Need Help....Machine Grounding

Hi, I am getting a new to me Milltronics ML-20 Monday. My concern is preparing a proper ground for the machine. This is the existing set-up I have:

My shop is next door to the main shop where the 3 phase power comes in.
From that point the power comes under ground and up into the service box in my shop. There is an existing run of 3 wires (no ground wire) to the spot I want to set the machine.
The landlord says the previous machine that ran there was a large surface grinder that used the conduit as a ground. (I've read enough on here to know thats not a good idea.)
The building has a ground rod in the front left of the building with a wire that runs across the top of the building to another ground rod in the right rear of the building and along that wire span
on top of the building are 3 equally spaced small rods going up from the wire. I assume for lightning protection. The landlord says it was that way when he bought the place.
Any suggestions would be greatly appreciated.

What you describe is a lighting protection system. It has nothing to do with the electrical supply top the building, it least it shouldn't.

Is the 3 phase 220V/240V or 480V? If 480V, there may not be a ground. Otherwise there should be a ground rod and wire at the meter or main disconnect. When I was schooled in electrical systems, every branch circuit that required a ground, was grounded back to that rod with a ground wire. When I first became involved with CNC equipment and saw the requirement to have a ground rod at each machine, I didn't understand that. I still don't, even after 25 years of reading arguments pro and con.

My shop and all buildings that I have worked on over the years are grounded to the ground rod at the meter/disconnect. Never had any strange events that could be traced back to this grounding system.

Here is one of the best video's I've ever seen, done by people who know what they are talking about. There are others in this series equally helpful. While I don't know if it will answer your questions, I'd be surprised if it doesn't.
https://www.youtube.com/watch?v=qNZC782SzAQ
Gus

If this is your first CNC in the building, you will probably want to get your ground network tested by an electrician that knows his stuff. It is not a trivial thing to test it, so unless you are a whiz with electricity and have the tools to do it, it is best left up to the pros.

The controllers and communications to your computer if you have one wired up can get really cranky if you have too high of an earth resistance.

This is a pretty good document by Megger describing the situation of earth resistivity, it's effects and testing it.

http://www.weschler.com/_upload/sitepdfs/techref/gettingdowntoearth.pdf

A few months ago I was in a big shop in St. Paul where there are codes and this plant has union electricians and they were drilling a hole in the concrete next to a new CNC drill grinder. I asked why ? They said they were going to put a ground rod in the floor there as a precaution as the grounds in the wire feed were to small. Not knowing better I said "ahh ok" as I have never seen this before. The next day I came back and there was a copper ground rod with a green wire attached to the machine. I would say it's worth asking a professional electrician about this in your area as there maybe codes you need to follow. Seeing your renting the space. May have to get permission to drill a hole. Rich

If conduit is coming out of floor then conduits are in floor.

Careful where you drill.

A grounding rod close to the machine thru the floor is cheap insurance you only have to pay for once.

Seriously guys, if a ground rod affects the way your machine works you should buy different machines as the people that built it didn't have a clue! Driven earth grounds are useless as they can't possibly carry any useful amount of current. If your electrician is installing them you ought to be wary of their work because they don't have a clue either. I spent 20 years doing industrial electrical work and can count on the fingers of one hand a with a couple left over the electricians I've known that actually knew what they were doing!

One problem with ground rods under large slabs that have been in place for a long time is that the earth drys out underneath the slab and you don't have the grounding that you think you do. This is one of the reasons that the code for grounding has changed in recent years.

I was having some arc stability issues on our plasma table at work. The table and controls had been grounded with a dedicated ground rod as the manufacturer of the plasma table had instructed. Problem was that the ground rod was actually loose in the ground that had been driven into dry clay. Poor install from years previous. I measured the resistance between electrical ground and the ground rod itself and got an open circuit. Connected the ground rod to a column base at a footing pad. This pad went into the ground to 4ft. and was about 3ft. square with re-bar. Re-measured the resistance and was good to go.
End of arc stability issues. Now the resident know it alls were telling me, "But that's not what the manual says to do."

My point is that there is a lot to properly grounding electrical circuits and equipment. Just because you ground to code by the book without checking that the ground system is in fact properly functioning, you very likely might have almost nothing.

One of the many good things about the evolution of controls technology is that the current control technology is much more tolerant of grounding issues than the old technology from years ago when we had multiple + and - supply voltages all present on the control boards.

Putting in grounding rods all over the place is not only wrong, but unsafe. Please understand that ground is not ground everywhere. There is and forever will be potential differences from point to point and that difference is dynamic and changing all the time. The ground connection used within a building must be non-current carrying and referenced electrically at a single point. That point is at the power entrance point of the building and it is called the equalization buss. Grounding at multiple points create ground loops, extraneous currents and is electrically noisy. This is very dangerous to electronics and people. The ground circuit must be an electrically quiet voltage reference and a safety path for faults only. Return current from single phase loads must always be on a separate neutral circuit which can only be connected to the equalization buss (ground) at the building entrance point. Bit serial comm. interface standards that use non-differential signal return lines in the comm. cable, like RS232 and its variances, are especially susceptible to catastrophic and spectacular failures when these little return lines try to normalize ground potential differences.

Above all please don't tie into the lightning protection!!! Thats a real bad bad idea, but if you do, please post the video of the next strike on you tube, would be awesome to see just what lightning grounded to a VMC would do to it :-) Would the enclsure act like a cars - plane and be a viable fariday cage, just how fast would your typical brushed servo rapid at the several million volts of lightining!! I guess probaly fast enough to make even a brother speedio seam slow :-)

IMHO you guys over in the states seam to have some real wacky ideas about grounding and how that relates to the earth, which by and large is the irreverent bit, the idea of grounding is far more about tying everything down to the same voltage plane for the nominal 0V level. Then doing that well enough to ensure all fault currents are high enough to ensure a rapid trip. Meat the requirements for that properly and noise won't be a issue, noise is caused by a fault, because yets face it, if a few micro amps of HF noise can't be brought down to zero voltage because of the earth’s systems resistance, just how well do you think its going to do with a multi thousand amp fault current?

adama said:

IMHO you guys over in the states seam to have some real wacky ideas about grounding and how that relates to the earth, which by and large is the irreverent bit, the idea of grounding is far more about tying everything down to the same voltage plane for the nominal 0V level. Then doing that well enough to ensure all fault currents are high enough to ensure a rapid trip. Meat the requirements for that properly and noise won't be a issue, noise is caused by a fault, because yets face it, if a few micro amps of HF noise can't be brought down to zero voltage because of the earth’s systems resistance, just how well do you think its going to do with a multi thousand amp fault current?

Click to expand...

You are correct about the common voltage plane, but not so correct about noise. Every wire has inductance and reactance and at certain frequencies, can be an open circuit. It's not just about resistance. It's about being electrically quiet across the largest frequency spectrum as possible. The lower the logic voltage of anything electronic, the more important quiet becomes. In this respect, multi-pathing, using wires of differing lengths, to the equalization buss is a common way to broaden the bandwidth of the earth bond.

steve-l said:

Putting in grounding rods all over the place is not only wrong, but unsafe. Please understand that ground is not ground everywhere. There is and forever will be potential differences from point to point and that difference is dynamic and changing all the time. The ground connection used within a building must be non-current carrying and referenced electrically at a single point. That point is at the power entrance point of the building and it is called the equalization buss. Grounding at multiple points create ground loops, extraneous currents and is electrically noisy. This is very dangerous to electronics and people. The ground circuit must be an electrically quiet voltage reference and a safety path for faults only. Return current from single phase loads must always be on a separate neutral circuit which can only be connected to the equalization buss (ground) at the building entrance point. Bit serial comm. interface standards that use non-differential signal return lines in the comm. cable, like RS232 and its variances, are especially susceptible to catastrophic and spectacular failures when these little return lines try to normalize ground potential differences.

Click to expand...

Yesterday I put in another conduit run with a dedicated ground wire back to my service box which is grounded to the power entrance.
After a lot of discussion about grounding rods...etc. This make the most sense to me.

Ground potential is not always the same between different places. For instance the difference in ground voltage between NYC and NJ is about 3.5 volts.

garyhlucas said:

Seriously guys, if a ground rod affects the way your machine works you should buy different machines as the people that built it didn't have a clue! Driven earth grounds are useless as they can't possibly carry any useful amount of current. If your electrician is installing them you ought to be wary of their work because they don't have a clue either. I spent 20 years doing industrial electrical work and can count on the fingers of one hand a with a couple left over the electricians I've known that actually knew what they were doing!

Click to expand...

You must not have done any work on sensitive electrical equipment, or have been fortunate enough to not have ever come across a suspect ground network. As mentioned by the subsequent posts, the primary intent of a ground network *with respect to electronics* is not to deal with ground faults or other high current situations(obviously the safety part of it is still important). It is to eliminate potential differences within the ground network and stabilize it, so that your fancy electronics have a nice environment to operate. (especially when there are wired communications involved).

Yes, it would be nice to just blame it on the machine builders for not making them robust enough electronically, but you are going to pay out the nose for that kind of electronic protection. All that power control circuitry is a lot harder to isolate than your laptop.

Get an electrician that knows how to do earth resistance testing, it's a couple thousand dollars well spent.