Transformer saturation-driven neutral instability

[Bronzeado]

Hi folks,

Has anybody experience on neutral instability during the energizing of a wye-delta transformer?

The transformer is energized throught the high voltage (230 kV) of which windings are wye connected and the neutral grounded.

The low voltage (69 kV) windings are conected in delta and, therefore, have a "ficticious" neutral point.

My questions are:

1. How this "ficticious" neutral point behaves with relation to the grounding during the transformer energizing?

2. If there is a TP on each phase of the delta side with its primary winding connected between the phase and the grounding, what kind of voltage I will get from this TP?

3. Is it possible to get a phase to grounding voltage bigger the phase to phase voltage?

I would appreciate your thoughts!

Best regards,

Herivelto Bronzeado

 

[waross]

If the delta secondary is open circuited, and not grounded, you may be getting capacitive coupling with the 230 kV winding.
This could result in voltages to ground greater than the phase to phase voltage.
Also,a wye-delta transformer experiences switching surges when energized. I understand that these surges can reach 200%, but I may be corrected here. If your wye grounding does not have good continuity or has a high impedance back to the source, you may be getting switching surges. These surges are transient and will dampen out.
Is your 230 kV/69 kV transformer large enough that the potential transformers may not be able to drain the capacitive voltage from the 69 kV winding?

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

 

[magoo2]

From my standpoint, that's a pretty unusual connection for a 230 kV transformer. Normally, 230 kV is a 3-wire circuit. Most transformer connections from 230 kV that I'm familiar with are delta wye.

How do you establish a ground connection at the transformer? There's no way to tie it back to the source unless you use a static wire. In any event, it's probably a fairly high impedance path to ground. For this reason, I would be tempted to treat it more like a floating wye delta connection.

On the distribution scene, this connection has been shown to cause overvoltages due to single-pole energization, with phase to ground voltages of 2.65 per unit possible. How are you energizing the transformer? I'd imagine you're using a circuit breaker or gand operated switch. If that's the case, then I wouldn't worry about getting the 2.65 pu value.

Did you have any arrester failures associated with this?

Finally, what's a TP? Terminal pole doesn't seem to fit.

 

[Bronzeado]

Dear Bill and Magoo2,

For your information, I am talking about a 100MVA, 230/69kV transformer wye-delta connected. The neutral point of the wye (230 kV) is solid grounding.

In general, we have more that one transformer with the same connection in a substation. Normally our delta system (69kV) has a grounding reference given by a zig-zag transformer.

When the second transformer (100MVA, 230/69kV, wye-delta) is energized by closing a 230kV circuit breaker, the delta (69 kV) will be open (including a bay of 50-100 m of cable from transformer bushing to circuit breaker) for some time before closing the 69kV circuit breaker that inserts the transformer to the system. My question is related to what happens with the "ficticious" neutral of the delta side during this period of time (transformer energizing).

May this "ficticious" neutral point reach some "positions" against grounding in which the phase-to-ground is bigger than phase-to-phase voltage?

Please, send me any literature or thougth on that. I need to anderstand this phenomenon.

Thanks Bill for you input. I will consider the capacitive coupling between the windings.

Best regards,

Herivelto

 

[davidbeach]

First off, I'd say that your transformer connection is quite common, 69kV systems tend to be delta sourced relative to the "infinite bus".

Your fictitious neutral will tend to be somewhere within the voltage triangle. The placement of that neutral will be determined by the capacitance to ground from each phase. To the extent that those capacitances are balanced (they won't be) your neutral will be in the middle of the triangle. The only way to get the neutral outside the triangle is to charge those capacitances, open and then reclose out of phase with the stored charge.

 

[Bronzeado]

Thank David for your reply.

Please, have in mind that the transformer is being energized, with the poles of the circuit breaker closing in different instants. So the "capacitances" involved may be charged differently of each other.

Also, inrush currents of high magnitude will flow throught the transformer primary windings (230 kV, wye) and may go into the ground.

Best regards,

H. Bronzeado

 

[davidbeach]

Are you trying to get to the root of specific misoperation or is this more general?

 

[Bronzeado]

David,

We have a lot of transformers wye-delta (230/69kV) in our system. During transformer energizing, it has been found phase-to-ground overvoltages in the delta side (69 kV) which remain for some time. These overvoltages seems to be caused by:

a) Neutral instability or

b) Ferro-resonance

What I expect to get from this forum of experts is some technical information, some light, to explain this phenomenon.

Bill (see above) has given some tips which I will work on that.

The voltage measurements comes from PTs with their primary windings connected betweem each phase of the delta system and the ground.

Best regards,

H. Bronzeado

 

[davidbeach]

What is the primary rating of your VTs? If it is 40kV, rather than 69kV, the VTs will be going into saturation with the peak of voltage of each phase. Do you have any physical evidence of overvoltage on the 69kV system or are you only seeing overvoltage on the secondary of the VT circuit?

 

[Bronzeado]

David,

The VT primary rating is 69kV.

So far, we have not seen any physical evidence of overvoltage on the 69 kV, only on the secondary of the VTs circuit.

Regards,

H. Bronzeado

 

[davidbeach]

Try adding some resistive loading to your VT secondaries. My guess is that you don't have a problem at the 69kV level, but you do have a problem on the VT secondary. Once the transformer is connected to the rest of the 69kV system you could even disconnect the loading resistance from the VT.

 

[Bronzeado]

Thanks David for your advice.

That is one solutionn for ferro-ressonance. Is it the phenomennon that is happening?

I would like to know a physical explaination on the phenomenon that is really happening.

My guess so far from discussions on "instability" of the neutral point of the delta side are:

1. Voltage rise of the substation grounding network during the transformer energizing due to the high inrush currents going into the ground network.

2. Voltage transfer due to the capacitance divider formed by the capacitances between the windings (high and low voltages) and the windings (delta) and ground (Bill suggestion as far I anderstood).

As a consequence, the delta neutral point will be at a "position" at which phase-to-ground overvoltages may impar the equipment, includind saturation of the PTs and, perhaps, the power transformer.

If ferro-ressonance appears, it should be the effect not the cause.

Do you agree with these comments?

Regards,

H. Bronzeado

 

[davidbeach]

Capacitive coupling at 60Hz should be quite low; at lightning type high frequencies the transformer looks like a capacitor between the two sides, but not at the power system frequency. Do you have relays on the 69kV VT secondary circuit that can be programmed to capture a waveform of the voltage as the transformer is energized? Knowing what the voltage looks like could help in figuring it out.

 

[slavag]

Hi Herivelto.
Are you have some news on the issue?
Could you please explain why you start with this Q's.
My collegue sent me today other case ( but on the same issue), he see this is ferroresonance problem on the VT. But in case of two VT's on the BB w/o any load ( 10kV). In case of one VT, no problem, and this case occure only on the one SS.
Are you have this effect in the special place or on the few installation?
Best Regards.
Slava

 

[slavag]

Hi Herivelto.
Are you have broken delta connection in the your VT?
I haven't documents on the English.
It's problem of small C ( you need add some 150-200m of cable for avoid this effect). In the Russia lot of MV grid is isolated neutral and this effect is very known.
Neutral point driven, but of VT.
It's effect of saturation of VT cores, and isn't same saturation per one phase VT, C and L are changes up to current resonance (C=L). This also move to increas of phase to ground voltage.
Hope that help.
Best Regards.
Slava

 

[Bronzeado]

Hi Slava,

I will try to get some voltage waveforms (69 kV) in order to show a better picture of the phenomenon we talk about.

I still thinking that ferro-ressonance (if it happens) may be an effect of the phase-to-ground overvoltage, NOT the cause.

Regards,

Herivelto

 

[slavag]

Hi Herivelto
We wrote togehter .
I haven't option now check this effect, but in the begining of 2009, we'll start commissioning of some SS, where we have this option, but it will be 11kV.
Regards.
Slava

 

[ePOWEReng]

This sounds like a good application for an EMTP like ATP. Have you tried modeling this in an EMTP program?

 

[davidbeach]

Without very good information on the saturation characteristics of the transformers there'd be no point in an EMTP study. Good saturation models can be very difficult to come by. Sure, you could use some generic saturation information, but it might not do a lot of good for the very specific case in question.

 

[prc]

Bronzeado,I came across this phenomena some 25 years back and had to struggle for several weeks to understand it,find a solution etc.Great American engineers of GE & WH had studied this and 4-5 AIEE papers ( of 1940's) describe this.They called it as neutral inversion- some form of feroresonnance.I had all of them,but at the moment I find it difficult to locate the titles.I shall try.The solution is to connect a capacitor to delta lines or close the secondary delta of CTs through a resistor.But once you have zig-zag trfs connected,this will not be a problem.