Very odd voltage issue (p-g higher than p-p)

Status
Not open for further replies.
S

SkyWarnEngineer

Member
Location
Texas
I just joined the forum to solicit feedback on a problem I have never before experienced. It?s a bit long so please stay with me. I have a 2.5MVA transformer that is 7.2kV primary Y and a 480V delta secondary. Our secondary uses a high resistance grounding system that consists of three small 2kVa transformers and a series of lights and a relay to indicate when we have a ground. This way we can ground any one phase and not stop any equipment (pumps) from running. An electrician walked into a block house (building with MCCs) and noticed a strange smell. He found the ?B? phase 2kVa transformer of one of our grounding systems deformed with resin running out of the case and down the wall it was mounted on. Upon investigating he found the ?B? phase fuse of the grounding system blown. He threw the disconnect switch to the ground system and checked voltage of the buss work. He found approx. 480V phase-to-phase, 410V A & C phase to earth ground, and 750V B phase to ground. His current measurements indicated no more than 3% imbalance between the phases, each were around 1500A. At this point he contacts the area electrical engineer who then contacts the rest of us to get the entire plant team together to help figure out what to do. We decide to use our 480V tie system to switch to a spare transformer, expecting the issue to be something with the transformer. We decide because of the wonky voltages to not tie the two transformers together and to get with operations and plan a brief minute or two power outage while we executed the switching order. The night before the transformer swap, operations was shedding unnecessary load. When we convened the morning of the switch we found voltages had become more normal. We measured 480V phase-to-phase and around 277V phase-to-ground for each phase. We decided to make the swap, but tie the transformers together so operations did not suffer any interruption. The swap goes flawlessly. We isolate the suspected sick transformer and call for a Doble test set and an emergency SD Myers DGA. We got with operations and got a list of all equipment that had been shed the night before the transformer swap. We then went and while measuring voltages switched everything back on and off. We found that a sump pump station, when started, gave us the same wonky voltages on the spare transformer. We were able to get consistent repeatable results turning the sump on and off. We called for a Baker test (megohm, high pot, PI, DA, etc.) test of the motor from the starter so we could verify the cable and motor. Everything passed. Somehow, a motor that runs fine, and passed all of the standard motor tests, when started, elevates phase to earth ground voltages up to 420V on ?A? and ?C? phases, and 720V on ?B? phase without effecting the current balance on the transformer, or the phase-to-phase voltage. We are at a complete loss to explain what is going on. Anyone seen anything like this before?
 
Have not seen this before, but an arcing connection to ground can "pump" the whole delta with respect to ground.
Make sure your measurements are not being confused by DC components.
A large capacitance to ground would not show up on the Megger but might result from something like wet insulation in the motor.
A failed contact transfer in the starter (wye-delta) might also cause bizzare results. On a grounded system it would immediately blow something.
 
SkyWarnEngineer said:
I just joined the forum to solicit feedback on a problem I have never before experienced. It?s a bit long so please stay with me.

I have a 2.5MVA transformer that is 7.2kV primary Y and a 480V delta secondary. Our secondary uses a high resistance grounding system that consists of three small 2kVa transformers and a series of lights and a relay to indicate when we have a ground. This way we can ground any one phase and not stop any equipment (pumps) from running.

An electrician walked into a block house (building with MCCs) and noticed a strange smell. He found the ?B? phase 2kVa transformer of one of our grounding systems deformed with resin running out of the case and down the wall it was mounted on. Upon investigating he found the ?B? phase fuse of the grounding system blown. He threw the disconnect switch to the ground system and checked voltage of the buss work. He found approx. 480V phase-to-phase, 410V A & C phase to earth ground, and 750V B phase to ground. His current measurements indicated no more than 3% imbalance between the phases, each were around 1500A. At this point he contacts the area electrical engineer who then contacts the rest of us to get the entire plant team together to help figure out what to do.

We decide to use our 480V tie system to switch to a spare transformer, expecting the issue to be something with the transformer. We decide because of the wonky voltages to not tie the two transformers together and to get with operations and plan a brief minute or two power outage while we executed the switching order. The night before the transformer swap, operations was shedding unnecessary load.

When we convened the morning of the switch we found voltages had become more normal. We measured 480V phase-to-phase and around 277V phase-to-ground for each phase. We decided to make the swap, but tie the transformers together so operations did not suffer any interruption. The swap goes flawlessly. We isolate the suspected sick transformer and call for a Doble test set and an emergency SD Myers DGA. We got with operations and got a list of all equipment that had been shed the night before the transformer swap. We then went and while measuring voltages switched everything back on and off.

We found that a sump pump station, when started, gave us the same wonky voltages on the spare transformer. We were able to get consistent repeatable results turning the sump on and off. We called for a Baker test (megohm, high pot, PI, DA, etc.) test of the motor from the starter so we could verify the cable and motor. Everything passed. Somehow, a motor that runs fine, and passed all of the standard motor tests, when started, elevates phase to earth ground voltages up to 420V on ?A? and ?C? phases, and 720V on ?B? phase without effecting the current balance on the transformer, or the phase-to-phase voltage.

We are at a complete loss to explain what is going on. Anyone seen anything like this before?
Click to expand...
I have not, but at least I can put some paragraphs in for you. Click "Reply With Quote" if you want to break the text up differently.
 
GoldDigger said:
OP said high resistance ground.
Stray path need only have a lower resistance than that (and the impedance of the zig-zag.)
Click to expand...


It could be disconnected (open resistor) or the resistor is to small. Just a guess. The values the OP describes are not abnormal for an ungrounded system with a arcing ground fault.
 
mbrooke said:
It could be disconnected (open resistor) or the resistor is to small. Just a guess. The values the OP describes are not abnormal for an ungrounded system with a arcing ground fault.
Click to expand...
Which would not account for burning out the grounding transformer, although that may have been a coincidence.
 
I could be wrong, but OP mentioned primary wye... with that said, is it grounded? Someone could correct me if I am wrong, but an ungrounded wye primary is ripe for issues. I have heard of ungrounded Y primaries being susceptible to ferroresonance and voltage shifts, Even with a delta secondary.
 
If the primary feed from POCO is actually a delta, you can use a Y primary with the center point left floating.
If you have a POCO wye and a delta secondary you MUST either leave the center of a wye open or use s delta primary winding instead.
 
GoldDigger said:
If the primary feed from POCO is actually a delta, you can use a Y primary with the center point left floating.
If you have a POCO wye and a delta secondary you MUST either leave the center of a wye open or use s delta primary winding instead.
Click to expand...

Depends on what your trying to accomplish. While an ungrounded wye primary will not blow fuses on distribution faults out on feeders, its very susceptible to ferroresonance. Pros and cons exist in both wye grounded and wye ungrounded. A delta primary is often the best choice.


In the ops case it could be anything.
 
SkyWarnEngineer said:
Our secondary uses a high resistance grounding system that consists of three small 2kVa transformers and a series of lights and a relay to indicate when we have a ground. This way we can ground any one phase and not stop any equipment (pumps) from running.
Click to expand...

Your service grounding description is not accurate. Not clearly defining your system will result in many mis-directed recommendations

You most likely are running an ungrounded system with a ground detection and annunciation scheme. The problems you describe fit this type of system.
While it is possible that you do have a high-resistance ground using a transformer style artificial neutral, nothing else you describe leads me to this assumption.

If your system is truly a 480V delta service, you need to get out of the habit of regularly performing L-G voltage measurements.

My first guess is that you had an arcing event involving the B-phase transformer used in your ground detection scheme. As others have mentioned, on ungrounded systems an arcing fault can create up to a 2x voltage condition. By opening the switch to the detection scheme you cleared the fault.
 
I have seen an arcing fault on an ungrounded system create high transients and that destroyed a lot of insulation in the system, but I have never seen it result in a sustained over-voltage, and my understanding is that is not possible.

It almost sounds like you have a totally ungrounded system and you're using an open-delta potential-transformer bank and relay to indicate a fault.

Your phase-to-phase voltages are fixed by the number of turns in each winding of the transformer secondary so those will never change unless a winding develops an inter-turn failure.

On an ungrounded system your phase to ground voltages will change based on the capacitance of each phase and the connected load on those phases, and this explains the different voltages you're seeing. Sometimes deliberately ungrounded systems are spec'd with 1kV insulation specifically because they can float so high above ground potential and this can definitely stress 600V insulation.

That said, I'll repeat I think you've always had an ungrounded system, and you're simply seeing a failure in the protection system. Why that is, I can't immediately explain. Is there a resistor on those three 2kVA transformers and did it burn open?
 
Last edited:
Sump Pump Control - Pilot Devices and controller?

Sump Pump Control - Pilot Devices and controller?

Did you test all of the control circuitry (pilot devices - starter, alternator, sequencer, float switches, cables ect. that operate the sump pump for grounds) when testing the pump motor circuit?
 
Last edited:
Sahib said:
I think it is to prevent 3rd harmonic current flow.
Click to expand...
No. It is to prevent high circulating current when there is any voltage imbalance in the line to neutral voltages on the primary side.
The existence of a closed delta secondary imposes a voltage balance constraint condition that does not exist on a wye primary to wye secondary situation.
Harmonics are a lesser concern and blocking triplen harmonic current depends only on the secondary being a delta.
 
Last edited:
Status
Not open for further replies.
Top