Transformer Vector Group effect on substation GPR 1

[acog]

Hi All,

Recently I have had a discussion with a fellow engineer in relation to the choice of winding configuration for a 132/33kV transformer in a wind farm application.

His sentiments were that a Yyn(d)0 (buried delta) should be used. I was proposing a Dyn1 transformer as I assume it will be cheaper even with the increased insulation on the HV side as it only has 2 windings as opposed to the 3 required for Yyn(d)0.

The other engineer said that there will be a problem with the substation EPR or earth(ground) potential rise during a 132kV fault if we employ a Dyn1 configuration. I can't see how placing a delta on the high voltage side downstream from a fault would increase the EPR at the substation.

Can anyone please comment on how the 132kV EPR would change due to winding configuration of a 132/33kV transformer (downstream of the fault location)?


Note: all wind turbines are connected with a Dyn11 33/0.69kV transformer.

Thanks

 

[prc]

A YNyn(d0) will be cheaper than a Dyn1 connected transformer. For this application, normally YNyn connection is used.

 

[Marmite]

The GPR is a function of the single phase fault level on the 132kV side and the earth grid resistance. It would be unusual for the GPR from a fault on the LV side to exceed that on the HV side unless you have lots of transformers on site, so the 132kV fault will be the worst case. I can't see how the vector group of the Tx downstream would have any bearing at all on the GPR with the system you have described. I agree with PRC that normally a YNyn connection would be used. The insulation on the 132kV winding can be graded towards the star point which saves money, assuming a solidly earthed 132kV winding.
Regards
Marmite

 

[7anoter4]

Agreeing with prc and Marmite, of course, I will try a simply explanation.
As I understand the wind farm it is a power plant that means the 132/33 KV transformer is the "source".
If the 132 KV neutral [of windings] is solidly grounded in a phase-to-ground short-circuit case
the GPR [or EPR] value will be the short-circuit current multiplied by Rground:
GPR =Isc*Rground.
If Rground is well designed, could be low and GPR close to zero.
The potential differences between phases and Ground will be close to 132/sqrt(3)=76.2 KV.
If the 132 KV windings are Delta connected in a phase-to-ground, short-circuit case the Ground potential will raise to this phase potential and the difference between Ground
and other 2 phases will be 132 KV.

 

[mgtrp]

My suspicion would be that your colleague is looking at how a YNynd transformer would affect the return path of your earth fault current. For GPR, the worst case would be a phase to earth fault inside the substation, whereby all of the fault current flows through the earth grid into the general mass of earth and back to the source.

With the installation of the YNynd transformer, a path for the fault current is provided within the substation so some (likely the majority) of the current will return via the neutral of that transformer, never leaving the substation and so never contributing to GPR. Try modelling in fault analysis software an infinite bus connected via a long transmission line to a DYn, and then a YNynd transformer, with a fault on the HV terminals of the transformer, you should see the difference in ground fault current flowing from the infinite source. Alternately, draw out the sequence diagram of the same.

For what it's worth, my experience is that the fault current is generally higher on the LV side of the transformer, especially if the source is not too strong, so the HV connection is not the determining factor.