any improvement when run straight at 12v? whats different?
Identical. is the fan air flow affecting anything? wire shake/air flow elsewhere/sensors? I've, through trouble shooting, have moved and tied down wires every where in multiple places with no effect. wish u were close enough, i'd come over and try to help. Me to. I need to see more details about how the fan speed control mechanism is wired. Evidently it is the speed control that is causing th noise, not the motors. The noise is probably coming from the speed controller producing fast edged pulses of 12 volts to the motors. The ferrite beads might work, but they would have to be placed near the controller to contain its noise. And it may help to put a capacitor across the 12 volt supply lines feeding the controller, so its pulses don't bleed back into the 12 volt distribution. No, the beads didn't help. They are about =BD inch or less from the controller. I don't know how the controller works. Its in a sealed metal box. I still don't know what type or size capacitor to use. I'm clueless on capacitors or what the actually even do.
OK, I think I have it visualized as to what you're doing. You have a PWM drive and it isn't connected directly to the terminals of the motors ? I don't think so. There are several wires going into the controller. Two are heavy wires (+ and -) from the battery to power the fans I'm guessing. There is also one other +/- that I think is the PWM contollers power source.
Assuming this is correct, in effect, you are radiating strong EMF waves under the hood. Adjacent wiring is becoming an R.F. receiving transformer and there for, influencing the lines of other equipment.! Especially things that have low current sensor transmission lines will mostly be whacked. I don't understand why this is only on low. What you need to do is have the output of the PWM circuit pass through an inductor circuit with a smoothing cap on the other side. I'm clueless about electronics. What you said just went way over my head. for proper size selection, knowing the carrier frequency of the PWM circuit helps. I have no Idea. This is the manufacture: