The 8bits side of Sporniket

[sporniket]

So, today I had some time, so I cleaned a little the main board, plugged the scart cable and plugged my usb adapter that I usually use on my STe (see there : viewtopic.php?f=9&t=5939)

2023-08-18--01-power-from-usb-charger.jpg

I switched on, and got the pleasure to get the boot screen

2023-08-18--02-switch-on-ok.jpg

I reconnected the keyboard and did a test again, no problem there too :

2023-08-18--03-keyboard-test-ok.jpg

So it seems I got a working TO8 :)

[sporniket]

Along the conversion to Kicad of the Atari 260/520 ST, I also started working on doing the same for the Thomson MO5...

[sporniket]

This week I got myself 3 more mo5s. In fact I was looking for one version with the EFCIS gate array and other differences with the "v1" I got before, so that I can also remake this version under Kicad. I took a chance on the net for a set of 2 (either v1 or v2, there was no picture of the inside) and then someone on the French forum proposed me his computer. It turns out the 3 are v2 variants, and the 3 have problem to run (not a problem for my project)

(...)

so next time I will install the oscilloscope to watch what happens.

I could finally set up my scope and one of those computer yesterday evening, so that first thing this morning was trying to troubleshoot it.

It was really just looking how each line was pulsing, and I pushed the reset button a lot of times to check that there was some pulse before getting stuck high or low.

I checked the CPU (the clock signals allowing me to setup the display), one of the logic IC that perform the decoding, the ROM and the DRAM.

At one point I noticed that the CS for the eprom was stuck to low, and that the data bus was then stuck to the value $F0.

But at another point, I could get one of those random start where the CPU was looping, and that's when I noticed that A0 was stuck to low.

With the computer switched off, I probed continuity between A0 and GND, and I got a buzz. Resistance measured was 1.4.

I removed each socketed chips (there are 3) turn by turn, but the short was still there. But near the north-west of the board, the resistance was 1.2, meaning that it was more around this part.

Then I went to the cartridge port, near the ROM, and I noticed that GND and A0 were next to each other, I checked inside to see if there was anything stuck (there was nothing), then I looked at the other side of the board, where I noticed a "stain" between GND and A0

2024-01-07--01--suspect.jpg

I got my digital microscope...

2024-01-07--02-zoom-avant.png

With a blade, I gently scrubed the thing that was laying there, blowed it and cleaned with isopropyl alcohol

2024-01-07--03-zoom-apres.png

Reinstalled the motherboard, switched on and ...

2024-01-07--04--fixed.jpg

Yeah !

[sporniket]

My TO8 computer was not used much as I did not address yet the supplying of power.

In the wait of doing something more refined, I decided to assemble some cables, a barrel jack 55/21 receptacle, an extension cord with switch, a DC 5V->12V converter, as well as a blue led.

This allows me to use my 5V power brick

2024-02-18--01-flying-wires-and-tapes.jpg

2024-02-18--02-close-up.jpg

The blue led is positionned so that I won't go blind when the computer is switched on.

2024-02-18--03-with-casing.jpg

[stephen_usher]

Looks like something scraped over that via and pushed some solder over to the other trace and also tore the solder mask.

[rubber_jonnie]

Looks like something scraped over that via and pushed some solder over to the other trace and also tore the solder mask.

I was thinking much the same thing.

@sporniket might be worth putting a splash of solder mask over it just to be sure.

[sporniket]

Thanks for the advice. I Will have to get soldermask then.

[stephen_usher]

Nail polish is good enough.

[sporniket]

Now that I finished remaking the motherboard of the MO5 with Motorola gate-array, I had some time to start looking at one of the not working unit I have.

2025-01-11--01-le-malade.jpg

On inspection, I see that the PLCC socket of the gatearray is broken

2025-01-11--04-inside-broken-socket.jpg

So today I removed the socket. Next time I will put a new one, and see how it goes

2025-01-19--02-after-removing-socket.jpg

Another thing I did was experimenting with the video output on my remade motherboard. It is a known problem that on the MO5, the stock circuit outputs a too high signal for R, G, and B, and thus on LCD screens that does not compensate, pastel colors are just white, and the orange is yellow.

2025-01-19--03-color-stock-on-lcd.jpg

After experimenting a little, I replaced the resistors that make a voltage divisor with the transistor (originally 1k) to a very weak value (22 Ohm). Now I can see some colors.

2025-01-19--04-color-mod-on-lcd.jpg

I may look for other ways to tweak the color rendering later, but for now it's good enough.

[mrbombermillzy]

This reminds me of the colour problems we had with the 'Hi colour' Twin shifter board. :)