I never thought the left-for-dead TRS-80 Model II would work again

well hello everyone and welcome back to adrian's digital basement yes if you can believe it what you see here on the bench is the trs-80 model 2 which i revived from basically left out in the trash on a whole bunch of videos back at the end of last year all the way into i think january of 2022. let me just do a little bit of a recap really quick in case you don't quite remember all of those video series but of course uh i'll have links in the description below so you can go watch those all in fact i myself had to watch some of those just to refresh my own memory on what i did on this machine because it's it's been quite a while and i've worked on a lot of machines since so when i found this machine it was given to me it was rescue it had the top off for a long time inside was tons of debris like leaves and bugs and stuff like that so i slowly took this thing apart which was interesting figuring out how to do that because there was a lot of rust and stuff and i slowly tested each of the parts like the crt and the analog board for it then the power supply then the disk drive here i hooked it up to a pc and accessed it in dos and then finally we got to all the cards and i have one right here this is the floppy drive controller a lot of these had bad corrosion on them just because of all the moisture that had gotten into this machine over the years well in that last video i thought i had this machine all back together it's basically how it was looking right here looking cosmetically pretty good and it seemed to be operating normally as well but after hooking up a gotek drive and booting into the trs-80 model 2 diagnostic disc this is an official disc image from radio shack i ran through the diagnostics and then on the reboot the computer stopped really working properly so in this video i'm going to try to figure out what exactly is going wrong with this machine and see if i can get it working now i'm recording this intro in august of 2022 when of course this machine's been sitting for quite a while and that's really because i had a problem with my right shoulder which prevented me from really lifting incredibly heavy machines like this so this computer sat on the back burner for a while i also recently released a video on the second channel where i showed a diagnostic rom that me and my friend david and frank came up with for this machine originally yes it's the same diagnostic rom for the model 1 and 3 but we ported it to run on the model 2. architecturally it's similar but not really it's just a z80 processor like the other machines but this thing supports up to 512k of memory the video controller is totally different he's a crtc stuff like that so there are quite a number of differences but david did the majority of the work and got that diagnostic rom running so i recommend you check out that video on the second channel first if you haven't seen it because i'm going to be using that diagnostic rom in this machine in this video to try to figure out what's wrong that's one more thing i want to mention before we finally get started with this video and is that most of the troubleshooting that you're gonna see around the keyboard problem and stuff like that was all done back in i think december of last year 2021 when i was actually doing all that the bulk of the work on this machine the problem is my shoulder thing crept up so i decided to not finish that video and not release it either and just shell this machine temporarily so first off we're going to start with me showing the problem with this machine just as a refresher that will be from the last video part 8 and then i will jump into the troubleshooting and i hope it's not done yet but i hope to get this machine working once again because it's so cool and i want to at least play around with it i felt a little ripped off after all that work i did on all those eight previous parts to never actually really play around with this computer at all because it just started misbehaving so quickly so anyhow without further ado let's get right to it [Music] now while we wait for this to boot people had mentioned that i need to make sure i take this disk out of the drive or at least open it um look at that it's not working still all right so this one is working fine uh whoa wait a second what's going on here no it's not working fine well this is very strange keyboard is having all sorts of issues like some of the keys give you the right key but like eight gives me a nine nine gives me a nine seven gives me a seven six gives me a six it's like one of the bits is bad or something let's try a reset again and look it just hung here booting up and this is off the original disc now what's happening with this machine oh okay um okay and it's acting very strange in a different way all right i'm turning the computer off i'm gonna go inside the machine and i'm going to manipulate the cables i'm just checking all the connections kind of wiggling the cards in in the saw in the slots maybe there's just a bad contact or something it's definitely kind of weird that things were working fine not to mention i had the gotek connected as it is right now yesterday and i was doing testing and stuff with the diagnostic disk and i didn't have any problems whatsoever okay it came up here all right but the keyboard is still having issues it types perfectly but instead of a d i get the e for instance and i get e for e and you know that doesn't work for f i get a g h is an i hmm i have never had a diagnostic routine actually cause damage now i don't really feel that that happened here it's just coincidence that something failed in the keyboard at the same time but pretty unusual that that happened all of a sudden just like that all right i'm over here in the technical manual for the machine now pretty much concluded this cannot be a problem with the keyboard this is a problem in the computer itself okay we look at the keyboard layout here these are the codes that it sends so over here on the number keypad 1 is 31. 2 is 32. and same thing over here so that's 31 32 i think this is like shift control shift and control like sends different codes but this is the regular code so when i push one i got a one but when i push two i got a three and pushing three i also got a three now i had that problem with the one two and three for both the number keypad and these keys here so this was key number two three and four where key number three was having an issue now this use this issue was happening all over the keyboard but if we look at three and i was also having a problem over here on key 56 it's hard to read that so there's key number 56 and there is key number three they're not even on the same lines at all on the matrix i really don't think this is a problem with the keyboard this has to be a problem with the card in the computer that receives the the signals from the keyboard now it's the video board that handles the keyboard now one thing i'm kind of interested about is it's 24 by 80 lines normally but it also has a 40 column mode as well it's kind of interesting but here is the block diagram of that card so here's the address control and data bus on the main backplane this stuff is all for video right here but right here is keyboard control and there is the keyboard itself communicating to that controller chip through these lines now i'm hoping that this keyboard control on the pcb even though it's just one square here has some normal chips on it and it's not some kind of custom microprocessor that might have failed because then we're in trouble on this page it shows the signaling that happens between the keyboard and the computer so data and clock comes from the keyboard and there's a busy signal so this is very typical for the way keyboards work on all pcs so i have two tabs open so i can jump between so okay so the data comes in it goes to this ic here which is an ls74 okay and it really just goes through a couple of these ics here i wouldn't be surprised if the ls244 is let's see is this the data bus yeah this is the data bus all right so it's connected right to the data bus so right away i am suspecting that this ic right here this ls 164 is the problem and indeed this is a parallel out shift register so it takes that signal that's coming from the keyboard which is probably just ascii like you push three and it sends a 3 and this just converts that serial 3 which is a string of bits with a clock signal into a parallel 3 the ascii representation of it now if we look in this column here and we look at the difference between two and three the only difference is this single bit right there that first bit so that's bit zero on the data bus is the only thing that's wrong so when i push 2 i get a 3 which implies that that bit right there is stuck high now the reason why i'm pretty sure that the problem is with data bit 0 that comes out of this chip this ls 164 is because if on the data bus of the whole computer or at least of the video card if data bit zero were stuck at one all the time we would have a non-functional video card the 6845 which is the crtc controller wouldn't be initializing properly there would be other issues but the problem is strictly related to the keyboard which means it either has to be this shift register it's possible this ls244 is bad the side of it that faces the shift register unfortunately it could also be the keyboard now i ruled out the matrix on the keyboard is not the problem because the symptom we're seeing is of a stuck data bit but it looks like it uses an 80 21 microprocessor in the keyboard to translate the matrix into the serial signal that goes to the computer so and unfortunately for me z4 here is the microprocessor and it generates the serial signal directly that goes out to the computer so here's the data in the clock signal so if this chip has gone bad that keyboard unfortunately is toast there'll be no way to replace this chip so i really really hope that the problem lies here on the video card i really hope oh boy all right so here is the video board out of the computer and that's the 244 that might be the problem and that is the shift register parallel shift register i have the power supply connected i'm going to feed this 5 volts i have the oscilloscope probe ready i unfortunately can't connect the keyboard because i don't have the right cable and it's all integrated into the machine otherwise i could just connect the keyboard up and actually type on it and look at the signals coming through on here but let me power up this board all right we're getting a nice five volts there on that chip which is as expected okay here we are on ls244 pin 12 that's data bus 0. it's a bit hard to see but pin 8 and 13 on the 164 are the two pins we want to look at you know now that i think about it it's quite possible let me just do the old temperature check on these no they're not hot it is quite possible that this has garbage as its default state so it has an asynchronous clear which is pin nine and it has a line over which means that it clears the chip when it's low okay and that signal is high right now so that chip might just be showing garbage i need to clear it okay and pin nine is connected to the u17 ls74 and pulled up to five volts a quick check to the 74 ls74 data sheet and it's pin one which is actually the clear pin for this ic as well and you notice it's an input here clear is an input and uh it's pulled high so this chip is functioning right it's working so that means i can pull pin 9 safely down to ground which will reset both of these chips and hopefully the ls 164 will output all zeroes at that point oh incidentally if you're wondering how the data from the ls 174 makes its way to the data bus there are a couple lines here on the ls244 right here fcrd which would be something like keyboard read i guess i don't know what fc stands for but when these go low i think that means that the 244 connects these lines to the data bus and then the computer can actually read the keyboard but while those are not enabled these signals here don't actually make their way past the ls244 so i've gone ahead and i've hooked the oscilloscope up to pin 13 which is right here i think that's the pin we're trying to look at and we turn the power back on okay it went back to four volts and i'm going to take this little jumper lead and i'm gonna ground it to the ground pin pin nine that is uh and let's see what happens ah it went low well that's problematic because that would imply that this chip is working no no no no no no no let me go through all the pins on here they should all be low at this point so that's three four five six ten is low eleven twelve and thirteen uh oh spaghettio this is not good here it is still technically possible that when this does the shifting function that it has a problem like reset resets it without an issue but the shifting out doesn't work properly all right so since i cannot connect the keyboard and test on the bench here i am going to pull these out both the 244 and the 164. we'll test them in the retro tip tester and let's hope that one of these two chips is bad all right we have the retro chip tester pro i've taken out the 164 i left the 244 in there i figured i'll just test one at a time so in with this see what happens oh no that is working this is bad news everyone really bad news all right so that chip's fine i'm gonna pull out this 244 next okay the 244 is out and i also installed two sockets in there while i was at it it's the moment of truth let's stick that into the chip tester oh no that's a good ship as well so that's really disheartening i'm just gonna reinstall this back into the board that's disheartening because i mean further testing is gonna be needed but i have a feeling it may well be the controller inside the keyboard that's given up the ghost a bad shift register here would have been super easy to fix because that's a very common part all right so the chips post test good in the retro chip tester pro i'm going to stick this back in the computer like this see if there's any change and then just for grins i'm going to swap out these two chips maybe there's a problem with the driver on the 244 that is not detected by the retro chip tester pro because it's not in circuit and let's see if if that's a problem it's technically possible that something else on the data bus after the 244 like the crt c controller is sort of holding the line high and maybe there's a problem with the driver where it's not able to pull it down to ground so that the transceivers which are these chips see it as a one i mean it could be something like that who knows okay with the keyboard connected it did boot back up to the diagnostic disc the one that wasn't working so let's see if these keys work one two is working three it's it's fixed itself this is not good i'd hate these types of problems what it's quite possible is that those ics were problematic and the act of removing them from the board and i used hot air to get them off the pcb after i used the hakko just to make sure i didn't lift any traces that does heat up the chip maybe there's something extremely marginal happening inside of there and the heat actually fixed it well that is very frustrating and disappointing all at the same time because i kind of expected to be able to find that fault and get it working and instead the computer fixed itself for no apparent reason okay the saga actually continues because i had recorded an outro to this video and everything and then after that i turned the computer on and it's not working again let's see if it's still broken hopefully it hasn't fixed itself in the few minutes it's been off since i recorded the outro so i am booting the diagnostic disk again and remember this was the one that was just freezing if this does freeze all the same i'm going to put the other disk in where it gives us the time prompt okay yeah frozen exactly the same way so let's reset the machine oops i meant to reset it but i turned it off and with the other boot disc the one that was at least letting me us type so if the keyboard problem is the same i have replacement 244 and 164 shift register here so we will see if those fix the problem okay and yep right away it's over here where it shouldn't be one and that not only hits the hits a tab and then puts the three over there let's see three three five yeah okay it's exactly the same problem so the two key gives us a three and the four key gives us a five and the six should give us a seven and it does eight should give us nine yeah okay so identical issue it looks like that data bit zero is frozen again look at this it's not even booting at all now what is happening all right um weird it's getting worse now this is where having spare cards would certainly help the situation because i could take out this video board for instance to put in another known good one see if that changes anything in the meantime i'm just going to swap these two chips over okay so i just swapped over the 244 because that booting problem not sure if the 164 could cause that issue because when you're not pushing a key it shouldn't be trying to read anything from the keyboard all right here we go see what happens at least it's starting to boot that didn't seem to make any difference at all the drive is still enabled but it unloaded the head so it's not even it's not even trying to boot anymore incidentally the gotek is not even connected at all so in case anyone thinks that that's somehow causing a problem it's unrelated to what's going on here all right just for grins i swapped over the 164 to that extra chip i had see if that makes any difference i i don't think that's going to make any difference whatsoever nope exactly the same so this problem is completely unrelated to the keyboard it just that was somehow one of the symptoms which is very very confusing to me the weird thing is this computer does work to some extent obviously it's initializing the video or you know getting some text it's able to read that the disc is in there or not in there but now i can't boot what i did is i swapped the bus transceivers on the video card so this is what connects the data bus swap two of them around i looked up which one is data bus zero or data bit zero the one we were having a problem with and i swapped that with the one next to it okay that was weird it actually is different in that it seems like it just rebooted itself sort of okay it says error 32 error 25 25 error 7. so this computer is a it's very unhappy machine look at that boot error rs okay i was about to record the outro but i have one more thing that i want to try i'm going to remove the ls244 for the keyboard from that board all together that should effectively disable the whole keyboard circuit but the rest of the video card should be working if mysteriously everything just starts working perfectly then i think that might give us a clue to what's going on there's something going on with the chip select logic for the keyboard section of this machine and i just removed the ls244 altogether which means that that ship has no possibility to erroneously drive the bus although that has me thinking that that shouldn't matter because there are there's a second set of bus transceivers between the video keyboard card and the rest of this machine okay so that made no difference it just stopped in exactly the same way so clearly something else is wrong all righty we are now back in august 2022 and as you just saw i was not able to really figure out what the problem was the two chips that i assumed were the issue on the video card for the keyboard or not actually a problem at all and that's really where i left off with this machine now if you watch that second channel video i think if i turn this on we should see the diagnostic rom running i still have it installed in the cpu card the diagnostic routines appear to be working properly on this machine so the question is if i try to boot this machine into dos or the trs dos that is are we still going to have that keyboard issue that i was experiencing before remember that when i first had this computer back together at basically the beginning of the last video this thing was working properly i was booting software and i had used it quite a bit and then it started acting up now in all the troubleshooting for a few hours the computer would work properly that was simply not the case it was days and days and days had gone by and this computer never started working again now of course i put this off to the side and it's been like eight or nine months since and maybe now it's actually working now here's the funny thing while i was making the videos for this diagnostic rom after i had completed the video the computer started acting up while running this actual diagnostic now the reason why you might have noticed when i turn this on it didn't click the floppy drive is because the floppy controller is out of the machine and i say that because i think the problem that might have been happening the whole time is actually due to the floppy controller and here's why now the video you're watching now is the diagnostic running on this machine i recorded it on my phone because i was done shooting the main video and it was running through the diagnostics and as it was going through all the different banks on the screen the floppy drive light was flashing on and off that absolutely should not happen there is no circumstance when that would happen in addition sometimes when i would turn on the computer it actually would just hang at the original memory test well i talked to david of course who wrote most of these enhanced routines for this rom for this machine well it turns out that basically what happens is when you first turn the computer on it actually issues a command to the disk drive controller on point the disk drive to the controller and it waits for that command to return a bit that says okay i'm done well what happens is if it doesn't get that right response it just sits there permanently indefinitely basically at that sort of a screen that shows all the characters on the full screen that is well that's what this machine started doing now it was kind of random though like sometimes it would work sometimes it wouldn't it was doing very similar things to last time where i was trying to boot the operating system and it just wasn't working well i found that when i took out the disk drive controller everything would work totally normally again nothing weird would happen on this machine oh there's one other little tidbit sometimes while i was running with this controller in there when it was acting up and the light was flashing and stuff where it shouldn't have been flashing occasionally the computer would crash it would reboot it would just do really weird things and again when i pulled the controller out everything went away so the diagnostics roms really did help me figure out that it probably is the disk controller and earlier in this video i was thinking that it was the keyboard controller on the video card so what's kind of cool is this really gives me a place to start and the stock diagnostic rom or stock rom on this thing is so primitive that it really gave me no indication what might be going on and because you can't just turn this computer on without the disk controller and have it do anything i couldn't really test properly and i guess what happened is while i recorded the video for the diagnostic rom i had this machine on for a number of hours and just like last time after a number of hours it started misbehaving now it's been a couple days since i shot that video on my phone with the computer screwing up let's put this card back in this machine and i want to see does it start misbehaving like last time okay the disc controller is back in the machine so when i turn this on if all is working well we should hear this click three times with the light flashing three times and then everything should be working normally let's see what happens okay well right away it's not doing anything and there it is it's just frozen at the at the screen here so the diagnostic rom issues a command to the disk drive to i think like select the drive and activate the heads or whatever and now it's waiting for a bit to be set to zero look at that it actually passed that screen all right so um oh okay it's doing it so here it is it's still doing it so this is days later okay i don't know did it once there goes this should not be happening now the thing is as i mentioned it's selecting the different banks of ram right now it's trying to as it goes through there and this is obviously selecting here the floppy drive when it shouldn't be so yeah i'm gonna say that absolutely for sure the fault that i was getting you know months and months ago is happening again on this thing and it all seems to point to that disk drive this diagnostic rom was super useful at figuring that out if the stock rom had like a basic prompt that if i turned it on it could go into basic if i didn't have the disk drive i might be able to issue some commands to see if there was a problem in fact i could have just taken out the disk drive controller and then if the keyboard suddenly started working and i put the controller back in and then i couldn't type anymore well right there i'd know that it was a floppy card but you can't do that so this rom that i came up with just like on the model 3 that was incredibly useful like we have some information now on what might be wrong because honestly after i found out that those keyboard decoder chips were good i was completely lost as to what might be wrong with this computer and i pretty much figured i'd need to grab cards out of another model too and substitute them in here to see if that fixed the problem all right there it is the floppy drive controller seems to be causing all our issues and actually something interesting i tried the original rom back in this machine this is actually the diagnostic rom i incidentally transferred it onto a 2716. i put the stock rom back in the computer doesn't even boot at all before as you saw earlier in the video it was actually trying to load the operating system and we had those weird issues with the keyboard and stuff like that now it's just attempting to boot the disk it like loads the heads i hear clicking and stuff never actually loads anything at all there's one little extra tidbit of information i want to add after recording that footage that you saw earlier in this video where i kind of gave up for a while at least before my shoulder went bad i actually did a whole bunch more troubleshooting on this disk drive controller trying to figure out if any of the ics were bad on it and because of that we have a couple extra socketed ics like this one here is socketed where it didn't used to be and i think that was part of that troubleshooting that i was doing with this controller i never actually got anywhere and i kind of felt very defeated and was totally stumped with the problem that this was having especially because there was just no easy way to do testing outside of booting an operating system and since that wasn't working and it was acting all weird i basically just had no idea where to go so the diagnostic rom here with that clicking of the heads as it did the bank switching really gave me an amazing clue that really pointed to the disk drive controller as being the fault so armed with the knowledge that the diagnostic rom gave us let's jump into the schematics for this card and try to figure out what might be going wrong that would actually explain that behavior we were seeing with this diagnostic rom here we are in the service manual and this is the schematic here for the disk drive controller in fact right here is a floppy drive controller chip the fd1791 so initially i was thinking that maybe this disk drive controller chip which is kind of a rare and expensive chip i mean it's not super expensive but it's not like a one dollar chip either i thought that this chip might be bad but i was sort of chatting with david about how the bank switching works the ram bank switching that is that seems to be triggering the head on and off and he indicated to me that it's not actually the floppy drive controller chip that selects the heads on this disk drive if we zoom out here on the schematics it's actually this ic right here which is a 74 ls 174 now unfortunately the quality of the scan of this service manual is not gray so you can't really read these very easily but there's the drive select pins output here sd select add an fm mfm pin here that comes out of this thing it turns out that it's actually this chip that turns on the led and activates the read write head on the disk drive it's not the floppy drive controller at all but i don't think that there's a problem with this chip specifically i think the problem lies in the circuitry that turns this chip on and if we look at this here this chip is actually connected directly to the data bus and i think what happens is when you write to the appropriate io port on the z80 what it does is it activates this chip and whatever's on the data bus which is these pins right here 0 through 7 get mirrored into the actual outputs here which go to the floppy drive so that's how you could do the drive selects for instance now the way this chip gets turned on is this pin 9 right here this is the input and it's the drive slt pin or whatever well it's really hard to read what this is coming from but we can just look for the drive slt and find that part of the circuitry and if we scroll up to this part of the schematic i think we're going to find it right there there it is drive slt and this gate has a couple inputs into it it has one input here from the right i o pin so that gets asserted when the processor is trying to write to an i o port it has this pin right here i o cyc this also comes from the processor and has to do with writing and reading to i o ports as well so that as you see goes into this gate it goes into here and that generates drive select but there's also one other signal that comes into here and it's this trace right here if we follow this along it comes up here is connected to the address lines a0 through a7 and right there those eight address lines coincide with the fact that there are exactly 256 i o address ports that are available on the z80 that you can write to so the way it would work is your code says write to i o port say 0 0 and the cpu asserts that line that says hey i'm writing to i o port 0 0 and it also sets the address bus up for zero zero and then what the select logic is supposed to do which is what we're looking at here is it's supposed to turn on the chip select line which is usually driving it low to whatever chip is going to be connecting to the bus to read the data bus out so whatever information the cpu writes to the data bus gets read in by that chip my hypothesis here is that the chip select logic here on the disk drive controller that reads a0 through a7 is actually activating erroneously possibly all the time or maybe the output is like somewhere floating or at a weird non-standard voltage where some chips actually get activated when they shouldn't at the wrong time and remember that drive select line right here is generated by gating it between this select logic up at the top and this i o right stuff so if this section up here is misbehaving it won't be turned on all the time it will only be turned on when there's also an i o right that's happening on the system so as the diagnostic rom goes through and bank switches all the memory it's writing to an i o register to tell that part of the ram card to select in and out those particular banks of memory even though they don't exist it's still writing to an i o port well it seems like that right to the i o ports also activating the disk drive controller there which is causing that chip to read the data bus when it shouldn't which is almost certainly why the disk drive gets turned on and off as the banks get selected because of course it's going through a bunch of different bits as i select the banks and the disk drive controller that particular ic is seeing the data bus when it shouldn't be and it's turning the disk drive on and off because that ic is also turning on the drive select for the other three drives that aren't actually physically connected to the computer then i feel that if those drives were connected we would actually see the led coming on on each of the drives consecutively like on off on off basically as it goes through the bank switching now let's think about the other problems that's happening remember when you type on the keyboard you're getting weird results i think what's happening with that is a bus conflict with the floppy drive controller because the read signal which is coming in here is decoded and mixed together with output from this address select stuff up at the top here and then that is going to the floppy drive controller chip which is down here which has a couple signals on it has an in signal and an out signal which i think are basically connected to the i o port read and write signals from the cpu and then it has a ce or a chip enable signal which is this signal right here cefdc which should be coming from up here and there it is right there cefdc and you notice here it's generated by this set of gates up there as well so that implies that during read operations the floppy drive controller is probably writing to the data bus at the same time as the keyboard part of the circuitry which probably is crashing bits together as a bus conflict which is resulting in those weird key presses now you might be asking the system did kind of boot but then it would crash like what was going on there you weren't typing any keys or anything well that's probably because as the system boots it writes to other io ports for the pio and the sio chips and that is probably causing a conflict with the floppy drive controller chip as the system boots so as it goes to like set or read the time it might cause a crash or who knows because the floppy drive controller chip might get out of sorts now at this point it seems like the system's not working at all and i think that's because the chip enable line to the floppy drive controller is probably just on all the time and that is causing issues as well and if the signal is sort of in the middle between five and zero volts it might be dropping getting closer to zero volts kind of like the pla failure on the c64 where yeah the chip is just being selected all the time as opposed to being selected sometimes so finally i think it's time to do some testing and of course i can't easily test in the machine well at least not right now because it's behind me there but let's do a little bit of bench testing with this particular card so if you remember in the past i did some testing with cards on the bench where i just powered them up with 5 volts directly from my bench power supply across i don't know one of the capacitors here we'll figure out which that is in a second i'm actually going to remove the floppy drive controller chip because i don't want to be powering up this potentially sensitive ic when there's no signals going to it and we don't really need to have this in there for testing of the chip select line anyways all right i have a feeling it's probably this capacitor right here that is connected to these ics directly so this one's labeled plus on that side let's just see right here yep 0.21 and right here we'll go to ground which is this pin and yep that's also continuity okay so this is u8 i'm on the 5 volt pin looking good we're getting 4.93 that's completely acceptable all right i am currently on u8 pin 11 which is this output right here and it's currently low which is unusual but we can't right away assume this is bad without like taking a look at the inputs so there is pin 12 and it's currently high it's 4.2 volts and there is pin 13 and it is also currently high let's go take a look at the data sheet and see what the output should be when the two inputs are high and we can see here it's a 74 ls 32 and that's confirmed that's exactly what's on there and here it is 74 ls 32 quad 2 input or gate and we were looking at the output on pin 11 and 13 and 12 were the two inputs and we take a look right here when inputs a and b are both high the output should also be high let's go back to the oscilloscope and double check that that's exactly what i'm seeing so we're on u8 pin 13 that is definitely high we are on u8 pin 12. that is definitely high and we are on u8 pin 11 here and that is definitely low so this chip this does not appear to be working properly i'll take this chip out and test it in the retro chip tester pro but i mean we can see right here it's violating its own truth table in fact we could probably check the other pins on here but um you know what no point we know this gate is bad now let's think logically about what's going on the output of u8 should only be low when the address lines match the correct address to actually select that particular i o port but it's always low and it's going into another or gate which means a low and another low coming from like a write signal from the cpu means it's going to select that chip so right there that confirms for sure u8 is bad that completely describes the symptoms of what's going on with this thing it's going to be selecting probably the floppy drive controller as well and that i see that connects the drive select pins when it shouldn't when any other thing on the computer writes to one of those i o addresses so what i'm going to do is i'm going to remove that chip which is going to be a bit of a pain due to that corrosion and see if that fixes things all right the chip is out it's a bit of a pain when the there's a lot of corrosion on the solder it's just really difficult to remove it from the board but uh no traces were lifted this should be no problem i have a right mind to just remove these chips as well they look especially this one it's all green and crusty on the pins here so i should probably just take that out along with these two but you know what let's just see what happens with that one and one thing i want to say before i hit the button to test this is that removing it with heat might have fixed it because obviously this was like a subtle failure because remember the system was working perfectly and then it slowly got worse and worse so i have a feeling this chip is marginal and the heat might have fixed it so let's just see okay right away it failed so so this chip is definitely 100 definitely bad and now i have the fun task of looking through my chips to try to find a replacement uh ls32 and by fun i mean completely not fun because i don't have a good way to organize my chips right now they're just on these sheets here and yeah it's tedious all right well that was actually relatively fast i found a 74 s 32 and the legs are all mangled here so i'll try to fix this i don't know the difference between s and ls is but i know i tested this in the retro chip tester pro and it does test okay there whether this is going to work in this system or not well you know i don't know because maybe it's too fast too slow one of those things all right there we go just like that that chip is in there so let's put the floppy drive controller chip back in here and i suppose why don't we go for broke and i'll just test this in the machine and see if it boots i mean could be this one i see the whole time was causing problems but like i said all that corrosion on here could be these other ones as well all right let's test all right here we go uh i have a disc in here i don't know if it's actually a boot disk or not but everything is connected it's got the stock rom back in here let's turn this on all right it's trying to boot all right this disc may not be oh it's freaking booting oh man i can't believe it it's freaking working what okay i was like oh look at that look at that let me grab the keyboard all right that looks better in the camera there okay here is the keyboard for it and uh thank you very much to the lovely viewer who sent in a new case for this keyboard so yeah same keys and everything that i restored last time so all new foam and foil but this keyboard case does not have a giant hole in the bottom here it actually mounts the keyboard connector properly so we'll turn that off and we'll balance the keyboard right there just like that and let's power this on again i cannot believe that chip you know what's really cool to me is i as i mentioned was totally stumped on what was wrong with this thing originally and then basically the diagnostic romp gave me the clues that i needed to figure out that it was particularly that select logic that was selecting i o ports when it shouldn't but okay so it's booted up there is the dos prompt or whatever the real test here is when i type on the keyboard am i gonna get the correct uh letters and stuff so one two three yep that is freaking i can't believe it oh wow i am ecstatic here so there it is there is the culprit chip the 74 ls32 the one chip that was bad on this machine i can't remember if any other chips were actually bad on here but this one was and it degraded and caused this machine to not work i have to say for me this is a real triumph of the diagnostic rom once again because for me the stock rom on here just gave no clues as to what was actually the problem because you can't remove the floppy drive controller and have the system work at all the stock rom just basically errors out immediately when the controller is missing there's just no way to validate the operation of the machine on the other hand the diagnostic rom does work when the disk driver controller is removed from the system and we very clearly saw that with the controller in there well it wasn't working and with the controller out it was so i mean that was the clue right there and then of course the other extra clue was the way the drive was selecting as the bank switching was happening on there which totally totally gave us the clues we needed to figure out that it was the chip select on the address lines that was causing the issue and that of course brought me to this freaking little chip here which uh yeah was the problem so yeah wow what a journey and i have to say when i was first doing troubleshooting on this machine when i wasn't recording with the camera too i spent so many hours so many hours trying to just wrap my brain around what could be the problem and i was just going down wrong paths remember the keyboard i thought the keyboard was wrong there's just there were red herrings there were things tricking me and i needed that diagnostic rom to work so the other trs-80 models do have a basic interpreter which would allow you to do some simple testing with like i o ports and stuff that of course is if the memory is working because we know on those systems when the ram is bad it won't even go into basic you just get that garbage screen as you saw on the model 3 rom anyhow this is a very long video and um but i am so happy so so happy to have been able to fix this thing and do it without having to swap cards around from another working system that certainly would have given me the indication i needed to know that it was the floppy drive card that was bad but even with that it wouldn't have given me the indication to know that it was this chip that was bad and i would have had to go down the path of taking all the chips out one by one or i have this which was sent in by seth and this would allow me you put this in the back of the computer like that into the card cage and then you can plug cards into the top here and they will run outside of the machine so at least you can probe the test points and stuff while the machine is powered up and running this might have been the trick i needed to figure out that this particular chip was bad in fact i think there were some test points right on the board for these chip select signals so i probably could have looked at those while the system was operating with a little logic analysis of the schematic figured out that that was the problem so anyhow if you like this video i'd appreciate a thumbs up but if you didn't you know what to do i would like to thank a lot of people with this video uh thank you to the original person who gave me this machine who rescued it that was a huge thanks and then of course thanks to seth for sending in this my big thanks to david and frank for helping with that diagnostic rom especially david he ported the model one three rom over to model two he did that himself basically so huge thank you for that and i'm really hoping that these diagnostic roms that exist now for the model one two and three will prove useful to others who are working on these machines give you some tips on how to get them working as a side note we're looking to port to the model 4 and the 4p because i have another one of those machines to work on although i don't even know if it needs the diagnostic rom i haven't tested it yet but just so that's out there and that should complete the majority of these machines i would like to ask if other people have the other models of this machine from the family like the 16 16 b 12 and the 6000 if you could test out the diagnostic rom on your machines just to let us know if it actually works or not we can update the documentation on the github repo to reflect that so anyhow i think that's going to be that comment down below thanks to my patrons huge thanks to their support their names are scrolling beside the screen if you become a patron there's early access to videos and other behind the scenes stuff on the channel things like that so okay that is gonna be that so stay healthy stay safe i will see you next time [Music]

Channel: Adrian's Digital Basement

Views: 57,989

Keywords:

Id: Qq9oRcgFnX0

Length: 49min 50sec (2990 seconds)

Published: Sat Sep 17 2022