Impedance matching of Trans. Banks-NOT paralleling

[wfowfo]

There is an on-going debate at our Coop about matching the impedances of single phase transformers in wye/delta (open and closed types) and wye/wye banks. Please keep in mind I am asking this in regards to banking transformers, not paralleling them.

An old REA document (Bulletin 161-22, page 13) states;

"Transformers used together in banks should have identical voltage ratings.... etc. Except in the seldom-used delta-delta connection, it is not necessary that the transformers have the same impedance."

We interpreted this to mean that it did not matter in the configurations we used. Yet in every Lineman's school we send our guys to, they are taught that it does matter. To complicate things further, they have been told to match the pots using figures anywhere from 2% to 10%.

Is there a definitive rule (ANSI, IEEE, etc.) anywhere that states this in clear and definable manner that I can use as company policy? If not, can anyone tell me what their own policy is?

 

[waross]

I can see several issues here.
Fault current: You will have different available fault currents on different phases.
Voltage drop; The voltage drop at full load is essentially equal to the percent impedance voltage. The difference between impedances will be the difference between voltage drops at full load.
If you use a 2.7% transformer with a 3% transformer the difference may be unnoticable.
If you use a 2% transformer with a 10% transformer, the voltage to neutral from the 10% transformer will be 8% low at full load.
This may be a problem for motors and three phase rectifier banks.
If you match within 10%, I understand that to mean that you can pair a 2% transformer with a 2.2% transformer. The voltage difference at full load would be about 0.2%. That sounds reasonable and acceptable.
yours

 

[stevenal]

Gotta disagree with the delta-delta statement. Unequal impedance will cause circulating current increasing system losses and decreasing available bank capacity. Is there an updated RUS version of this REA Bulletin? What's the title of this bulletin?

 

[ePOWEReng]

Most of the utilities I am familar with have the ability to mismatch transformers that they are banking up to a certain amount in their standards. I think I recall a limit of the large transformer being 1.667 times the smaller two. Probably assuming there is some lighting and misc. load on the larger phase and the smaller two phases just have a share of the three phase load. Waross is right that voltage imbalance can be an issue if you don't watch yourself.

 

[waross]

Hello stevenal;
I agree with you. However, I think that the statement is confusing.
I think the statement;
"Transformers used together in banks should have identical voltage ratings.... etc. Except in the seldom-used delta-delta connection, it is not necessary that the transformers have the same impedance."

Would be more understandable if it was rephrased as;
"Transformers used together in banks should have identical voltage ratings.... etc. It is not necessary that the transformers have the same impedance, except in the seldom-used delta-delta connection."
I would also be concerned with unbalanced and circulating currents in mismatched wye-delta banks. If you have different impedance voltages on any delta secondary you will have different voltage drops under load. With different voltage drops, your delta vector diagram will not close on the origin but will leave a small gap. Forcing this gap closed by bolting up a delta connection will affect the division of current between the transformers and may cause circulating currents, depending on the impedance/resistance ratios of the individual transformers.
respectfully

 

[stevenal]

Eng-tips converted the word "connection" to some completely non-relevant DSL link. With the underlining, my eyes missed the comma. With the comma it matches IEEE C57.105 also warning about circulating current in D-D transformers.
Wye-delta connections should be floated on the primary wye side. With no return path available for I0 current, no circulating current can occur even with unbalanced voltages and impedances. With the D-D connection no return path is needed. Current just circulates in both windings.

 

[wfowfo]

stevenal
On our wye/delta banks, we do float the neutral on the primary, but I take it that the voltage imbalance can still be a problem?

As for what waross said about a 10% difference (as in a 2.0% pot banking with a 2.2% pot) probably being acceptable, does everyone agree with this rule of thumb? Is there an actual standard?

By the way, thanks for all the responses. Looks like I'll be eating a little crow at the next meeting.

 

[ePOWEReng]

stevenal -- I agree with your accessment of having an ungrounded wye on the primary side of a wye-delta, but you also have to be careful of overvoltage do to transformer resonance if you are single phases switching on a ungrounded wye-delta bank. The wye should be temporarily GROUNDED during single phases switching! Just temporarily jumper the neutral to the wye point, until you close all three of the cutouts. This may not be a big deal unless you put lighting arresters on your Tx, if you do have lightning arresters and you don't temporarily ground then the LA might blow up in your face! I have seen this happen and the lineman got pretty pissed off when the same thing happened twice in a row!

 

[stevenal]

hold6448

Yes, I believe this is how we operate.

wfowfo

The floating neutral solves the loss and possible bank overload from circulating current. Imbalance can still be a problem for motor loads where the resulting negative sequence current produces a torque in the wrong direction tending to overheat the motor. I agree that a 0.2% voltage imbalance is acceptable.

ANSI C84 indicates the maximum no-load voltage unbalance at the meter should be 3%. Unfortunately they don't give a recommendation for the unbalance under load. 3% requires a motor to be derated to 90%. At 1.5% unbalance the motor derating factor improves to 97% and no derating is required below 1% unbalance. Of course this includes all sources of unbalance to the motor terminals.

I'd still like to know the details of the bulletin you referenced.

 

[waross]

Hi stevenal;
I happen to be in a part of the world where the prevailing wisdom is that the wye-delta connection is the prefered connection. They connect the primaries to neutral/ground. A lot of low voltage sensitive appliances like refrigerators and freezers burn out. A lot of distribution transformers burn out. A lot of delta transformer banks have one fuse-cutout hanging down or missing.

Hi wfowfo;
Does your co-op have voltage balance standards? What percent of unballanced voltage is considered acceptable on your system? I would suggest using that figure as a basis for discussion on what is an acceptable figure for impedance mismatch.
If your target for voltage balance is 5% consider what part of that tolerance you are willing to give up to allow mismatching.
In this context, a 10% mismatch based on a 3% impedance transformer will be 0.3%.
You can have a 0.3% unbalance due to mismatching and a 4.7% primary voltage mismatch and still be within your 5% target for overall voltage unbalance.
respectfully

 

[waross]

Hi stevenal
We were posting at the same time and I missed your post. Please revise my 5% figure downward.
respectfully