A500 Keyboard Not Working (except Ctrl+A+A)

Starglider 2 · 12 September 2018, 02:15

Hi guys, I have the exact same problem as I've seen elsewhere, but no solution was ever posted to those threads. Here's the symptoms:

- After I replaced 2 faulty CIAs with working ones (which fixed all the issues I was having with drives & other weird crashes) the keyboard inexplicably stopped working
- Reinstating the faulty CIAs doesn't return the keyboard to working state so something has changed
- CAPS flashes only once during boot (normal)
- CAPS can be pressed and depressed (lit and unlit) during the boot cycle but once the machine has booted, CAPS stays lit up (if lit when pressed) and can't be released
- No keys are responsive
- The exception is Ctrl + A + A to reset

What would you try next?

r.cade · 12 September 2018, 02:36

Check that the keyboard cable has not been damaged. Other than that, broken CIA socket or bad CIA or bent pin?

Also very likely is just keyboard CPU died. I have several like this. Don't know what kills them... static to the CPU chip, plugging cable in wrong... who knows.

Starglider 2 · 12 September 2018, 04:30

Thanks. The CIAs are both replacements, unless 4 are faulty.

Continuity on the cable passes.

I guess that leaves the CPU!

Jope · 12 September 2018, 06:06

I doubt it's the CPU. The problem lies either in the keyboard MCU on the keyboard PCB, or the CIA at U7, or in the connection between them.

How did you measure the continuity? Just the cable or all the way from the keyboard to the CIA's DAT and CLK pins 39 and 40?

Starglider 2 · 12 September 2018, 07:12

Quote:

Originally Posted by Jope

I doubt it's the CPU. The problem lies either in the keyboard MCU on the keyboard PCB, or the CIA at U7, or in the connection between them.

How did you measure the continuity? Just the cable or all the way from the keyboard to the CIA's DAT and CLK pins 39 and 40?

Edit: I get continuity right through from the keyboard PCB cable header to those 2 pins.

The CIA chips are new and I've tried 6 of them.

Jope · 12 September 2018, 07:47

Do you have another keyboard to try with?

Starglider 2 · 12 September 2018, 07:54

Quote:

Originally Posted by Jope

Do you have another keyboard to try with?

No but I've ordered a replacement control board. Any ideas in the meantime sir?

Jope · 12 September 2018, 08:01

I don't have that many unfortunately.. If you have a scope or a logic analyzer, you could measure the clock and data signals from the keyboard to see if it's shooting anything when you power it up and hit some keys.

The keyboard reset goes via a different wire, so the keyboard is partially or fully working. However if either the clock or data output pin is burned, it will not be able to transmit any keycodes.

Starglider 2 · 12 September 2018, 08:03

I have a logic probe. So which pins should I be trying it on specifically when I hit keys?

Daedalus · 12 September 2018, 12:28

Clock is pin 1, data is pin 2. Also worth checking their continuity to pins 40 and 39 respectively on CIA U7, and check both of those lines for shorts to each other, ground or 5V. You should get a byte on the data line each time you press or release a key, and matching clock pulses on the clock line.

The protocol is described in detail here, which should help to see what's going on.

Starglider 2 · 16 September 2018, 03:06

Hi guys, so the plot thickens...

I got a replacement keyboard PCB very fast from eBay. Hooked it up, and exact same issues...

On boot sometimes the CAPS lock starts stuck lit up, sometimes not.
No keys work.
After about 20 seconds, if I press CAPS on and off a few times, it will get stuck on/lit.

I've reflowed all the solder around the CIAs, the CPU, and the keyboard connector on the mainboard. I also recapped the board.

Obviously the keyboard PCB itself is new, so that rules out the keyboard chips or the connector cable and plug.

The CIAs on the A500 are also new (they're the 5th & 6th I tried after other issues.)

So if it's not the CIAs, not the solder joints, not the connector, not the plug, not the cable, and not the keyboard PCB and chips, what could it be? The keys themselves somehow? Some other OCS chip?

What would you try next?

Starglider 2 · 16 September 2018, 05:09

Quote:

Originally Posted by Daedalus

Clock is pin 1, data is pin 2. Also worth checking their continuity to pins 40 and 39 respectively on CIA U7, and check both of those lines for shorts to each other, ground or 5V. You should get a byte on the data line each time you press or release a key, and matching clock pulses on the clock line.

The protocol is described in detail here, which should help to see what's going on.

Where is the best place to connect the logic probe's crocodile clips on an A500?

Starglider 2 · 19 September 2018, 06:41

Hey guys, I have a new keyboard on the way but am on a time crunch... If the new keyboard doesn't fix things, what would you suspect is causing the non functioning topping and stuck caps lock light after boot?

I've tried:
Replacing CIAs
Reflowing solder on most of board
Testing continuity of keyboard to pins 39/40

Thanks.

nogginthenog · 19 September 2018, 19:48

Did you try cleaning the keyboard membranes? Maybe some liquid got in there.

Starglider 2 · 20 September 2018, 00:24

Quote:

Originally Posted by nogginthenog

Did you try cleaning the keyboard membranes? Maybe some liquid got in there.

Thanks, I actually tried a whole new keyboard today. Same problem ??? It's definitely something on the Mobo...

Starglider 2 · 20 September 2018, 05:33

The plot thickens... Or thins...

By today I had 2 keyboards + a third keyboard controller.

I mixed and matched and now have one almost working set.

But, occasionally, the keyboard does some of these things:

1. Stops working mid session and the Caps blinks 3 times repeatedly, which means a timing issue but I'm not sure what chip.

2. Random text starts appearing, usually:

````````

What chips does this point towards?

Thanks!

ajk · 20 September 2018, 07:41

The keyboard signals (clock and data) connect to the CIA at U7, there isn't really much else directly involved. Also I don't think there is a way for the motherboard to purposely reset the keyboard, so if that is indeed what happens it must be something to do with the keyboard itself.

Is an unstable power supply ruled out? Aside from that, I can't really think of any easy things to test. You'd need to hook up an oscilloscope and/or logic analyzer to the keyboard and see what goes on when the errors occur.

Starglider 2 · 09 September 2020, 00:36

Hi guys, just to update this thread in case it helps anyone else, I resolved the A500 keyboard weirdness with a Ray Carlsen PSU. His conclusion was that the Vampire card was drawing too much power; more than the 2 factory PSUs I tried could handle. His one offers more amperage, and the keyboard glitching seemed to have been caused by that and therefore went away. Finally after 2 years of being fearful to use my Amiga because I knew the keyboard would lock up, it's working! ??

Jope · 09 September 2020, 11:05

Good to hear you finally got it sorted. Old PSUs can indeed make life a pain. :-)

12 September 2018, 02:15	#1
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	A500 Keyboard Not Working (except Ctrl+A+A) Hi guys, I have the exact same problem as I've seen elsewhere, but no solution was ever posted to those threads. Here's the symptoms: - After I replaced 2 faulty CIAs with working ones (which fixed all the issues I was having with drives & other weird crashes) the keyboard inexplicably stopped working - Reinstating the faulty CIAs doesn't return the keyboard to working state so something has changed - CAPS flashes only once during boot (normal) - CAPS can be pressed and depressed (lit and unlit) during the boot cycle but once the machine has booted, CAPS stays lit up (if lit when pressed) and can't be released - No keys are responsive - The exception is Ctrl + A + A to reset What would you try next?

16 September 2018, 03:06	#11
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	Hi guys, so the plot thickens... I got a replacement keyboard PCB very fast from eBay. Hooked it up, and exact same issues... On boot sometimes the CAPS lock starts stuck lit up, sometimes not. No keys work. After about 20 seconds, if I press CAPS on and off a few times, it will get stuck on/lit. I've reflowed all the solder around the CIAs, the CPU, and the keyboard connector on the mainboard. I also recapped the board. Obviously the keyboard PCB itself is new, so that rules out the keyboard chips or the connector cable and plug. The CIAs on the A500 are also new (they're the 5th & 6th I tried after other issues.) So if it's not the CIAs, not the solder joints, not the connector, not the plug, not the cable, and not the keyboard PCB and chips, what could it be? The keys themselves somehow? Some other OCS chip? What would you try next? Last edited by Starglider 2; 16 September 2018 at 05:09.

19 September 2018, 06:41	#13
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	Hey guys, I have a new keyboard on the way but am on a time crunch... If the new keyboard doesn't fix things, what would you suspect is causing the non functioning topping and stuck caps lock light after boot? I've tried: Replacing CIAs Reflowing solder on most of board Testing continuity of keyboard to pins 39/40 Thanks. Last edited by Starglider 2; 19 September 2018 at 06:59.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
A500 dead keyboard replaced but still not working properly	Amiga500Forever	support.Hardware	8	12 February 2017 14:12
A500+ Keyboard Mostly Not Working	RobTurbo	support.Hardware	4	01 February 2016 19:07
A500 keyboard, column of keys not working	amigaosx	support.Hardware	1	16 September 2015 15:04
A500 keyboard 2 keys not working	Zippy Zapp	support.Hardware	39	24 December 2014 16:57
Amiga 4000D Ctrl-A-A not working.	Turran	support.Hardware	10	28 January 2013 10:42

12 September 2018, 02:36	#2
r.cade Registered User Join Date: Aug 2006 Location: Augusta, Georgia, USA Posts: 548	Check that the keyboard cable has not been damaged. Other than that, broken CIA socket or bad CIA or bent pin? Also very likely is just keyboard CPU died. I have several like this. Don't know what kills them... static to the CPU chip, plugging cable in wrong... who knows.

12 September 2018, 04:30	#3
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	Thanks. The CIAs are both replacements, unless 4 are faulty. Continuity on the cable passes. I guess that leaves the CPU!

12 September 2018, 06:06	#4
Jope - Join Date: Jul 2003 Location: Helsinki / Finland Age: 43 Posts: 9,861	I doubt it's the CPU. The problem lies either in the keyboard MCU on the keyboard PCB, or the CIA at U7, or in the connection between them. How did you measure the continuity? Just the cable or all the way from the keyboard to the CIA's DAT and CLK pins 39 and 40?

12 September 2018, 07:47	#6
Jope - Join Date: Jul 2003 Location: Helsinki / Finland Age: 43 Posts: 9,861	Do you have another keyboard to try with?

12 September 2018, 08:01	#8
Jope - Join Date: Jul 2003 Location: Helsinki / Finland Age: 43 Posts: 9,861	I don't have that many unfortunately.. If you have a scope or a logic analyzer, you could measure the clock and data signals from the keyboard to see if it's shooting anything when you power it up and hit some keys. The keyboard reset goes via a different wire, so the keyboard is partially or fully working. However if either the clock or data output pin is burned, it will not be able to transmit any keycodes.

12 September 2018, 08:03	#9
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	I have a logic probe. So which pins should I be trying it on specifically when I hit keys?

12 September 2018, 12:28	#10
Daedalus Registered User Join Date: Jun 2009 Location: Dublin, then Glasgow Posts: 6,334	Clock is pin 1, data is pin 2. Also worth checking their continuity to pins 40 and 39 respectively on CIA U7, and check both of those lines for shorts to each other, ground or 5V. You should get a byte on the data line each time you press or release a key, and matching clock pulses on the clock line. The protocol is described in detail here, which should help to see what's going on.

19 September 2018, 19:48	#14
nogginthenog Amigan Join Date: Feb 2012 Location: London Posts: 1,309	Did you try cleaning the keyboard membranes? Maybe some liquid got in there.

20 September 2018, 05:33	#16
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	The plot thickens... Or thins... By today I had 2 keyboards + a third keyboard controller. I mixed and matched and now have one almost working set. But, occasionally, the keyboard does some of these things: 1. Stops working mid session and the Caps blinks 3 times repeatedly, which means a timing issue but I'm not sure what chip. 2. Random text starts appearing, usually: ```````` What chips does this point towards? Thanks!

20 September 2018, 07:41	#17
ajk Registered User Join Date: May 2010 Location: Helsinki, Finland Posts: 1,341	The keyboard signals (clock and data) connect to the CIA at U7, there isn't really much else directly involved. Also I don't think there is a way for the motherboard to purposely reset the keyboard, so if that is indeed what happens it must be something to do with the keyboard itself. Is an unstable power supply ruled out? Aside from that, I can't really think of any easy things to test. You'd need to hook up an oscilloscope and/or logic analyzer to the keyboard and see what goes on when the errors occur.

09 September 2020, 00:36	#18
Starglider 2 Registered User Join Date: Jan 2014 Location: California Posts: 1,146	Hi guys, just to update this thread in case it helps anyone else, I resolved the A500 keyboard weirdness with a Ray Carlsen PSU. His conclusion was that the Vampire card was drawing too much power; more than the 2 factory PSUs I tried could handle. His one offers more amperage, and the keyboard glitching seemed to have been caused by that and therefore went away. Finally after 2 years of being fearful to use my Amiga because I knew the keyboard would lock up, it's working! ??

09 September 2020, 11:05	#19
Jope - Join Date: Jul 2003 Location: Helsinki / Finland Age: 43 Posts: 9,861	Good to hear you finally got it sorted. Old PSUs can indeed make life a pain. :-)

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)