At my facility we have a 3 phase 480v delta wye transformer( no neutral) feeding our plant with a high resistance ground package with an analog meter on it reading about 25 volts to ground. Measuring line to ground voltages on phase a is 283v, phase b 274v, phase c 295v. But measuring phase to phase I have roughly 480v +- 2 or 3 volts across all 3 phases. We have been having a few vfds and 480 to 24v power supplies going bad lately and I’m not sure if the phase to ground imbalance would cause that. I have had no luck tracing down the minor ground fault due to misreadings from the analog meter after isolating the 4 main switch gear breakers that it powers and also lack of down time. Is there a possibility that the meter needs to be recalibrated? Or could any one give advice on tracing down the ground fault with limited down time.
Finding minor ground fault on a 480v system
- Thread starter Rob442
- Start date
- Status
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I don't work on large HRG systems. I was in charge of a single small (75kVA) HRG system for our lab.
I don't think you can use voltage measurements to reliably trace small faults.
If you had a solid fault, then one phase would be at 0V to ground, the neutral would be at 277V to ground, and the other phases would be 480V to ground. I'm sure you could trace such with voltage.
But in the situation you describe, things such as harmonic content on the circuit (introduced by input rectifiers on your VFDs, for example) and capacitive coupling could raise the neutral voltage without a true insulation fault. Or the common mode rectifier ripple of the VFDs capacitively coupled through the motors could place a common mode voltage on your system, seen as a neutral to ground voltage.
I think that to reliably trace faults in your system you need to measure common mode currents, possibly with some tool that would induce a common mode signal on your grounding resistance. I know such systems exist, but I've never had to research them in detail nor use them. On my small system we would just take a clamp on oscilloscope probe around the three phase conductors to search for net current.
-Jon
I don't think you can use voltage measurements to reliably trace small faults.
If you had a solid fault, then one phase would be at 0V to ground, the neutral would be at 277V to ground, and the other phases would be 480V to ground. I'm sure you could trace such with voltage.
But in the situation you describe, things such as harmonic content on the circuit (introduced by input rectifiers on your VFDs, for example) and capacitive coupling could raise the neutral voltage without a true insulation fault. Or the common mode rectifier ripple of the VFDs capacitively coupled through the motors could place a common mode voltage on your system, seen as a neutral to ground voltage.
I think that to reliably trace faults in your system you need to measure common mode currents, possibly with some tool that would induce a common mode signal on your grounding resistance. I know such systems exist, but I've never had to research them in detail nor use them. On my small system we would just take a clamp on oscilloscope probe around the three phase conductors to search for net current.
-Jon
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
Do the VFDs have neutral terminals? What do they use to reference ground?
In other words, how do they "know" what the relative line voltages to ground are?
In other words, how do they "know" what the relative line voltages to ground are?
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
Most VFDs have transient voltage suppression devices connected line-ground. These can cause problems when the VFDs are used on anything other than a solidly grounded wye system.
Additionally the DC rail to ground voltages fluctuate differently depending on the supply grounding. The voltage from + to - rail depends only on L-L voltage and has the same expected ripple for all the different systems, but the voltage from + to G and - to G is different. This DC rail fluctuation is capacitively coupled to ground via the switching PWM and the motor winding to frame capacitance.
So the important functional bits of the VFD are blind to grounding, but a bunch of second order aspects are very much ground dependent.
-Jon
Additionally the DC rail to ground voltages fluctuate differently depending on the supply grounding. The voltage from + to - rail depends only on L-L voltage and has the same expected ripple for all the different systems, but the voltage from + to G and - to G is different. This DC rail fluctuation is capacitively coupled to ground via the switching PWM and the motor winding to frame capacitance.
So the important functional bits of the VFD are blind to grounding, but a bunch of second order aspects are very much ground dependent.
-Jon
No they do not. So I guess you’re hinting that it’s irrelevant
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
Actually, asking, as I'm only slightly familiar with VFDs. I'm trying to understand enough to contribute.
- Location
- Wisconsin
- Occupation
- PE (Retired) - Power Systems
HRG resistors are typically chosen to limit the ground current to about 5 amps, but the actual value does depend on your system characteristics.
Do to the low current it is common to find voltmeter as part of the HRG package. Just use Ohm's law to determine the current, assuming the OP has a typical 55.4 ohm resistor with 25V on the meter, they have about .45 amps on the ground. This may be a normal amount of leakage current for their system.
Many HRG systems are supplied with a 'fault detection contactor which varies the resistance so the ground current pulses between two values. An ammeter which clamps around all of the circuit conductors (zero sequence) can be used to locate which circuit contains the fault simply by looking for a pulsing current value.
Most of the large HRG manufacturers have good on-line technical bulletins.
Do to the low current it is common to find voltmeter as part of the HRG package. Just use Ohm's law to determine the current, assuming the OP has a typical 55.4 ohm resistor with 25V on the meter, they have about .45 amps on the ground. This may be a normal amount of leakage current for their system.
Many HRG systems are supplied with a 'fault detection contactor which varies the resistance so the ground current pulses between two values. An ammeter which clamps around all of the circuit conductors (zero sequence) can be used to locate which circuit contains the fault simply by looking for a pulsing current value.
Most of the large HRG manufacturers have good on-line technical bulletins.
We have 3 of these HRG units for each of our 3 transformers, the other 2 are reading 0v. That’s a good idea to verify the meter is reading correctly. And yes there is pulse option for locating a ground fault but would it pick up such a minute voltage? Also would it be harmful to any sensitive electronics?
The vfds that I work on (mostly powerflex 5hp and under) are all 3 phase with no neutral and do not reference any voltage to ground on the line side that I am aware of. Someone might prove otherwise
Yes we had a hard ground fault on one of our hrg packages before and it was very easy to locate the short. As far as common mode noise we do have roughly 100 or so drives in which the motor leads are unshielded so that’s very possible.
- Location
- Wisconsin
- Occupation
- PE (Retired) - Power Systems
It is not uncommon to find up to 1 or 2 amps, on a system, much less the fraction you are showing.
There is always some amount of coupling capacitance, primarily due to the cables, which causes a base amount of leakage current to always be present. There is a fairly simple measurement you can perform, which is typically done when an HRG is first commissioned. The test involves disconnecting the resistor and then measuring the amount of current that flows in a direct short from L-G. A current limiting fuse is normally used in the test circuit, just in case there really is a fault to ground.
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
Has the voltage across the resistor been measured with an oscilloscope to see if it's mostly 60 Hz and/or low order harmonics, or if there's substantial higher frequency content from VFDs, etc.?
Flicker Index
Senior Member
- Location
- Pac NW
- Occupation
- Lights
What's the resistance value where the 25v is present? That will give you how much the leakage current is
No I have not verified that as there are 3 resistors in the unit and I can’t find a wiring diagram on the hrg package. I just need to find the time to go over it. Thanks for the advice
- Status