PARALLEL TRANSFORMERS - LOAD/IMPEDANCE CALCULATIONS 1

[genhead]

I want to parallel two transformers with the same voltage ratios, kva rating, and vector group.
Can anyone tell me how to calculate the load carried by parallel transformers with difference impedances. In this case 5.97 and 6.1 %
What aout transformers with different kva ratings ?

 

[bigamp]

From the J&P Transformer book, for two transformers, A and B having the same or different kVA and the same or different impedance voltages AND THE SAME X/R ratio:

Current in A (Ia), in Amps:

Ia = (Il x Na)/(Na + Nb)

Current in B (Ib), in Amps:

Ib = (Il x Nb)/(Na + Nb

Na = kVAa/Vza

Nb = kVAb/Vzb

Il = total load current (in Amps)
kVAa = Transformer A kVA rating (in kVA)
Vza = Transformer A impedance voltage at full load (%)
kVAb = Transformer B kVA rating (in kVA)
Vzb = Transformer B impedance voltage at full load (%)

There is a more complicated formula (with drawings) for transformers with different X/R ratios but it is a bit of a pig to write!

Regards

 

[Alex68]

I suggest to study the equivalent 1ph circuit of your system in which you consider the absolut (ohm. nor p.u. neither %) impedance values. After you solve this circuit, you will know the power flowing in each transformer.

 

[tgott]

Since both transformers have the same rating, to figure out the distribution between the two, simply divide the impedance of the other by their sum (.0597 + .061 = .1207). The transformer with 5.97% impedance will carry .061/.1207 = 50.538% of the load; the 6.1% transformer will carry .0597/.1207 = 49.47% of the load; ie. the transformer with the lower impedance will carry more of the load.

For transformers of different kVA, you'll need to convert them to the same kVA

Znew p.u. = (Zold p.u.) x (kVA newbase/ kVA oldbase)

Keep in mind that in paralleling transformers with different impedances that you will need to be aware of circulating currents.

 

[cuky2000]

Tagott has a good point since parallel transformers could be closely modelled as a current division circuit with two reactance in parallels.

The above method is simple and provide sufficient accuracy, particularly if X/R are the same with small deviation otherwise.

See the enclose site for additional info.

http://www.usbr.gov/power/data/fist/fist1~8/1~8_1.htm