480V Wye Service Voltage Issues

[vtpower]

Folks,
We are having an interesting voltage problem at a water pumping facility and I was hoping someone might be able to shed some light. The setup is a a three phase overhead power bank, made up of three 167KVA, 12.47kv / 277/480Y service. The service enters the building into the main disconnect, then into an automatic transfer switch which is tied to a generator on the other end. The station has 100HP lift pumps for water supply, and they are all 480V delta pumps. There is also a 480V / 120/208 dry type stepdown to pick up lighting load throughout the building. The issue I am having is, there is a Siemens phase detectiond device that monitors all three phase to phase voltages and will shut the motors down on undervoltage, over voltage and imbalance. It continously is tripping the motors off on Imbalance, and I look at the line voltage and have appx 484v, 484v and 506v, but the funny thing is the phase to ground voltages are all right on at around 280, 280 and 280. Ther is also a set of primary regulation about 10spans back and I know the primary system is balanced and has good voltage. How is this Phase to Phase voltage imbalance occuring while the phase to ground voltages are balanced? All three overhead transformers are exactly the same at 4% impedance, and are even the same batch. Any help would be greatly appreciated.

Thanks.

 

[dpc]

You have less than 3% imbalance, based on the phase voltages you indicate. I would think the trip setpoint should be somewhat higher than that. Do you know what the setpoint is?

I would be somewhat suspicious of the voltage regulators. I have seen one regulator control get out of synch with the others. You should be able to see what tap each regulator is on via the dial indicator on each regulator. Are there other large customers on this feeder?

When you measured the 480V voltages, were the pumps on or off?

Are the pumps running off VFDs?

What were you using to measure the voltages?

 

[vtpower]

Thanks for the response DPC. The relay is set at 5% imbalance, or 25 volts at 480. I did review the regulators and the taps were even, along with the datalogging in the control indicated they have been working properly. The confusing thing is, the line to ground voltages (277) are fairly balanced, which leads me to believe the primary voltages are well balanced. There are no other large customers between the regulators and this customer.

We measured the 480 with a bunch of different scenarios, pumps on, pumps off, transfer switch in generator mode so absolutely no load on the incoming and continued to have this strange imbalance. I was wondering if there were issues in the pumps but when the generator comes on the voltages are well balanced and there are three seperate 100HP pumps that are there for redundancy and the same happens to all.

We also had a variety of voltage measurements devices, there are built in analog meters on the panel, we also used our own voltage meters, and we have a high resolution Rustrak voltage recording device that takes 30snapshots a second of voltage and current. All three are showing the same situation.

The issue I can't understand, is what would cause a line to line voltage imbalance, but not effect the line to neutral voltages?

 

[davidbeach]

There is something going on, buried in your "appx." that is hiding the real values. If you have all three phase-neutral voltages at 280V and all three are separated by 120[°] you will get three phase-phase voltages of 485V. If the phase angles are not all 120[°] you can have three phase-neutral voltages that are all the same but have different phase-phase voltages, possibly all three different.

To get about 506V, you would have to have phase angles of 115.5[°], 115.5[°], and 129[°]. That would produce 505V, 474V, and 474V. Unfortunately that 474V is 10V less than your measurement.

Something does not compute.

 

[vtpower]

I agree with you there. I was wondering if I had a phase angle shift caused by an impedance difference or a transformer ratio difference. We did TTR the transformers and got same ratio's and the nameplate has a percent Z of 4% on all three. I will be receiving the results tommorow of our Rustrak power quality device and will be able to compare the exact voltages on all three legs.

Any ideas on what could cause a voltage phase shift? This has got to be fairly tough to do. I can understand shifting the powerfactor of each leg, where the current phase angle changes, but actually changing the phase angle of the voltage seems impossible without having some sort of source or differently connected tranformers.

 

[davidbeach]

Getting to 115.5[°], 115.5[°], and 129[°] would probably be asking too much of normal conditions, but a distribution feeder with lots of single phase loads not even distributed will develop a voltage phase shift from source to end; with out regulators there would also be a difference in voltage magnitude but the regulators mask that. I've seen as much as 6[°] or 7[°] difference between the smallest phase separation and the largest phase separation.

 

[apowerengr]

vtpower - have you got a capacitor with a blown fuse?

 

[waross]

I agree davidbeach. May I elaborate a little on your post?
I suspect that a heavy neutral current somewhere upstream in the distribution system may be causing a neutral offset.
A voltage drop in the neutral has to act in some direction, so it causes a shift in the neutral point in the vector representation. I suspect that such a shift is causing the effects that you describe.
The effect will originally cause unequal line to line voltages and probably unequal line to neutral voltages. I suspect that the problem is quite far away from your installation. I suspect that a set of voltage regulators downstream of the original problem is adjusting the line to neutral voltages. That would explain the regulators being on the same tap at your location.
If the regulators are wye connected they should produce equal line to neutral voltages. If your transformer primaries are wye connected they will reproduce equal line to neutral voltages.
The phase shifted neutral will, however, cause unequal line to line voltages.
I had some problems at a sawmill with a 400 Hp. motor on a soft start. The utility had very uneven loading and a very serious neutral shift. The soft start would shut down on an excess phase angle condition. The mill would sometimes lose several hours in a day when the neutral loading down the road was heavy. We didn't have much to work with but we found a circuit breaker that was suitable to install as a bypass. Once we started the motor we closed the breaker and then turned the soft starter off.
respectfully

 

[jghrist]

Is the primary connection delta or wye? Where is the unbalance measured, at the transformer or at the load? How balanced is the 120/208 lighting load? How much lighting load is there compared to 3Ø motor load?

 

[vtpower]

Thanks for the responses.

I did check all the capacitors on the circuit, and all are in with no blown fuses.

Waross- I am going to look closely at the circuit loading and neutral current on the system to see if we in fact do have some heavy neutral current upstream. So, with the breaker, were you bypassing the soft starter?

Jghrist - The primary connection is wye. The unbalance line to line voltage is measured on the pots themselves, and at the transfer switch in the building. I have checked with and without load on the system. I am not sure how balanced the lighting load is yet, but that is fed from a dry type that is connected 480 delta on the high side. The only reason I might rule out a problem with loading, is that when the generator is running the voltages are right on.

 

[Snoball]

Is the transfer switch a 4 pole switch?

 

[jghrist]

I would suspect either heavy unbalanced load and neutral current upstream on the 12.47 kV line or a bad neutral connection somewhere. Either would cause a neutral offset. A neutral offset would normally make the Ø-N voltages unbalanced. It would take an odd combination of loads to leave the Ø-N voltages balanced but unbalance the Ø-Ø voltages.

Can you connect some VTs to the primary to see if the primary voltage is unbalanced?

 

[waross]

Hi vtpower;
We started on the soft start and then the plant electrician closed the bypass breaker across the soft starter and turned the soft starter off.
The breaker was close enough to the motor full load rating to give some overload protection.
I left instructions to connect an overload relay to the shunt trip on the breaker but it probably wasn't done.
jghrist and vtpower;
I suspect that the cause of the problem is some distance away, probably miles.
I suspect that there is a set of voltage regulators near the problem that are equalizing the phase to neutral voltages. That would explain the equal line to neutral voltages at your station and explain why the local regulators appear to be on the same taps.
respectfully

 

[davidbeach]

Yeah, but you can't actually get those numbers. You can get five of the six, any five, and they will determine the sixth. The numbers given just don't represent a 3-phase voltage unless the voltage was changing fast enough that those sequential readings all represent different conditions on the system.

 

[wbd]

vtpower -
did I miss something? In one post you stated that the voltage is 506V and then in another the relay is set at 25V over 480V. That would make the setpoint below the actual voltage of 506V. there's your problem.
By the way, where are you in vermont? I'm in Rutland

 

[vtpower]

wbd - The relay is actually set for 5% imbalance, so it shifts depending on the phase to phase voltages. The facility manager is looking into a similar relay with a 10% imbalance trip, but of course we still have the strange voltage issue that will cause motor winding heating.

I am at Green Mountain Power in Colchester, you CV or VELCO?

Thanks for the response