The strange case of the Commodore 64 colour bands.

[stephen_usher]

Yep. So much for "Next working day delivery."

[rubber_jonnie]

Yep. So much for "Next working day delivery."

Bloody liars!!

[BennehBoy]

What motherboard revision is your machine?

[stephen_usher]

250407 rev. 1 I think.

[stephen_usher]

Well, I recovered the new capacitors from work after getting home from Cornwall.

Testing them with the multimeter shows that the meter is only slightly out when measuring in the picoFarad range, about 1pF. This shows that the original capacitor was ~15pF, which actually tallys with the markings. The schematic shows 10pF.

Anyway, I've swapped the caps and now I can't adjust the clock to precisely the correct rate, it's slightly too high. However, the display does seem slightly sharper. Unfortunately it's not helped with the colour banding, so I've ordered an R3 VIC-II chip from Retroleum and we'll see what happens with that.

[exxos]

I don't know if you saw my thread here https://www.exxosforum.co.uk/forum/viewt ... f=44&t=701 granted 2600 but I went to great lengths to solve svideo issues. While I got some improvement , my TV wasn't great either, but ultimately I gave up. But was lucky someone did a RGB kit.. So not sure if there is one for the C64 or not.

I haven't read the whole thread. But assuming those 15pf caps are on the video lines., you can use higher values. But while it reduces noise, you get more blur the higher you go. A lot just seemed to be interference . So tinfoil screening everywhere might help... Even a wire to the osc case to 0v helped on the 2600.

[stephen_usher]

On the C64 the VIC-II chip does everything, including all system clock generation and outputs the luma/chroma signal directly. From the look of things the chaos signal’s clock is varying across the scan line meaning that the PAL colours regularly swing from a magenta tint to a cyan tint and back again.

The capacitor I swapped is the one which resonates with the main clock crystal. I changed it as according to the Tektronix ‘scope the clock signal’s frequency was varying by -1%/+10% which I thought might be a harmonic caused by the oscillation circuit being out of tune. The schematic says that the crystal had to be paired with a 10pF +=5% ceramic capacitor but the original was reading far higher than this. (After using the new caps to calibrate the multimeter’s reading, the old cap was 16pF but is marked as 15pF.)

[rubber_jonnie]

So the caps didn't help, but at least you've been able to secure a new VIC II, fingers crossed it's like the VIC in my VIC-20 and marginal and replacing it will resolve the problem.

Out of curiosity, do you have any freeze spray you could use to cool the VIC-II down? Be interesting to see if it's a thermal issue.

[stephen_usher]

Temperature doesn't seem to make any difference.

[rubber_jonnie]

Temperature doesn't seem to make any difference.

Fair enough.