Thanks Bill, I hadn’t seen that particular page. As I mentioned already, the
formerly clear “stuff” was so deteriorated I could just pull the glass plate
off with gentle fingertip pressure.
I ran a bead of clear silicone around the outside of the clean plate and CRT
face and bonded them back together.
Display looks great!
However, I can now see that every other line, starting with line 2, is
showing a full line of double quotes (0x22) instead of spaces (0x20). I read
the circuit description and schematic, and it appears that bit “2” is stuck
high on the even-line RAM – for some reason the designer decided to call the
LSB bit 1 instead of bit 0.
Typing (for example) “abcdef123” shows the correct text on the blank odd
lines, but on the even lines it echoes as “cbcfef323”. Confirming that stuck
bit.
Looks like the RAM at location H15 should be the bad one... we’re having a
heat wave and it’s too hot upstairs to work on it until tomorrow morning at
the earliest.
ETA: Now it's tomorrow morning and just cool enough (although 100% humidity,
at least outside) but my replies aren't showing up in the archive - filtered
somehow.
Anyway. I did a bit more Googling and discovered that plain water dissolves
the PVA goop just fine. No need to use a lot of expensive alcohol which
seems to be a less effective solvent anyway!
So I took the board out and scrubbed it in the kitchen sink with running
warm water and an old toothbrush. Rinse with distilled water, now gently
baking in the oven at around 140F to get the water out of the keyboard. Then
onto the RAM replacement.
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
