GadgetUK164 - Retro Gaming Repairs & Mods Another quality video. Watching now. Still humbled and cannot express my thanks enough. This would of cost me £££ - I'm celebrating by 'eaching' (eating and watching) with some garlic bread and spag-bol 😛😎

You're welcome mate =D LOL @ eaching - I always "each"!

GadgetUK164 - Retro Gaming Repairs & Mods HAHA 😂

Thanks mate =D

Thanks =D Yes, I love those heatsinks - just waiting on some more to arrive! LOL @ 5 C64s! I have a similar crazy number!!

Great news =D

Thanks, much appreciated =D

=D

No worries =D Yes, faulty RAM indicated by the Dead Test could be the multiplexers. There's no way to test for that type of problem in software.

Cheers my friend =D I will do more with the scope, I've got a number of videos to do on some of the basics of electronics and repairs.

Thanks =D You missed a 4 off - Commodore 64 =D

Thanks =D Yes, some red herrings - I hate those lol =D I've still got the faulty chips and will do a follow up video later!

Thanks! =D Yes, I wish I had an EPROM programmer that could test TTL (I've got 5 different EPROM programmers and not one of them tests TTL lol). I was going to create my own TTL tester using a PIC, I've not got around to it yet but I will do at some point! In the meantime I might buy a TL866!

Thanks, no problem! Let me know how you get on with yours!

Thanks =D I always reply where I see a message!

Thanks =D It's annoying it didn't end after the first faulty chip was found :P

=D

Thanks =D

Many thanks =D

Thanks =D Yes, was a bit of a pain, shame it didn't end after the first 7708 was replaced =D

Yes - that's ideal if you have a device that can test TTL etc. I've got 5 EPROM programmers and not one of them will test either DRAM or TTL =/

Yes, I had forgotten that the clock circuit can vary between boards too - that's a very good point! But in general, assuming you dont have clock problems, you can use pretty much any schematics to check connectivity. The IC numbering can confuse you too I guess if you get schematics for the wrong board.

Thanks =D

Thanks =D

Possibly dirty switch or power socket!

I might be misremembering but I recall going back and forth between the output pin of the 7805 voltage regulator showing 4.99V and the input pin on several chips showing 2.2V. Not sure if there could be a failing component somewhere bringing down the voltage or a cracked solder joint somewhere that got better when it heated up. Anyway the video helped a lot and none of the caps looked damaged so not going to repair further that which is no longer broken or it might never work again =)

There are 2 5V rails! The regulator on the C64 motherboard provides 5v ONLY for the VIC and chips next to it. The PSU provides a second 5v rail which feeds most of the other chips on the board. So if you see a low 5v elsewhere on the board (other than the VIC or associated chips in that area), then either your PSU has a low 5v level, or more often than not its the power switch or power socket that is dirty. But you could also have a short somewhere, that would pull the 5v down, but shorts of that magnitude don't generally fix themselves.

🤣

Hehe =D I guessed that!

Thanks =D Just be aware that other things can cause black screen (like the PLA, CPU, CIAs, SID etc). But if its not the PLA or RAM then the 7708s are very possible.

yep! i've had the PLA tested in another machine and it works fine. the SID is non existent and the CIAs have been socketed and removed. am told the RAM was replaced (but unsure) so will try some 257s and see how we go.

Thanks! Hehe, trying to find faulty Amigas is half of the problem lol.

Thanks

I think it would be better to probably just have a dedicated heatsink and remove the top tbh - but... That top joins the side piece etc and acts as a larger heatsink. So I guess it would be worth testing in both scenarios to see which way the VIC is actually cooler. Personally, after seeing how hot my VIC chip gets I prefer a fan tbh. On the C64C it doesn't get so hot though.

Thanks =D Very much appreciated =D I think the cart I used to use at Eteson Electronincs was a homebrew cart - it had that appearance tbh, and I've searched high and low and not found anything that looks like it (either on the board with RAM, on regards how it displayed on screen).

Check the solder points on the caps you've replaced - maybe something is bridged where it shouldnt be. Beyond that it could be that a chip has died from ESD, or something has just come to end of life (maybe PLA or something). Also check your voltages before doing anything else too, make sure 5v and 12v from regulators are correct, check the fuse hasn't blown, also check there's 5v on the main ICs

In this video I used a "Logic Probe" - which indicates a high / low / pulsing / high impedance signal, and I also use an Oscilloscope to measure clocks.

Thx I could understand the word probe in de video but not the word logic

Are you saying you measured between the 5v rail and ground and measure 6 ohms?

If there's no PHI2 from the CPU, I would suspect the CPU tbh! I could be wrong, but I think the VIC2 provides the clock to the CPU, and the PHI2 should be created by the CPU.

Yeah, that's what I've gathered too from reading the schematics. I'll order 40 pin sockets and test the dead(?) CPU in my working machine then :)

Thanks, yes - I do say it correctly these days! That was filmed a while back!

Whoops my bad. I haven’t watched your newer stuff yet as much as the older videos. As a recent subscriber I’d like to get the older things out of the way first and work my way up to the new stuff in case I miss out on some hidden pearls of wisdom.

Yes, and it can make a difference. -20 is 200nS -15 is 150nS -12 120nS etc. But the C64 only needs -20 or faster. So you can mix them without issue.

Sorry about that! I will try to do less swaying and camera movement when up close! =D

There is no need to apologise ;-) Its just your content is so good I want to watch but sometimes when there is too much movement in front of camera the DMS kicks in. I used to think it was just me but its quite common - well.blogs.nytimes.com/2015/11/14/feeling-woozy-it-may-be-cyber-sickness/ for instance I don't get the problem with say Jay's stuff for instance, he had the camera just that little further back.

I am sure they fail a fair bit, but I've got loads of C64s with those chips on that work fine atm! Ultimately I am sure they will die before other makes perhaps though.

Provided you don't use too much heat they are OK to work on! Sinclair Spectrums are much more difficult. If you have little desoldering experience its best to get a cheap desoldering station as it makes chip removal much easier. Something like the ZD-915.

I find that if I cut the chips out when I replace them it makes desoldering easier and you don’t spend forever heating the area trying to get chips out.

It's unlikely heat in the attic would cause any problems tbh. Let us know how you get on =D

I've heard that some PLA replacements (OTP etc) can have problems with the cart port, but I haven't found that with the Atmel 27C512R at all. I would test connectivity between the EXROM and GAME connections that come from the PLA to the cart slot - make sure those are correct. Maybe you have a damaged trace somewhere around the PLA.

Just for the sake of putting it out there, Commodore *is* known on occasion to have used whatever was left in inventory to make some C64's. Just ask Jan Beta, he'll tell you. :) However, I agree with GadgetUK164... in this case, the first place I would start is to check the EXROM and GAME traces for continuity.

Pretty sure I've worked on more C64's over the years tbh! I repaired tonnes of them when I was in the trade. And whilst you do occasionally see one with an odd RAM chip, it's not that common.

I have a fully socketed board build coming up =D Regards testing the SID, you really need to familarise yourself with how a 6581 and 8580 sound with a number of different games. Then always use those games to test. It's easy to spot when there's a missing channel or bad filter etc. But sometimes problems with ADSR stuff can be subtle.

@@GadgetUK164 great, thanks

=D

The power supply itself provides the 5v for the other TTL chips! The regulator only powers the VIC chip and the surrounding chips in the shielded area. Check the 5v on the power connector on the PSU, and check the fuse in the PSU hasn't blown.

I had saw a video, maybe yours, where you measured 5VDC at the 5VDC regulator and then measured 5VDC at the TTL chips.. I assumed that voltage for the TTL's was supplied from that regulator! My mistake. Ah Ha! Just checked..Less than 2 volts coming off of C91 (and the TTL chips). Maybe something is loading it down(?). Good power supply tested in a working unit. Thanks! EDIT------------------------------------------- Just watched the video again and you explained the separate 5 volt sources in the text. I should pay more attention. 3:42 4:10

Feel the tops of the chips - if something is dragging the 5v down to 2v, something will be super hot! My bet is the SID or a CIA.

That's a nice tip! But the main reason I never connect the 1541U2 up whilst fault finding is the risk that a fault could damage the 1541U2. In this case it wasn't likely, but there are some black screen faults where say a data or address bus is stuck high, which can kill other things connected to that bus. I've seen 64C's with a dead CPU, and 2 x dead CIAs - all with the same data bus connection killed. If I connected up a 1541U2 to that, it could damage it! The only time I will use the 1541U2 is when I know its working lol.

Yes, I think some pre-checks are usefull, but I've succesfully used the 1541U2 in 2 black screen repairs until now: In the first machine, all ROMs were socketed, I removed all of them (even char ROM not needed to 1541U2). After removing all of them, 1541U2 menu worked fine and using the altenate KERNAL function, I was able to get empty blue screen. So RAM was fine, no need to worry about that, and one of the ROMs had to be the cause. It turned out that the BASIC ROM was spoilng the bus, replaced it. C64 working fine. For the other machine, the 1541U2 menu did show immedeately that one of the RAM chips was defect. Replaced it, machine was working fine for 2 weeks, then broke again :( I'm sure the RAM is correct, I have socketed all RAM chips and tried many things there, but as the 1541U2 menu displays perfectly the RAM is unlikely the cause now. I'm considering desoldering the CPU, but it'll have to wait until I'm in the mood again to look at it.

Great way to test the ROMS there for sure!!! =D

Yep, that seems to be what Jan found. Spike on the 12v side after power turned off (probably due to capacitors). It did make me wonder if a similar spike was created on the 5v side too though.

Investigation is ongoing. I replaced the 12V Recom one with a regular 7812 for now. I'm using the C64 a lot and haven't found any more issues yet. At least nothing you can tell from the outside except the whole machine is obviously running cooler. I'll watch for spikes on the 5V rail when I open it up again.

So I tried an rf cable directly in to the aerial socket on my tv, tuned it in and to my surprise all works normally. Either there is something wrong with the s-video scart or it just doesn’t play well with the cheap scart to hdmi converter I have with the LGC1. odd as the speccy 128 works perfect with it!

Haha =D It's dust in the air - I really need to get an air purifier in that room I think!

Did you try swapping the 7708 chips (like in this video)?

@@GadgetUK164 That is next on my list, but fortunately they are not MOS brand and look to be Fairchild's. While I do solder and desolder frequently on modern circuits, I get leary of these 1980's circuits with the sort of tin roll coated traces...damn traces lift if you even look at them. Someone has reworked this board in a couple places and burnt the solder mask off but traces seem to have continuity. The IC's you refer to are memory multiplexers? So would that cause a data line to be held either high or low? I don't have a logic probe but that sounds like a smart thing to test

I think address lines go through those don't they? If you've got a stuck databit, it could be something else entirely, like a bad CIA for instance, although i wouldn't expect it to boot the diagnostic ROM cart if you had a stuck databit all the time.

The cart I've seen was homebrew I think - I am sure that it was created by a friend of the store owner. It had RAM in there, and the test screen was completely different to any of the ones I've seen.

Plenty of diag carts around, World of Jani lists a few, you can download them and fire them up in VICE to see how they work: blog.worldofjani.com/?p=164

Thanks! I've not found it yet =/

Putting the wrong sized fuse in shouldn't be a problem - unless something fails and draws too much current, but even then it would be rare for that failed component to kill anything else. That said, I have seen where the CPU fails, pulling the 5v to ~3v and the failed databus then kills the CIA chips too due to the CPU databus failure.

@@GadgetUK164 I replaced two ram chips that were getting very hot. Still won't boot. I'm now beginning to think it's either the PLA or the VIC, but I can't test that cause this is the only c64 I have and after searching on eBay it doesn't seem to make any sense to buy expensive replacement chips if I don't even know if that will get it working again... PS: I probably caused a short circuit while I had the case open with the wrong fuse in now that I think about it