You say that the voltage drop doesn't start until A1J2 is plugged in. Looking at the manual, A1J2 shouldn't carry 5VDC to the displays, just segment signals unless an IC is shorted out. Normally, with that connector unplugged, you shouldn't see anything in any of the displays. Plugging the A1J2 connector back causes the 5VDC short conflict.
I see this was suggested at the beginning of this thread but I didn't see where it was addressed. I know this has one of those "new" CPU boards in it but have you tried running it in a different game (since you've owned it) to see if the problem is in the CPU board itself? I understand that Fred always tests and videos his boards to prove they work, that's smart business and verifies that the problem is elsewhere.
However, it may be that a unknown existing display issue shorted out something on the new board when you plugged it in and turned it on the first time. This isn't that uncommon as anyone that fixes boards can tell you, even Fred knows this. I always tell anyone I fix boards for that once the board is plugged in and the game turned on, the board can short out again if the cause of the problem isn't corrected too. It may be the same issue that took out the original CPU board and the reason it was replaced by you.
My suggestion would be to send the board back to Fred for testing and have him repair it for you if needed. If it turns out that there is a problem, then, he'll be able to tell you where to look in your game to correct it. Then, don't plug the board back in until you do. If it turns out that the CPU board is good, all you've lost at this point is shipping costs and you're back to looking around for the issue again.
System 80, not just a job, it's an adventure