Strange problem with MK3 Atom....

emulators, hardware and classic software for atom + system machines
Post Reply
Prime
Posts: 2869
Joined: Mon Jun 01, 2009 12:52 am
Contact:

Strange problem with MK3 Atom....

Post by Prime » Tue Jan 05, 2016 10:34 pm

Hi all,

I'm working on preparing two of Arcadian's Atoms to be upgraded with the new disk interface. The one I'm working on currently is the MK 3 machine. This seems pretty much the same as the MK4 board, the one difference I have seen is that there is no LK7, the outputs of the two regulators where commoned, though that track had been cut on this board as both regulators where fitted.

Anyway this machine has been composite modded (comp video to modulator out), and has been patched for 5V operation. I have also replaced all the 22µf caps with modern ones.

The machine boots to the Acorn Atom prompt, but the keyboard is not functional most of the time. However if I unplug IC3 (the middle 81LS95), then the keyboard works. I have checked the chip in question in one of the video 81LS95 sockets and it appears to be working correctly.

This seems very odd, I guess the socket could be faulty this machine seems to be fitted out with black plastic sockets manufactured by TI. The other thing that is odd is that all the TI chips on the board seem to have some black oxidization on the legs but the non TI chips don't.

I have also checked the /CS on the 8255 which seems to be strobing as you would expect for a periodic keyboard scan. However the PA0..Pa3 (pin 4..1) seem not to be toggling which I would expect them to do as the keyboard rows are scanned, they seem stuck at about 1.6V.

This is weird.

Cheers.

Phill.

User avatar
oss003
Posts: 3326
Joined: Tue Jul 14, 2009 12:57 pm
Location: Netherlands
Contact:

Re: Strange problem with MK3 Atom....

Post by oss003 » Wed Jan 06, 2016 7:37 am

Hi Phill,

the only simularity I notice between IC3 and the 8255 are A0 and A1.
If you remove the floppy interface, does the keyboard work ok then with IC3 installed?

Greetings
Kees

Prime
Posts: 2869
Joined: Mon Jun 01, 2009 12:52 am
Contact:

Re: Strange problem with MK3 Atom....

Post by Prime » Wed Jan 06, 2016 8:30 am

oss003 wrote:Hi Phill,

the only simularity I notice between IC3 and the 8255 are A0 and A1.
If you remove the floppy interface, does the keyboard work ok then with IC3 installed?
No, as the floppy interface is not yet fitted :) It only seems to work with IC3 unplugged.

I'm going to try changing the socket later on to see if that is the problem.

Cheers.

Phill.

PhilYoung
Posts: 205
Joined: Sun Jun 12, 2011 5:55 pm
Contact:

Re: Strange problem with MK3 Atom....

Post by PhilYoung » Wed Jan 06, 2016 12:19 pm

Prime wrote:Hi all,

The machine boots to the Acorn Atom prompt, but the keyboard is not functional most of the time. However if I unplug IC3 (the middle 81LS95), then the keyboard works. I have checked the chip in question in one of the video 81LS95 sockets and it appears to be working correctly.


Phill.
That's exactly the problem I had with my Atom, which is also an Issue 3. With IC2, 3 and 4 fitted the keyboard was mostly not working, but 1 in 10 times after BREAK it would work for a period. Also the screen would occasionally become 'sparkly' (composite video output, UHF modulator not used). The original white sockets are still fitted in that area.

IC4 (8304) is back in and didn't seem to be causing the problem. Both or either 81LS95 fitted in either position does reintroduce the problem. So I've just left them out for the moment. I should probably retry after replacing the sockets with turned pin ones but never got round to it.

Removing these ICs was someones suggested solution to the keyboard problem, so they presumably had the same symptoms. I don't recall who that was though,

Phil Young

User avatar
oss003
Posts: 3326
Joined: Tue Jul 14, 2009 12:57 pm
Location: Netherlands
Contact:

Re: Strange problem with MK3 Atom....

Post by oss003 » Wed Jan 06, 2016 12:41 pm

If the floppy controller has buffered lines, you can insert a socket with shortcuts for the address- and datalines in IC2,3 and 4.

Greetings
Kees

User avatar
hoglet
Posts: 9385
Joined: Sat Oct 13, 2012 7:21 pm
Location: Bristol
Contact:

Re: Strange problem with MK3 Atom....

Post by hoglet » Wed Jan 06, 2016 12:44 pm

PhilYoung wrote: Removing these ICs was someones suggested solution to the keyboard problem, so they presumably had the same symptoms. I don't recall who that was though,
It was Daniel:
http://www.stardot.org.uk/forums/viewto ... 768#p51617
http://www.stardot.org.uk/forums/viewto ... 873#p71873

Dave

User avatar
danielj
Posts: 8338
Joined: Thu Oct 02, 2008 5:51 pm
Location: Manchester
Contact:

Re: Strange problem with MK3 Atom....

Post by danielj » Wed Jan 06, 2016 1:02 pm

I was just searching for that thread! - Also, the black tarnish, I had that on loads of ICs in my atom... They were all fine once polished up.

d.

User avatar
1024MAK
Posts: 10230
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Strange problem with MK3 Atom....

Post by 1024MAK » Wed Jan 06, 2016 4:05 pm

danielj wrote:Also, the black tarnish, I had that on loads of ICs in my atom... They were all fine once polished up.
The black tarnish affects various old chips. It is due to the metal used reacting over time in the same way that anything made of silver tarnishes. Chips that are soldered, or which are tightly fitted in quality sockets (and have not been disturbed) can be left and the tarnish will not affect their operation.
However, if the chip is removed and replaced in a socket, the tarnish does need cleaning from the part of the pin that makes electrical contact in the socket.

Mark

Prime
Posts: 2869
Joined: Mon Jun 01, 2009 12:52 am
Contact:

Re: Strange problem with MK3 Atom....

Post by Prime » Wed Jan 06, 2016 4:59 pm

Bingo!

I've worked out what the problem with the keyboard was : The reset circuit!

This machine didn't have C11 fitted so on Kees' advice in another thread I fitted a 22µf cap there, but didn't notice that R38 was 4.7K and not 1K, so what I think was happening is that the CPU was getting reset correctly but the 8255 was not.

I've also managed to solve another problem that this machine was exhibiting which was that the display was garbled, by garbled I mean that it was smaller and repeated in two columns of 3 copies of the display. That turned out to be a bad IC9 (or it's socket, I replaced both). Interestingly enough the 6847 was interpreting the low amplitude of it's clock signal as being 3x too fast, I checked the /HS and /FS and was getting approx 48MHz for /HS and 180Hz for /FS. I wonder if something like this is causing lurch666's problem in the other thread.

Cheers.

Phill.

User avatar
CMcDougall
Posts: 7049
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland
Contact:

Re: Strange problem with MK3 Atom....

Post by CMcDougall » Wed Jan 06, 2016 4:59 pm

^yip, mine had that effect too. Fine sandpaper sorted that out, but don't over do it, or exposed copper on pins...
ImageImageImage

Post Reply

Return to “acorn atom and acorn system series”