The strange case of the Commodore 64 colour bands.

Any 8bit repairs, images, cool things, upgrades etc.
Post Reply
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

Well, the pristine C64 I picked up a couple of weeks ago does actually have a problem, the display output.

It has a very rare (it seems) issue where the chroma signal has jailbars, alternating between a red tint and a bluish tint across the screen. Also, just last night I found that it has issues if you have a dark screen and then change to a light screen. For about 5 seconds the light screen is dim and then it suddenly lightens.

You can see the issue in this video:



My first guess is the two electrolytic capacitors in the VIC-2 tin as they are part of the video clock circuitry, and they're almost 40 years old.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
rubber_jonnie
Site Admin
Site Admin
Posts: 10462
Joined: Thu Aug 17, 2017 7:40 pm
Location: Essex
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by rubber_jonnie »

Definitely worth a try changing the caps. Must check my 64C.
Collector of many retro things!
800XL and 65XE both with Ultimate1MB,VBXL/XE & PokeyMax, SIDE3, SDrive Max, 2x 1010 cassette, 2x 1050 one with Happy mod, 3x 2600 Jr, 7800 and Lynx II
Approx 20 STs, including a 520 STM, 520 STFMs, 3x Mega ST, MSTE & 2x 32 Mhz boosted STEs
Plus the rest, totalling around 50 machines including a QL, 3x BBC Model B, Electron, Spectrums, ZX81 etc...
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

No, the caps are fine and they seem to be there just for power smoothing.

At one point, after removing, Electrolubing and refitting the VIC-2 I thought I'd fried it as nothing was working, but it was due to the very poor single-wipe socket. Phew!

Anyway, popping off the top of the modulator helped. Firstly, as per Adrian Black's sharpness mod video, I removed the capacitor on the luminance line and that sharpened the display massively, but it didn't change the colour banding.

So, as I test I thought I'd put a composite video cable onto the connections to the modulator. Composite video gave the normal results. I then connected the luminance only. This should give a black and white image and sometimes it did, but more often than not I got colour! Not only that but the coloured banding was even more prominent than in the composite signal. I couldn't take a picture as I needed two hands just to get it to work.

So, I think the issue is somewhere in the modulator. This is a pain as getting the modulator out will be difficult. But first I need to take a look at the modulator schematic.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

It looks like the design merely has a resistor between the composite output and the luma, so that's why the colour information is bleeding over.

Looking at the pure chroma signal in the same way shows the bands, so that's where I'm going to have to concentrate.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

I wonder if the 74LS629N is iffy. The colour clock adjustment voltage seems to make no difference to the clock frequency, though this could be the Hantek 'scope as it's not good at frequency measurement. It's bouncing between 7.75MHz and 8MHz but the trace looks stable. I might get the Tektronix down for this.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

OK, got the Tektronix down and the colour clock is now set perfectly. The clock circuits are fine.

It looks like it's either the VIC-2 itself or the circuitry in the modulator.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
rubber_jonnie
Site Admin
Site Admin
Posts: 10462
Joined: Thu Aug 17, 2017 7:40 pm
Location: Essex
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by rubber_jonnie »

Interesting stuff, be good if you can find a fix for it, I'm sure you're not the only person suffering this.

Could be the VIC-II for sure too, they do die, or like the VIC in my VIC-20, just stop behaving properly.
Collector of many retro things!
800XL and 65XE both with Ultimate1MB,VBXL/XE & PokeyMax, SIDE3, SDrive Max, 2x 1010 cassette, 2x 1050 one with Happy mod, 3x 2600 Jr, 7800 and Lynx II
Approx 20 STs, including a 520 STM, 520 STFMs, 3x Mega ST, MSTE & 2x 32 Mhz boosted STEs
Plus the rest, totalling around 50 machines including a QL, 3x BBC Model B, Electron, Spectrums, ZX81 etc...
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

After more investigation with the Tektronix oscilloscope, which I'm sort of getting the hang of despite the illogical menu system, I've discovered that the 17.7344MHz crystal is not that stable. Although it's averaging roughly that value it's actually swinging between 17.4MHz and 18.3MHz. That variation would definitely cause the colour tint to change markedly as the frequency of the PAL chroma signal shifted frequency.

Now, to try to find a replacement. I can see some on Mouser for about 51p +VAT each, but that's Mouser in the USA and the cost of buying just one thing would be high. I'll hae to make a list of things to buy. I know they sell dual-port SRAMs which would be useful for my Apricot hard disk project to store the fake WD1010A registers, and they're not cheap, so that might be a good item to put into the basket.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
rubber_jonnie
Site Admin
Site Admin
Posts: 10462
Joined: Thu Aug 17, 2017 7:40 pm
Location: Essex
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by rubber_jonnie »

stephen_usher wrote: Sat Apr 02, 2022 10:17 pm After more investigation with the Tektronix oscilloscope, which I'm sort of getting the hang of despite the illogical menu system, I've discovered that the 17.7344MHz crystal is not that stable. Although it's averaging roughly that value it's actually swinging between 17.4MHz and 18.3MHz. That variation would definitely cause the colour tint to change markedly as the frequency of the PAL chroma signal shifted frequency.

Now, to try to find a replacement. I can see some on Mouser for about 51p +VAT each, but that's Mouser in the USA and the cost of buying just one thing would be high. I'll hae to make a list of things to buy. I know they sell dual-port SRAMs which would be useful for my Apricot hard disk project to store the fake WD1010A registers, and they're not cheap, so that might be a good item to put into the basket.
Do you have a signal generator that you could use to replace the crystal as a temporary test? What about the caps in the oscillator circuit? C70 perhaps?
Collector of many retro things!
800XL and 65XE both with Ultimate1MB,VBXL/XE & PokeyMax, SIDE3, SDrive Max, 2x 1010 cassette, 2x 1050 one with Happy mod, 3x 2600 Jr, 7800 and Lynx II
Approx 20 STs, including a 520 STM, 520 STFMs, 3x Mega ST, MSTE & 2x 32 Mhz boosted STEs
Plus the rest, totalling around 50 machines including a QL, 3x BBC Model B, Electron, Spectrums, ZX81 etc...
User avatar
stephen_usher
Posts: 5578
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: The strange case of the Commodore 64 colour bands.

Post by stephen_usher »

From what I understand, if the capacitance is changed slightly you just get a slightly different resonant frequency rather than one which varies, but I s’pose you could get further resonance at a lower frequency too. I haven’t got any 10pF caps at the moment. However, I could desolder it and test it in my multimeter, which has a capacitor tester. Having said that, Commodore probably sourced the cheapest possible crystals they could. :-)
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
Post Reply

Return to “8 BIT CORNER”