The mysterious tripping during transformer's energizing

[crisleon]

Hello, please consider and help me to solve my puzzle. All went well during first NO-load energizing of the 15kV/6kV 2,5MVA transformer via cable line (of L=400m, 3x150mm2 Alu, probably with inrush-current about 10*FLA, Sepam protection). However we got problem with the other loaded cable line supplied by the distant substation which is laid parallelly in the earth with this first cable about 200m along. The circuit breaker in this distant substation was tripped-off simultaneously with this 2,5MVA-trafo-energizing (by the protection unit VIP35 of Merlin Gerin, with settings Is=80A and Io=12A/0,6sec). The suspicion is towards this earth short-circuit protection because of lack of better proposal (the log of events from VIP is not accessible; we got the kick into the ass so the replay of the experiment is not possible to perform). Is this possible such an explanation and why? Maybe another cause?

 

[purplepete]

I am not a protection engineer but have seen three-phase cables (240 sq mm, one per phase) laid in pefect trefoil to a 415-volt motor with an earth conductor nested against the same phase for about 150 metres run. With about 400 amps phase current the earth conductor had 90 amps induced. We only picked up on this after the first motor bearing failure. I wonder if you have an earth conductor from the alternate supply in close proximity to a phase of the new one? Or whether induction is a possibility.

 

[sibeen]

It may have been caused by a "sympathetic Inrush".

 

[crisleon]

Thank you for your suggestions, I have to investigate this case once more.
Regards

 

[wcarter]

crisleon,

What kind of cabling is this MV line? It should have a external shielding that is grounded, preventing any inductance. The only time this is not true is if there is an issue with the shielding. Aka too high of a turn radius, or a nick in it. Has this cabling been resistance checked or hi-poted recently?

 

[crisleon]

Wcarter - certainly, these are 2 cables with 3 * single 150mm2-Alu cores, each one in XLPE insulation longitudinally sealed, with the copper 35mm2-shielding cores grounded on both ends of the line (and the common copper shielding core going back through the window of the Ferranti I0 transformer in the upstream feeding station). Cables were laid in 2008 and 2010, and hi-potting tests were mandatory in time of commissioning (probing U=37,5kV) as well as measuring of resistance of cores (all tests were positive).

 

[wcarter]

Even assuming they used pvc, vice wtr-xlpe insulation, there should not be a breakdown so soon in the cabling. With the proper grounding, the cables should not be an issue, nor should there be anything induced into or out of them.

I'd perform standard transformer tests initially. Whereas I am unsure if the downstream transformer is a dry or oil-filled transformer, I will state the basic tests for a dry (which are the same for an oil, just with some additions).

Meggar (resistance) high side to low side, High to ground, low to ground. Turns ratio. A doble (pf) test CAN be done in a dry type, though the results from these are question. The primary and secondary connections will need to be take off for these tests. Comparing the test results with your local code. ANSI/NETA in the US.

Whereas you are utilizing a Merlin Gerlin relay, I am assuming you are outside the US (which is my speciality). In the US, we will typically utilize a HRG or LRG in many cases with MV/HV transformers. But from this relay, the amperage is much higher than you would have from utilizing one of these systems.

Is the "inrush current delay" on?

Is this transformer a new installation with this being the initial energization? If so, I have seen some rather large inrushes in this case.

Idle curiousity, but what is the configuration of the supplying (distant) transformer, and the transformer having issues?

A distant odd other possibility is if someone say, DC hi-poted this transformer, your core could have become polarized. Rare, but not unheard of.

 

[Kiribanda]

crisleon,
Before answering I want to clarify the follwings please.
1)Is your VIP-35 feeder protection installed on the circuit feeding the 2.5 MVA transformer or on the adjoining feeder cable?
2)Because of the residual CT connection VIP-35 might trip on transformer inrush.Have you verified whether "inrush current delay" is set to ON?
3)Can you upload a sketch or SLD showing how VIP-35 connected to the feeder?

 

[crisleon]

Gentelmen,
2,5MVA trafo (dry type Trihal) is protected by Sepam. It is completely new and was tested by the producer's representative (Schneider) and seems to be OK. We can't test it onece more now because it is still working (after the energizing problematic start we have no troubles with it).
VIP35 protects the outlet in completely another trafo station via the circuit with the CRc relays (a type of Ferranti transformer, as far as I know) and it trips the circuit-breaker by means of the mitop realease. Inrush current delay is set to ON. Earth fault setting is 12A/0,6sec, overcurrent setting is 80A.
The installation is in Poland (Warsaw's region). I am an electrical designer in charge of the proper installation and protection of this 2,5MVA trafo which is intended to the future cooperation with the generator-set.
Nowadays the problem is sound-proofed, however I am afraid of the next case when the energizing will be necessary to perform once more, in time of connecting this generator.
Regards

 

[wcarter]

Schneider is usually pretty good with their FSR's. Do you have the test reports from their onsite testing? If you were able to re-energize with no issues, I would look toward's the core of the trafo having become polarized (which is not unknown). This would also account for a trip on the first energization (which depolarizes the transformer, but causes a large currst spike), but no trip on subsequest energizations.If they had gotten the pasing wrong, that could have also shown the initial issue, but it would then also show up in your subsequent energizations.