Page 27 of 41

Re: OSI 600D Superboard Build

Posted: 16 Feb 2024 10:13
by rubber_jonnie
stephen_usher wrote: 16 Feb 2024 09:49 Hopefully the new box does the trick. It'll have to have a frame buffer to read the frame from the source, store it and then write it out at a different frequency and number of lines (and potentially colour encoding).
That's what it seems to do, it's one of these:

Screenshot 2024-02-16 at 10.05.03.png

Which was the cheapest one I could find late last night. I don't doubt it will have caveats, such as frame drops etc, but TBH it's not like I'm going to be running FMV on it :) so for a basic text screen it should be fine.

stephen_usher wrote: 16 Feb 2024 09:49 With respect to "composite video", you have to think of it as just the TV signal before it gets modulated for transmission over RF, so it would be whatever the TVs in the region expect for their input. So not only will the number of frames per second change but the number of scan lines and the colour encoding (not an issue here).
I had a good read of the composite Wiki and TBH I'd always seen composite as a 'Standard', not realising it was different in other regions. I suppose that's what you get for assuming. Having said that, given that Klyball has said there are some differences with the UK-101 video circuit, so it may be possible to mod it to be the same as that.

Re: OSI 600D Superboard Build

Posted: 16 Feb 2024 10:25
by rubber_jonnie
stephen_usher wrote: 16 Feb 2024 09:53 By the way, if you have one of these devices it may be able to do the business:


61IWL4KW7CL._AC_SL1000_.jpg
I've got a lot of adapters, Retrotink, OSSC, weird little cheap composite to HDMI adapters that I've gathered over the years, but to date not one of those, though I have seen them on Amazon.

I'll keep that one on my back pocket.

Re: OSI 600D Superboard Build

Posted: 17 Feb 2024 17:34
by rubber_jonnie
Well, good news and bad. The PAL/NTSC converter arrived, but I still don't get anything out of the OSI... Not on my LCD or on my 5.5" TV.

:pullhair: :pullhair: :pullhair:

However I have managed to prove something interesting, which is by connecting the output of the adapter to the LCD I'm using, it shows the LCD does accept an NTSC signal as well as PAL, which is good news, so I'm happy about that.

So I think there is some sort of issue with the video circuit so time to get the schematics and service manual out again to see what's what.

Re: OSI 600D Superboard Build

Posted: 18 Feb 2024 12:38
by rubber_jonnie
So, quick stop and status check.

I have a video signal, it does look like text is displayed on the B&W TV, however it's not usable for whatever reason. Probably because it's NTSC and the TV is PAL. If I feed the B&W TV NTSC from my adapter it is rapid scrolling, a bit like the OSI.

Next up, the Dell LCD display I have does support NTSC & PAL via composite as shown when I hook up the PAL <=> NTSC adapter and flip between the 2 standards:

IMG20240218102916.jpg
IMG20240218102942.jpg

So my adapter isn't strictly necessary, but at least I have it to hand now.

So that suggests an issue with the video circuit itself, so I'll have to dive into the service manual and run through all of the checks for the video circuit. At least now the clocks work and I seem to get address and databus activity, it's probably just likely an issue within the video generation circuit itself.

Re: OSI 600D Superboard Build

Posted: 17 Mar 2024 19:23
by rubber_jonnie
OK, so my successes over the weekend have made me want to get back to this project, so time to dig out the schematics and work out what IC's make up the video circuit to see if I can get a stable signal.

I know that there is a video signal that changes based on what I do with reset, but by the same token, it doesn't look to have a back porch.

I know that somebody else who built one had it work just fine with an LCD display, so I thin k it should work, but I also know that I can see what looks like text on the micro CRT I have, so even if it isn't running properly, I should see something if the video circuit works, and I just don't.

I love a walk through schematics, don't you :)

Re: OSI 600D Superboard Build

Posted: 17 Mar 2024 20:09
by stephen_usher
Yeah... if they're decent ones without massive errors. (Not thinking about Bomarc Mac schematics at all, honest! ;-) )

Re: OSI 600D Superboard Build

Posted: 17 Mar 2024 20:24
by rubber_jonnie
stephen_usher wrote: 17 Mar 2024 20:09 Yeah... if they're decent ones without massive errors. (Not thinking about Bomarc Mac schematics at all, honest! ;-) )
Having used the Bomarc Mac Classic schematics, I know exactly where you're coming from...

Thankfully there is a good set in the service manual, the only problem being that they are split across multiple pages. Being in colour though they do actually specifically identify what IC's do what, including the video circuit, so fingers crossed.

Re: OSI 600D Superboard Build

Posted: 17 Mar 2024 20:32
by stephen_usher
By the way, this board which inserts a backporch into a ZX81's video outut may be useful:

https://github.com/redhawk668/ZX81-Composite

Re: OSI 600D Superboard Build

Posted: 17 Mar 2024 20:57
by rubber_jonnie
stephen_usher wrote: 17 Mar 2024 20:32 By the way, this board which inserts a backporch into a ZX81's video outut may be useful:

https://github.com/redhawk668/ZX81-Composite
I saw that on 'More fun making it' on YT the other day, may be worth a try.

Re: OSI 600D Superboard Build

Posted: 19 Mar 2024 10:03
by rubber_jonnie
OK, so I have ID'd the ICs that form the video circuit, so since there aren't many of them I think the first port of call will be to check all have power and GND and that there are no shorts between pins.

Since there are a relatively small number of ICs and they are all 74LS IC's I'll pull them and test them in my TL866 too, and if all are good, then it'll time to break out the scope.

Thankfully somebody posted some scope pics of the video signal from a working machine, and it has a front & back porch, which I'm not sure mine has, so I at least have something to work on.

Given I can see pixels in the fast rolling screen I have a suspicion that the HSYNC is off somewhere, so hopefully the schematics will help ID where that comes from.