In a previous episode, I showed this poor Commodore VIC-20 that was bought at a flea market and covered in dirt and oil. I made it my mission to restore this computer back to something that somebody could be proud to own. I worked for days to clean up the case and the keyboard, and now it’s time to see how it all worked out. I retrieved the top case from the hydrogen peroxide solution. And it worked. You can see a difference between the top and bottom pieces, keeping in mind that the VIC-20 always has a beige tint to it, so it wasn’t massively discolored to begin with.
I’ll treat the bottom piece next. While that was going on, I spent some time working on this horrible RF shield. My angle grinder died recently, so I’m just going to use the wire brush on my bench grinder. One interesting thing here is that the camera is recording at 30 frames per second and I guess it syncs almost perfectly with the movement of my grinder, so it looks like it is moving really slow in the video, but it’s actually spinning really fast! Anyway, I had some luck removing some of the corrosion in some areas. But most of the really dark stuff was just way too deep into the metal.
It is at least smoother now and doesn’t’ leave any residue on my fingers. But some areas were so thin that trying to remove the corrosion just ended up breaking off what was left. I also figured it was time to re-install the keyboard mechanism. After all, if we’re going to test the motherboard it will be helpful to have a keyboard. I’m not entirely sure what purpose the clear tape serves covering these wires, but since it came that way from the factory, I decided to go ahead and put a new piece in its place. I’ll also start re-attaching some of the keys, but I want to put some lithium grease on this metal rod for the space bar.
The space bar is the only key on this particular keyboard that requires any extra work to install. All of the rest of the keys, even the larger ones, just pop straight down like this. OK, now that those are all done, let me show you this trick. So you may notice that the T key it still missing, but I have taken the plunger and key from this spot here. What is that key? Well, it’s the up-arrow key. This is not a cursor key, and it is rarely used on the VIC-20. The only real purpose it serves is when making math formulas, this key and symbol is used as an exponent to raise a number to a power, so to speak. Also when combined with shift, It produces the greek letter PI for use in math as well. Not something you need in every day use, though. Plus, it sort of resembles the T character anyway.
And so I’ll just stick this right down in there. OK, Now we’ve got to deal with another sort of problem. OK, so this needs to go in here. And since this is in such bad shape that I’m tempted to just throw it out. The trouble is, this is actually, it’s not necessary for the computer to operate, it’s supposed to be just an RF shield.
But, the motherboard actually mounts to that and the holes, the screw holes are up here and there’s nothing to actually hold the front of the board on. I could go ahead and put this back in here, as ugly as it is, but then there’s the problem of this. This is supposed to be a cardboard separator, that would keep the pins on the bottom of the board from shorting out with that RF shield. Well, the trouble is, this, there’s no way to fix it. This is just disgusting and it’s covered in oil. So it’s just got to go in the trash. So, I’ve come up with a sort of solution. I’m just going to use this vinyl tape and essentially cover this entire thing. That should prevent any accidental short circuits and will also cover up that nasty corrosion, and still give me the ability to mount the board to the case properly. Of course, there are supposed to be these vent holes to help with air flow, so I’ll just use a razor knife and carefully cut out these spots. And yeah, I think that’s going to work out fine.
I also want to replace this sticker on the bottom now. But one thing I find annoying is there is this big hump here, probably left over from injection molding. I’m tempted to grind this down so the sticker will lay flush. But at the same time, this is how it came from the factory, so maybe I should just stick with that. For this task I’m going to use this thin double-sided tape, and I’ll just cut it to the exact size of the recessed area using a razor knife. And there we go. I love that MADE IN USA label, you don’t see that anymore. I also took this opportunity to put new feet on the bottom. These are some I had bought at the hardware store recently and they happened to fit several different types of computers that I’ve worked on recently. Sure, they aren’t the same color, but the are a perfect fit otherwise and since they are on the bottom, I’m not going to complain about the color.
OK, so now comes the point I’ve been secretly dreading. Now that it is fully dried off, I need to test this board and see if it works. This computer didn’t come with a power supply, so I’m going to use my NuBrick, which is a modern power supply replacement for Commodore machines. The only thing I need at the moment is power and composite video. That should tell me if the board is working or not. And here goes. OK, well, that’s sort of good and bad news. Obviously the board doesn’t work, but the fact that there is video and the border and background colors are set correctly tells us that a lot of important components are actually working. So hopefully this won’t be too hard to fix. Thankfully most of the important chips are socketed from the factory. And the first order of business should always be to re-seat these chips and make sure they are making good contact. I also discovered the chips were still covered with oil on the bottom where the brake cleaner missed.
So I’ll need to clean these up individually. And back in the board they go. OK, so I have now cleaned and reseated all of the chips, and I’m leaving these two out completely as they are not necessary to get a screen to appear. However, since they are required to use the keyboard, disk drives, and things like that, they’ll eventually need to go back in. OK, so let’s try this again. And there’s no change. Some of the possible suspects I’m going to have to start looking for are, a bad chip of some kind, a cracked, corroded, or otherwise broken trace.
That can include a bad solder joint. Or it could be a bad component like a resistor, diode, or capacitor. The first thing I should do is test all of these socketed chips. These include everything from ROM, the video chip, and the main processor. Now the easiest way for me to do that is by using this other spare VIC-20 I have. I have two VIC-20s, one is the pristine boxed version I usually show in videos, and then this is the one I’ll actually use for myself since it is in less perfect condition. Anyway, this one has the same revision motherboard so everything is identical, so I’m just going to pull these chips one at a time and try them in the other machine and see if anything changes. However, after checking all of the chips, I found that none of these were the problem. I also tried these chips in the good VIC-20 to make sure they were all good, and they are.
So, next I want to start checking traces. All right, so what I’m going to do now is, I have printed out two data sheets. This is the microprocessor, and this is one of the RAM chips. And so this is the microprocessor, and the RAM chip is over here. What I want to do is check the address and data lines between the CPU and the other side of the board to see if there is any break. I’m going to start with data line zero, which is pin 33. All right, so if I want to check data line zero, I can do it here on these ROM chips. That one works. That one works. However, when I come over to this side of the board, well, it doesn’t work.
And none of them do. But, it’s actually not necessarily a problem. One of the things I realized is that there’s a bus transceiver right here, possibly two more here, so that’s going to kind of attenuate my signal here. So that being the case, I’m going to have to check this area here independently. So what is the purpose of these bus transceivers? Well, the CPU and the video chip both share access to the RAM chips. And so the purpose of these is to literally cut off the CPU’s access to the RAM during the times that the video chip needs to access it. And I was not able to find any bad traces on the board. Next I’m going to move to the logic probe.
Now this probe needs to be powered by the board itself. And a good place to get power on the Commodore boards is from the 5 volt and ground on the cassette port like this. All right, so what’s going to happen is I’m going to touch this tiny tip to the different pins on this chip. Now, if it flashes high, that means there is 5 volts at that location. If it flashes low, then that means that location is grounded. Now, if this light here flashes, that means its constantly flip flopping back and forth, which is good because that means there is activity on that. Some things should be always high, always low, and some things should be flashing. So, that will give us an idea of what’s going on. OK, so this first chip up here, this is the actual power to the chip, and it’s showing to be high and that’s all it is which is good because that’s exactly what it’s supposed to be.
So, now we’re going to check some of these other ones. See, this one’s flashing. That means that there’s activity on that line. This one’s flashing. So on and so forth. So, basically, what I’m going to have to do is check all of these and check if there is activity where there is supposed to be or see if something Is grounded that’s not supposed to be, et cetera. OK, since I have ruled out all of these chips being bad, that leaves possibly any one of these chips. This is both good and bad news. The good news is most of these chips are just logic chips or RAM chips and they are either still made today or they are at least still available as old stock. The bad news is, they are all soldered in place. And since I don’t have very good desoldering tools, it would be best if I could test them in place. But it would certainly help to narrow the list down. So I checked Ray Carlson’s VIC-20 troubleshooting guide, and looked for any scenarios that match the symptoms we’re getting on this machine.
I was able to narrow it down to these RAM chips, which I don’t really have any way to test, and these 3 bus transceivers, this hex inverter, and this 555 timer up here. The 555 timer is probably the easiest to test. It’s job is to hold the reset line low for 4 seconds. So to test that, I can just check the reset line on the CPU and see if it does what it is supposed to. And it does. So we can eliminate that one from our list of things to check. Next we’ll look at this hex inverter. So how this works is there are actually a total of 6 inverters on this single chip.
So, just to make this easier to understand, we’ll just look at one of the inverters, which is pin 1 and 2. Basically, whatever signal goes in pin one, will be the opposite on pin 2. So if it is high, then it gets turned to low or vice-versa. Notice that I’ve removed the CPU and the video chip so that there should be no activity on this board at the moment. So what I’ll do is check pin one and I see that it is low, and so pin two should be high and it is. So I’ll move down to the next one, it is high, so it’s counterpart should be low, and it is. And so on. So, I verified that the inverter seems to be working correctly. So that’s another one I can mark off my list. Next I’ll take a look at these bus transceivers. The way these work is that the signals from the CPU are essentially blocked. But when you pull pin 19 low, it should allow signals to pass through. And since I’ve removed the CPU there won’t be any activity, so I will have to make some of my own.
OK, so what I’ve done is I’ve attacked a ground wire to this chip, pin 19 to make everything go low. This unit here creates pulses. You can see the little light flashing that it’s creating a pulse. And this of course here is the logic probe and it’s light will flash when it sees that pulse. Let’s see what happens. Ok, You can clearly see the pulses traveling through the chip. So now we’re going to move over to the next line. Still working. And I was able to verify that all 3 bus transceivers were working. And so realistically, the only major culprits left were the RAM chips here.
And I don’t have any way to test these in the board with the equipment I have. So I took the board over to my friend Raymond’s shop and he used his magical tools to desolder the first RAM chip in no time at all. Of course, as you might expect, anytime you remove a chip like this from a board you might as well replace it with a socket. So that’s what we did. ——NEW STUFF And with a new RAM chip in place, as you can see the board booted right up! It was a glorious sight to see after hours of troubleshooting. Since there was no keyboard attached, I thought I’d try out a game on it. Omega Race was the first VIC-20 cartridge game I ever played, so I felt it was fitting here. And while I had no keyboard, I could at least attach a joystick. And here we go! I love that logo! It’s very simple, but it’s so elegant! Its interesting to think this may actually be the first game that’s ever been played on this computer.
OK, before I put this thing back together, there’s one other problem I want to solve. I noticed that this jumper wire here which powers the user port is apparently slightly broken so I need to solder that back on. In order to do this right, I needed to unsolder the good wire, then clean all of the solder out of the holes. And fortunately there was just enough slack to trim back the wires a little bit to get some clean wire exposed. And there we go, that was probably the easiest fix of the whole project. Next I’ll re-attach this RF shield. I mainly needed the shield as a mounting point and it does have several screws that hold the board to it, but I thought I’d at least solder a few of these parts back so the shield is properly grounded. And now I can finally install the board back into the case, And before we go any further, I thought I would power it up in the case and make sure nothing was grounding or shorting out. And everything looks good. Now I’ll attach the keyboard and try it out. And it works.
I can type on it just fine and all of the keys are working, except the one. And while doing weird things like this is perfectly acceptable when there’s a lack of any replacement keys, I had always planned this to be a temporary solution, I just didn’t know how temporary it would be. When I was walking out of Raymond’s, he pointed out to me some things he was going to throw away and this was one of the things that was there. And notice it’s the same key mechanism that’s in the VIC-20. It’s already missing a couple of keys too, but it does have a T key and so I’m going to go ahead and take that. And here we go. And here’s the new key and plunger and fortunately, they do appear to be compatible with the VIC-20. And so I took the keyboard back apart and put this plunger inside.
And now I can pull off this up arrow key and place it back where it belongs. And then I can install the new T key from the C64. And now I can close this thing up for hopefully the last time. I think it’s time for some Cheese and Onion. Seeing this VIC-20 cleaned up and working again and playing games, brings the same joy to me as seeing rescued animals being cleaned up and sent home to new families for a second chance in life.
And I suspect this VIC-20 has had a really hard life of work in a grimy industrial environment, followed by probably years of being stuffed away, forgotten, in some storage shed or something like that. And now it’s getting a second chance at life and hopefully it will spend the rest of its life being appreciated by someone. .
As found on Youtube