Motor current imbalance with balanced voltages 1

[pwrengrds]

I was asked to investigate imbalanced currents in a new installation. Location is in rural NW USA, estimated about 7-10 miles from the substation. Loads are mostly motors and lighting, it's a brick plant.

There are two buildings, with two services, the transformer (500kVA; 20.5kV/480; Ygnd/Ygnd) feeds both services. The voltages are balanced 274/276/276 volts but the currents 19/29/32 amps show excessive imbalance. The motors were started from a separate source (generator) and were balanced at 25 amps.

The other building load was 274/277/277 Volts and 43/54/57 on a motor.

I have been told that another motor fed from a different transformer was balanced and when they fed this load from that transformer it exhibited the same imbalance.

Any ideas what would cause the current imbalance?

 

[sibeen]

What are the Line - Line voltages?

 

[waross]

Somewhere between your plant and the substation will be a set of single phase voltage regulators. They will be seeing unbalanced line to neutral voltages due to uneven loading of the distribution circuit. The voltage regulators will correct the line to neutral voltages and by so doing will render the line to line voltages unequal and will also introduce phase angle errors when considered line to line.
The back EMF of the motor will present equal voltages and equal phase angles. This will cause a transfer of energy between phases.
As well as taking energy to drive the load, the motor will be transferring energy between phases to try to correct the unbalances. The degree of motor over-current will depend on a combination of the magnitude of the voltage and phase angle errors, the impedance of the transformer, the impedance of the distribution system and the impedance of the motor.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[zlatkodo]

Sometimes, this happens if there are lots of VFD's in the same electrical grid.

Motor Repair and Winding Design

 

[pwrengrds]

Voltages high side are 20.7/12kV; Voltage low side are 480/277 volts

waross; I agree with your assessment, what do I do about it? I have an appointment with the utility engineers this week.

Attached is a snapshot of the motor running. I thought that the phasors were 90 degrees out of phase due to unloaded operation except that B phase was 120 degrees out. It stumped me and made me think my measurements were wrong (taken with a fluke 435).

 

[waross]

I am curious as to the line to line voltages at the 480 Volt level. That will tell if you have unbalanced voltages.
As to a mitigation scheme, the first step is to request the utility to review and re-arrange the single phase load balance. This may not be possible and it may not be possible at all times of the day. It may call for ingenuity and innovation to find the best compromize.
Another mitigation scheme that I have seen is the connection of a wye/delta transformer bank downstream of the voltage regulator bank. The wye/delta bank will try to the limit of its impedance to balance the line to line voltages and correct the phase angles. The Y bank will transfer energy from the higher voltage phase(s) to the lower voltage phase(s). It will also generate a neutral current that will work to correct the offset neutral and in so doing correct the phase angles.
The Y neutral must be connected to the primary neutral conductor. I am aware that some standards warn against connecting the primary neutral on a Y bank but the neutral must be connected if the bank is to be used for unbalance correction.
Unfortunately the 4 wire, Y connection has some serious downfalls.
The bank will back-feed into a lost primary phase. This may cause fuse blowing and/or transformer burn-out.
If two primary phases are lost, the bank will back-feed approximately 50% voltage into the two missing phases. This condition is a known killer of residential refrigeration equipment.
In the event of a line to ground fault in the distribution circuit the Y bank will contribute a fault current limited by twice the impedance of the transformer bank. (Not three times, although the current is zero sequence current, and the normal zero sequence impedance is three times the positive sequence impedance, the supply to the back-feeding transformer is from two sources at differing phase angles.)
A side effect of this is that a line to ground fault anywhere on the distribution circuit may result in blown fuses at the Y bank.
I should mention that the installation that I saw was an accidental situation. A three phase distribution line ran for about 10 or 15 miles. At that point it split in two directions and and neither direction carried all three phases.
At the end of the three phase section there was a business using a four wire wye/delta service. The business ceased to be, but the transformer bank stayed in place for a few years. Then an engineer noticed the unused bank and had it disconnected. At that time, what had been a well balanced circuit according to the substation metering became a badly unbalanced circuit.
It was common in that general area to see Y banks with one fused cutout hanging down and the bank functioning "Open delta".

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[pwrengrds]

The voltage phase to phase was 473.6/485.3/477.0. We took data at the motors again, the previous one had the CT's reversed. What stands out is that while the motor is unloaded two of the phases lag by about 90 degrees, the third lags by 60 degrees, see attached. This is true of motors in both buildings.

The utility has a 3 phase tap changer on the transformer feeding the circuit. The only voltage regulator is about 3 miles away, is connected Y, and feeds a separate branch circuit. It's not on that circuit.

They did another test off a separate building with it's own transformer, this one connected ungrounded Y/delta, ran the motor off that, still had the unbalanced load. The main load on that is a single phase transformer, about 50 amps, equal to 8-9 amps primary.

 

[waross]

Yup. Voltage regulator problem.
Disregard the Voltage regulators for a moment.
Unbalanced voltages and unbalanced currents go together.
If you have unbalanced currents that implies a neutral current. With a neutral current you will have a voltage drop on the neutral.
That voltage drop will generally act to reduce the voltage on the most loaded phase. That voltage drop acts to displace the neutral so that the current neutral point and the voltage neutral point are not congruent. This results in phase angle errors.
Now take a branch off of the line and install voltage regulators on the branch. In acting to restore the voltage on the most loaded phase, the regulator will sense and correct both the line drop and the neutral drop. This means even more current on the neutral and an even greater neutral displacement. With this neutral displacement comes greater phase angle errors and more line to line unbalances.
The first solution is to improve the load balance both on the main distribution line from the sub to your plant and on the branch line fed by the voltage regulators.
Ask the utility for a waiver on power factor charges, a reduction in your energy billing and a generous adjustment on demand charges as part of your demand may be occasioned by the unbalanced currents in the motors. This unbalanced motor current is working to correct the utility's voltage and phase angle issues so the utility is both benefiting from and charging you for partial mitigation of issues that they are responsible for.
Your motors are transferring energy from the lighter loaded phases to the heavier loaded phase and partially mitigating the problem.
Your motors are running hotter and hotter insulation means reduced motor life. That heat is also energy that you must pay for and energy demand that you may be charged for.
Try measuring line to neutral voltages and line to line voltages with no motors running and with as many motors as possible running.
I would like to see a comparison.
The Public Utilities Commission may be interested also.

Bill
--------------------
"Why not the best?"
Jimmy Carter