Keyboard Dead

[flashjazzcat]

Need some troubleshooting advice. Just re-socketed Pokey on my 130XE, since I replaced the square pins on the stereo board with round ones (the board had previously been pushed into the subsequently ruined machine socket). Work on the stereo board went well, and I replaced the machine socket on the motherboard with a nice precision one. No lifted traces or any other nasties... very clean job. However, when I came to test everything after reassembly, the keyboard's stone dead. Well - not quite: pressing either shift key actually registers as the "Break" key, which is rather disconcerting. All other keys - apart from Reset - are totally unresponsive. I've tried different keyboards, but the problem's definitely on the motherboard. Checked for shorts, continuity, etc... found nothing. I'm wondering if the Break / Shift thing offers a clue; maybe one of the support chips gone? I've very closely scrutinized the keyboard scan pins which are connected to the 4051s... no visual or continuity issues there. Seems everything I touch turns to s**t at the moment, but I'd like to get this fixed as I put a hell of a lot of work into this machine.

[flashjazzcat]

Socketed and replaced both 4051s. Still the same.

[+bob1200xl]

POKEY does the keyboard - have you tried that? (silly question?) Can you un-do the stereo board?

 

Bob

[flashjazzcat]

Stereo board's removed, Bob. So far, I've gone to the extreme lengths of grinding off the precision socket, checking the traces (all good),washing the motherboard, soldering in a new precision socket... and the thing's still exactly the same. Shift produces "Break", all other keys dead.

 

Fresh out of spare 65XE motherboards I'd like to mod up, and fresh out of ideas. I'm just wondering if the fact the Shift key triggers Break is any help???

[erazmus]

grinding off the precision socket

Ouch, ouch, ouch. The XE boards are terrible for traces lifting with too much soldering iron heat. Did the through plating survive? Perhaps in the process of removing the socket, you tore out a through hole plate, thereby disconnecting a top trace from a bottom trace?

[flashjazzcat]

Ouch, ouch, ouch. The XE boards are terrible for traces lifting with too much soldering iron heat. Did the through plating survive? Perhaps in the process of removing the socket, you tore out a through hole plate, thereby disconnecting a top trace from a bottom trace?

 

It's a fair question. Actually, this XE board is remarkably robust, though. If I have to "jump ship" to another board, I'll be worried: I have a Rev-5 which is VERY flimsy, and an 800XE board (which looks the more resilient of the two).

 

In any case: thanks for the suggestion. It's something to check.

[erazmus]

Looking at a 130xe schematic which I found here:

 

http://atariage.com/forums/topic/127716-good-quality-130xe-schematics/?do=findComment&comment=1541087

 

I would use a meter to check continuity from the keyboard connector to U24 and U25 (4051) (page 6 of the schematic). Then I would check from U24,U25 to POKEY. If you check continuity from the top of the board, with the chips plugged in, by touching the top of the IC leg, this will make sure that there's no break in the trace from a bad via or anything. Possibly U24, U25 is bad? Do other POKEY functions work, like the pot inputs, sound, etc?

[flashjazzcat]

Can't disable SDX on the U1MB, but Pokey is still producing sound when the machine boots up. I'd already checked continuity between Pokey and U24/25, but I've since checked between U24/25 and the keyboard connector, and found no breaks. U24 and U25 have already been swapped out, but symptoms remained identical even after desoldering and socketing those two chips.

 

Candle suggested a short might be to blame, but I simply can't find anything. Removing and replacing the Pokey socket didn't change anything, so it's unlikely any further damage was done at that stage, since symptoms remained identical. Exasperating...

[bandit]

I would try, (if you have not yet) insert a index card into keyboard socket, then install keyboard ribbon .

[flashjazzcat]

I would try, (if you have not yet) insert a index card into keyboard socket, then install keyboard ribbon .

 

Good idea... however, it appears to be fixed.

 

Having tried everything else, I did the (umpteenth) visual inspection of the underside of the Pokey socket. I'd been a good boy and used plenty of flux and wick (as well as solder pump) when removing the original machine socket (and first precision socket), but it looks like a SINGLE STRAND of wick had survived both socket replacements, creating a tiny, fragile bridge between two of the keyboard matrix pins, embedded in flux residue. I scraped this away and - hey presto.

 

Thanks for all the help and suggestions, guys. This just turned out to be one of those flukey things which bite you in the ass even when you're sure you did everything right.

[erazmus]

Result! Congrats.