> > One thing I have found though - a *lot* of chips in this machine are
> > replacements, and the solder quality is total crud.
> Ugh. Add "Solder sucker", "Desolder wick" and "RoadRunner wiring pen" to the
> "Tools Required" list too. I hope you've got a good soldering iron and a
> powerful light (60W or so).
I would actuaslly be tempted to do a complete resoldering job on this
board. I don't like shotgun debugging as you know, but bad connections
will come back and bite you at the wrong time (i.e. when you're doing a
demo and don't have a soldering iron with you).
>
> > All the RAM chips and both MC6821's are replacements, and the sockets
> > for the LM1889 and MC6847 look to be non-original too, implying that
> > those suffered some form of destruction and were replaced. Given solder
> > quality, I'm going to check continuity on the address / data lines for
> > all the RAM chips and check for shorted tracks too.
> Watch out - the tracks to the RAM may be mixed up, i.e. A0 going to A8 and so
> forth. Check the schematics, then buzz out the board with a continuity
> tester.
Well, you can at least check that every address output on the SAM goes
to _an_ address pin on each of the DRAMs and that there are no shorts
between them.
>
> > For some reason one of the RAM chips has a 220n decoupling cap, whilst
> > the rest are 100n.
> Shouldn't make much difference. If it looks like the 220n is a replacement,
> swap it for a 100n.
I wouldn't worry about this.
-tony
Received on Wed Jul 21 2004 - 16:01:38 BST
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:36:52 BST