[the spyder]

Im just the computer geek, so bare with me...
We had a horrible powersurge at work the other day. The machine is a matsura h plus 630. Now do not ask me WHY I did this, but we like to fix what ever we can around here. Our local service techs like to screw us in the 4 figure range every time.

The whole building lost power, but when it came back on, the machine loaded then POP. The main 100amp breaker in the wall panel tripped. With some close inspection I found codes were being thrown on the Fanuc controller power supplies. There are 2. The first one was giving 7, a over voltage alarm, the second was giving 0, a over voltage alarm detected in the system alarm. Anyways. Some how I decided to tear apart the first unit, finding scorched components. The units are a three phase 230v to DC 325V converter. After removing the power supply, I found arcing carbon marks on the rear of the circuit board, between the massive transistor modules. These are directly tied into the 3 phase. The only damage on the circuit boards is around the smaller control interface for the transistors. Example, around the first transistor, the small control circuit has 3 resisters, 2 caps, and 1 diode blown.

What I really do not understand is there are 4 of these modules in a row. The 3phase inputs at the bottom of the unit. The first transistor is blown. The second is untouched. The third is again fried. The fourth is good. They are all chained together, one after the other on a buss bar. The controllers again are the same as above, 1st fried, 2nd fine, third fried. How is this possible? Now the machine was sitting idle when this happened. No load. Is it possible for a surge to "jump" between the legs of the 3 phase? There are inline power filters, rectifiers, but no real surge suppressor.

Has anyone else had this happen? I tested the transistors today and one is completly fried, the other 3 look good. I get to put this whole thing back together tomorrow morning. YAY.

[the spyder]

Well, surprise enough everything worked out smoothly. We ordered a new board and 2 transistors, I installed them and reassembled the machine. Stood there for about 20 minutes going.... ummm do I really want to do this, with my hand on the power switch. Its been running since.

[NC Cams]

Possible explanation: Transistors are voltage rated for a maximum voltage potential difference between the collector and emitter. Exceed that and you let the smoke out of the transistor and it quits working. A power surge does that sort of thing. Makes you wish you had a surge suppressor, no????

BTW: For some of their DC drives, Fanuc "sequentially fires" their transistors and/or SCR's at the proper time. Thus, they can convert 3 phase AC into rectified DC. Since it is 3 phase rectified, it is smoother than full wave rectified single phase.

Such DC drives are unique which makes their stuff so pricey to repair - they use "ASIC's" (application specifice IC's - custom modules) that they make EXCLUSIVELY. Like something like $125 for $5 worth of potted parts on/in a APC (whatever) package size. Good work if you can get/hold it.

Consider yourself lucky that the replacement of obviously fried/molten parts fixed the system. Others have NOT been so lucky....

[Al_The_Man]

Stood there for about 20 minutes going.... ummm do I really want to do this, with my hand on the power switch.

He He... know how you feel, with some IGBT's at $250 a pop, you tend to get nervous and pray alot.
Al.