I am not sure where this post belongs but I am an industrial electrician working in the mining industry. We have ground fault indicator lights on our 3 phase delta system which makes it easier to locate a ground fault issue. . . . one of the trhee lights go out and we just start shutting equipnebt down until we isolate the problem and all three lights are on. Recently we have had a situation where all three lights flicker for several minutes at a time which is something new. I understand the theory ofthe ground fault circuit but am puzzled as to what might cause all three lights to flicker as though there might be some tracking on all three phases. The problem doesn't appear to be isolated to just one transformer in our substation but has been showing up on a couple of different ones. Anyone else have experience with this?

jmaclean:
Good morning & welcome to ECN forums from one of the 'Jersey guys'

A light goes out when that phase gets grounded. The lights go out when there is no power too. Could the flickering ground lights be due to intermintant connections?

Thanks for the replies, guys. The scenario of the intermittent connections is possible but this occurs in several different MCC's throughout the mill and on at least two subtation transformers.

You wouldn't have any VFDs bouncing back harmonic echoes?

====

While there are various ways...

One VFD scheme uses a 'low pass filter' to establish a 'DC tank' -- in the form of a very hefty capacitor, or two... which is then drained by IGBTs to craft AC wave forms by stacking DC transient impedances.

If these massive capacitors cause enough phase shift then the indicator lights would read 'dead.'

A scope or strobe set up would discover if these transitory outages are synch'd to 60Hz or some harmonic. ( 120 Hz = low pass ripple frequency )

Hmmm... We have dozens of VFD's throughout the mill, most, if not all of them with inductors to filter noise from the system. This just started happening in the last 3 months or so. This is something we haven't considered before.

Clarification... most of the hve line reactors installed. . .

Most of the 3 phase ground detectors are using neon lamps. Is it possible you are seeing the normal flickering of neon bulbs? All 3 are going out in Sync? sequentially or random? Is there a set pattern? Any effect on how long it has been on? IE if you turn the power on and the lights flicker for a little while and stop. You turn the power off and on and it starts up again or do you have to turn the power off for a few seconds before it will start again? might indicate a bad capacitor.
I am also trying to imagine what normal machine could have a rotating ground fault? Maybe a wire round armature or rotor in a machine?

Assuming it isn't possible for all three phases to ground at the same time, the other explanation would be power loss. If you lose a phase, do the remaining two bulbs get dimmer?

I looked on Ebay and they have Dranitz monitors for $100 or less.
It can sure help you figure out what is going on with your power. I used mine a lot when I was working for IBM and I took it with me when I left.

Have you checked your ground bonding? Ground detectors need a good ground reference to work.
Joe

The solution of associating which motor starts with the
ground fault light is good. One solution I used in a paper
mill was to use a ground tracer, which put a buzz tone on the
particular PDC, with the ground fault. Using the earphones
and the probe coil, you could hear which drive might have
a bad motor or feeder, if you opened the starter doors,
and got the probe coil close to the load side terminals.

This ground tester had 3 lamps, as ground detectors,
and the buzz could be switched to any phase. The tester's
ground wire had to be placed carefully.

Grounds were not too high a priority in an operating
plant, and done on the fly. Of course if a second ground
happened on another phase, something would blow, perhaps
clear the first ground, or shut down a motor with blown
fuses.

The part of the plant with the worst grounds was the
wood mill division, with timber up out of the saltchuck.
Water and vibration was everywhere.
The biggest motor on the 575 volt side
was 250 HP, everything bigger was 2300 Volt.

Steam rams were used to flip logs and there always
was sawdust burning around the cylinders in the basement
and on the cable trays, so sometimes it was a cable fault.

One caveat here, I worked in pulp and paper, and what
we did may not be safe or customary in mining.