13 Sep
2013
13 Sep
'13
12:19 p.m.
On Fri, Sep 13, 2013 at 3:03 PM, David Riley <fraveydank at gmail.com> wrote:
Did you stick a scope on the DOUT pin of the chip,
both installed and
not?
I have not yet pulled out a scope. That's in the next round.
It's possible you have leakage in there from
something that is
just below threshold when the RAM isn't installed and sticks it above
threshold when it is.
Good thing to watch for.
I was going to suggest checking the same on the DIN
pin, but then I
looked at the schematic and realized that it's connected directly to
the 68K without a 245 in between. Nonetheless, it's worth checking
anyway.
Sure.
The other thing I could suggest is checking the
'AS253s which serve
as the address muxes. It's entirely possible that there's a flaky
bit on the output of those, which would explain the regular pattern,
especially if they're blinking (which doesn't necessarily mean
software is modifying it, just that it's not consistent). Given that
those are early make AS parts, I wouldn't be surprised if one of them
has gone off, especially given how warm I've seen them get.
Good tip. I hadn't (yet) noticed they were AS parts. *Those* I
wouldn't have in my box of spares. I may have to order a
handful if my probing indicates they are suspect. I was considering
them because of the regular nature of the ghost bits.
Thanks!
-ethan