Why was a dead VIC-II was causing a clock not generated by the VIC-II to go way off-spec? Because the clock generator in the C64 is a beast: The crystal is the colorburst frequency * 4, and comes out of one half of the 629 VCO, tuned by the trimmer in the can. That is used by the VIC-II for chroma generation. It is also divided by 7(NTSC) or 9(PAL) in the 193 and again in the 74 to get ~1MHz, but that is not used as the system clock. It is only fed into the MC4044 PLL. The dot clock is generated by the other VCO in the 629, and is initially free-running. The VIC-II divides the dot clock by 8 to generate 1MHz phi-0 clock. The phi-0 clock is fed into the MC4044 PLL and compared to the divided color clock. The PLL error is fed to the 629 and adjusts the frequency of the dot clock to ensure it stays locked relative to the color clock. That keeps the luma-chroma phase relationship correct and allows the C64 to generate proper luma/chroma signals, instead of relying on artifact color like the Apple II. This also means if the VIC-II isn't outputting the phi-0 clock, the PLL never locks and the VCO generating the dot clock drifts off to some weird frequency.
Also, I think you've had more than one C64 where those caps leaked. Fascinating as I haven't run into that yet -- but if any caps were going to leak, it would be those in there baking next to the VIC-II. Mmmm yum, hot capacitor juice!
Apparently it's common on the KU boards, but it's probably worth checking on the 326298 and 250407 boards. The previous one that had leaky caps was a '407 with a ceramic VIC.
I too had PCB corrosion damage on my original 1983 Silver Label C64 (KU-14194HB board) in the VIC-II/clock area from leaky electrolytic capacitors. The contained heat from the VIC-II certainly hastened the capacitors failure. This damage needed a few discreet enameled copper wires to replace the damaged copper. My KU-14194HB had the CPU/VIC-II/SID/PLA/ROMS and CIAs in sockets too! I blew up more than a few CIA chips hooking various electronic experiments up to the User Port. My VIC-II is a ceramic 6569R1 dated 0383.
Haha, yeah the CIAs will die just from looking at them the wrong way. I think it took them until the 250469 boards to actually put in some protection for the CIAs, I now have 3 dead ceramic VIC-IIs, and only one working one left
Hi, what you have is a genuine PAL silver label with card KU 14 194 hb. All the signs you mentioned are spot on: labels lifting, controller ports sticking out. I am a collector of these c64 SLs, I have many both European and American, the latter with card 326298 and metal label. From what I understand, the video signal of the KU14194hb cuts out due to those capacitors you took out, but this is just the initial stage, i.e. the RF signal is not transmitted properly, but you can see the screen very well with the 8-pin connection. The capacitors then start to leak and then you have damage like the one you found in your c64 SL. Also I would have to remove those 2 capacitors and replace them to eliminate the root problem. I also put heatsinks on the chips, which I then removed to respect the originality of the machine as much as possible. In any case, for daily use, I use a c64c (all dissipated) and a c64 breadbox with heatsinks and fan which, if you're curious, you can see in my very first video made years ago, when I found that rainbow c64 in a materials recovery center, where they were about to shred it... at the time the collecting craze didn't exist yet and I paid about 2.5 euros for it. I wasn't subscribed to your channel, even though I already left a like, but I'll subscribe now.
I recently bought a pal silver label as well, something interesting is that the SID in my silver label was made in the same week as yours! 49th of 1982, and it sounds exactly the same as in the video, maybe that the early SIDs were made in a different way than later SIDs? Thats my theory anyway, while I also did find the sound a bit weird compared to my other C64's, I did not bother to change/swap the 1982 SID because of the rarity of the machine, and its not my main machine that I'm gonna use everyday, Nice video! I enjoyed watching it!
Congrats. Yeah this won't be my daily driver machine, so I don't plan on changing anything on it. Funny the previous machine I showed on the channel with leaky caps in the VIC can also has a SID from 4982. It also sounds the same, so certainly just a quirk with these old SIDs
After watching your video I checked my ku-14194 board and guess what caps c107 and c108 were leaky ! My board is still working but I‘m going to replace the caps as fast as possible.
Interesting about the capacitors.. i wouldn't have even thought to check that as I've not really experienced this in those older boards but will be something i will try to check out going forward. Thanks!
Nice! A real Silver Label with a KU board and everything! Yeah, that SID sounds just a tiny bit glitchy. Probably not bad enough to warrant a replacement yet IMHO, but definitely something to watch.
Now I just need a 64G 😉 There's certainly something up with that SID but as you say, it's not enough to replace it. Although I did write an M for marginal on the underside, which I'll no doubt forget about
Haha, true. Although I did try out the CoCoSID demo on the CoCo2 and it came close. Of course the CPU can't do anything else while the CoCoSID demo is playing but it's a neat trick
Maybe you could try sun brighting instead. It won't be as efficient, but gentler on the plastic. Would be interesting how bright it will get. Greetings, Doc64!
Good Job. The Sid may be on the way out. It doesn't sound terrible though. I would wait before you replace it. I do want to try something called the X-Sid. Supposedly sounds much closer to the original compared to the SwinSid which I'm currently using. The corrosion would have caused another failure eventually, so it was good to repair it. When I had my Vic-II failure last year I replaced those caps. I figured I might as well future proof it.
Interesting, looks like the X-SID is based on the SwinSID so you may end up with something that only sounds marginally better. Who knows maybe it will sound great, but personally I cannot recommend the ARMSID enough. Yes they're more expensive but you'll never need to worry about looking or buying anything else 👍
Nice work in the repair! On Bwack's page for the interactive BOM, it says, "interactive-bom generated using plugin". What plugin does it require? When I try to view the interactive BOM all I can see is the HTML code. I don't see any way to make the page display it as an interactive webpage.
I've never had an issue with the interactive BOM using Chrome. I don't know what plugins it requires, but perhaps your browser is blocking one of them?
When removing non-SMD ICs I would just cut all the legs close to package then remove each leg. This minimises risk of track pads lifting from the PCB. The ICs are cheap & can be destroyed without any tears shed.
Wish I would not have sold my C64 it was a store demo my dad paid $900 for instead of waiting for them to start shipping. It had the orange VIC 20 F keys. Serial number on it was 235.
Nice work. I wouldn't retrobright it due to the marks and digs in the case. Instead I would paint it after "panel beating" it. My breadbin was like this so I used filler and sandpaper then painted it. I got some Ironlak paint from Officeworks which is good paint but cracks if left out in the heat of the sun. If You paint it mask the labels first. If You try retrobrighting the tops of the keys be extremely careful, I did this and ruined them from marbling which I then had to spend hours polishing it out, fortunately it was on the top only. Probably 3 hours max if rtrobrighting keys but make sure they are covered completely in creme peroxide. Officeworks has better variety of paint colours than Bunnings.
Good to know. I'm not a fan of painting it, but I guess if I fk it up retrobrighting it then I can always paint over it. But yeah, I'll retrobright first, then panel beat it. Learnt my lesson doing it the other way around on a 1541C - the panel beated spots ended up going lighter than the rest. Ended up sanding it down so I now have a smooth finish 1541C haha. And yep, the keys can be risky for sure. The ZIF socketed C64 is a good example of that, the keys are all marbled but I'm not too worried about that one, it's a beater anyway
@@TheRetroChannel if you go the paint method, Adrian and I found a paint that they sell in home Depot (assuming you have one of those) that seems like an original color. I don't like painting either, but i had to do it once on a failed cream retrobright.
No home depot here, but if it comes down to it I'll ask you guys and maybe I can find or import that colour. There's still many things we don't get here, no deoxit, no naval jelly - I know Adrian loves that stuff 😄
I am hesitant about retrobriting the case on mine as well. I might just try leaving it in the Sun for a couple of days first. I have had good results with this method in the past. It’s a lot slower than retrobriting, but much safer.
I have very bad experiences with those Fairchild PLAs, they cause cartridge compatibility issues and also a C64 that I keep with a Fairchild PLA inside has trouble running Internal Karate.
Yeah I have another KU board with slightly older chips but with a rainbow label. I think Commodore just threw the boards in whatever cases they had on hand that day
I don't lknow about the value of putting heatsinks on these chips. As Bill Herd says, the main method of sinking heat from these plastic case micros is via the legs and the solder connection to the traces on the motherboard. I think retro preservationists would be better off installing a fan in the case to stimulate better air flow since these finned heatsinks really don;t really move heat faster than the slowest interface which is plastic, or worse, ceramic which is typically used as a heatshield.
I'm sure Bil knows what he's talking about. But also keep in mind the VIC-II normally includes a heatsink in the lid of its cage, and the 64C and Bil's 128 uses the RF shield as a heatsink for a number of ICs. I think Commodore wouldn't have bothered if they saw no value in it. More surface area either by a direct connection to the mainboard or on the outside of the DIP will help sink heat. Sure a fan would help but then you need to mount and power it., I think the small convection effect is enough to push warm air out the top of the case and draw cool air in through the bottom
Why was a dead VIC-II was causing a clock not generated by the VIC-II to go way off-spec? Because the clock generator in the C64 is a beast:
The crystal is the colorburst frequency * 4, and comes out of one half of the 629 VCO, tuned by the trimmer in the can. That is used by the VIC-II for chroma generation.
It is also divided by 7(NTSC) or 9(PAL) in the 193 and again in the 74 to get ~1MHz, but that is not used as the system clock. It is only fed into the MC4044 PLL.
The dot clock is generated by the other VCO in the 629, and is initially free-running. The VIC-II divides the dot clock by 8 to generate 1MHz phi-0 clock. The phi-0 clock is fed into the MC4044 PLL and compared to the divided color clock. The PLL error is fed to the 629 and adjusts the frequency of the dot clock to ensure it stays locked relative to the color clock. That keeps the luma-chroma phase relationship correct and allows the C64 to generate proper luma/chroma signals, instead of relying on artifact color like the Apple II.
This also means if the VIC-II isn't outputting the phi-0 clock, the PLL never locks and the VCO generating the dot clock drifts off to some weird frequency.
Heh, I heard oddities on the SID tests too but the #8-bit-NODANCING-party sounded fine.
Also, I think you've had more than one C64 where those caps leaked. Fascinating as I haven't run into that yet -- but if any caps were going to leak, it would be those in there baking next to the VIC-II. Mmmm yum, hot capacitor juice!
Apparently it's common on the KU boards, but it's probably worth checking on the 326298 and 250407 boards. The previous one that had leaky caps was a '407 with a ceramic VIC.
I too had PCB corrosion damage on my original 1983 Silver Label C64 (KU-14194HB board) in the VIC-II/clock area from leaky electrolytic capacitors. The contained heat from the VIC-II certainly hastened the capacitors failure.
This damage needed a few discreet enameled copper wires to replace the damaged copper.
My KU-14194HB had the CPU/VIC-II/SID/PLA/ROMS and CIAs in sockets too! I blew up more than a few CIA chips hooking various electronic experiments up to the User Port. My VIC-II is a ceramic 6569R1 dated 0383.
Haha, yeah the CIAs will die just from looking at them the wrong way. I think it took them until the 250469 boards to actually put in some protection for the CIAs, I now have 3 dead ceramic VIC-IIs, and only one working one left
Hi, what you have is a genuine PAL silver label with card KU 14 194 hb. All the signs you mentioned are spot on: labels lifting, controller ports sticking out. I am a collector of these c64 SLs, I have many both European and American, the latter with card 326298 and metal label. From what I understand, the video signal of the KU14194hb cuts out due to those capacitors you took out, but this is just the initial stage, i.e. the RF signal is not transmitted properly, but you can see the screen very well with the 8-pin connection. The capacitors then start to leak and then you have damage like the one you found in your c64 SL. Also I would have to remove those 2 capacitors and replace them to eliminate the root problem.
I also put heatsinks on the chips, which I then removed to respect the originality of the machine as much as possible. In any case, for daily use, I use a c64c (all dissipated) and a c64 breadbox with heatsinks and fan which, if you're curious, you can see in my very first video made years ago, when I found that rainbow c64 in a materials recovery center, where they were about to shred it... at the time the collecting craze didn't exist yet and I paid about 2.5 euros for it. I wasn't subscribed to your channel, even though I already left a like, but I'll subscribe now.
very neat channel mate, im a recent new sub and ive really been glued to your videos. well done and thanks for the quality!
Wow, it's a really old one! Nice that you were able to get it working again and mount heat sinks for future heat dissipation.
I recently bought a pal silver label as well, something interesting is that the SID in my silver label was made in the same week as yours! 49th of 1982, and it sounds exactly the same as in the video, maybe that the early SIDs were made in a different way than later SIDs? Thats my theory anyway, while I also did find the sound a bit weird compared to my other C64's, I did not bother to change/swap the 1982 SID because of the rarity of the machine, and its not my main machine that I'm gonna use everyday, Nice video! I enjoyed watching it!
Congrats. Yeah this won't be my daily driver machine, so I don't plan on changing anything on it. Funny the previous machine I showed on the channel with leaky caps in the VIC can also has a SID from 4982. It also sounds the same, so certainly just a quirk with these old SIDs
My favorite computer of all time. Mine has a bad vic chip also sitting in my basement. Thank You.
I'll have to check the caps on my KU boards, thanks.
After watching your video I checked my ku-14194 board and guess what caps c107 and c108 were leaky ! My board is still working but I‘m going to replace the caps as fast as possible.
Another GREAT video!!! I'm loving all the love you have been giving to Commodore on your channel!!
Interesting about the capacitors.. i wouldn't have even thought to check that as I've not really experienced this in those older boards but will be something i will try to check out going forward. Thanks!
Nice! A real Silver Label with a KU board and everything!
Yeah, that SID sounds just a tiny bit glitchy. Probably not bad enough to warrant a replacement yet IMHO, but definitely something to watch.
Now I just need a 64G 😉
There's certainly something up with that SID but as you say, it's not enough to replace it. Although I did write an M for marginal on the underside, which I'll no doubt forget about
Not even remotely an expert, but was it the noise channel? It sounded off to me...
Nice work!
That C64 sounded great to me! But I'm a TRS-80 CoCo guy, so any noise the SID chip makes sounds fantastic to me.
Haha, true. Although I did try out the CoCoSID demo on the CoCo2 and it came close. Of course the CPU can't do anything else while the CoCoSID demo is playing but it's a neat trick
Maybe you could try sun brighting instead. It won't be as efficient, but gentler on the plastic. Would be interesting how bright it will get.
Greetings, Doc64!
great video!
I REALY LOVE MY C64 ❤❤❤❤❤ FOREVER ❤❤❤❤❤❤❤❤❤❤
Good Job. The Sid may be on the way out. It doesn't sound terrible though. I would wait before you replace it. I do want to try something called the X-Sid. Supposedly sounds much closer to the original compared to the SwinSid which I'm currently using. The corrosion would have caused another failure eventually, so it was good to repair it. When I had my Vic-II failure last year I replaced those caps. I figured I might as well future proof it.
Interesting, looks like the X-SID is based on the SwinSID so you may end up with something that only sounds marginally better. Who knows maybe it will sound great, but personally I cannot recommend the ARMSID enough. Yes they're more expensive but you'll never need to worry about looking or buying anything else 👍
It's so weird listening to Adrians "Dance Party" play soooooo slow.... LOL, yeah i'm in NTSC land.
Nice work in the repair! On Bwack's page for the interactive BOM, it says, "interactive-bom generated using plugin". What plugin does it require? When I try to view the interactive BOM all I can see is the HTML code. I don't see any way to make the page display it as an interactive webpage.
I've never had an issue with the interactive BOM using Chrome. I don't know what plugins it requires, but perhaps your browser is blocking one of them?
When removing non-SMD ICs I would just cut all the legs close to package then remove each leg. This minimises risk of track pads lifting from the PCB. The ICs are cheap & can be destroyed without any tears shed.
Wish I would not have sold my C64 it was a store demo my dad paid $900 for instead of waiting for them to start shipping. It had the orange VIC 20 F keys. Serial number on it was 235.
Nice work. I wouldn't retrobright it due to the marks and digs in the case. Instead I would paint it after "panel beating" it. My breadbin was like this so I used filler and sandpaper then painted it. I got some Ironlak paint from Officeworks which is good paint but cracks if left out in the heat of the sun. If You paint it mask the labels first. If You try retrobrighting the tops of the keys be extremely careful, I did this and ruined them from marbling which I then had to spend hours polishing it out, fortunately it was on the top only. Probably 3 hours max if rtrobrighting keys but make sure they are covered completely in creme peroxide. Officeworks has better variety of paint colours than Bunnings.
Good to know. I'm not a fan of painting it, but I guess if I fk it up retrobrighting it then I can always paint over it. But yeah, I'll retrobright first, then panel beat it. Learnt my lesson doing it the other way around on a 1541C - the panel beated spots ended up going lighter than the rest. Ended up sanding it down so I now have a smooth finish 1541C haha.
And yep, the keys can be risky for sure. The ZIF socketed C64 is a good example of that, the keys are all marbled but I'm not too worried about that one, it's a beater anyway
@@TheRetroChannel if you go the paint method, Adrian and I found a paint that they sell in home Depot (assuming you have one of those) that seems like an original color. I don't like painting either, but i had to do it once on a failed cream retrobright.
No home depot here, but if it comes down to it I'll ask you guys and maybe I can find or import that colour. There's still many things we don't get here, no deoxit, no naval jelly - I know Adrian loves that stuff 😄
What is that program you're using for the schematics? I remember you mentioned in a previous video but forgot to take note.
It's the interactive BoM by bwack. There's a link to his github page in the video description
@@TheRetroChannel Thank you. I heard the mention throughout this video but assumed he plotted out the schematic. This will come in extremely handy.
I am hesitant about retrobriting the case on mine as well. I might just try leaving it in the Sun for a couple of days first. I have had good results with this method in the past. It’s a lot slower than retrobriting, but much safer.
Sunbrighting might be the way to go, maybe I'll hold off until it warms back up down here.
@@TheRetroChannel send it up to me if you like. I’m in North Queensland. One hour of sun up here will do the trick! Lol
I have very bad experiences with those Fairchild PLAs, they cause cartridge compatibility issues and also a C64 that I keep with a Fairchild PLA inside has trouble running Internal Karate.
Interesting, I'll have to try out international karate and some carts to see if it also has issues.
@TheRetroChannel are u from Australia or New Zealand
Hey man, where do you get those nice looking shiney heatsinks from?
A store on AliExpress called "we do heatsink"
I have a PAL C64 with that exact same board, 8pin DIN video, Kernal Rev02, made in West Germany... but the case has rainbow label, not silver
Yeah I have another KU board with slightly older chips but with a rainbow label. I think Commodore just threw the boards in whatever cases they had on hand that day
I have 5 silver labels lol bought them all between 10-20 years ago for peanuts.
Retro bright it and do a video on it
I don't lknow about the value of putting heatsinks on these chips. As Bill Herd says, the main method of sinking heat from these plastic case micros is via the legs and the solder connection to the traces on the motherboard. I think retro preservationists would be better off installing a fan in the case to stimulate better air flow since these finned heatsinks really don;t really move heat faster than the slowest interface which is plastic, or worse, ceramic which is typically used as a heatshield.
I'm sure Bil knows what he's talking about. But also keep in mind the VIC-II normally includes a heatsink in the lid of its cage, and the 64C and Bil's 128 uses the RF shield as a heatsink for a number of ICs. I think Commodore wouldn't have bothered if they saw no value in it.
More surface area either by a direct connection to the mainboard or on the outside of the DIP will help sink heat. Sure a fan would help but then you need to mount and power it., I think the small convection effect is enough to push warm air out the top of the case and draw cool air in through the bottom