News:

Printed Amstrad Addict magazine announced, check it out here!

Main Menu
avatar_Rabs

Black Screen 664

Started by Rabs, 21:23, 20 December 22

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

SerErris

Yeh that looks like random stuff going on on your AY or PIO. 

I assume keyboard is still disconnected, right?

So that would mean that it is just a ghost effect from the AY to PIO communication. 

This can be:
1. Pullups on AY are mad. All the inputs for the AY on Port A should be pulled up (meanining nothing is pressed) and the AY should always put out FF. That is because all the input pins to Port A are not connected to anything, so nothing will be able to pull them down to 0 volt. 
2. However the AY anyhow looks suspekt outputting strange stuff on the one line of its Data bus. So it also could be that it is not the Pullups or Inputs wrong, but actually something wrong with the data output. The data output should not look like that and the uptime of the signal is pretty short. 

I do not find any two sources to describe the correct connection between the PIO, the keyboard connector and the AY. All of them have numbers on them, that actually do not make any sense, or cannot be read.

I have a copy of the 664 service manual, that I would assume is in itself correct, but actually it is impossible that it is correct.


It connects the matrix between pin1 and pin 2 of the keyboard connector (that is how I read it), but according to the schematics of the same manual of the PIN 1 and PIN 2 of the connector are both connected to the PIO, which does not make any sense.

PIO is output and and AY is input - so two connected lines cannot sit on the same chip.

Anyone has a single schematic and keyboard matrix for the 664 that actually makes any sense?
Proud owner of 2 Schneider CPC 464, 1 Schneider CPC 6128, GT65 and lots of books
Still learning all the details on how things work.

SerErris

Maybe someone can untangle that for me.

I used this service manual:
https://www.cpcwiki.eu/imgs/1/1f/Service.manual.cpc664.ctm644.gt65.pdf

And I have been looking at those two pictures, scratching my head:
You cannot view this attachment. You cannot view this attachment.

So lets try to apply some logic:
Pin1 of the matrix applys only to a single key (DEL key and that actually connects on the other side also to a single key. 

So Pin 1 is obviously an Y row key and it should not get in conflict with anything else on the keyboard. So it would be best option to assume that pin1 from the Matrix diagram actually is Y10 on the Schematic. logically then the upper row of the matrix corresponds to the top most (Y10 and following down). 

Pin2 on the other hand needs to be connected to the AY. One can see that there are only 8 inputs on the AY but 9 lines from the matrix. Howe they did it? They connected two lines from the matrix to the same input line (IO7). So the only reasonable way to not interfere anything here is to have this Pin2 connected to the keyboard with another line, but you can still distinguish the keys because you have different input lines to the matrix. So Del key actually sits on the same sending line as the joystick. And there is no other key from the keyboard connected to this sending line. So that my assumption would be (cannot validate), that actually Pin 2 and Pin3 from the matrix (X8,X9 in the schematic) do share both IO7 on the AY.  ...

Then we could assume that Pin4 of the matrix is Y7 and that goes on mirrored .. (left of the matrix is right in the schematic). 

But unfortunately I do think you would need to measure this. 

I still believe that the AY is the culprit for your keyboard issues - so if you do have another one - just check if that goes away. 
Proud owner of 2 Schneider CPC 464, 1 Schneider CPC 6128, GT65 and lots of books
Still learning all the details on how things work.

Rabs

Quote from: SerErris on 14:40, 14 January 23Maybe someone can untangle that for me.

I used this service manual:
https://www.cpcwiki.eu/imgs/1/1f/Service.manual.cpc664.ctm644.gt65.pdf

And I have been looking at those two pictures, scratching my head:
You cannot view this attachment. You cannot view this attachment.

So lets try to apply some logic:
Pin1 of the matrix applys only to a single key (DEL key and that actually connects on the other side also to a single key.

So Pin 1 is obviously an Y row key and it should not get in conflict with anything else on the keyboard. So it would be best option to assume that pin1 from the Matrix diagram actually is Y10 on the Schematic. logically then the upper row of the matrix corresponds to the top most (Y10 and following down).

Pin2 on the other hand needs to be connected to the AY. One can see that there are only 8 inputs on the AY but 9 lines from the matrix. Howe they did it? They connected two lines from the matrix to the same input line (IO7). So the only reasonable way to not interfere anything here is to have this Pin2 connected to the keyboard with another line, but you can still distinguish the keys because you have different input lines to the matrix. So Del key actually sits on the same sending line as the joystick. And there is no other key from the keyboard connected to this sending line. So that my assumption would be (cannot validate), that actually Pin 2 and Pin3 from the matrix (X8,X9 in the schematic) do share both IO7 on the AY.  ...

Then we could assume that Pin4 of the matrix is Y7 and that goes on mirrored .. (left of the matrix is right in the schematic).

But unfortunately I do think you would need to measure this.

I still believe that the AY is the culprit for your keyboard issues - so if you do have another one - just check if that goes away.

Thanks, I will have a read and physically check out the connections. 

I don't have a spare AY at the moment but as a last resort I can take one out of my spare 6128. Wonder if there is a small test board out there for the AY?

Thanks  again for the help, very much appreciated. Will post again when I have a replacement AY.

Animalgril987

Hi @SerErris.
From the Wiki:
 https://www.cpcwiki.eu/index.php/Programming:Keyboard_scanning#Hardware_scancode_table
 it can be confirmed that matrix pin 1 is indeed Y10 ( and also matrix pin 13 is Y7 ). I'm sure that more detective work will uncover how the rest work out.
And, effectively, the first joystick is also on matrix pin 1/ Y10.

WacKEDmaN

#54
Quote from: SerErris on 14:40, 14 January 23Maybe someone can untangle that for me.

I used this service manual:
https://www.cpcwiki.eu/imgs/1/1f/Service.manual.cpc664.ctm644.gt65.pdf

And I have been looking at those two pictures, scratching my head:


So lets try to apply some logic:
Pin1 of the matrix applys only to a single key (DEL key and that actually connects on the other side also to a single key.

So Pin 1 is obviously an Y row key and it should not get in conflict with anything else on the keyboard. So it would be best option to assume that pin1 from the Matrix diagram actually is Y10 on the Schematic. logically then the upper row of the matrix corresponds to the top most (Y10 and following down).

Pin2 on the other hand needs to be connected to the AY. One can see that there are only 8 inputs on the AY but 9 lines from the matrix. Howe they did it? They connected two lines from the matrix to the same input line (IO7). So the only reasonable way to not interfere anything here is to have this Pin2 connected to the keyboard with another line, but you can still distinguish the keys because you have different input lines to the matrix. So Del key actually sits on the same sending line as the joystick. And there is no other key from the keyboard connected to this sending line. So that my assumption would be (cannot validate), that actually Pin 2 and Pin3 from the matrix (X8,X9 in the schematic) do share both IO7 on the AY.  ...

Then we could assume that Pin4 of the matrix is Y7 and that goes on mirrored .. (left of the matrix is right in the schematic).

But unfortunately I do think you would need to measure this.

I still believe that the AY is the culprit for your keyboard issues - so if you do have another one - just check if that goes away.

it is a bit of a head scratcher... 
the numbers on the matrix are keyboard connector pins....they are also listed on the schematic in the middle...
eg pin 1 on the matrix is pin 3 on the 145 or Y3.... pin 4 on the matrix is pin 6 (not 4!) on the 145 and Y6...
X8 and X9 share the same line on the AY (pin7)and they are pins 19 and 16 on the matrix (not pin 2 and 3!)

ive circled the matrix pin numbers in the attached image...(also the arrows on the joystick and X# are pointing the wrong way!)

TotO

Quote from: WacKEDmaN on 07:55, 17 January 23also the arrows on the joystick and X# are pointing the wrong way!
The black arrows show the direction of the signal while the white arrows show where it connects.
"You make one mistake in your life and the internet will never let you live it down" (Keith Goodyer)

Rabs

New AY and now working.   :) next the disk drive.

You cannot view this attachment.

TotO

Hum... I'm sure that will follow a Syntax Error... :-\
"You make one mistake in your life and the internet will never let you live it down" (Keith Goodyer)

WacKEDmaN

Quote from: TotO on 12:37, 17 January 23
Quote from: WacKEDmaN on 07:55, 17 January 23also the arrows on the joystick and X# are pointing the wrong way!
The black arrows show the direction of the signal while the white arrows show where it connects.
ahh thanks Tot0... i knew there had to be a reason for it! :P

Rabs

Quote from: TotO on 15:53, 17 January 23Hum... I'm sure that will follow a Syntax Error... :-\
Hopefully otherwise I've got more problems  :)

Rabs

Still interested to see how the AY failed. It just looked like the io inputs to the ÀY were floating and I did think about adding external pull-ups but these should be in the AY.

But for now I have a working keyboard and sound. Needs more testing though.

So far;
1. AMSDOS ROM dead
2. OS BASIC ROM dead
3. ÇRTC dead
4. AY bad
5. Number of broken tracks
6. Broken keyboard membrane 
7. Key contact high resistance issues

Just can't get over how good the key caps looked compared to the PCB and ICs.

WacKEDmaN

Quote from: Rabs on 19:42, 17 January 23So far;
1. AMSDOS ROM dead
2. OS BASIC ROM dead
3. ÇRTC dead
4. AY bad
5. Number of broken tracks
6. Broken keyboard membrane
7. Key contact high resistance issues

Just can't get over how good the key caps looked compared to the PCB and ICs.

thats alot of issues!... id still be at it! :P
really good job getting it back up and running...

Bryce

Quote from: Rabs on 19:42, 17 January 23Still interested to see how the AY failed. It just looked like the io inputs to the ÀY were floating and I did think about adding external pull-ups but these should be in the AY.

But for now I have a working keyboard and sound. Needs more testing though.

So far;
1. AMSDOS ROM dead
2. OS BASIC ROM dead
3. ÇRTC dead
4. AY bad
5. Number of broken tracks
6. Broken keyboard membrane
7. Key contact high resistance issues

Just can't get over how good the key caps looked compared to the PCB and ICs.


The AY is known to be fragile and it's directly connected to external pins (the joystick port). So a simple static discharge could have caused the chip to fail.

Bryce.

Rabs

OK, almost done. Disk is working, keyboard and joystick are working and tape is loading, BUT the remote to the tape is not working.

You cannot view this attachment.

So last bit, probably another broken track but I can't quite work out the tape circuit from the service manual. Is the relay not involved somewhere?

Rabs

Oh and my the way, I don't hear the relay click when loading a tape. To load I disconnect the remote cable.

Rabs

This one is learning experience. It is almost like it knows, like it has some sense to stay dead.

So all working, tape remote fixed, another broken track. Even ran a long soak test.

But Games not loading completely so swapped with my other 664 to check the tape and all OK. So back to this one to try again....and...grey screen....   :picard:

You cannot view this attachment.

Can't believe all the chips failed suddenly but don't think I did anything to cause a failure like this.

Will come back to it tomorrow...maybe it will fix itself   :laugh:

WacKEDmaN

gray screen and no memory shown could be the Mux ICs... 74LS153 ...theres 4 of them...finding the bad one is a PITA!

Rabs

Quote from: WacKEDmaN on 07:22, 27 January 23gray screen and no memory shown could be the Mux ICs... 74LS153 ...theres 4 of them...finding the bad one is a PITA!
Thanks. I suspect it is another track. I did originally fix one track between the CRTC and the multiplexers. I guess, that the tracks are very fragile and continue to deteriorate. On the plus side I have pretty much now learnt how every part of the board operates  :laugh:

eto

or you have created a short somewhere.

I just had exactly that when playing around with a new extension and I accidentally created a short - gray screen, all RAMS seemed to be dead. Then I found and fixed the short and all was fine again. 

Rabs

Quote from: eto on 12:14, 27 January 23or you have created a short somewhere.

I just had exactly that when playing around with a new extension and I accidentally created a short - gray screen, all RAMS seemed to be dead. Then I found and fixed the short and all was fine again.
All I did was power it back on! Where do I start? My assumptions are that the main data and address bus are fine because the Diagnostic ROM is running. So any problem lies with either the RAM, the Multiplexers (74LS153) or the the two data buffers (74337 and 74244). I know the pcb is problematic (so I could be chasing ghosts on the tracks) and I have already replaced the CRTC and AY and the two ROMs (so it has had a hard life and something else may have failed).

Think I need to scope the RAM data in, out and address lines and see if anything turns up...anything...

Good job I have nothing else to do   :laugh:

eto

Quote from: Rabs on 15:44, 27 January 23Think I need to scope the RAM data in, out and address lines and see if anything turns up...anything...
Also check CAS and RAMDIS. In my case, I managed to shorten CAS.

Rabs

I have traced RAS, CAS, WE and RAMDIS and they all look to have activity. I have also traced all Data outputs and again I can see activity. All Address lines also look to have activity.

But A11 from IC104 looks a bit odd to me, like there is another signal superimposed on it and none of the other address lines look like this.

You cannot view this attachment.

and when I trace this with A10, again from IC104, it looks to align.

You cannot view this attachment.

I have loads of the 74153 ICs so I am tempted to just replace IC104.

Rabs

It's possessed  :o

You cannot view this attachment.

I desoldered IC104, tested the chip and it is OK but replaced anyway. No difference   :picard:

Is it me or are the PCB holes not quite wide enough for a standard socket?

Anyway, I decided to look more closely at ÌC112 and RAMDIS. 

All of a sudden it started working. I did nothing   :o

SerErris

Maybe a correded pin on a chip or a cold solder joint ... You pressed on the culprit, now it works ... until it breaks again.
Proud owner of 2 Schneider CPC 464, 1 Schneider CPC 6128, GT65 and lots of books
Still learning all the details on how things work.

Rabs

Quote from: SerErris on 02:34, 28 January 23Maybe a correded pin on a chip or a cold solder joint ... You pressed on the culprit, now it works ... until it breaks again.
Thanks. I noticed that, if I waggled the power connector on the PCB, I could get the Diagnostics ROM to crash. Reflowed the joints and solid now. Hopefully this is it.

Powered by SMFPacks Menu Editor Mod