Osbourne OCC1 problem

[Philip Pemberton]

Ade Vickers wrote:
> 	MOV format (4mb): http://www.solutionengineer.com/ozzie/occ1_prb.mov

Something in the video generator circuit on the digital side by the looks of 
things. Probably pre-character-generator -- those characters look fine and 
they're in the right place, but they're the wrong characters. So that's 
anything pre-character-generator basically -- buffers, latches, RAM, not 
necessarily in that order.

Point me to a PDF of the schematic and I'll see if my overworked brain can 
come up with any other ideas :)

> Any ideas where to start looking? I have an oscilloscope (albeit I've
> forgotten how to use it, and am not 100% sure where the probes are), and a
> multimeter... beyond that, not a lot.

Have a read through the Tektronix "ABCs of Oscilloscopes" guide - that's on 
<http://www.tek.com/> somewhere (their search engine usually works fairly 
well, or try Google). That should be enough of a reminder of the terminology, 
and how to make the scope work.

Now the things I'd try in order:
   Set the scope to 0.5V/div, AC coupled and connect it between +5V and 
ground. See if there's any significant noise on there. If there's around 0.1V 
or less of noise / pulsing on the power line, you should be fine. Any more 
than that and you want to check the power filtering. Check using various 
ground and +5V points -- this might expose a bad track or via.

   Does the keyboard work, or is the machine totally crashed after booting 
(assuming you leave the terminator where it is)? Does a CAT (or DIR, or 
whatever the relevant CP/M command is) show something similar to a directory list?

   Check the address lines for the display RAM. If it's DRAM you're pretty 
much screwed, but with SRAM you can see if the count is incrementing normally 
or if there's an output line stuck low.

   If the RAMs are socketed (and the same type), try swapping them around. If 
you know the main RAM is good, swap each VRAM chip with a main-RAM chip until 
either the main RAM test fails or the video stops glitching. This, of course, 
only works if the main and video RAMs are the same type.

> Basically, everything works except for the shaky video. Sometimes you see a
> whole page full of 1s, or 0s; essentially, it's all a bit random. Bad
> connection somewhere, perhaps, or maybe a failing chip?

A screen full of one character may well be a power-on test or something 
similar... If it's happening after power up then something is very badly wrong.

-- 
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/