Page 11 of 16
Re: The strange case of the Commodore 64 colour bands.
Posted: 09 May 2022 18:46
by stephen_usher
Looking at the circuit, it's more likely to be the 74LS629 as that's the main clock generator chip.
P.S. bought one from RS
Re: The strange case of the Commodore 64 colour bands.
Posted: 09 May 2022 20:11
by stephen_usher
Modulator re-fitted and the SN74LS629N put into a socket. Tested and working. The display looks as it did before without the ragged vertical edges.
Re: The strange case of the Commodore 64 colour bands.
Posted: 09 May 2022 21:39
by rubber_jonnie
stephen_usher wrote: 09 May 2022 18:34
The beads are pristine. I'll have a look at the circuit to see which chips to replace.
I didn't really think the ferrite beads would be a problem, but worth a look.
Probably worth changing the 74LS193 too while you're at it.
Re: The strange case of the Commodore 64 colour bands.
Posted: 09 May 2022 22:54
by stephen_usher
The 74LS193 is part of the clock divider circuit for the dot clock (and hence the main system clock) and has no influence over the chroma output, but does take the chroma reference clock as an input. Given what I'm seeing is a variation in the chroma reference clock, tinting the colours due to the PAL colour encoding frequency shifting, it's best to start there. Once the SN74LS629N has been replaced that's all of the core clock done. Given that RS don't sell the 74LS193 I couldn't get it at the same time. Farnell do have it but the handling charge makes it rather uneconomic to get it from there.
(The other components around there are purely DC voltage components and the DC voltages are good and very clean with tiny amounts of noise, even in the MHz range, which is where I need to worry for this.)
Anyway, we'll see what happens when I replace the VCO chip.
All the ferrite cores are pristine, as can be seen here:
IMG_2659.jpg
Re: The strange case of the Commodore 64 colour bands.
Posted: 10 May 2022 09:49
by rubber_jonnie
stephen_usher wrote: 09 May 2022 22:54
The 74LS193 is part of the clock divider circuit for the dot clock (and hence the main system clock) and has no influence over the chroma output, but does take the chroma reference clock as an input. Given what I'm seeing is a variation in the chroma reference clock, tinting the colours due to the PAL colour encoding frequency shifting, it's best to start there. Once the SN74LS629N has been replaced that's all of the core clock done. Given that RS don't sell the 74LS193 I couldn't get it at the same time. Farnell do have it but the handling charge makes it rather uneconomic to get it from there.
(The other components around there are purely DC voltage components and the DC voltages are good and very clean with tiny amounts of noise, even in the MHz range, which is where I need to worry for this.)
Anyway, we'll see what happens when I replace the VCO chip.
All the ferrite cores are pristine, as can be seen here:
IMG_2659.jpg
Yeah, I think you're right, get the 629 done and review after that.
Everything looks pristine, be nice to get that banding fixed, it seems to be all that's left spoiling the machine.
Re: The strange case of the Commodore 64 colour bands.
Posted: 10 May 2022 10:47
by stephen_usher
Yeah, other than the banding it's working perfectly.
Re: The strange case of the Commodore 64 colour bands.
Posted: 10 May 2022 11:15
by rubber_jonnie
I might be wrong, but C66, the left leg might have a cracked joint. Probably nothing, maybe just a trick of the light bit it seems to have a line around the bottom edge.
Not that I expect it to fix anything.
Did you say you'd swapped the xtal out?
Re: The strange case of the Commodore 64 colour bands.
Posted: 10 May 2022 12:17
by stephen_usher
I can try reflowing C66, but I think it is a trick of the light.
Yes, I've changed the crystal and the resonant capacitor, C70.
C107 and C108 are power line smoothing caps and the power is nice, smooth and correct, so there's literally nothing else which can affect the Colour clock once the 629 is replaced.
I suppose that it could be a phase issue between the colour signal and the pixels/dot clock. If this still doesn't work I can try going through the clock divider circuit.
Re: The strange case of the Commodore 64 colour bands.
Posted: 10 May 2022 13:12
by rubber_jonnie
stephen_usher wrote: 10 May 2022 12:17
I can try reflowing C66, but I think it is a trick of the light.
Yes, I've changed the crystal and the resonant capacitor, C70.
C107 and C108 are power line smoothing caps and the power is nice, smooth and correct, so there's literally nothing else which can affect the Colour clock once the 629 is replaced.
I suppose that it could be a phase issue between the colour signal and the pixels/dot clock. If this still doesn't work I can try going through the clock divider circuit.
Bit frustrating though when nothing seems to work! It does seem like something is out of phase though.
Fingers crossed that 629 will arrive relatively quickly.
Re: The strange case of the Commodore 64 colour bands.
Posted: 10 May 2022 13:28
by stephen_usher
It should be here before 2pm. :-)