OK, here is the solution, very commonly used in a successful commercial product: Use a twisted pair

Had thought that myself. But unsure if that would make a difference I didn't say anything. I remember pulling a twisted pair of wires from the cockpit to the wingtips for some sort of sensors that were mounted on each wing tip. I asked why they had to be twisted and it was explained to me that the twisted pair prevents interference - or something like that. (back in 1978) The number of twists were specified with a tolerance. While pulling the wires through clamps that would tend to push the twists together and cause a rejection of the insulation. We had to undo every single clamp and put the wire inside it then refasten it. It turned out to be a miserable job. But yeah, a twisted pair might just solve the problem. I don't know it WILL or WON'T, but it's worth a try if your wires are not already twisted.

 

Use a twisted pair

If the question is about leakage inductance, then there are many ways. I have built transformers with Primary/Secondary as a twisted pair.

 

If the question is about leakage inductance, then there are many ways. I have built transformers with Primary/Secondary as a twisted pair.

Twisted pairs in a transformer are mostly to provide the minimum possible leakage inductance.

 

I asked why they had to be twisted and it was explained to me that the twisted pair prevents interference -

Untwisted conductors that carry significant current can radiate EMI. Twisting together can cancel this effect to a great degree.

 

Untwisted conductors that carry significant current can radiate EMI. Twisting together can cancel this effect to a great degree.

That is exactly why the four pairs in the data cable are individually twisted. So that they do not radiate into each other.