Southern Fried Transformer

Our 3,000-HP diesel generator is there for emergencies. Once a week we start and run it at full power for about an hour, and the few times we've actually used it for temporary power, it worked just fine. In a real emergency you'd only need full power for a little while, and we never had one of those.

Well, a few years back I decided to make a 24-hour test run at full power. About seven hours into the test, the generator tripped on loss of field. The little transformer that taps 3 phase power for the field had burned up. When I got there it would still "fry spit" -- to use our technical terminology. Melted insulation was dripping from it onto the floor. We found nothing else wrong.

I figured something else should be hot, but the generator field was only lukewarm to the touch, as were the magnetic amplifiers between it and the crispy transformer, so where did that high current go? Where did it come from?

Everyone was puzzled -- especially me. It clearly wasn't poor ventilation for the transformer was far hotter than anything else around except maybe the engine exhaust manifolds 20 ft away.

This is a gleaming gray locomotive-sized diesel engine nearly 50-ft-long, 20 cylinders, and absolutely beautiful. The electric generator on one end is ~2500 kw, the size of a small sedan. The transformer that burned up was a few KVA, the size of a big toaster oven, and located in a cabinet nearby. Its connection to the generator was the usual, wye generator to wye primary, delta secondary.

So the transformer was replaced with a somewhat better rated one having lower internal impedance. Ventilation holes were added to the cabinets and a second test run was begun. This time the transformer burned up in about two-and-a-half hours.

Now that was a head scratcher.

Well, a question well stated is half answered. It looked like the question to be answered might be: "How can current go into a perfectly good transformer and never come back out, and in fact overheat and burn up the transformer?"

There are a of couple possibilities:

Naturally, opinion divided into two camps, the overvoltage guys and the distortion guys. So we instrumented every winding of every transformer involved. The primary side neutral wire normally carries little or no current so we very nearly ignored it. It is fortuitous that the peskiest of the harmonics that give three phase trouble, the third, is easily measured in the primary side neutral wire. So we hung a current transformer and oscilloscope there, too.