C64 repair, help needed from experienced C64 enthusiasts

[RazSlack]

Alright, so I have an old C64 I got from my school after we found it in an old store room, at the moment it doesn't really wish to work, it produces a pattern similar to this:
http://personalpages.tds.net/~rcarlsen/cbm/misc/pict-o-guide/C64%20garbage%20screen.jpg
or just simply a black screen, depending on how it feels when I start it up. Now I've read around about this, and someone said it's because the caps are old, and there's too much ripple which messes up the logic circuitry.

Well, just for the heck of it, I added some caps for the filters, but still the same. Then after some more reading, I found that the VIC or memory can be faulty, or the PLA as well, or maybe the kernal rom.

The key word here is -maybe-, I don't really want to replace every single chip on the board, so I decided I'd check if the solder joints have gotten loose, I resoldered the aforementioned IC pins, and resocketed every chip that had a socket, but to no avail. Still the same screen.

Then I thought, hell I have a scope! Let's do some probing! Well I found something weird on the clock signals, shown on the pictures. One of them shows the clock signal on pin 1 of the MPU relative to GND (the shielding):

http://oi62.tinypic.com/30iexza.jpg

The second picture shows both clocks, on pin 1 and on pin 39 of the MPU:

http://i58.tinypic.com/2hrpeuo.jpg

Now I don't have much knowledge of how these old 8-bit multi-phase MPU's work, but I don't think that the clock on pin 39 (it's Phase 0 on the schematics) is healthy. Anyone have some knowledge about this, or perhaps could tell me what -should- be the signal there by taking out a working C64 and doing a probing on that pin, so I could check what's the issue?


Also according to the schematics, Phase 0 is generated by the VIC chip, so if that clock's the culprit, I'll know it's because of the VIC.

EDIT:
Ignore the DC offset, it's because I forgot to turn the scope to DC coupling.

[uwezi]

The whole video signal is generated by the VIC, so as long as you can see characters on the screen I would assume that the VIC is ok. But then yes, in the C64 the two-phase CPU clock was generated by the VIC, this is why US and European C64 were running at different clock frequencies, because the color burst frequencies for NTSC and PAL are different.

Let me hook up my C64 to a scope and I can check the clock signals...

[mos6502]

Sounds like a dead PLA. That's also the most common failure. Could also be bad RAM.

Do you have a cartridge? If so, what happens when you power it with a cartridge inserted?

You could also hack a cartridge and put a dead test EPROM in it.

The best and quickest way to fix a C64 is really to have a working one, and then test the chips of the dead one by putting them in the working machine one by one. But, never put working chips into a non working board.

[uwezi]

Well, it seems that the phi_2 signal is supposed to look as crappy as it does on your C64.

This is how the two signals look in my (working) C64:


And this is the video signal, but since you - at least sometimes - get an image on the screen, it should be ok:


Be careful not to run the C64 for too long without heatsink on the CPU and VIC. I remember that an overheated VIC was a common cause of death at those times.

Since the PROMs were masked rather than OTPROMs I guess you can exclude those from your search - I would probably suspect a power supply or RAM problem.

Do you know this site: http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/

[RazSlack]

Yeah, I've checked that site, also thanks for the pictures, I'm now assuming it's not the VIC, neither is the VCO, since I checked that as well.

Sadly I don't have another C64, only this one, neither any cartridges, but maybe I can get a few from a friend to test. I really hope it's not the PLA that's faulty, because it's a bitch to replace that is nigh impossible to me due to lack of EPROMs or spare working ones, since my friend's busted C64 has full black screen all time, so I guess his one's PLA is faulty too. How could I test if the RAM is working or not? The voltages on them seem to be fine.

[jucole]

Hi,   not sure if you know this site forum but it's a great place to get advice on C64's  [url=http://www.lemon64.com/forum/index.php]http://www.lemon64.com/forum/index.php] [/url]  mind you they tend to refer you first to follow Ray Calson's notes,  which can be found here.  http://personalpages.tds.net/~rcarlsen/cbm/c64/

The VIC runs the show with it's master timing and also refreshes the DRAM whilst the CPU is on phase 2 of it's clock; the VIC will also steal clock cycles from the CPU in order to grab screen data when it needs it.   The PLA is a common failure, but so is the RAM / ROMs and the logic,  so might be best to use a dead-test cart which might give you a good place to start!

But if you do de-solder any IC's be careful and take your time, as the PCB traces lift-up very easily.

[AmmoJammo]

Start by replacing the electrolytic capacitors on the power supply section, and in the external PSU if its one you can open.
It won't fix the issue, but they are 20+ year old capacitors, that were cheap as chips to begin with, and probably barely work at all anymore.
Plus, I believe a bad power supply/capacitors is often considered the cause of failed memory in the C64's 

[jucole]

This is also a cracking read...
[url=http://personalpages.tds.net/~rcarlsen/cbm/misc/diagcart/64-128%20cart/Technical%20manual%20clearscan.pdf]http://personalpages.tds.net/~rcarlsen/cbm/misc/diagcart/64-128%20cart/Technical%20manual%20clearscan.pdf [/url]

[krivx]

Apologies if you've already checked, but what do the supply voltages look like? The original c64 power supplies were notorious for failure, all the potting tend to cook things.

[RazSlack]

Power supply looks fine, the 7805 is getting a bit hot on its plate, but considering it's regulating from 11 volts instead of about 7, I think it's normal behavior.

The 7812 is perfectly cool, I soldered it to the ground plane below it (seems they didn't do it in the factory, lousy bums didn't even put a screw there to at least give it some thermal transfer).

I find it weird though that the regulators get so much input voltage reserve, the 7812 is regulating from 20 volts as well. Think this might be the cause for their tendency to fail?

Also something interesting to see, the ceramic caps have some kind of white gunk on the top of them:

http://oi62.tinypic.com/osgwh4.jpg

Best seen on the cap to the left of C27.

[SeanB]

Normal for those wax dipped ceramics to have with age. It is the wax that has come out of the ceramic covering. Not a worry.

[uwezi]

This is also a cracking read...
[url=http://personalpages.tds.net/~rcarlsen/cbm/misc/diagcart/64-128%20cart/Technical%20manual%20clearscan.pdf]http://personalpages.tds.net/~rcarlsen/cbm/misc/diagcart/64-128%20cart/Technical%20manual%20clearscan.pdf [/url]

especially this text seems interesting:
http://personalpages.tds.net/~rcarlsen/cbm/c64/one%20case%20history.txt