OSI 600D Superboard Build

[rubber_jonnie]

For certain values of "running".

I think 'running ish'

[rubber_jonnie]

OK, so odd.

I removed R58 and I can see both the 15Khz pulse and what I believe is the video signal at the PCB, but they aren't mixing, which is odd...

[stephen_usher]

Checked the pull-ups R57 and R62(?)?

You need to check the pull-down resistor on the sync line too.

[stephen_usher]

Just out of interest, if you have the video circuit complete but don't have the monitor connected what does the output look like?

I'm wondering if there's a DC offset from the Dell monitor causing the signal to be destroyed as there's no decoupling via a capacitor and/or clamping. It's an extremely poor video circuit design. They could have combined the sync and luminance using an OR gate rather than hoping one signal wouldn't flood the other using just a resistor.

[rubber_jonnie]

Checked the pull-ups R57 and R62(?)?

You need to check the pull-down resistor on the sync line too.

Yes, I've checked that whole group of resistors, all seem to be fine. R57 is for the cassette port, so doesn't play a part here.

With regards to your comments regarding the Dell display, yes it's certainly possible. I would also say that the Dell is pretty good at displaying almost all composite signals that I throw at it, though my spare Apple IIe mainboard and it don't get on so well from a composite point of view. I am going to see if I can get it hooked up to another composite device.

However, if I disconnect from the Dell, I'm still getting just the 15Khz sync signal at the video connection at the back of the board. It's really odd that the video signal from U7 just vanishes when they are mixed.

I'm no electronics designer, but the mixing method does seem rather agricultural!!

I did notice one other weird thing earlier, and I'm not sure if it's my scope or a genuine thing, but sometimes the clock circuit starts up at just 2Mhz and it takes some time to get up to 4Mhz. I can't help but have a nagging feeling that there is still a clock issue.

Ho Hum.

For now I just need to focus on the video circuit, because without that I can't really progress.

[stephen_usher]

R58 is the trim for regulating the relative size of the signals. It's between 0 ohm and 5K ohm. If you have the variable resistor with the wiper (sync) at the zero end then it'll swamp the luminance signal. In the UK101 it just has a fixed 560 ohm resistor instead.

manual-6.jpg

[rubber_jonnie]

R58 is the trim for regulating the relative size of the signals. It's between 0 ohm and 5K ohm. If you have the variable resistor with the wiper (sync) at the zero end then it'll swamp the luminance signal. In the UK101 it just has a fixed 560 ohm resistor instead.



manual-6.jpg

No reason why I can't try a 560 ohm, think I have some.

My pot has the wiper in the centre, which is where I believe it should be.

[chronicthehedgehog]

You might find a couple of Don Lancaster's books interesting:

Cheap Video Cookbook
Son of Cheap Video

Both available on his site. They cover video circuits very much like this of the era.

[rubber_jonnie]

You might find a couple of Don Lancaster's books interesting:

Cheap Video Cookbook
Son of Cheap Video

Both available on his site. They cover video circuits very much like this of the era.

Thanks, I'll take a look.

[rubber_jonnie]

Oh boy, this is driving me nuts!!

I have had 'images' over the course of the day:

PXL_20240127_150301174.jpg

But I then noticed something odd, and that was the clock at the CPU on pin 37 and 39 (39 is an output) was only at 500Khz when this appeared, not 1Mhz as expected. So I checked the clock input on the LS93, and sure enough there was 4Mhz as expected, and in fact it completely matched what the service manual pictured (Image top left on the paper, waveform 1.

OK, over to pin 11 which is directly connected over to pin 37 (Clock in) of the CPU, and sure enough it's showing 500Khz. OK, so why is my clock wonky again!! It's dividing by 8 instead of 4.

I swapped the crystal for a 3.579Mhz one I had in my spares, just to check the result, and sure enough, I got 447Khz. Divided by 8 again. Damn it.

So, at that point I decided to go and play Duke Nukem 1 on my Evercade and come back to it another day.

:dizzy: