Trouble with a 5KV circuit.

[SteveO NE]

Hi,

Hopefully you guys have some ideas about what's going on because I'm hoping we can get to the bottom of this problem sooner than later.

So, here is the skinny:

We have a 5KV "transmission" circuit that we are unable to energize.

Background and story: the purpose of the circuit is to span about 1500ft with minimal line loss/vdrop out to a PV array. The circuit is comprised of a 1MVA 480v wye primary to 4.8kv delta secondary, 1500ft of 500kcmil AL 5kv rated conductor, then terminating in a 4800v primary, 480 secondary 1MVA transformer.

The first test of the circuit failed with the breaker holding about a minute and tripping out. Over the phone my thought was an issue with cascading transformers (possibly inrush or zero crossing) so we tested lifting the mv conductor and seeing if we could energize just the first transformer. This worked without an issue. Next test. We megged the lines and we definitely had a problem, turns out the installer had created a high impedance fault with a conductive layer of the conductor to the concentric shield. The initial closing in on the system had smoked the ends of the conductor so new conductor was pulled.

The new conductor megs out good, but everything has been cautious in testing here because its not a cheap mistake to fault out another pull. Tested first xfmr alone and it powers up. Tested transformer with conductor attached ( floating on opposite end and not attached to the transformer) and it trips out instantly. Again megging phase to phase and phase to shield tests out OK. An instant trip to me would seem to indicate the GFCI is tripping vs. Current overload.

The transformer has it's own GEC and both the 480v and 4.8kv equipment (including the shield conductors) all come to a common ground bus in the transformer and the 480V EGC also goes back to the service GEC. Not quite as designed but it's where we are currently.

I have thoughts about the issue and put together some procedures for testing but I wanted to see if anyone else has any ideas without the bias of what I am already trying.

Thoughts/questions?

 

[GoldDigger]

First thought: Do not connect, nor even ground, the 480V wye primary wye point/neutral. A delta-delta transformer at the POCO end might be more appropriate. Or else a delta-wye to allow the transmission line to be symmetrically grounded. You probably need a wye on the inverter side if they source line to neutral or monitor line-neutral voltage.
Next: Are you corner grounding the delta transmission line or floating it? Floating is a very bad idea since a re-striking ground fault can develop damaging voltages to ground (multiples of 4.8K !).
If connecting the transmission cables causes a problem yet there is no measurable cable fault, it might be possible that the unloaded cable capacitance is causing a ferro-resonant condition on the source transformer.

 

[gar]

190621-1652 EDT

SteveO NE:

Start testing with a greatly reduced input voltage, and look for abnormal current and voltage. Start at a voltage where no tripping occurs.

.

 

[GoldDigger]

190621-1652 EDT

SteveO NE:

Start testing with a greatly reduced input voltage, and look for abnormal current and voltage. Start at a voltage where no tripping occurs.

.

:thumbsup:

 

[GoldDigger]

First thought: Do not connect, nor even ground, the 480V wye primary wye point/neutral. A delta-delta transformer at the POCO end might be more appropriate. Or else a delta-wye to allow the transmission line to be symmetrically grounded. You probably need a wye on the inverter side if they source line to neutral or monitor line-neutral voltage.
Next: Are you corner grounding the delta transmission line or floating it? Floating is a very bad idea since a re-striking ground fault can develop damaging voltages to ground (multiples of 4.8K !).
If connecting the transmission cables causes a problem yet there is no measurable cable fault, it might be possible that the unloaded cable capacitance is causing a ferro-resonant condition on the source transformer.

To elaborate on the primary neutral connection issue: If the line to neutral voltages from POCO are not perfectly balanced and you connect the wye point of the winding, the unbalance will cause circulating currents in the delta secondary and in the primary windings that can be many times the full load current. This could cause a very fast trip on energizing. Do you have a cause of trip indication on the source breaker?

 

[electrofelon]

A few questions:

1. The transformer closest to the utility, you didnt specific the winding configuration.

2. I assume the 480 stuff has GFPE. Is it tripping on GFPE or magnetic? What exactly is tripping and on which end?

3. I am a bit unclear about this, perhaps you could clarify:

The transformer has it's own GEC and both the 480v and 4.8kv equipment (including the shield conductors) all come to a common ground bus in the transformer and the 480V EGC also goes back to the service GEC. Not quite as designed but it's where we are currently.

 

[synchro]

I can understand being cautious after having to replace the mv conductor, and therefore not connecting the second transformer yet. But if ferroresonance is occuring as GoldDigger pointed out, it may be possible that connecting the second transformer would introduce enough loss into the system that ferroresonance and subsequent breaker tripping would no longer occur. If so then more testing with the second transformer disconnected might not be the best way to go forward.

GoldDigger also makes a good point about the problems of connecting the neutral point on wye primary.
Quoting from the Eaton Power Distribution Systems document about the wye-delta connection:
"Grounding the primary neutral of this connection would create a ground source for the primary system. This could subject the transformer to severe overloading during a primary system disturbance or load unbalance."

 

[iceworm]

Following Ethan's post:
Exactly which breaker is tripping?
Are you energizing this from the PV end?

What does the one-line look like?

PV (with inverter) (480V, 3ph output - Is this Wye or D? Grounded or un-grounded?) -->>-- CB -->>-- xfm (480V wye Primary - Hopefully the Wye point is un-grounded - floated, not connected to anything)/( 4800V D secondary - Un-grounded with detectors? Corner grounded?) -->>-- 4800V fused disconnect -->>-- Line to second transformer -->>--

xfm (4800V D primary)(

480V wye Primary ) -->>-- 480V CB -->>-- connection to power system (possibly utility)

If not something like this, what then?


​

​

 

[Cow]

Who is doing the megging? Are they qualified to do so? Are they testing at the correct voltage, time, etc?

We hire a 3rd party testing agency for our MV installs and they will give us a report when they're finished, if we request it.

 

[Julius Right]

In my opinion, the capacitive currents are symmetric if the shield of all phase is grounded.
If a phase shield is not grounded then only a single capacitive current exist between the two
grounded shields.
The single core cable capacitance is: C=2.п.ε/ln(Dins/dcond) F/m
For 5/8 kV cable 500 MCM Dins=1.01 inch dcond=0.748 inch
ε=εr.εo=2.3x8.85E-12 F/m
C=4.25885E-10 F/m .For 1500 ft=304.8*1.5 m CRG=1.94714E-07 F
Two capacitance connected series will be 1.947e-07/2=9.73572E-08 F
The capacitive reactance is 1/( 2.п.f*C)= 27245.9 ohm
IcapHV=4800/27245.9=0.176 A
At 480 V the current-flowing in a single phase induced from HV-it is
4800/(480/sqrt(3))*0.176=3.05 A
GFCI is tripping at 30 mA

 

[electrofelon]

Hello OP?? Any updates or responses?

 

[sameguy]

Hi-pot test?