A600 keyboard dead

[Jope]

I have an A600 that only lights up the caps lock led on the keyboard (and it stays on for ever), doesn't accept any other keyboard input than keyboard reset. (I'll have to check that kbd reset again, I might remember wrong)

I checked the schematics and verified all the components near the keyboard controller, they are ok. Also no short circuits anywhere, the connections match the schematics.

I couldn't reliably check the oscillator - I have no scope.

The keyboard controller gets extremely hot, which leads me to believe it's fried.

However the question here is, could a fried CIA chip be the real culprit here? I don't want to swap surface mounted components just for the fun of it..

It seems strange that kbd reset would work on the keyboard processor, but nothing else does.

In case someone gets stuck on the fact that this is an A600, the keyboard controller is the same part number as is on big-box and A500 keyboards. Just in a PLCC case and mounted on the motherboard.

[Jope]

yoodoo wrote:
Dodgy or unseated ribbon cable, perhaps?

Not this one. :-)

Noster wrote:
keyboard controller?
If I remember right, the signals of the keyboard are processed from a CIA, Paula, and Gary. Or is the controller chip the one that is found inside the keyboard of a "big-box" Amiga (e.g. A2000/A3000/A4000)?

The A600's keyboard is very much like an external keyboard, except it's all integrated to the motherboard.. So the controller is just plopped on the motherboard and wired directly to the CIA, instead of having it in the (separately boxed) keyboard and having connectors in between too.

Isn't the keyboard reset processed from Gary instead of the CIA? So it would also work if the CIA is dead.

The keyboard controller is where the reset signal comes from in the A500/600/1200. Thus I find it hard to believe that it can't send stuff down clk and dat but it CAN decode a keyboard reset?

In big box amigas, the controller could also give this signal, but there is no wire for it in the cable.

The A500 has the same keyboard controller and it's keyboard also has one extra transistor (missing from big box keyboards) to send the signal down the reset wire.

This is why I'm lost now. It's either the CIA (since data doesn't seem to flow from the kbd controller) or the "keyboard" (since it's totally unresponsive and caps lock is always on), but in this case it's a real pain to swap either one, since they're surface mounted.

Unless someone has a better idea, I'll soon disconnect the keyboard controller's clock and data outputs and wire an A500 keyboard directly to the CIA. That will give a definite answer, but I always like to investigate things fully before starting to make modifications. :-)

[Jope]

alexh wrote:
The three fingered salute (keyboard reset) is almost certainly a hard wired input to the processor. It should be separate to standard keyboard I/O. Might explain why that works and nothing else does.

It is indeed a separate line going from the kbd controller to the general reset lines that criss cross the motherboard.

I just find it odd that the chip can "half fry", with enough key matrix decoding happening that the reset is still working but nothing else works. It does not even acknowledge that it should switch the caps lock mode. :-)

Oh and I have had dozens of Amigas, with various keyboard failures in some of them, but never anything like this.. Actually I've never even read of these kind of symptoms happen to anyone else on the Amiga forums I frequent.

[Jope]

It was the keyboard controller that was fried.

I wired an A500 keyboard directly to the motherboard and that works ok. (except for caps lock, I think this keyboard is a bit broken too. :-)

Keyboard reset also works by wiring the A500 keyboard's kbreset line to the A600's reset plane. :-)

I think I'm going to hide this A600 in some industrial enclosure.. Just a reset switch + leds on top and all the ports sticking out from the rear. (also joy/mouse).

Perhaps also build in a ps2 keyboard adapter.