Cause of fault that's producing low current on one phase 1

[electrickiwi]

I have a 3-ph synchronous generator, 500 kW 415 V, feeding directly onto a (weak) grid through a standard 800 kVA, 415/11 kV distribution transformer (Delta HV, solidly grounded Star LV) via a contactor and circuit breaker. The system is new and from first synchronisation, current on phase L1 is 80 to 100 A lower than the other two (this is with output between 60 and 250 kW). See attached waveform. Protection relay trips on 50Q (negative phase seq overcurrent).

No significant voltage imbalance, either on or off-load, and everything looks to be in phase. I only have access to LV measurements.

The only other time I've seen something similar has been due to a transformer phase fault on the HV side (blown fuse I think), and because of the Delta-Star winding everything looked fine on LV until it came under load. I need to prove the cause of the fault here though, because it may not be the same.

Is there anything I can tell from protection relay event captures (see screenshots attached) or by simple on-site tests that would confirm the most likely cause? The transformer belongs to the distribution network operator and it's remote so not much option for direct investigation there at short notice.

My next step will be to bolt a 3-phase short between generator and transformer, then manually excite the generator to 300-400 A and see if the problem persists. That should isolate whether it's a generator-side (my problem) or transformer-side (not my problem) fault... Anything simpler that might narrow things down??

I'd appreciate any troubleshooting help on this one. Cheers.

 

[waross]

The voltages and phase angles near the end of long distribution circuits tend to get a little ratty. A combination of little or no transposition, distributed single phase loads and the neutral voltage drops that they cause give rise to unbalanced voltage drops and phase angle errors. Voltage regulators are often culprits. Downstream from a bank of phase to neutral voltage regulators, the phase to neutral voltages will be very close to equal, but if the voltages on the upstream side of the voltage regulators are not balanced you will have phase angle errors.
This is easy to check; equal phase to neutral voltages but unequal phase to phase voltages is a sure sign of phase angle errors.
Check your grid voltages where you have access. Check both line to line and line to neutral.
If possible roll your phases. If the problem is with the generator, the current abnormality will stay with the generator phase. If the problem is with the utility the current abnormality should stay with the grid phase.
Note: If the four wire wye delta transformer may be energized without being connected to the grid, it should show up any generator fault. The wye primary with neutral connected:delta secondary transformer abhors voltage or angle inequalities. If you are able to energize this from the generator with no issues before connecting to the grid then the issue is almost certainly a grid issue. Any inequality of voltage or phase angle on the wye primary will drive a circulating current through the delta limited by three times the transformer PU impedance and the generator impedance.



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

 

[VTer]

What type of loads do you have on your LV system and what are the voltage and current readings on the LV side when the generator is not running?

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic — and this we know it is, for certain — then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". – Nikola Tesla

 

[stevenal]

New system with new CTs? I'd be checking the CTs for correct ratio and shorts removed.

 

[electrickiwi]

stevenal, that was indeed considered but have confirmed our measurements with a separate power quality meter and current clamps.

VTer, very low loading, 10-15 A/phase. At that level (i.e. when not generating) currents are balanced within a couple of A.

Bill, thank you, after submitting my post I realised that rolling the phases would give the same result for less effort so that'll be the first step. Unfortunately the transformer is network operator equipment (and generator can't black-start) so can't energise it offline.
Voltages on the LV are balanced (L-N and L-L) when we're not generating (and even when we are) within 2-3%. I don't have access to MV measurements at the moment.
My transformer knowledge is limited - for one in this configuration, what would the LV side look like (V mags and angles) if a phase was lost on the HV side? Could one tell simply from LV measurements? Would the LV look different under load vs no-load?

Thanks!

 

[VTer]

So when you run the generator, what additional loading is placed on the generator? Are you exporting to the utility?

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic — and this we know it is, for certain — then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". – Nikola Tesla

 

[electrickiwi]

Inconclusive testing from the phase rolling... fault looked much different to original (significantly more harmonics and varying phase-angle shifting of current waveforms that make me suspect something inadvertently altered on site - unfortunately I'm only able to be involved remotely).

To try for a more conclusive result, will be hooking up a large load bank to independently load the generator and the transformer (and confirm measurements with an independent meter). This should isolate which end is at the root of the problem.

QUESTION: If it happens to be a grid imbalance problem or a transformer fault, should we expect to see the same fault symptoms on the LV side when DRAWING 300-500 A into a balanced resistive load as we saw when PUSHING that current out from the synchronous generator? That is, could the load bank test still mask/hide a problem??

Thanks.

Ps. VTer, we export up to 500 kW to utility when synchronised. Approx 10 kVA additional in auxiliary station loads supplied by generator when online.

 

[waross]

Try floating the wye point or temporarily resistance grounding it. Do you have current flowing on the neutral?

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

 

[electrickiwi]

To update on where things got to, a large load bank was hooked up and used to sink current from the grid, and separately, current from the isolated generator. The generator test showed textbook waveforms, and the grid test showed minimal imbalance of 2-3%.

So the issue is only apparent when pushing power onto the grid via the transformer. Nothing apparent when drawing same amount of power off the grid. Does that point to a specific grid issue?
Still need to get a reply from the network operator...

 

[waross]

Are all three taps on the same position on the transformer?
Can you check the kW and KVA exported on the individual phases?


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

 

[PHovnanian]

Weak grid?

Any chance there is a capacitor bank down the road with an open phase?

I'd be a little careful about reflecting the voltage and current readings made on the wye side of a delta-wye transformer to the delta side.

 

[pwrtran]

1) There are negative and zero sequential components involved, which indicates there is a path between a phase or two phases to the ground

2) The phase angle between the negative sequence voltage and current cannot be greater than 90 degree, they might be due to either not from the same phase or the current vector need flipped. So find out individual matching phase voltage and current sequential vectors would help

3) 0.7% voltage unbalance right from the xfmr should not pass the factory test, so it probably from somewhere else in the field

4) if the harmonic patten is non-characteristic it then could be caused by intermittent arcing or kind of conducting, so I would also check the LA's

 

[fx007]

I could bet the neutral conductor from your transformer to the motor is too small. Pls check and let us know...

 

[iceworm]

electrickiwi -
I did read all of the posts - here's my understanding:
The gen is 415V Wye, solidly grounded
The GSU xfm primary is 415V Wye, neutral connected.
The GSU secondary is 11kv Delta
The 11kV system is what? impedance grounded? (doesn't matter much - just curious)

If so -
waross hit it on his march 10 post:

Try floating the wye point or temporarily resistance grounding it. Do you have current flowing on the neutral?

My experience is mostly off grid, self-generated industrial - and I'd say you have the wrong transformer. Your's was made for stepdown. The one you would like to have is 415D/11kvY - or maybe a 415D/11kvD depending on your 11kv system. But, the crew installing the gen/GSU xfm put in what they had available and 11kvD/415Y is likely way more common than the other direction. So, you deal with what you got. So float the transformer primary neutral.

This is just a guess, if it was 13.8kv and 480V - I could tell you for sure. Although I highly suspect that 11kv/415V, 50 hz follow the same physics.

Or, my assumptions are not true, and I'm all wet - that happens sometimes.

ice

 

[VTer]

When you are exporting power you have an 11kV delta ungrounded system but the utility system might be a grounded wye 11kV/6.35kV system feeding other single phase loads. Check with the utility what their load distribution is on that feeder.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic — and this we know it is, for certain — then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". – Nikola Tesla