 I have some new plastic bags inside which are capacitors. What I'm going to do today is to recap my old Amiga power supply, which is this repulsive piece of yellowed plastic. Because it is old and repulsive and I've already had to repair something inside. Really I should get a new power supply but I can't be bothered. So I am just going to replace all the capacitors in it, which should hopefully make it a little less. Well you'll see when I open it. So this is probably going to be a rather boring workbench video, but I'm recording it anyway just in case something interestingly hilarious happens. This is for an Amiga 600 as I got a while back. It's shown up in some videos. I actually have a full capacitor replacement kit for that, which I'm going to be doing at some point, but there's a lot of surface mount stuff that I'm a bit nervous about, so I'm taking my time preparing there. But this is boring simple through-hole stuff with big thick capacitors and you may notice that the board is somewhat warped and has some heat marks on it. I have discharged this by the way, so it is safe. So what we've got is... I should actually unbag my capacitors. I've been through this and should hopefully have all the right ones. This is a 400 volt 47 microfarad that is replacing this one. You can see that the new one is significantly smaller. These are 4,716 volts and these are replacing these ones. 4,716, just double checking everything. These are from a different manufacturer, 22 microfarads 25 volts. These should be replacing these little ones here, which are very difficult to... 22 microfarad 25 volts, so that's good. And then we've got this one and this one, which is a 4,716 volts and another 4,716 volts, which should be these. And there's actually three in here because I'm using one for something else. That might be the A600 actually. So, let us start removing them. Soldering iron is hot, so this should be relatively straightforward, even though desoldering is one of my least favourite workbench activities. Doing that quite badly. There we go. And the negative side with the black stripes is marked, but the type is not, so let us write that down. Is this pen writing on this thing? Not very well, but that is readable just 385. Okay, so let's try and do this one next. Lovely through-hole stuff, so easy. Negative marks in the right place, 4,716. I have to remove the fluff though. Interestingly, this is also a capacitor, 6.8 nanofarad ceramic job. As far as I can tell, that does not need to be replaced. Okay, so now these two grounding plane means that you can actually get the thing hot, which is nice. Is that way round? Yep. That is a 22. And I'm always checking to make sure the capacitor is actually the same way round that the board says it should be, because otherwise, well, sometimes they print the board markings incorrectly. And sometimes they put the capacitors on backwards. But as this is a power supply, I would have expected it to explode if that was going to happen that way round. This is also a 22. You can see that this board has actually seen rather better days and all the corners have broken off. I really don't know what's happened to this, why it's in such a bad state. Maybe it's seen a lot of heavy use. Okay, come on, someone doesn't want to move. Okay, and that one, negative down 4,700. I've already replaced one resistor more or less on spec. And when it was behaving oddly and that seemed to have made it better. But it's not in particularly great shape. I don't normally touch mains voltage stuff due to knowing my limits, but a simple recap should be okay. It's also 4,700. And this one, heating up one leg at a time and slowly rocking it out. That way around, yes, and this is a 470. Last one, two more to go. There's also a 4,700 and the last one. This might be a rather quick job, which would be nice. This is almost certainly soldered with lead-free solder, which is a higher melting point than leaded solder. Okay, and that is also a 4,700. All right, so that would be all the caps. So let us now apply some fresh solder to the solder joints and then clean them. This will make it much easier to work with. These joints are one of the other capacitors. This is the mains input side. You can see from the big, thick tracks that this is the high voltage side of things. This board actually has decent separation between the high voltage and the low voltage side. Was that one there? No, that's actually a missing component. These two pads here, you see there's a little square there, but nothing in it. Resistor maybe? This is high voltage. This is low voltage with much thinner tracks. This is, what is this? Yeah, this is the electronics behind the switch mode power supply. There's a chip here that spans the high and low sides. So what this is doing is it's measuring the voltage of the low side in order to control the transistor output from the high side. There should be, if I understand things correctly, a capacitor that spans high and low. But I don't see one. Maybe it doesn't work like that. Anyway, this is the last. Okay. So now we get the solder sucker out and we take all that solder off again, hopefully freeing up the holes. This will allow us to get the new capacitors in. I haven't cleaned these ones. It seems a waste to put fresh solder on just to take it off again, but it does make a big difference. See, this is leaving nice clean solder holes. And the new solder is noticeably shinier. So a single pin there. Ah, there's one side of, seems to be an inductor. Inductors have always been a bit of a black art to me. I don't really know what they do. I just don't know why they do it. Or why one would want them to do it. Good evidence this thing is easy to work on. It's not having a ground plane that makes a difference. This is a sheet of copper down the middle of PCBs that's used to, well, for grounding and to prevent RF interference. And it sucks the heat out when you try to solder. And this one doesn't have one, which means that I can heat up a pad very quickly. Okay, well, that seems to be all the joints clean. So let's start placing some. Let me see. This is a 70. That goes here. This is another 470 that goes here. I've bend the legs slightly so they don't fall out. These are the 4700s. Okay, the leg spacing is slightly different, but it will work. We also have to be careful not to accidentally place one backwards. So I'm double checking as I go. This one, this one. Finally, this big one. Oh, hang on. There's the two little ones I haven't done. I'm just thinking these two little ones. Just thinking the best order in which to do them. Normally small first. That is in 2225. So here that goes here. Yeah, let's do those, I think. They are sufficiently solid. They don't need support. So I can just solder those straight on. The others will all need a bit of support. Those are adequate joints. Okay, let's take that out because it's huge. Okay, this is not staying in firmly. So I'm going to have to push it with one finger. That means that it's not going... As I need three hands, one for the solder, one for the soldering iron and one to push the component. We'll just tack this on for now. But the capacitor is in the wrong place. Now I can push from this side and remelt this. Like so. Now the solder joint will hold this component in place while I solder this side. Perfect. Okay, so let's do the other 4700. This one, which will need to do the same way. So the end is shaking. I'll get it into place. The other leg, like so. So now we can do these ones. Now the legs are not quite spaced as well. They don't quite have the same spacing as the originals, which means that they won't quite go flush with the board. But the plus side is that this will actually wedge them into place. It should make it easier to solder. That is, wedge them into place while unsoldered. Like so. That one's a bit looser. Yeah, one of the legs has bent so that it's now sort of straighter. So let's just do the tack thing. Okay. Just position like so. And do the other leg. Let's just re-solder this one, I think. Okay. Only two left to do, both of which will need tacking. And do the legs. Oh, yes, and this one, the big one. So that goes in this way around. This will also need tacking. I thought there was enough solder left on the leg, left on the pad not to need more. Do it like that. Adjust the position. Do this leg. Do that leg again. And we should be finished. Okay, so now I remove the stray bits of wire. Actually, before I remove the stray bits of wire, just want a last double check to make sure they're all in the right way round, which it looks like it. Looks reasonable. And also when removing these bits of wire, you do want to make sure that you get them all rather than leaving stray bits of wire sitting on the board. Particularly with the main stuff, that would be bad. And these bits of wire all end up in a little pot, which are used for useful things later. So I think that is now done. So let's reassemble it. This is the... Hang on, this is the main side. Wait a minute. The small tracks are the main side. The big tracks are the computer side, the low-voltage side. That's an interesting way round of doing it. Maybe they're using the big tracks because the low-voltage side has higher current. That would make sense. But that is not how Switch Mode supplies I've seen before work. Well, these power supplies do have a certain reputation for not being particularly good. So maybe there's a reason why it's like that. Okay. Actually, before doing anything else, is my tester still working? Yes, it is. Let's just... The wires are long enough. Let's just see if I can measure some of these. Okay. That's telling me it's a... 47-microfaric capacitor. Looks okay. 5188 microfarad. Well, that's kind of out of spec. It's supposed to be 4700. These are taking a while to test as well. 49, that's a little better. And by the way, you're going to throw these away. I'm not going to try and keep them for anything else. That one's better. Some of these numbers, I don't really know how to interpret such as V loss and the resistance figure. I believe that the resistance, the ESR, gets too high that indicates the capacitor's not working. I don't know about V loss. Yeah, it looks like these aren't terrible, but they are rather elderly, so new ones will be definitely worth it. That's a high ESR. And the value of 25, it's supposed to be a 22, so that one I think is suspect. They're both the same. Okay, well, I now have figures on record, so let's get the Amiga out and then we will smoke test this. Okay, I have the Amiga out. I'm using this phone to record the video and by the magic of compositing, that will appear. Yeah, smoke test time. Power is not plugged into the Amiga. Let's just turn it on and see what happens. Well, no funny noises, no disturbing smells, none of my fuses have blown. That sounds good. Let's just take a quick look at the voltages. The meter is somewhere where it can be seen. Now, I believe that this power supply needs to be pinged by the computer before it turns on, but let's just see what we get. DC, not a lot, not a lot, not much, very little, and nothing. Okay, so, nothing seems terrible or seems to have happened, so let's turn that off again, plug it into the computer and see what happens. Okay, moment of truth time. Go. Got power lights. I believe it's coming to life. It's trying to boot off floppy. There is no floppy in it. Okay, that looks good. I need to reinstall this. I'll probably make a video about that sometime. I have not plugged the mouse in, which was intelligent. Can I move the mouse around? Using key combinations. No, I can't. Okay, well, this seems very plausible. The screen image looks decently solid. The drive is working. Let's just stick a random disk in and see what happens. Okay, and reboot. This one now tried to boot off the floppy. I actually forget the precise status of this machine, which workbench is in it, so which kickstart is in it. That's not so great. Okay, well, as power supplies go, I think that is working. The floppy drive when it was operating seemed decently positive, but I think that is a success. Actually getting the thing up and running again is going to be a different project. So here is the brick itself. It smells warm, nothing more than that. Yeah, I think that's a success. Excellent. It's nice to have a project that actually hasn't turned out to be more complicated and or more failing than I was actually expecting. Also, I now notice that I actually forgot to press the record button on the phone, so the video you're seeing is not captured on the phone. It is in fact just zoomed in from the phone screen, which is why the quality will look terrible. I mean, I could just reshoot this section again, but then it would be fake. Anyway, I hope you enjoyed this video. Please let me know what you think in the comments.