HELP - Amiga 600 with non working keyboard

[D94pam]

Hi,

I would really need some help troubleshooting an Amiga 600 where the keyboard stopped working (after being stored for a while).

The machine has been recapped some 2-3 years ago and back then the keyboard worked just fine.

What I have done is the following:
- Tested the keyboard in another 600 where it works fine (ribbon cable is in other words ok).
- Measured for continuity from all the traces leading up to the white keyboard connector.
- Reflowed the solder on the keyboard controller and connector in case of a micro crack.

The keyboard is completely dead. Caps lock does not even lit up when I press the button.

Thanks!!

[solarmon]

What is your A600 motherboard revision and what is your keyboard - has it got a blue or green membrane?

You can simulate the keys presses by shorting certain pins on the keyboard connector or on the keyboard MCU - I would test this to rule out the membrane, keyboard connector or keyboard MCU.

[D94pam]

Yeah, that sounds like a good approach.
Motherboard is a Rev 1.5 and the membrane is green.

[solarmon]

OK, here's what I would do.

Use Amiga Test Kit for the keyboard test feature.

Then use the following resources to understand the keyboard matrix:

https://github.com/BleuLlama/AmigaSc...yb_Mitsumi.pdf

and translate that to the A600 schematics:

https://www.amigawiki.org/dnl/schematics/A600_R1.5.pdf

And use amigapcb.org to figure out where those signals are on the A600 motherboard:

https://www.amigapcb.org/

(although, this only has it for rev 2 A600, the pinouts for the keyboarfd connector and MCU should be the same).

For example, the Left Amiga key (LAMI) is in the PB6 trace in the keyboard matrix. This is active (i.e. pressed) when it is connected to ground. PB6 is connected to pin 14 on the keyboard connector (CN13) and pin 26 on the keyboard MCU (U13). Use amigapcb.org to locate where these pins are and you can short it to ground to simulate that key press. If the keyboard MCU is working then the Left Amiga key in the Amiga Test Kit keyboard test should be highlighted. Repeat this for other keys if you want, by most of the keys require for the 'X' and 'Y' signals to be shorted, as per the keyboard matrix.

[D94pam]

Thank you very much for valuable information. The route to check functionality using the keyboard matrix I hadn't thought of. Brilliant!

Unfortunately everything is still dead. I tried a couple of different keypresses using this method.

But one thing that struck me when looking at the schematics (https://www.amigawiki.org/dnl/schematics/A600_R1.5.pdf) is the circuitry handling the power up reset. One of the caps there is an electrytic one and even though the machine has been recapped and the area looks at least ok (the problem area in this machine was caps located on the right side of the mainboard handling audio where pads/traces had been eaten up). But there might still be some issue here, so I will dig out my oscilloscope checking the signals are doing what they are supposed to here.

Maybe you have any other idea though that I can also look into? Or am I on the right track you think?

Thanks!!

[solarmon]

Just to be clear, maybe I misunderstood - is your A600 booting up at all? I.e. can it boot to the Kickstart screen and you can load Amiga Test Kit either via floppy disk or as an ADF using a Gotek?

Or is the A600 not booting up at all? If so, then that is a whole different issue and the focus shouldn't be about the keyboard!

[D94pam]

Yes, sure, absolutely! Apart from a dead keyboard everything else works just fine. So no misunderstanding.

[solarmon]

If your A600 is booting and working apart from this keyboard issue, then that could suggest an issue with the keyboard MCU (U13) or an issue with the _KB_DATA and _KB_CLOCK signals/tracs going to the ODD CIA (U7).

Have you run health checks using Amiga Test Kit - especially a CIA test?

[D94pam]

I did a health check using Hertell's DiagROM and now after your suggestion also Amiga Test Kit. All CIA tests are passing.

Interesting thing (using ATK) is that when doing the keyboard test a number of times (going in and out of the test) I managed to get some kind of signal (bouncy though with a bunch of zeroes being sent) one single time from the ~ key.

So it is very much leaning towards a faulty MCU or the signals from the ODD CIA. That will be my next deep dive. Just wanted to let you know my status.

[vk3heg]

I had to replace the membrane in my A600. I tried a replacement flex one, and it only lasted six months,. Then came across the hard-keyboard-membrane. Ordered one, and it has been a delight.

https://www.sordan.ie/product/712/ne...0-green--blue/


Stephen

[KONEY]

I also got the Sordan membrane and the solution is great!

Only problem for me are keys far from centre like ESC, HELP, DEL, BACK and directional arrows. They need to be pressed with a lot of force, too much. Not sure what to do, screws are already tight.

[amiman99]

Originally Posted by KONEY

I also got the Sordan membrane and the solution is great!

Only problem for me are keys far from centre like ESC, HELP, DEL, BACK and directional arrows. They need to be pressed with a lot of force, too much. Not sure what to do, screws are already tight.

I have a similar problem as you, but it's a SPACE key and some other key. I'm thinking it might be the plunger, it may need a cleaning or a new one.
I don't have time to disassemble the keyboard to see if I'm correct, to many screws.

[solarmon]

Check the plungers. Swap the suspect keys with known working ones. If the issue follows then it is the plungers - which you can clean/wipe (not with isopropyl alcohol), or what I have done is to use a thin layer of silver conductive paint.

[KONEY]

I was thinking the same but before fitting the new membrane they worked OK plus I double cleaned everything before putting it back together. So I don't really know...

[KONEY]

I'll give it a try!

Originally Posted by solarmon

Check the plungers. Swap the suspect keys...

[KONEY]

I fixed it! Solution was to apply a layer of duct tape inside the metal tray, actually 2-3 layers near not working keys. This extra thickness did the trick.