How to fix an 8-bit computer with only cheap tools

well hello everyone and welcome back to adrian's digital basement this is actually the first video i've made in a while because i had surgery on my right hand and put me out of commission so the videos you've been seeing on the channel both channels actually were ones that i made before the surgery on my hand i've been giving updates to my patrons about the progress of what's been happening with the hand and recovery and stuff there's even a funny video from me waking up after the operation in the hospital so if you want to see that stuff if you're a patron you can go check that out on the post feed there on today's video though i'm going to be following up from last weekend's video i'm going to be fixing this commodore 64. and i'm going to try to do it entirely with this very inexpensive scope that i reviewed on last week's video sixty dollars all in shipped for this and we're gonna see if that's gonna be good enough to fix this machine last week i only looked at like kind of theoretical performance of this oscilloscope here and not actual use case on a real machine i won't be using the three thousand dollar national instruments virtual bench that i normally use and if during the repair of this motherboard i need a multimeter i'm not going to use my ev blog one i'm going to use this one annang an 8008 i think this thing was 20 or maybe it was 30. i've had it for a few years it's a very inexpensive multimeter but it's actually pretty good performing all things considered i'm not sure i will need it though because typically the oscilloscope is all you really need to do troubleshooting on a board like this so without further ado let's get right to it [Music] all right this is the 64 board i'll be working on today this one also comes from henning as were the last batch of them i had done some testing at some point when i opened these i think and i wrote for this one number four black screen with and without the dead test no sid burning up ram okay so this one might be an easy fix all right unlike the blowtorch motherboard this one it looks like to be in really really good shape i'm actually not seeing any rework whatsoever on it and on the front side this is one of the very latest versions of the short board so when i say shortboard it means that commodore consolidated a lot of chips into the larger pla chip here so you notice this thing has way more pins than the earlier one and it's even more consolidated than the earlier version of this short board in that this had a color ram chip off to the side and commodore actually went and integrated right into the pla for this particular board the date on the pla is 1991 as is the date on the rom chips oh and this 6526 so 1982 the 64 came out and this was getting near the very end of the run the top of this capacitor here's a little bit torn up this thing was obviously stacked at some point but that shouldn't really cause any kind of issue and if it is a problem i can just replace it okay so as i indicated in the note the ram is burning hot and that really does seem to be a very common thread on these short boards i am not sure if we've had one with the burning up ram in this series yet like this recent batch of repairs but i know for sure i have fixed short boards with that problem so before i actually do any further testing i'm just going to go out on a limb and say that this board probably works perfectly once we replace both of these ram chips but this is more about testing the hand tick here so let's see how the signals look at least on the two ram chips with this thing powered up if you're interested in knowing more about my specific thoughts on this scope i recommend you check out this first video which actually i just posted today if you haven't watched the video on the hand tech and you don't intend to i'll just give a really short version here 60 scope very inexpensive but surprisingly capable for the price for only sixty dollars a lot of the really inexpensive scopes you see floating around the internet the ones they're like 10 20 they're junk they're not useful for this type of work trying to fix a board that runs at one megahertz or two megahertz like a 64. you need something a little bit more capable than that and something like this hand tech actually should be able to do the business but i haven't actually done repairs with it yet so of course that's what we're going to be doing here in the video on the scope i talked about how the software that came with it just has a lot of shortcomings and some idiosyncrasies that are really annoying well it turns out that there is an open source piece of software that can actually run the scope and it's this openhandtech6022 it's designed specifically for this oscilloscope and what's even more cool is that this software works on linux mac and windows so if you are a non-windows user and you want to use the 60 scope no problem just use this open source software when i first took a look at the project i was surprised that the software looks actually quite a bit better than the uh the software that comes with the scope plus it's of course maintained and updated and open hand tech has quite good and thorough documentation which i recommend you check out one of the things that i talked about in the first video about this scope is that the triggering is really inadequate and i have to read about this a little bit on certain forum posts and actually on this project the reason why is evident this thing has no hardware triggering built in the hand tech scope entirely relies on software for triggering basically it looks like this device just digitizes the signal and then takes that digitized signal and streams it over the usb cable it doesn't have any ram or buffer memory in here unlike the virtual bench which does have sample memory built into the scope so what you're looking at on the screen of the computer is just the relevant portion that you're interested in and of course the virtual bench like regular bench scope and i'm pointing up to my rygal scope there has actual hardware triggering where when you set up the trigger thresholds that is happening in hardware so you're going to get really quick and accurate triggering not reliant on software to do the triggering like with this but that's what enables this thing to be so cheap to get rid of that hardware triggering on the two channels makes this thing very inexpensive and this is it this is open hand tech running right now a couple cool things about it you can actually alter the calibration output the default is one kilohertz we're now looking at a 100 kilohertz signal in addition has this split view which i really like so this top view is actually what's being streamed over usb and you can see the cursors here and that's basically zoomed in on this lower section here and i can use the keyboard and the scroll wheel to zoom in and out and notice it actually stays on the trigger point as it zooms um i do have the persistence mode turned on so there you go notice it's moving a little more jittery like and i think that's because the frequency generator running at 100 kilohertz is just not that stable of an output but you turn that on it's kind of like a little bit of persistence in the split view which you can turn off by clicking on the little magnifying glass here if i just use the scroll wheel alone it scrolls around in the waveform but if i use control that's when it zooms in and out what i haven't figured out is a quick way to change the time base other than pushing these up and down arrows which i don't really love you can actually control the sample rate from my understanding reading the documentation the higher sample rates are actually really glitchy and we saw that in the official software because i actually ran at 48 mega samples per second and you got really weird results and even in this we're seeing that so at 30 mega samples a second look how it's all over the place but if i lower it down to 10 we're getting a much more stable signal maybe that's a problem with the usb not being able to transfer enough data i don't know but i do have a plug directly into a port on the motherboard it's not going through a hub or anything so who knows in fact this software i think locks out anything over 30 mega samples because it says basically it's unusable which was what we were seeing with the official hantech software at 30 mega samples a second definitely we're losing data or something over usb because this little glitch here that i've zoomed up on definitely doesn't exist in reality so let's put that back down to 10 hit play and while we don't have as much resolution here on this rising edge of this 100 kilohertz signal we certainly have just overall a more stable data stream one more thing i'm going to talk about is the measurement capabilities and that is on the bottom here we're at one volt per division so these are the division lines we are at around 2.2 volts peak to peak 1.45 volts rms and 100 kilohertz so there's a little bit of a frequency counter remember all of this is happening in software so it's definitely not going to be accurate especially if your sampling is causing all that weird glitching which those higher sample rates do cost all right so the motherboard is plugged into a power source i have a ground lead connected to the ground on the motherboard and i'm ready to start probing on my pc here's the pin out for the ram chips and of course the scope is running okay i'm on pin two which is the first of the four data bits that are on each of these chips so each of these chips has 64 kilobytes in them but only four bits so that's why you have two chips to get a total of 64k across eight bits so let's turn this on so we're getting activity and i'd say that that looks okay but then i'm on pin three and this looks really bad pausing it it looks like something is trying to hold the data bus high or it's it's falling very slowly back on pin two this is a little bit more how it should look kind of goes high and then low with these little glitches which are normal i'm on pin 17 here which is data bus three and there's pretty much nothing happening there and then this one here is pin 15 and notice how low it is so peak to peak 2.4 volts that's completely wrong so whatever's trying to drive the bus here which is probably the cpu is being held much lower than it should now let's look at the other pins on this other chip here uh we're getting 3.36 volts peak to peak it's also getting very hot uh that pin here which is ow sorry that's burning my finger uh that pin there looks bad okay this one here is also 2.9 volts and this one here is 3.44 let's look at this one i didn't even notice that yeah 3.6 i don't know all of these are really low this one looks really screwy this one's not so bad peak to peak is 4.23 but we're getting all this weird kind of droopy stuff which on this one we are not getting so yeah hot chips like this probably what's happening is the cpu or other things is trying to drive the bus like the rom chips and these ram chips are just shorting those lines directly to ground i don't think the heat that we're getting out of the two ram chips is only coming from the fact that they are shorting the data bus lines while i'm theorizing they're shorting the data bus lines down to ground because whatever's driving the bus like if it's the cpu or the roms there's not much current in that so if you just create a short circuit there you're not going to get super hot chips and the cpu and the rom chips themselves aren't that hot now shorting the outputs of chips completely to ground is not good for them so this actually could damage things like the cpu or the rom chips so it's better to not run the machine too long while it's in this state but sometimes you need to run it long enough to figure out which chips are shorting it to ground okay so for methodology to see if these ram chips are causing this five volt signal here on the data bus to be all the way down here around 2.8 volts is we are going to cut the leg on one of these so just get some really sharp snips like this flip them over and i'm just gonna cut pin two which is data line zero and the reason why cutting like this isn't a bad idea is because it's not difficult to put a blob of solder to reconnect that if you're wrong as soon as i cut it the pin sort of pulled away from the chip there so it should not be connected anymore so now if i put the scope back on there on the motherboard side and we turn this on okay and look how different the signal looks now now i know it's jumping around a lot let's pause this see if i can get it where there's actually a waveform there you know i'm going to adjust the trigger a little higher play there we go look at that 4 volts peak-to-peak oh and the software crashed all right and we're back so it's not the most stable it did say on the github repo that it's really thoroughly tested on linux windows is kind of not so much testing and on os x or mac os it's really not tested at all so we turn this on again there we are four four volts peak to peak don't worry about this middle part that is just because the data line is floating there we go look at that that looks great far better than how it looked before with it connected to the ram chip now if i touch on the top part of the pin on the ram chip of course all we're going to see is it down around ground but let's do the same thing on the next pin okay so it's looking like that let me adjust this trigger level here up a bit all right so it's um actually not looking terrible i don't think this line is actually shorted because uh take a look here uh see these signals here they're actually decent i mean it's not up to four volts but it's nearly at 4 volts it's posit so yeah there it is 3.84 so i don't think that pin is actually a problem but let's find another one that looks bad and shorted this one looks weird but it's still going up all the way here to four point something volts so i don't think that is a problem okay 17 that's another bad pin oh we're not getting anything there on either of them okay so that pin i may be just a dead short let's check this one okay this one here this looks really low and bad see there we go 2.4 volts that is pin 15 let's cut that pin all right turn back on there we go looking much much better now of course it's hard for this thing to trigger on there but we are still getting when i pause it with an actual signal come on oh i keep pausing it at the wrong time give me something to stop on i really wish there was a single shot button that just did like one capture but there it doesn't appear to be anyhow you saw there for a second i just turned it off that we were getting the full height all the way up to like four point something volts all right next pin i want to test is this one which is pin 17 and that's giving almost nothing how about on this chip here we're almost getting nothing let's just cut that pin and let's see if that if that returns to something actually we're going to need to do it on both chips i think it might be shorter on both of them all right and we hit play on the yep and look at that see it jumped right up remember this was stuck all the way down around ground and now we're up around 4.58 volts we're not getting a lot of activity on it if i turn the power off and on but i think that's okay anyhow i think that pretty much confirms that these two chips are the problem so what i'm going to do now is i'm just going to show the method for getting chips out the safe way of doing it without causing any damage just cut all the legs on them with these side cutters and that's going to free up the entire body of the chip and then we could just use the soldering iron to pull out the pins and then we just have to clean up what's on the motherboard this is by far the safest way to get chips out and if you know they're bad like these hot chips then it is the way to go and unless you're 100 sure of what you're doing getting chips out i would always use this method there it is so that's it so the chip has been removed what's left is just the pins we just do this one as well quickly and there we go both of the bad ram chips have been cut out just leaving the legs for removing the pins from this i'm going to use this the pine sill direct heat soldering iron it's usbc powered i have a video on this it's 25 for this thing i really really recommend it and if you order one please don't forget to order one of these it's a silicone usb-c cable these are really flexible plus they don't burn very easily either regular ones will burn because they have pvc coatings and they're stiff so get one of these they're on their website they're pretty cheap as well i'll put a link in the description to the review i have of this so check that out if you haven't seen it but basically i have this set for around 350 degrees grab some tweezers and then you just heat up the pin and then you just pull them out one by one when you've heated the pad enough the pin just slides out super easily there's no resistance or anything and if there is then you haven't heated it enough one other thing to consider is when you're cutting the chip away make sure you cut the pin up near the body of the chip and not down by the motherboard because if you don't cut it down by the motherboard you're not going to give yourself anything to grab onto with the tweezers that is also the ground pins can be a little tough sometimes so turn up the heat a little if you need to for these ones maybe push the boost feature on the soldering iron and then you should be able to free it all right another tip is it's easier sometimes to turn the motherboard around 180 degrees to do the other row of pins all right so one of the pins i accidentally cut too low so i'm going to have to try to get it out from the other side that was my mistake which i just warned you about even i did it uh okay all the legs are out there they are the one that was cut too low what i had to do is on the back side of the motherboard i had to push it this direction through the top because it's wider on the top side when it connects to the chip so you can't push it through the via that way you have to push it through the motherboard but there wasn't enough to grab on this side so on the back side he heated it up and using the tweezers i pushed it up and then i was able to grab it on this side now so okay so the pins are out but there's still solder in all the holes we got to clean that up so we can get a socket in right a couple ways to do that well three actually you can use uh braid you can use a pump a hand pump or you can use a detoldering pump like the hakko 301 which i normally use that thing's like 300 so it's certainly not cheap if you're just doing a one-off machine braid is like this stuff here this is mg chem braid or solder wick i barely have any left so i'm going to run out shortly this is really good i do recommend this but avoid the cheap stuff get the good name brand stuff it works a whole lot better and then this right here is a hand pump these actually work quite well too i think if you get good at these this is the solder solder pulse then they work well but i personally i'm not very good at using these so i kind of struggle with them i do better with the wick why don't i give the pump a try what i'm going to do is i'm going to add fresh solder to all of these holes or at least a few of them all right here we go i'm going to heat up one of these and then try to suck it out oh that worked okay so i just got to re-prime it next one oh okay all right this thing is working great again just like removing the pins the ones that are on the ground plane or the 5 volt plane are going to be more difficult so i kind of recommend you definitely have to put a lot more heat into those first so i'm just going to add a good blob of solder onto these so i'm going to use the boost to really get some good heat into there and then here we go let's try it that did not work try again okay i think i got it let's try this one up here boost by the way is holding a button down it goes up to like 420 degrees celsius gets a lot of heat into it ah that was my fault i missed let's try again nope there we go oh that i got it let's try this one here nope not successful this is where the hakko 301 i normally use would make super quick work of these and for me who does this a lot it's totally worth it look at that i am unable to get this one to to do it it's just you got to get enough heat into it so that it stays molten long enough to for me to take the tip of the soldering iron off of it and then uh you know then it sucks it up into this solder pulse here look at this i'm still here trying so it's uh it's created a little like a little like a mountain of solder there as it tried to get it out now that's really frustrating for my sanity i'm going to move on to these other ones which will be a lot easier obviously when you're using something like this you get a practice and then it becomes a lot easier to do and i just don't have a lot of practice because i just don't have a lot of need for it because i have the hakko 301 that's what i use okay so everything is done except for these two which i think it's five volts and these are the top uh these four pins there those are ground actually you know what i'm just gonna try with the wick maybe that's gonna work better now you might be saying to yourself why is this so hard well it's just because it is and that's why specialized tools like the hakko exist to make this doable look at that that actually made quick work of that using the boost temperature on here plus the wick so i guess this is a good method for uh getting these ground ones out nope that one did not work though heat is your enemy on the ones that go to traces and that's because it's very easy to lift those but these ground plane ones it's a lot less risky so you don't have to be as careful with these ones ah see this is not easy okay i think it's done this one looks like it's cleared out on the top but not down at the bottom so i'm going to try again just add some fresh solder into there okay there we go i think those top four are done let's do these bottom ones so it looks like it's done it's a little crusty because uh this has flux in it so that gets on the board so you have to clean that off with some alcohol what you got to do now is i just need to check the back side and the back side looks great so i think we're good to go to put sockets in there it's going to clean this up first put a little bit of 99 ipa get something like an old toothbrush and just clean that up get all that extra flex off and it also gives you a good opportunity to take a look at it with a little magnifying glass or something just spray on the alcohol and then i recommend you take a little bit of paper towel and you just sop it up basically so any kind of flex residue that's left behind you should hopefully get on there you'll see a little bit of brown and we'll just do that on the back as well so there we go two ram chips removed now i always use these types of sockets these round hole types you can get these from aliexpress or ebay very inexpensively but you could use any socket those are too big i personally prefer this kind of socket over this kind which are like the double wipe ones just because i feel it's a little harder to damage these these can actually get damaged relatively easily because they're kind of cheaply made these days if you install this in the board and then you put a chip into it that you've previously desoldered from the board the the legs are a little covered in solder they're not perfectly smooth i've actually had these get damaged and i had to take them out and put new ones in and it's a real pain in the butt and i really don't like to do it so these type that doesn't happen but this type is actually harder to get chips into because the holes are less forgiving than these these dual wipe ones you have to have everything aligned just perfectly but for me i find in the long run these i prefer but i think it's personal preference which you will use all right so the two sockets are in and they're loose you see they just move around very easily if i flip the board over these are just going to fall out so of course you can just put a piece of tape across it i like to use this which is blue tack it just keeps them handy near the bench and i just basically push it down on top of the sockets like that to keep them in i just solder a couple of pins on the outside so i'll do like one right here and i'll do one on the other side as well and that's just to hold it in place and i'm going to pull the blue tack off and then i'm just going to double check that the socket is completely flush against the board so there we go now i can just take that off if i solder these middle pins with this in there kind of melts the blue tack i don't recommend that so now what i'm doing is i'm holding this socket with my fingers i'm holding on that pin and that pin so i don't burn my fingers and i'm just going to melt the solder again and i'm just going to push the socket into the board and it was all the way in just making sure that it's flush against the pcb and the same for this that one that one okay we're good and now i'm just going to resolder all the rest of the pins and it does help to get a little length of solder i just burned that off just as it's easier to handle so i'm just going to go through these again 350 degrees and reattach all the pins there we go i'm all done it's a good time now to inspect your work use something like this a little i loop you can get these from ebay aliexpress amazon wherever for just dollars this is like a cheap version of it there's expensive ones but this was literally like two dollars so those are great for also inspecting traces and stuff you should have inspected the traces first before installing the sockets again i took a close look everything was perfect and the final thing you might want to do is notice there's some flux on the board here just because that was rosin core solder so again i'm just going to put on a little bit of ipa there i'm going to use the toothbrush here just sort of get that off this cleaning part is completely optional it doesn't hurt it to leave it it just looks nicer in the long run and again i'm just dabbing with a paper towel here to get up any of that flux residue cleaning it just makes it look nice you're a long one if someone's looking at the board in the future they won't see some really ugly rework has been done i can still tell that i hand soldered this versus the rest of the board which was wave soldered or whatever it was done with this manufacturer you could still tell a human did this not to mention on this side of course you're going to see two sockets there so commodore doesn't use that but yeah a little cleaning goes a long way and now to reinstall some fresh ram chips that i know work and even before we plug in video to see if this thing is working why don't we use the oscilloscope here to check those data bus pins see how they look so here we are on pin two data bus line zero and that looks much better peak-to-peak over 5 volts next one same thing lots and lots of activity and if we go on to this side they're looking really good we're on data bus line three and this is data bus line two also looking perfect and we'll just check the other chip i mean it's gonna be the same good good that is good and that is good and like just let's take a look at this app here if i hit pause we're getting really nice signals out of this uh cheap oscilloscope i'm i'm really impressed this this piece of software works so much better than the official handtech stuff of course if we touch those chips they're not even warm at all they're just cold to the touch even without any video output connected to this thing we could look at the signaling on the vic-2 chip here to make sure that we're getting what looks like a good video signal and we can look for that line on the 64 that you get when you first turn it on that during a black screen all you'll get is a line but once it starts to display text and other stuff you'll start to see other parts of the signal so i have up here 6567 it's it's it's a similar it's a different part number to this but that is the vic 2 pin 15 is the luma signal so that's what we need to look at 19 18 17 16 15. and there it is i've adjusted the trigger so it's triggering off of the white line and this rocks this software is unbelievably better see this constantly changing information here that's because we're looking at one scan line here with this capture so that spike and then that data is one scan line well it's triggering off random scan lines so of course as it goes down the screen you're going to have parts that are just solid blue and then you have parts with text so it's going to be constantly changing if we turn the machine off and back on we're going to see the line which appears while it has a black screen and then it's going to start drawing content here we go there it is there's the line and there it is this is the border right here on one side and the border on the other side and normally it's that darker blue color and the text is lighter blue so anytime you see the little peak there in fact if we pause it hopefully we're going to get nope that's just the solid border at the top or bottom let's see hopefully we can get one where we have text of course the text when you just first boot up is only at the top there is the text so this is part of the scan line that's making up the text and this cheap scope allows you to perfectly see that we're getting a good video signal out of this machine and we do not have a black screen and this is before i've even hooked up a monitor to it alrighty so i'm going to hook up a video output cable that goes into the retro tank and i think i'm capturing the right input so here we go maybe there there we go and it works without a doubt bad ram chips i think the the fact that they were hot gave away the fact that these were bad but i wanted to show the entire troubleshooting process of validating that those were causing the problem and cutting those pin legs showed us as soon as we cut it the signal started returning to normal kind of validated the fact that those specific chips were causing the problem if they weren't hot and i had no idea what i could do is look at one of the data bus lines that didn't look right and i could start cutting a line on every chip that's connected to it you gotta look at the schematics for that until you find the one that's bad this machine seems to be working so what i need to do is get a sid chip in here and plug in the test harness i've mentioned this before that you don't need to use the test harness or the diagnostic cartridge if you don't have one if the computer is booting up to basic then just use it and hopefully it's going to be working perfectly and then you're good this is just a way for me to quickly validate that all the functions of the machine are working in case there's anything else that's wrong i can go work on fixing it keep in mind if you're running the diagnostic cartridge and you don't have the harness connected you are going to come back with errors oh and actually look at that i got some errors already now with the commodore 64 the cassette which is connected to this port right here is actually driven partially from the cpu and i think one of the two 65 26s although i'm not completely sure about that i have to go check the schematics to see that so the reason why it's saying the cpu is big bad is because i don't think this cassette is working correctly but it could easily be a bad connection on here or the cable wasn't plugged in all the way it wasn't so let me just push those down um the reason why actually now i think about it that it was saying one of the 6526 is as bad is because the diagnostic harness is connected to the user port and the user port is driven from the 6526s so it's trying to say that it sees a cassette port problem but it's not totally sure if it's a user port problem or if it's the cpu not driving the cassette port properly let's turn this on and actually we're not getting an led flash on there and that should happen when you first turn on the machine so i might have to do a little bit of troubleshooting but i don't think the cassette is actually going to work on this machine as it is right now now one thing to consider if you were working on this board and the cassette didn't work and you don't use the cassette port it doesn't really matter like you don't need to try to fix it um yeah it's coming back with a bad interrupt as well which is why 6526 keep in mind these diagnostics are really inaccurate and it says stuff is bad on here i think i've showed this on previous videos where totally was pointing to the wrong thing as the problem all right let's find the schematics for this board we're here on zimmers we have the pcb assembly number down here 250 469 and the pcb number 252 311 252 311 and here it is corrected rev b let me quickly find the cassette port there it is right here so we got five volts uh we got a sense pin notice that's running down here into the mpu which of course is the cpu it is connected here to this line which is going here to the 65 26 the flag pin but keep in mind i do have a loopback connector connected right here to the iec port and it was happy with that so i don't know if it's actually connected to pin one there cassette read the cassette reading seems to happen through srq in so they share that pig pin and that does go through here but the motor control that does not appear to be working this led which has a current limiting resistor here comes into pin three the led turns on when the tape motor is instructed to turn on pin 24 is the one that actually turns on the motor so there's unregulated nine volts and two transistors here so this is actually a switch to turn on the voltage to the motor which is what's happening here this is the sense line here so this one right here so i should be able to see pin 24 changes when you first turn the machine on because this is going to basically be spinning the motor a little bit let's just grab the multimeter oh this is all tangled together you know it's like your old headphones before we use bluetooth headphones oh it's getting tangled up one two three four okay so this is set to volts let me turn it on yep weird 3.1 volts and starts at zero though hmm that's weird i don't i never really looked at this circuit so i'm not sure if that's how this should be working so this pin right here is what goes to the led and there's a dropper resistor and that's what turns on when the motor is being told to turn on and when i turn on the power of this machine like right now that led does not light up so basically this motor on signal which does come from the cpu and goes through a couple transistors wherever they are on here that's not working that's not to say the cpu is bad it could be something with that circuit now when we follow this circuit through we end up at r3 right here that goes to the cpu i should lift that pin i'm going to lift this resistor off the board and that's going to break the connection between the cpu and that circuit so i can actually try to drive that circuit directly with 5 volts or 0 volts and we'll see if the motor turns on the motor signal turns on if it's not being driven properly because the cpu is damaged then we know that cbu is bad but if it doesn't turn on even when i give it five volts or i give it ground then we know for sure the problem exists with that circuit and the cpu is probably fine so when we take a look at the circuit here we have q1 and q2 there's q1 there's q2 there's the resistor there's cr1 which is the zener diode right there six point or is a 5.8 volts i think it's hard to see it's a little voltage regulator so to speak so that's part of the circuit as well so let me just tone out which side of the resistor goes to the cpu here i think it's pin 24 or 2524 it's that side and that's the side i am going to lift off the board there it is i just heated it up and pulled that out so the cpu is no longer connected let's go back to the oscilloscope just because we can because we have it let me test that pin right there coming from the cpu i'm gonna turn this on all right no we're getting i mean it's noisy but that's okay uh what are we getting 4.25 volts here that indicates to me that the cpu is working fine that this problem exists in this circuit right here for the cassette so i'm going to clip this clip lead on to the left side of this resistor this is the side that normally goes to cpu and i'm going to turn the power on and i'm going to short this to ground and we're not getting the led turning on and i'm pretty sure that should be happening taking a look at the oscilloscope here on the side of the resistor that's facing the circuit obviously grounding it doesn't really do anything so let's give it five volts and see if anything happens nope we're getting a lot of voltage drop we're at 700 millivolts and that's across this 1k resistor something is shorted one one of these parts is shorted and now we're going to use the cheap multimeter here to figure that out so let's uh put that on ohms and it is so let's see about this one here so emitter collector base nope that's not shorted the big one is fine how about this little guy here nope that seems fine as well definitely in continuity mode no no shorts what doesn't make sense to me is the right side of the resistor is going to this pin here on what is this q2 and the other pins aren't shorted so why when i hook five volts up to that we only get 800 millivolts because we're getting so much voltage drop the transistor here is acting as if it's shorted i'm a bit perplexed at what's happening here so i think i'm going to remove this transistor here what is that q2 2sc1815 i'm just heating up with a soldering iron again a good way to get this out is you just heat two pins up at once and you kind of tilt it out and then heat up the other pins to try to get it all out there we go so it kind of mangles the pins up a little bit but you can get that out to continue with the theme of testing with cheap stuff i'm going to use this part tester i've had this forever the battery's so old uh you just hook things up to it and it can test components and i found it to be relatively reliable of course this does the testing out of circuit so there we go clip that on yeah there we go an npn transistor that looks fine i guess i'm very confused at what's happening here okay i think what i'm going to do next before i put that part back in let's see what we're seeing for our voltage right here between r1 and cr1 i think we should be seeing i think it says plus 5.8 volts there it's kind of hard to read we should be seeing that there no we are getting 1.15 volts i don't totally understand that all right so since that transistor is good i put it back in the board the problem lies somewhere else now i just realized on q1 which is this large package here we should be seeing unregulated nine volts which i think is something like 10 11 12 13. so we should be seeing that on the collector there which is the middle pin so we turn this on okay so there it is on the collector 13 volts that's fine now on the emitter uh because it's not working right i think when i tested earlier i i saw nothing oh wait what 6.4 volts dc wait what so the resistor is still disconnected so i guess wait a second i think that's what it should be this is a little bit of a voltage regulator of sorts it's using this this larger q1 package here to regulate this unregulated down to i guess about 6 volts wait i'm sure i tested this before and i was not getting 6.4 volts i i'll have to review footage i don't remember if i recorded that or not but uh okay so wait a second so at this point if i connect up this which has the led on it that wait did you see that it blinked the power's off let's turn the power on wait a second the led is on what what is happening let me connect this clip lead here to the resistor that normally goes to the cpu and when i ground this nothing happens but when i touch 5 volts it shuts off the motor it's working now i didn't do anything i literally removed this transistor i tested it and then i put it back in the board and now this thing is working what [Laughter] this doesn't make any sense okay i'm going to reconnect the resistor back to cpu and theoretically we should have a working tape circuit now magically so if i plug this back in now the resistor is connected what should happen is when i power on the machine we should see the light come on for a second and then it goes off like that this doesn't make sense the machine has fixed itself i have to think that this this transistor here is marginal and heating it up removing it and then reinstalling it in the board has resulted in it just magically working while editing i realized another possibility that could have caused this problem that unregulated nine volts there might not have been working and maybe in my manipulation that started working like maybe the fuse wasn't in there properly or something was wrong over on the side of the board by the power switch on this type of board the only other thing that i'm aware of that uses that supply is the sid chip which isn't actually installed in this board so if that weren't working maybe the city would have no audio but i didn't actually get to testing that all right to know for sure test harness is reinstall and plug the video back in let's power this up the light blinked so that's implying the cassette is working so again i think i talked about this in the last 64 repair video i did or maybe it was two ago the diagnostic cartridge can tell you that things are bad like the cpu is bad or the 65 26 is bad i have a feeling that everything on this test is going to pass now although that interrupt was showing as bad so let's see what happens all right cassette past interrupt okay there we go everything fast false readings on the diagnostic line so i think again it goes back to what i was talking about with ram always try to understand the faults if i had gone removing the cpu and the whatever it said it was one of the 65 26s was bad i could have caused more damage to the board not to mention if i had cut the legs off those parts that would have been expensive because those chips are quite valuable now they're hard to find both of those and they would have both been ruined so in the end i didn't exactly figure out what the problem was but removing that one transistor and putting it back in fixed the problem so i'm going to mark on the can of this probably that that is a suspect part so in the future if i work on this board or say someone else works on this board and the cassette's not working i'm gonna hopefully give them the indicator to suspect that that transistor has failed and to replace it and put a new one in so i think everyone knows what time it is now it's time for an 8-bit dance party so i'm just reconnecting my speakers to the bird's nest of wires that i have down here because i have to split the audio signal and send it into the dac or the adc that is so i can record it for everyone to hear uh let's see if this is all hooked up properly i think it is all right keyboard is hooked up i'll just prop that carefully arm sid is in the board adrian's tools and this is a pal board i don't think i really mentioned this so we're going to hear the music at a slower rate than we're normally used to hearing it and here we go [Music] all right well anyways that is working i think this machine is fixed it was just ram in the end and then a mysterious problem with the cassette port that who knows i mean i wouldn't even have noticed if it wasn't for the diagnostic cartridge right because uh without that harness i would have never tested the cassette drive to know that it didn't work whoever has this machine in the future if it's not me they may have a cassette drive hooked up to it and want to load an old tape and it wouldn't work so it's working for now why don't i go ahead before i forget and mark up here uh which one was it here q2 suspect q2 i know people don't like it when i write on the can here but believe me you could just use a little 99 ipa and this modern sharpie stuff comes immediately right off all right summary time it was my intent with this video to show that it is totally possible to use pretty inexpensive tools to do nice solid repair work on a commodore 64. the hand tech 60 oscilloscope turns out that with that open hand tech software i'm mighty impressed that open hand tick software did crash a couple times is much much better than the junky software that this thing comes with so if you have one of these absolutely stop using that included software and use openhandtech it is totally worth it and then it's really the first time in earnest that i used this solder pulse and it worked really well except for of course those ground pins on there and that's where i ended up using the braid and i think these aren't that much either though they're not as cheap as those ones you can find everywhere for like a few bucks i think this costs like 10 or 20 dollars but it really is quite a good tool in between takes while i was copying footage off the camera i did open this up and clean it out that's sort of key and why these are so much better than the cheap ones is you can easily just take them apart you just turn that and this comes right out like that there's an o-ring on here and i use some silicone grease to basically re-lubricate it inside here and just helps it create a much better seal you can find these things all over ebay aliexpress amazon as well these component testers i've had this for years and years it was a kit it works fine just get any of these cheap ones and they all work pretty much the same and it's fantastic it works really well it's great for telling it apart a pmp versus an npn transistor for example and then next up was this the really cheap multimeter i've had this thing for quite a while now and i don't know what it even costs right now but i totally recommend this or any of these really expensive ones this is a huge step up from like the two dollar ones you can get at harbor freight here in the us and probably elsewhere like on aliexpress those things work and they they're okay in a pinch but they're super junky this thing on the other hand is actually a pretty decent piece of kit considering uh the low cost of what it actually is so that's it successful fix with cheap stuff and another working commodore 64. thank you very much henning for sending these in it's just i'm slowly going through i'm fixing them and uh it's great education i think for everyone just to see the typical problems that happen on these and how you go about fixing them i'll try to put links in the description below so the various things that i have shown here like the various tools and stuff if i can find uh current links like i said check out the video on the han tech or han tech if you haven't seen that already because uh it's a decent piece of kit for 60 bucks i'm i'm mighty surprised and next up i want to thank my patrons their names just from the side of the screen if you want to become a patron you can do so at the side of the link i mean in the description as i think i mentioned the beginning of the video i do post extra stuff on my patreon page for my patrons so like i had updates about my hand surgery i will have some other surgeries coming up so they'll get the updates there but i also post uh patreon exclusive videos just like little off the cuff videos now and then that they get to see so it's kind of one of the perks and benefits from being a patron not to mention you get to see videos before they are released to the general public so if you see comments on videos on my main channel or second channel it's coming from my patrons they're the ones who get access to that ahead of time and hit that subscribe button if you haven't already all the usual youtube stuff check out my second channel a subscribe really helps me out if you haven't already subscribed to my channel and i think i've gone on for long enough so uh stay healthy stay safe and i will see you next time bye [Music]

Channel: Adrian's Digital Basement

Views: 104,004

Keywords:

Id: PyFeBAcg7ls

Length: 52min 54sec (3174 seconds)

Published: Sat Feb 12 2022