The application might be for slower speed positioning, with the linear sensors used for the angle. (my guess)
Filtering instead of shielding may demand a whole lot of really good luck, in addition to some serious skills.

 

A current-mode signal, such as the industrial 4-20mA current-loop is relatively insensitive to voltage noise pickup, so often doesn't need shielding as compared to a voltage-mode signal, although the wires are usually twisted pair to minimize loop area and noise from that.

 

Most 4-20mA sensors terminate either in a current input of a device or in a 250 ohm resistor,to give a 1-5 volt signal Back in post #6 the TS mentioned a 1 meg input resistance, so the current loop does not seem likely. And while the current loop scheme is certainly quite noise resistant, I have had to correct errors made by those who did not understand at all what makes it noise resistant. Another incident of the "Jim Moore Syndrome".

 

That seems very unusual that linear sensors are used for rotor position (DRV5055) !!?

It's likely using sinusoidal field drive and balancing. Our "maglev" motor systems all used analog positioning with sinusoidal drive signals.


https://www.mdpi.com/1424-8220/23/13/6168

Normal shielded cables for each drive and sensor pair should be all you need if you drive controller signals are clean.

Four maglev pumps on a RF Linac. I've repaired lots of bad cables from pump (mechanical rotor) crashes. Wiring is usually not too critical but the source quality of the drives/sensor signals are. GIGO

 

OK, not production line equipment. AND, so the actual goal is to produce a better control box. Usually that means eliminating whatever was causing a complaint. So now my question shifts to: what needed improving??