6128 keyboard woes *** SOLVED ***

[ervin]

I guess that multimeter is earning its keep! Congratulations on at least restoring some functionality, any victory in these matters represents a good learning experience. The keyboard I'm using on my GX4000 plus modification is getting a little problematic I think it may be a faulty membrane as all the connections seem ok so I may have to use the keyboard from the faulty 6128 I just bought. Well at least as a temporary solution. I suppose the keyboard membrane gets the most mechanical punishment so some problems are to be expected. I might have a try making a membrane from scratch when I have the time, just for the hell of it 

Yeah not bad for a cheap and cheerful unit!
(It was only $10).

Making a membrane from scratch would be fascinating. What would you use to create the lines for each key group?

[beaker]

Anyway, where did you learn about the Aussie Holden ? (woops, off topic again) 

I know they brought the Monaro over to the UK a few years back, not sure if they released it in Europe via Opel?... wouldn't mind owning one to hoon around in 
Vauxhall Monaro

[Bryce]

Depending on which flat cable you're actually fixing/testing - One of the sockets is connected directly to IC101 (74LS145) pins 1 to 11 (skipping pin . These are the output signals to the keyboard. When the CPC is running, the keyboard is scanned by switching one of these pins to GND one after another. The other flat cable is connected directly to the AY, pins 7 to 14. If a key is pressed, the corresponding pin of the AY will go low and the CPC knows what you've just pressed. Test each pin of both connectors to the ICs to make sure each connection is really working.

Bryce. 

[ervin]

Depending on which flat cable you're actually fixing/testing - One of the sockets is connected directly to IC101 (74LS145) pins 1 to 11 (skipping pin . These are the output signals to the keyboard. When the CPC is running, the keyboard is scanned by switching one of these pins to GND one after another. The other flat cable is connected directly to the AY, pins 7 to 14. If a key is pressed, the corresponding pin of the AY will go low and the CPC knows what you've just pressed. Test each pin of both connectors to the ICs to make sure each connection is really working.

Bryce.

Alrighty, I've been doing just that.
And it looks as though all the messing around with the keyboard has been a waste of time.

One of the pins of the AY appears to be dead. So it's hopeless. I have no idea how to fix this. 
I ruined a perfectly good membrane because I didn't know what I was doing.

I went along the AY pins, and matched up the pins with the pins on the keyboard connector.
All of them had current except the one that (I think) connects to the line that SPACE etc are on.

Apologies for the photo quality:

[protek]

@ervin: You could try to smudge some graphite over the remaining lines with a pencil. Worked for my +3 membrane and it is a commonly known way of unlocking old Athlon and Duron cpu multipliers.

[ervin]

@ervin: You could try to smudge some graphite over the remaining lines with a pencil. Worked for my +3 membrane and it is a commonly known way of unlocking old Athlon and Duron cpu multipliers.

Great idea! I'm off to try it right now!

[Bryce]

That's not going to help if it's the AY that's dead. Are you sure that the pin isn't working? Define it "not having current"? Do you mean it doesn't connect to the connector?

Bryce.

[ervin]

That's not going to help if it's the AY that's dead. Are you sure that the pin isn't working? Define it "not having current"? Do you mean it doesn't connect to the connector?

Bryce.

I think the AY itself is ok.
Certainly when I press CLR, I get that familiar beep.

When I multimeter between the faulty little silver dot (techie term  ) and the AY leg thing, 7 out of the 8 that I presume connect to the keyboard connector show current.

It's when I try to check the little silver dot (second silver dot from the left in the second photo above), there is no current between it and the corresponding AY leg.

EDIT: I've just read that post back to myself - I sound so clueless!
Ah well, let's face it, I am!

EDIT 2: I've had a look at the AY underneath the motherboard.
Everything seems clean and intact.
It's just the bit on top that stuffed.

Incidentally, I'm amazed at how compact the motherboard is. Just seems so small compared to modern boards.
Of course, it is around 1 million times simpler!

[Bryce]

Why are you measuring the little dots, you should be measuring between the cable connector and the AY pins.

Bryce.

[ervin]

Why are you measuring the little dots, you should be measuring between the cable connector and the AY pins.

Bryce.

I tried that, and there is no current between the cable connector and the bad pin.
(However, there IS current between the cable connector and the dot in front of the bad pin).

The other pins do show a reading between themselves and the cable connector.
I just noticed that on the other AY pins, there is current between the little dots and their respective pins, but the bad pin doesn't.

[Bryce]

Is there any damage to the PCB in that area (check the bottom side too)? Maybe the track is broken? If the AY is ok, you may have to make a wire link from the AY pin to the correct connector pin.

Bryce.

[00WReX]

Just a quick question from me,

Ervin has indicated that the "X" on the keyboard is not working but when the joystick fire button is pressed it produces an "X".

I would have though this would have been using the same data line into the AY at some point ?

Cheers,
Shane

[Bryce]

Yes, but the pin he mentioned isn't responsible for the X either, so I assume the AY is ok and there is a break somewhere else.

Ok erwin, here's something to try. Get a piece of wire and starting with the first pin of output connector (keyboard doesn't need to be connected) hold the wire on pin 1 and then run the other end along the contacts of the input connector (the one attached to the AY), a row of characters will appear on the screen. Now do the same for pin 2 and so on. That way you will be 100% sure whether the PCB or AY really has a problem.

Bryce.

[ervin]

Is there any damage to the PCB in that area (check the bottom side too)? Maybe the track is broken? If the AY is ok, you may have to make a wire link from the AY pin to the correct connector pin.

Bryce.

The PCB appears to be ok. To my inexperienced eyes, at least.

[ervin]

Yes, but the pin he mentioned isn't responsible for the X either, so I assume the AY is ok and there is a break somewhere else.

Ok erwin, here's something to try. Get a piece of wire and starting with the first pin of output connector (keyboard doesn't need to be connected) hold the wire on pin 1 and then run the other end along the contacts of the input connector (the one attached to the AY), a row of characters will appear on the screen. Now do the same for pin 2 and so on. That way you will be 100% sure whether the PCB or AY really has a problem.

Bryce.

Thanks for the tip Bryce.
I've done as you have suggested, and here are some results:

In this picture, I've tested the line from the keyboard that isn't producing anything.
I presume the weird characters are related to the CTRL and DEL keys etc?
Anyway, V X etc are appearing, and they are the characters that aren't working with normal keyboard use.



Then I tested the next line in the keyboard connector.
The keys on this line work with normal keyboard use.



Does this mean that the AY is ok?

[Badstarr]

Yikes, the old keyboard continuity connection checking dance! Something I'm all too familiar with! Fear not though, from my brief skim through the recent posts on here I can see that a fix will most likely be possible! The worst case scenario is that you might have to do a little soldering to reinstate the connection that looks to be missing at the AY. It may be worth getting a decent soldering kit to keep your new multimeter company, you seem to be prepared to get your hands dirty which is a good thing! Everyone makes mistakes along the way to learning a new skill.


The AY connections should go direct to the poly connector where you plug the keyboard in, so you just need to trace the correct pin on the connector and solder a fly cable to the correct pin on the AY assuming its not just a simple dry joint situation, which seems fairly likely based on your description. Also check the schematics section on the main wiki as it's always good to have a map to guide you where you need to be, it may be all a bit confusing if you aren't familiar with electronics but take some time to study them and you will soon start making sense of them! But do always exercise caution though when working with electricity!


As for building a membrane I was thinking about going for a more robust option, maybe using some old fashioned wires and tougher plastic, I think Amstrad used a similar solution for one of the 464 keyboards and from what I remember they are considered practically indestructible!

[Bryce]

I'm all with you badstarr, sound advice, but I think he should wait until the new keyboard has arrived. It's still not clear whether there's a problem on the board at all. If he plugs the good board in and it stil has problems, then he can start soldering IC pins, but until then, I'd hold back on the PCB experiments. You don't want to break anything further if it's not necessary.

Bryce.

[ervin]

@ervin: You could try to smudge some graphite over the remaining lines with a pencil. Worked for my +3 membrane and it is a commonly known way of unlocking old Athlon and Duron cpu multipliers.

That trick didn't work, but it did lead to something else...

[ervin]

IT IS WORKING!!!


I dabbed a bit of conductive ink over the offending broken line (from the little dot to the AY pin), waited for it to dry, cured it under a bright light, and now SPACE, Z, X etc work for the first time!!!

I am so chuffed right now!

With my aluminum foil repair, I'm not too sure that the key repeat rate is consistent, but it's only a temporary solution, so that's ok.

THANKS EVERYONE for all of your help. I could never have done this without you.

[00WReX]

WooHoo...well done, your persistence with this has been impressive.
Good to see you did not throw your hands in the air and give up. 


Hopefully when the replacement membrane arrives it will all work as good as new.


Also just noticed that from the screenshot your 6128 is an "Awa" and not the later "Amstrad"...it's only a single link on the motherboard that changes this, but it indicates yours is an earlier Australian delivered 6128. Cool. 

Cheers,
shane

[ervin]

Thanks Shane.

It's been a heck of a ride, and it isn't over yet, but there is light at the end of the tunnel!
The hacky way I've fixed the keyboard isn't really ideal, as the key repeat rate seems a bit wrong, but it'll tide me over until a proper working membrane arrives.

Yeah, I noticed the "Awa". It means that this computer is 26 or 27 years old, I believe.
It's amazing that these things still work after all these years!

[Gryzor]

 Oh man, this is touching indeed =) Have lots of fun! And, hey, what about the HxC?

[ervin]

Thanks Gryzor.

The HxC hasn't arrived yet.
Hopefully it'll arrive this week, I can't wait to try it.

[ervin]

Hi all.

Just wanted to let you know how this has all ended up.

My aluminium foil method turned out to be a bit of a dud.
I just couldn't get all 8 keyboard lines to work consistently, and without "cross-contaminating" each other with current.

So in a fit of rage I ripped them all off.
(Actually, I removed them really carefully).

I then tried various ideas, related to conductive ink, foil and conductive ink, conductive silver paint, more ink, more paint, more foil etc. etc. etc. I'm surprised that I had the patience to try so many things, quite frankly!

Anyway, in the end I came up with a successful solution!

The 8 tips of the membrane were re-created using aluminium foil tape, cut into short, narrow strips.

I then painted over the entire narrow part of the membrane with conductive silver paint. 
Once that dried, I etched gaps between the 8 lines using a sharp object. 

In the end, I had 8 lines connected to the main membrane body with the paint, and at the end of narrow part of the membrane the aluminium foil strips connected with the paint.

And it worked! Some keys had a sort-of-rubbery feel to them for some reason, and (as I've mentioned before) the key repeat rate seemed slightly off and inconsistent, but that's okay, because I had a fully working keyboard! (Even the keys that didn't work and led me on this adventure in the first place).

Here's how it looks.









It isn't a perfect solution of course.

And that's where ynot.zer0's generosity has come in.
The VERY SAME DAY that I got the above solution finished, my replacement membrane arrived.

I gave it a good thrashing, and it all works perfectly.
No rubbery key response, no key repeat problems.

Just a perfect 6128 keyboard. AT LAST.


Okay, that's enough from me.
Thanks again everyone for all your help and suggestions.
This is a wonderful community that I feel very privileged to be a part of.

[protek]

Good to hear your keyboard is now finctioning perfectly.