14 Mar
1998
14 Mar
'98
7:14 p.m.
] From: Wirehead Prime <wirehead(a)retrocomputing.com>
]
] In repairing the SWTPC 6800 MP-A CPU board, I've discovered I THINK that
] a 7474 which is used to generate NOT HALT to the CPU (and to halt it
] under certain circumstances) is keeping the CPU halted all the time even
] though the bus says there's no reason for it to be halted.
]
...deletia
]
] Thanks...
]
] Anthony Clifton - Wirehead
From: "Jack Peacock"
<peacock(a)simconv.com>
considering the low speed of a 6800, an LS (or even an HCT)
should work...but, are you sure it isn't the case of a fast
pulse on the D input when CLK is hit (on the falling edge if
memory is correct)? and you aren't seeing it (using a logic
probe or a triggered scope)? did you check the voltage in to D?
maybe it's in no mans land (i.e. around 2v). Maybe its the IC
driving the flip flop D input thats bad, or a fast pulse is
hitting the R* input
Anyway, I'd take it out and put in a socket if you think its
most likely cause
Jack Peacock
Do you have anything funky plugged into that box? Later SWTPC systems
had disk controllers doing DMA, but I'm not sure about the 6800.
Does it lock up all the time? If so, can you yank everything except
the MP-A and an MP-C or MP-S in port 1, and see if it still happens?
In that case, there should be nothing wanting to do DMA, so the HALT*
line should be permanently high - no possibility of sneaky fast pulses
on the HALT line that you might overlook. Then everything is simple:
If HALT* is still high on the SS-50 and low on the output of that
7474, then that 7474 is definitely ill. If the problem goes away
when you do this, then maybe the 7474 is okay and maybe you've got
some board doing spurious HALTs.
OTOH, to keep the CPU halted, you'd have to be getting one of those
sneaky pulses on the HALT line in every clock cycle; just one here
and there wouldn't keep the CPU totally halted. Just on that basis
alone, I'd say it's almost certain that the 7474's flip-flopping
days are over.
As for 7474 vs. 74LS74, since the output of this is only driving the
one CPU input, I expect it should be okay. I think the main concern
would be whether or not the replacement chip can drive as many inputs
as the chip it is replacing, and an LS output can certainly drive one
standard input. (Experts are welcome to correct me here!)
Cheers,
Bill.