3-winding Trafo Zero-Sequence Data

[drchaos]

Hi,

I'm trying to determine zero-sequence data for a 3winding trafo 220kV/110kV/10kV Ynynd5 250MVA ABB transformer to put into PSSE.
In the transformer test sheets, under zero-sequence tests, there are 4 sets of numbers (Z, R, X):
HV with LV open -
HV with LV shorted
LV with HV shorted
LV with HV open
What is confusing us is that there is no mention of the tertiary - is it open or shorted in these tests? Can I use the much-cited formulae Zp = 1/2(Zps + Zpt - Zst) etc? And if so, which set of data should I leave out - are two of the above four sets of data essentially the same?
Thanks in advance for any help.

 

[drchaos]

Actually I'd like to clarify the above questions:
When you solve for the zero-sequence equivalent impedances Zp, Zs, Zt, we have four measurements, and so get four equations in 3 unknowns, and so get non-unique solutions.
Does anyone know why this is, or how to resolve this paradox? Which solution is the correct one?

 

[opmgr1]

I would make the following assumption:
HV with LV open; I would assume that the teritary winding is shorted and so I would label this Zpt.
HV with LV shorted; I would assume the teritary winding is open and so I would labe this Zps.
LV with HV open, I would assume that the teritary winding is shorted and so this would be Zst.
LV with HV shorted not sure what to make of this, my best guess is that would be Zsp. Not sure if there is a relationship with Zps.
You can combine the equations Zp=1/2(Zps+Zpt-Zst) etc. with the set of equations Zps=Zp+Zs etc.

 

[drchaos]

Thanks opmgr1 - what you say is what we started out thinking, and that's what's in all the textbooks. However, in the test data, there is no mention of the tertiary, and I found a presentation on 3W Trafo testing from Siemens on the net, where the tertiary is essentially left untouched for these tests. The four equations become:
Z1 = Zp + Zt
Z2 = Zp + Zt||Zs
Z3 = Zs + Zt||Zp
Z4 = Zs + Zt, where Z1-Z4 are the four measurements I mentioned at the start. There is more than one solution to these equations though the one we'll probably use is:
Zt = sqrt(Z2(Z1-Z4))
Zp = Z1 -Zt
Zs = Z4 -Zt....has anyone else come across this?

 

[prc]

Is it an auto transforme? From the vector symbols it looks like a three winding transformer.When you say LV trust it is 110 kV winding.