 Well, here's something I've been putting off doing for a while. This is my trusty Commodore Amiga A600, it's shown up on this channel several times before. As you can see, it is currently playing the Xenon 2 demo disc. If I turn the sound on, you can hear the excellent soundtrack, but if you know it well, you may be able to tell that something is not quite right. Now, if I switch from the Xenon 2 to Kyeref's Amiga test suite, through the magic of video editing I can do this like this, I can demonstrate in rather more detail what's going on. It's currently set to play a sine wave through the four Amiga sound channels. All currently off. If I turn on the left channel, you get an annoying beep. If I turn on the right channel, you also get an annoying beep, but it's a quieter one. If you turn on both at once, you might be able to make out the fact that the left channel is much louder than the right. So what I think is going on is this machine needs recapping. It hasn't been done yet, all Amigas need it eventually, so I'm just going to have to do it. It also doesn't help that Commodore kind of screwed up with the circuitry, the audio circuitry on the A600, and you have the wrong type of capacitor to decouple the output audio. So it chances are that those two caps are failing, and that's why we're seeing this problem with the audio, which is amazingly annoying. I don't like recapping machines. I'm not particularly good at it. I have actually done someone else's A500, and it does seem to work. So I'm going to give it a try on this A600, which will be fun. Okay, here's the A600 motherboard removed from the machine. The capacitors we need to replace are these electrolytics here. They contain on the inside a tightly wound coil of paper and foil, all soaked in disgusting electrolytic fluid, and the fluid evaporates over time, causing the capacitors to dry out, and that changes the electrolytic properties of them. If you're unlucky, a capacitor will actually start leaking fluid onto the PCB, which is really bad, because the stuff is noxious and corrosive and can wreck your PCB. So this is actually a fairly straightforward job. The most difficult bit is going to be these two capacitors here, because they're sandwiched between the audio jacks and the keyboard connector, which is this thing here. Having people on YouTube who've done this, mostly what they seem to do is to desolder this. However, you really need a vacuum desoldering gun to make that work, and I do not have one. So I am going to attempt to do this without desoldering this thing. Now we can get a bit more room by unclipping the top part of it, but it's not a lot. Now, this is the point where I need to explain what I'm going to do, because there's a right way to remove capacitors, and there's a wrong way. The right way is to use a hot air gun or hot tweezers. Now I don't have any hot tweezers, and while I do have a hot air gun, I've tried removing capacitors with them before, and it just doesn't seem to work. I'm just very bad at it. And what tends to happen is that I melt everything on the board around the capacitor, and I blow off all these epically tiny surface mount components, and these are actually quite big ones. So I'm going to take these capacitors off the wrong way, and this is where my comment section will fill up with people screaming that this is bad. This is correct. The way I'm going to remove these capacitors is bad, and you shouldn't do it. However, I have practiced, and it is safer for me to do it this way than it is to do the right way. So I'm going to demonstrate, but not on the Amiga. So that goes over to the side, and we get out a junk board, and let me demonstrate what I'm going to do. So here we have some zoom onto the board, and here is a capacitor that we're going to remove. This is a surface mount capacitor. It's fastened down by these two metal pads onto the surface of the board. Now, it's tempting just to grab the thing in the pliers and pull. This will ruin the board. The copper on the surface of the PCB is glued to the fiberglass backing, and these pads are very small, and as a result, there's not a lot holding the copper pads down. So vertical force is likely to tear the pad off the board, and that's a bad thing. So instead, what I'm going to do is to grab the cap firmly in the pliers and then do this. I am pushing down on the board. What I'm doing is using metal fatigue to break the wires inside the capacitor. There we go. Let me put the upside down cap here so that you can see it. So one leg has actually come off inside the capacitor. The other has come off a bit further down. This has left the soldered on parts of the legs safely still attached to the pads on the board. I think this one, which you can see sticking out, may have actually pulled off the solder, but it's left the pad behind, so that was a little bit risky, but seems to have worked. I can then use the soldering iron to clean up the pads and remove the torn legs of the capacitor. And I've done a ton of practicing on this board, and this seems to work and be reliable. So that is what I am going to do. Okay, moment of truth time. I'm going to start with a nice, easy, big one. And this trick is smaller movements because you're not trying to break anything with brute force. You're trying to use metal fatigue to break the wires. It is coming loose. There we go. Perfect. So I'm now going to make a note of what this is. So that is C303 is a 22 microfarad. Looks like 35 volts. I've got a list somewhere else, but it's best to be sure. And interesting. That doesn't look great. The solder is dull and there's some goop on the board. Yeah, you see the discolored tracks there? I think that capacitor was leaking. Yeah, this machine is definitely overdue a recap. Okay, so I am just going to power through these and you're going to see it all in time-lapse. Because this is really boring. That is all the surface mount caps taken off. Some of them have left behind the bottom plastic shell. They'll come off when I clean up the pads. However first we need to take off these electrolytics. These are through-hole capacitors and should be relatively straight forward. Next step is to remove these plastic things. I don't think the best way to do this. I pulled it in my quite expensive and extremely nice soldering tweezers. Heat up the pin and it just lifts off. Okay, and now we go and do all the others. Some of those melted rather than coming off with the pins, but this now takes us to the next stage of the process, which is to add copious amounts of flux and then clean up the pads and removing the capacitors has left fragments of wire behind, so we have to remove those. I'm going to clean the flux off with IPA. Most of them look clean, but now we're going to attempt to use a piece of desoldering ribbon to remove any excess solder from these pads. So now let's try putting one of these new caps on the board. So this stuff is pick and place machine ribbon. It's not really designed to be used by humans. So we remove a capacitor and this came from the bag labelled 220 microfarad and I'm a little perturbed by the fact it says 220 on the can, but I have no way to measure it, so I don't know if it's correct. We'll just assume that our mouse did the right thing. This mouse, digikey, where the digikey did the right thing. So let's bend the legs up just a little. Now I'm going to do 304 first, so let's see if I can actually get this right. Let's solder on the positive side, nice fresh leaded solder. So actually holding the soldering iron in my left hand here, we want to apply just a little bit of heat here, so that end of the cap goes down and gets put into place. Positive is the side that does not have the stripe and now I can see that the other end of the cap needs soldering, yeah that's locked in place, yeah that wasn't great. Normally with surface mount components you put the solder on one end, fasten that down and then go to the other end. I think that with these I need to try putting solder on both pads first and it is in place. Okay, so now we just repeat this process hopefully with focus, focus. So these two are now fairly firmly soldered on. So we repeat this process with all the other caps on the board, except for these two. I'll come to these. A bit of a problem has shown up. This is C321 which is a 10 microfarrow to 25 volt capacitor. These are the ones I bought. These are 10 microfarrowed 250 volt capacitors because I couldn't get 25 volt ones. I didn't realize but these are substantially bigger than the pad is. So these ain't going to go on. Now what am I going to do about this? These are just way too big. I could make them go on by making them stand proud of the board. The legs on the bottom will bend down. It'll be a bit awkward to solder and of course they will stick up. I think there's room. There's a big electrolytic here so yeah there will be room. The other thing is to use through holes. They'll have to go sideways and have the wires soldered on here. I think that is actually worse than using the surface mounts. I'll go and do a bit of poking around. Okay through hole it is. Now the floppy drive goes up over all this so there's not a lot of clearance. So this is going to have to go more or less sideways. So we're going to bend these legs. Okay positive is to the south of the board. So this is going to go on like that. So let's trim the legs down. Okay and a stripe goes to the north. So let's put some solder on the pads and put some solder on the legs. Should it be better off this way around? Certainly easier to put on this way around. It just needs to be at a slight angle. Beautiful more or less. Okay so now we just need to do the same for this one. This one is further back. There is more clearance so it will stand upright I hope. Okay I think this is working. It's a little bit ugly but it should be fine. So let's go over to the other side of the board because now we're done with the the right hand side to these. If you look at the positioning this thing is awkwardly in the way so I can put the component on but then there's no space to solder the top leg. That's awkward. I can attempt to melt this pad and then slide this in before it solidifies but that does not seem like a good idea. If it doesn't work I'll never get it off again. So I think if I bend these legs way down I can then stand the cap up and get in underneath to solder the thing in place. Now this plastic shroud here is where it's going to come off completely. What this is for is to stop the rim of the capacitor touching the PCB. I actually don't think we need it. So I'm actually going to start with the one over here because I need to be able to get in on the right hand side. So positive south stripe north and then it will bend over to go upright because it's slightly projecting off the left side of the case. That's kind of terrible but it does seem to be working. So we can do the same thing with this one. Okay like so. Hmm not very happy also not happy with the focus but it looks like it's working. Okay the next two C235 and 236. These are 100 microfarad jobs and two of these and these should be the right spec ones. Okay so now we need to do C239 the one next to it. This is the chart says that is a 100 microfarad job. If I put these on the wrong place 235 at 236 100 and 100. Right so this one is also a one of these hundreds and that's pretty short on space. So this one is actually too far over to this side. So let's see if we can adjust it. This is one of the advantages of this horrible hack. Oops okay well that wasn't very well soldered on then. It's important to know that this one actually has the the plus sign to the north so this is inverted relative to the others. So I can make this go in but the question is will I still have room to attach the IDE connector because it's going to end up canted over rather. There's one way to figure this out. Okay here is the IDE connector so if we plug this in then how much clearance do we actually have? Enough just now I was originally thinking that one of these would be one of the 10 microfarads but looking at my chart I see that these are 321 and 306 we've done 460 and 214. C214 is this one the label is here and 460 is this one and it's also worth noticing that these are also the opposite way around. So we can easily put our two standing upright ones here and here. So it's this one that's the awkward bit but I think this can be made to work. Okay wow I hope this will work. I don't think I managed to show this on camera. Yes that goes on neatly there. This one is C459 and it is a 22 microparad surface mount and this is one of these little red ones so this one should actually fit which will be nice. Any other end goes down easy. So now we have this one I think that's actually all the surface mounts. Yeah we've got the through holes to do now. Yeah this is as this is through hole stuff it's going to have to be both sides of the board which is we're going to have to hold the thing more or less vertically to work on which means none of it's going to show up on camera so I'm just going to do this bit offline sorry. Okay I've managed to attach the big through hole capacitors with varying degrees of goodness. This one just wouldn't fit so it's standing up which brings me to the last two which are these. These are the audio capacitors. Amiga, sorry Commodore, what's this? That's a metal freak. Amiga cocked up with these. These are decoupling, smoothing, whatever the output audio signals so the voltage range these have to cope with goes positive and negative and you're not supposed to use electrolytics for that. You're supposed to use unpolarized capacitors, ceramics that sort of thing and this is probably why these have failed on my Amiga. I mean they have failed it's not something else. So I am going to attempt to replace these with come on actual surface mount capacitors. I'm not sure this will work but let's give it a try. These things are tiny so the first thing you need to do is to add a solder to just one pad like so then we remove our capacitor from the package just one. I'm not actually sure these are big enough so I want to put it there so it spans across the two pads. I think it'll work. Yeah I think that will actually work so we hold it down and apply some heat. That should have soldered it down on one side it has actually moved so let's see if we can get this back on track. Yeah that is still fastened down so now we apply some solder to this end. Once you actually have the knack of it surface mount is supposed to be much faster and easier to solder than through hole I do not have the knack of it. Okay let's see the other one so that goes here okay now do the other side okay and do actually seem to be soldered joints both ends. Yep I think that's done. Excellent and I actually forgot there are two more I need to do which are these two which are some more of these we're going to stand these upright. Okay I believe it's done so now I just need to check things and then we start reassembling and then see if it works. Okay everything looks all right so moment of truth time let's turn it on and see what happens. Well the computer seems to be working up to a point there we go nothing's exploded nothing's making nasty hissing noises. Okay we should have some stuff hooked up so I turn on and it should boot okay so we go to the audio menu and it suddenly occurs to me that I can't actually hear what's going on when this is hooked up to my capture device so after plugging in my headphones I have discovered that sound is only working on the left hand channel so that suggests that I've cocked up something here the left hand channel is this one that the red connector because for some reason these are reversed which is odd. Okay I'm going to have to do a bit of setup in order to debug this because I can't do it with the the capture card plugged in because you know I need to be able to hear what the result is. Well I did some poking with the oscilloscope and made a few discoveries so here I've got the schematic of the audio system for the Amiga 600 and the audio comes out of the Paula chip here to the right channel and then goes through this series of filters and then it goes to the right jack circuitry with C334 the capacitor that we replaced with the ceramic here. Now this is the Paula and the right pin is I believe is this one and I managed to trace this on the board that is here to these two op amps that make up the filter and all the four audio op amps are in this chip here and I managed to trace the signal so that it was arriving at pin it was coming out of pin eight which is here that is this pin here on the op amp and then it goes through some stuff and it goes through the other op amp to pin seven which is this one and I was observing that the signal was going in to pin eight well rather out of pin eight but not out of pin seven so my immediate thought was that the op amp failed and that's easily replaceable if it was partially failing ahead of time then possibly the stress of replacing these two caps next to it caused something to fail however this didn't seem entirely likely and indeed further tracing showed up something rather interesting so the signal comes out of pin seven and goes to C334 so here is pin C334 and if I do continuity test of course we get nothing through the capacitor because capacitors don't conduct through them however if I touch the metal piece on the top of the capacitor and this pad here we get nothing but if we do it for the one next to it we do get a signal this solder joint is bad it's not hooked up to the capacitor I'm not sure why this makes a difference so if we zoom out back again I can see that there is a contact between pin seven here the output of the op amp and the bottom end of the cap but of course it's not connected at the top end so as far as the op amps concerned this pin is just not connected to anything well there's a bit of other circuitry there but I can't see why we're not getting any signal coming out of this pin the only thing I can think of is something on the on the output side is drawing that line down but I can't find anything like that and I don't know why having this capacitor hooked up would make a difference but this is something which is wrong and therefore it is fixable so let's go and fix it so fixing it should be straightforward we just need to heat this joint up and apply solder I think that will probably do it so let's turn the meter on and try that interesting still nothing so that is metal this is me touching both ends of the the metal thing on the end of the capacitor okay it doesn't help that I'm not actually seeing what's going on very well okay I saw the solder actually flow over that metal end cap so that should now be working just testing the other connections yep okay so that's not a great joint but it should at least work so that's not a great joint but I think that what I need to do now is to hook everything back up again and see if that has made a difference and what do you know it doesn't work it's doing exactly the same things it was before I recapped it this is a good thing in some ways listen so the left channel is working fine the right channel is very quiet so this is suggesting that something else is wrong with this possibly one of the components in the output stage this is all nasty little surface mount stuff so I've identified these four components here and they are all surface mount jobs on the other side of the board trying to debug this is going to be tricky because everything is connected to everything else I mean you notice that the these filters this is a feedback affair so the output of the op amp is actually being routed to the input so trying to decide whether the op amp has failed will depend on whether the components downstream of the op amp have failed I think that given that current cannot flow through a capacitor I'm wondering if it's worth trying to take c337 off the board that will isolate the the right hand stage apart from these cross-connects here which I'm not quite sure what they're for and that should let me rule out any problems with these components here I found some references that occasionally these do die so that's interesting and I should really find out what these are for these are each of the jacks has a switch in it so that it knows when something's plugged in so I think that what this could be doing is trying to mix the two together there is an actual mono output but that doesn't go to the jacks yes I'm going to have to do some reading up on this