Can I finally fix this TRS-80 Model III motherboard? (When I tried and failed before)

well hello everyone and welcome back to adrian's digital basement on today's video i'm going to try to repair this trcd model 2 or sorry model 3 motherboard i've had this motherboard for a while and i tried to repair it once before and it completely stumped me and i wasn't able to fix it now this was before i was even making videos on youtube and i hadn't learned a lot of what i've learned since that point so i'm thinking now i have a better chance of actually getting this thing working this time so without further ado let's get right to it [Music] before i get started with the repair let's talk about my test setup a little bit because normally if you're working on a trs-80 model 3 motherboard you need the computer sort of split open with the monitor and all that stuff not exactly ideal for doing repairs if you don't have a lot of space like me but also not great for someone trying to make a video about it because then you're recording the screen with the camera and all that kind of ridiculousness so for this setup obviously i have the motherboard here on the bench and i have it connected to both power supply and the video goes to the rgb to hdmi first let's talk about the power supply and now don't worry it's unplugged currently so this is an actual trcd model 3 power supply out of a machine and i had a couple of these because in my trc model 3 upstairs both of the power supplies didn't work originally thing just appeared totally dead so i actually removed them and this is one of the two and i replaced it with an atx power supply to get this power supply working all i had to do was replace this one capacitor right here so that cap is what powers up the switching controller i know you can't really see it due to insufficient light there's a tiny little package right there it looks like a transistor but that's actually the switching controller that runs this entire switch mode power supply and that cap there is its power supply it's a very common part to fail because it's typically near a giant power resistor which is this green one right here and basically takes line voltage like the mains line voltage and it slowly charges up that capacitor until there's enough voltage in it that that little power supply controller thing can start running and then once that is running it takes over and it's powered off of like the main part of the power supply like from 5 volts or 12 volts or whatever because the cap is critical for the thing even starting and it's right next to a very hot resistor that's hot all the time high usage power supplies often bakes that cap and it kills it so yeah i swapped it out and this power supply immediately came back to life because the original wiring harness for the power supply to go to the motherboard was still in the computer upstairs i had to just make a new one so i just did that right here luckily on the power supply in the corner there actually has the voltage outputs and of course i tested that with the multimeter just to make sure i did hook up also an led right here with a resistor so i could tell when this thing is running since of course it's open and it's dangerous there are high voltages as it says right here while this is running mains voltages and things so do not touch this of course while it's powered up and for the last little thing i had this it was off like an old macintosh or something i cut this out to size and i added little plastic standoffs so that the high voltage on the bottom of this board is not just touching the desk right here it actually sits just like that the power cord i made is this one and it goes into this connector right here which is the power connector for the motherboard beware that the pin out is slightly different on this than it is on the power supply it's like reversed actually before powering this up i used continuity check to make sure that the 5 volts from the power supply was actually showing up on the right pins right here on the ram which is not installed at this time along with the 12 volts that also shows up here the negative voltage is negative 12 out of the power supply and is regulated down to -5 volts onto the motherboard with a little tiny voltage regulator here it's just a bias voltage so it's super low current that's why you just have a tiny little voltage regulator here but i was able to find the minus 12 volts that comes in the power cable on one of the leads on this little regulator that's right there now of course that totally sorts out the power now next up is video and i didn't want to lug the monitor down here i wanted to hook it up to the rgb to hdmi which is sitting right here and i know it's difficult to keep talking about this thing on my videos it's very hard to get them right now because the raspberry pi's themselves are nearly impossible to get this is running on an original pi zero but it does work with the new pi zero two i think it's called which is fifteen dollars if it's in stock and it never is anywhere anyhow on the model three motherboard this connector right here next to the power connector is actually for the video output and you can check in the radio shack service manual for the pin out but i've written it down this post-it note just for my help it's connector j5 you have a video signal vertical sync horizontal sync and ground of course if you download the latest firmware packages for the rgb to hdmi there is already a model three and a model one profile in there ready to go someone else has obviously done this work actually the funny thing is i hooked this thing up to the motherboard and i created my own profile and i saved it and then i went to go add it into the main profile directory as opposed to the saved profiles and then i noticed the real one was there oops for connecting it to the rgb data she might go from this pin header here and i had created this a long time ago it goes to a nine pin and on the side right here i had written the pin out like what color means what like vertical horizontal stuff like that going to this nine pin connector which goes into the rgb hdmi so i'll be capturing the video of the rgb to hdmi through obs here i have the oscilloscope which will be capturing as well along with a service manual which i'll have down here and that should be it that's how i'm gonna do my troubleshooting okay so here is the motherboard so we have the z80 cpu right here we have some rom chips this is the character generator rom we have ram which would be installed but because i had tried to fix this in the past then i couldn't i just removed the ram it's a 4-1-6 to use elsewhere i thought this motherboard would be sort of a goner but in the end obviously hopefully we're going to try to get this working so i will have to reinstall some ram there we have two expansion headers down here on the bottom i can't quite remember what is what i think the one on the left here is like a parallel port like a printer port and this one is a bus connector like an expansion bus connector hooking up a external hard drive things like that and then there are two expansion connectors up at the top here for plugging in like these clip-on boards which i think there are some holes here here and here that basically hold those in place along with being pushed in there and there's some holes there for whatever little clips and stuff like that pretty sure one of these is for a floppy drive controller and one is for a serial cart the trs-80 model 3 works perfectly without a disk drive controller and becomes like a cassette system for saving basic programs onto tapes and stuff like that now if you're familiar with model 3 motherboards you might be thinking there's something wrong with this particular one it's missing a couple other parts besides the ram and yeah it's these cables here so for whatever reason when tandy designed this thing they needed you to install these ribbon cables on the motherboard and let's see how this one goes on this one goes let me find where it goes uh wow why am i having trouble here well this one goes there are two connectors here and here so that goes in that position ah yes okay and then this cable goes from here to there now unfortunately take a look at this one see how it's a blue color it should not be like that it should just be a silver color that is corrosion this thing has corroded so badly i don't think this is actually going to work anymore this smaller one is only partially corroded see the blue is on the ends there and also on these outer traces but this inner part looks okay and the other side is the same thing so let's take the multimeter on continuity mode just double check it is and let's just see if any of these are actually making contact through okay it actually is making contact let's just see if all of these are working yep yep okay well as shocking as this one looks it is actually working now i i assumed these weren't working and it was going to be my intent actually just on the back of the board create badge wires to connect these two connectors together i was just about to go plug this in and i noticed it's all delaminating here which i think is exactly why it's getting all corroded and this is also a common problem by the way on the model one the keyboard connector uses the same exact kind of ribbon cable and it totally falls apart and just toning this out from this side of the connector to up here definitely this connector is connected directly to this cable and let's see over on this far side here yep also this one is the same thing okay anyhow both of these are fully functional now so these cables as bad as they look they actually work i've been talking a lot about this motherboard let me actually power it up and let's take a look at what it's doing right now in this current state all right so there's the output as you can see it's just a bunch of garbage now because this particular motherboard uses ttl logic for display circuitry it doesn't use a crtc chip like the later pets as an example it doesn't require that the computer actually be operating and running code out of rom to initialize the display the display will initialize no matter what's going on with the cpu you could remove the cpu and it would still actually display information these two chips right here are 2114 s rams which are going to be the display buffer and as i said that this is going to be the character rom so i'm sure if i remove the character rom which is going to get a full white screen but we can tell it's working because all the characters look correctly formed and then the static ram here is currently not initialized so it's just displaying what's in these chips as an example i'm just going to unplug that to turn it off and i'm going to take one of these out and let's take a look at what happens to the display so there it is notice that the characters are no longer formed properly there's a bunch of question marks and that's because with this chip out things are just floating now and the character rom doesn't exactly know what to do with that or whatever buffer chips there are so that is completely normal and expected i think if i removed both of the chips we'd probably get a full screen of just mostly question marks all right so we're approaching a system that is not working this is the equivalent of a black screen on a commodore 64. what's the first thing we check well power clock and reset signal and that's on the cpu so let's do that now all right so the cpu is up here the z80 cpu let's see what we're looking for first uh let's check out 5 volts pin 11. all right we're getting 4.82 volts which of course is exactly um what i expected because i already tested the power supply and the 5 volt rail on the chip is just connected directly to the power supply next up is reset line now when you have a line over it here or you have a slash in front of it that means that that signal is active when it is low so when the pin 26 is held at 0 volts that means that the cpu is in reset so when you turn on a computer you should see that low for a split second and then it should go high to indicate that the reset circuit has started execution or is telling the cpu to start executing plug it back in again and we should see it stay low for a second and go high okay there we go so that's normal behavior right there low then high okay next up clock signal and that's going to be this little symbol right here pin six okay this is what we're seeing we need to zoom in because it's going to be high speed and there it is frequency of two megahertz which i'm pretty sure is exactly what the trs-80 runs at so at this point it seems like the system should be at least trying to run so what it should be doing at this point is go to the reset vector which i think on the z80 is all zeros and that should select the rom which will then start reading back code onto the data bus and then basically start executing code you don't need the ram to even be working for that to work so we should see something happening on the data bus and on the address bus on this cpu let's take a look okay so what we are on right now is pin 30 which is an address line zero and that looks like absolutely normal activity right there let's just go through the line here that's address line one two three four i'm just going through all of these and they all look fine nothing weird nothing stuck in like the middle or stuck high or stuck low we're seeing normal activity on all of them all right so i checked up to address line 10 that looks fine let's try the data bus now that is data bus line 0 on pin 14 so i'm just going to go through them again make sure that everything looks good and it does that looks like normal activity we have a five volt line in the middle which is what we got six five three four and then we have the clock signal okay data bus lines look good and the address lines look good so the cpu is trying to do stuff now of course there's no ram in this thing so it's not likely it'd be able to really do anything like it's not going to clear the screen we're still going to get that that garbage i think what i want to do is take out the rom chips that should cause the data bus just kind of get stuck like maybe float even and the address bus just sit there because the cpu is trying to execute it's not getting anything or who knows what it's going to do might act weird so i'm just going to write uh one two and three on these so i can keep them apart i always draw a little arrow just so i know the correct orientation of the chips i put them back in correctly so to be honest i don't know if these roms even work and when i was troubleshooting this thing i'm not sure i actually had replacement roms to even test or so it could well be just one of these roms it's bad and that might be the entire problem with this machine i don't think i did much troubleshooting taking that working motherboard and swapping parts with this bad motherboard to try to figure out what the problem is but we'll see if we can do that maybe i'll get the retro chip tester pro out and we'll see if these roms are are working or not okay so i powered up the board again and now we're looking at address line zero it's just sitting there with a perfectly repeating pattern which definitely indicates the cpu was running code before and now it's just sort of not and i'm looking at address line one it's the same address line two same it's just it's sort of stuck yeah this is kind of normal behavior the cpu is just not able to do anything and looking at the data bus same thing is going on in fact it's kind of at 3.9 volts sort of floating and that's because the roms on here were pretty much the only thing that were driving the data bus because the cpu is trying to read out of the roms and then the roms need to write what they have in their contents onto the data bus so the cpu can execute it so that's why we saw activity on the bus that was coming from the rom chips and now we're not seeing anything because the cpu is not writing to the bus it's reading from the bus and because it's floating it's not going to get anything useful all right i know we have a really small window here but i am looking at the service manual and these are the three rom chips in the machine i can see that one of them is a 2k by 8 and the next one is a 4k and then there's an 8k okay and there's actually a hint on which is which so this one says 316b 332 and 364. this one here is the 8k this one here is the 4k and this one here is the 2k and according to the schematics the 2k goes into you is that a 106. because this one is in u105 and u106 is that one there and then yeah that one goes here this one goes in there so they were definitely in the right position now the thing is when we looked at the oscilloscope we saw that the roms were definitely doing something because if they weren't being selected like say the select logic was bad then what would happen is you would still see floating data on the data bus probably one of these is used for like initial booting of the system there's probably a basic rom and then i don't know what the other one does so whichever one is used for booting the system that one needs to get selected by the ttl logic on here for selecting the chip and then it does that reading from it well we saw activity on the bus coming from the rom chips so we know for sure that something is being selected there could still be a problem with select logic i'm going to kind of go to words that maybe the rom chip is bad so i think what i'm gonna do is break out the chip tester pro and actually just had a thought you know there's no ram in this system so like i said it's not even gonna execute i'm just gonna pause right here i'm editing the video and i notice that i keep talking about the fact that if there's no ram the computer is not really going to run and the reason for that is most 8-bit processors just don't have a lot of registers so there's a lot of pushing and popping to the stack which is part of system memory typically that is used to temporarily store registers and allow subroutines to work most code is written in a way that requires a stack to work and for the processor to have a working stack you have to have working system memory that is used for the stack and you might be thinking maybe this motherboard works i can tell you that it definitely doesn't and the ram that was on here the board was working exactly like this when i first tested it a few years ago and with the ram on there it did the same thing and i took the ram off and i used it somewhere else i thought maybe the ram was bad so i tested on another machine it didn't improve the situation at all so i don't think it's actually getting to the point where it's either initializing the video could be a problem with the cpu talking to the video ram actually that could well be a problem but i think i need to do two things right now i'm going to get the rich retro chip tester pro see if these roms read properly and then i'm going to put ram on this thing and we're going to see what that does as you can see this one says 16 b so we know this is like a 23 16 type rom but there's something very interesting that you have to consider the pin out of the 2316 and say the 2332 and 2364 is somewhat standardized but what's not standardized between the different rom chips is what combination of select lines need to be enabled to read out the rom and you can definitely come a gutser as dave jones from eevblog says if you try to read this incorrectly by using like say the commodore the way of selecting roms may not be the same on the trs-80 and you will get a blank rom back okay so we're going to pop this one the 16 into here and if you notice there are actually two 23 16s an a and a b and i'm assuming there's a slight difference in pin out i don't know what's what so let's just start with the a one now here's the active state thing i was talking about and um this thing has an auto detect mode which you get to by pushing that button and now it says jump to detect so i'm going to push jump and there it is it says not possible i think that means that it can't figure out how to read this so i hit the reset button to get out of there all right so let's go to the b1 and here it is these are the active states again in different lines commodore apple ii and then auto detect so i'm going to hit the select on there okay there it is figured it out low is 18 and 20 and high is 21. with that combination the chip is outputting valid data or some kind of data so it determined it thinks that is the correct select logic for that chip i think that happens to match what commodore does but even commodore is not consistent from one ship to another so again be very careful with that all right so we're going to hit okay it's going to read the chip it's going to calculate it on the sd card i have an index file which see that's interesting um it it should have read that i think try that again okay 23 16. is this in there perfectly make sure it's in there jump to detect there it is now we're going to read it okay so the index file which i think you saw pop up there for a second is the main database of like all the roms that are in maine and the crcs of that are for all those roms so you stick in a rom chip like a commodore vic-20 kernel rom and it will say vic-20 colonel on the screen there let's try a different one because it's it shows that crc there but it's not giving us the trs-80 thing there okay so this chip here is the 2332 so we have to exit out of this and go to that which is this and let's see auto detect not possible oh because the chip is only half in the socket let's try that again okay there it is so low is 20 and high is 21. let's see what that gives us aha look trs80 model 3 so this 32 chip is actually reading and in fact there is the file so u105 that would be like what it's called on maim if we just move this for a second this would have been the middle one yeah u105 there okay so i'm going to draw a tick mark on there to say that that chip is probably good not 100 sure that it's good that retro chip tester pro is not perfect at testing things doesn't load the outputs the same way as the computer but at least it's able to read it and it sees it as what it is this is the next chip here this is the 64 that is from the socket over so we're gonna go to 64 and it's a 24 pin one oops 24 pin let's see does this have an auto detect yes it does there it is low is 20. let's read this one out okay there it is u104 model 3. so this one i'm going to draw a tick mark on it that one looks good as well but let's go back to this one here and let's auto detect okay there it is and we'll read that oh wait there it goes why wasn't it reading properly that's weird so is the chip marginal i mean certainly looks good it's from 106. let's try again because the thing is it may read one time properly but maybe it doesn't read multiple times properly so let's do this auto detect again hit okay okay it's working so maybe that was just a glitch in this or something i am not quite sure about that okay so after three reads that chip seemed like it was good so i'm gonna say that all three of these rom chips appear to be good now it's not for sure but the fact is they give the check some this thing is very useful for that now i could have easily read it out myself done a checksum maybe that main database i assume has is available on the internet somewhere and you can find the checksums i don't remember where i got it i think the creator of this thing had a link to it that i just downloaded the whole like daily database that's on the sd card so there should be a way to for you to at least dump the rom with a regular e-prompt programmer if you create an adapter to do that and then find the roms for the model 3 and then compare them you know use a hex editor or whatever compare the two files and that would be probably one way to also see if they were good but of course thanks to stuart who sent this in this thing speeds that up a lot all right so i think i need to stop dorking around here and actually just put ram in this thing put the roms back in so we can continue troubleshooting my all right the ram is all installed unfortunately only this ram is tested good and the retro chip tester pro this other stuff i just had i found that this ram is relatively reliable and when i find bad chips i always throw them away or they're in the dead part spin so let's power this baby up i put the roms back in as well in the correct spot let me just double check is everything yeah everything is facing the correct direction okay here we go all right well we're getting a little bit of a difference here so now we're in 80 column mode so trs-80s both the model 1 and the model 3 because they're both the same really have both 40 and 80 column switchable now that the ram is in and the roms are back in and it's switched into 80 column mode because before it was 40 that actually implies something to me that implies that perhaps the rom is executing code right now and like it's looking for a keyboard um you know input or whatever and it is able to command the the display circuitry to 80 column mode well it's not 80 actually it's like 32 and 64 columns i think i i kind of have a feeling now that maybe the problem is actually the cpu talking to the display memory because if you think about what happens you turn on computer very first thing that happens it does some bootstrappy stuff eventually what happens as it executes rom code and it's using the ram to do that what it needs to then do is clear the display memory erase it you know fill it with spaces basically which is the equivalent of erasing it and then put the welcome message if the cpu is not able to write to the display memory which is these two chips right here well that's not going to happen but the computer might be actually executing and working normally otherwise i think one thing we can do to see what's happening with the ram is to just look at the ram chips here and see if it's writings of them so here is ram on this particular motherboard right here and notice here on this first chip here we have raz we have cast and we have a wr pin that's the right pin if it's actually writing to memory then i'm going to go out on a limb and say that the roms are actually running so i'm on pin three and i don't know if that's a correct write signal i mean it's it's you know this this bouncing up and down is just because other stuff is happening but it's not writing to memory right now all right let's change the trigger mode to falling and i'm going to switch this to normal mode which means it won't show anything until it actually gets triggered so if i unplug this notice there it showed something so when i plug it in here it looks like it was writing to memory let's uh let's do that again power cycle it there it is so those are all rights you see all this up here it was actually writing to memory it was doing it in like bursts and then i guess what happens is once it gets to like a cursor prompt maybe it's running rom code like it's waiting for keyboard input and um there's no more ram access at that point okay so definitely it's writing to memory when it powers up which would be normal it's probably doing like a little bit of a ram check or whatever and then it sets the display to 80 columns which again i think is something that's happening because the rom is actually executing i think if i pull these out that won't happen anymore so the next tactic is let's go to the schematics and look for the display circuitry and try to figure out how the cpu is able to write to display memory all right so this right here is the video output there's the sync output uh this is really hard to read right this is a 74 ls 166. this right here is the character rom so this is cycled through to display the characters so around here somewhere we're gonna find the video memory okay here they are two one one fours two one fours they are four bits each there are vd7s here's the data bus that goes to the video memory we know these chips are good because it's currently displaying memory out of them that's what the garbage is on the screen and when i pull them out remember we're seeing those question marks that implies that definitely those chips are well i mean at least they're being read out by the video circuitry and that happens automatically but right here u67 this has got to be it doesn't say what china chip it is but it has the vd lines on one side and it has the d lines on the other here it is right here 74 ls 645 i got to look up what that is exactly all right so it's a bus transceiver but what exactly is different about this than the 245s that we normally see it's a bi-directional bus transceiver 20 pin package that's all pretty normal let's bring up the ls245 to compare so this is the ls245 datasheet so there's the direction pin output enable uh i mean this is all looking very similar all right so propagation delay 15 nanosecond versus 12 volts the same i mean current drive all similar i'm very confused it literally doesn't look different to me well back on the schematics here so whatever this chip here pin one is the direction signal vrd so that's the direction signal and vrw this is the right signal for the memory let's look at where those are coming from exactly or you know what why don't i scope those signals and let's look at what that looks like okay so there we go pin one on the um 645 here it's changing direction quite a bit and then this is the enable signal and it's also active quite a bit that chip is being seemingly instructed to turn off and on so it's like doing its thing so why don't we look at this right signal on the video memory because of course there's a flashing cursor if this thing is actually running which should indicate that it's updating video memory okay so on the video ram on the right signal that's definitely not writing so i'm going to do what i did before switch the trigger mode to normal and we're going to power cycle this and we're going to look to see if we see any writing to the video around no there is no writing going on here at all okay i'm making a lot of assumptions here like i'm making an assumption that the computer is actually working like it's running like it's even trying to clear the memory um so what do i do here let's see what do i do let's see what i need to do is find out where that signal is coming from and then try to look at the circuit to see like if it's a logic chip that generates it which it will be what exactly what conditions are being set to generate it maybe i can probe the other pins to see if that chip is is just bad and not actually outputting the right signal the schematics here are not helpful because everything here like see these arrows point to things that go elsewhere on the thing and there's no lines and it's really not helpful so v wr i have to kind of look through everything here it's gonna be an arrow pointing out because see how it's like an in arrow there it's an out arrow so i gotta look at all the out arrows here and find where that is now i just had a thought i still haven't found that signal but i did find in the address decoding page this the keyboard signal i think this like goes to the keyboard it's probably some ics on there to basically um read from the keyboard let's take a look at that pin i don't think it would be doing anything if the computer weren't waiting for keyboard input this is just going to try to validate my hunch that the machine is actually running we're going to go to u60 i think that's a 60 and we're going to look at pin 6. and as this ic right here it looks like it is the correct 145 so we're gonna be looking at pin seven oh it's confusing because look at this zero one two three four five six this is actually the decoded um signal and that is the pin number so it is seven hmm okay i'm not seeing anything let's power cycle the computer yeah that does not look quite correct and if we change it to single nothing there's nothing happening on that pin what about these other signals here so i'm checking the inputs here so i'm on pin that's pin 12 which is the d input pin 13 there's activity 14 15. so we're getting activity but it doesn't look like any of the outputs on this chip are actually doing anything i mean that doesn't look right this chip might be bad no all of these this this chip is selecting things and it just those outputs don't look correct let's look at this little neighbor here uh which is similar not quite the exact same thing but if we scroll over we can see that this is also selecting like well there's like rom a for instance so we look at pin 12 or yeah pin 12 on 58 so there's pin 12 and you notice there's quite a bit of activity on it looks normal for like selecting the a-rom chip uh but then this comes out this is like a little cascade selector thing and then this using some address bus input here is supposed to select these other chips right here and it does have pull-up resistors rom b is all of those c keyboard and video so i mean could i have just stumbled onto the bad ship accidentally but one thing is for sure if we look at the schematics here see this vid signal that is probably necessary for any kind of activity to the video circuitry so for like writing to the video circuitry that probably needs to be low or high or whatever probably low because these this has pull-up resistors see pull-up resistors right here so vid needs to go low to select the video circuitry and if this ic is bad that's not going to happen i'm thinking that these little drops here are it trying to drive it to ground and not working so like it's trying to select things so i am going to say right now this tip this tip is bad right there now funny thing is you saw i was trying to drive it to ground but it wasn't didn't look like it was able to so what i'm going to try to do is i'm going to try to find another one of those chips i'm going to piggyback it on there because basically if the chip i piggyback on is able to drive it to ground this bad chip's not going to do anything it's kind of like dead there on the board and maybe this thing will come to life let me go find a ttl chip all right so the bad chip right here this is a bcd to decimal decoder so i'm just going to put that chip right on top there i've just piggybacked it and let's take a look at this pin here which i think is pin 8 for the video all right so i'm going to keep my finger on the chip here i'm going to plug in the machine and let's take a look at what we see on the output well would you look at that that is very different and let's switch over to the hdmi and we're getting nothing that's interesting why are we getting nothing let's unplug this and take the chip off and see if we get the garbage again like we should okay there it is garbage now let's put that chip back on and we'll plug it back in well we don't get garbage but we're not getting anything that well could be because it's trying to boot a floppy disk do you have to basically hit the brake key to get to the basic prompt on a model 3 i don't really remember because there's obviously no keyboard hooked up and we don't have a disk drive hooked up i think before i grab a keyboard so we can do some more testing to try and break out of this thing let me replace this chip i can't believe i just stumbled onto it pretty hilarious all right there it is a socket there's the bad chip removed from the motherboard and there's the replacement so let's pop this one in and uh see if we have a nice working machine of course i'm going to need the keyboard which i have grabbed as well all right here we go i'm going to plug it in oh we have garbage that is interesting so that implies that was not the fix okay well that's so weird that when the chip was piggybacked see when i plug it in there was 40 columns for a second and then it went to 80. what did i do wrong here let's uh get the scope first off let's check out the input so i'm on 12 that's 13 14 and 15. so inputs all look good video output pin nine okay there it is so that definitely looks a million times better um than it was to me with that other chip i wonder why when i piggybacked this bad one on well the good one on top of the bad one that we at least had a black screen wow that's uh kind of disconcerting well let's look through these so pin one is rom b oh that's interesting it's not really dropping down hmm okay let's look at rom c which is four and five same thing and then pin seven is the keyboard now it's not doing anything i guess because it's not looking for the keyboard and then we have the video which is working all right well let's see here let's pop this one out this is the new one and the original chip is back in okay we're not really seeing anything now on any of these pins let's recycle the power again okay wait a second so that is the video output and then there is the rom b select and yep it's uh doing the same thing so maybe i accused the chip of being bad when it wasn't now okay we're getting garbage with a little bit of static hmm okay what i want to do is replicate what we had happened the first time when i piggybacked the good chip on top of the supposedly bad one okay it's piggybacked and we hit the power so that was like some kind of a red herring okay so i just tried something and i took the chip out in both of them and now we're getting 40 columns so it's not even going to 80 column mode so that's interesting so i don't think this chip is actually bad this 145 that i took out luckily it's not damaged and should still work in the motherboard here alrighty so definitely this chip is not bad i don't know why it looks like that on its outputs the only slight decrease and i also don't know why when i piggybacked it originally it did result in a black screen it's very confusing to me i mean it was it was pretty clear i'm pretty sure i did it like put it on it was black took it off it wasn't put it back on it was black again and now it's not doing that so it's all a little confusing to me um all right lots of time has passed and i've been doing lots of troubleshooting off camera because at this point i don't really know which direction to go there's a few things that i figured out since i turned the cameras off to do some troubleshooting first off this ribbon cable here is pretty important actually this connects the data bus between the cpu and its buffer and a lot of the other components including the video circuitry so with this out of there the video circuitry can't possibly work but i did double check that i'm getting connection from this side which is this buffer chip over to the buffer chip here that's on the ram circuitry so this is a good connection this thing there's a set of jumpers on the board that are for whether you're selecting like 4 or 16k ram chips things like that i want to double check that all of them are set correctly and they are i took out all of the additional ram so we just have 16k in here this is the first bank up here at the top though and what i did is i double checked that these two extra chips here worked i checked them in the retro chip tester pro so all of these are known good chips there's a ribbon cable coming off the board here this is actually an ide cable because i couldn't find the original cable but i do have the keyboard for the computer this is a model 3 keyboard i just had a spare one in my storage bin and yeah this is so i could push the reset button to help with troubleshooting i learned that the 74 ls 645s where there were three of them one here one here and one there are the same as 74 ls 245s and in the service documentation it actually calls them 245s so clearly those are interchangeable parts i have the three 645s here which all test good in the retro chip tester pro but um they get hot when they're running that just i kept put touching them with my hand and i was annoyed by that so i put in modern chips um these are modern versions that just don't get hot and lastly on the schematics i was looking for this right signal video right and i don't know how i missed it but it's generated right here by this ic there it is video right and there's video read it is kind of interesting on these 2114s rams that the chip enable line is grounded which means these chips are always active but you have this chip here which is one of those 245s or 645s and the video write signal or video read signal i'm sorry is what connects these to the actual data bus and when i say database d0 through d7a is this side the left side of this ribbon cable this side is without the a and this side is with the a the ribbon cable that goes up to this side here is the b cable like data buzz b and from what i can tell it is only used for these two connectors here but basically i've been going over this board reading through the theory of operation for the machine checking everything and nothing really makes sense to me what is happening is things are kind of inconsistent now that i have a reset button on the keyboard sometimes i push the reset button and the screen flickers a little bit there's still garbage but it flickers a bit sometimes i push the reset and then it does nothing i can look at the various signals on some of these chips like the video read and the video right and depending on which reset cycle i'm in sometimes there's no activity on those and sometimes there is so things are just weirdly inconsistent besides swapping out those bus transceivers i did double check that these video ram chips work but i mean i know it's not that i did swap out a different cpu and that didn't make any difference whatsoever so it's not the cpu either so this has all kind of led me down the path that maybe this rom chip is bad now while it tests perfectly in the retro chip tester pro maybe when it's tested at speed or red at speed by the cpu it's not really returning the right results i'm not sure i remember when i first well when i first showed on camera in the retro chip tester pro this was actually not reading correctly and then it just started working again so the fact that things are just inconsistent when i hit the reset button like the states of the machine the various state machines on here just seem different every time that screams that maybe the rom is not working so what i went ahead and did is i built an adapter i happen to have some of these pcbs and it goes from a 2364 to a 27 128 and there's actually a jumper location right here that allows you to select between two different images in here so you could have two versions of the rom and i figured i'd need to build one of these anyways because i'm thinking that the next step for this machine or this motherboard is going to be maybe trying to write a diagnostic rom that's gonna do something like it will do a little ram test or maybe it will just immediately start writing to video ram over and over again and it just can let me validate you know what's going on exactly i i don't know um to test that this actually is wired up correctly because i've never used one of these i put this into the retro chip tester pro which happens to be sitting right here is that going to focus and that is the reading it actually read it out as the correct rom so i know that this is working as a 23 64. so i haven't tried this in the machine yet alrighty there we go i'm pretty sure that if this is installed on the motherboard due to the stacked nature of it i can't install the rf shield that is in the actual machine the motherboard will fit without a problem i don't think there'll be any issue without the shield but definitely with the shield i think that's going to be too tall okay moment of truth i'm capturing video see what happens here no no change okay so and see when i hit the reset button occasionally the relay which is this little blue thing here goes like click it makes weird noises the whole computer is just not not happy something is going on on here that i am just struggling to find i don't think anymore that this machine is actually executing code properly well it's starting to boot clearly but i don't think it's actually getting to like a basic prompt that i just can't see the states of all the various signals are too inconsistent to really be a working machine something is causing this machine to not work all right there it is that's the trc model 3 repair video that was not of repair i got stumped once again so i apologize for the unrewarding video i figured i'd leave it in though because i wanted to show that the trials and tribulations of trying to repair a board like this it's just natural for to go down various paths unless you're super familiar with the way it works like i am with commodore 64 as an example it takes a lot longer for me to kind of digest and try to break down and reverse engineer from the schematics the way this thing is supposed to operate now you can imagine if you're working on an unfamiliar board for the first time you don't even have schematics it can be very very difficult to try to even figure out the problem but even when you have good schematics like i do in this case and i even have like theory of operation and stuff like that that's in the service manual it still didn't result in a working board but don't worry i have some more ideas and i'll be taking a break between the repair sessions and i think sometimes that's what you need to just sort of clear your mind and start again from a better perspective on things i think i was going down some of the traps so that the wrong paths i was going down the last time i repaired this board even though it was years ago and i think i just need to kind of like clear my mind and start again which i'll be doing in the next video and i guarantee you i can't guarantee 100 but i guarantee you that almost certainly i will have this board running in the next video and and no i have not recorded the next video yet while i'm doing this outro so if you enjoyed watching me flounder around not being able to fix this board thumbs up but if you didn't know what to do subscribe if you haven't already check out the second channel if you haven't already and thanks to my patrons i really appreciate their support you can do so you can become patron down below if you feel like it and um that is going to be that so stay healthy stay safe i will see you next time [Music]

Channel: Adrian's Digital Basement

Views: 80,204

Keywords:

Id: EGFKjjlvKf4

Length: 48min 28sec (2908 seconds)

Published: Sat Aug 13 2022