New guy, "new" mute 6128 and a broken keyboard

[Velktron]

Hello everyone, I'm a long time Amstrad CPC user even though I never really had one of my very own, until now. The first home computer I've seen must have been my cousin's Schneider CPC464, with an awesome collection of games on tape. I kept visiting my cousin all the time and pestering him just to use that little green bitty box 


Other friends of mine were more "elite" and had spiffy 6128s with their posh 3" disk drives and all....but only one guy I knew had a color monitor, heh.


Anyway, after years and years, I finally found my first very own 6128 in a junk store, for 5 Euros (just the computer though, no discs and no monitor). Not a great problem though, being the hardware tinkerer that I am, I didn't have much trouble cobbling together a power supply from one of my spare PC ATX ones, and even connecting a 3.5" drive to it (excellent guide there on the CPCwiki BTW, I wish I knew it was this simple back in 1996, when I had the chance to tinker with a couple of borrowed CPC!).


As for the monitor, being too lazy to make a RGB SCART lead for it (besides, I haven't found a 6-pin DIN for that yet...), I tried with a simple luma + sync composite cable, good enough to get a B&W pic to see if it, at least, functions.


Well, it did turn on, I got a  B&W picture on a TFT TV, as expected, the internal floppy reacts (no idea if it will need a belt though, I have no 3" discs to try), however some of the keys were dead (in particular, the E, R, U, I, O, |@ , 7, Return (!) and CLR keys). Yup, not an ideal situation, as many commands depend on them 


After some keyboard schematic tracing, I discovered that all of those keys belonged to the same matrix line, however after testing the keyboard membrane I concluded that there was nothing electrically wrong with it: all key pads were conductive, and test currents from an ohmmeter reached all the way to the ribbon connector from any of them, and resistance was no different than that of the other lines.

Then, I tried probing the mainboard connectors directly with an ampmeter (with the CPC ON  ) and discovered that the keyboard matrix line that controlled those keys (on the left-hand ribbon cable connector) was completely dead, and would not respond even to direct probing/shorting it to any of the right-hand connector: it was not possible to get any of those keycodes to register, while I managed to get almost every other keystroke by using this method.

Two other quite worrying symptoms: no sound output at all, either through the speaker or the headphone port, and even plugging the joystick in results in stray/bogus keycodes being entered along with the usual arrows and "X" inputs... I suspect that there is some deeper damage here, e.g. a cracked PCB, the AY-3-8910 chip...which indidentally also does keyboard decoding..

So, to my question: is there anyone here with experience facing a similar problem? How probable is it that just the AY-3-8910 (or some of its internal lines) went bad? Is there any way to check the "health" of the signal lines to the AY chip, or a utility to check whether it reacts correctly to being probed? I would like to exhaust all options before going down on the PCB with a soldering iron and trying to manually bridge connections....ugh...

BTW, I'm able to load some programs from the 3.5" floppy in this state, as long of course as their filenames don't contain any of the offending characters
The hardware seems stable enough to run something demanding like Batman Forever (yup, tried that one as soon as I got the drive going ;-) so it would seem a bit odd that physical hardware damage could have settled in...

[TFM]

Welcome here! I'm sure you will like it

[Velktron]

Thank you :-)


This is just my second day of testing my "new" 6128, and so far it went better than expected, if we exclude the sound and keyboard problem (e.g. I didn't expect that it was so easy to get software on 3.5" "CPC" disks, I still had the dreaded 22disks MS-DOS utility and 5.25" drives in mind, and horror stories about drives without ready signals, while a couple of jumpers and a cable did the trick, go figure  ). Even this way, it made me quite happy :-)

[Bryce]

Hi Velktron,
     The fact that you are having problems with the sound AND keyboard would suggest that the AY is dodgy, but also check whether it's decoupling capacitor is damaged (although unlikely). You'll need to swap the AY, but they are relatively easy to obtain. If your soldering skills aren't up to swapping the AY, then let us know.

Bryce.

[ralferoo]

Have a bit more of a check, testing all the keys in the same rows/columns of the matrix: Programming:Keyboard scanning - CPCWiki

If it wasn't that you said CLR didn't work, all the other keys that failed were column 2 and 3, so test to make sure if it's the whole column or just those keys.

The fact that columns 3 and 2 are both dodgy might mean that the keyboard IO for bits 3 and 2 are somehow damaged or possibly the data lines themselves are damaged - possibly corrosion or failed solder joint. If you download the schematic, you could do a continuity test between the Z80 data lines and the AY data lines.

[Gryzor]

 Welcome, mate! Hops you stay and have fun here


Just one question: are you trying to fix it just for the fun of it? Because it's relatively easy and cheap to find a working one...[nb]Of course, each and every CPC is worth saving [/nb]

[Velktron]

Well, it's true that they weren't exactly rarities back in their heydays, but it's the only one I've seen for sale in my favourite thrift shops (and believe me, there's a lot of junk and rarities to be found there  ). I probably just want to fix it for the fun of it, though. If I simply wanted to play some CPC games with the least amount of trouble possible, I'd just get an emulator.

[Gryzor]

If you're indeed in Greece as your location states, you could do worse than trying the Retromaniax forum. Where in Greece are you?


*sigh* I miss ELX, which was looted in the riots a couple of years back...