 I have come back from holiday and I come bearing a new printer. This is a Radio Shack CGP115 colour plotter. It was originally intended for use with the TRS80, hence the branding, but in fact it has a standard printer interface on the back and with luck I should be able to make it work on any hardware even USB. There were actually a ton of these made for all the different 8-bit micros. They're all based on the same Alps plotter mechanism which you can just see through the lid here. Now this is a plotter, not a printer. Plotters work by physically drawing on the paper with a pen. This one uses these nifty little ballpoint pens. They physically draw on the paper. This uses two stepper motors, one to wind the paper backwards and forwards, and one to move the printhead backwards and forwards, which is this. There's also a solenoid that makes the pen physically touch the paper or not, which moves a little bar I'll show you later. It can change pens by rotating the wheel, which is somewhat delicate. And in fact this whole thing is incredibly delicate and in fact does not work. So I got this off eBay as usual, and it is in absolutely beautiful order. It is very nearly mint out of the box. That wonderful 1980s plastic texture is intact, so I would really like to actually make this work to demonstrate it. However there are a few problems. One of which I have actually managed to fix, which was one of the impossibly delicate little plastic levers inside as broken. But the other I shall attempt to demonstrate, although it might be hard to see on the camera. So we plug it in and when we turn it on you'll be able to see it try and do some test drawing. And I'm not going to put the pens in right now because it's just going to make a mess, but so that's it homing the head. It's now trying to draw squares in each colour for each of the four different pens. And you might be able to notice that the paper wheel here, the paper roller, is not moving. Therefore it's unable to feed paper up and down. And if I press and hold the paper feed button you can hear something happen, but this thing doesn't move. In fact it seems to want to move a little, which is odd, and you notice that there is in fact a noise being made. Some of this is coming from the step motor that drives the printhead, but some of it's coming from the step motor that drives the roller. And I'm not entirely sure why. I can tell it to move the pen to the right and that's nice and smooth, back to the left again. And that tries to change heads. So that one's working all right, but there's something wrong with the paper feed mechanism. So I'm going to take this thing apart and actually find out what's wrong. I hope. Hello, me from the future here. So this project turned out to be substantially more of a mess than I was expecting. Huge numbers of things went wrong and some of them weren't even caused by my own incompetence. I ended up with five and a half hours of video and some of the work I did with the camera turned off. So rather than upload all of that lot, watch them all and make people wade through it. I think I'm going to try and edit down to the highlights and see what happens. Hey, anything's an improvement. Let's get on with it, shall we? So the first thing is to remove the paper. It uses this big roll of generic receipt paper. So put that aside and you see there's actually not a lot to it. The Alps print mechanism lives in here. The PCB with the micro controller that runs the whole thing lives in here and there's a ribbon cable running behind. So first thing is to take the lid off. So here is the actual plotter mechanism, this miniscule little thing. There is a step motor in here behind the shield and another one in here. They drive these little wheels. So this is the one that drives the printhead. You can see it moving. That is the cog that tends to fail. This is the one that drives the print roller. So I can turn this by hand and that feels nice and smooth. Does it feel nice and smooth? Felt a bit odd. I think that's fine. If the gear here had split, I think it would have bound a bit more visibly. This is the solenoid that moves the printhead back and forwards. It moves this thing here, which pushes this lever, which moves this bar. It's not actually moving the printhead at the moment because the return spring for the printhead is this tiny little foil thing here and it will only work if there is actually a pen in the printhead. So I'm not going to demonstrate that but I'll put a little draw on the roller. This was the piece that broke. I managed to fix this by fusing the two halves together using a soldering iron and I'm still amazed that seemed to have worked. Right, so here is the PCB that makes it all work. Try and get that under the camera. The actual electronics are in here under the shield. We have some power transistors. Over here we have some more transistors. These are connected to the bundle of cables that goes to the print mechanism itself. So I believe that these are the amplifier transistors that drive the stepper motor signals. I'm sorry, these are not power transistors. These will be the voltage regulators that drive that power the board off the DC supply. I see that they are both 17805s. I can go and look up to find out what it is. DIP switch is here for control. I don't have the manual but luckily the DIP switches are documented on the bottom. So now this all strikes me as extremely strange because this is a stepper motor. It gets driven forwards or backwards in steps based on the polarity of the signals you provide it. There's like four of them, I think maybe six. The bundle of wires to the stepper motor is invisible underneath the plotter mechanism. So I wouldn't expect it to drive weakly. I would expect it to jiggle madly backwards and forwards if the signalling was wrong. But it does actually seem to be moving a bit. Trying to, which is weird, is possible that nasty noise is in fact the stepper motor jiggling madly backwards and forwards. So this is what makes me think it's got nothing to do with this little cog here. I think whatever's wrong is an electronics problem caused by the something on the PCB which actually makes me cautiously optimistic about figuring out what's wrong. Now the interesting thing was when I got it it actually did work. Apart from this lever being wrong. Now I've fixed the lever and while I was testing it by just powering the thing on and off it drew its boxes. The boxes it drew became smaller and smaller and more and more rectangular until they were just lines as it began to move the the roller less and less. And that's very strange. That suggests some kind of ongoing failure which was weird. So the first thing I'm going to do is remove the plotter. I've done this so I know it works. I undo a couple of screws while this is mint in the box. The box is no longer mint because it suffered a small mishap on the way back caused by a jar of my nephew's homemade apple and clove jelly. So while the printer is fine the box itself has definitely seen better days and now smells fairly strongly of cloves which is a shame because it was a quite a nice box but that is what protective boxes are for so I'm not too sorry. This thing actually came with two complete sets of the very hard to find pens which is great. They all seem to work which is also great. I was so lucky with this thing. I thought I was paying slightly over the odds when I bought it but given what I got that I'm extremely satisfied. So here you can see the bottom of the print mechanism. Here are the bundles of wires going to the step motors which are behind the shield. You see there is no electronics. It is just completely dumb. Everything is driven from the PCB over here. So this bundle of wires is driving the step motor. These two wires here are driving the solenoid. No they're not. Ah that's interesting. These two wires are sense wires. They come from a read switch under here which is used to detect which pen is at the top. So if I rotate to here you see there is a metal bar on the top of the printhead that will be magnetized so when that's at the top it turns a bit too far and it only turns in one direction so let's go around the top again. So when that's at the top it will trigger the read switch so that the computer knows that it's looking at the top pen. That's neat. This is the solenoid and that is driven by these two wires the red and black ones. So these four bundles of wires are driving the the paper roller. All right so let me just rearrange things a little. I want this to go flat like so and let's just power it up again see what happens. There's everything on camera more or less that's working making funny noises. I don't think that worked. I don't think that was behaving the way it expects. Let me try this again without it resting on the table and push some buttons. That's not working. That's supposed to be changing the printhead. Okay the let me just power cycle it to calibrate it. It changes the printhead by moving the pen left and the little needle in there pushes the roller around a bit which is why it jiggles backwards and forwards. If I press this button it should wind. Right that was supposed to be winding all the way to the end. If I push this one again it winds all the way back. Now the other stepper motor is behaving the same way. That's very interesting. Right now the other stepper motor is gone so whatever progressive failure caused the failure of this one has now caused the failure of this one too. That's actually a little bit problematic because I was hoping to compare the signals to the two to verify what was going on. I'll tell you what it does suggest. The stepper motor is getting the right waveforms from the controllers just not at a high enough voltage so this suggests that the power supply to the stepper motors may be failing and it's been slowly degrading over time and the paper feed stepper motor needs more oomph than the stepper motor that runs the printhead which is why that continued to work for a bit. This is an RF shield intended to prevent interference to the outside world from the microcontroller inside. They don't do that anymore. Right we've got a 6805 processor, some 74ls logic and a thing. I don't know what that thing is. This will most likely have the software burnt into an internal ROM which is a shame as I have a nice shiny new EEPROM program that I was wanting to read the ROM off. Oh well these they're all different. I can go and look them up later and attempt to reverse engineer the circuit which doesn't look complicated. This metal really that was plasticized on the inside. This metal shield is clipped in. So I'm going to assume that this copper thing is ground so let's probe some stuff and see what happens. Let's try the voltage regulators first. 5 volts dead on, bit that ground, 11.5 input. 11.5 is the DC supply, 5 volts is the output that is working just fine. So let's try this one. 5.7, 0.7. Those should be identical by the looks of it. Is this the power supply regulator for the quite hot power supply regulator for the for the stepper motor and solenoid? Is it dead which is why it's not working? That would explain everything. So I got to use one of these when they were new. When I was at school my school had one of these plotters. It was used for on a BBC micro. I don't know whether it was the same brand. I said they made a bunch of these. Some of them had their own boxes. So the Commodore version, I think the Commodore was the 1520. That used the Commodore box. And I got to play with the one my school had for a good bit. Eventually the nasty nylon gears split and they threw it away. Which is a shame because these days they're easy to fix. There's something going on there with the solder mask. If only I still had a lot of the old hardware that wasn't considered special back in the day. I'd be able to repair it. I have thrown away so much stuff that I now bitterly regret because it's easy to repair and is now quite rare. Never throw anything away. You'll know you'll miss it later. The HA 17805 is a completely standard 5-volt regulator. And in my junk drawer here is a completely standard 5-volt regulator. So I think I should just remove this and find out what's and replace it and see what happens. Okay, yeah that's warm. So let's clean this up a bit. And have I managed to strip? Yes I have managed to lift a pad blast. It's ground so a bodge wire is easy. Yeah that was just too much for some afraid. That's ranking competent. I need to get better at sorting. Yeah I've actually managed to lift several pads. The ones on the top here at each end are not connected to anything. There's just a pad printed on the top of the board to anchor the component. And then actually get them all off. The bottom looks okay. I'll need to run a bodge wire from the middle to the ground. This pin here is the closest to the ground. Here's the replacement component. And yes I did check that the pinout was compatible. Maybe I even did it right. So I we soldered the bottom with nice new joints that are actually not too bad. That one wasn't. Looks okay to me. Right now bodge wire. Now before I do anything else let's just double check that that connection is going where I thought it was. So this pin here to and no it's not making contact with the ground pin of the component. Up here that pin there. You want to connect that to that. All right so let's clear some of the debris out of the way and turn off the soldering iron and plug everything back in again and power it on and see what happens. Briefly because this is not to actually think okay moment of truth of time and power. Oh interesting. Right that's doing exactly the same thing it was doing before and that is very hot. Right I think that was a complete waste of time. I think something else is wrong. Something down this end. Interesting. Very very interesting. Okay so I'm going to turn this on and measure the voltage across here and turn it off again. Work better with power. Not a lot. See this one's not warming up. This one is almost red hot. I actually forgot to try powering it up without this thing attached. That is worth doing just in case I was completely wrong about what the regulators do. So I don't think this is going to work again so given that this is probably a fairly high current device I think I may let it this off. Okay well I managed to get the component off eventually. I may have fried the component by getting it too hot and I have managed to lift another track another pad rather so the board's not in great shape but it's now off so let's just do a few quick multimeter tests. All right the next step is that now that's disconnected let's power on and see what we see because if I see the stepper motors try to move I know that I'm wrong about what that thing does and what we see. Great fantastic right that voltage regulator its job was to move this. That's what it's for because that's a big coil it will use lots of power there for it has a regulator of its own. Okay I'm going to guess that this whole complex of transistors here is for driving the AC supply for this. I am not going to bother trying to decode it in any great detail. I know that bit worked at least it did until I broke the voltage regulator so the output from the output to the stepper motors comes from this LB1257. This is then connected to the GPIO pins of the processor so an LB1257 is a driver circuit it's just an array of transistor networks in fact the data sheet says it is specifically designed for driving XY plotters and it names the ALPS plotter mechanism specifically so it's dead simple voltages come in voltages come out the 9 and 10 here are ground and power respectively so we've got ground which is the ground plane power is this is connected to this which is a via oh oh possibly the other way around right we've got so we've got a big fat track here we've got a big fat track here but they're not connected together this does kind of suggest that something's wrong here or possibly here maybe this is gone really okay well can have to do some more debugging so what we can do and actually this is what I thought it was originally going to do before I got into the red herring with the voltage regulator is I can measure the waveforms in and out using an oscilloscope and I just spent some time setting up my oscilloscope so I can record it and that's now working so we can do that and see what comes out I hope the stepper motors haven't died I mean is it such a simple device what could possibly have gone wrong yeah okay well I'm going to take a break and get back to this so you will see me in a few seconds but for me it will be probably a couple of days so wish me luck I suppose well let's get back onto this thing so a quick summary for all you people who haven't been paying attention which mainly means me because the last my touch this was yesterday this is not working it is not it is failing to feed the two stepper motors in the plotter correctly yesterday I attempted to fix it I completely misdiagnosed what was going wrong removed probably a perfectly functioning regulator replaced it with another regulator removed that regulator again in the process slightly mangled the board lifting a couple of pads so that turns out not to be the problem at all so what I'm going to do now is use my oscilloscope to try and get more information about what's going on and I'm not going to touch the board until I have more idea of what and why is happening so what I'm going to do is power the thing up and make it do things and then watch the waveforms that are produced and then driven to the stepper motors and the way this works is the processor here generates the signals via its GPIO pins here which get routed to this chip which is a transistor array used for driving stepper motors and the outputs of this then go to the feed to the stepper motors so hopefully I should be able to see something interesting using the oscilloscope and I have managed to hook the oscilloscope up to a camera so through the magic of video editing you might be able to see what I'm doing so let me just go and get that all turned on and we'll see how it works okay here hopefully it should be and it's probably going to go up there somewhere depending on exactly what I'm doing the oscilloscope is a extremely old and beautiful tectonic 7603 weighing about 13 kilos it is completely analog with no computer elements whatsoever those numbers you can see at the top of the oscilloscope are drawn using complex analog logic rather than anything resembling a computer it is beautiful I love it I can barely lift it and I'm using enormous amounts of image processing on the picture you're seeing because it suffers badly from screen reflection and it's got a bit of a tendency to reflect anything you can see including myself so you should be hopefully seeing a basic image of what's on the oscilloscope and here's the oscilloscope probe it's got a thing single probe thing and a earth connection okay so I should be able to press this button right it should be driving this step motion but I can see it going around a little just not enough to actually do anything so now what I want to do is to do that while touching these contacts and see what comes out this is going to be more hands than I usually have that one interesting so it does seem to be glitching when I press the button let's try this one similar this one this one oops this one there we go right so I bet the four down this end of the other motor yeah the reason why I see it glitching is you see that the head moves slight sideways slightly what that's supposed to be doing is moving the pen onto a little metal shield which you can't see in there so that when the paper moves the pen doesn't get doesn't accidentally touch the paper and draw so that's producing a waveform let me just adjust the time base a bit interesting looking waveform that I wasn't really expecting okay let's try the other's end of the thing so what you were seeing there was the output that was the result after amplification so let's touch it here do the same thing this is what the processor is emitting interesting so that's a nice TTL signal well what we see at this end is not a nice TTL signal it's floating high when it's not connected so that's that is two and a half intervals which is exactly what I expect but that's going almost to six volts okay that's not actually what I expect to come out of this thing so either the driver chip is bad or there's something wrong with the stepper motors which is interesting so that hasn't got us anywhere closer to figuring out what's gone wrong with the thing other than the waveform output weird so when I started taking it apart this motor worked fine but this one had failed and I was rather hoping that that would continue because then I could compare signals from the two but that's really not really not how it worked out okay so let's do some signal tracing I suppose here's the bottom of the plotter you can see the the wires just come into pads here and are routed directly to the stepper motor via these five five six six signals and notice that two of them are black and the other four are multi-colored which makes me oh yeah I see that these two black ones are connected together right this is this is presumably ground this one that's the middle wire it's it's white what a fantastic color for a ground wire if it's ground so let's just move these screws out over a short anything plug it in power it up again and let's check that voltage okay so we know that the shield is is zero volts and this is my ground and it says five volts right okay it's an active low device so again where has this got us not sure I think that driver chip is maybe bad we could try replacing it I can get a replacement for about three to four francs plus shipping and it's just a matter of removing the old ones sticking in a socket and plugging it in I mean that's probably worth doing if I'm going to do that then I might as well get a couple of modern switching regulators and replace these this one on the left is already like pretty mangled the advantage of switching regulators they produce far less heat so I wouldn't need the massive heat sink anymore and uh there'd be much less stress on the board plus less worries about uh thing overheating if I don't get the thermal paste right but other than that I'm not really sure I mean the one bit that was working fine was the bit that I broke by removing this regulator which is the step the solenoid now that's the bit I would expect to fail the solenoid is this this just clicks this backwards and forwards so I think that's probably the next thing to do if the stepper motors themselves are faulty I'm essentially stuffed so I think I'm probably just going to have to call a break and come back with some parts that have been ordered oh well see you next time then so I did a bit of research and I discovered that the waveform I observed on the oscilloscope is in fact completely normal the mysterious hook at the beginning is caused by as far as I can tell back emf from the rotating motor it induces a voltage in the coils that you then feed back and you see it in the graphs so that's fine this suggests that the driver circuit is completely normal and operating correctly so all that work I did on the board including kind of damaging it was a complete waste of time good to know the person I talked to also suggested some things I really should have thought of myself which is simply dry lubricating the plotter now I didn't actually do this because everything feels quite smooth when I rotate it by hand but it occurs to me that the motors being very small and being operated at 100 hertz are probably very low torque so maybe it's just that the 40 year old lubricant in the bearings has slowly gummed up over time and it worked for a bit and then something got sticky and it stopped working so let's give that a go I have two lubricants we have contact flon this is safe to use on nylon which I'm going to use for most of the gears and the the metal track that this thing moves on you can see the bar in there so I also have a light lubricating oil that I'm going to use for some of the bearings unless contact flon will do fine and interesting this does not actually fit into the nozzle well I wasn't going to spray it directly onto the thing anyway it's going to spray a little into this yeah it's just it there's put a bit of contact flon into the pot and then I'm going to use this syringe this is not a hypodermic syringe it is a inquiry filling syringe it's got a blunt tip it sucks some into the syringe yeah we've got a little in there because this will then allow me to put it very precisely onto the under this so we add some here we're in there here there we go right and I can actually see it flow down the inside of the the head joint okay so this does not feel any different I think that's smoother right there is one other joint I need to do which is there's a little tiny metal roller in here which moves against the pen load bar and that rubs against metal this is probably the wrong lubricant for this but it'll do there looks lubricant everywhere is that smoother maybe oh now let's power it on to see what happens no it's still not working okay so that wasn't a particular success it may take time to uh like wear in in fact when I moved this it is a bit gritty to be honest and I can see a dark patch on the yeah the grating actually corresponds to one rotation of the motor bearing so possibly the motor itself needs lubricating you see there are these brass bearings four and a half of the motor so let's clean this off get rid of the rest of the lubricant put that aside and let's get out the other one lube oil it is a light lubricant used for lubricating light things and do the same thing again so there that's not better hmm so what about this side uh this side is terrifying so the print head is moved with this metal wire which loops over this pulley round this round uh yeah round another pulley and back down so and there's another pulley this side as well and if I were to undo anything I'm quite certain that I would never get it back together again one end of the wire is attached in here to this screw which I'm not going to undo but there's a spring clip on this big wheel and I can take that off and then lift this thing off and probably all the wires would come loose and I never get it back on again so what is wrong with the thing so does not appear to be lubrication I think something mechanical has actually gone wrong I think that the more I retake this the wait a minute wait a minute there are two gears here and they're not lined up that's never going to work what is this doing is that a is this a clutch right because the two gear wheels both move against this piece here but if they're not lined up then they're just going to grate are they different sizes no they're the same size just one of them is skew I think I am going to have to take this spring clip off oh well all right uh there it is I can feel the wires tensioning as I pull I don't want to do this I think the what I think the wires are captured by the these two pulleys not stretching the spring and the spring up to the other end the spring stretching here we go what is this what's going on here I think one of these was warped oh that's sprung it is a clutch because both sets of gears are meshed with this thing so it won't they cannot rotate independently but there it's split they can rotate like this so maybe do I just set it and push it on like so yeah I don't understand what's going on here I am going to actually reassemble this and I think seek more assistance quick update I discovered something so I tried removing this gear and then I tried making the motor drive by pressing the button button that button and you can see that it does indeed go around but if I rest my finger on it the motor stalls and has allowed grating noise you may not be able to hear on the microphone there is very very little torque available I'm not pressing hard on the motor at all if there's just not enough torque there to operate the mechanism so it's possible that this is all working absolutely fine and there's nothing mechanically wrong it's just the motors have failed which is weird I hear a rumor that occasionally the head of the rotor which is a permanent magnet can become demagnetized I have no idea so I suppose I could replace the motors this is not getting any easier to rotate I have to say so let's just try and remove the motor and see what happens so this is a simple crosshead so I need a small phillips bit all this work don't want to knock any of the gears or heaven forbid spring clip out of the way okay that shifted it I should be able to go back to the small screwdriver and there goes the motor incidentally I figured out what the clicking noise is the thing retakes is there are fine teeth in the roller here that grip the paper and it's bumping over this thing so that's actually all as intended there's a brass baffle thing bearing or baffle bearing here I bet there's one sticking out the other side and that's what you can see here you see it does move a little but the position is such that I the only way to get it out is to move it this way to get the bearing out from underneath this but I can't do that because it's wedged under there so insert the screwdriver and just leave a yes yes I do wow what's that okay so here is a motor I don't know what this thing that came off is it's small and bronze still go around find that button this one it goes around still stalls and I put my finger on it well I can replace the motors if I can find a replacement I have no idea what this is the only labeling is h4 h4 3 4 here which is not very helpful stepper motors are supposed to have a fairly standard set of mounts but I I bet this is not one my best bet probably to find another plotter and replace the mechanism wholesale at least the gearing all seems to work so I would have spare part including the the nasty nylon bearing by bearing gear this is the gear that tends to split so having spares is a good thing I reckon this is pointing fairly conclusively at motor failure I have never heard of this happening with one of these particularly both motors at the same time okay so I think this motor is faulty it feels okay sort of but it's PSP generating very little torque so let us try to actually measure how much torque it's emitting now I have a rather dubious way of doing this let's stick some blue tack on power the thing up wait for it to stop moving this is it doing the self-test drawing and let us insert this nail now hopefully the blue tack is stiff enough to hold it to the gear wheel let's try pushing the button I need to be upright that's not great it's too off-center you can feel it wobbling let's try that that's better okay so put on the table and press the button the table yeah put it somewhere you can see it the table stops it moving and it's going to apply a little bit of torque a little bit of force to the table so I'm going to try and measure it using this set of kitchen scales this is kitchen science kind of make it so you can see the screen just it's like going to be super super inaccurate but may give us a number so tear it down to zero it's a fairly sensitive here press button it doesn't seem to generate enough to measure so it doesn't really that doesn't give us any numbers but does at least does at least give a lower bound I suppose okay well what's the smallest amount that this thing can measure I've got some of these five grams let's try it over here still five grams so I know it's less than five grams let's try one of these nothing this is a bit bigger two grams two grams right well we know now that at a distance of call it 20 millimeters it's producing force of less than one gram of force so you measure torque in Newton meters so the number of Newtons is so I'm getting a torque of less than 0.0004 Newton meters I think that is not enough and it is broken so let's try the other piece of debugging I was wanting to try I still haven't entirely ruled out electrical failures I still don't like that that feels way too stiff but it could be problems with cables and things so let's unplug the motor okay it's set to five volts so here is the motor as I want we're going to set this to five volts and then energize one of the coils it moves a little I'm just so moving it randomly by energizing the coils but you can see it turn now kind of what I want to do is to permanently energize it and then try and turn it by hand see how much resistance there is storing 20 milliamps that rotates freely that requires a bit of force to push but not a lot so I think this is probably generating the same amount of torque which is interesting well that hasn't actually got me any further but if I can replace these motors with similar ones then maybe I can get this thing working more likely I will have to find a whole new plotter assembly and install it into the chassis that shouldn't be too hard these things do hit eBay and many of them are broken because the spindle gears are dead but I have two working ones here that I can pull off yes interesting oh well so more news I took the x-axis motor off in order to try lubricating it I wanted to get right into the bearings only to discover that it grates when it goes round so this motor is basically dead I think they've both failed probably for the same reason whatever that is I am just going to have to get new motors or more realistically an entire new plotter mechanism well damn still it was 40 years old so look what I've managed to do so if I press this one yep so it is actually now moving correctly more or less correctly see that's no longer changing pens so what I did was I pushed the axle not with that screwdriver didn't I pushed the axle in this way and the motor started working again so the shaft is slipping on the in the bearing you can actually see that happened a little so well let's do the calibration thing and now it fails and I push it in it works again now it's failing let's try it like that yep it's trying to draw squares the other motor actually behaves similarly but weirdly in reverse so I need to push against this side this end and suddenly it starts working so what's happened is the motor bearings have died and well something in the motor bearings have died this is actual genuine motor failure and I can try and get replacement motors but these are so old and so unlabeled that the only realistic way to do this is to find another plotter module and cannibalize it um so I wonder if I can bodge something for example something to keep the shafts keep pressure on the shafts this one is fairly straightforward because something that just pushes against both sides or rather pushes against this side I don't want to push against this one you know oh dear well at least I know that the bulk of the plotter works and I can attempt to fix the board and repair the damage I did to it but what am I going to do about this I was all set to just give up you see so what can I put here I want something that applies a bit of pressure to the shaft but not too much and not too much friction is there any way to fasten anything to not really this shield is supposed to go over it but it's just clips on it's not actually connected to anything so I can't screw anything on there I'll have to go away and think about that I think so I have a cunning plan to bodge this thing into submission and hopefully make it work again now uh the plotter assembly I reckon is a write-off ideally I would replace the motors I don't think this is going to be possible because I'm not going to be able to find replacement motors that will fit in the unlikely event which I can actually installing them will be a nightmare I am in fact much more likely to find a complete replacement plotter assembly which I shall just swap into my rather good-looking chassis so I am going to make some modifications to the plotter now the problem is that the bearings inside the motor appear to have collapsed and they only work properly if I apply force to the shaft this one I need to push the push the shaft in this way to make it work this one I need to push the shaft in this way to make it work and I have a cunning plan for doing this so for this one I need to whoops let's do not do that I need to push it this way now I am going to use this bolt to do this the end of the bolt is slightly concave so it will very nicely go on the end of the metal shaft a spot of oil there and I'll end up with a fairly low friction contact surface it won't slide off and I can adjust the pressure easily enough by moving the bolt in and out but what's the bolt going to fasten to the answer is this 20 millimeter washer I am going to mount this here inside this empty space and the bolt is going to go into the washer with a nut on the inside so that I can adjust the position by rotating the washer how am I going to fasten the washer in epoxy this side is easier I need to push the shaft in this way now a similar mechanism would involve pushing against the side of the case and I don't want to modify the case because the case is in really good condition so instead I'm going to use this this is actually a key puller I made by crudely bending a paperclip and it turns out that I can just do this and do this rather get into position like so and that provides enough force to the motor I have actually tried it so that bit's relatively straightforward I'm actually going to make a new one that's more custom design apart from anything else I want to keep the key puller for pulling keys with now chances are that this will not actually make this work any length of time I think that the bearings are just going to continue to fail so it will work for a bit and then die however one of the things I will be able to do is to measure stuff like the step size of the motor which I don't know yet which will help me find replacement motors if such things exist so let us get started first thing is the washer so the bolt is going to go here the washer needed to go about here and I'm going to use a little bit of blue tack to just temporarily hold the washer in place get that rough view vertical and more or less aligned so the inside of this is nicely rough so the epoxy should work fine and this will go in like so so that looks fine so next step just do a bit of swabbing with some IPA just to clean up the metal a bit because I've had my greasy fingers all over it okay I've had my greasy fingers all over it down here and that's all I need so let me just put this away before I start hallucinating while the epoxy epoxy while the IPA evaporates I am going to get the epoxy ready for which I need my two part epoxy this is ancient and you fairly horrible yeah I can actually tell the difference in the plastic and the half congealed rotten epoxy but we squirt some quite a lot actually into the pot back on hopefully the right way around like so so we now have unstirred epoxy I need a match that is now out because you always use matchsticks to serve epoxy those are the rules mix this together good and hard like so and onto my washer there's probably going to be quite a lot of force involved here so I'm actually going to put the flat bit on I made a lot of epoxy because it's hard to make small quantities of epoxy okay that should probably do this bit I'm not going to do now I need to wait for the epoxy to cure but there is something that we can do while we're waiting which I'm going to do now and that's to deal with the power supply issue because I mistakenly removed one of the regulators from the board and made a bit of a hatch of the board I lifted some pads and the other regulator is also a linear regulator and I don't like these they get really hot I don't like the fact it needs heat sink just running this thing on the bench without the heat sink attached this has been getting pretty warm so I'm just going to replace them both with modern buck regulators here is one of them it's a cheaper Chinese adjustable power supply module you feed voltage in outcomes voltage at a particular voltage you can adjust the voltage by rotating this pot they're cheap they're cheerful they work reasonably well so I'm going to use a couple of these this board uses two regulators this is the one for the electronics and stepper motors this is the one for the the head the the pen solenoid because that uses tons of current so I'm going to stick with the same plan and use two different power supplies so because I'm going to replace them with the buck regulators I don't need the you can't see it because it's not on the camera the big chunky heat sink anymore so I'm going to take that off it just screws on which is nice and I have an ulterior motive for this which is it gives me two screws and two screw mounts ideally suited for placing the regulators so here's the first one in is this way out is this way so we're going to put in towards the back of the case here very straightforward and the second one needs to be the same way around all right and they're nice and firmly attached so the next step is to actually wire things up and this involves working on the board for a bit so the first thing is I want to remove the second buck second linear regulator and this time I'm going to do it properly as opposed to the first one so how do you get these things off properly well what you do is you sacrifice the component and just chop it off because this means you now have three separate legs which you can desolder one at a time where ease visible behind the shield comes out easily enough so this now gives me three holes which is v out v in two do two different v outs this is the one for the power supply this is the one for the solenoid okay first is to remove this patch wires I don't need this anymore and it's just kind of getting in the way that's the three things I need to hook up are grounds see I was hoping that these would be connected together the the electronics ground and the solenoid ground but apparently they are not I thought they were from last time so they are going to need separate grounds one to each power supply the other thing I need to hook up is of course the five hole output one of which goes to the electronics the other which goes to the solenoid and the third thing is the v in and these should be connected together they are connected together yeah and that's just straight to the the input from the power supply yeah and in fact this is the input here it's this big track which goes to here to here they're both connected directly together and one ends were smoothing capacitor so the two grounds are a little bit more interesting because one of them is on the top here and the other is the other is connected directly to this big copper pad so let us attempt to clean out some of these holes so I'm going to need to put wires in them right so we have three holes for the electronics supply now what am I going to do with the solenoid supply I managed to lift a pad for this is v out this big chunky track here connects down to the power transistors that generate the solenoid output so I think can I put a wire through there I think I need to clean that to be honest or I can just sort the wire on the bottom there's also I need ground but the ground track is at the top so I will in fact wait a minute what is this connected to there is just the one track it runs from this hole to this diode and nowhere else ah right right yeah I think what this is doing is it's protecting the electronics from big induced reverse current spikes so the solenoid has a big coil in it and it stores a lot of energy and when when you stop energizing the coil you get that energy back out again as reverse voltage and I think that this is protecting allows that the two grounds be connected together but it prevents current flowing backwards from the solenoid into the electronics so I think I do need to connect the ground for the solenoids supply to here somehow the other thing I would need to connect is v in to the solenoid power supply but I can actually just common that from this one I would rather use a separate wire to be honest actually I'm going to play this smart rather than solder the wires directly to the board I am going to attach these headers because that will then allow me to actually have a connector because the power supplies are screwed to the other side of the chassis ah every time I want to take this apart I'll end up with two sections connected to each other by soldered wires so these headers will actually help there so I just need the appropriate size of one of these to get them in line it get them lined up okay that has them all held in place so now I just need to solder the bottoms on so this is a joint let me just that managed to push the component a bit out of the board so the top end is spiky the reason for plugging this on is to get all the pins lined up with each other it's also a cheap and easy way to hold the header in the board while you work on it okay let's do this one next double check positioning not brilliant this one is actually skew so let's see if I can deal with that that looks better does the plug still go on no it doesn't because they no longer line up they are not the two sets of connectors are in fact not a tenth of an inch apart so no this plug would never have gone on gone on that's why it wasn't working right never mind they're both reasonably neat now so let's solder that other joints done that one try and get access under the shield to this one and this one luckily the shield appears to be heatproof plastic okay now this middle pin here is the the solenoids ground wire and it's not connected to anything on the inside because I lifted the pad and the track is broken so I'm going to need that patch wire back again now I actually screwed up I mean one of the many times I screwed up with this last time I soldered this on because the ends of the wire were too long and I ended up shorting the ground to power which is why the first attempt linear regulator it overheated and didn't work so to be careful that reason that's not solid enough I'm going to need to put some solder on this pin because there's no pad to connect the pin to the solder just blobs up around the pin and it's really unpleasant reasonably solid yes and then this can go around and connect to this diode okay so we now have some header connections let's beep that out and just make sure that it does appear to be connected to sensible things so this is ground we know it's ground so this should be connected to the middle pin yes but not this one this has its own ground the five volt line we know is available down here on the driver on the motor driver and that should be connected to here yes but not to this one that's correct we know that these two pins should be connected together because they are V in yes and we know that this is the solenoid ground and that should be connected to this yes this is the solenoid power line and it comes down here to the top end of this power transistor so let's check that this is the solenoid here is the power transistor correct all right so we now have headers on the board next step is to come up with a couple of connectors for which plug on to these this involves crimping things for these things i have a kit somewhere i will go and fetch that or alternatively i have a million of these i can just pull off set to three so we have red orange and yellow here we have a matching set of red orange and yellow they've got connectors on one end and pins on the other all very easy no need to crimp anything so this will go on the board we're going to use red for for the actual v out orange for ground and yellow for and yellow for v in they're all the wrong colors but that's what i've got so i'm going to go on there we go and the other side use the same colors same order so okay now we can rotate the shield back into position and we can even stick these things back on those things these plastic things in to hold it in place one two and out three there's one and that side of things is done oh yeah i want to put the the rf box back on where i put it here it is you can't see the rest of my workbench which is a blessing really on somewhere all done so this there will then get mounted inside the board like i have made a very stupid mistake yep i have i've put these connectors on the wrong side of the board go me so the correct thing to do would be to desolder these but i really don't think i'm up to doing that without ruining the board so can we just bend them instead the answer is yes poorly yeah okay right that should work so now the board will go in like this it's still not brilliant because there's not a lot of clearance here but it should work anyway the next bit i do is on these so unplug all this lot and put this aside because we now need to solder these pins on to the board now the colors we i had decided on was red for v out orange for ground and yellow for v in so we have in minus in plus so these two just go on like this and red comes up here these are commons together the two in minus out minus all right now this one i've actually used before in a previous project so it's got this stuff all over it it means the pads are already tinned but anyway quite a lot of gunk on that one actually actually i think i have a spare i may use a fresh one no i don't this one already has pins attached to it uh just while the soldering iron heats up i have a number of cheaper modules these are voltage converters 3.3 to 5.5 this is a 5 volt 3.3 volt unit that's supposed to plug on to breadboard except turns out that my breadboard is a different size it doesn't work these are 3.3 volts uh regulators uh annoyingly these are not the same pinout at the 7.805 and these are voltage inverters you feed in say ground and three volts and what you get out is uh minus three volts and ground except the only time i ever used one it ended up destroying an elderly computer of mine so either i don't know how to work it right or it's faulty i'm not going to touch them again for a while they'll drift like mad but that should work okay so i suppose the next thing is the smoke test which is to plug in the 5 volt lines and then turn it on and see what happens so this head is now moving fine however the solenoid is not working why is the solenoid not working if i started up with paper feed press it actually does more of a test this is actually writing text solenoid is not working why is the solenoid not working uh isn't there a thing missing here wait what was that missing all along well it thinks it's generating five volts i believe that's just a smoothing capacitor well the epoxy is decently solid so it'll take a little bit to cure so i will just actually move the blue tech yeah the blue tech is slightly epoxied to the chassis and just do this thing so that goes on like this like so and i just got a bit of attention feeling that's not actually going to work i think the cup shape on the bottom of the bolt is not actually quite cuppy enough yeah that ain't working uh this this principle will work i just need something else under there some kind of uh low friction thing so let's just wind that back a bit just get that to the way and let's have another look at the solenoid let's try that what the hell won't these supposed to be attached to something sheared off the flipping board i can actually see you brown purple gray okay i suppose we better put those back on then so i should be uh clearing the holes and putting the wires through but i'm not going to because that will involve way too much assembly in the board just putting nice clean solder blobs on the broken pads like so i do not believe this is the cause of my solenoid problem as i can always hope okay and power on see what it does no solenoid it's doing something so i would expect the solenoid to be producing spikes in each direction in order to uh either flip the solenoid flip the lever in one direction or the other you see it actually stays when i put when i push it and annoyingly this bit actually worked before i started taking things apart one other possible thing that's worth looking into is how much current that solenoid is drawing does my cheapo uh buck regulator is it actually capable of drawing that much current of generating that much current rather that's easily measurable i simply need to look at the output voltage to see whether it drops is that five volts oh that's not smooth yeah uh you know what i'm going to take this off the board and replace it with the one with the smoothing capacitor i have no idea why this is missing with smoothing capacitor but the spikiness of output voltage makes me think there are bad things happening so let us do that and power on oh that moved it's not moving again it's going out but it's not going in i think this may not be generating enough oomph i have to try the the scope again i don't want to have to go back to that bloody linear regulator given all the effort i took to replace these things i was completely rock solid so if the if this was running out of voltage i'd expect to see a voltage drop here but i'm not so that doesn't appear to be what's happening so the other thing is to check the uh the output voltage from the that's actually going to the solenoid let's do that again first it draws the boxes right now it's okay you see lots pulses down to zero and if i push this it's not popping back either the driver logic driver transistors here that are trying to make step motor work the solenoid work have failed or something in the solenoid has failed i don't know what but frankly i've been at these hours today so i am going to give it a break and try and figure out more they're so nearly working too update i have news i tried this this is the original regulator that took off the board wired in to the board and so what's what happens let me just plug this thing right on get the clicking noise and this is it drawing text so this regulator works this one does not but this was generating rock solid five volts so why is this working but this one wasn't let's wait for this to finish okay so i suppose it might not have been this might not actually be a five volt regulator that seems extremely dubious to me i mean it does say 7805 right there on the label but let's actually check that so this should be ground and this should be put that way you can see it five volts so i am deeply confused well i suppose i could always put the regulator back then i would need a heatsink i mean it's not getting warm but it's not drawing much this solenoid draws an amp at five volts but the the whole transistor logic here the whole purpose of that appears to be that it only energizes the solenoid impulses to flip it from one direction to another okay i've done a bit more work on this so if this thing wants a linear regulator to run the solenoid it can get a linear regulator to run the solenoid and so i have done this this is the linear regulator that i pulled off the board this is not the the original one for the pen solenoid which is this one it's the one that ran the electronics because i snipped this one off the board and therefore it's less likely to be overheated and wrenched around and in lieu of the big metal heatsink i have attached this little aluminium thing now this should be safe because the head seems to draw lots of current in very very short bursts the overall load on the regulator is quite small it's the other one which seems to use most of the power because the other regulator is powering both the electronics and the stepper motors which run continuously so this seems to be okay and what i've done is i've soldered some pin sockets onto it so it just pushes on to my socket here which is reasonably robust and means i don't have to think fiddled with the board anymore i really don't want to get to take those things off i might break the board irrecoverably and what i've done here for the the Y stepper motor is i'm using this little piece of plastic this is actually the top of a milk bottle because there's this little indentation here which fits neatly over the axle and is self-centering so this then provides a bit of pressure and it does actually seem to work it's bodgy as hell and i'm not convinced it will continue to work but it will do for now so i can demonstrate this working so this is all moving normally so we should actually now have enough in place to actually give this a try okay so let's pass power off power on again and let's see what happens right it's it drew something but there's just not quite enough to make the paper feed work yeah you see that's too much but the alternative appears to be too little let's try that now after a square that's your sort of rectangle another rectangle another rectangle yeah that's not worked so well this is it drawing text so it's working it's just not working very well i can see small text here about the right sort of shape it hasn't fed correctly onto the next line yeah i'm going to have to come up with a different solution for this so a bit of a marathon later i haven't all put back together i actually made several changes including replacing the paper clip that was holding the tension on the x motor mainly because i lost the black one there's a bit more tension now and it does seem to be working rather better you can see it there so i did from shape on this side i drilled a cavity in the end of the bolt and added this rubber washer to give more consistent pressure that seems to be working better as well you can see that i put it all back in the case another thing i made was that i actually managed to solder the brown and gray wires onto the motherboard the wrong way round so these two buttons were swapped that is now fixed and it is now working a lot better as i shall hopefully demonstrate so hold down the paper feed button to do the self test turn on so it draws the black square it draws a blue square it draws a green square it does not draw a red square there so it has you can see that uh it's drawn squares are actually square um it's drawn black text blue text the green text is not very good the red text is missing completely and then we're back to black and blue again this is the 40 column mode so the characters are twice as big as they are in 80 column mode i don't know what's wrong with the red pen it just doesn't seem to be drawing on the paper when it's in the printhead if i take the pen out and scribble it on the paper it works fine so something there needs adjusting to try and uh increase the pressure on the paper the same is probably true with the blue pen the uh the pressure on the paper is actually done by a combination of a spring in here and a minuscule foil thing which the end of the pen slips into so possibly something there needs tinkering with i don't actually know suspect ones which is red flip this forwards and i push the head across it should leave a line but it doesn't so i'm not sure what's going on there and i don't know what i can adjust i don't know if you can make it out but there is a red spot here from where the pen touched so it is it's making a bit of a mark so it's now in a mostly good working order apart from that i don't know how long it will continue in working order so i think i think i will wrap this up as it's enormously long i have managed to barge the thing back into working order for at least black which is fine to be honest i'll continue to tinker a bit to see if i can make the other pens work possibly they're just old not particularly good anymore they are like all 40 years old and i think i shall attempt to set up a proper demo video of this thing which is not part of the repair video anyway well that took way more hours than i was expecting and now i need to try and figure out some way to edit this all into shape to stick this on youtube that will be exciting and probably the hardest part of the entire job anyway i hope you enjoyed this video please let me know what you think in the comments