Conflicting desires in a transformer...

[davidbeach]

Historically we've stepped down from transmission/sub-transmission to distribution voltages with delta-wye transformers with the low-side wye solidly grounded. Worked great for decades. Still works great for much of the system.

But, we're seeing increasing amounts of distribution connected generations, mostly PV, on some of the more rural portions of the system. Enough in some cases that once it's all built the distribution system will become a source feeding the transmission. Now those delta windings begin to become a liability in that they don't provide an effectively grounded source to the transmission. We're adding 59N protection to trip the transformers if they are feeding a ground fault on the transmission system.

An effectively grounded source would be highly desirable. The most straightforward way of getting there, while also maintaining an effectively grounded source for the distribution would be to replace the delta-wye transformer with a wye-delta-wye transformer. Many advantages in doing so.

One huge disadvantage though, feeders need to be capable of being tied to adjacent feeders and the 30 degree phase difference between a feeder sourced by a wye-delta-wye and a feeder sourced by a delta-wye makes it difficult to tie the feeders together.

What I want seems to be a purple flying cow, aka a wye-delta-wye where the low side lags the high side by 30 degrees; they just don't seem to be stock items. ;-)

Can a wye-delta-wye ish transformer be wound with a bit of zig and/or zag and provide a 30 degree shift while still having decent ground performance? Is that something that scales up to 16.8MVA (self cooled rating) with primary voltages up to 115kV? I know various manufacturers make harmonic cancelling transformers with all sorts of weird phase shifts zigging part of one winding and zagging part of a different winding, but at sizes two orders of magnitude smaller and voltages over 2 orders of magnitude lower.

Are there distribution level (13kV or so) phase shifting devices that could sit at the boundary between the two phase angles and allow tying the two sides together?

Has anybody else had to deal with this type of situation? How did you solve it?

 

[waross]

The problems that I encountered with grounded wye:delta transformer banks were circulating currents in the event of an unbalance or fault on the wye side and the transfer of energy from phase to phase caused by the circulating current.
One aspect of this is fault currents that are higher than shown by a calculation that ignores the back feed.
A serious safety issue is that when one phase is opened both sides of the open are still at line potential.
I still see an issue as long as the delta is present.
You may know better, but this should be considered.

A possible solution based on harmonic mitigation techniques.
A suggestion that may not work, but:
Consider an application where one of the working voltages is 100 kV.
In place of the 100 kV winding wind place two 86.5 kV windings on each leg.
If my sketch is correct, by connecting the windings to adjacent phases you will get 100 kV at a 30 degree shift.
If my numbers are off, correct my numbers and consider the concept.
The windings will be 73% oversized compared to a straight winding. The winding on the other voltage will be standard so the size increase will be less than 73%.
It's up to you if you want to include the delta, given the possible issues.
Note: with the higher impedances typical of transmission transformers compared to distribution transformers, the delta issues are not nearly as severe as they are at a distribution level.

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

 

[Mbrooke]

How I would solve it:


115kv Wye grounded primary, 13.8kv delta secondary. A zig-zag or grounded wye delta transformer attached to the 13.8kv bus to provide a place for fault current and neutral current on the distribution system. In many cases the grounding transformer does not even have to be custom made- standard pole pigs of sufficient size can do it. This scheme common in some countries and works well- and I figure if it works in almost every generation station built around the globe, why not distribution?

 

[dpc]

I think Mbrooke's idea would be the most economical, if adequate space exists. This would make your main transformer simpler and less expensive. The grounding xfmr would have to be sized to handle the neutral imbalance. Phase shift across feeders and subs should work out, since it is still a 30 degree shift. There would be some costs related to switching and protection of the grounding transformer, but probably not a deal killer.

(First step for you should be to upgrade all that 57.5 kV transmission line to 115 kV so you don't have to buy dual voltage primary. )

As usual, I'm probably overlooking something....