Half voltage usually indicates a faulty bridge rectifier with one of it's internal diodes going open circuit. It's bridge rectifier BR2 in this case so you're on the right track.

With the machine off, disconnect the J3 connector from the rectifier board and do a diode check on that BR2 which basically has four internal diodes.
If the board is in very toasty condition you might need to resolder the bridge.

What's a diode test with your multi-meter saying about the replacement bridge?
What are you measuring at TP3 when the J3 connector (backbox harness) is disconnected from the rectifier board?

Nice pic, you have another dud bridge rectifier. Two open circuit internal diodes, both on the same cycle.

We're talking about the rectifier board, not the solenoid driver board.

When J3 (backbox harness) is NOT connected to the rectifier board, TP3 on the rectifier board should measure around 11.9 volts.
When J3 is connected to the rectifier board, TP3 on the rectifier board will now measure between 14.5 volts and 16.5 volts thanks to the large capacitor at C23 on the solenoid driver board.

Are you actually measuring any short circuits across any BR2 pins?

Does rectifier board test point TP5 measure a dead short to ground? If yes the problem might be downstream from the bridge. Check the coin lockout coil on the coin door and make sure the green wire isn't also touching the coin door frame.

Yeah sorry, BR3 is the solenoid bridge that's fused by F4.

What bridges are you using?

Are you by any chance overheating them for too long when soldering them?

Not sure if it helps, but I always solder the bridge after screwing the bridge to the rectifier board/plate, and also wait at least 5 minutes between soldering each leg.

Does it happen if you disconnect J4 from the SDB (Solenoid Driver Board)?

With these two connectors removed:

* What voltage are you measuring at TP3 on the SDB?
* what voltage are you measuring at pin 24 of the U2 chip on the SDB
* Is the flipper enable relay clicking on at powerup? i.e. do the flippers always flip when you press the flipper buttons regardless of being in game over mode?

Sorry, I just re-read your posts and you mentioned it's all working properly again. So those three diagnostic questions I asked are only relevant when the game is locking on the coils at powerup.

The only way all coils can lock on at powerup is a failure of some sort with the 74L154 chip at U2 on the SDB. I would carefully inspect the solder joints of pins 12 and 24 on that chip which is ground and 5V power respectively. Follow the connection between pin 24 to TP3 since it goes through some jumper links and carefully inspect their soldering too.

When do those transistors fail? At first poweron after replacement, or first activation during attract/game mode?

Yes, the green wires go to the coil lug where the banded side of the diode is.

If the transistor driving that coil keeps blowing, the coil diode might be shorted. Before you solder that white wire on the other coil lug, measure the resistance across the coil - should measure around 53 ohms. If you measure a dead short circuit, replace the diode on that coil. Also check that the diode is properly soldered because if one leg isn't, the transistor will always blow on first activation of that coil.

You might just want to replace the diode on that coil anyway.

Hmm, if you still have the diodes that were removed from the coil, measure if those diodes are shorted. The problem might actually be an intermittent short in the coil winding..