The shield shocked me!!!

F

Thread Starter

Fluoronator

While replacing some discrete I/O modules I took the opportunity to neaten some of the incoming wiring to our control panel. I quickly learned that disconnecting the shield (from ground) of a 120vac twisted pair cable will deliver a potent shock. I was under the assumption that the twists in the cable would cancel out any inductance into the shield. I thought the shield should only carry minuscule noise and prevent static buildup.

However, the disconnected shield remains hot. Is this normal or is it an indication of a problem?
 
welcome to the world of ground loops! They are especially common and trouble shooting is always fun...

You have to start with you electrical grounding design drawings, if one was prepared, otherwise start the search.
 
Just to follow up, instrumentation grounding is kept separate from the electrical power grounds for this reason. It has to do with ground potentials, grounding resistance, etc. It is not unusual to have ground potentials between power and instrument grounds.

The power transformer used for instrument power is tied to the instrument ground to prevent the situation you describe.

Ground coordination is quite an exercise...
 
I agree with d; this isn't all that uncommon. I doubt that it's caused by the wires inside of the shield, unless they are constantly carrying a fair amount of current.

Where is the shield drain wire grounded? It should be grounded at only one end of the cable--and by default, that is usually the control system end. (Most people say it has to be grounded at the source of power for the circuit--but most sites I have worked at the source of power is usually, 98% of the time, coming from the control system and they just choose to ground all shield drain wires in the control system cabinet just for "consistency."

I have seen shield drain wires grounded at both ends and have a high circulating current--but that was caused by proximity to other high current-carrying cables (improper signal level separation).

One of the things I usually do is have a jumper with an in-line fuse of 1/4-1/2 A and use that to jumper the shield to ground. If the fuse blows, then there's a more serious problem--like the insulation of one of the conductors inside the shield is nicked and shorted to the shield somewhere. Or, again, there is proximity to high current-carrying conductors (440 VAC, three-phase motor leads, for example) which is causing a high induced current on the shield/drain wire.

Usually if the shield is grounded at both ends simply lifting the shield at one end will resolve the problem. Sometimes, though, it doesn't. And that's when the fun begins.

Sometimes it's necessary to disconnect all the conductors involved and meggar them to the other conductors and to the shield drain wire. Sometimes, it's pretty obvious the cables were not routed properly (insufficient signal level separation)--but even then, that can be difficult to rectify if the cable run is over a long distance or passes through one or more vaults which are difficult to access.

Don't fear the worst--just use a logical process of elimination until you find the problem. It may be relatively easy to solve--like finding the shield drain wire terminated at both ends. Sometimes, even the best of electricians get a little over-anxious especially if the job is behind schedule.

Please write back to let us know what you find.
 
F
Thanks a lot! I'll start by checking for grounding on both ends and if that's not the problem I'll use the 1/4 amp fuse trick. I let you know...
 
W
I can't say that I agree with below statement regarding grounding of instrument grounds. The proposed arrangement will also give you problems with lightning or thunderstorms. While the instrument ground is generally a single point ground, it should be connected to the power (safety/protective) ground at one place (Code requirement under the NEC). One of the reasons is just this - getting a shock on a shield.

It is unlikely that you will get shocked from something coupled onto a signal shield. If the shield is connected at one end (which it should be for low frequency grounding, e.g. <100 KHz) , the coupling will be capacitive in nature and will have high source impedance. If the shield is connected at both ends, you can get both inductive and capacitive coupling to the shield in addition to circulating ground currents due varying earth ground potentials. The shield being connected at both ends also should not cause a shock under normal environmental conditions but could if the shield is very long, there are large telluric currents in the area, or there are lightning or thunderstorms in the area.

First, you should not get a shock from a signal shield. The likely culprit is that the shield is in contact with an AC source (which should blow a fuse if grounded properly but may not if there is a high resistant current return path) or an AC source is using the shield as a return path, e.g. in parallel or in lieu of a neutral (which will probably not blow a fuse).

One of the first things I would do is measure the voltage to ground and connect the shield and measure the resulting current, which may give us clues to where it is coming from. If it is not a standard voltage, I would look at it with a scope and see if it 60/50 Hz. This may also may give us a clue as to where it is coming from. Then I would trace the shield looking for places where it could be connected to an AC source hot or neutral. This would also allow you to look for unusual wire routing or connections. If you are still having trouble finding the culprit, consider using one of the circuit tracers where on clip a signal generator device at your shield end, and then use the tracer to find what circuit you are connected to. If you find the shield grounded in two or more places, disconnect all but one ground and see if it goes away. Leave the shield grounded to the zero potential reference point of the signal circuit.

> Just to follow up, instrumentation grounding is kept separate from the electrical power grounds
>for this reason. It has to do with ground potentials, grounding resistance, etc. It is not unusual to
>have ground potentials between power and instrument grounds.

> The power transformer used for instrument power is tied to the instrument ground to prevent the
>situation you describe.

William (Bill) L. Mostia, Jr. PE
ISA Fellow, SIS-TECH Fellow,
FS Eng. (TUV Rheinland)
SIS-TECH Solutions, LP

Any information is provided on Caveat Emptor basis.
 
Agreed the grounds are ultimately connected as they must be for safety, at a single point.

I have to wonder about the use of shielded power wiring. It sounds like they wiring is not properly segregated as required by most codes.
 
Top