On sys7 driver boards the col drive section had resistors that were present in earlier versions replaced with wire links. Looks to me like you've got an older version of the driver broad in the game and the one you're swapping in from BK has the links so it works fine.

On early (sys3-6) schematics they are 1k resistors R204 to R211, on the schematics with the updated version they show as 0 ohm links W9 to W16. What's on your functioning board may look like a resistor because they are machine placed at assembly and normally have just a single black band, but they are a wire link of zero ohms.

This explanation would make sense because all the drops on Barracora are on the same switch column, so any combination of five drops prevents the switch read from functioning.

So, to solve your problem ensure R204 to R211 (or W9 to W16) are wire links and I think you'll be fine.

Only speculation on my part, but I believe it is tied up with when the WMS drops stop using the horseshoe contacts and went to those with the switch blades at the bottom. If I'm correct the issue is too many simultaneously closed switches causing the column drive to run out of puff (that's a technical term - i guess I mean current ) If you look at the switch matrix circuit the column drive splits to a parallel path at each row (with a closed switch) which would divide the current.

Remember the horseshoe drop contacts had a momentary contact on the way down for scoring and a single series "all in bank down" switch - which very typically just reset the drop bank to up. So with this arrangement there would be relatively few simultaneously closed switches.

I guess for WMS another approach to this problem, other than removing the resistors, would have been to spread the individual drops across different switch columns, but this would make a dogs' breakfast of under playfield wiring.

Does anyone know which game WMS went to the new style drops, the Barracora and Jungle Lord I have use the new style, my Firepower has the old style - maybe Alien Poker or at sys7?

I have noticed some driver boards with zero ohm resistors are more likely to have switch bouncing in system 6 games. Those resistors and cap would be "decoupling" purpose i'd assume.

I'm not sure how a series resistor would do any debouncing? No caps were changed in the design as far as I understand. I imagine the blade / leaf style switches used in pinball are fairly grubby, but that a matrix read method would have some inherent debouncing in it just due to the timing? How are you detecting the bounce you observe?