exxos's Atari, Amiga & retro forum

stephen_usher wrote: 21 Dec 2025 11:53 :crossed1:

Indeed...

stephen_usher wrote: 21 Dec 2025 11:33 Oh! Hopefully not one of the riveted in keyboards!

New keyboard and backlight coming tomorrow...

Watched a few replacement vids and it's a case of ripping the old keyboard out and putting in a new one. There are screws around the edge and I expect that will hold things steady. There are also 2 lots of fan screws and mainboard screws holding it together so I expect it'll be ok without the rivets. May be a bit more deck flex than before but as long as it works I'm happy.

Surprisingly easy machine to disassemble, jus a LOT of screws. we'll see how it goes.

Back to the 800 then.

I currently have the borrowed 3.546894Mhz crystal from my 200Jr in the 800. All seems to work OK.

I have ordered parts to make an oscillator out of a 14.187576Mhz crystal but not sure when they'll get here, same for the miniature signal generator. So likely on hold until the weekend at the earliest now.

@olivier.jan Do you by any chance have the schematics for your oscillator circuit? I've been looking at this stuff all afternoon and it seems there are many ways to do it. Thanks.

If it's a simple divide by four then push it through two flip-flops, such as two of them in a 74LS74, e.g.

Pin 2 -> Pin 6
Crystal oscillator -> Pin 3
Pin 5 -> Pin 11
Pin 8 -> Pin 12
Pin 9 -> Clock output

Just found an old schematic from 2020, here is the clock generation I used. This actually divides by 16, if you to divide by 4, connect to QB instead of QC. And don't forget to connect VCC on 163 like I forgot on this schematic :(

stephen_usher wrote: 21 Dec 2025 20:59 If it's a simple divide by four then push it through two flip-flops, such as two of them in a 74LS74, e.g.

Pin 2 -> Pin 6
Crystal oscillator -> Pin 3
Pin 5 -> Pin 11
Pin 8 -> Pin 12
Pin 9 -> Clock output

I was thinking more along the lines of how the crystal needs to be set up to oscillate before feeding into an IC. I'm good at tracing circuits and troubleshooting, just not good at putting circuits together...

Got to go and run some errands shortly, so not much happening today and later I have to fit a new keyboard to my Mac (If it arrives!)

olivier.jan wrote: 21 Dec 2025 21:33 Just found an old schematic from 2020, here is the clock generation I used. This actually divides by 16, if you to divide by 4, connect to QB instead of QC. And don't forget to connect VCC on 163 like I forgot on this schematic :(
Screenshot 2025-12-21 at 22.28.08.png

Thanks for that. You're using an oscillator there (X2) rather than an XTAL right?

Yes, I had a bunch of oscillators available. I guess you just need to add the proper capacitors to use a crystal directly.

olivier.jan wrote: 22 Dec 2025 10:12 Yes, I had a bunch of oscillators available. I guess you just need to add the proper capacitors to use a crystal directly.

Yep, I was thinking that since the crystal came from an XE the schematics might show me the circuit and capacitor values 😁

rubber_jonnie wrote: 22 Dec 2025 09:48 I was thinking more along the lines of how the crystal needs to be set up to oscillate before feeding into an IC. I'm good at tracing circuits and troubleshooting, just not good at putting circuits together...

Got to go and run some errands shortly, so not much happening today and later I have to fit a new keyboard to my Mac (If it arrives!)

Fair enough. The combination of ceramic caps needed for a crystal does seem to be a bit of black magic. I often see a 74LS04 in the circuit to square up the output, but I think a 74LS74 would do that using the flip-flops anyway.