Williams High Speed (Williams, 1986) System 11 CPU help please

[Nighter]

Hey, having some issues with a system 11 CPU board from a High Speed. Specifically random switch closures. Lead me to investigate the switch matrix.
I found that two of the eight matrix drive lines(1J8-4 and 1J8-7) are pulsing at twice the frequency as the rest.
I suspect the PIA at U38, so I replace it. No change.
I suspect something up with U40 or the transistor drive stuff, so I bend up the 8 legs of the PIA that feed U40 (pb0-pb7) to not make contact with the socket and retest at pins 13 and 15 on U38 (pb3-pb5). No change. (this has eliminated every thing after the PIA)
I’m at a loss now, so I try a new set of game roms at U26 and U27, no change.
<O</O
Now I really don’t know where to go. Any Ideas what could cause two of the lines to cycle at twice the frequency? Something corrupting the data or address bus?
<O</O
Any thoughts would be appreciated, I’m pulling my hair out on this one.

 

[pinballdaveh]

have you ruled out U30 and U39? does the switch edges test show any faults or codes? what happens when all rows and columns jacks are disconnected?

 

[Nighter]

have you ruled out U30 and U39? does the switch edges test show any faults or codes? what happens when all rows and columns jacks are disconnected?

Yes, U39 and U30 have been ruled out by bending pins 2-9 on U38 up out of the socket. I did this at the same time I did pins 10-17 so I'm fairly certian the issue is before U38. The board is on the bench so p/f or conectors are not causing the issue. I'm looking at the outputs on a scope, its a nice clean signal, just the two drives (pb3 and pb5) pulse twice as much as the rest.

 

[pinballdaveh]

do all the inputs and outputs of scr6 read the same? ive never heard of one of those rc packs going bad. are the values of c54 and c55 read the same as the others? are R71-R78 all within spec? are Q47 and Q48 the correct part#s