What's new
What's new

How would you diagnose/remedy this grounding issue?

APD

Stainless
Joined
Nov 5, 2005
Location
Hudson Valley, NY
Today I was working on my lathe at home, and as I reached over to grab the air gun off the mill table, I felt a little tickle.
I got out the multi-meter and discovered I had 17 volts AC coming from the lathe when running.

I then ran a ground wire from the junction box on the back of the lathe to the ground wire in the wall box.

The stray voltage is gone but what is the cause and how would you try to track it down?

The lathe is a 3HP Harrison M300, running off 220v into a teco N3 VFD.
It has been set up like this for years with no issues.
 
The stray voltage is gone but what is the cause and how would you try to track it down?
The lathe is a 3HP Harrison M300, running off 220v into a teco N3 VFD.

The cause is the normal stray energy form the VFD and motor operating from it, with no grounding to dissipate it.
Track it down? Turn off the vfd output and measure, then disconnect the vfd input and measure.

SAF Ω
 
IF YOU DO NOT GROUND the machine, you can EXPECT a certain amount of leakage voltage. Especially if using a VFD, but even with native 3 phase or even single phase.

A little capacitance exists, unbalanced, to the motor from the coils in it, and the wiring, etc. That is the "cause".

NOT HAVING A GROUND IS JUST WRONG, and you found out one reason. Your 17V would have been a good deal more if there was a real fault.
 
IF YOU DO NOT GROUND the machine, you can EXPECT a certain amount of leakage voltage. Especially if using a VFD, but even with native 3 phase or even single phase.

A little capacitance exists, unbalanced, to the motor from the coils in it, and the wiring, etc. That is the "cause".

NOT HAVING A GROUND IS JUST WRONG, and you found out one reason. Your 17V would have been a good deal more if there was a real fault.

Yes I know the machine needs to be grounded.
The 240 supply to VFD is grounded in the box on the wall. And there are 4 wires in the junction box on the machine, connected to 4 wires from the VFD. So it appears to be grounded. I added another ground because it obviously wasnt making a good ground connection today.
I guess Im just looking for ideas on how something could suddenly lose its ground or anything else to check.
The lathe has probably been installed at that location for about 4 years, and I cant imagine it has not been grounded the entire time. I would have gotten shocked a long time ago.
 
If you can feel a "tickle" at 17V you are a very "sensitive man", and may want to be careful around 12V wet cells. They may "set you off" and you don't even know why ;-)

A "very good ground" could display 17V differential between another "very good ground" only a few yards distant.

Lightning doesn't strike from the sky down to the earth, It is DRAWN down by a moving charge WITHIN the earth.

But good grounds, dedicated grounds, that have been "tested" by running a single wire "to earth" and evaluating the potentials and currents are never a bad thing.
 
To be fair, the VFD itself can create local voltages if it is not grounded per the instructions. It can have radio frequency voltages on the wires and on it, with peaks and nulls along the wires, even on ground.. Not that I think that is what you felt, but it is possible.

The other thing is that I doubt it was really 17V. Probably it was pulses, and a good deal higher in voltage. The meter has some averaging in it, and very likely was under-reading the fast pulses of voltage, so you may have had 60 or more volts as pulses, which most people might well feel.

The backs of my fingers are sensitive to leakage currents if brushed lightly over a surface, I find that sort of current easily as a sort of "buzzing". But only the backs, the rest of my hand is not over-sensitive. .
 
Most switch mode power supplies leak due to the somewhat high magnetic fields.

Your main supply cable should ground to the equipment frame first then pigtails from that point to the vfd.

Any electrical box like one used to complain the vfd also grounded.

Sent from my SAMSUNG-SGH-I337Z using Tapatalk
 
To be fair, the VFD itself can create local voltages if it is not grounded per the instructions. It can have radio frequency voltages on the wires and on it, with peaks and nulls along the wires, even on ground.. Not that I think that is what you felt, but it is possible.

The other thing is that I doubt it was really 17V. Probably it was pulses, and a good deal higher in voltage. The meter has some averaging in it, and very likely was under-reading the fast pulses of voltage, so you may have had 60 or more volts as pulses, which most people might well feel.

The backs of my fingers are sensitive to leakage currents if brushed lightly over a surface, I find that sort of current easily as a sort of "buzzing". But only the backs, the rest of my hand is not over-sensitive. .

That reading from a multimeter may not mean much if it is a high impedance source. That is why some electricians use their Wiggins testers even though they have better quality equipment. (Jerry, I know you know that, I'm posting it for the unknown readers out there.)

When I first had three phase power installed, it read 500 volts on a 240 V line with an analog multimeter. Turned out that the meter had a compensating circuit to correct for the slight drop off at 400 cycles compared to 60. The power line was picking up the signal from an AM radio station nearby and the compensating circuit compensated the megacycle signal into a 250 volt increase in reading.

RE backs of fingers, same reason mothers check the temperature of baby's formula on their wrists where there is no callus built up as there is on finger tips. High line clearance workers are taught to test whether a tree is hot with the backs of fingers. Besides the sensitivity issue, a shock activates all the muscles and the stronger ones dominate, causing the hand to close and the forearm pulls up to the shoulder, away from the offending tree.

Grounding can be a tricky issue, what with voltage built up across ground wire impedance and currents induced by magnetic fields. Chris Sarros told me that to get a really cold ground for KMOX AM's 50 kw transmitter they ran a coaxial cable to a solidly earth grounded rod, in effect a shielded ground wire. How weird is that?

Bill
 
High line clearance workers are taught to test whether a tree is hot with the backs of fingers. Besides the sensitivity issue, a shock activates all the muscles and the stronger ones dominate, causing the hand to close and the forearm pulls up to the shoulder, away from the offending tree.
Bill

Makes sense when you think about it. But never would have, good to know.
Thank you.
 
Yes... per the meter reading

Most digital voltmeters have a 10 megohm impedance. That's pretty high, and deliberately so, with the intent that it will not "load" the voltage and drag it down, which would lead to a wrong reading. The problem when checking a power line for being "live", is that just a very small leakage can show up as a large number in volts.

It takes only around 400 picofarads of capacitance (0.0000000004 farad) to show a voltage of half line voltage using a 10 megohm meter. That can be due to just the fact that one wire runs next to a hot wire for some distance.

With a higher frequency source, like a VFD that uses pulses at several thousand Hz, way above 60 Hz, it takes a good deal LESS capacitance to get a high voltage reading due to leakage. Radio does not need any capacitance, it can directly "induce" voltages onto a wire, the wire acts as an antenna. The VFD has "harmonics" that are well up into the radio and TV broadcast frequency area, so it can have all sorts of weird effects, depending on circumstances. But a "Wiggins" tester is not affected.

The "Wiggins" tester (known as a "Wiggy") actually draws power from the line, to operate a solenoid. The solenoid pulls a pointer along a scale to indicate the voltage on the wire etc being measured. That takes some actual current (the whole tester vibrates), and leakage will not have any effect on a "Wiggy".

When you find voltage on a machine using a meter to test for it, you MIGHT have a serious shock hazard problem, or you might have leakage (and no ground) or you might have a radio effect. If you find a voltage showing on the case or frame of a machine with a Wiggy, you almost certainly have a definite shock hazard problem. Either way you may have a bad ground, though.
 
Last edited:








 
Back
Top