adding a 1 - 10uF cap (50Vdc) power to ground should help the sag in the power supply voltage which matches up with the start of the analog noise. Put it as close to the unit as you can, called bypass cap. You could also add a 0.01 cap to the analog signal (other end ground) to help smooth it out.