Variable Frequency Drive Noise Reduction

[rgsparber]

My Variable Frequency Drive (VFD) powers my mill's spindle motor. The mill is controlled by a Computer Numerical Control (CNC) system. As I used my mill in CNC mode, it quickly became evident that noise from the VFD was crashing the CNC. This article explains how I stopped the crashes and why the fix worked.

If you are interested, please see

https://rick.sparber.org/VFD_Noise_Reduction.pdf


Your comments are welcome. All of us are smarter than any one of us.


Thanks,

Rick

[metric_taper]

Ground straps that are made up of very fine braided wires can have the opposite results relative to high frequency (and expectations they work). It may appear they are all shorted together, but typically they have oxide, and oil film that insulates them. So these tiny wires with the RF skin depth makes this look like a large impedance. That is the same issue with stranded 12AWG SO cord ground.
Your solid strap is the way to go. Should be interesting with installing motor leads in a coaxial Faraday shield. The only design issue is how to connect this shield with minimal inductance. Typical STP with ground drains (pigtails), will have around 25ohm impedance per inch at 100MHz (22AWG STP signal wire and drain wire gauge).
There probably is some conducted emissions on the 220VAC power to the VFD. I don't know how good the EMI front end filters are in the VFD.
It looks like you fixed your EMC issue.
Very nice write up that should be helpful to many.

[rgsparber]

Ground straps that are made up of very fine braided wires can have the opposite results relative to high frequency (and expectations they work). It may appear they are all shorted together, but typically they have oxide, and oil film that insulates them. So these tiny wires with the RF skin depth makes this look like a large impedance. That is the same issue with stranded 12AWG SO cord ground.
Your solid strap is the way to go. Should be interesting with installing motor leads in a coaxial Faraday shield. The only design issue is how to connect this shield with minimal inductance. Typical STP with ground drains (pigtails), will have around 25ohm impedance per inch at 100MHz (22AWG STP signal wire and drain wire gauge).
There probably is some conducted emissions on the 220VAC power to the VFD. I don't know how good the EMI front end filters are in the VFD.
It looks like you fixed your EMC issue.
Very nice write up that should be helpful to many.

My plan is to feed the braided tube through a hole in a sheet of copper foil and solder it all the way around. The copper will attach to the two ground screws inside the VFD. At the motor end, I will do the same but bolt the copper sheet to the box (paint removed at point of contact). Only problem at that end is that the cover has a gasket so it floats. May need to run a ground wire to it so the cover doesn't radiate.

I know that soldering both ends of the braid will make it painful to remove but should give the best possible connection.

[rbjracing1]

Hi Rick, I get a screeching noise though my radio on some stations when I turn the VFD that powers my knee mill on. I wonder if this would fix that also I will have to give it a try.

John

[rgsparber]

Hi Rick, I get a screeching noise though my radio on some stations when I turn the VFD that powers my knee mill on. I wonder if this would fix that also I will have to give it a try.

John

That would be an interesting test. Just be sure to scrap all of the paint off on the mating surfaces.

Rick

[DIYer]

Thanks Rick! We've added your VFD Noise Reduction to our CNC category, as well as to your builder page: Rick's Homemade Tools. Your receipt:

VFD Noise Reduction by Rick Sparber

tags: CNC, electronics