Mutual Inductance 1

[deetz]

So we have 8 exhaust fans in our heat treat area. They all are on their own drive systems. One of the motors fried. We turned off the drive setup and were going to unhook everything but noticed 60-103 volts from any of the lines to ground. So I started to check things out a little closer to see if maybe it took the drive system out or one of the switches to allow volts through. I disconnected the wires from everything and still had volts to ground. I tried disconnecting all the grounds to see if that helped. Nope. So can you get that much voltage just by mutual inductance? I do get a tickle when touching the wires and a ground. Thoughts?

 

[electricpete]

I assume this is not causing any problem, just a curiosity.

I don't know the physical configuration of the cables but from what you've described I'd be more inclined to suspect capacitive coupling than inductive coupling. If you've been using a digital multimeter, try putting an analog meter (or light bulb) on it instead to see if the voltage comes down.

If you feel really curious, put a scope on it for an idea of the frequency content.

Do be extra careful for your safety until you understand the source of the voltage better.

 

[deetz]

Good idea. Thanks. It's kind of wierd how I have both ends of the wires disconnected and it still has voltage. We did find out that if we shut down the other drives and motors that the voltage goes down. Half, or four, of the motors are are run in the same 1 inch emt.

 

[bacon4life]

Touching ungrounded wires is dangerous. If it is not grounded, it is not dead. Any further testing should be performed with equipment rated for full line voltage with the appropriate CAT rating.

There can be voltage induced by both inductive coupling and capacitive coupling. On utility transmission and distribution lines, it is common to thousands of volts on wire with no metallic ground connections, or 10s of amps of current induced on wires with grounds at each end of a long line. On ungrounded wires, it possible for capacitive coupling to cause the ungrounded wire to be at nearly full line potential.

Assuming the motors are operating at 208-480 volts, the measured values seem typical. The ungrounded wire behaves as a capacitive voltage divider between the energized wire and the nearest ground plain. Having the energized wire very close to the isolated wire or the ground very far from the wires will increase the voltage on the isolated wire.

In order to measure induced voltages, you would need to ground one end of the wire and the measure the voltage at the far end of the wire.

Inductive coupling will be proportional to the current flowing on energize wires, whereas capacitive coupling is independent of current flow.

This is a perfect example of why control wires are run separately from power conductors.

 

[davidbeach]

"...drive systems..." Multi kHz carrier frequencies. Of course there are induced voltages. As the bacon guy says, not dead if not grounded. Better yet is grounded and any drives on any adjacent circuits shut down.

When one this sentence into the German to translate wanted, would one the fact exploit, that the word order and the punctuation already with the German conventions agree.

-- Douglas Hofstadter, Jan 1982

 

[deetz]

Thanks. Everything is grounded. So the only way to get rid of the voltage, when working on changing out a motor, is either shut down the other motors or rerun the line in seperate emt.

 

[deetz]

Or so it would seem that everything is grounded. As I was rereading my reply I should really double check everything.

 

[deetz]

On another note. The conductors for all the drives and motors are fed from one Buss plug, 480v. They go to a distribution block. When the one motor fried it took out the 70 amp fuse supplying this setup. The other motors were still working, by what I assume, from the VFDs generating the third leg. I am looking at installing a CB for each of the supply lines to the drives. I'm not sure why this wasn't setup this way from the begining. In any case, do I need to be looking at a Time Inverse style breaker or would just a regular mini breaker be sufficiant as long as I size it for the inrush, unless the drive really takes care of that. Yes the conductor size is a factor as well. Thoughts? And thanks for the support that you provide. It is much appreciated.

 

[electricpete]

> Thanks. Everything is grounded. So the only way to get rid of the voltage, when working on changing out a motor, is either shut down the other motors or rerun the line in seperate emt.

Hang safety grounds to ensure it is grounded.

EDIT - and do a live/dead/live check to make sure the equipment is deenergized before you come in contact with it.

 

[wcaseyharman]

I’ve encountered this phenomena a few times:
- in an old pan alarm inducing alarm panel powered by 120V. As it was unwired 70V appeared.

- at a house near a 230KV line. A fluke on the house gutters showed 600V, but you could walk up and touch it without getting shocked.

One technique is to put a something like a good quality 10Kohm resistor between the probes on your meter. Your meter is likely very high impedance. Connecting a resistor across will load the circuit and if it’s an induced voltage it will likely drop to a couple of volts.

 

[7anoter4]

Calculated for a group of 3*150 mm^2 cables 1000 m length , the voltage induced for 70 A it is only
8 V phase-to-ground.
The resistance of the insulation is very high so even a voltage meter of some MΩ resistance it will not measure such a small current.
However ,Fluke sustains that capacitance is the cause of such elevated voltage readings.
See

https://www.fluke.com/en/learn/blog/digital-multimeters/dual-impedance-digital-multimeters