neoncluster.com

Problems with striped screen

With character ROM removed

With character ROM removed (close-up of cursor)

Close up showing cursor with pixel inverted in the centre where the line passes through it. I was hoping that this was going to be some kind of insight for me to fix the machine...It wasn’t.

Started looking around the address lines D0 thru D7 that feed the video signal. Pin 15 of IC12 wasn’t looking quite right.

GOT YA! 

IC12 (74LS166) was looking mighty dodgy. Whipped it out. Dropped in a socket and it was fixed

I often pull out machines that I haven’t used for a while and give them a thorough working out to check if they need any maintenance and to keep the keys moving and functioning. My two Microbees are no exception.

This time the 'W' key on my Microbee PC-85 was funky and didn't work at all, but nothing a drop of 98% alcohol and about 100000000 key presses couldn't fix. Microbee key-switches are notorious for failing, and I find it a constant battle to keep them all working. Every time I power them up I find some keys that aren’t functioning correctly. Fortunately this time it was a simple fix, with no need to break out the soldering iron.

However my Microbee 32k IC was another story. On power up the screen appeared striped. I could read the characters through the striping, and it otherwise seemed to function correctly.

First I swapped the core-board (that contains the ROM and RAM) from my PC-85 model into the 34k IC model, but the problem persisted. That eliminates the core-board as the problem. Therefore the issue must lie on the main motherboard.

I then removed the character ROM. The screen was white, and I could see the cursor, no stripes were visible, but that could mean that white stripes just don’t show on a white background...that test didn’t seem to give me any useful clues.

I then swapped the character ROM, although by now I had guessed it probably wasn’t the problem, and the 6546 CRCT between the 32k IC and the PC-85...still no change.

My assumption at this point was that there is a stuck bit somewhere, probably in one of the address lines, so I grabbed the schematics and started looking around the address lines D0 thru D7 that feed the video signal.

After about 40 mins I found that the S/R signal (pin 15) from IC12 (74LS166) was looking mighty dodgy, My logic probe was showing high and low signals, but no pulse! I cross checked with my working PC-85 and this pin had a strong pulse on pin 15. I have plenty of spare 74LS166 IC’s so I piggy backed one over the suspect IC and on powered up. The picture was clear!

I have been told, and have read that there is a certain amount of risk with piggybacking some IC’s (although I don't actually know the rules), but if I did damage the new 74LS166, I had plenty more so the risk seemed fairly minimal. A 50 cent part vs risking damage to the board during removal of a suspect part. That risk seemed worth it, and in this case It paned out.

Finally I whipped out the bad IC, and dropped in a nice machined socket. Done!

Fixed!

Fixed!