Yet Another Acorn Atom Repair

emulators, hardware and classic software for atom + system machines
User avatar
CommanderCoder
Posts: 81
Joined: Wed Nov 06, 2019 5:50 pm
Location: Royal Leamington Spa
Contact:

Re: Yet Another Acorn Atom Repair

Post by CommanderCoder » Fri Sep 11, 2020 12:47 pm

Looked okay but quite fuzzy but I can't tell now. Sadly my 60 year old military-grade oscilloscope has just broken. The beam is stuck in the top right.

I think it is time to get a logic analyzer. One of these is cheap and should help with fixing the atom. Or with a bit more browsing .. https://www.amazon.co.uk/AZDelivery-Log ... B082MF1Z1C


I'll solder up some of the dodgy connections on the board while I wait, and do more continuity testing.

User avatar
1024MAK
Posts: 10229
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Yet Another Acorn Atom Repair

Post by 1024MAK » Fri Sep 11, 2020 1:32 pm

If you can afford a 16 channel logic analyser, that’s far more useful than a eight channel type.
Often you need to monitor eight data or address lines plus various control signals, or eight data lines and at least six address lines plus two control lines...

Mark

User avatar
1024MAK
Posts: 10229
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Yet Another Acorn Atom Repair

Post by 1024MAK » Fri Sep 11, 2020 1:37 pm

CommanderCoder wrote:
Fri Sep 11, 2020 12:47 pm
Looked okay but quite fuzzy but I can't tell now. Sadly my 60 year old military-grade oscilloscope has just broken. The beam is stuck in the top right.
That could be something relatively simple, like a loss of the power rail for the horizontal and vertical deflection system. Do you have documentation for it, or are you able to find technical details on-line?

Mark

User avatar
oss003
Posts: 3325
Joined: Tue Jul 14, 2009 12:57 pm
Location: Netherlands
Contact:

Re: Yet Another Acorn Atom Repair

Post by oss003 » Fri Sep 11, 2020 3:04 pm

Did you try pressing BREAK and then CTRL-G?
If the processor is running, sound should be heard.

Greetings
Kees

User avatar
CommanderCoder
Posts: 81
Joined: Wed Nov 06, 2019 5:50 pm
Location: Royal Leamington Spa
Contact:

Re: Yet Another Acorn Atom Repair

Post by CommanderCoder » Fri Sep 11, 2020 4:58 pm

There isn't a loudspeaker attached, but I'll attach one and give it a go!

User avatar
CommanderCoder
Posts: 81
Joined: Wed Nov 06, 2019 5:50 pm
Location: Royal Leamington Spa
Contact:

Re: Yet Another Acorn Atom Repair

Post by CommanderCoder » Sun Sep 13, 2020 11:03 am

I've only got 4 hook clips so this is a bit minimal. I think I'll order another analyser since it works well and run them in tandem.. and some more hook clips.

So.. The processor and VDG are working. The A/G line is low (as are GM0,1,2) on the VDG and it is getting a good clock.

The CPU is getting 1Mhz and normally it does nothing on the SYNC / RW lines. If I press BREAK a few times it eventually gets going and this is the output. The three rows are (top to bottom):
D0 - Clock
D1 - RW
D2 - SYNC
IMG_8089.jpg
In this picture, this is just the SYNC after the CPU gets going.
IMG_8087.jpg

I'll try moving the RAM around and see how that helps.

User avatar
CommanderCoder
Posts: 81
Joined: Wed Nov 06, 2019 5:50 pm
Location: Royal Leamington Spa
Contact:

Re: Yet Another Acorn Atom Repair

Post by CommanderCoder » Wed Sep 16, 2020 11:49 pm

Moving the RAM around didn't help so I started the long task of checking every address and data line to the RAM from CPU or VDG. They were all okay.

I then started checking the control lines specifically leading to IC51&IC52, or IC42/43 - these are the minimal RAM chips needed. The connections to IC30 were good for IC42/43. I checked IC23 too to be sure that the ROM was being accessed. The stream of data that the logic analyser picked up suggested it was okay.

Then I started checking connections for IC6, IC7 and IC8 to IC51&IC52. These select the #00xx and #03xx range of RAM. Specifically there is one line between pin 11 of IC8 and pin 8 of IC51 & IC52 that selects these chips.

SUCCESS! There was no connection along this line. So these RAM chips were never getting selected. The connection between pin 8 of both the chips was good, and the connection to pin 12 of IC5 was good. However the connection to pin 11 of IC8 was not good. The tracks on the board look good front and back but definitely no connection.

So I made a temporary modification and switched on.
IMG_8106.jpg
IMG_8105.jpg

The picture is really bad as it is using a composite video cable plugged into the empty modulator and this is wired directly to R11. However, it is definitely a working ATOM from a CPU, RAM, PPIA, and VDG point of view.

Switching to the video pins on the PL4, and fiddling with the wires causes the picture to jump around and occasionally gives a good image but it doesn't last long enough to take a photo.

Next question: any advice on solving my picture problem?

User avatar
trixster
Posts: 1057
Joined: Wed May 06, 2015 12:45 pm
Location: York
Contact:

Re: Yet Another Acorn Atom Repair

Post by trixster » Thu Sep 17, 2020 7:56 am

It lives!

Great work :D

User avatar
roland
Posts: 3997
Joined: Thu Aug 29, 2013 9:29 pm
Location: Born (NL)
Contact:

Re: Yet Another Acorn Atom Repair

Post by roland » Thu Sep 17, 2020 8:02 am

CommanderCoder wrote:
Wed Sep 16, 2020 11:49 pm
The picture is really bad as it is using a composite video cable plugged into the empty modulator and this is wired directly to R11. However, it is definitely a working ATOM from a CPU, RAM, PPIA, and VDG point of view.
Good work =D> Normally the video output is taken from R6 but you did that in the next step.
CommanderCoder wrote:
Wed Sep 16, 2020 11:49 pm
Switching to the video pins on the PL4, and fiddling with the wires causes the picture to jump around and occasionally gives a good image but it doesn't last long enough to take a photo.
Next question: any advice on solving my picture problem?
Since it occasionally gives a good image I suggest to check the solder joints, tracks and components in the video circuit. IIRC there were some issues before in that area so maybe you overlooked one.
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:

User avatar
oss003
Posts: 3325
Joined: Tue Jul 14, 2009 12:57 pm
Location: Netherlands
Contact:

Re: Yet Another Acorn Atom Repair

Post by oss003 » Thu Sep 17, 2020 9:04 am

Nice job ..... the jumping picture could be because the Atom is running at 60Hz ....

Bluring can be created by bad capacitors (C2,3,4).
Also check if the repaired videocircuit with the wires is done ok.

Greetings
Kees

User avatar
CommanderCoder
Posts: 81
Joined: Wed Nov 06, 2019 5:50 pm
Location: Royal Leamington Spa
Contact:

Re: Yet Another Acorn Atom Repair

Post by CommanderCoder » Thu Sep 17, 2020 9:15 pm

I checked all the connections around the video circuit. All looked good. I improved the connection to my video phono so it is now a good connection. The picture came on and was really stable but odd colours.

Then I had a sudden spark of inspiration. My TV has different picture modes for composite video - ah.. the old days of PAL, NTSC, SECAM...

Here are the different outputs. AUTO is not the one I want.. PAL is!
IMG_8120.jpg
IMG_8119.jpg
IMG_8118.jpg
IMG_8117.jpg
IMG_8116.jpg
IMG_8115.jpg
So just some tidying up, making the powersocket more rigid, screwing it all together.

What can I do about getting colour and/or the AtomMMC in it?

User avatar
oss003
Posts: 3325
Joined: Tue Jul 14, 2009 12:57 pm
Location: Netherlands
Contact:

Re: Yet Another Acorn Atom Repair

Post by oss003 » Thu Sep 17, 2020 9:53 pm

There are a few options for expansions .....

http://stardot.org.uk/forums/viewtopic. ... cts#p86259

Another good HDMI colour option is hoglets RPi video converter which only needs a noisekiller board internally.

viewtopic.php?t=18747

viewtopic.php?t=16907

Greetings
Kees

User avatar
CommanderCoder
Posts: 81
Joined: Wed Nov 06, 2019 5:50 pm
Location: Royal Leamington Spa
Contact:

Re: Yet Another Acorn Atom Repair

Post by CommanderCoder » Sat Sep 19, 2020 11:04 am

Had a bit of a panic last night. I put all the chips that roland sent to me back in the Atom, and the video was corrupted again. After some searching it looks like the 81LS95 tri-state buffers are interfering with the data bus. With either of them in IC2 or IC3 in any order, the image is just random characters. Without either of them, the Atom is fine. Is that a problem with the Buffer chips or something with the motherboard? I'll try switching them with those in IC27/IC28

I ran the RAM Memory Faults test from Atomic Theory and Practice (12.3.1, page 92) and the RAM is okay. I ran CLEAR 4 and draw some lines and that looks good too.

I like the idea of HDMI output. I've used the Raspberry Pi for lots of things and a Raspberry PI MMC and HDMI board would be brilliant. I suppose no-one has looked at combining the PiMMC2 board and the 6-bit RGB-to-HDMI adapter have they?

Has anyone tried running the Atom and Pi adapter from a single 5v USB supply? Then all the adapters could be inside the Atom case with a single USB power socket on the back; and USB joystick on the side. Software could be uploaded to the Raspberry Pi (and thus AtoMMC) via WiFi & SSH.

That would be ideal but put the Pi under pressure.

http://stardot.org.uk/forums/viewtopic.php?t=19359
https://github.com/hoglet67/RGBtoHDMI/w ... tart-Guide

User avatar
oss003
Posts: 3325
Joined: Tue Jul 14, 2009 12:57 pm
Location: Netherlands
Contact:

Re: Yet Another Acorn Atom Repair

Post by oss003 » Sat Sep 19, 2020 2:23 pm

CommanderCoder wrote:
Sat Sep 19, 2020 11:04 am
Has anyone tried running the Atom and Pi adapter from a single 5v USB supply? Then all the adapters could be inside the Atom case with a single USB power socket on the back; and USB joystick on the side. Software could be uploaded to the Raspberry Pi (and thus AtoMMC) via WiFi & SSH.
The Pi-adapter gets it's power from the Atom 5V on PL4 so doesn't need a seperate power supply.
A 2 amps or more USB power supply should be ok.

IC's 2,3 and 4 are buffers for the expansion bus PL6/7.
To use PL6/7, you need IC5 for databus dis/enabling.

Greetings
Kees

Post Reply

Return to “acorn atom and acorn system series”