Phase imbalance on transformer energization?

[peebee]

I'm dealing with a 1500kVA 12kV:277/480v step-down transformer with a bolted pressure switch (BPS) installed nearby. This is on a generator distribution system which is energized for about 10 hours daily & shut down the rest of the time.

There is a phase loss/imbalance relay in the BPS which trips about half the time on energization, even when set to max time delay of 6 seconds. The relay holds in when the switch is closed after energization. The relay was replaced once, so it's unlikely that the relay was the problem.

Any thoughts? This is a purely academic question now as the relay has been disconnected.

 

[1x1x1x]

Peebee,
Because the transformer inrush current has, a very high content of DC component and a negative sequence second harmonic the bus voltage will be distorted during switch on.
The relay was probably tripping, because the large percent of second harmonic current caused a second harmonic voltage (negative sequence) distortion on the bus.

 

[peebee]

You think that'd go past 6 seconds? Any better ideas than disconnecting the relay?

Also, by the way, we have another similar installation on site, but the tx is smaller and there's more cable between the tx and BPS, and no such problem has ever been experienced.

 

[DanDel]

I'm not clear on your system. I think you are saying that the BPS is on the 480V side of the transformer, and it is usually closed(when the system is de-energized). Do you energize the transformer from the primary side(you said step-down transformer)? Where is the generator in this system, on the 12kV side? What is the manufacturer and model of the relay, and what is the load on the BPS?

 

[peebee]

There's three paralleled 765kW, 277/480v generators which are stepped up to 12.47kV for distribution to four step-down transformers. The 1500kVA TX is one of the four step-downs. The BPS is on the 480v secondary of the TX, normally closed, manual operation, and typically would stay closed indefinitely. The BPS feeds a downstream two-breaker automatic transfer switchgear (TS). Demand load on the TS is about 1000kVA The three other locations are similar but smaller TX's.

Normal daily operation is to start & synch the gens, close to the step-up tx, and sequentially transfer the four TS's to gen on 20-sec intervals. The BPS trips out from the phase-loss/rotation relay on tx energization prior to transfer to generator while the tx is still unloaded (some uncertainty here due to distances & communications issues involved). Once the tx is energized, the BPS may be manually reclosed and will hold in.

I don't have info at hand on the relay mfgr., but can get it. We are planning on temporarily installing a PQA, this has not yet been done. I mistakenly referred to the relay as phase-loss/imbalance in my original post, it is actually a phase-loss/rotation relay.

 

[peebee]

As a slight aside. . . .

This relay is not critical as it is not required by code. It would not break my heart if our final solution is to keep it disconnected.

There's some argument over how bad phase loss would be. I'm under the impression that single-phasing could damage motors prior to motor overload heaters tripping. Others swear that heaters will adequately protect the motors. Any thoughts?

 

[busbar]

The relay producer should be able to provide specific out-of-circuit instructions for testing. Whether or not the device is operating as claimed should be verified.

 

[DanDel]

peebee, so when the BPS trips from the relay, there is actually no load on it because the TS is still on the normal position?
This phase loss/rotation relay sounds like it might be a 47 (Phase-sequence voltage), which checks for correct voltage sequence. Please provide nameplate info if possible.
What is the distance from the BPS to the TS?

 

[busbar]

Some 47 {negative-sequence overvoltage} devices have an inverse-time-voltage characteristic; id est, the greater the imbalance, the faster it operates. Is that the case here?

 

[1x1x1x]

Peebee,


According to the Applied Protective Relaying Book by J.L. Blackburn (Westinghouse):
The resistance from source to transformer determines the damping of the inrush current.
Transformers close to the generators will have a much longer inrush because the resistance is low.
The inrush current varies from 10 cycles for small units to much as 1 minute for large units.

Based on my experience the inrush current can be easily few second long at power generating station.
However if a nearby capacitor bank forms a parallel resonance in the region of any integer harmonics of the inrush current, the resonance effect will amplify the voltage distortion, and the distortion will exist much longer.

If you can block or bypass the tripping for few second of the unbalance relay during transformer energizing it probably will mask the problem.
Another possibility is to measure the voltage to the relay using differential probes and oscilloscope. The oscilloscope shall be “pre-triggered” by the unbalance relay tripping. The waveforms will show if the unbalance relay is triggered by harmonic distortion.

 

[jbartos]

Suggestion: An oscilloscope or similar device can reveal the transient waveforms. The DC offset is expected to be noticeable within short proximities of generators. However, other causes potentially exist as indicated in some postings above.