MFM drives are really unreliable
Вставка
- Опубліковано 5 лют 2025
- #retro #mfm #seagate
Last week, we took a look at five MFM hard drives and controllers used to hook them up to a computer. In this week's video, let's try to get those drives working.
Part 1: • Not forgotten: MFM har...
Part 2: This video
--- Info
Drives in this video:
Seagate ST-225
Micropolis 1325
Computer Memories Inc. CM 3426
PTI - Peripheral Technology Inc.
Seagate ST-4051
--- Video Links
Seagate ST-225 Service Manual:
www.minuszerod...
MFM hard drive emulator:
• This thing can backup ...
IBM PC AT 5170 Series (talking all about BIOS patching):
Part 1: • IBM PC/AT Model 5170: ...
Part 2: • IBM PC/AT Model 5170: ...
Part 3: • Fixing and improving t...
IBM PC AT 5170 Patched BIOS for faster hard drives:
archive.org/do...
SpeedStor:
www.minuszerod...
Spinrite II
winworldpc.com...
Mac 512k with a MFM drive inside:
• Macintosh 512K with an...
Adrian's Digital Basement Merch store:
my-store-c82bd...
Adrian's Digital Basement ][ (Second Channel)
/ @adriansdigitalbasement2
Support the channel on Patreon:
/ adriansdigitalbasement
-- 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
--- Links
My GitHub repository:
github.com/mis...
Commodore Computer Club / Vancouver, WA - Portland, OR - PDX Commodore Users Group
www.commodorec...
--- Instructional videos
My video on damage-free chip removal:
• How to remove chips wi...
--- Music
Intro music and other tracks by:
Nathan Divino
@itsnathandivino
Thanks to the resourcefulness of my viewers, I've been pointed to the service manual for the ST-225! I love this community. www.minuszerodegrees.net/manuals/Seagate/Seagate%20ST225%20-%20OEM%20Manual%20-%20Oct85.pdf
Hey Adrian, thanks for the video; I was suspecting my hard drive to have scratches as well, so I followed your instruction to take the disks out in order to polish them - they do look fine now, after the polishing - but I forgot the right order to put them back in... does it even matter in which order I put them back in if I low format it?
...I'm obviously just kidding... 😂 great video mate! 😁👍
A lot of those drives even from that era may need to be actually 'PARKED' which could be a large cause of the not spinning up properly, or the heads not reading properly.
I have a few SCA SCSI drives and they make that sound quiet!
@@llwellyncuhfwarthen this also explains the high number of broken drives from the era... people turned off their computers without prior parking the drive properly.
Oh, I just facebooked you the same thing, and a slightly different one too. I hate trying to put web links onto youtube comments, as they tend to get flagged as spam and then removed. Anyway, nice video Ade!!
Many years ago I was given an ST225 with bad sectors in the boot tracks. With nothing to loose, I opened the drive and moved the track 0 sensor inwards a little bit. Closed the drive, formatted it OK and used it for years!
I had a 40 mb ide with a stuck stepper motor. I took the top off, gave the heads a nudge and put back together. Never had a problem after 😳.
My hvac class had a ps/2 that had bad sectors in the boot track as well. Just made a floppy boot disk, and all data and programs were accessible (including the games) 🤣
@@kd7cwg OMG yes I also had an old Conner drive that was pretty big for its time that I got at a computer market / hall with a bad sector 0. Booting from floppy then accessing it worked fine indeed! Man thinking about this I had so much old broken stuff I just bought from computer market junk piles for next to nothing that I got working through creative means. None of which I'd really rely on very hard, but eh it was good enough for hobby stuff. Also getting super cheap old SGI workstation monitors with weird connectors like 13W3 and getting them to work in linux by manually crafting X modelines until it gave a stable picture..20" CRT's that would heat your house for the price of a beer or two lol.
ST-225s are immortal.
The main trick besides not breathing over or talking near the drive while the cover is off is to *always* put the cover on with the drive spinning.
@@kd7cwg The system disk of my first digital recording studio's Atari refused to start one day, so I opened it up, gave a little spin to the platters and it restarted like nothing had happened. For the following few years the morning routine was open the drive, turn everything on, help the drive spin, put the cover back and just pretend it was normal :D
Hard drive: It has been so long since I have been given a voice... allow me to sing you the song of my people.
You should save the logic boards from the dead drives, so that if you see the same models in the future, you can swap them to see if that fixes any problem. At the very least there might some valuable parts you can reuse.
I remember when I was a child in the late 90's we were given a free Philips 8088 PC that had a rather peculiar start up procedure. You'd have to heat up the hard drive with a hair dryer or it would not spin up at all, but once you got it hot enough it would run perfectly fine. We used to play a ton of old games on it for a year or two and it was still working just the same when I gave it to a friend of mine some time in mid to late 2000's
It was nice to hear those old Seagate drive sounds again. Growing up, our computers were made almost entirely of broken parts. The only reason we had hard drives at all was because they had died for someone else (mostly at my dad's work). For the most part, the old Seagates could be brought back with a low level format. Stiction was a thing, and we got in the habit of rotating the entire PC case while flipping the switch. They would occasionally get flakey again, but we kept backups. Later, Spinrite became a thing and it brought back those drives in-place, data and all. That tool was a godsend. We also had a 600MB Micropolis SCSI drive at one point that took so long to spin up that the controller would timeout. It was very loud, but was very reliable. Ah memories...
Upvote for the spin rite reference. That was an awesome tool
i saved many 'dead' drives using spinrite. stiction was definitely a thing with seagate st-225's made during a 4-6 month period, i have 2 st-225's in my 10mhz xt clone. the boot drive always starts up, the second one always sticks if it sits for more than a week. i was told that the stiction problem was due to the heads being too-well lapped and smooth, which causes them to stick to the platters. if the drive doesn't stick, it's because the heads are rough. go figure...
Seagate and “most likely to work” two things I never thought I’d hear together.
At a place i used to work, we had some old S-100 buss computers that had hard drives (yes i know, it's weird). And these hard drives would get to where they would not spin up. We had to remove the drive, remove the cover over the platters, spin them by hand, then re-assemble. These drives lived through that so many times it isn't funny. For at least 6 years we kept having to manually spin those drives, and they never died or lost data. Which is just amazing, considering the head clearance on a HDD.
The useless old stuff in my mind.. debug g=c800:5.. The good old days! Thanks for the great content Adrian. Keep it up!!
I came to make the same comment. LOL
Now that triggers a memory in me.
DTC 5150
Ah yes, to trigger the hard drive controller's ROM utilities...
I was born in 1983, which makes me effectively a "90s kid"... my first computer (386SX 25MHz) was already using IDE for the hard drive and at least the beginnings of BIOSes as we understand them today. I missed out on the MFM era of storage, although thrift stores started dumping 5150s / XTs / 286s / etc. en masse somewhere in the later 90s / early 00s. (For cheap! Imagine getting a fully intact IBM XT off the shelf for $5!) Although I bought a lot of then-"old" hard drive controllers and MFM hard drives, I assumed they were all "bad" because they just reported error codes. I wish back then I had known about triggering hard drive controllers internal low-level formatting utilities as such (and how critical that was!), but I wouldn't figure that out until somewhere in my late 20s when I started messing with this stuff again and of course the Internet was more prevalent with such information. I've made a couple ST-225s work since then in proper 8088 based PCs.
2:15 has to be my favorite of all horrible, suffering, aged-out hard drives. It's so sad! (And noisy.)
Adrian's reactions are priceless, least of all realizing he once worked repairing (or "salvaging", as may be more accurate) people's hard drives at some point.
I used to rebuild those ST225 drives. They have a track 0 - 1 track and I think another track (or cylinder if you like) beyond the last cylinder 612 with a data pattern on it. After replacing the platter and cleaning/inspecting the heads we would then write the track -1 pattern with a very home-made box connected to the ST506 connectors it would step the head stepper motor back (on which we drew a dot on the spindle) each time you pressed a button then you pressed the write button I think there was another button to get to the other track or there was some combination of presses. The drive would then come ready. We then did a surface analysis with a dedicated Winchester testing and diagnosing machine which would mark any bad sectors. Then we would attach it to a PC with a controller and do the debug g=c800:5 to low level format the drive and then we would partition and do a high level format. When rebuilding the drives we used little plastic spacers to separate the heads for removing from the platters, we inserted the plastic spacers with tweezers, really fiddly. All this while wearing a paper suite, gloves and a mask while a blower in the clean room tried to blow everything away! I bet all that kit went to the scrap yard years ago. For stiction we used to gently flick the heads (off the media) with the plastic spacers, which would demagnetize them and usually this would get rid of the stiction.
Hi, I have an ST225 which is only capable of showing its catalogue, when I access it through a floppy boot disc... ever had this symptom? What would be your recommended approach? Cheers!
@@JulienMR HI. Are you interested in recovering the data or do you just want to get it going? I would ensure you have the correct cyls/heads set up and also that you can read other drives with the method you are using first before attempting anything else. My guess is you're able to read the contents tables and the data isn't visible because the heads/cyls are set wrong. Should be 615 cyls and 4 heads which I think is the old IBM type 2 size 20MB (21MB formatted 26MB unformatted) If my memory serves me right. If you have no luck with that try formatting with DOS. If that doesn;t work then a low level format using debug could be needed prior to the DOS format. Run debug from DOS using the command line: debug g=c800:5 from the HDD ST506 based controller card ROM if that works when complete run the DOS format again. Obvs this will destroy any data on the drive but will make it usable in DOS again if the board, heads and platter are OK. As it seems to be coming ready and you can read the directory I assume at least the hardware is working. Good luck. BTW this is only my opinion I haven't seen an ST225 for over 40 years so I may have forgotten a few things. So do this at your own risk.🙂
Oh - if you low level format you will need to create a partition before the DOS format using FDISK I think.
@@radio-ged4626 Thanks ! I'm pretty sure than the HDD Type is good, as I checked the switches on the controler board : Off/On/Off/Off, and the first two being Off/On means Type 2 for Drive 0 (the next 2 values are for Drive 1, and I don't have any). I didn't try any Format command yet as I hope to save the drive content, but I'm thinking more and more about it !
@@JulienMR Yes, it could be the drive was set differently on a previous controller so that's why you can't read the content.
this certainly was a blast from the past. I recall servicing drives suffering from "sticktion" or however you'd spell it. we called it that as well and the root cause I was given was that the spindle lubricant would cool and harden beyond the initial torque point of the spindle motor. as crazy as it sounds, the fix that worked most consistently for us was to actually drop the drive flat onto a table from a few inches up. well within the engineering g-force tolerances but enough of a bump to knock the spindle loose and boot the system for imaging.
Years ago, I had a maxtor IDE hard drive with bad bearings, and oh my god that thing was loud. I'd have it running in a machine in the basement, somehow it ran windows 98 ok, and you could hear it clear from upstairs with the basement door closed. I also have an ST-225 MFM hard drive I'm trying to get working again, didn't think to use speedstor until watching this video. As always, thanks for the quality content!
@Adrian 17:20 already 2 or 3 mins i hear a schuss parasitic noise in my bluetooth speaker (is off grid rightnow) and some in synch with visuals on your debug screen. Your install has some electronic noise immunity issues, i believe 😅 Nothing a few X and Y filters can correct, for sure ! The second you stop it, like 17:25, noise disappears...
Great channel, thx from a 51yo young man happy he was there in the 80's 👍 very refreshing, lol 😅🖖
Take the platters, drill a small hole near the edge. Take one of them and drill evenly spaced holes around it's edge. Suspend the single hole platters from the multi-hole one with light fishing line. suspend the whole contraption by fishing line strands attached to the multi-hole disk, and tied together. This gives you a great sounding wind chime. You can even drop one through the center with some salvaged polyamid cable to catch the wind & act as a clapper.
With the ST225, try pulling the PCBA off it and cleaning the contacts that connect the HDA to the PCBA. If I recall correctly there will be spring connectors for the hub motor and an elastomeric connector for the heads. Worth pulling the elastomer out, cleaning with IPA, and flipping it so the "squished for 30 years" parts are no longer aligned and it makes slightly tighter contact.
Thanks for this video! I remember the sound of the old ST-225 and similar drives. The good old days! You put forth a great effort on these. Much appreciated.
This reminds me of the efforts we make to un-stick antique refrigerant compressors which have become stuck from dried up oil or other problems. The only difference is that we use 240 or even 480 volts of raw power to get them broken free. Thankfully, opening them up doesn't equal death with these devices!
Yes, unless you burn out the windings, an old compressor is probably a lot hardier than a delicate hard drive of this vintage
The Commodore A590 was the first 20 MB hard drive in 1989 for the Amiga 500 (Autoboot did only work if the Amiga used Kickstart v1.3) and i believe it used one of these MFM drives which was a Western Digital XT model (you will find pictures at Google because those XT drives are fairly big). The A590 was having a XT and SCSI interface. The read and write noises are funny and very mechanical. If a floopy disk was in the disk drive the Amiga did autoboot from the disk drive otherwise the hard drive took over. Amiga was having a very clever auto config system just like modern PCI from the PC.
It was great hearing the sounds of those old drives, brings back memories. I still have my original SCSI version of the ST-225 (the ST-225N), would be great to spin it up and dig through all the old '80s software and long forgotten programming efforts.
Adrian sets out to make a video showing how unreliable MFM era drives are.
Every drive doesn’t work.
Message received Adrian 😁👍🏼👍🏼
Your video brings back memories of the good(???) old days building/fixing/optimizing PC’s. Have forgotten how much an issue those old MFM/RLL disk drives had, esp. when larger drives came out that were not directly supported by the BIOS. Good walk down memory lanes. Thanks
Sad to see none of those drives worked. Was fun to see the insides of them though! I’ve actually had incredibly good luck with MFM drives, and most of them I’ve gotten have worked fine with little effort. I actually just tested all of my spares recently and they all still work fine. Some of them I’ve had for years too.
I will go to great lengths to repair them sometimes as well. For example, I did a spindle motor replacement on a Miniscribe recently, which was 100% successful!
Obviously, I’ve definitely gotten some bad ones over the years though, but I’d say probably 80 percent or more of them I’ve bought worked fine in the end!
I'm glad you made a video showing how often they just can't be fixed. I've got one that doesn't work and I was trying to work out why I couldn't fix it as easy as some of the other videos that i had seen.
Brings back memories of what a PIA setting up a hard drive could be!
Indeed, heads, cylinders and sectors was a PITA. Varied from BIOS to BIOS
ruined the data on 0 track on someone one time thinking the settings on the PC was wrong and wasn't utilizing the whole drive.
Turned out back then the CHS on the BIOS was skewed to what the HD/controller expected.
Believe it or not, I was able to recover almost all of what was on the drive.
How you ask?
Folders are actually files, almost all subfolders and files were in folder, and almost everything on the drive was one cluster or contiguous clusters.
Using tools that could read the drive raw, searched for sectors containing '.' and '..'. Those would be folder entries.
Get the size of the folder, calculate the # of clusters, and slowly made a map of what the cluster table/File Allocation Table (FAT) should be. Get the file entries in the folder entries and repeat.
I only had a handful of situations where if I assumed contiguous clusters, I ended up with file chains that would appear to have cluster collisions.
Analyzed the data in the sectors assumed to belong to the files, and I could figure out where the next cluster actually was an for which file.
But when you only have a few 10's of MB, you could do that. Forget trying that with GB or TB!
Thanks for the fun MFM video! - Also, I love the half-smirk you got right while holding the screwdriver menacingly saying 'percussive maintenance!'
When I worked at a computer store in the late 90's we would put the drives in a fridge when they had sticksion overnight and if they would spin the next day we got the data off of them with Laplink and a Laplink cable. Good times to be sure! We called it "Shake and Role", 1st you shake the drive side to side, 2nd you would shake and role it in a circle that the same time and if that did now work them it was Thump time with a screwdriver... Lol
We used HardDisk TecSpect v6 to get 99% of the drives specs we worked on back then. I believe it cost $100.00 US for a 2 licenses copy back then.
Great video Adrian!
LLAP 🖖
I love your channel, Adrian, thanks so much for your hard work. I've opened up quite a few non-working old drives where stiction had been 'fixed', by force or by itself, to find the heads still stuck to the platters, ripped clean off their mounts. The noise it made when the drive was running was unmistakable. Happy days! 😂
Haha yeah that's happened to me too. Slamming into the now naked arms as the drive spins up. Bad times indeed!
Whether the hardware works or not it is never a failure when you learn something useful, I call this a success with a twist.
Back in the day, Micropolis Hard Drives were mostly large, high volume disk drives. They were known for being noisy, but were usually larger (storage space wise as well as physically) than other drives. Seagate MFM drives were some of the most reliable back in the 80's. They were also the largest producer of MFM drives at the time. Miniscribe were primarily hard drives used in business. Western Digital had a few, but no one could compete with Seagate. Things changed when IDE took over. Western Digital and Maxtor took over the top 2 spots.
You ever hear the miniscribe brick story?
@@MrZorbatron are you referring to the fraud stuff they were involved in, to attempt to keep stock price from falling?
@@rdh2059 Exactly. If I remember correctly, they shipped cartons with pieces of masonry that approximated the weight of the drives that should have been in them, ostensibly with the intention of recalling the items in short order, before they would have made it into the distribution channel. Unfortunately, things didn't go this way and it was caught. I might (probably) be missing something because this is something I learned about probably 15 years ago. I understanding was that the books were cooked well worse than was initially apparent.
I have never seen an MFM drive before. Never used them. My first PC was from a company I was working. I did have a a hard drive in my computer, but not sure which one. Man I wish I had not sold mine. I had to build it myself. Our tech department were building PC's and putting them out, and they had a large amount to build for the companies around ours. So, I financed mine and the only way to get mine that Friday was for me to build it myself. And that is where I began learning and built lots of PC;s in my life. I have taken many bad hard drives and keep the very strong magnets!
I don't know how to thank you for showing the jumpers on that ST-225.
I've had a 225 laying around thought to be dead for years now but it was just set to the wrong drive select, the thing actually works, I am ecstatic right now. Thank you so much for your video.
You know it's gonna be good when Adrian has to drop regular noise warnings in!
Enjoyed seeing inside the drive while it was spinning and moving the head. Fun stuff!
I enjoyed your video up to the last minute. I remember other HDU technology, called RLL or Run Lenght Limited, even more unreliably than MFM. Great video.
I had a similar model Micropolis HDD to that one and it was quite loud too. Seeing the motor PCB bought back memories. Compared to the original ball bearing based Seagate 3.5" 7200 RPM Barracuda hard disks the Micropolis was quiet. Those things were really loud. They were so loud that it was uncomfortable to be nearby. They ran roastingly hot too. It really makes you appreciate the later FBD designs.
Thank you for the trip down memory lane! (there's a pun in there somewhere). I still have a couple of these drives but no mfm controller. I still have my very first hard drive I used on my ColorComputer. It's a 5 meg - yes 5 meg shugart full height drive. It still spins up and self tests.. Now you make me want to find a controller!
Thanks!
Stiction was also an issue in large CDC sealed drives. We discovered that if the heads didn't move, the Teflon surface of the platter would get hot and effectively become an adhesive so when the heads landed, it would effectively "glue" the heads to the disk surface. What we finally ended up doing was modify the HD driver to rapidly move the heads around any time we went some number of minutes (I forget the actual time) without being accessed. We would that move the heads to a random sector.
I had a ST-225 WAY back in the day. I'll note that those often get SUPER hot, which is part of what causes "sticktion". Even the best of them got hot enough to effectively melt and then boil away any lubricant used. I had one that after running for a day or two solid got hot enough to literally to unsolder it's own components. Seagate's answer to this was to always mount it with the metal body pressed against the cage to act as a massive heatsink. Always good to see older technology of yester-year.
Also, most of the ST-225 were from the age where you could use MFM or RLL. Given then RLL would make the same drive about 38M (vs 23M), I'd bet the one you had was formatted as RLL.
Not a bust - second half interesting and informative for those who haven't seen the inside before.
Brings back memories from 20-30 years ago.
Micropolis drives were notorious for running extremely hot and failing fairly quickly. I remember back in the day we were backing up one of our server's Micropolis drive RAID arrays to tape so that we could restore the contents to a new WD RAID array, and it took forever. The drives were getting extremely hot, even with a fan in the RAID array case. So we took the cover off the RAID array enclosure and stuck a box fan in front of it to keep it cool. Once we had the data restored to the new RAID array, and making sure everything was working properly, we sledgehammered the Micropolis drives and sent them and the old RAID array case to recycling.
Tbh the best part was seeing you have fun taking those drives apart haha!
Thanks Adrian. I spent the last 30 years working at an industrial site that used the micropolis RD53/54 hard drives with microvax 2 mini computers running a propriety software. They were all MFM drives and used Dec M7555 controllers. We had up to 13 running in the field and at least 3 backups running in our maintenance shop. As well we would sends them out for repair at W.M. Farris in California. There were a number that would scream during startup and then get quieter after running for a few hours. Just before I retired I replaced them with solid state MFM drives made by DREM. The DREM drives worked great. Thanks for the memories...
This video was not a bust-it was hilarious. Thanks for sacrificing all those cilia for us!
I always kept a computer around back in the day with AMI BIOS and ran the AMI Diag utilities on the hard drives. It was an excellent way to get drives going and even I think in some versions did speed tests to determine your ideal interleave number.
I live in maryland and there was a rock station out of DC and up to maybe 10 years ago they had some old computer running in the studio using a Seagate st-225 maybe ... just funny driving to work and I could hear that drive running in 20teens
Thanks for the video!
I remember a TSB about the stiction issue back in my CompuAdd days. The paper described the cause of stiction to be from the epoxy that affixed the head to its associated arm. When heated up, the epoxy would outgas and form a layer on all the surfaces inside the drive. Over time that build up would become dense enough to temporarily glue a head, or heads, to the platter(s). For drives without the automatic head parking feature, running a Park program before shutting down the computer would move the head assembly to the innermost track where the motor had more of a torque advantage over the stiction. Unfortunately the build-up of residue continued until even this wouldn't work. Snapping the drive housing to force it to break loose was much easier with the 3.5" drives but required a bit more vigorous action due to the lower mass of their larger ancestors. We were told to replace all drives by Tandon from a certain manufacturing date upon the first customer complaint. The drives would be replaced with either Miniscribe, Seagate, Fujitsu, or Western Digital as these had either gone through this phase of failure already or, had so few related issues, weren't considered to be a future issue.
Uh, that brings back painful memories. The frustration, incredulity, and sadness. As a poor student, affording a drive was bad enough, forget about backups. When these drives died, they did without warning, and just like here, with teaser sounds that makes you think they might be a little alive. I do not pine for those days, hardware is so much more reliable today and spinning rust so cheap can you can raid it and have backups.
Back in the 80's to 90's I frequented a local repair shop to buy parts...I bought a bushel basket of MFM drive pulls for 10 bucks. There were 10 or 12 CMI drives and 6 0r 7 Seagates in there. I found out later IBM had to recall all the AT computers with the CMI drives - I took the tops off and the rust was slung off the platters on the inside periphery of the case..but I got 5 of the Seagates working.
Early PC hard disk drives used MFM (Modified Frequency Modulation) encoding to store data on the disk surface, the same as PC floppy drives did, and always have. Shortly after hard disks became more common, a similar encoding scheme came around, called RLL (Run Length Limited).
You need a burst writer: Kinda like a servo writer, but those just needed the burst to be written. After it starts the YOUR heads are reset to track zero and knock against the backstop or inner stop, the burst is meant to stop this. I can't remember the actual name of the device off hand, but the inner tracks are written with a servo burst, the problem with these drives is that the head park area is over the burst area and so you tend to get head slap which is transit or shock damage caused by the heads leaving a group of four indentations. Without the burst data the head location when reset may end up several tracks in and not over the the outer track zero, if that occurs the drive will refuse to format. You can maybe adjust the head end-to-end backstops but this means opening the drive and definitely means rewriting the burst.
I remember the days of MFM hard drives. Started off with a trusty 20MB! Then somewhere along the line I remember an 80MB drive where a pair of heads had failed so we simply configured it in the BIOS as having less heads and it worked like a treat! Still got 60MB out of it. In those days, that was a win.
Absolutely take apart bad hard drives! The parts are cool, the platters are decorative, and in more modern drives there are neodymium/"rare earth" magnets to recover. I've sometimes found the bad drives are cheaper to get than the magnets by themselves. 😅
Your videos are never a bust! Learn so much. Thank you again.
Back in the 1980s my school had a network of BBC Micro computers (6502-based) connected to a file server in a cupboard. This was a 20MB drive, probably one of these, connected via SCSI to a twin-processor BBC Master (also 6502-based). It worked fantastically well... except the drive had a squeaky bearing and howled like a damned soul ALL THE TIME. Maybe it was; later it got struck by lightning and stopped working...
I've heard of an old hard drive cramming the head through its side, so I would not count on the electronics not allowing to move the head too far. We had a Seagate ST-157A at home (3,5" IDE). Long story short, due to some percussive maintenance, it got a bit of the case shaved into chips due to the platters hitting it. My dad opened the drive case, blew the shavings off, closed the case and it still worked flawlessly, not a single bad sector. Those things were rock solid, too bad Seagate hasn't seen that kind of build quality in pretty much decades...
Usually the failure in such old drives is mechanical. I have repaired a few at some point. You have to remove the cover, check, lubricate where necessary... in one in particular I actually removed and physically cleaned the discs. And it was working again. To remove the noise, change the ball bearings.
Ah, the ST 251 was the drive I remember having.... 40MB and I remember the sound of it self parking when turning off the computer.
38:00 Oh, contrare, Dear Adrian, this is a MOST Excellent instructional video!
I'm surprised this isn't a "II" video, but I'M HERE FOR IT. This episode borders on ASMR. Perhaps clip some out for shorts! LOL @ the QC Pass sticker.
Enjoyed your view and expertise on this old equipment. I sold my IBM XT years ago when moving. I've regretted it. I loved the sound of the old full height hard drive. When yours go bad and you recycle them. Do you ever sell the broken ones that still spin up?? I just love the startup noise and spin sounds on vintage tech. Not to be used for data storage.
Something I always used to do when testing hard drives, is I would always write the date when I wrote notes on the drive. Makes it easier on you in the future! Let's not forget, when you open these babies up, be sure to grab the neodymiums... Free magnets are always great. Fun stuff...
those don't use voice coils, watch the previous video
Those platters are made of the purest aluminum of any mass produced parts ever made. The oxide coating is so thin as to be completely invisible. A pigment is added only so that in manufacturing they could tell the coated platters from the raw ones. In later years there was no need to add the pigment, so the platters appear to be bare metal. The surfaces are also extremely flat, usually within 25 microns across the entire surface.
A hard drive platter makes an extremely nice signalling mirror, for primitive hiking. The sun's reflection will be about the size of the platter over considerable distance and you can look through the center hole to aim the reflection.
I remember in the late 90s I got an old MFM similar to your type 2. It wouldn’t be detected and quit. I ended up opening it and using speaker coil wire to connect the head to the board: ie the flexible cable toward the head was broken. I closed it and it worked. Out of 20mb, I lost some 2mb.
One advantage in these old drives was that you could totally erase the platters, then low-level format them with speedstore and they would be back to normal operation.
Modern HDD's have servo tracks and other lovely things written on the platters in the factory, that once lost the drive will never work again. Some software says it's low-level formatting them, but in reality, the drive just lies, it can only be low-levelled in the factory on a special jig. The head actuators are voice coil based, with their only reference being those factory servo tracks. once lost the drive has no idea where the heads are.
One trick on modern drives is that outer tracks have more sectors than inner track, however the onboard controller translates all this into a nice linear pattern, so while the computer thinks the drive has so many tracks and heads etc, this has nothing to do with what is physically in the drive - in modern systems these days the whole concept of heads and cylinders etc has been abandoned, its all just a big long string of data blocks.
The word "interleave" sends a shiver down the spine of anyone who has ever tried to set it optimally.
I worked as a technician in the late '80s through mid-90s and we called it stiction too. I would lift the anti-static strap underneath and twist the spindle with a pair of needle nose pliers to get it started and then immediately back up the drive.
I had a Seagate ST225N SCSI hard disk connected to my Atari ST and it looked and sounded just like those MFM drives, probably the same mechanism with a different driver board. Never knew new hard disks had glass platters.
I have had a lot of those ST-225 and don't think I've ever seen one with the SEAGATE logo on the bezel! Neat.
As for the loud drives I've had luck just letting them spin for an hour to loosen up the old lubricants. Plus I just love listening to them 😄
If I remember correctly for an MFM hard drive to work the drive has to be formated with the specific MFM controler card. I was building computers back in the days when MFM and RLL drives were in use but I only sold new IDE drives. I had to recover a lot of data off MFM drives. Most of these were 10 to 20Mb and it could take several hours to pull the data off a drive.
Those old drives that had voice coils for the head movement have excellent magnets. I've got a couple of voice coil magnets from an old Micropolis 9GB SCSI drive and those magnets are super strong.
I had a seagate 20MB HD, probably the 225, that stopped working. So went out and bought another one. I think I paid $400 in '86 or so.
I too had the idea to swap the boards on the HD. BOTH WORKED! So then I had 40MB of storage. Whoo Hooo!
wow. this really took me back to my early tech days. the sounds, the slow spin up, parking the drives, termination jumpers, 'thumping' a stuck drive, the really slow error checking etc Reminds me of many things I don't miss about older PCs. I forget what model of HD I got it out of but I had a head arm (desk toy leftover part) that looked like you could drive a vehicle over it. The old drives had so much more metal in them.
Those irregular drive noises give me the slightest pang of anxiety, even knowing they're not here...
That first drive sounded like something out of a horror movie, which would be appropriate if it had important data and started failing.
I sincerely thank you for not talking and letting us enjoy the rare motor sounds!
Of all the MFM drive manufacturers at the time, there was one who was by far the definition of unreliable. Enter Kalok. Kalok was and is by far the definition of unreliable due to band slippage and high failure rates. Even after they left the industry in 1994, Kalok still holds that title as the most unreliable MFM hard drive and you'd be lucky if the thing ever worked at all. Let alone finding two that worked!
I remember all those drives. The st225's always had problems, but could always swap the boards to get working drives out of batches of drives I would pick up. Still remember the days of nothing but noisy drives. My ears are still bleeding to this day from using whatever came in the door to give more storage space. Gone are the noisy, but fun days.
That first drive, ans your face as it squealed.. lol. I was testing 16 IDE drives two weeks ago and I had one that sounded just like that. I shut it off.. thought it was going to blow up my bench.
You forget how noisy those old drives were. Even your home computer sounded more like a data centre.
MFM drives were always terrible, but cheap. Easily serviced too. Considering the tolerances were gigantic by today's standards. For mobile computing, we tied them to our bicycles and rode all over town. You could basically fill them with sand, throw them at the wall, and still get them to work again. It's nice to see all the old fixes again, and so many ideas we never thought of.
Back in the day, people used to take the old platters out, and use them as coasters..... Considering that these drives are over 35 years old, it's not surprising that they no longer work......
Any two surfaces in contact display two coefficients of friction at the interface: the static coefficient of friction (stiction) and the sliding coefficient of friction. In this case, stiction is the holding force that must be overcome before the axis can begin to move.
I am so happy that my MFM HDD is Working Well. My ST-225 Makes the Knocking noise just at the beginning, and then it runs well.
33:44 = I had a hard drive like this earlier salvaged from flea market. I dismantled it and took out the central motor and the stepper motor used to move head assembly. Also I kept the discs thinking, they are made of copper. Later I realized, they are made of aluminium and with a layer of oxide of copper.
Those disc plates from modern hdd's are quite useful as a mirror when you shawing head with straight razor.
We ALL called it stiction, it was very common with video tape recorders and rotating heads. The tape would stick to the head as it flew by, and make a very large mess inside the transport!
We used to get those stuck drives spinning again, by rotating them clockwise on the floor or a table and stopping the drive case from spinning very quickly with a phone book* to break loose the platter. *(those existed before the "internet" too... :) )
My father had one MFM drive that lasted 20 years of daily use in his "writing apparatus" Amstrad as it was coined in the taxes. I still have it somewhere and it contains his entire bibliography of thousands of articles and dozens of books.
Back that up ASAP if it even still works. As you see in this video these drives can be very temperamental…
@@dogecode386 I still have that drive somewhere though i have of course backed it up long ago. That writing apparatus of his lost its power supply in 2005-2006, but the disk survived. I don't think it's the first disk that machine had, but certainly the longest living. I think it was amstrad 1512, but not sure since it had AMD 8086, not intel and hercules graphics for the sharp fonts.
On the one hand, it was fascinating to see the insides of those drives, and what had gone wrong with them. On the the other, I couldn't help cringing and wincing at how you were touching the platters and heads with your bare hands. I know they were bad anyway, but I had an awful visceral reaction to that. 😬
I had a Micropolis 1.2 G on my Amiga back in the '90s. One day I came home from work and I could hear it squealing before I got to the front door. Fortunately I was able to get all the data off of it by wearing tight headphones.
Wait... if it was running before you got home, who was using it? And HOW???
I never turn off my computers.
The 1st hard drive I saw was a winchester made for an Apple 2 computer star network. It required one whole apple 2e 48KB computer to run the drive which was about a shoe box size and you could see the heads moving on the 6" platter.. it's capacity? 5 Mb!
"what am I doing watching someone mic up old hard drives" continues to watch.
I owned a stepper-band driven SCSI hard drive for my Amiga 500. Yes they WERE that loud. I could hear it 2 rooms away, both the roaring of the fan/spindle and the chittering like a typewriter of the heads over the platters. They were bloody LOUD, Adrian. 🤢
When I learned how to use MFM disks, Spinrite was the tool that they always told us to use to determine the interleave.
In 1994 I put a ST-251 40 meg with Doublespace compression in a Pentium 100.
It worked perfectly, no errors never lost data. Running Windows 3.1 and Autocad 13.
My dad has stories about the stiction - Just drop the drive like an inch on the desk and boom its free. Also I had a SCSI Full 5.25" drive that was so loud you could hear it outside the building.
The drive platters make a pretty decorative mobile, I have one hanging in my window with a sign that says "Got Tape?"
Hello Ardrian . Fro ST-225 - Faulty track 0 signal ? all the time trying to calibrate , try to make that trick :
-normaly connect the disk , don't setup disk in the bios . Run MS-DOS after that run program SGATFMT4
from the list chose geometry (CUSTOM = 615 ,4, 17 or "ST-225") , after that Option 1 Low level (blind formatt without verify).Turn off the pc and turn on setup in the bios reset if pass calibration it's ok if not change the PCB and do the same .Common problems :
-Head crash ,faulty IC , PCB read/write error because of age .
This brings back so many memories! A fun watch. Thanks!
Hello Adrian
I enjoyed this video very much. 😊
My first pc at age 16 had a ST-225 and i also had that PTI Drive in my Fingers. :)
Moving inwards track 0 and sticktion… memories :)
Thank you!
I like your entertaining comments about "unhappy harddrives" and " belly up" (I did not know this phrase yet) companies, it makes me grin 😄 The sound reminded me of my first 20MB Kalok harddrive.😂 I have found an interesting "manual", it is called "CSC Hard Drive Bible 7th Edition 1994".
32:50 Oh come on, it just needs a little love and some oil, do not kill it. 🤣