Parallel transformers should have the same terminal voltage.
This generally means that they should be on the same tap.
In that case there should be no circulating current.
Differing impedance voltages will affect the load sharing ratio but should not cause circulating currents.
That said, delta transformers and delta banks may experience circulating currents even when not connected in parallel.
But;
The circulating current should be seen by both CTs in each differential pair and should not cause a problem.
However, circulating currents do add to transformer heating and for proper protection you may consider monitoring the current in each winding as well as the line currents.
I have seen delta connected transformers badly overheated by circulating currents even though the line and load currents were quite low.
In parallel single phase transformers, a voltage unbalance of twice the impedance voltage will cause a circulating current equal to full load rated current.
What is your application?
Bill
--------------------
"Why not the best?"
Jimmy Carter
These are three delta wye grounded 20/25/35MVA 3 phase units being run in parallel. My understanding (if correct) being transformers of different makes or ages will never have truly identical primary to secondary ratios which leads to slight secondary voltage difference. The plan is to keep all units the same impedance when anyone is replaced to prevent overloading from asymmetrical load division across them.
While it is currently planned to keep the tap changers in the neutral position full time, if ever activated all the units must be lead in sequence with a leader follower scheme?
Lagging on one, leading on the other; reversed at the other winding.
Master-follower is one way to coordinate load tap-changing. Other methods use measured circulating current or circulating reactive power to bias the controls in a way that keeps them together. See beckwithelectric.com for more information.
There is also the case of equal percent impedances but unequal X/R ratios. Current sharing will be good but the KVA with rated current through both transformers will be less than the sum of the individual KVAs.
Bill
--------------------
"Why not the best?"
Jimmy Carter
When the X/R ratios are different the phase angles of the current will be slightly different through each transformer. Even though each transformer may be loaded to its rated KVA, the current vectors add, well vectorily, not arithmetically. The result is sum of the transformer KVAs will be slightly more than the load KVAs.
This may not arise in practice very often. The slight difference is more for the purists and those who wish to have a complete understanding of the subject.
In practice, the actual load so seldom matches exactly the transformer rating on a day when the ambient temperature is up to design maximum and there is no wind, that the slight reduction in capacity is not an issue.
If you have occasion to parallel two transformers with greatly different X/R ratios, then you may want to calculate the phase angles.
It may be easier and quite safe to be not aware of the issue.
Bill
--------------------
"Why not the best?"
Jimmy Carter
It will probably not even be an issue if truly slight. The application calls for 3 units in parallel, however the normal load will only be 2/3 of the total banked rating 4-5 hours a day during peak load. The 1/3 remaining capacity is when (or if) one unit is taken out of service the remaining two will be loaded to their full OFAF rating.
My biggest concern however are replacement units. I plan on specking a common impedance and X/R ratio (one the has been around for a long time and will continue to be for the given MVA). Does 12% Z sound typical?
Better to go by the impedance of existing transformers. Don't look at X/R.As already discussed, its impact is negligible. But if you specify it, based on old transformers, you will end up with very poor efficiency transformers. R % (depends on load losses only) is coming down all these years, as transformer losses are coming down.
For a 35 MVA transformer,you will find %Z specified from 8-20 % depending on the region and utility. Common range is 10-15 % on max rating.
C57.1210- 2010 for transformers recommend 7% (with OLTC) 7.5% with out tap changer for 200 BIL Transformer. It is 9&9.5 % at 550 kV BIL. It is on self cooled MVA base ie at 20MVA base.
IEC/TR 60909-2-2008 (data of equipment for short circuit calculation) gives the range of %Z and X/r ratio at different parts of the world. These are on maximum rating base ie at 50 MVA.
