Sometimes it's best to just leave well-enough alone. Editorial Director John DeDad takes a look at the effects — both good and bad — of swapping out an oversized transformer for an undersized one, and how to compensate for the change.

Q. Our service consists of a 1,500kVA, 12.47kV-480V utility-owned transformer with 6% impedance (Fig. 1). We have two large, 6-pulse adjustable speed drives. One is connected to a 660-hp dynamometer, and the other powers a 440-hp fan. Aside from these pieces of electrical equipment, we have virtually no other loads, except for a small amount of lighting.

We conducted a power system harmonic analysis and found that, as we suspected, the drives were producing 5th, 7th, 11th, and 13th harmonic currents. But we also found that the total harmonic distortion (THD) on the voltage was well within the guidelines specified by IEEE 519. In fact, the harmonic voltage distortion at the utility transformer secondary was less than 5% of the total of all harmonics, and less than 3% on any single harmonic.

Through several billing periods, our utility found that the total demand for our metered facility was only 315kW. According to our utility, the dynamometer, being a regenerative device, was not drawing power most of the time but instead was feeding power back into the system.

Feeling that its billing was accurate, our utility believed its transformer was oversized and decided to replace the 1,500kVA unit with a 500kVA unit. After all, with a maximum 315kW demand, the 500kVA unit was thought to be more suitable.

Shortly after replacing the original transformer, we began to see significant problems, mostly resulting from some interaction between our facility's power system and the new transformer. Basically, we found that the THD at the point of common coupling (PCC) was now in excess of 12% on the 5th harmonic alone, and over 9% for our entire power system. Any idea what happened?

DeDad's answer: Without access to the actual measurements of your harmonic analysis at your site, I can only guess what took place with the transformer change.

A major characteristic of harmonic currents and voltage distortion is that the voltage distortion increases, even when the same currents are flowing, as the source impedance increases. And this increased voltage distortion will be in direct proportion to any multiple of source impedance increase. In other words, the amount of voltage distortion that will occur varies with the magnitude of the source impedance, when distorted currents are flowing through this source.

In your situation, it could be that as the utility reduced the transformer size, it may have inadvertently increased the apparent impedance in the same proportion. In other words, the utility, in effect, reduced the transformer capacity by a factor of three (1,500kVA down to 500kVA) and possibly introduced a new apparent impedance of 18%.

When that happened, those harmonic currents now flow into what appears to be a very high-impedance source (the 500kVA transformer), which causes the voltage distortion to dramatically increase.

You have two options to eliminate the problem. The first is to have your utility replace the 500kVA (“high-impedance”) transformer with a much lower one, possibly the original 1,500kVA unit. This will change the THD back to the original, acceptable levels.

The second option is keep the 500kVA transformer and to install a harmonic trap on its secondary (Fig. 2 above). This should remove the harmonic currents from the line. So without distorted currents, you should have only minimal voltage distortion.

Got a PQ problem? Send your questions to