Killed by magnet | HDD examination
HTML-код
- Опубликовано: 7 дек 2015
- This HDD was destroyed in my previous video. In this video I take it apart to investigate if I can tell exactly what went wrong. And then I just fool around with it...
Full music credits:
'EDM Detection Mode' by Kevin MacLeod is licensed under CC Attribution 3.0.
Direct Link: incompetech.com/music/royalty-....
ISRC: USUAN1500026
© 2015 Kevin MacLeod
'Carefree' by Kevin MacLeod is licensed under CC Attribution 3.0.
Direct Link: incompetech.com/music/royalty-....
ISRC: USUAN1400037
© 2014 Kevin MacLeod 
Наука
The magnet removed the low-level formatting on the platter, that the drive's hardware needs to determine the characteristics of the drive, such as its actual RPM, velocity of the platter at the heads, and the spacing and location of the tracks. With those marks gone or damaged, the hard drive circuitry can't tell for sure if the platter is even turning, or how far the heads are from the platter's hub.
What might be interesting is determining if you've erased _all_ of the factory formatting: The heads appear to linger at certain locations before sweeping the disk's surface: is that because there's still some data there for the heads to lock onto, or is it just a physical coincidence or hard-wired spot? Does applying more magnetic disruption change the heads' behaviour?
+Nevin Williams
*That.*
Only other thing I can think of, but would say it's already improbable: the strong magnetic field permanently magnetized something in the heads.
superdau That'd also be interesting to see: what all parts of a hard-drive are susceptible (to a significant degree) to magnetic fields.
I know some parts are shielding, such as the mounts for the neodymium magnets that swing the heads: They're still attracted to a magnet, but they constrain the flux away from the platters. It might be interesting to see this property explored too.
+Nevin Williams It could also be that the head did crash into the platter and that destroyed the head. It could be a very slow crash (for example near the middle of the drive) or a very swift touch but it could be enough to destroy it, I think.
+Nevin Williams well he could take an hdd, open it and then hold a much smaller magnet close to it whilst spinnig, thus erasing only that particular area
+GiantPen23 +Nevin Williams or even carefully take one apart, then re assemble (to make sure you can) test, then if it is ok, disassemble again and only test one platter, use paper to protect it, and make sure it cant hit the magnet, after each test try the next plater, once all are tested you have a definite answer, i thought that the head position was determined by the force applied
The magnet erased the "SA" or "Service area". The SA is a track before the MBR that you cannot access without special expensive hardware or reverse engineering.
The SA contain usually the drive geometry, a bad sector list, smart data, firmware extension, LBA translator (which include bad sector reallocation address)... Without the SA, the hard drive can't boot. The SA is usually at the very beginning or end of the disk, that's why the head in your dead drive stay most of his time in the centre.
Also, like Nevin Williams mentionned, you probably have erased the servo (data at the beginning of each sectors) that help the drive know his LBA position and rotating speed.
So a super strong magnet is not necessarily a sure way to perma-wipe a HDD?
@@TaiganTundra If you want to also damage the drive, than go for it. But keep in mind that the corruption caused to the SA / Servo will prevent you to run the drive after, to confirm you properly actually wiped everything.
If you want to perma wipe the drive and render it unusable, a other good solution you can do and actually visually inspect, would be to use that strong magnet, but then, you get flat screw driver, spin the disk then remove the oxide layer with the screw driver. You do all the surface until all the brown stuff is gone. Then you turn platter over and do it on the other side.
It hurts to see you touching the platters
ikr?
+Psycedelic Dakke dak How?
+the mre nerd oh lots of ways
+the mre nerd It's like touching a snowflake. You're gonna destroy it. Now go watch videos of people scrapping HDD's and laugh sadistically, knowing it causes great pain to your fellow nerds.
+Psycedelic Dakke dak They are broken anyways so who gives a shit.
4:10 so what's interesting about that parking mechanism is you'll find it on MANY older hard drives. Back in that era, hard drives were all standardized to have a region of tracks designated as "the landing zone", located around the center of the platters closest to the spindle. This is because for a very short period of time, the heads themselves are actually making contact with the platters. They float above the surface of the platters when they spin fast enough to generate enough lift, but until they spin up quickly enough the heads will be dragging along the surface. There are spacers on the heads to keep actual head-to-platter contact to levels of non-existent, but the spacers will still make contact with the platter surface.
This "landing zone" is more or less a region on the platter that contains no data, and is the designated region where the heads will park and "take off" from when spinning up.
On more modern and recent drives, they simply park in the parking lock away from the platters and only cross the platters when they are rotating sufficiently fast enough to generate enough lift for the heads.
its likely erased the servo tracking information
the drive could easily b erepaired if you have the million dollar equitment lol
That or something similar to that was my theory as well.
It's corrupted so it needed to be formatted then it woud work
@Creeper1105 no
HDD platters make nice mirrors!
+SteelSkin667 I know right, I think the first time ive ever truely seen myself was in a corrupted notebook hard drive platter... Sat there for like 25 minutes...
spelling bees must be tough where you're from lol.
isaac10231 Polish is even worse lol
What about hungarian?
What about filipino?
What about finnish?
B's
Thanks for all the kind comments and suggestions on what might have gone wrong with the drive!
In my video where I opened another magnetically killed HDD, I concluded that the data on the platters were simply wiped away by the magnet.
I still have a feeling that is the case - but I can't conclude it with certainty.
Especially after reading this from the pro's: www.supermagnete.de/eng/faq/What-is-the-safe-distance-that-I-need-to-keep-to-my-devices#hard-drive
But the 6x2" magnet is ridicously strong - even at some distance. Much more than my second-largest neo-magnet and the sizes that Supermagnete sells. So I wouldn't be surprised if the 6x2" is able to wipe data from the platters at some distance (~10 cm from the laptop in my case). It's just hard to prove :) I may need to revisit this magnet vs. HDD in the future...
the magnet messed up the low level format so the reader/head don't even know where it is
+brainiac75 Please do, this awesome!
did you check if it showed up in disk or device manager.
+brainiac75 I think you bend the head's arms, that's why you can hear a tiny weird sound when the head is moving. Old HDD contains a lot of magnetic elements, even on the disk itself, nowadays engineer make them less sensitive to magnetic fields, but with a your powerfull magnet, I think you can still kill one.
You clearly wipe everything on it (every single bit is a polarisation state so...) but this is not the major failure.
I'm not a pro, so I don't tell you "THIS IS WHAT HAPPENS !!" and I'm French so forgive me my english skills aren't very good.
Bonne continuation :)
+brainiac75 I took one of my old harddrives apart to find the read/write arm was bent on the top platter which was why it was making noises and likely why it was causing windows to crash. It wasn't even my boot drive and it was causing massive system instability. Your magnet may have not only wiped the data but also bent the read/write head even just slightly.
Probably the reader heads are damaged. I once took apart a hard drive (mainly to see how a hard drive worked). I pretty much killed it from my stupidity. the drive was then used to create an xbox prank video. after that, I smashed the deceased drive with a hammer.
Your videos are well made, informative and clearly you are passionate for the subject matter. Thank you for sharing this
+YOA Ownage HD
Thanks for watching and commenting. Nice to know I'm on the kind side of RUclips :)
+Brainiac75 magnet vs hard drive with the top off to see what it would really do
George Murray That's actually a good idea. A working HDD should be able to run without problems for a short time with the lid off. Maybe with some acrylic glass over to keep the worst dust particles out. Maybe it could visualize what effect the magnet has on the HDD.
do it plz
hmm youtube comments are broken again i cant see all replys
Tracking data on the platters is most likely damaged.
Love your voice , it makes it so easy to listen to and take in what you're saying.
With all due respect to the magnet, Seagate drives really don't need much encouragement to get to the "click of death." There's a reason data centre admins steer well clear of them.
This is really well researched and filmed, good job dude!
the sound of the head scraping against the platters made me cringe
It is a horrible sound... I turned the volume way down in editing to avoid maddening people. In real life the sound was unbearable.
I love your videos! It's really interesting seeing magnetism demonstrated on camera with so much stuff, from HDDs to chemical elements. Cheers from Mexico!
+luneki
Thank you very much! Greetings from Denmark.
You messed up the data on the hard drive, the heads don't know where to go.
Did you try looking into disk manager? Windows only shows actual usable partitions in My Computer. But now that you've opened it it's destroyed, forever.
I'm pretty sure the drive was damaged beyond repair. The drive would just constantly click and spin the disk up and down - even before I opened it :)
These tests recorded via video are very well done and interesting to say the least. Thanks and continue to share similar experiments.
+Joseph Sunday
Thank you! More to come so stay tuned.
Funny!! I learned Danish today and I learned something about HDDs! Subscribed for the long run, mate!
+Zurhairi Zohreen
Awesome! Glad to hear.
+brainiac75 say "fnugfri klud" 3 times in your next video :-P.
Brian, your videos are very instructive about neo-mag's and with PC Hard Drives! I indeed like your videos and I wish you very good luck with it!
+[OFFICIAL] xXxFTM2XxX (FranciscoTheMan2)
Thank you very much :)
I think that the magnet somehow managed to misalign the reading heads (the clearance is really small, usually in the nano-scale). Perhaps that it is possible that the magnet affected some magnetic particles in the reading heads.
If you have seen a hard drive in shape under a strong microscope, you can see lots of little dots. When it is affected by a magnet, then you can see (under a microscope) a few wiped lines. The structure of the platters are away and the bootsector too
HDD stores not only user's data but also some additional information about it's sector and track layout, part of its firmware (in most cases) and some miscellanous data (SMART data, defect and relocation map, etc.). If this data is corrupted, then that disk is cactus. Very old HDDs could be low level formatted, so they could be fixed after such failure. But modern HDDs can't (In theory they can but no vendor gives you specialized software required to do this).
ok so let me ask does this mean that an individual could make the software required to fix the drive or would the company have to do that?
Probably manufacturer would be able to repair it as long as there is no physical damage to the platters and heads. They have to somehow create sectors on new blank platters. I don't know if that strong magnetic field didn't affect magnetic properties of materials used but if not, then it should be possible for them to fix it.
pvc It's cheaper to get a new HDD then repair a broken one
no. it would cost nothing to connect a specialized controller onto the drive and let it to rewrite the missing blank information
Gabe Merritt Only because the tools that would be required to do so are not commonplace, just like with screens.
Heya, loved your experiment. As for the examination, my guess would be the reading head's got magnetic and thus unable to read at all. But the data on that disk wouldn't be exactly readable either. Keep up the fun and informative videos, certainly makes my Sundays fun. regards warukata
It's good to learn about stuff that you didn't know about before and how they work. Thanks Brainiac keep it up 😊💜
+KillerPhoenix 25. (KillerPhoenix25)
Thanks for watching and commenting! More 'educational' videos to come :)
This is pretty cool, I have a 3TB Seagate Barracuda that shit itself a while ago and I kinda want to dismantle it now
Subscribed for the superficial informative videos!!! Love the way you murdered 2 hard drives with magnets...
I'm a little late, but I can tell you that the "clicking" sound is the R/W head going back to the parked position to attempt to fix a problem. This can be caused mostly by 2 factors.
The first factor is physical damage to the platter or head that stop it from reading the disk. The second factor is the destruction of the servo-information and partition map on the disk. They are usually inaccessible by software and they tell the disk controller where the head is located on the platters and where files are located. When you placed the magnet on the hard disk, you probably erased the whole drive including the servo-information. When you spin the drive up, the head doesn't know where it's at, and attempts to fix it by reseting the head position.
nerd
as soon as I saw clicking realized the heads were fucked
Man i just wanted to stick a magnet on that disc when it was spinning xD . I just wouldn't be able to resist doing that if i were in your place lol. Cool stuff man.I enjoy watching your videos. :D
If you could, do a video on how you store all your magnets, it would be really cool and I think a lot of people will like it!
Nevin Williams is correct, the lower platter is pre -formatted with addressing and vital information necessary for the heads to know where they are at and fine tuning for reading and writing. Once erased all the ques necessary for the transfer of information and first transmission and hand shaking is lost. Heads do retries and verification and that is about it.
it almost looks like the read heads were bent up
I'm not sure I understand how harddrive RPMs work...if the platters spin at 120 RPM, how does that equal 7200? Are there 60 layers to all the platters in total?
I've taken about 100 hard drives apart. I love the platters and the magnets. I found some really cool small tubular magnets and some small square ones that are really powerful for their size. This doesn't include the larger magnets that are inside.
Now I know what sounds I don't want to ever hear :D Great vids btw!
Fantastiske og meget lærerige videoer! Har set med i et par år nu, og jeg synes dine videoer er lige spændende hver gang. Du laver bare dsv. ikke så mange :/ Men bliv endelig ved, for du er i mine øjne en af de få RUclipsre tilbage som ikke laver videoer for pengenes skyld, men for at lære fra dig.
Tak skal du ha'! Mit dagjob sørger for økonomien, så indtjening fra RUclips er kun for at købe bedre udstyr og mere interessante ting til videoer. Jeg laver videoer, fordi jeg elsker det, og videoerne bliver godt modtaget af de fleste.
Det også det rigtige at gøre! Se nu at få lavet nogle flere videoer, ved det kan være svært at finde tid, men syntes det lang tid at vente ;(
It made me laugh when you ran the drive without the head guides. You're such a boss.
why do you need sata to usb converter? isnt it easier to buy just a ready for use cable?
Love your videos. And appreciated the dannish lesson. A lot. Thanks!
What strikes me most about Danish is the simplicity of the words.
You should have taken the magnets out of the HDDs and demonstrated them!
+RazorBeamz
Yeah those are pretty strong as well, at least if they still are like in the old days... Those could bite you pretty bad!
+RazorBeamz
They are nice magnets but a little small in a fairly modern 2½" HDD. But I did it in a previous video with an old 3½" HDD: ruclips.net/video/ep_VCotA1ZY/видео.html
Thanks for watching!
I'm think that the part of data that makes the disk be visible was washed out
Hey I found your channel a few days ago and I think it's awesome! I like your videos they are very interesting and educational. I hope you will upload new videos soon!
+kubus7653
Thank you very much! And welcome aboard if you have subscribed :)
I do upload videos 'regularly'. But I spend an enormous amount of time on my videos making them as good as I can. And they are not blog/vlog/gaming videos that often can be made in a day ;)
So usually there are 1-3 months between my videos. I try my best to upload every month but I'm an amateur RUclipsr with a dayjob, family, friends etc. to take care of besides my channel. But I promise my next video will always be worth waiting for. Thanks for watching!
+brainiac75 I actually didn't expect a reply from you, I'm surprised, because most youtubers don't reply to their viewers' comments. And you are right. It's better to spend more time on a vid to make it flawless than make a lot of videos but low quality.
Interesting video as always, great job!
It could be tiny little floaters that hover very very close to the platter that are not the heads, but near the location of the heads that might've slipped, and flew away resulting the actual read/write heads to actually touch the platters (the read/write heads aren't actually supposed to touch the platters, it's only supposed to hover very close above them)
Your massive magnet magnetized the core (the spinny thing), and the head reader is being attracted to it. You can see it getting stuck @ 2:18 cool video!
The magnetic feild caused the disk to get wiped since there is also a magnetic feild on the tip on the arm that aligns the magnetic particles on the disk creating 0" or 1". The monster magnet caused the aligned 0" and 1" on the disk to get moved around rendering them usless for the arm with needle to reed
My guess would be that the basic magnetic structure of the platters has been damaged or destroyed. Mechanically it looks fine but magnetically it probably looks like a total mess. The computer is probably detecting it as a drive initially but when it comes to actually read the parameters of the drive it's getting rubbish so it doesn't show it in the drive listing. You might be able to plug it into a linux machine and get a bit more information about what's going wrong.
The noise from the hard drive is because the heads have been bent by the magnet, and when passing to the disc, the heads are not aligning correctly. Also one of the heads may not be touching the disc and there for not reading all required bits. The hard drive may have not shown in in Explorer because it needed to be initialised.
the hdd's controller also store some data on the platter itself if im not mistaken. if that data is wiped the controller will fail to bootstrap
Probably the service area got erased by the strong magnetic field. The strange head movement is due to the heads trying all the various backup service area locations on the HDD, and finding them all corrupted (a HDD always has multiple copies of the service area, in case a bad sector develops in one).
If I were to take a guess I would say that the large neodymium magnet permanently changed the polarity of a ferromagnetic piece of the drive. As I don't know for sure, it would be cool to look for field lines in all of the pieces with some magnet viewing film.
The dreaded HDD click noise. Still triggers anxiety attacks after all these years.
hey dude, how do i get the magnet inside the hdd out? one of em broke and i tried to get the other but it aint budging, thanks.
+GanonSlayer
They are fragile so there are no guaranteed way to get them of the bracket unharmed. The best way seems to be this: ruclips.net/video/GQhSy-Kz5DM/видео.html
Did you get that USB to SATA/IDE thing from Aldi? Because my dad has that from aldi.
No, not Aldi. I ordered it from a British eBayer.
OK. Could you give me a link to the item?
Tux Ace I bought it from hkstoredirect but they don't seem to stock them anymore. This is similar: www.ebay.co.uk/itm/High-Speed-USB-3-0-to-2-5-3-5-SATA-IDE-HDD-Adapter-Cable-UK-/131386282284?hash=item1e973b912c:g:6~gAAOSwnDxUmSSX
Hi,
Either the header to disk gap is misaligned(increased) due to the pull of the magnet or the magnet which moves the header in the assembly lost its power due to domain misalignment.
Can you attach a glass plate over the HDD and run it with a magnet ?
Every millimeter surface of the disk have negative and e
positive attractions that are encoded like 101010 etc. When a powerful magnet gets to close to the shiny disk parts, the plates become demagnetized, rendering them useless to read or save info.
I watched this video years ago and now it gives me this in my recommended
Do you think that your Giant magnet might of breeched the shielding of the HDD's magnet, thereby altering the internal HDD's own Nxx value?
That's unlikely. Neodymium magnets have high coercivity, so they are very hard to demagnetize/degauss with another magnet.
Thanks for watching!
My pleasure and thank you.
Try connecting it to your laptop and opening Disk Partitioning Program to see if it detects it. Run a cmd and start disk part program, check if it's working there, usually windows doesn't recognizes not working disks from the normal gui.
I feel like the pcb on the bottom side of the drive may have had a ram chip that was damaged (I've heard that at least some are controlled magnetically) because to my knowledge, no operation crucial data are stored on they platters. I believe it should be on rom chips and the computer should be able to detect it if the platters are blank.
Always great Brainiac75. Make's me want to experiment with magnets.
The reading/writing head was bend up, so that it is too far away to read or write any Data to the disc. This distance is still to small to see anything.
I still have a couple of those magnets "sticking" around from HDD I smashed open like 20 years ago.
Maybe magnetized or misaligned the read/write heads causing the disk to be unable to locate tracks hence the repeated seek. Could have burned out heads as well.
Only partitioned drives will appear in Windows Explorer, it's possible the magnate erased the MBR or GPT partitioning data casing it to not appear in Windows Explorer. However if you check in Windows Disk Management or in the BIOS it might appear, another potential issue that the Vista system could have be installed on GPT format and when connected to the older XP simply doesn't appear because the BIOS is not forward compatible with GPT, the drive may have appeared had it been tested on a GPT enable UEFI based computer. If you have any questions about my rambling just ask
I'd like to see if it's how long you're holding the magnet to the disc drive, try taking a laptop without a ssd and just pass over it quickly and if it has no effect slow down on the next pass and see if it's the amount of time near the magnet breaking them
"Alright, enough Danish lessons for today." 😂😂😂😂
Hard drive stores diagnostic and running information on it's platters. If that's wiped out by the magnet the hard drive won't be able to function correctly, if at all. Newer hard drives have those plastic "skids" to keep the heads from scratching the disk during transport.
Basically what happened, is that the magnet set all of the poles on the platter to one polarity (north or south) erasing all data. There is a spot on the platter near the center of the platter that tells the head where it is, and that was erased. So, the head couldn't find where it was and read any data, effectively destroying it.
1st thing i would have tried would be to reformat. That is what a magnet messes up, the magnetically stored data. But if the tracks are completely gone I don't think a regular reformat will create new ones, it can only diddle the data in them.
I think like others said, the data and formatting have been destroyed / damaged. But I think you could have repaired the hard drive if you didn't open it, you could have completely wiped the whole disk with dban data destruction tool, then reformatted it. You said it was not detected by the computer but if you plugged it in the computer and booted into bios mode, then use dban to completely erase everything, then reformat it, that could work.
Last time I did the magnet to an HDD it bend the little prong on the heads so it didn't read the platters, but it also formatted the platters to nothing
Running it without the head guide just makes it beep at you
The big magnet maked the screws magnetic because of this, the reader arm going to the magnetic parts ant makes the "click" sound.
What about the faraday cage that is the HDD case? Maybe you erased the eeprom memory that is mounted outside the case on the board. They use this eeproms with the firmware for determine the sectors and so on...
The hunting suggests it is trying to find the servo tracks which you successfully erased.
3:54 disk sounds like a heart monitor machine or what ever you call the beeping heart noise
If you had not opened it up you could probably find a method to run a low level format by connecting some diag port. The drive normally has preformatted sectors and couldn't find anything it expects anymore.
Maybe the extreme magnetic field just threw the HDDs servo motors out of alignment? It might think it's moved the read head to the right position, but in reality it's off by a few millimeters.
It possible recovery data, I accidentally sticker one magnetron magnet in my hdd about 15 min?
The magnet erased the magnetic data on the drive. Just like it can do to a video or audio tape.
The magnets demagnetized the magnets (or really screwed them up) that are used to control the read write heads and the magnet probably destroyed the low level on platter firmware. As a result the hard drive was completely rendered helpless. Data recycling companies often use these kind of magnets to completely destroy the low level formatted sector tracks and then the data so it will be gone forever. The magnet also probably warped the read write heads just enough short before a head crash but to prevent them from ever being able to read or write data anymore.
I think if you put the magnet near a floppy disk it will also erase data since floppy disks store data the same way like a HDD . Just that floppy disks are made of a different material.
As a CompTIA Certified IT Technician, your Neodymium Magnets caused permanent erasure of the blocks in the platters themselves, and also made the Drive Head unreadable.
This kind of damage is reminiscent of what happens when a hard drive is dropped. The head may or may not crash, and if it does it's unlikely to cause scratches on the platter as the platter is a hard material. It's possible that the head might be damaged by a crash, the arm is almost certainly misaligned (as when a rotational drive is dropped) and there may be other signs of damage such as traces, magnetic components and components sensitive to static electricity since you were probably causing electromagnetic induction throughout the laptop... It might even be possible that other components within the laptop have been damaged, for example I'd love to see memtest86 results! Though that'll probably require a new fan as the system will get HOT... I also noticed that both screens were reacting strangely. Was this due to the GPU overheating when the fan was exposed (and probably stopped)?
hard drives and USB drives have "files" on them that allow basic formatting and detection on computers. your magnet removed those basic files needed for detection. hard drives use magnetic discs with ether a positive attraction and a negative attraction... a 1 or a 0. these attractions make up the basic binary, so that's why a magnet wipes the data, it disrupts all the set attractions to a 1 or a 0 depending on what side of the magnet is facing it.
3:44 i am not an expert in hdds but i heard a scratching noise from the readwrite heads
My guess is that the hard drive deleted the data by switching the two poles North and South. The platters deleted data simply.
So the head if is on the platter it can detect data but by just on or off of the platter thats why it can detect the data.
U should try magnet vs SSHD or solid state hybrid drive, its basically a hdd and sdd bond together
I think the giant magnet loosened a screw on the writehead's rotatey-joint and made it less aligned, causing the clicks.
When it did that click, it seemed like it was trying to scan but couldn't.
I've seen a video before, though it was long ago, where someone opened up a hard drive, powered it up, and released the discs while they were spinning at full speed, you could do that to still get some fun out of this old busted hard drive!
+XANApwns
Hehe, that actually sounds a little scary. They could travel quite far and fast at the rate they're spinning.
I think in the setup he used, he had a small room where he put together a quarterpipe against the wall so they would travel up it and get some air
check in "create and format hard disk partions" to see if any weird disks show up
1:54 why did that poor paper make me laugh xD
The carbon pads that make 'contact' with the platters are actually rather sensitive, so it is possible that the heads were offset. Another possibility is that the magnet corrupted the firmware, though that is unlikely as it is usually stored on solid state chips.
+SPEEDYxArcher
You could be right. The head/plate-distance has incredible low tolerances and is crucial for the HDD to work.
+SPEEDYxArcher My guess is that the magnet corrupted the boot sector of the drive - the firmware is fine, it's trying to scan properly, and the head stops where it needs to, but it just can't find the thing it's looking for.
+Ardusk if there were no boot sectors then drive would report no os erorr to the bios
Grzes kop I'm pretty sure that the "Error 2100: HDD0 (HardDisk Drive) initialization error" that we got in the last video is more serious than that. For a "No OS" error I imagine the computer needs to boot the hard drive, check for an OS, and then get back with a negative. If the boot sector is malformed, then it can't check for the OS because it simply doesn't know where to start looking.
Those USB sata things don't work as well as an actual motherboard ... maybe it would have still worked.
Atleast you could have seen if the drive was recognised in the BIOS. And if it did ,you could have used a disk utility in Ubuntu to save it.
Maybe the magnetisation was weakened by the magnet, and was strong enough for a big jolt outwards, but only retracted back in by the disk spinning and pushing it back. Another theory is that (as a fact) the distance between the disk and needle is thinner than a human hair (hence why its game over as soon as you open it) but the magnet bent it enough so it touches, however that theory would make sounds as soon as it started, so to refine that theory, the airflow that the disk produces is enough to create a cushion around the needles, until it retracts back inwards, but pushes when going out.
the reason why im saying inwards all the time is because if you listen and watch carefuly, the noise only happedns when the needle resets