LCI problem

[oftenlost]

Anyone,
34000 HP synchronous motor, two identical windings delivering half total horsepower driven by LCI drives at 2800 VAC in marine environment. One winding flashed phase to phase at the buss rings. Windings and stator cabling determined to be contaminated with salt water. Windings given superfician cleaning and the motor restarted. Flashed again at 1500 VAC. Affected half disconnected and the lead terminations insulated by shrink tubing and the motor restarted. Operated normally at low speeds and at approximately 2100 VAC, flashed over again. Buss connector rings approximately 80 MM apart and disconnected windings flashed phase to phase. The drive is fed from an ungrounded delta secondary and the Y point is grounded through 5 parallel 1600 ohm resistors. It was discovered that 4 of the 5 had been broken so that the Y ground was 1600 ohms instead of the parallel resistance of 5 of them.
Question is, could the disconnected windings have developed the 20K volts or so required to flash across the distance between the buss rings or is there something else at play here. I would think that transformer action would generate 2100 volts plus or minus and the Y resistor was designed to limit ground currents and even an increase of 5X would not generate the requisite voltage. What am I missing here? Megger, PI and surge comparison tests indicate a good winding and the low megger cables have been replaced and the motor appears to be operating normally after a two day cruise but I am concerned about transients that the OEM engineer indicated could be the problem.

 

[oftenlost]

Make that superficial cleaning

 

[davidbeach]

One field or two? If two, was the field of the disconnected stator also disconnected? If the disconnected winding had full field, its terminal voltage would be the internal voltage and could easily be significantly higher than the rated terminal voltage.

 

[oftenlost]

One field about three quarters excited based on the screw speed recorded. The stator windings are wound 30 degrees offset on a common core. The stator windings were open. I would have thought that the motor, if operated as a generator would generate something like the voltage required as a motor and if it were transformer coupling, a 1::1 ratio. The megger reading between the windings were approximately 1K ohms so there was that factor too.

 

[TomG33]

Hey Oftenlost,
Maybe you have fixed this by now, Just saw the post today.
But anyway i have a few questions/comments to put up.

I guess there can be two reasons for a line to line flash in the motor,
1 = too many volts
2 = Too little resistance.

Given the description you supplied I do not really think you have a situation where you have too many volts.
This could only be possible if for some reason the field is over excited. But seeing it is the same field that works for both halves of the LCI, If something was wrong with the field or it's excitation then it would be common to both sets of motor windings and not just the one.

Were all three flashovers between the same two terminals? I am guessing that you are talking about the motor terminals.

Given that the flashing was Line to Line, I doubt that the grounding resistor connection problem was a contributing factor. If it was, the flash would have been line to ground.

You said Superficial CLeaning, My guess is that your problems are that this type of cleaning was not enough to get the insulation rating back up where it needs to be. Originally the failure was due to the salt water contamination, but after the first flash, the whole area would have been contaminated with the metal splatterings as well as a copious amount of carbon soot. unless it is all removed you will continue to get flashovers.

Tom