When I finished my apprenticeship as a TV Service Engineer back in 1986 the company I worked for had no vacancies for me - even though I passed all my exams and had a very good end of apprenticeship report. The TV rental market was dying so I re-trained as a computer repair engineer. My very first job as a computer engineer was aligning and repairing the very Tandon floppy drive you're working on in this video. It's all I did for the first 6 months before working my way around all the different peripheral devices and eventually repairing the actual computers. I've done this job for 37 years and now I am a network consultant. I am very happy to have got 37 years out of repairing electronics and consider myself very lucky. I continue to work on electronics as a hobby now - but I have reverted to working on valve (tube) radios (as part of my apprenticeship education - even in 1983 was based on valve technology) I still look at the occasional laptop or PC when I feel like it.
People that have the skills and experience on the old tube stuff are not getting more by the day. So great that you can help to preserve more historic electronics
Just shows how useful an oscilloscope can be, when you can see the data getting lost along the way it really helps with figuring out what's gone south... :)
Adrian's Digital Basement videos should replace most computer class teachers in elementary and high schools. If I had this growing up I would have learned so much more.
@@jeremylindemann5117 And some “higher ed”.. I had a “class” on computer science after high school AND after working at a drive repair shop, ( yes we repaired both hard and floppy drives ) Any way.. the book stated that the index sensor was the track 0 sensor.. That and just about every other “fact” was WRONG. I ended up teaching the storage section. At first when I was correcting everything, the instructor was questioning where I was getting my information. When I stated I just came from a shop in Beaverton Or called “ Disk Drive Repair”, he got the idea.
I've got one of those! Manufacture date: December 27th 1982. It, um, somehow made its way home from IBM, where I worked, one day and never went back. I don't even remember if it worked when I obtained it, but it was almost certainly heading for the landfill if I hadn't rescued it. It's a fun relic to have around to show kids how far we've come in 40 years.
Wow, this brought back memories. In the 80s I worked for a Canadian dedicated word processor manufacturer that used SA400s and Tandons. Did tons of alignments. It was always either dirty heads or alignments. Every now and then a dead motor. Don't think I ever encountered any other issue.
The challenge of debugging and fixing things is most of what I enjoy about working on old devices, I almost never end up using them regularly afterwards… so many consoles collecting dust.
The diagnosing and repairing of broken electronics is my favourite part of your channel. I actually don't have any special attachment to the retro tech itself. So, in the nicest way possible, I hope you've always got a pile of broken stuff to work on. Thanks for the interesting videos.
Every time I have encountered a fault with that multi vibrator in the design I have found it to be bad. They go bad frequently. I don't know why they go bad so often.. this video was a really good one! 😊
Ah that's interesting! I can't recall running into a bad one but I'm not sure how many times I've worked on something with one. I will need to store that tidbit somewhere in my brain :-) I know the fault on the Plexus was a dual flip flop now I think about it.
AHAHAHA! The first disk drive for my home-brew Nat Semi 32008 using a Western Digital 1770 controller was a dodgy Shugart SA-400 fished out of a bin. The work around to dodgy msmv was wiring a bypass with an OR (or was it AND) gate!
Very nice fix! If you look at the datasheet of a floppy drive control IC (not to be confused with the floppy driver controller on the PC side) for 3.5" drives you'll find for example in the AN-917 datasheet page 3 figure 5 a schematic showing what the signal path is. 1. The head 2. Preamp (differential) 3. Low pass filter (the one on your schematic the opamp with the inductor) 4. Differentiator, turns the sine waves in more pointed pulses but still sine wave like 5. Zero crossing detector, turns the signal into short pulses 6. Time domain folter & pulse generator, these are the flip flops that transform the time between short pulses positive edge pulses into the final square wave where the edge going negative to the next edge represents time where 4us, 6us and 8us represent the flux transitions 10, 100 and 1000. Nice to see this in discrete components instead of one IC that does it all! If such an IC fails then it's game over for the drive, these can't be found anymore. I've done decoding of analogue signals of the flux transitions captured using an oscilloscope programmatically in an application that I wrote. Interesting stuff! That the grease weasel has some retries could be that their decoding doesn't have much of a compensation for timing ie if there's variations in motor speed or a bump on the disk (the well known clicking sound) detection of the proper timings may be off. The hardware decoder in the floppy controller card is more robust in that regard.
utterly fascinating, got a bit lost in the logic area but don't care it was just brilliant to listen to someone who really knows what they are talking about and is so comfortable with it, well done
The power connector on the GW can be used to power a 5v floppydrive. Relatively new 3.5” drives are 5volt only. (I seem to remember coming across one that required 12v as well but it was a long time ago and I can be mistaken. I created an external GW drive using only USB-c that way. (I have a few type-c weazles). Furthermore let me take a moment to express my appreciation for you, Adrian; your videos are always very entertaining and informative. Not just “this is the fault and now it’s fixed” but also the backstory etc. Thank you!
This is a gold mine of useful information! I wish that I'd known this stuff when I had a Tandy Model 4, back in the early 90s. Its disk drive stopped reading information and what did I do? I threw the whole computer away. What a bozo.
Im very impressed with your videos. Im trying to repair a C64 with 1541 and an Apple2e with duel floppy module. Im learning SO much that I could have benefited from 30 years ago.
It is a reminder of how difficult things could be back then. On the other hand you could fix stuff ... Now we have to replace entire massive components or deal with OS bugs that might never be fixed.
@@adriansdigitalbasement The worst thing I remember was buying a new board, plugging it in, and the computer wouldn't even boot. BUT I put it in my sister's computer and it worked perfectly. There was simply no explanation and no fix. You simply returned the board.
I just repaired my old 25 disc file CD changer and was proud of myself. Had to order belts from Belgium and clean up the rails. But had to disassemble the whole thing. But watching your channel and others gave me the drive to try. They don't make anything like this anymore
Looking forward to watching this one. Just repaired a pair of Tandon TM100s from a Kaypro II. Bad TIP110s, speed controller, seized bearings. One of them was rotating at 900rpm 😂
@14:09 not sure if it will get tested further into the video, but before I forget: measuring the resistance of the read head is also one of my first checks. Usually each wire pair should be between 10-20Ohms. If you see 0 or infinity or even kOhms, one of the coils is broken. Then you will have a very bad time indeed.
you just boggled my mind thanks i learned a lot i love my computers but most of them are broken i dont know if i will get back to my hobby you gave me inspiration
Yes, Adrian's determination in the face of the problems gave me hope for my "museum". I like that he labels the new issue with the 286 not as frustrating but an opportunity for more fun in the future.
I actually have a late revision 5150 made in 1983 and it shipped with a single sided drive. It also just had 64K of RAM and MDA. I managed to find mine as an unused boxed example so I know the drive is original. The manual in the box also contains DOS 2.1 on single sided disks. I believe IBM continued to offer a single sided drive late into production as an ultra low cost option before finally being phased out in the XT
I started at Computerland in late’83 and had my hands in hundreds, maybe thousands of these machines before they went to customers. I can tell you none went through the config area with single-sided drives by then. I find the April ‘83 date interesting.
@@adriansdigitalbasement Agreed. I didn’t consider that. I do remember seeing only a few machines with the black power supplies (recessed nuts) come in for service but don’t recall the type of drives. This is all sort of a fun rediscovery for me. As I recall, a few years in IBM started using another floppy drive manufacturer (I believe they called it a type 2). Haven’t seen one of those around but they were not good and would be an interesting repair. If memory serves, they had no track zero switch.
This seems very similar to an issue I had with an old and defective Chinon FB-354E in my very early Amiga 500 (with space invaders keyboard). It stopped working, and after probing around I found after amplifying the signals, they run it through a Nand gate, a simple 7438. Probably to make a proper digital signal from it. It turned out to be the issue and in the end it was a 50 cent repair. I repair lots of stuff, but I was so happy to get this fixed as I definitely wanted to keep the original drive in there, even if I have several replacements. Great work in video too, that greaseweasel seems like a nifty little thing!
Reminds me of working on the c64 1541 when reading disks was intermittent; after pulling down the case and looking at the head assy, a small pop out felt pad would wear over time and not compress the disk against the head causing intermittent reads. Couldn't find anyone or place to get a new pad, bought some felt, hole punch, and some double sticky clear tape. Good to go. Great memories.
Wish i was like you and had tons of retro computers lol. Very late too the game if wanted to start now, after getting a house few years back pretty much any extra cash goes into it. So thanks for all the great video's of the things i grew up around and never really new how they worked. I was i outdoor kid was in the woods more then inside growing up. PS never got a tick as a kid with no spray on even now go too cut my grass and have gotten them on me.
It would have been easy to edit that part out to make the repair appear all neat and tidy and I'm so glad it was left in. It showed the reality of our passion where the equipment is so old there can be multiple simultaneous faults. I'm glad we live in a time where these computers are old and rare enough to be interesting but still common enough that folks with knowledge and grit can repair them. Thank you Adrian for showing us how it is done!
my hardest lesson in electronics happened when I stepped barefoot on a 555 that fell on the floor... It hurt more to pull it off my heel than it hurt when coming in!
the Microsoft Sam voiceover made me do a spit take. i have too many interests and i end up watching several videos of various you tube channels in a row. i greatly enjoy Your content Adrian. And i look forward to your next repair adventure.
This sort of stuff gives me flashbacks of when I had to learn about disk formatting low level in the 90s by books from the library because nobody I my friend group knew why mac disks didn't work in IBM compatibles. Ended up finding a System 7 program that would read IBM formatted disks on my Powerbook 100 so I could do homework on it because Word on the Mac was the GOAT.
U can use the decal on the bottom of the spindle. When at the correct speed the pattern will seem to stop. There are two patterns. One for 60hz and one for 50hz
I grew up in Boca Raton, home of the IBM division that created the PC, and saw loads of 5150s. I owned four of them over the years, too, all acquired used, but somehow I never even encountered a single-sided drive, ever, in Boca Raton. So, interesting to see.
That's cool. My 5150 is an original early 16-64k machine I bought from the original owner here in Portland. Even my machine had the drives upgraded at some point or came with double sided drives... So in other words I've never seen a single sided IBM branded drive either.
My Tandon TM-100 is still sitting on the shelf, so it's always fun to grab it and follow along with the video. It's a -2A model with a "48 TPI DSR" label under the latch, and a build date of 3-9-84. I've decided that's a US American date format, which means it's a little older than I am, so I don't feel quite as old as I would if it used the correct date format. My guess the problems at the end were related to the XTIDE BIOS. If they've patched the floppy routines, it's possible those routines are bugged in a way that gets the drive into a bad state, and it's returning unexpected data that DOS doesn't understand.
Very nice repair of an old floppy drive. It occurred to me that this old floppy drive makes a lot of noise and buzz and the newer, faster computers can't filter it out and it interferes, causing errors and freezes.
3.5 inches floppy disk drives also do have terminator resistors on-board as well. I took a few apart and indeed they're terminated so if you remove that SIP resistor pack, the issues would basically be the same as older 5.5 inches floppy disk drives - the small floppy disk drive ASIC chip still use open drain driver circuits mainly to keep backwards compatibility with older computers.
When I first got into the IT field, I had one of those type drives in m,y work PC. Then we endeed up going with the much smaller type. They were pretty good drives.
@48:45 my most self-loathing way of wrecking electronics is trying to put screws in and such while it is powered on ;) Did that several times in the late 1980's
Another great video! Fascinating watching you fault find, some great tips. It would be fantastic to see a similar video using a branded Watford or Cumana disk drive on one of your acorn systems (model B or Master).
This isn't an insult.. but does anyone else put Adrian on sometimes in the background and listen to an entire hour long video and if someone tested you would have zero idea what was going on? Guess I just appreciate his passion for vintage hardware, and for talking lol.
after a first full watch through, he goes in the playlist for back ground audio....usually catch missed info during the replays that comes in helpful later....
Compared to the size of the drive in this video, it is amazing just how small the floppy drive got to. Those 3.5" USB floppy drives on the market these days contain a full Teac HD drive that is only 1cm in height. Given a 3.5" disk is only 3mm this is a huge achievement. The floppy disk was one of the last century's greatest inventions.
Fantastic video Adrian, however when you performed the first format after replacing the faulty IC, the process failed with a fatal error on track 38. Your subsequent read attempts therefor failed at that track until you formatted the floppy in another drive. (and cleaning the drive probably helped a lot as well). Keep up the great work, very entertaining videos.
My TM100-2A actually has a metal retaining pin on the door. It came out of an old industrial enclosure together with a TM-503 hard disk and a couple of very nice PowerOne linear power supplies. I don't think it was ever used before I bought it.
It’s been 40 years but I’m pretty sure they redesigned the latch due to failures in the original design. I recall we had a bunch of spare latches in-stock and there was a change made along the way to make them more robust.
Lots of variants of the TM100, the most common: TM100-1 Single sided TM100-2 Double sided (there was a cost reduction A revision) TM100-4 Quad density 80 tracks 96tpi TM100-4M 100tpi Micropolis format variant, (could store almost a megabyte)
Im actually following along with this. Im restoring a 5150 and testing some parts out on another machine. I have a tm100-2a and the underside of the latch says 48tpi DSR...which makes mine a later double sided 360k unit.
Nice scientific-method troubleshooting the drive! Really appreciated the side-by-side testing/swapping cables with the working drive. Quite curious to see a followup with such rigorous troubleshooting between the two mobos, the greaseweazle, the XT-IDE, and the bios-extension ROM. I got lost, there, it seemed like all those problems were related...?
Nice fix! I wonder if the motherboard / freeze issue has to do with BIOS settings or compatibility? I think floppy controllers use an interrupt that might result in a system hang if set up incorrectly for the drive type.
That 74LS221 is known as a "one shot". When triggered at its input it produces a pulse of a preset duration set by the resistor-capacitor network attached to other pins on the chip. Looks like that 221 completely failed internally and stopped producing pulses.
That sounds similar to when in school I learned to make a 555 timer produce a single pulse with tuning the capacitors and resistors. Fun to know it exists in a single LS chip :)
Checking the unused output of the one shot is an excellent troubleshooting tip. Another useful tip for checking timer chips is to check the analogue waveform at the RC network. You should see the period of the delay.
5:20 FWIW, the ICs have date codes 8231, 8222, 8221, P245 etc. That latter one specifies 1982 week 45, so the PCBA would post date that. The equivalent time stamp for the finished drive (23rd April 1983) would be 8317 - that narrows down the manufacture of the PCBA to 24 week slot, probably early '83.
I had a Tandon TM-100-2A in my IBM PC, and about a week after I got it, EVERYTHING mechanically was out of alignment: both drive heads, the track 0 sensor, etc, and better yet I didn't have an alignment floppy. Yet somehow, over the course of over 2 weeks of on and off adjusting and testing it with both imd and my scope, I somehow aligned it all close enough to read even disks formatted in a 1.2MB drive every time. So yeah, they do go out of alignment, and it's almost not worth it to fix unless you have a lot of time on your hands to work on it.
I'll have to get one of those GreaseWeazel cards. I do have a CatWeazel (newest version) but had issues with system stability back when I was using that. I did have some success reading C64 disks, though, and formatting disks with various capacities, etc.. Also, at 55:55 there are a few typos on that NFORMAT! screen!
IDK I'm afraid of my stuff breaking. I probably don't know enough to fix any problems that aren't just bad caps. 😥 I recapped my Mac Classic. It still works, but i didn't manage to get off a single cap without lifting a pad. 😢 So i had to use hot glue to keep the new ones in place. But i guess it's better that having electrolyte leak onto the motherboard. You hadn't released your video about removing SMD caps at the time. I still need to recap the PSU. I also need to recap my PS/2 286 IBM PC, but I'm afraid of damaging it. I think i should be able to do it using the techniques you showed in that video though.
Yeah, I was just going to say this. It's completely ghetto all the chips popping out, the overclocked crystal bugging.😂😂 I mean it keeps us entertaint but maybe someone should sponsor him a new favorite 286. One that works.
You need a PAT2 from Dysan. Was a simple Z80 SBC that had all of the tester routines in firmware! I used to work at a DRIVE repair shop for years.. the PAT would allow for over 30 drives to be cleaned and aligned in an eight hour shift..
I encountered this "media unusable" followed by refusing to access the disk anymore repeatedly whilst trying to create 720k floppies to read in an Atari SF354. Even though the ST can read PC floppies, the PC didn't like to make them single-sided. Whenever I gave the FORMAT /F command it would do exactly what happened to you, and then I wouldn't be able to use that disk anymore :P -- including not being able to format it back to 1.44mb or write over it with imd/rawrite. Since I don't have an endless supply of 720k floppies, I finally ended up "fixing" those disks by reformatting them to IBM in the ST and/or using makedisk. It would be nice to solve this mystery -- the /F option looks quite straightforward but just didn't seem to work AFAICT, and it's just a weird nerd sniping kind of issue - I usually think of floppies as just be a block device, I wonder what could be getting written that screws up further writes? btw I don't have an XT-IDE in the 586 I was using for this work, that one's just your typical Win95 box with a generic 1.44mb drive, but in my 286 I do have one and whenever I boot off the C drive, the A drive is no longer accessible for some reason, so fwiw yes definitely there are still bugs in the XTIDE bios apparently related to floppies. I assume it has to do with it hooking int 13h somehow but haven't quite figured that one out yet either.
You probably missed my comment from the video where you had problems with the video card, it seems like the ISA bus connector fourth slot over had a socket separation problem. Wondering if you put the disk controller in the slot...
When I finished my apprenticeship as a TV Service Engineer back in 1986 the company I worked for had no vacancies for me - even though I passed all my exams and had a very good end of apprenticeship report. The TV rental market was dying so I re-trained as a computer repair engineer. My very first job as a computer engineer was aligning and repairing the very Tandon floppy drive you're working on in this video. It's all I did for the first 6 months before working my way around all the different peripheral devices and eventually repairing the actual computers. I've done this job for 37 years and now I am a network consultant. I am very happy to have got 37 years out of repairing electronics and consider myself very lucky. I continue to work on electronics as a hobby now - but I have reverted to working on valve (tube) radios (as part of my apprenticeship education - even in 1983 was based on valve technology) I still look at the occasional laptop or PC when I feel like it.
People that have the skills and experience on the old tube stuff are not getting more by the day. So great that you can help to preserve more historic electronics
Just shows how useful an oscilloscope can be, when you can see the data getting lost along the way it really helps with figuring out what's gone south... :)
Yes, it emphasises exactly how the real world is analogue, with noise and all the rest, and not just digital levels!
@@Zadster Continuous rather. Not really analog (or analogue), that term stems from the signal being an analogy to something else.
Adrian's Digital Basement videos should replace most computer class teachers in elementary and high schools. If I had this growing up I would have learned so much more.
@@jeremylindemann5117 And some “higher ed”.. I had a “class” on computer science after high school AND after working at a drive repair shop, ( yes we repaired both hard and floppy drives )
Any way.. the book stated that the index sensor was the track 0 sensor.. That and just about every other “fact” was WRONG.
I ended up teaching the storage section.
At first when I was correcting everything, the instructor was questioning where I was getting my information. When I stated I just came from a shop in Beaverton Or called “ Disk Drive Repair”, he got the idea.
I've got one of those! Manufacture date: December 27th 1982.
It, um, somehow made its way home from IBM, where I worked, one day and never went back. I don't even remember if it worked when I obtained it, but it was almost certainly heading for the landfill if I hadn't rescued it.
It's a fun relic to have around to show kids how far we've come in 40 years.
Was that from around the Greenock facilities, or had you moved over to America for IBM? :) (Inferring from your username!)
Nah, they’ll just ask, “Where’s the USB port?” 🤣
@@kaitlyn__L IBM Hursley, near Winchester in the 1980s, then Boca Raton, FL, and finally Austin, TX.
@@EnglishMike sounds like quite a journey! Thanks for indulging my curiosity.
@@kaitlyn__L No problem!
Wow, this brought back memories. In the 80s I worked for a Canadian dedicated word processor manufacturer that used SA400s and Tandons. Did tons of alignments. It was always either dirty heads or alignments. Every now and then a dead motor. Don't think I ever encountered any other issue.
The challenge of debugging and fixing things is most of what I enjoy about working on old devices, I almost never end up using them regularly afterwards… so many consoles collecting dust.
I'll buy something thats broken from Goodwill just to have a few hours of fun fixing it, good way to spend 5 bucks.
A constructive hobby, instead of getting high or drunk in your free time is key.
What a wealth of history, knowledge, exploration and education packed into one video. Thank you!
Wow, how did you know I was trying to repair one of these? Your psychic abilities could not be better timed.
The diagnosing and repairing of broken electronics is my favourite part of your channel. I actually don't have any special attachment to the retro tech itself. So, in the nicest way possible, I hope you've always got a pile of broken stuff to work on.
Thanks for the interesting videos.
Another great deep-dive diagnostic. Great use of Grease Weasel as a dignostic tool. Cheers Adrian!
The fact than you can fit an entire, powerful, system inside the chassis of that floppy driver, nowaday, it's incredible...
Every time I have encountered a fault with that multi vibrator in the design I have found it to be bad. They go bad frequently. I don't know why they go bad so often.. this video was a really good one! 😊
Ah that's interesting! I can't recall running into a bad one but I'm not sure how many times I've worked on something with one. I will need to store that tidbit somewhere in my brain :-)
I know the fault on the Plexus was a dual flip flop now I think about it.
AHAHAHA! The first disk drive for my home-brew Nat Semi 32008 using a Western Digital 1770 controller was a dodgy Shugart SA-400 fished out of a bin. The work around to dodgy msmv was wiring a bypass with an OR (or was it AND) gate!
Very nice fix! If you look at the datasheet of a floppy drive control IC (not to be confused with the floppy driver controller on the PC side) for 3.5" drives you'll find for example in the AN-917 datasheet page 3 figure 5 a schematic showing what the signal path is.
1. The head
2. Preamp (differential)
3. Low pass filter (the one on your schematic the opamp with the inductor)
4. Differentiator, turns the sine waves in more pointed pulses but still sine wave like
5. Zero crossing detector, turns the signal into short pulses
6. Time domain folter & pulse generator, these are the flip flops that transform the time between short pulses positive edge pulses into the final square wave where the edge going negative to the next edge represents time where 4us, 6us and 8us represent the flux transitions 10, 100 and 1000.
Nice to see this in discrete components instead of one IC that does it all! If such an IC fails then it's game over for the drive, these can't be found anymore.
I've done decoding of analogue signals of the flux transitions captured using an oscilloscope programmatically in an application that I wrote. Interesting stuff!
That the grease weasel has some retries could be that their decoding doesn't have much of a compensation for timing ie if there's variations in motor speed or a bump on the disk (the well known clicking sound) detection of the proper timings may be off. The hardware decoder in the floppy controller card is more robust in that regard.
utterly fascinating, got a bit lost in the logic area but don't care it was just brilliant to listen to someone who really knows what they are talking about and is so comfortable with it, well done
The power connector on the GW can be used to power a 5v floppydrive. Relatively new 3.5” drives are 5volt only. (I seem to remember coming across one that required 12v as well but it was a long time ago and I can be mistaken. I created an external GW drive using only USB-c that way. (I have a few type-c weazles). Furthermore let me take a moment to express my appreciation for you, Adrian; your videos are always very entertaining and informative. Not just “this is the fault and now it’s fixed” but also the backstory etc. Thank you!
Oh and of course, you can power the GW from a pc power supply using the connector, just disable USB power with the appropriate jumper.
I find early IBM machines quite fascinating, I have never used one before.
This is a gold mine of useful information! I wish that I'd known this stuff when I had a Tandy Model 4, back in the early 90s. Its disk drive stopped reading information and what did I do? I threw the whole computer away. What a bozo.
Hey back then we all did the same. The machines were basically junk (in our minds) so almost everyone just tossed them. Myself included!!!
Im very impressed with your videos.
Im trying to repair a C64 with 1541 and an Apple2e with duel floppy module.
Im learning SO much that I could have benefited from 30 years ago.
These retro computer videos remind me of so many things I don't miss about old computers.
It is a reminder of how difficult things could be back then. On the other hand you could fix stuff ... Now we have to replace entire massive components or deal with OS bugs that might never be fixed.
@@adriansdigitalbasement The worst thing I remember was buying a new board, plugging it in, and the computer wouldn't even boot. BUT I put it in my sister's computer and it worked perfectly. There was simply no explanation and no fix. You simply returned the board.
I just repaired my old 25 disc file CD changer and was proud of myself. Had to order belts from Belgium and clean up the rails. But had to disassemble the whole thing.
But watching your channel and others gave me the drive to try. They don't make anything like this anymore
Looking forward to watching this one. Just repaired a pair of Tandon TM100s from a Kaypro II. Bad TIP110s, speed controller, seized bearings.
One of them was rotating at 900rpm 😂
Overclocking a floppy. 😄
@@jasonhaman4670 🤣
Tune your floppy with nitro.😁
Ah yes I've had at least one drive lose speed control due to the TIP IC.
@14:09 not sure if it will get tested further into the video, but before I forget: measuring the resistance of the read head is also one of my first checks. Usually each wire pair should be between 10-20Ohms. If you see 0 or infinity or even kOhms, one of the coils is broken. Then you will have a very bad time indeed.
That audio edit though...seamless 😂
you just boggled my mind thanks i learned a lot i love my computers but most of them are broken i dont know if i will get back to my hobby you gave me inspiration
Yes, Adrian's determination in the face of the problems gave me hope for my "museum". I like that he labels the new issue with the 286 not as frustrating but an opportunity for more fun in the future.
1:02:34. Never seen anyone so excited to have more broken things to fix.
I actually have a late revision 5150 made in 1983 and it shipped with a single sided drive. It also just had 64K of RAM and MDA.
I managed to find mine as an unused boxed example so I know the drive is original. The manual in the box also contains DOS 2.1 on single sided disks.
I believe IBM continued to offer a single sided drive late into production as an ultra low cost option before finally being phased out in the XT
Very interesting! It feels like the exact details on the single sided drive has been lost to time.
Excellent video! I have purchased a Greaseweasel, and intend to learn more about it than just to use it for transferring files.
the st johns + atari mashup...
0:43 - nice voiceover 😉
Good ol' Festival TTS.
I wasn't expecting that at all and I busted out laughing 😆
I have no retro computers in my collection, I watch you so I don't have to, but I used some of this stuff back in the day. 😁
This is a better drama than a lot of the ones coming out today!
I started at Computerland in late’83 and had my hands in hundreds, maybe thousands of these machines before they went to customers. I can tell you none went through the config area with single-sided drives by then. I find the April ‘83 date interesting.
Perhaps this was a spare for replacing an original machine's broken single sided drive? That really is the only thing that makes sense at this point.
@@adriansdigitalbasement Agreed. I didn’t consider that. I do remember seeing only a few machines with the black power supplies (recessed nuts) come in for service but don’t recall the type of drives. This is all sort of a fun rediscovery for me.
As I recall, a few years in IBM started using another floppy drive manufacturer (I believe they called it a type 2). Haven’t seen one of those around but they were not good and would be an interesting repair. If memory serves, they had no track zero switch.
This seems very similar to an issue I had with an old and defective Chinon FB-354E in my very early Amiga 500 (with space invaders keyboard). It stopped working, and after probing around I found after amplifying the signals, they run it through a Nand gate, a simple 7438. Probably to make a proper digital signal from it. It turned out to be the issue and in the end it was a 50 cent repair. I repair lots of stuff, but I was so happy to get this fixed as I definitely wanted to keep the original drive in there, even if I have several replacements.
Great work in video too, that greaseweasel seems like a nifty little thing!
Man I’ve just been loving your videos the past year. Everyone is great before bedtime :)
Reminds me of working on the c64 1541 when reading disks was intermittent; after pulling down the case and looking at the head assy, a small pop out felt pad would wear over time and not compress the disk against the head causing intermittent reads. Couldn't find anyone or place to get a new pad, bought some felt, hole punch, and some double sticky clear tape. Good to go. Great memories.
Having watched 2:55 I can say that I very much look forward to seeing your video on the Grease Veasel (?) 👍
Wish i was like you and had tons of retro computers lol. Very late too the game if wanted to start now, after getting a house few years back pretty much any extra cash goes into it. So thanks for all the great video's of the things i grew up around and never really new how they worked. I was i outdoor kid was in the woods more then inside growing up.
PS never got a tick as a kid with no spray on even now go too cut my grass and have gotten them on me.
That was fun! And the last 10 minutes of your video sums up what every retro-enthusiast just has to learn on their own with "quirks"! 🙂
It would have been easy to edit that part out to make the repair appear all neat and tidy and I'm so glad it was left in. It showed the reality of our passion where the equipment is so old there can be multiple simultaneous faults. I'm glad we live in a time where these computers are old and rare enough to be interesting but still common enough that folks with knowledge and grit can repair them. Thank you Adrian for showing us how it is done!
my hardest lesson in electronics happened when I stepped barefoot on a 555 that fell on the floor... It hurt more to pull it off my heel than it hurt when coming in!
Nice workshop!
Thanks for the explanation of the floppy drive and the troubleshooting.
11:30 "modern disk drives" lol, great video as always, very informative, you're a great teacher
Stepping on upside down ICs might actually be worse than legos.
Bad for both the foot and the chip. LEGO be like "I'm still alive mofo."
@@rommix0haha yeah feels like you could drive a steam roller over a LEGO block and the block would be just fine. 😮
@@adriansdigitalbasement Indeed. it'd be the steam roller that would get crushed lol.
Great troubleshooting and tutorial!! The mother board issues at the end added more "fun" 🤔
Awesome utility and board! Not that I have a lot of legacy floppy drives around (a few) - but this is an awesome project - thanks for showing it!
the Microsoft Sam voiceover made me do a spit take. i have too many interests and i end up watching several videos of various you tube channels in a row. i greatly enjoy Your content Adrian. And i look forward to your next repair adventure.
This sort of stuff gives me flashbacks of when I had to learn about disk formatting low level in the 90s by books from the library because nobody I my friend group knew why mac disks didn't work in IBM compatibles.
Ended up finding a System 7 program that would read IBM formatted disks on my Powerbook 100 so I could do homework on it because Word on the Mac was the GOAT.
These kinds of repairs are so satisfying!
U can use the decal on the bottom of the spindle. When at the correct speed the pattern will seem to stop.
There are two patterns. One for 60hz and one for 50hz
I grew up in Boca Raton, home of the IBM division that created the PC, and saw loads of 5150s. I owned four of them over the years, too, all acquired used, but somehow I never even encountered a single-sided drive, ever, in Boca Raton. So, interesting to see.
That's cool. My 5150 is an original early 16-64k machine I bought from the original owner here in Portland. Even my machine had the drives upgraded at some point or came with double sided drives... So in other words I've never seen a single sided IBM branded drive either.
My Tandon TM-100 is still sitting on the shelf, so it's always fun to grab it and follow along with the video. It's a -2A model with a "48 TPI DSR" label under the latch, and a build date of 3-9-84. I've decided that's a US American date format, which means it's a little older than I am, so I don't feel quite as old as I would if it used the correct date format.
My guess the problems at the end were related to the XTIDE BIOS. If they've patched the floppy routines, it's possible those routines are bugged in a way that gets the drive into a bad state, and it's returning unexpected data that DOS doesn't understand.
Very nice repair of an old floppy drive.
It occurred to me that this old floppy drive makes a lot of noise and buzz and the newer, faster computers can't filter it out and it interferes, causing errors and freezes.
Great floppy drive diagnostic and repair video!
22:19 just goes to show that this drive gives zero flux about anything....
😅
This is more exiting as any Detective Film i have seen.
3.5 inches floppy disk drives also do have terminator resistors on-board as well. I took a few apart and indeed they're terminated so if you remove that SIP resistor pack, the issues would basically be the same as older 5.5 inches floppy disk drives - the small floppy disk drive ASIC chip still use open drain driver circuits mainly to keep backwards compatibility with older computers.
With all those weird issues it seems more like Adrian's Diabolic Basement
When I first got into the IT field, I had one of those type drives in m,y work PC. Then we endeed up going with the much smaller type. They were pretty good drives.
I only found your channel a few weeks ago and I'd like to thank you for uploading some great content on vintage computers. Cheers!!
@48:45 my most self-loathing way of wrecking electronics is trying to put screws in and such while it is powered on ;) Did that several times in the late 1980's
Another great video! Fascinating watching you fault find, some great tips. It would be fantastic to see a similar video using a branded Watford or Cumana disk drive on one of your acorn systems (model B or Master).
This isn't an insult.. but does anyone else put Adrian on sometimes in the background and listen to an entire hour long video and if someone tested you would have zero idea what was going on? Guess I just appreciate his passion for vintage hardware, and for talking lol.
Plenty of other channels for that. Political commentary as background noise ftw.
after a first full watch through, he goes in the playlist for back ground audio....usually catch missed info during the replays that comes in helpful later....
Compared to the size of the drive in this video, it is amazing just how small the floppy drive got to. Those 3.5" USB floppy drives on the market these days contain a full Teac HD drive that is only 1cm in height. Given a 3.5" disk is only 3mm this is a huge achievement. The floppy disk was one of the last century's greatest inventions.
I look forward to you working on the high speed tape drive in the Coleco Adam! And the floppy drive
Fantastic video Adrian, however when you performed the first format after replacing the faulty IC, the process failed with a fatal error on track 38. Your subsequent read attempts therefor failed at that track until you formatted the floppy in another drive. (and cleaning the drive probably helped a lot as well). Keep up the great work, very entertaining videos.
Turn off ROM shadowing whenever you get weird hardware issues like this. Accept the performance hit for greater compatibility.
Extremely helpful! Thanks for all the good info.
My TM100-2A actually has a metal retaining pin on the door. It came out of an old industrial enclosure together with a TM-503 hard disk and a couple of very nice PowerOne linear power supplies. I don't think it was ever used before I bought it.
It’s been 40 years but I’m pretty sure they redesigned the latch due to failures in the original design. I recall we had a bunch of spare latches in-stock and there was a change made along the way to make them more robust.
Lots of variants of the TM100, the most common:
TM100-1 Single sided
TM100-2 Double sided (there was a cost reduction A revision)
TM100-4 Quad density 80 tracks 96tpi
TM100-4M 100tpi Micropolis format variant, (could store almost a megabyte)
I lol'd at the TTS correction voice at 0:44
Great Job, Adrian!!
I have learned a lot from you again. Thank you very much ;-)
Georg from Germany
Im actually following along with this. Im restoring a 5150 and testing some parts out on another machine. I have a tm100-2a and the underside of the latch says 48tpi DSR...which makes mine a later double sided 360k unit.
Nice scientific-method troubleshooting the drive! Really appreciated the side-by-side testing/swapping cables with the working drive.
Quite curious to see a followup with such rigorous troubleshooting between the two mobos, the greaseweazle, the XT-IDE, and the bios-extension ROM. I got lost, there, it seemed like all those problems were related...?
power jack on the greaseweasel is probably for passing 5v to a 3.5" floppy drive from USB
Yes, that's what it does. Many "modern" 3.5" drives don't need 12 volts at all.
*You are a wizard!*
Nice fix! I wonder if the motherboard / freeze issue has to do with BIOS settings or compatibility? I think floppy controllers use an interrupt that might result in a system hang if set up incorrectly for the drive type.
Awesome video! thank you
That 74LS221 is known as a "one shot". When triggered at its input it produces a pulse of a preset duration set by the resistor-capacitor network attached to other pins on the chip. Looks like that 221 completely failed internally and stopped producing pulses.
That sounds similar to when in school I learned to make a 555 timer produce a single pulse with tuning the capacitors and resistors. Fun to know it exists in a single LS chip :)
I can build a one-shot out of redstone, if anyone needs me for their Minecraft floppy disk drive.
So that's what they're called! Hopefully I'll remember that search time for the next time I need this.
Checking the unused output of the one shot is an excellent troubleshooting tip. Another useful tip for checking timer chips is to check the analogue waveform at the RC network. You should see the period of the delay.
The best Channel
5:20 FWIW, the ICs have date codes 8231, 8222, 8221, P245 etc. That latter one specifies 1982 week 45, so the PCBA would post date that. The equivalent time stamp for the finished drive (23rd April 1983) would be 8317 - that narrows down the manufacture of the PCBA to 24 week slot, probably early '83.
Hoping for an Apple II floppy disk drive repair so I get mine working! Got some hints from this video.
We are in the golden age of Adrian's basement, thanks Adrian.
I had a Tandon TM-100-2A in my IBM PC, and about a week after I got it, EVERYTHING mechanically was out of alignment: both drive heads, the track 0 sensor, etc, and better yet I didn't have an alignment floppy. Yet somehow, over the course of over 2 weeks of on and off adjusting and testing it with both imd and my scope, I somehow aligned it all close enough to read even disks formatted in a 1.2MB drive every time.
So yeah, they do go out of alignment, and it's almost not worth it to fix unless you have a lot of time on your hands to work on it.
Dousingle sided voiceover dub hits different.
I'll have to get one of those GreaseWeazel cards. I do have a CatWeazel (newest version) but had issues with system stability back when I was using that. I did have some success reading C64 disks, though, and formatting disks with various capacities, etc..
Also, at 55:55 there are a few typos on that NFORMAT! screen!
IDK I'm afraid of my stuff breaking. I probably don't know enough to fix any problems that aren't just bad caps. 😥 I recapped my Mac Classic. It still works, but i didn't manage to get off a single cap without lifting a pad. 😢 So i had to use hot glue to keep the new ones in place. But i guess it's better that having electrolyte leak onto the motherboard. You hadn't released your video about removing SMD caps at the time. I still need to recap the PSU. I also need to recap my PS/2 286 IBM PC, but I'm afraid of damaging it. I think i should be able to do it using the techniques you showed in that video though.
Is that the 286 board that has had it's clock tinkered with, and has everything been set to default when running these controllers and such?
Heh heh, for a second there, it looked like you created the world's first hardware virus.
Enjoyed this, and learned! Big Like Button!!
I don't know why that thing is your "favorite 286 motherboard" (59:08). It seems to have nothing but problems.
Yeah, I was just going to say this. It's completely ghetto all the chips popping out, the overclocked crystal bugging.😂😂 I mean it keeps us entertaint but maybe someone should sponsor him a new favorite 286. One that works.
Cable offset can usually erase the current track. The write gate get turns on
Is that the same 286 board that you had to reseat the keyboard controller 3 times? It might be cursed.
Neat, I figured it would have been the flip flop, with all the issues you've had in the past with PAL encoders going bad 😅
You need a PAT2 from Dysan.
Was a simple Z80 SBC that had all of the tester routines in firmware!
I used to work at a DRIVE repair shop for years.. the PAT would allow for over 30 drives to be cleaned and aligned in an eight hour shift..
The greaseweazle does everything the PAT2 can, but it also does imaging. And it isn’t unobtanium!
I encountered this "media unusable" followed by refusing to access the disk anymore repeatedly whilst trying to create 720k floppies to read in an Atari SF354. Even though the ST can read PC floppies, the PC didn't like to make them single-sided. Whenever I gave the FORMAT /F command it would do exactly what happened to you, and then I wouldn't be able to use that disk anymore :P -- including not being able to format it back to 1.44mb or write over it with imd/rawrite. Since I don't have an endless supply of 720k floppies, I finally ended up "fixing" those disks by reformatting them to IBM in the ST and/or using makedisk. It would be nice to solve this mystery -- the /F option looks quite straightforward but just didn't seem to work AFAICT, and it's just a weird nerd sniping kind of issue - I usually think of floppies as just be a block device, I wonder what could be getting written that screws up further writes?
btw I don't have an XT-IDE in the 586 I was using for this work, that one's just your typical Win95 box with a generic 1.44mb drive, but in my 286 I do have one and whenever I boot off the C drive, the A drive is no longer accessible for some reason, so fwiw yes definitely there are still bugs in the XTIDE bios apparently related to floppies. I assume it has to do with it hooking int 13h somehow but haven't quite figured that one out yet either.
I never saw a TRS-80 MIII with out of alignment drives but I have come across Commodore drives out of alignment. All that knocking does add up.
It appeared you got an invalid low level format on the first track that confused DOS. When you read the track it was showing invalid 254 track number.
Mr. Shugart s suckerpunch: his hdd company is still around, but you know their new company name: Seagate ❤
You probably missed my comment from the video where you had problems with the video card, it seems like the ISA bus connector fourth slot over had a socket separation problem. Wondering if you put the disk controller in the slot...
Now do an Apple Twiggy drive, Adrian! 🤓
Great video!