Zero sequence impedance of transformers 4

[HiSet]

Should we considering zero seq. impedance of YN/yno power transformer(HV star point unearthed but LV directly earthed)in calculting ground fault current contribution ? or will positive sequence impedance will suffice.

 

[jbartos]

Suggestion: This transformer circuit for short circuit study normally has the HV ungrounded wye infinite impedance (open) circuit or very high zero sequence impedance in the zero sequence electrical equivalent circuit.
Reference
1. W.D. Stevenson, Jr. "Elements of Power System Analysis," 3rd Ed., McGraw-Hill Book Co., 1975

 

[RajT]

If the prim is grounded as you state then you should consider Z1+Z2+Zo when calculating ground fault curr5ents.

 

[bigamp]

HiSet,

For a YN/yn configuration, with no tertiary winding, and the YN starpoint unearthed the zero sequence impedance will depend on the type of transformer construction. For shell type, the zero sequence impedance will be very high (some 100's %) and the earth fault current available on the LV side will be very low. For core type transformers, the zero sequence impedance will be around 50%, so you will get reasonable earth fault current. To answer your question, you can not just use the positive sequence impedance. If the HV starpoint was earthed then you could say that the zero sequence impedance was about the same as the positive sequence impedance.

There is an old book, but a very good one by Alvin H Knable "Electrical Power System Engineering", McGraw-Hill 1967 that discusses the subject well. A lot of texts make the statement that for a YN/yn transformer with ungrounded HV starpoint the zero sequence impedance is infinite and in this they are not correct. A number of electrical analysis software packages make the same assumption too.

 

[jbartos]

Suggestion: Reference:
Alvin H Knable "Electrical Power Systems Engineering", McGraw-Hill 1967 provides in Figure 3-2 on page 76
a fictitious tertiary windings to arrive at the model for wye-grounded wye transformer. Namely, the fault current IFA=3I1=3 x 1 /1.20=2.5 p.u.
This appears to be somewhat empirical result.
Compare the result with Reference:
Electrical Transmission and Distribution Reference Book, by Central Station Engineers of the Westinghouse Electric Corporation, East Pittsburgh, Pennsylvania, 1964,
Table 7 Description A-2 and A-3 approaches

 

[HiSet]

Bigamp,
Thanks for info.Just to let you know that this is a core type transformer without delta tertiary and I did following measurements to arrive at four different zero seq ohmic values namely
Voltage injection on primary winding between unearthed neutral and all three phase shorted with secondary open and in the second test, keeping secondary winding shorted.Third and fourth test is same as above but with voltage injection done from secondary side. Question I guess is how to compute one zero sequence value from these four set of readings.

 

[rbulsara]

HiSet:

What 'contribution' are you talking about? Which side is the ground fault?

If the primary is ungrounded (or even grounded), there will not be any phase to ground fault current on the primary side, if a ground fault takes place on the secondary side! Current only return to its source.

There will only be positive or negative sequence currents on the primary lines, due to a seconday side fault.

Sequence impedances have no physical meaning, they are just 'invented' for mathematical modeling. If there is no phase to ground current, zero sequence impedance is infinite , it has to be.

 

[HiSet]

rbulsara
For sake of clarity I am assuming a feeder fault on transformer LV side which is solidly grounded.Primary star in ungrounded.Zero sequence current on primary side has to flow to maintain amp turn balance,if it's grounded.But what happens to HV contribution if it's ungrounded as now amp turns balance is not maintained?

 

[jghrist]

How can there be zero-sequence current flowing in the secondary of an ungrd wye - grd wye transformer? If there is no connection to the primary neutral, then the currents flowing into the neutral (the primary phase currents) must add to zero. Zero-sequence current is 1/3 the sum of the phase currents, and therefore is zero.

The secondary currents are related to the primary currents by the turns ration n, a scalar which is the same for all phases. If primary currents are:

IA + IB + IC = 0, then the secondary currents are:
Ia + Ib + Ic = n·(IA + IB + IC) = 0
I0 (secondary) = (Ia + Ib + Ic)/3 = 0

Therefore, the zero-sequence impedance must be infinite. This is consistent with Westinghouse T&D Handbook Table 7, Fig A-2 and A-3 with infinite Z_GP.

 

[davidbeach]

If you have a grounded wye winding of a transformer, you can have zero sequence currents on that winding. Does not matter what the other winding is. It is true that there would not be a zero sequence to the primary in the situation outlined. The zero sequence currents exist on the secondary side, and there will be corresponding negative sequence currents on the primary side. But the transformer has a zero sequence impedance on the secondary side, and it is far from being infinite.

 

[bigamp]

davidbeach, you are spot on.

jghirst, If you are able to get a hold of the text I referenced (Alvin H Knable "Electrical Power System Engineering", McGraw-Hill 1967) he explains it quite well, the tank of the transformer acts as a tertiary winding of sorts. Sorry but I do not have my copy at work so can not quote word for word (can do tomorrow however). From memory, Knable says that the zero sequence impedance is in the range between 50% to 200% for core type transformers and 3000% or more (i.e. effectively infinity) for shell type. The J&P transformer book is in agreement with Knable.

There are many texts that say the wye/grounded wye connection can not produce ground fault current, including the GEC Protective Relays Application Guide. I believe they have got it wrong.

 

[jghrist]

bigamp,
I doubt if I can get the referenced text and none of mine cover this subject. I'll buy what you are saying, however. It makes sense in a core type tranformer. When I reviewed the Westinghouse T&D Table 7 Fig A-3 (core type) closer, I see that the Z_GP term that is infinite would be paralleled by another impedance for a ø-grd fault. The value of this impedance is determined by a zero-sequence impedance test.

Is there any advantage to using an ungrd wye - grd wye transformer connection? I guess it would provide a zero phase shift transformation without the third harmonic problems of a grd wye-grd wye connection. I'll stick with grd wye-grd wye with a delta tertiary.

 

[bigamp]

jghirst,

It is possible to get the text second hand through Abe books, they currently have two or three available. It is surprisingly expensive though, in the $US mid 40's. Not a bad book.

I have come across wye-wye transformers where these have been used in preference to delta-wye transformers purely for cost reasons (no other advantages). It appears that a wye winding may cost a bit less than a delta one.

I'm with you though, it pays to stick with the conventional.

 

[stevenal]

In defense of the texts which simply show an open circuit: None of the ones I've seen mention the form. The open circuit is valid for shell form and banked single phase units. The're not wrong, but they are incomplete.

 

[HiSet]

Bigamp,
Please clarify following points.
For core type trf: Is Zo value is 50-200% of Z1 ?
For shell type trf:Is Zo value is 3000% of Z1?

 

[bigamp]

HiSet,

The impedance values are quoted from Knable and are not a percentage of Z1, but actual percentage impedance on the transformer base, where 100% impedance = 1 per unit (pu).

For core type, Zo = 50%-200% (i.e. 0.5pu - 2pu) and for shell type Zo = 3000% (30pu).

 

[SidiropoulosM]

Positive and negative sequence impedances are independent of whether the sequence currents are injected into the primary or secondary winding. The zero sequence impedance, however, will have very different values depending on what terminal we view.

In this particular case, for zero sequence currents the ungrounded primary will be seen from the grounded secondary as an open circuit magnetizing impedance. The zero sequence impedance, as seen from the grounded secondary, will have a finite but very high value, equal to the magnetizing impedance.

The zero sequence impedance as seen from the primary is infinite.

 

[jbartos]

Suggestion: Knable Reference, mentioned above, presents relationships, e.g. IFA=3I1=3x1/1.20=2.5, which are not documented by references or proved or derived. This is sometimes referred to as "not traceable." Some more rigorous projects would not accept these relationships or concepts for a reference. Generally, books do have mistakes and erratas.