U20 on a WPC is a 2803 column drive for the matrix. If that was bad; you would have had other switches on the column acting up too.
Quoted from Castro28:
I ended up blowing a transistor (u20) on the CPU board. Luckily Williams had it socketed in so I just pulled it out and put in the new chip, fixed.
It's not a transistor, it's actually 8 darlingtons. I also have *never* seen one with a socketed U20; someone has changed that along the way. U20 was always getting popped.
Quoted from jazzmaster:
I am fairly certain the switches are engaged. I thought this might be a ground issue but I would expect that the high power would indeed be turned off, if that were the case, but all of the solenoids fire as usual.
It wouldn't be ground. One of the switches is connected to the switch matrix; used to detect if the door is closed. The other just has one side of the solenoid secondary going through it. If the switch that detects the door being closed failed; machine thinks it's open but the solenoids continue to work; the CPU/driver doesn't control the voltage; the interlock physically disconnects the AC before it gets to the driver board.
At least that's how I remember all the WPC machines I had my head in worked.