Power Quality Considerations for CNC Machines: Grounding, or why you don't add a ground rod at the machine

[tom baker]

From the Electric Power Reliability Institute, this is an well written and correct research paper from that discusses recommended practice for CNC grounding . There is an interesting statement on page 10 "these practices be followed....and they are not based on myths about ground loops or misconceptions about the need for an isolated ground".

I sent this report to Mike Holt about 20 years ago and it became the basis for this graphic.

 

[Joe.B]

Interesting. Thanks for sharing. I've watched, read, and confounded over the concepts and apparent contradictions related to this. There have been a few threads about this here, one very recently. It seems that there is still no clear answer and as with many things, "It depends." So to summarize: The CNC manufacture's/designers are looking for ways to clean up the "noise" in the electrical system in order to make their machinery perform better under normal operating conditions, but the solution can cause a potentially catastrophic failure under extremely rare circumstances. Specifically, adding a local ground rod at the CNC can clean up the electrical system "noise" and will help the machine run better unless you happen to get a nearby lightning strike that can cause induced voltage to travel through the delicate circuitry, causing failure. That about right?

 

[LarryFine]

The risk isn't due to an additional ground rod, it's due to that rod not being directly connected to the rest of the GEC system.

 

[electrofelon]

Specifically, adding a local ground rod at the CNC can clean up the electrical system "noise" and will help the machine run better

I don't see how. There is a fundamental flaw in logic that has developed here. Proponents of the "cnc ground rod" seem to put the burden of proof and explanation on the opponents, wanting us to explain why it will NOT do any good, when they should be ones explaining how it will do any good.

Another thing: according to the article, nearly all manufacturers recommend or propose a ground rod. Do you really think the people who work at say, Haas, really understand grounding? They may be good at designing and building CNC machines but I don't see why they would know anything about earthing really. They should just stick to building CNC machines and stop making recommendations on things they know nothing about.

 

[zbang]

There's also the view that a properly designed & built machine wouldn't need "additional grounding" to deal with fuzzy power; very similar to the view of many audiophile power systems.

 

[electrofelon]

There's also the view that a properly designed & built machine wouldn't need "additional grounding" to deal with fuzzy power; very similar to the view of many audiophile power systems.

Yeah I would hazard a guess that any issues would be because of bonding/shielding issues from inductive and capacitive effects. I don't see how a ground rod would do anything whatsoever.

 

[Joe.B]

Yes there is a lot of gray area here. Did somebody at some point suggest that a better design would be to provide a local transformer to truly isolate the power supply for the CNC, would an independent ground function better in that scenario? I don't know enough to claim that it would or would not, just curious. Thanks for the great topic.