Apple IIc floppy drives can be fickle things
Вставка
- Опубліковано 7 лис 2024
- Apple II disk drives are interesting things that don't work in exactly the same was as normal and common PC floppy drives. So let's try to fix this drive that doesn't seem to be able to read any floppy drives.
-- Links
Locksmith:
mirrors.apple2...
Greaseweazle:
github.com/kei...
Barn found Apple IIc machines:
• Apple Archeology: Four...
Tandon TM-100 Reapir:
• Repair of an IBM PC fl...
Adrian's Digital Basement Merch store:
my-store-c82bd...
Adrian's Digital Basement ][ (Second Channel)
/ @adriansdigitalbasement2
Support the channel on Patreon:
/ adriansdigitalbasement
My GitHub repository:
github.com/mis...
-- Tools
Deoxit D5:
amzn.to/2VvOKy1
store.caig.com/...
O-Ring Pick Set: (I use these to lift chips off boards)
amzn.to/3a9x54J
Elenco Electronics LP-560 Logic Probe:
amzn.to/2VrT5lW
Hakko FR301 Desoldering Iron:
amzn.to/2ye6xC0
Rigol DS1054Z Four Channel Oscilloscope:
www.rigolna.co...
Head Worn Magnifying Goggles / Dual Lens Flip-In Head Magnifier:
amzn.to/3adRbuy
TL866II Plus Chip Tester and EPROM Programmer: (The MiniPro)
amzn.to/2wG4tlP
www.aliexpress...
TS100 Soldering Iron:
amzn.to/2K36dJ5
www.ebay.com/i...
EEVBlog 121GW Multimeter:
www.eevblog.co...
DSLogic Basic Logic Analyzer:
amzn.to/2RDSDQw
www.ebay.com/i...
Magnetic Screw Holder:
amzn.to/3b8LOhG
www.harborfrei...
Universal ZIP sockets: (clones, used on my ZIF-64 test machine)
www.ebay.com/i...
RetroTink 2X Upconverter: (to hook up something like a C64 to HDMI)
www.retrotink.com/
Plato (Clone) Side Cutters: (Order Five)
www.ebay.com/i...
Heat Sinks:
www.aliexpress...
Little squeezy bottles: (available elsewhere too)
amzn.to/3b8LOOI
--- Instructional videos
My video on chip removal without damage:
• How to remove chips wi...
--- Music
Intro music and other tracks by:
Nathan Divino
@itsnathandivino
If I had to wager a guess, I'd say that the lubrication on the stepper motor was super sticky and exorcising the drive loosened it up....
My thoughs exactly
Even if you slid the carridge back and forth a few times and it looks like it now works perfectly, it in fact often is not perfect enough to work. You need the drive itself to seek from min to max a few dozen of times before it starts doing its thing. The same goes with some early harddrives.
If we are correct, this drive will work fine unless it's left in dusty environment for years again
Since it sometimes struggles with innermost track I suggest trying to manually slide it in that edge area with power off at least a dozen of times, sliding past where the innermost track normally is if the drive allows that.
Always make sure you get the demons out of your Apple II floppy drives, that's usually what is stopping them from working.
@@DavidMarvin I CAST YOU OUT
@@DavidMarvinI’m so happy you posted this. I know they meant exercising, but imagining Adrian performing an exorcism on some old and haunted floppy drive made me laugh out loud.
"The power of Woz compells you!" 😂
My dad used to repair televisions for a living and I often helped. He always used a cold spray on components to find a bad transistor or chip. Works wonders.
Anyways, love your channel. Always great videos.
My ex's dad used to do the same also car radios etc, the amount of times she would come home to find cords cut (he needed one!), or lose cassettes in tape players he would test with!
Great episode as always. The Plexus in the background really is killing me though. We need an update about the project and the company engineer you got in contact with
I'm the Plexus, and I don't work anymore. Sorry. Too tired.
It's fascinating to me every time. I have no interest in retro computing, not in Apple devices or electronics.
But I love watching you, inspect this old things and repair them. ❤❤
I used to repair recent-ish computers - like last 20 years when you could replace ram, a screen, even a GPU on a laptop if I put in alot of work...this old stuff is beyond me! I love listening to and learning though
Even "nothing-burgers" are tasty when you're the chef, Adrian! 🙂
Just make sure you don't invite Trump to the cookout.
In locksmith, A stands for Address field error (missing or bad checksum). D is for the Data field error (missing or bad data checksum).
The Address field starts with D5 AA 96. The Data field starts with D5 AA AD. Both Address field and Data field end with D5 AA EB iirc.
The deep dive into floppy drive head signals was well done. Thanks again Adrian.
I love troubleshooting videos that leave you flabbergasted because... well, that's real life sometimes. Thanks for not editing out the real life moments, Adrian!
I am watching your video while repairing a commodore 1541 disk drive !
Keep up the good work !
Despite not catching the exact issue this was a really comprehensive and interesting video so don't feel bad Adrian.
woz is amazing!, never would have been an apple with out him.
Don’t forget to re-create that Fat City disk with the Greaseweasel so you don’t get false positives when testing Apple II’s down the road.
Yeah good call! I need to image and recreate that floppy again -- as it has a bunch of saved high scores on it I want to preserve
This was a funny video with the 2nd floppy drive finally working after it was cleaned then running a several times back-to-back. That's how things go at times, as you know. Technology can be quite temperamental at times. glad to see that it does work for you now.
Hi Adrian, love your videos, I like vintage computer videos, my first computer was a zx spectrum, later an amiga 500 and then all kinds of pc computers, never use an apple computer, the best thing for me of your videos it's your English spoken it's so clear that I can understood all that you say with no effort. Thanks for spend your time with all of us. Cheers.
I wonder if what "self fixed" was the steppers for the head which may have been slightly gummed and was not aligning properly with the track when it was going to some of the inner tracks, that would explain why it was going from outer to inner with less and less error reading the more you exercised it
Apple tracking is weird, which is why companies came up with some very interesting copy protection schemes. For example, when the head "steps," it is actually being stepped twice. So what some companies did was write a custom dos, that once booted into, would zero the head, then step it 3 times for track two, instead of only 2. Then it could step twice like normal, but the heads are halfway between the normal tracks. Which is why this method was called "half tracking."
And there was enough play in the head mechanism, that if you single stepped (half track) then very quickly stepped back, you could actually get 1/4 tracks.
And then some bright bulb thought up "track arcing," which starts a track, then partway through quarter steps it, then again, then again. So you'd end up with on track, +1/4, half way, +3/4, then the next track. Some copy programs would scan the entire drive in 1/4 track increments, so you could see where the data actually was.
Copy ][+ was the best, IMHO (i think that was the name, it's been a very long time). But even locksmith 6.0 (which is what Adrian was using) also had 1/4 track indications. Going by 1/4 tracks and looking at the results, you can tell which "track" has the strongest signal/fewest errors. Once you know this, you know which way to adjust the heads. Then it's just a very tedious process of trial and error till you get the alignment "right." But it will only be "right" for that specific disk.
Which is why if you really want to align a floppy disk properly, you needed a calibrated disk and an o-scope. The calibrated disk is not readable by the machine, because it has analog tracks written to it instead of data, which is why you needed the o-scope. I used to do a LOT of PC 5-1/4 disks back in the day (Of all types, SS, DS SD, DD). Saved many a full height drive from the scrap yard simply by aligning it. We did so many, that eventually i had a whole pile that customers just wanted replaced, and we swapped for a really low price. When i had a lot of dead time, i would just calibrate floppy drives.
I wonder if the bearings on the motor were a little dry and loosened up, causing the speed to be more consistent? Still a fun time watching this!
It wasn't that as when I first tested the IIc machines I checked the drive speed via the sticker and it was spot on. That was when I lubricated the head sliders and made sure that was working well.
I think because of the drive fixing itself it made your video even more interesting. Makes for a interesting story imho. Please keep posting those kind of videos, even if they don't go as planned while recording.
As for why it fixed itself, I would suspect old grease in some head moving component.
The replacement keycaps on that IIc are perfect!
When the //c came out, Apple introduced the DuoDisk for the //e at the same time, with the DB-19 connector (same connector as the//c external floppy). Later, Apple introduced the "UniDisk" 5.25, which had a daisy chain DB-19 connector, and could also be used as drive 2 on the //c. All of these drives used the same reduced height drive mechanism, compared to the Disk ][,
What amazes me is that the 4 barn find apple II work 100% basically without repairs.
Indeed!!! Just really some bad key switches which is obvious why those would be bad!
Not sure if anyone has mentioned this - but this kind of looks like its related to the "reforming" of the electrolytics once current started flowing again...
I love this guys videos and have been watching them from time to time. I have an Apple II+ floppy drive that needs some love and get working as it hasn't worked in years, My goal is restore it and the whole machine.
It's fixing itself Adrian, because it's scared of your repair skills. It has no chance staying broken and is just giving up.
So you're saying Adrian is the Chuck Norris of vintage computer repairs? Yeah, I guess you're right...
It's the Tech Aura.
It might be a mechanical issue rather than electronic. You cleaned the heads, but what about the rails? I've seen the same sort of issue crop up on other hardware when after kinda warming up and working the old grease, it starts to work again.
@24:00 it may be the caps re-forming a bit from the power up time.
I don't think so, the drive started working when it started reading disks, improving with reading disks, if it were the caps it would do it just by being plugged in, and improvements would not be *this* quick
@@jwhite5008 it depends were the caps are connected, you are assuming they are directly on the supply rails, they may be on sub rails that are not always active, or may even be used in R/C circuitry.
I genuinely look forward to each and every video you produce, man!
Another great video. Sometimes this happens when you work with physical devices. They have a will of their own. 😀 Sometimes these old units just need a little workout to get the kinks out just like us.
For fine tuning sensitive potentiometers a simple aid is to cut out a cardboard arrow and push the screwdriver through it. That way you can see a tiny rotation translated into a much larger motion at the arrow tip to help with micro adjustments.
The "A" and "D" means errors reading sector address or sector data, so the "D" means that the drive could read the address of the sector but the data has a bad checksum. Sectors in the disk are stored in two parts with a gap in between, so when you write one sector it search for the matching address and then switch to write mode to write the data only.
I also think that the electrolytic caps were the problem 🙂
i am a fan of theme music. yours is right up there with TNG, Sanford and Son, and a host of other shows that have the perfect theme music.
YUP - this was a great vid!!! Even "nothing burgers" are awesome....."Adrian's Troubleshooting Mind".....good stuff.....
I kinda wanna watch you screw with the alignment just so we can watch you fix it. I'm sure you can fix it! It'd make a great video!
Same thing happened to me on my IBM 5155's Tandon drive B. It was so seldom excercised that it got "unhealthy". Thorough tests for one full afternoon got it in back in top shape again.
On the first disk drive you connected with a external cable: my suspicion was that it was spinning a bit slow due to gummed up bearings and when you ran a lot of tests you put some heat into the mechanism and the stiffened up lubricants got working again and thus it sped up to its rated speed. Writing this while you are still working on it tho so might be you find another conclusion.
When something like this "fixes itself", it's usually something on the mechanical side that wasn't working properly. My guess is as others have said, the heads were not tracking properly. What we know/saw:
1. This drive does not have any feedback from head positioning. The stepper motor is connected to the head assembly via a clutch, which is why the "head banging" is needed since the clutch is designed to slip.
2. We had dirt. And grunge. In abundance.
3. The drive started reading correctly from track 0, and got worse the higher the track number.
4. The more it was tested, the higher the first bad track became, until all tracks read correctly.
My theory is that there was some part in the head positioning system that had a bit more drag than the clutch would allow, meaning that the clutch would slip just a tiny bit on each head move. As the head assembly was exercised, whatever grunge was causing the drag was being cleaned off, reducing the drag (and clutch slippage), and allowing higher number tracks to be read.
I would still recommend cleaning every part of the head positioning mechanism, then re-lubricating. This drive should be good for another 40 years.
that color is amazing!
I'd wager the issue with that drive is the mechanism moving the head assembly. I think this because the more it moved the better the reading got, I think I could also hear the difference in it while it was reading. It'll probably get gummed up again after sitting for some time.
For drive alignment there are special disks with analog signals on and you need to connect up a scope to a test point. For an example of this, curiousmarc does it on a 8” drive.
Sometimes with clogged heads you just have to run it a while to sweep out the blockage. Also, usually I clean laterally, not perpendicular. I once read that there is less risk of fibers from the swab getting caught in the head gap.... and scrub a lot more than you think you have to. Over decades the contaminants seem to form a film on there.
On some industrial kit I used to have a special disk to exercise the drive before we downloaded data or systems upgrade due to these problems. The other issue is the belt as not being used for a long time losing elasticity or creating uneven running by a bump where the motor was.
I just think your drive was the head movement as the lubricant goes off & drives motor bearings having the same issues!
Even Adrian's nothing burgers are awesome videos
Does anybody else recite " hello everyone welcome back to Adrian's digital basement" with Adrian as he says it in the intro?
most informative and educational "nothingburger" :D
Capacitors!
On thing that amazes me about Apple II drives is the belts. They are incredibly robust. I've never seen on break or disintegrate. Incredibly, I've never even seen one slip off the large rotor, even though there is no rim and no ridges to hold it on.
Regarding the drive that fixed it self. Probably gunk on the rails. As the head would approach the center it would get slightly out of alignment because it cannot move as freely.
I think Adrian cleaned and lubricated the rails in a previous video. And the typical mechanical maintenance is something he does to all the drives by default..
@@Colaholiker Even if you lubricate the rails, if it was really crusty you need to work it back and forth many many times, a few dozen times at least, and then it gradually smoothes out. Which is exactly what happened.
@@jwhite5008 you do. But I am sure, Adrian is well aware of this. He has serviced so many floppy drives on and off camera... 😉
Curiousmarc has some videos aligning floppy disk drives (8" ones in this case) using an oscilloscope. Interesting stuff, comes with free elevator music.
Great video. I "repaired" my apple iic drive by finding a heap of write protedt stickers which had fallen off and were preventing the heads from reaching some tracks!
Looking forward to a greasweazel video. Are you using a 360k or 1.2Mb drive to write the fat city disk? I dont have a 360k PC drive, and can't write a perfect image using the 1.2Mb PC drive I have using greaseweazel.
I'm coming to this a few days late so apologies if this has been mentioned but my guess would be the capacitors 'reforming'. This can happen after long periods and just switching on for a while causes them to reform which I am sure is something Adrian has mentioned on CRT videos so I am surpised he didn't suggest that.
those computers were at the tail end of life when used them in grade school
If you notice it started to read the disk better after you use locksmith without a disk, I’m wondering it the stepper motor was sticking and now you have exercise the stepper motor it starting to work right
Mecanical problem the belt slightly slipping but the rollers get progressively cleaner and the belts glazed surface scuffed up.
Great episode 😊
But, Adrian, we LIKE those silly old things!
Maybe an idea for another episode: It would be interesting to know how the Amiga could read and write MS-DOS or Apple 3,5"-inch disks and how this has been made possible. And how could the Amiga boot from Non-Amiga-DOS disks?
Nice video as always brother, keep updating us😊
Adrian, are your test programs uploaded to an easily accessible site? My wife bought an Apple III (yep, 3) at a yard sale and I suspect that its floppy drive will need attention/testing, *after* the PSU is recapped.
As noted in other comments you might have had some debris on the rails which slowly got shifted as you tested the drive. Another issue I’ve had before is too much downwards pressure from the felt pad - I had to loosen up the spring at the back of the head.
in the late 80's 90's i remember using a scope and look for a cats eye pattern to align the drive
"I had a terrible dream! Just a bunch of ones and zeroes... and I think I saw a two!"
I wonder by allowing it to run some electricity through the boards it may have reformed the caps and exercised the motors allowing them to relubricate which allowed things to start working.
I was thinking maybe the drive head was magnetised in some weird way, but reading from disk a bunch of times somehow made it wear off? That or gremlins.
ALPS switches tend to need cleaning after so many years. they weren't very dust and dirt resistant.
I know you do retro computers... but if you ever watch "will it run" videos of old cars that have been exposed to the elements. It frequently occurs that heat cycles will clear up gummed up parts. I suspect this to be the case with the drive. Oh, and I'm just a self-professed armchair expert, pay me little mind. ;)
When I was a kid I ended up with a zenith ez pc. I believe it's double density drives. They are 3.5". They sometimes made very odd noises and didn't work at all. It wasn't headbanging. I know what that sounds like. But every time I had it looked at, it was perfectly fine. lol
Better a nothing burger than an air biscuit. 😂
My bet is caps reforming during use. I'm no expert, but seems likely.
It's really the only explanation! I need to find some schematics to see how they are used on the drive. There are only two or maybe three of them on the drive.
I've seen that happen. Leave the power on for awhile and the capacitors seem to acquire the correct values again. But I thought that only applied to electrolytic capacitors. I could easily be wrong about that.😊
@@adriansdigitalbasement Electrolytics are usually relatively slow, so I doubt they would be in the direct signal path. More likely they are part of some power rail regulation/stabilization circuit.
However, if they needed reforming, they could have leaked (electrically, not in the corrosive sense) in the beginning, and maybe a voltage rail needed by the analog voodoo on the board was therefore running on the lower side. With them reforming, the voltage came up and things went back to working normally.
Since it didn't go back to bad even when turned off for a while, I doubt that it's a thermal problem with any of the ICs not working well while cold.
But I'm just assuming and drawing conclusions, I have never even seen an Apple //c in person. They were not particularly common in my neck of the woods.
@@adriansdigitalbasement I say it's the rail lubrication smoothing itself out
With caps it goes a little worse when you not run it for a while, here we could not see it.
Furthermore, it started to fix itself when the disk started to be read, not when the drive started spinning, i.e. it went better with every read, and not just as it sat there.
I highly doubt it's the caps.
The witch would check lubrication (bearings and stuff, gunk plainly speaking) and then she'll move to mechanical stuff that might be not properly aligned. She'd wait for one or more day, and then some weeks, to check if the 'getting better' phenomena would disappear.
39:48 It may seem obvious to say, was Broderbund Software had such a huge presence for us millennials and gen x alike.
My favorite creative tools from them were Kid Pix and Fantavision (the vector animation software).
I'm at about 30 minutes in, where the drive is "starting to behave" and my thought is that there's some kind of heat soak going on. The drive warms up, the chips warm up, the caps warm up, or something gets the grease going and it's able to pick up on the signals.
Either that, there's some very slight sticktion that's working itself out with moving the drive head?
Hi Adrian. Maybe some day, you could test those “Mac” mice and find one which does work on the //c, and the one you have now (which doesn’t). Then see if you can take them apart and find out why. Maybe they have a different chip on the inside, or maybe the mouse chip has a different clock speed? Maybe bad caps? Bad pull-up or pull-down resistors? Who knows, but it would be interesting to know and do a video on it. 😊
It's was figured out a while ago (hence the adaptor)
Mac shorts the data line to ground with pullup, and 2c shorts it to VCC, some revisions kinda work enough, some are confused to be a joystick.
If interested, google (with quotes): "Using a Macintosh 9-pin mouse on the Apple IIc"
Headbang🤘
26:19 I’m betting there is a cold solder joint or something on it that is working once it gets warm. I bet if it’s allowed to cool down/be unplugged a while it will go back to not working at first… unless dust in a track or motor is being worked out.
Electrolytic capacitors will reform when used. Probably what happened. I remember replacing a tantalum with an electrolytic in my telescope mount, and it took it a long time to boot the first go. After that it was normal.
The stepper motor had old lube that loosened up in time. Seen it before.
I freakin' click 'like' the second the video starts!
Are you ever going to try resurrecting a machine that uses the S100 bus?
5:14 That little write protect opto reminded me of a discussion that I had with Trixter(Jim Leonard) some years ago. His PCjr's FDD was stuck in write protection and I asked him if he checked that circuit, but his response was that he already got rid of the drive. The conversation was probably a decade ago at this point, but the loss still makes me feel sad.
Maybe a capacitor reformed or the head was not aligned and then as it was exercised it started correcting the alignment
drive probably needed demagnetization
I had issues with the head being slightly magnetized. I made a makeshift coil, hooked it up to an ac 9V transformer and waved it over the r/w head. It started reading disks again and worked ever since. Maybe, you had something like this here.
I don't think you can really demagnitize the head (the metallic housing itself) this quickly by just reading disks?
I've seen apple take off the standard controller boards and put on their own. Could you just put a standard board back on and use imagedisk ?
The stepper motor might have needed some exercise to get it accurate again. That's my two cents.
The more Apple II gear saved the better!
That drive might have the auto aligning circuitry that allows for very MINOR adjustment of the read/write head (which allows for the error allowances)
The Drive: "Who the frak you callin' a faulty drive??? I'll show you faulty!! Motherboard Fragger!!"
Sometimes the floppies and the heads themselves can become oxidized and repeated reads can “polish” things up and they will start working again.
Thanks for sharing your knowledge
I had something similar happen to me. I had an Apple II C I was cleaning up and the disk drive didn’t work. Tried everything to get it to work, cleaning, plugging connector in and out etc. Ended up swapping the internal drive with an external Apple disk drive and putting the internal one into the external case. Both drives ended up working just fine. I have no clue what fixed it but I wasn’t gonna complain.
Could it be that there is a capacitor reforming? Otherwise I like the stepper motor explanation...
since the voltages and signals on those read heads are so tiny, could it be that the wires being wrapped under that capacitor might have been picking up some noise? Caps that haven't been exercised in a while could maybe smooth out after repeated use? You know way better than me. Just thinking.
is it working because components are warming up ?
Sometimes drive need "rehersal" more to work in optimal condition.
I dont know if there is a amplifier IC on the PCB, maybe due long time no use the ic is going bad.
But after getting power again it cure`s its self.. Think the head should be fine.
Oh no, Step Motor, you're stuck!
I wouldn't be surprised if the disk running over the head several time is burnishing the drive head. Now that it's polished it's reading better.
I do wonder if the head-banging was actually doing something to knock itself back into life, like a bad connection that the vibration made reconnect, kind of like how you could tap on an incandescent bulb to revive it when it had a broken filament, or it shook some corrosion loose that was shorting the head's soldered connections, given how it improved after each head-reset, but that's pure guesswork on my part...
27:07 could be the read amp is “self repairing” when it warms up
That drive was not able to move the heads properly. The more it was tested the more the head was moving properly. I am sure that if you had pushed the heads to the last tracks and back to tracj zero, 5-6 time it would had worked faster.
I love the fact that did not trusted the copy of the game you had as a good write. Always check and recheck. Never fall to conclusion too fast. Been doing lots of debugging of complex device since my teen. now close to 60 I am still doing it and I really like how to debug. Must be a Montreal thing.. ( I am from Montreal.. ;-)
I am thinking that stepper motor for the head was gummy , and needed a work out , if you watch it in the beginning , its step are not the same ( look at 18:25 ish) then after its been working it steps are more clear and are the same (27:07)