Question Noisy hall sensor pot data - how to solve? Advice needed.

I have an issue with the noisy feedback from the hall sensors. Essentially I have a DIY frictionless Hall sensor potentiometers and they are working just fine in my 2DOF. However in the 6DOF's first test as soon as I switch on the main power - I get a lot of erratic movements and in SMC3Tools I can see the noisy data. Here's what it looks like with the power on - forward 2 motors (Arduino #1) are the noisiest:


As soon as I turn the main power off - so only the Arduinos are powered by USB - the noise virtually disappears, settling to normal levels:


That is in the state of rest. Once it starts moving it goes haywire just trying to stay in place, the data is just too noisy:


There is also some crosstalk so when one motor moves, the one next to it twitches slightly.

What could be the problem and how do I solve it? I have the Arduinos inside the ECB connected but few plain wires to an insulated VGA cable, where the ground goes to shielding, voltage to one wire and 2 sensor signals - to 2 other wires. On the motor end the wire splits into 2 300mm ribbon cables that get the voltage, ground and signal to/from the sensors. Obviously this is not working well enough. The ground and voltage is provided by one of the Arduinos.

- Is the noise RF interference or is it "dirty" power or ground?
- How can I mitigate it?
- If it's the wires, what kind of wires would work better?

I'd appreciate all advice.

Video of the issue:

Hello, it look extreem… soldering a
104pf 50V keramisch condensator to the output of your pots helps. Between GND and Vout
But I think there is more needed.

Use shielding the cables.

Maybe first check your connections
And make sure your pots are solid. Check the Vout. If it moves then it is no good.
last I know.. check you power supply. You should put your Vout cables of the powersupply through a iron -material 13 - ring a few times (winding). If you have an old monitor cable, you get it from the thickening on the cable. Cut it out. Or buy on internet search for material 13.

these things solved it for me.

Yeah, I solved it by using different wires with independent shielding and 104 capacitors.