autotransformer with impedence ground

[badasselectric]

I have a 750kva autotransformer 480v/1000v, it has a 15a 600v resistor connected to the xo/ho terminal. The autotransformer feeds 2 400a/1000v breakers equipped with shunt trips and ground monitoring. I have experienced 3 faults with damage to a phase conductor in the trailing cable and the ground fault monitor has not tripped. I don't believe this is the proper application for an autotransformer. What really bothers me is even if the fault went all the way to our distribution transformer why didn't the ground fault monitor trip? its looking for 5a.

 

[PWR]

Is the autotransformer supplied by a 480Y/277V Source? How is that source grounded? How does the "ground monitor" sense a fault? If the source is solidly grounded, I doubt if hardly any current will appear in the 15A resistor. In any case, it does sound like it will be tough to make it work without isolation of the 1000V system from the 480V system(which, of course, an autotransformer does not accomplish).

 

[Bung]

Assuming no delta winding, the Zo of the transformer is

Zo%=((N-1)*Zps%/N)+3Zn%*((N-1)/N)^2
(ref- old copy of Westinghouse T&D book)

Zn is the per unit value of the grounding resistor.

If a '15a 600V' resistor translates to 600/15 ohm, Zn is 3000% (100%*40/(1000^2/750000))

This suggests a very low earth fault current (<14A, or 1/30-th of 750kVA at 1000V by my reckoning), probably way below pickup of your ground monitoring.


Bung
Life is non-linear...

 

[mc5w]

If the X0 terminal (neutral) of your supply source is not connected to H0X0 of the autotransformer, the autotransformer will act as a choke that tend to keep downstream ground faults from drawing any current.

If this is an installation that is subject to U.S. National Electrical Code, it is an illegal installation. The supply neutral can really only be used to supply revenue meters and power monitoring devices that need a true Blondel configuration for accuracy purposes.

 

[badasselectric]

The autotransformer is supplied by a solidly grounded 480Y/277V 2000kva transformer. The ground fault monitors have a c.t. around all three phase conductors for that circuit, they are calibrated to trip at 3-5 amps. With this type of ground fault sensing shouldn't the relay trip even if the current is going back to the distribution transformer? With this configuration it would seem the grond fault current will not be limited to 15A by that resistor.

 

[PWR]

You probably have not carried the neutral of the 2000KVA transformer to the neutral of the autotransformer as mc5w commented. Thus, when a ground fault occurs, you have no completed path back to the source. If you do complete the neutral path it will be a solidly grounded one and the 15A resistor will continue to have no effect.

 

[Bung]

I seem to missing something here. Can you confirm my understanding of the circuit :
1] source is Y connected, neutral (star point) solidly grounded.
2] autotransformer has the star point connected to ground through a resistor - please clarify its resistance
3] ground fault is occurring on the load side of the autotransformer.
4] autotransformer has no delta winding. Is it in a steel tank?
If this is how it is, then the grounding resistor will very definitely have a marked effect. Also, the higher the value of the resistor, the greater the (relative) effect of the "tank delta".

Where the current is "going back to" is only half the story. It also has to come from somewhere, and that somewhere is the phases. It will result in a zero phase sequence current flow which your core balance CT arrangement will pick up if there is any ground fault current flow.


Bung
Life is non-linear...

 

[badasselectric]

Bung,
I'll do the best I can, thanks for the input so far.
1. correct
2. correct, the nameplate on the resistor says 15A,600V only.
3. correct
4. correct, it is a dry-type winding.

I understand that the current is coming from the phases, I fail to see what role the grounding resistor plays in limiting GF current when the phases are not isolated in the autotransformer. The ground fault monitor is setup to trip between 3-5 amps flowing through the ct, when i inject 5amps through the ct it trips like clockwork. It may be that I am not getting enough fault current during a fault, or my relay should be setup for .25A . just a thought.

still pondering,

 

[Bung]

Auto-transformers are messy things - the primary and secondary may not be electrically isolated, but the magnetic circuit means that you must still have the correct ampere-turn balance in each winding. And at very high values of grounding resistance, the eddy currents induced in the tank (if there is one) make the tank appear as a pseudo delta winding, modifying the ampere turns balances somewhat. The old Westinghouse T&D book is a good reference.

The 15A refernece on the resistor is ambiguous.Presumably it is intended to limit the earth fault current to 15A, with a maximum working voltage of 600V (ie maximum dissipation of 9kW).

Your ground fault setting sounds a bit on the high side, as it is only 1/3 of the maximum ground fault current (which I am assuming is the 15A from the resistor spec). I would be more inclined to look for a 1/6 - 1/10 ratio. But don't try changing it without proper, competent analysis by somebody who has access to all the facts! I can only see what you have told us here!

Bung
Life is non-linear...

 

[davidbeach]

...the primary and secondary may not be electrically isolated...

By definition of an autotransformer, the primary and seconday WILL NOT be electrically isolated. Because the two windings are electrically connected, there will be zero sequence currents flowing through the transformer winding from primary to secondary. Absent a delta tertiary the zero sequence network does not have a connection to neutral, looking instead much more like the zero sequence network of a grounded-wye grounded-wye two winding transformer.

 

[PowerfulStuff]

My experience with Neutral Earthing resistors are that the nameplate doesn't imply anything about the resistance i.e. in this case the resistor may be rated to carry 15A continuously and insulated from the case up to 600V. The resistive element itself probably cannot handle 600V directly across it except for a very short time. The resistance would have to be measured.

 

[badasselectric]

we are going to correct the problem as follows:
1. install a 277V 15A resistor on the xo of the supply transformer,as there are no 277V loads on this transformer.
2. change out existing zero sequence monitors with trip setting of 3-5A, to an adjustable trip setting from .25 to 2A.
3. float xo at autotransformer.
4. set GF on 2000A breaker in service disconnect to 5A.

Thanks for all the help everyone. badass