Disk drive fired 'Frisbees of death' across data centre after storage admin crossed his wires The working week's winding down once again and that means it's time for another edition of On-Call, The Register's Friday tech support tale recounted by readers. This week, meet “Tim” who in the 1980s worked for Data General. Yes, that Data General, the one that EMC acquired for US$1.1bn in 1999 so it could make hay with the …
Not an uncommon incident.
The university I went to did not repair the door of the computing room after a similar incident for many years. It had several shards from a plate embedded into it. Someone apparently did not fix the pack properly. It was a living reminder - computers can be dangerous :)
There were other funny stories from the same period - like the engineering school VAX had a well established boot mark on one of the cabinets. The operators were told - if it gets stuck, you reBOOT it. BOOT there and exactly there. Geddit?
Unless you saw it yourself, and can give me the make & model of the drive in question. But you heard the story from someone else, who heard it from ... It's "a friend of a friend", all the way down, with no details. I was one of the guys who used to go out and trouble-shoot these stories (for DEC in my case). It always turned out that somebody got pissed off, ripped the drive out, and threw it at the wall/door/whatever.
If you kicked a VAX on my watch, you'd get exactly what you asked for: A voided warranty. We had shock indicators inside for a reason.
Unless you saw it yourself, and can give me the make & model of the drive in question. But you heard the story from someone else, who heard it from ... It's "a friend of a friend", all the way down, with no details. I was one of the guys who used to go out and trouble-shoot these stories (for DEC in my case). It always turned out that somebody got pissed off, ripped the drive out, and threw it at the wall/door/whatever.
Quite. Looking at those packs you'll see that they won't come apart and fling their platters around without a serious amount of energy being required. Way more than a linear drive actuator can deliver. Orders of magnitude more. One thing, the actual mass of the head carriage is pretty low. For obvious reasons. There's a hefty magnet as part of the positioner, but that one's stationary.
Yes, you'll ruin the pack, and quite likely the positioner as it rams into the end stop. But separating the platters from the spindle? Not bloody likely. And that would have to happen before the drive would start flinging the platters, frisbee-like, across the room.
One of the early DEC RK-series drives had a positioner that used a photocell and a striped glass plate as feedback. Unfortunately, the glue used to affix the glass to the positioner wasn't quite up to the task, and under heavy load (such as the drive exerciser) the glass had a tendency to become loose. Ergo, no position feedback, resulting in the positioner trying to join the heads with the spindle with great gusto.
This post has been deleted by its author
I was not there but saw the effects of a rotating drum (preceded the spinning disks) that had come off its thimble and crashed through the cabinet, down a ramp, through a major door and across the hall.
I don't know the speed of this thing (attached to a GE-635) or its weight but I can imagine it was quite substantial.
As far as those possibly apocryphal stories about disks of death, we did used to dare each other to stand in front of a drive unit and press the emergency off button. This button was supposed to apply an immediate brake to the disks, probably making them useless but also possibly causing them to fly off at various angles. The height of this rotating mass of rusty bits was just below waist level. In those days most of us cared about the equipment down there.
I've head both the platters and the boot story many times from unrelated stories.
Those packs have a pretty solid spindle and the heads pretty much vaporise under the conditions described. I have in my spare room a platter with a deep gouge in the edge from a DEC RA60 that an operator crossthreaded swapping disks in a hurry. The people present recount a sound like a gunshot when he mounted the drive from RSTS. The heads had to be vacuumed out of the unit. The possibility that a poorly maintained or mistreated pack could come apart can't be ruled out though.
As for the boot, the networking consultant Bill Hancock told that story about his own VAX/750 at a DECUS UK convention in the mid/late 1980s and knowing him it wasn't original either.
Unless you saw it yourself,
I have seen the aftermath. The door, the shards and whatever was left of the drive cabinet.
I forgot to add a detail here - a couple of years prior one lazy sod (I know who too) got too tired of going through all the motions when changing disk packs and disabled most of the safeties.
Have been there, did not do that.
As a graduate student in EE, I managed to get a little independent work experience by hiring on as a replacement assembly line tech at the local DEC plant in Westfield MA (during the summers of 1976 and 1977). It was there that LA36 DECWriters, VT7x video terminals and RK06 disk drives were manufactured and tested.
I was hired as a final test tech on the RK06 (The RK06 was short lived, as the double density RK07 came out as we were ramping up the production line.) disk line. I was given a test procedure, schematics and a theory of operation manual, and told to test the drives and debug as required. I cannot stress enough how valuable this experience was, as I look back on it 40 years later. I got to work on a real design, see how real engineers did things, learn about manufacturability and testability, and get paid for doing it.
In any case, I was warned about the linear motor, the heads and the alignment pack. The RK06 had three or four "quad" modules, one of which was a linear board. It drove the linear motor, getting velocity and direction feedback from an optical slide and a couple of infrared sensors shining through it.
There was an emergency retract system, which applied battery power to the linear motor voice coil when feedback was lost. I think you can see where this is going.
While testing, there was a switch which could be used to disengage the voice coil drive. This was used to test the optical feedback sensors before closing the loop. There were other checks required of the wiring between the linear motor and the drive circuitry. Only after the feedback circuitry and the linear motor connections were verified good, was the switch to be closed.
I have personally seen, and heard, the results of miswired voice coil connectors. IN becomes OUT and vice versa. With speed too fast for human reflexes to react. It was common to rest one's hand on the top of the magnet housing, while using fingers to move the head carriage, while testing the feedback circuitry. We were warned multiple times to keep our fingers clear of the gap between the carriage and the housing. MOST of us managed to avoid getting our finger tips nipped by an unexpected emergency retract (feedback and wiring good, problem in the linear board). Some of us enabled the feedback circuitry without having adequately checked the connector. The result was usually heads driven into the alignment pack spindle, or nipped fingers.
As the platters were aluminium, and the spindle speed was a constant 3600 RPM, determined by power line frequency, there wasn't any possibility of the platters shattering (except, perhaps, by mechanical defect). The whole pack (2 platters, three data and one servo surface, 30MB?) was pretty solidly build, as I recall. But heads were routinely sacrificed.
Good times. I found my documentation last year, sitting in the bottom of an old file cabinet.
"We had shock indicators inside for a reason."
Indeed; and it's all of that hugely costly tenuous and superflouous value add stuff that DEC added that eventually put them out of business rather than keeping it all lean, but under a fantastic customer support policy.
"hugely costly tenuous and superflouous value add stuff that DEC added that eventually put them out of business"
Actually, I think DEC went out of business (or value lowered until they were bought up, more realistically) because they were "old tech" and the computer market had moved along. They sunk a LOT of effort into "Alpha with NT" with little effort in making PCs and cutting edge peripherals. HP, on the other hand, "did that" and survived the 90's.
And Sun won where DEC once ruled. Maybe Solaris helped?
[shock indicators are cheap - a cost benefit if it helps you prove that someone kicked the cabinet causing a head crash, thus voiding the warranty]
Amen to that. Still scarred by an incident over forty years ago, when I stood on the front of a narrowboat stuck on a sandbank, paying out a rope to a kind passing boatperson which offered to tow us off. Attached rope to the front of our boat, then handed other end to kind boatperson, who attached it to their stern, and set off. Rope gradually emerges from water (therefore dripping wet and even heavier), staightens, tensions, then abruptly detaches from rear of other boat and flies directly back at me, impacting between waist and mid-thigh.
Good thing I never intended to breed.
(Facepalm icon because it was a "D'oh" incident, but also because there's never a crotch-palm icon when you need one.)
Trying to stop an approx 4" mooring line disappearing out the fairlead in my naive yoof I did the time tested method of slowing it with my boot. Until my boot snagged and got pulled complete with leg into the fairlead.
A testicle stopper on one side, and a large friction burn up my right butt cheek for my trouble. Stopped the dinghy with outboard that was pulling it to the buoy in a hurry....
Ah well, luckier than the docker in the Gulf on the previous trip who lost his head. Literally, as the polyprop multiplat rope, as above, let go (from a fellow officer present at the time... not pretty)
I won't mention anchor chains parting on the winch inches from your face....
At least wire sings before it goes. Quite sparkly as the strands flick against one another. Polyprop tends to stretch a lot, and then give up in a rush.
Saw plenty more incidents and accidents. I always gave anything under tension a wide berth.....
I used to commute on the Mersey ferry(*) some score of years ago.
They would tie one 4 inch thick rope round a bollard and then swing the ferry boat against the motors to align the stern with the Seacombe landing stage. The ropes complained loudly when the tide was running. I tended to stay back a little until the drawbridge went down.
(*) 1200 tonnes, four engines.
This post has been deleted by its author
Some years ago I went to a steam fair. A traction engine had descended a moderate slope to a stream (presumably to take on water) and had sunk to its axles.
A heavy Scammell equipped with a hydraulic winch was called to the rescue, it sank a barn-door sized spade anchor into the ground and connected up a rusty 3" wire rope. As the winch bit, the wire rope pinged and 6"lengths of broken wire stuck out a right-angles to the rope. vibrating. There were loads of people around, including kids, and no-one seemed bothered. I beat a very hasty retreat and hoped for the best.
"Establish no go safety zones for assistants and onlookers."
When an RAC man was winching my broken down car onto the back of his truck, he made a point of asking me to stand at the front of his truck with him.
It only took half a millisecond for the image of a snapped winch cable in motion to register in my mind.
Yup - have seen a neat(ish) hole punched right through the back of the tailgate of an old Range Rover. from exactly the same situation. Looked like a cannon shell had gone through it.
There are plenty of videos around where people use the tow ball for vehicle recovery rather than a recovery hitch. Some of them show deaths as a result of said tow ball travelling like a cannon ball.
"a lovely spiral "rope" pattern embossed"
A guy I knew years ago had a similar mark diagonally across his chest from trying to use a retired "Gold Line" climbing rope to pull his dad's car out of an exceptionally large pothole formed from a minor flash flood. Warning: If water is running over the road, it also might be running under the road.
I was leading a horse into a horse box and tied it to the ring behind me. Before the ramp could be shut the horse decided to back out. The nylon rope stretched longer and longer until the brass loop on the horses head collar snapped. The catch on the end of the rope shot past my head and punched through the sheet behind me pulling the rope through the hole.
On my first day in broadcasting I was introduced to the reel to reel tape machine - a Lyrec from memory. I was then introduced to the razor blade and how to edit audio using the two. Next piece of information was that razor blade did not live on the tape machine. To explain why I was shown how quick the rewind and fast forward spun the tape and reels. "Imagine that going at speed and a razor blade going with it". I agreed and was actually very nervous I'd do it. Then I thought that the razor blade was quite light. It would therefore be unlikely to stay on the reel whilst it got up to speed.
It was only later in the day that I spotted the some marks in the wall at desk height. Looking at the size and shape they were obviously from razor blades that had been left on the machine. Someone told me subsequently that if you left your blade towards the centre of the reel it could get up to some speed before coming off. I bought a magnet that sat in my desk drawer and I covered one end with some loosely held hard plastic. This held the razor blade securely covering the sharp edge when I wasn't using it.
I notice you do not report on the efficacy of your splices.
Using a razor blade that has been stored on a magnet, means that you are using a magnetized razor blade, on magnetic tape.
That produces loud bang, or pop noises in the resulting tape
But then I note that you only referred to your work there on the one day.
I only used the magnet stored blade for when we were reclaiming tape reels. So much easier to just slice through worn tape on a reel rather than unspooling it. The first time I did it I unspooled about an hours worth of tape and filled the small office bin and then some. A lot more fun to do it that way but far messier.
Otherwise I had a bit of bluetack and a oil paper holder (to prevent rust) stuck to my the other side of the drawer tray. That held the working blade.
The first year I helped out with the tape-store clear out at a studio in north London, I just took one side off the reels and let the tape unwind into bin bags.
Spending the next morning collecting lengths of 2 inch tape that stretched the length of Highbury New Park from tree to tree taught me that cutting the tape off the reels was the way to go.
Wait. ‘Magnet’ plus ‘audio tape’? I’m surprised you kept the job. The last thing in the world that you want is a magnetised razor blade being used for editing tape! Must have have made a lovely thump as the edit passed the heads...
And never in all my many years in recording studios did I ever ever see an engineer or assistant leave a razor blade on top of a reel. Ever.
A ‘budget’ studio I worked in once had a Brenell 1” 8 track machine, this being the cheapest on the market. This meant it lacked some sophistication in the transport department. So you could go directly from fast wind to play without the transport slowing the reels down before engaging the pinch wheel and capstan. This produced what was called a ‘bootlace’, which was a few feet of tape stretched from being 1” wide to about a 1/4”. Needless to say this would always occur somewhere in the middle of a master take.
A full reel of tape on fast wind is a sight (and sound) to behold.
When I was at university somebody forgot to tighten the 10kg chuck on the machine shop lathe. When the lathe spun up to speed the chuck came loose, hit the floor at 2000rpm and took off like a rocket.
Years later the hole in the wall of the machine shop was still there. Pour les encourager les actress.
"the hole in the wall of the machine shop"
According to legend something similar happened to a 20k rpm biology centrifuge at a university I worked in when the pulse counter went wonky. Only it was on the 5th floor.
I read a similar story in the late 1970s about a 2MW hydro generator which lost its oil supply sometime in the past. Before the operators noticed and could (manually) stop it, it levered itself out of the cradle and thankfully went downstream for a couple of miles (upstream was the dam wall). The article noted that contemporary hydro turbines were 600MW with automatic emergency shutdown systems and in the one case of oil failure documented up to that point, the ends of the bearing holders were visibly glowing red hot before they managed to stop it.
Almost all those floor standing washing machine drives used mains-driven synchronous motors for the platters, so I'm surprised about overspeeding. About the only thing which sounds possible is that the head carriage shattered the spindle and freed the platters. Even at 3600RPM they'd be scary things.
About the only thing which sounds possible is that the head carriage shattered the spindle and freed the platters. Even at 3600RPM they'd be scary things.
It's just that their relative energies and robustness would make it extremely, if not totally unlikely that the positioner could cause the platters to break free from the spindle. It's feasible with a pack not locked down on the spindle correctly, but in that case just spinning up the drive would cause havoc already, no need for the positioner to join in the fun.
I've witnessed one incident of a chuck key launched into a cinder block wall where it ended up leaving a nice hole. The wall had several other scars in the vicinity indicating that this was not the first such incident, just one of the "better" ones as it managed to find a spot where it could make a hole.
This was in shop class and the idiot student that left the key in his chuck got a boot in the pants from the instructor. You can't get away with that anymore, but then you can hardly find machine shop classes in high school anymore either.
You can't get away with that anymore, but then you can hardly find machine shop classes in high school anymore either.
Careful! Even mentioning the term 'machine shop" is almost grounds for imprisonment these days, in case one of the poor wee dears has nightmares about getting a splinter in their finger.
I don't know about that. Recently had a tour of the high school my young 'un will be going to next year and they had all sorts of cool stuff in the workshops.
Did you notice the castor wheels underneath, where they roll the stuff out during parental unit tours and put it back into hiding at term time? Did you notice how the female parental units were steered into one room while all you male ones were steered into another (rumour has it the rooms the ladies see have large amounts of cotton wool everywhere, so while the menfolk are seeing nice toys they can only drool over using (and hope their sprogs come away with nothing worse than bone-deep cuts that miss nerves and arteries), the ladies see a very protected environment where a paper cut on a student results in the immediate firing and life imprisonment of the teacher)1
1 Implied stereotypical sexism not intended, even if it appears this post was written as such!
That sort of thing really is common. In ten weeks of workshop practice we had the same idiot put one piece through the roof while turning between centers and then, not two weeks later, explode the stone on the cylindrical grinder. When a stone breaks at 5-9,000 rpm it has real energy, and the parts head out of the machine at speed. Even old 14" hard disks had nothing like enough energy to get out of the case.
"Not an uncommon incident."
It should be. Assuming this story is correct and no apochryphal, the linear motor should have had a speed limiter and a cut out that operated as soon as the linear motor had rotated enough times (if it was geared) or beyond a certain angle (if it was direct drive) that the head would be off the disk or squashed up against the spindle. These are the sort of things industrial systems designers have to consider, obviously the world of HD manufacturers was different back then.
the linear motor should have had a speed limiter and a cut out that operated as soon as the linear motor had rotated enough times (if it was geared) or beyond a certain angle (if it was direct drive) that the head would be off the disk or squashed up against the spindle.
A linear positioner in those disk drives is not something that rotates, and it's not geared either. It runs on rails or sliders, driven by a voice coil enclosed in a magnet. Not unlike a loudspeaker. The limits are stops on the rails of some sort, usually its mounting points.
And if the feedback is knackered, there's no way the controller can tell the positioner is moving too fast or past its range.
if _I_ had designed the thing, cutting ANY wire would detonate it.
I can think up many ways to defeat someone trying to disarm an explosive device by cutting a wire, like "cut any exposed wire, or remove any cover, and it goes BOOM". [real bomb disposal would probably just safely transport it in an armored container, then detonate it someplace away from people+things]
Good thing I'm not a terrorist, yeah - but it's the way white-hat hackers think - anticipate what is being done by thinking about how YOU would do it, muahahahaha!
/me thinking of 'Lethal Weapon 3' at the moment...
afterthought - I think the military deliberately detonates IEDs by use of claymore mines and/or C4
You might wish to have a safe method of disarming the bomb in case your own aircraft has to land again without using it
No, because if you have a method of disarming it - then so does your enemy. And from watching varuous documentaries on TV, it's clear that the Germans did booby trap the detonators in the stuff they dropped on the UK. So we had to develop various methods of disarming the bombs without triggering them - one of which was to physically cut a hole in the bomb and take the explosives out (IIRC it was steamed out and then shovelled up off the floor or something like that).
If you have armed the bombs and then can't drop them on the target (or any secondary target) - you simply ditch them in the sea on the way home. I believe a heck of a lot of UK bombs were dropped in a specified zone in the Channel - and there was a theory that Glenn Miller was killed when his plane was hit by a bomb being dumped after an aborted raid, but it seems that theory has since been debunked.
It is possible to design a bomb with 'collapsing circuits' so that if any wire is cut the bomb detonates. That has been true for a long time.You can even include fluoroscopic detectors that might identify attempts to x-ray the device and detonate it. One way the EOD teams used in WW2 to get round this and timers was to pour liquid oxygen into the fuse cavity to freeze clockwork mechanisms and stop current flowing to the detonators.
Modern techniques usually use a container of water that is used to 'disrupt' the device using an small explosive charge to drive a wall of water through the device, disrupting it without sparks an removing thed explosive from the detonator and timing mechanism.
So this is where someone called "Stumpy" makes his debut On-Call post ?
I still have scars on my arm from where the brown workshop coat I was wearing got caught in the lathe spindle at college.
Just as well that (like most of the stuff at college) the coat was old and worn and so the sleeve stitching came apart quickly. Otherwise I'd have been typing with only my left hand for the last 32 years.. Also luckily, I was cutting stuff at a fairly low speed.
(Moral of the story - always check that opening the safety guard *actually* stops the lathe like it's supposed to and that some previous student hasn't disabled the safeties. I've never seen a college lecturer go so white so quickly.)
No - but I was once working inside a small 19" ethernet switch to clean or replace the fan.
Since my workspace was rather cluttered, I had it on my lap while I was sitting in a chair.... (you see where this is going.)
I powered it up to check the fan plug orientation was correct, and the case went live with 220V. Its rather hard to have a complex coordinated muscular response when sitting down.
So from memory I leaned back, pushing hips forward at speed, sending whoile switch flying through the air to crash into the solid concrete floor. Surprisingly only two ports ceased working due to this abuse.
Blasted dot matrix printed tried to take my hand off.
Got a new dot matrix printer machine in (6 years ago), went to press the Online button and the panel fell in much to my surprise (This Lexmark machine is certainly not made to the same standard of their dot matrix printers from years earlier, with the previous one only being replaced because no official printer driver for it came with Windows 7/Server 2012). I put my hand in to push the button panel back and of course I hit the Online button.
The printer did as instructed and I got away with some cuts and bruises (And a reminder to turn off any machine being worked on if you put your hand inside it).
P.S sorry but thankfully I am not called Stumpy.
> *Because once bitten....
A chemistry lecturer I know used to wander around the hall whilst mixing up black powder in a mortar and pestle. Cue nervous students trying not to be near him.
He stopped doing it after a former student showed up minus 2 fingers. Said student had become a science teacher and tried the same stunt without realising that you need to keep the mixture wet to prevent ignition...
A chemistry lecturer I know used to wander around the hall whilst mixing up black powder in a mortar and pestle. Cue nervous students trying not to be near him.
One of my teachers many years ago recalled how he "cured" a student of being over inquisitive and always fiddling with stuff. He deliberately left a pestle and mortar on a side bench with something unstable in it. Needless to say, when the over inquisitive student came in for the next lesson, he couldn't resist giving it a bit of a grind ...
Mind you, we found that most of the cupboards in our form room (a physics lab) weren't locked - oh what fun we had with the Wimshurst machine. Could get some real sparks off that one ! Then one day someone said "what happens if you put a polo mint between the balls ?", so we tried it - put a polo mint in the gap, held by the balls which were adjusted to grip it and wound it up. There was the usual crack as it sparked - and we could find no trace of polo mint, no bits, no dust, we had no idea what happened to it.
But little Calum knows he's not allowed to touch the video recorder so it couldn't have been him that posted the chewed Duplo man through the slot.
Which does rather suggest that all the foreign objects jammed under the tape carriage are the work of one of the adults in the house...
I know... I once had to pull a banana out of one... And a peanut butter sandwich out of another...
Buttered toast with marmite on it in the case of our family VCR - but it survived for another 6 months before I finally built a compact (ish - this was 15 years ago) PVR. That did solve this particular problem - in fact it took almost a year before my better half asked out loud where the VCR slot was.....
For those imagining VHS or Betamax or Umatic, a VCR before then had to be moved very gently. Especially the Philips N1500 which was like a proof of concept prototype made of Meccano. Even the chassis. Really it's amazing it didn't have valves (tubes) in it. The two reels were stacked maybe to make the cassette smaller or to make the helical path on the drum easier.
Though there was a Panasonic that used a single reel cartridge (based on EIAJ 1/2") and a stiff leader. One way of ensuring tapes were always rewound.
A 1950s EMI giant BBC audio recorder full of valves was less insane inside.
Called a "portable" because it had a lid and handles. Weighed 29kg and mains only.
EMI portable like BBC used in mid 1950s
HiFi mono by using full width of tape.
I used a near washing machine sized disk drive with the large "cake cover" style disc packs. Part of an Intel system running ISIS II. I can't imagine how the accident can happen and surely the heavy metal cover and lid must have been off?
I worked on lots of N1500, 1700s and V2000s from the late 70s into the early 90s and I can assure you they were all hellish, arms, levers, motors, pulleys, bastardo springs & unknown electricikty circuits that somehow managed to work.
The modular versions added extra fun where (ex)rental jocks(tm) would spray everything with servisol (in an attempt to get high or fix it) and then that would degrade the plastic around the module mountings and the modules would then flap about.
Cheers R*Jocks for that extra workload.
We had I believe these style drives for the backup kit and there were trusted *nix nerds that were only allowed to handle anything to do with the kit and they lived & worked within the halon environment ...
Called a "portable" because it had a lid and handles. Weighed 29kg and mains only.
I had a Pye TV like that. Cream metal cabinet with side panels that opened up like a WW2 lorry bonnet for service access.
On that subject, anyone ever try to move an AR88 radio?
Looong, Loong ago I went up to Lisle Street in Soho & bought this wonderful radio - an "upgrade" from the R1155 radio I'd been using for years.
Now came the problem of getting it home from Soho back to my then home in Norbury, SW16.
Luckily, in those days, it was possible to phone up for a mini-cab at a Shilling a mile & get it into the back of the vehicle for the ~20 mile home.
I completely forget just how I got it upstairs to a very sturdy wooden table in my bedroom - or even when, years later, I sold it sometime prior to hitch-hiking around the world. Well at 77 years old - it was back in the mists of antiquity, so I can't recall those time now.
I serviced Creed T150 teleprinters in the R.A.F. We had an instructor with a dent in his forehead, received when he was fixing one. He had removed the airbrake (yes, really), from the carriage return, then fired the print carriage into his head. It probably weighed about a kilogram and was made of cast iron.
What I find interesting about video recorders is how much simpler the later ones were in terms of design.
Granted, a lot of that would have been electronics boards, many of which would have been replaced with a few chips after 25-30 years, but even *mechanically* my parents c. 1990 video was- IIRC- quite a lot simpler inside than my neighbours c. 1980(?) video (which was passed on to them when theirs broke down). (#)
And- again IIRC- the one I bought circa 2004 (##) was simpler still and ridiculously shallow at the back.
Is this all down simply to refinement of the mechanical design over 30 years or so?
(#) My Dad did mention something about the plastic construction quality of some parts of the 1990 model when he was repairing it, but I've no idea if that was a factor in their ability to simply things mechnically. (Why would it be?)
(##) i.e. The last few years of video recorders, when DVRs were already an option. I only bought the former as a stopgap because the price/capaciousness of DVRs wasn't *quite* there for what I could afford, but I knew it would be in two or three years.
"Jesus wept."
Oh, no tears, please - it's a waste of good suffering!
They are smaller but spin way faster. But when an hard disk dies, I meticulously open it, recover the platters and the magnets. I am not sure what I will do with the magnets, but I plan to build a nice mobile out f them: they are shiny and should sound very clearly when they hit eachother.
The field is very narrow and on the surface of the exposed face, the other side of the field is mostly within the metal. It means you have a very uniform field contained between the pair of plates
You don't want a field spilling out over the platters for obvious reasons.
You cannot make a perpetual motion machine with permanent magnets. All that happens is the magnets move until the attractive and repulsive forces cancel one another out, and the system then sits in its stable equilibrium.
In an electric motor, either the supply is an alternating current which keeps breaking the equilibrium; or there is an arrangement with brushes energising different coils in turn via the commutator so as the armature moves, a new set of coils is brought into play such that the net forces change to make it carry on moving in the same direction it just moved -- which would bring it back into equilibrium but for that pesky commutator spoiling it.
The point is, it's the changing of one of the magnetic fields -- whether in the stator as in an AC motor, or in the armature as in a DC motor -- that is what keeps the motor going. If you try to build a motor using only permanent magnets, then there will always be at least one stable equilibrium point, and it will come to rest there. You have to do some work moving everything into its initial position, and this gets stored as potential energy in the system. This may be enough to make it run for a good while, depending on friction and any load applied (and notice, no "perpetual motion machine" is ever shown turning an actual machine of any kind, not even a tiny generator and a single, dim, high-efficiency LED ..... funny, that, innit?) but it will always come to rest eventually, at a point where the static attraction and repulsion balance one another out exactly. And there will always be at least one such position -- if you can't find it, it means you have made a mistake in your working-out. The magnets will find it, every time .....
Oh, and I have neglected to mention weakening of the magnets, which would happen in fairly short order if you ever did somehow managed to extract any usable energy out of a permanent-magnet-only motor .....
This post has been deleted by its author
by cutting the platter and bending it
I think a lot of the modern spinning-rust drives now use a glass platter (or at least the last 3 2.5" drives I've "investigated" have..)
Cutting those might be possible with appropriate tools (although the police might want to know why you have a high-intensity laser cutter lying around) but bending it might be slightly more challenging.
I suppose appropriate amounts of heat might do it.
"I think a lot of the modern spinning-rust drives now use a glass platter"
a couple of decades ago I had a Kalok drive die on me, while beta-testing Win '95 [which was a BIT more disk-intensive than earlier winders]. It sounded like breaking glass when it ate itself. So even in 1994 I think this was being tried...
"I am not sure what I will do with the magnets"
Most from 3.5" drives will hold up, from above, a 16 ounce hammer handily. And, of course, fridge magnets -- but a tiny dab of hot-melt glue or silicone centrally on the magnet will provide a bit of rocker, and easier removal.
Shame there's no mechanism to make it work as told.
The linear motor is fast, true, but it has no power to speak of. It also has practically zero mass compared to the disk pack. A stack of ten of those platters weighs probably 1,000 times that of the entire R/W assembly ... and you'd have to fling them all at the same time because the disks are fixed together so the drive only needs to read a single sector indexing.slit (cut into the bottom hub) for the entire stack.
You clearly have never encountered one of these behemoths. The platters were over 35 cm in diameter, and a pack weighed about 10 kilograms as did the read/write head assembly. It was perfectly possible to "walk" a 200 kg machine across the room, simply by slamming the heads back and forth. Any fingers that go in there, probably ain't coming out.
I used to be paid to work on them. Today, I restore them because somebody has to[0]. The read/write assembly weighs quite a bit, yes. The actual moving mass, however, not so much.
Yes, you can make the drives walk. On well waxed floors. It is very hard on the equipment, though ... regardless, there is no way to fling the platters across the room, alas.
[0] Or I'm a glutton for punishment ...
I've used drives like this (although they were obsolescent by the time I had much to do with them: Fuji Eagles were the common thing by then, which were much more modern, and two people could carry one). You could feel and see the cabinet move when the heads loaded, and when doing any kind of intensive activity like checking the disks. I'm not sure about the ejecting-the-platters thing though: I imagine the mechanism for that might be the (remains of the) head assembly getting wedged between the edge of the platters and the casing of the drive.
"It also has practically zero mass compared to the disk pack."
Really, because from the pictures it looks like it's made of cast iron. What did the disc packs weigh in that case?
Really, because from the pictures it looks like it's made of cast iron. What did the disc packs weigh in that case?
That's the entire head positioner assembly. It's a LARGE (stationary) permanent magnet with the read/write heads moving on a linear guide, driven by a voice coil. It's quite similar to a loudspeaker.
The moving part of the positioner, as with the speaker, should be as low as possible so that the heads can move at the highest possible speed (so, shortest possible seek times), without requiring kilowatts of energy.
I was told tales of guys manfully wrestling out of balance disk drives that would bounce around the floor like bucking broncos.
And was shown a picture of the university's 128kb ram pack being delivered by cargo plane. It was the size of a shipping container.
I'm a little sad I missed that era. But maybe not that much.
Not a drive but I witnessed an massive SMT pick-and-place machine hopping a good couple of feet in the air.
We'd moved factories and someone had the bright idea to put the SMT equipment on the mezzanine floor, you know, the one made of springy plywood.
When these machines start up they run an exercise cycle with the placement heads moving between min and max coordinates at maximum speed. The servomotors involved are enormous as the machines can place nearly 4 components per second. This motion coupled with the springy floor to make for a definite brown-trousers moment for everyone involved.
After power was removed they were swiftly lifted back downstairs and onto a nice, solid, concrete floor.
Not drives, but lab centrifuges. - We were taught to "balance" these by adding a tube opposite the one we were spinning containing a carefully measured (by eye) level of water. One day somebody (else) forgot... The centrifuge walked off the bench and was wandering across the floor when somebody had the presence of mind to turn it off at the wall.
And don't talk to me about ultra-centrifuges...
Christ on a bike I forgot about them.
We undergrads weren't even allowed to go near the ultra-centrifuges. Did they not put out a million G or something crazy like that.
Even the normal centrifuges we used could be terrifying and would regularly crush the sample vessels.
industrial washing machines have (or had) a self-balancing feature that causes a flow of water into 'opposing' positions during the spin cycle. Basically, as the thing spins, it trips a switch when out of balance causing appropriate water flow to re-balance. As the thing spins up, you can hear the water jets balancing them, and then they smooth out. Maybe a similar system for centrifuges? Just a thought...
/me wonders where the IT angle is on this one...
But how much does an "on" bit weigh vs an "off" bit? Does that mean that the drive needs to calculate how many bits are "on" at a particular angular section? And move some of those to other sectors as the unit is spinning at 10,000RPM?
No, that was satire.
But does anyone remember the hole punched in the 8" floppies (really floppy) to synch the read heads? There were times when those large froppy dicks would seem to wobble because of the holes. (The term "froppy dicks" was relayed by a female co-worker commenting on a non-european-descent instructor teaching a class. Does that get me out of the #MeToo toilet?)
There were times when those large froppy dicks would seem to wobble because of the holes.
Extremely unlikely; froppy drives run at 300 or 360 RPM, and the mass of the actual disk is just a few grams.
Sometimes people managed to get the disk into the slot in such a way that the central hole didn't quite line up with the drive hub, and forcibly closing the door would put a nice dent in the side of the hole, or even cut a segment out of it. Usually that was the end of the disk, although I can imagine people reformatting, and reformatting again, and again and again, to save those precious froppies.
Well, I know of one drive that took a dive.
That was when a cable-laying monkey had removed an entire row of floor tiles behind one of those washing-machine drives. An entire row. Not staggered, or leaving every other tile in (only every third tile if you were feeling very adventurous). An entire, contiguous row.
Shortly after, the drive started seeking. And shaking. And with it, the tile supports underneath the drive. And predictably, that bit of floor, with the drive on it, collapsed backwards into the chasm.
After hauling the drive back on to terra firma (it is unknown whether the cable monkey was made part of the floor support during the process) and powering the unit back on, VMS uttered the memorable line "Drive improperly dismounted".
We had a CadCam system that ran on a Nova6 mini, with some IBM 3350 removable disk packs, similar to this story. An air conditioner service guy was checking for refrigerant leaks with a bottle of soapy water in a squeeze bottle, which would produce bubbles when applied to a leaky refrigerant line.
He left the bottle sitting on top of a convenient washing machine size disk box, without knowing the box quivered when the disks seeked. A few minutes later, the bottle fell over, spilling on the removable cover, and dripping into the disk mechanism. A few minutes after that the CadCam crashed, with disk errors.
Unbelievably, removing the disk pack, drying out the mechanism, and replacing with a recent backup disk pack worked.
A somewhat larger disc than the ones described in the article, and probably rotating at a much more sedate pace...
In this image, leaning against the cabinet of a smaller disc unit
http://linux.orbi-online.co.uk/webcams/main.jpg
That looks like the platter from the large disk drive made IIRC by CDC and rebadged by other mainframe companies in the late 1960s and early 1970s..
The 600MByte (yes - megabyte) drive consisted of two side by side vertical stacks of several of those platters. The head positioning actuator was positioned in between them. It was so large that both heads moved the same distance in opposite directions simultaneously to neutralise the forces.
The disk bearings were water cooled. If the mains pressure dropped then the unit would start to shut down - which allowed the water pressure to rise above the threshold. The result was a regular loud groaning sound on those occasions
The whole thing was in an enormous cabinet about 1m (d) 2m (l) 2m (h) and weighed 1.5 ton(ne)s. It had mostly transparent sides so you could see it operating.
When it was installed it was squeezed through the building's access doors and wheeled down the computer room along a central aisle that was solid concrete underneath the tiles. As it was moved off the aisle the false floor started to collapse. Luckily they managed to retrieve it until the floor was strengthened - otherwise it would have meant taking the roof off the building for crane access.
Initially we had regular problems with head crashes - which would carve a bright silver band about 2cm wide round the platter. There appeared to be only one engineer in the company (country? world?) who could set them up reliably.
IBM 3270 drives, I assume relatively similar from the description with all the adjustments made by turning screws 25% or a turn clockwise or similar...
I also recall an NCR minicomputer where air-con failed leaving the room too hot to enter to switch the thing off - that was more fun, and marginally closer to human maiming, if not death.
Another close call was with the same NCR attempting to recycle the rather large, and very solid metal drive housings at decommissioning which near tipped over the forklift attempting to put them in the scrap metal skip. Took a few chunks out of a concrete staircase while on a porters trolly made out of welded scaffolding too. They were certainly too heavy to move by an single, even fairly strong person...This also could have been hazardous to the fork lift driver if he had not been so ready to swap it for the heavy one at the back of the factory as the handy one simply tipped itself towards the resolutely immobile drive housing...
Oh air-con hell! Back in the day I went out to client site (hospital) to work out why their DG Aviion kept crashing. The temperature in the room (shared with a Strowger exchange) was like nothing I'd ever experienced. They had put a fairly heft portable air-con unit in there but failed to notice that the exhaust hose had cracked... And so they kept it on, expecting the room to somehow magically cool.
seeing all of these "grampa had to walk to school, in the snow, uphill, both ways" stories makes me wonder if it wasn't so bad just to have to toggle in the correct bus address before pressing the 'boot' button...
(on a DG Eclipse, in 1980)
> IBM 3270 drives, I assume relatively similar from the description with all the adjustments made by turning screws 25% or a turn clockwise or similar...
The IBM 3270 was a CRT terminal. You may be thinking of a 3370? Take your pick from the drives featured in the Wikipedia article: https://en.wikipedia.org/wiki/History_of_IBM_magnetic_disk_drives
Back in the late 70's, I was working in the oil industry and using DG Nova 1200's and Nova 3's. They brought out the Micronova with really small (for the time) circuit boards.
We needed a small waterproof computer that we could take on the oil rigs. So, I stripped out a Micronova, put it in a waterproof box with a small (around 8", if I remember) screen and two floppy disks (does this design sound familiar?). I attached a keyboard in the lid. It all worked very nicely. I tested the waterproofness by throwing it in the river Yare (attached to a rope, of course), fishing it out - and it started up with no issues.
Now, we didn't have the resources to build many of these ourselves, so I went to DG and asked them if they would build them for us. The answer? "Sorry, but we cannot see a market for a portable computer".
2 years later, Compaq arrived on the scene. To say I was a bit miffed, was an understatement.
Seems they had a change of heart. See 1984's Data General-One. Unfortunately, it ran MS-DOS ... Neither Coherent or Minix of the era work on it, sadly. Hardware issues, of course.
I worked there when the DG/One came out.
DG only wanted to sell batches of them to large corporate customers.
They had an all-hands meeting where someone asked why they weren't advertising the DG/One to individuals in the computer magazines. The answer, from "Captain" Ed DeCastro himself was, "Because they're not our market"
Jaws dropped. They were certainly IBM, Compaq, etc's market!
// had my own DG/One for a while. Slow, with an incompatible serial port.
Notable in the early '70s, perhaps, when EDS4 and EDS8 packs were normal (detect the ICL experience) but by the mid to late '70s EDS200 was the standard size.
I have never seen a disc drive 'walk'. Of course they may have been fixed down, but given the delicacy of alignment I would have thought a vibration severe enough to shift the drive cabinet would have had an impact on the platters.
These were pretty sensitive; cautionary tale told to me was about an operator who dropped a disc pack (bloody heavy things) but was afraid to own up. Next time it was placed in a drive it failed, and also buggered the heads. Cue diagnostics 101. Roughly, "Well, it worked last time in the drive over there. Let's swap the discs over to check if it is the disc or the drive." Result is a disc pack buggered by the now damaged drive, and a new drive buggered by the original damaged disc pack. Allegedly they repeated the diagnostic test several times before someone called a halt and got the ICL engineer to check everything. They didn't damage ALL the drives and disc packs......
"They didn't damage ALL the drives and disc packs......"
IIRC on set of operators did just that with the System Disk - going along all the 8MB drives - and also ruined the reserve "master" System Disk.
In the Kidsgrove Systems Test commissioning the 8MB exchangeable disk drives were subjected to a thrashing that moved the heads to their extremities at high speed for some time. The unit vibration was definitely severe.
When the first 8MB disk units arrived for the prototype System 4-70 mainframe they had to be commissioned by union approved engineers from the factory. By the time the second prototype arrived the local engineers were allowed to install the disk units.
There then came an day when it was discovered that disks formatted on the second machine wouldn't work on the first one. Much head scratching until someone spotted that the disks on the first machine rotated in the wrong direction. The heads were symmetrical - so they flew ok - and the prototype's disk O/S was initialised from magnetic tapes.
The reason was well known to local engineers. The System Test area had its 3 phase wires wrongly coloured. There had never been a time when there was no equipment installed - to allow the problem to be corrected. The local engineers had to remember the correct wiring. The visiting engineers had come in at a weekend and no one thought to tell them about the discrepancy.
Apparently card punch mechanisms did not take kindly to going in reverse.
"Notable in the early '70s, perhaps, when EDS4 and EDS8 packs were normal (detect the ICL experience) but by the mid to late '70s EDS200 was the standard size."
Back in the early to mid 80s we had 200 and 300 MB disk packs. The older guys said that the 200s replaced 50 MB and smaller packs.
"I have never seen a disc drive 'walk'."
Lucky you.I have. I also had to fi the resulting mess.
"Of course they may have been fixed down, but given the delicacy of alignment I would have thought a vibration severe enough to shift the drive cabinet would have had an impact on the platters."
You'd be perfectly correct. Oh, was there ever an impact.
Same story - as a newbie I was keen to follow the rulebook and that said you didn't call an engineer until you had done your own first-line diagnostic work to identify the exact problem. I loaded an EDS200 pack, wouldn't come on line, so I tried the same pack on another drive unit to see whether the fault followed. In fact just to be sure I tried it on several other drive units.
Of course the defective disk pack had broken all of them and I made the mistake of writing up the engineer call with "description of defect: can hear tinkling noise from drive units." My resulting nickname "Tinkle" took a long time to live down.
That same scenario was quoted to me by my chief programmer at my very first job as the reason a certain operations manager was told never to touch actual machinery again and his position respecified to an all-paperwork and supervisory role.
Same kit too. Must have been an institutional thing.
I was told by an engineer that if the big red "FAULT" light ever came on with an ICL eleven-high pack drive, it was critical to leave the room at once and punch the Big Red Off Switch as you went (machine tool factory, cast iron Big Red Off Switches were everywhere).
So while I find the original story to have many questionable details, exploding disk drives are possible under *some* conditions. At least, our chief ICL site engineer felt so.
Tape spools would fall to bits at speed at the drop of a hat, as I recall.
I used to work with Pertec 5MB fixed + 5MB removable platter drives back in the day. These drives would 'emergency unload' the heads from the disk if errors were detected. This was done by discharging a capacitor to the head voice-coil via a relay. A common engineering fix for repeated unloads was to remove the relay...
... and that is why I cannot resist buying ever smaller size, but larger capacity memory devices. The marvel of the ever growing data storage density never ceases to amaze me. Recently bought 200GB microSD SanDisk U1 card. Even though I do not need it (at the moment).
In the 1980s I ran a DG Nova with a 5+5 removable Phoenix drive attached to a very large, expensive, scientific instrument. I got funding to replace the drive with a Winchester. The new drive was mounted in a 19" rack enclosure contained in one side of a of a knee-hole desk unit that housed the CPU etc on the other side. The drive could be slid out on rails for servicing and we all admired the platter and drive head beneath its transparent top cover. The read-write head had an electromagnet "voice coil (So called because it made a pleasant melodious sound when it operated?). The head had an aerofoil cross-section and "flew" just above the surface of the rotating disk.
I found out the reason why a common failure of a hard disk was called a "crash": I was listening to the musical hum as I ran an experiment when there was a faint bang followed by a loud scraping sound. When I turned the DG off and opened up the rack, the inside of the transparent plastic cover was coated with oxide and I could see a large spiral radial gouge mark on the platter. I called in the service engineer who told me that the repair/replacement cost was our problem as it was an "normal" event. I think the cost was about the same as my new car.
In the late 1977's or early 1980's we had a CDC 9790 (I think) disk that had 40 surfaces and read 4 surfaces at a time for a 38 MBit/sec transfer rate. Running diagnostics would cause the entire cabinet to really shake when the seek rate matched the resonate frequency of the case. One day a user came to me with funny images from the machine that looked like tool marks crossing the image at regular intervals. I got an uneasy feeling. She added that there was a funny noise coming from the disk. I immediately powered down the unit. You could see a ring of dust on the enclosure where the head crash had occurred.
For real excitement - has anyone seen an IBM RAMAC 305 disk unit in operation? There was a single head that was moved between the 10 platters via (air) pneumatics. It was about 6 feet tall and made a lot of noise running. The single r/w head would seek horizontally and vertically! Amazing what good mechanical engineering can accomplish.
When I started in the business (1968), we all wore ties and jackets to work. You could recognize the REAL old-time engineers amongst the tea-time rabble because they all wore bow ties. They'd been trained on card sorters, collators and other large, high speed mechanical kit. Wearing a normal tie was dangerous if you serviced those: bend over a running card sorter and if your tie dangled into it, you were instantly part of the machinery.
The ICL engineers at my first place wore chemistry teacher jackets (burn marks, threadbare and leather patches on elbows) , ties (probably clip on) and carpet slippers while they worked.
Comfy feet were more important than industrial safety.
"...they all wore bow ties."
A co-worker used to repair photocopiers for a living, and was required to wear a tie. One day, a machine started up while he was working on it, the end of his tie got caught in a roller, and started pulling him in. He managed to flail around the back of the machine and get the power plug out of the wall. Upon returning to his office, he asked his co-workers how they kept this from happening to them. One yanked his own tie - which promptly popped right off. "Clip-ons" he said.
Indeed! My (recently deceased) father was a lab manager in Glasgow. He managed an PDP 11-45 which did pioneering work in what we now call expert systems in medical diagnosis. IT had removeable disk packs I digress though.
Other parts of his job were preparing the new fangled endoscopes for diagnosis of GI problems. And the other end (he said the Japanese manufacturers included a bite guard with a colonoscope...)
He always wore a clip-on tie - he said that when patients came round from the sedative they could be unpredictabl and try to throttle you with your tie.
There's a bit in the tom bower book about how Mr Fayed wanted to sell House of Fraser. They went to the most respected merchant bank at the time Warburgs to do this. The bloke in charge of the deal from the bank side had to go and meet Mr Fayed at his office. He was then presented with a clip on tie and told to wear it to every meeting. When he asked why he was told that when giving the Egyptian bad news and Mr Fayed didn't like it if he grabbed the tie it would come away and no one got hurt.
Sounds a charming bloke Mr Fayed.
I used to work at PYE TVT in Cambridge. We did not have to wear ties while repairing picture monitors - basically large CRT TVs in metal boxes. Rumour had it that shortly before I arrived, a salesman had turned up and had a look inside one wit its case off. His tie had a gold thread in it, and it touched the 5kV tube anode supply. Loud yell resulted.
As for myself, I was wiring one up - the video cable was 1/2" thick coax with a connector about 1" across on the end, and screwed into the chassis. The other end was connected to a steel frame bolted to the ground, with a huge plaited cable going to grounding rods outside. The mains supply also had a metal connector. On the occasion in question, the connector had been wired up by a colour blind engineer, with the red live wire instead of the green earth connected to the metal cast connector shell. Holding the chassis under my left arm, I grabbed the mains connector - the mains obviously went strait across my chest - and my yell stopped the entire factory! No earth leakage trips in those days! Colour blind electricians ARE a problem.
From roughly 1978 to the present, I've not worn a tie at work. My first work was in engineering and often had to go to the shop floor or take equipment into the field for testing. Ties were a no-no as a) they got in the way and 2) a safety hazard. When I changed career paths to IT, I kept the same philosophy... no tie as it's a safety hazard and the bosses agreed.
From roughly 1978 to the present, I've not worn a tie at work.
A former manager, about me (as FS tech, mind you) not wearing ties: "What do you think a customer prefers, a tech wearing a tie or one without?"
Me: "One who fixes the problem."
And that was the last ever time that matter was discussed.
"Ahh the good old days, when real engineers had beards [...]"
In my mainly UK experience beards were a rarity in the computer industry in the 1960s. Possibly many engineers were ex-military - who usually had had a ban on beards. The programming staff were mostly fresh-faced youngsters - and half of those were women. Many men sprouted beards in the 1970s though.
We had drives like this (I think the ones we had were known as 'tridents' although that may have just been what Xerox rebadged them as).
My wife once had to physically eject a service engineer from the site for *smoking* over one of them with the pack out of it.
And the saga of the scavenge: the Xerox fsck process was called 'scavenging' and it was something you did occasionally, and certainly after a power failure or crash. So, there was one or both of those and the system duly started running a scavenge on a drive (I presume not the drive that had the system on). Hours later it was still running it, and Xerox were rung up: they said it was normal for scavenges to sometimes take a very long time. More hours (in my memory, a day) passed and someone finally had the wit to go down to the machine room. It was immediately apparent that the drive was very unhappy from the noise: it turned out that it had had a head crash, and the scavenge was now blindly scraping the remains of the head assembly over the platters, presumably repeatedly trying to find out where the heads were which I think was done by seeking to some special magic track at the edge/centre of the platter. The drive was full of crud which was once either platter surface or bits of the heads. I was told that they never found some of the heads, as they'd been abraded entirely away.
The drive was full of crud which was once either platter surface or bits of the heads.
A colleague of mine specialised in disk drives was busy rebuilding one of them after a disk crash, when the system manager came in and looked at the brownish deposit on the inside of the disk tub. The answer to "What's this stuff then?" made him go rather pale, and silent.
"Your data"
We had plug-compatible (mostly) 3330 disk drives (100 MB each) on our IBM mainframes. They were the top-loading washing machine format. Our Itel AS/5 (370/158) really swapped a lot due to our cheap manager not buying enough real memory and the swap disk used to vibrate so much that it would walk across the floor occasionally and the CUDI (control-unit device interface) cable would occasionally unplug itself.
Every Saturday I would come in and reseat the cables to reduce the probability of it happening during prime time.
Ah, they joys of working on real hardware!
ISTR working on 80mb SMD drives and they dumped a charged capacitor onto the voicecoil to yank the heads back and off the disk for various faults detected. The carriage and heads shot back at a Hell of a speed.One of the labs were I did a bit of training still had an errant 6" bar magnet, as used with a coil for speed feedback ) buried in the breeze block wall behind the drive and was used as a warning not to stand directly behind the drive when first powering up after any repairs in the carriage area.
Yes I remember the 80MB and later 300MB SMD drives (made by CDC as I recall ), and being very wary of the emergency retract, especially when replacing a head and doing the head alignment - watching the 'cats eye' on the scope with your hand on the tool ( like an elongated wind up tin opener on cans of sardines ) currently attached to the head carrier , you touched that tool very gingerly indeed !
Is the size of (each) of the disks in a pack. Until well into the 1980s the standard physical size of one of the disks (in a stack) was 14" and about 3/16" thick and made of reasonably high grade aluminium. And they were not light. The capacity of disk changed, but the size didn't until sealed "Winchester" disks became a "thing" that would lead to the small drives that we have today.
Imagine one (or even a salami slicing set of them) flying at anything (fleshy or hard). Makes Oddjob and his bowler hat look like a ineffective amateur.
Less common than disks were drums - metal lumps about the size of washing machine drums coated with magnetic oxide and an array of fixed heads - one per track. They were faster than disks because there was no seek time.
Some of these could weigh 60lbs and rotate at 3000rpm. A story that I heard (but could not verify) was that on one of these a bearing failed and the drum departed from the mechanism at speed and left a trail of destruction before it finally came to rest.
When I visited the machine room in Claremont Tower at Newcastle University in 1978 or '79, they had a drum acting as the swap space on the IBM System/360 Model 65.
What I remember is that the side was replaced by a perspex panel, and you could see the multiple fixed heads arranged around the spinning drum, so there was no seek component of the access time, merely the rotation time of the drum.
At Durham, we had a PDP11/34e with RK05 drives in a DEC 19" rack , and when the drives were busy, the whole rack rocked forward and backward quite violently as the voice-coils moved.
Later, I looked after a system with 80MB SMD drives. The worst that we had happen was the platter brakes seizing, making one hell of a racket, and a minor head-crash. We did have one pack that had the bottom guard platter bent making it a little unbalanced, which used to sing, but we only used that to hold an infrequently updated system backup.
> When I visited the machine room in Claremont Tower at Newcastle University in 1978 or '79, they had a drum acting as the swap space on the IBM System/360 Model 65.
I did my MSc Computing course using this machine in 1972/3 - it was actually a 360/67 (a 65 with the IBM 2301 drum) running the Michigan Terminal System (MTS), a quite decent time sharing system of the time, and only superseded later by VM. [I assume that it lasted until 1978/9?!]
They were still using MTS when I started there in '89 - just for the 1st years, I seem to remember the machine was an Amdahl, no idea what model it was though (I suspect we night have been told but I don't have any notes from back then any more). I also seem to remember they had a cluster of *nix machines that the 2nd and 3rd years were allowed to play with instead of being hobbled by the old system.
I also remember an old HP analyser system when I started at RR with a fixed/removable platter on it - I only saw the cover off it once when it was being serviced and was told in no uncertain terms never to get your watch or anything that might be magnetic anywhere near the voice coil, even when it was off. Didn't need telling twice on that one.
The analyser was 'donated' to a college (who also had another virtually identical one) when we were granted funding to purchase one that could be moved by one person and was considerably more powerful (it even used 3 1/2" Floppies!!!)
When i first started with a certain employer years ago they had Unisys 494's running in 490 mode with Fastrand drums as well as 432 and 1782 drums. The Fastrand were two huge drums with a controller cabinet and in total weighed in around 6000 pounds. The floor tiles buckled a bit when they got moved, but i don't recall any collapses. BIggest danger was always getting a tie caught in a printer (or in my case at one place, getting a hand caught in a decollater).
Fastrand drums were actually made by a sewer pipe manufacturer. They were the only people who could make a drum of the size true enough to spin up to the high speeds needed.
The cabinet for them included mount points for a hydraulic crane that would be required if you ever needed to remove the drum.
The Fastrand standards were still echoing through OS2200 15 years ago in the various options available.
I think my favorite drum device was the ICL one that used 255mag cards in a hopper and a drum transport. The cards would drop and secure to the drum, then you could read and write to them, and they would eject back to the hopper.
Of course, as time went on the mechanism would loosen up and the cards wear and then would come the day two or three cards wold drop at once. The operators said it sounded like when you were a kid and used clothes pegs and cigarette cards (ask yer grandad) to make your bike sound like a motorbike. Often wished I'd seen that one up close, but I came in ten years too late.
I've seen manuals for exchangeable drums that were conical - you screwed them into the drive small end first, pushed the go button and they would be lowered until the heads would fly. Never seen it in the wild though. I think they were ICL equipment, but can't be certain. Working from memory here.
"Fastrand drums were actually made by a sewer pipe manufacturer."
Ah, probably centrifugal casting. Fascinating process but I imagine you wouldn't want to be anywhere near it when it went wrong. (While hot molten iron + heavy moulds rotating at high speed = what could possibly go wrong.)
Workplace environment a tad different from my bijoux office: https://www.youtube.com/watch?v=113cm_U2EBk
Here's one for the weekend, and thanks for sharing your war stories.
Some of these could weigh 60lbs and rotate at 3000rpm. A story that I heard (but could not verify) was that on one of these a bearing failed and the drum departed from the mechanism at speed and left a trail of destruction before it finally came to rest.
This happened with an US reconnaissance plane during flight. The equipment operators reportedly took refuge in the rear compartment, hoping that the bulkhead would hold.
The drum trashed most of the equipment before it came to rest. And it was probably a good thing that it didn't exit the airframe.
I used a DG Nova 3 when I started at GEC in 1979. Usually a nice, quiet system although taller than me and wider than I was at the time, perhaps not now though. Even when an engineer was checking out the disk it stayed relatively quiet but with the cabinet open and the disk slid out for access, the movement of the heads caused the machine to shake from side to side.
I worked for a DEC house in the late 80s. We had some of these in a room in the middle of the office. We came in one Monday to find that the bearings had gone on one of the disk packs and the platters had gone through the wall and across the office like they were the frisbee versions of angle grinder discs. If it hadn't happened over a weekend there would have been blood on the floor. They put sheet steel round the walls after that.
I only used the magnet stored blade for when we were reclaiming tape reels. So much easier to just slice through worn tape on a reel rather than unspooling it. The first time I did it I unspooled about an hours worth of tape and filled the small office bin and then some. A lot more fun to do it that way but far messier.
Otherwise I had a bit of bluetack and a oil paper holder (to prevent rust) stuck to my the other side of the drawer tray.
"Yes, that Data General, the one that EMC acquired for US$1.1bn in 1999 so it could make hay with the CLARiiON iSCSI SANs"
The original CLARiiON disk arrays were SCSI, not iSCSI, and had migrated to Fibre Channel by the time EMC acquired them. iSCSI came much later, I think with the AX series.
[Source: I had a bunch of 1st generation EMC DAEs in a 42U cabinet in my kitchen, until a few years ago when my long-suffering partner realised how much electricity they were drinking.]
I used to work on removable pack disk drives, once spent a long time replacing the heads in a drive aligning it etc. Looked at my watch and wondered why it hadn't taken me much time, working close to the magnet on the head mechanism was stopping the watch. Bad practice to wear a metal strapped watch anyway.
I've seen disk drives walking across the floor, and the sound of a head crash wasn't very nice.
The Burroughs On-Line Disk File (as it was known) had massive 36" platters spinning in the vertical plane. My B5500 field engineer said that the advantage of this design was that if an earthquake occurred, and the platters shattered, the probability of injuring people in the machine room was much lower. You can see the product brochure at http://s3data.computerhistory.org/brochures/burroughs.onlinedisk.1962.102646217.pdf
The Illiac IV computer, also built by Burroughs, had disk drives with a similar vertical platter design. Pictures at http://www.computerhistory.org/collections/catalog/X1629.99
"You can see the product brochure at http://s3data.computerhistory.org/brochures"
I love it: "... enables this small disk to store almost two and a half million characters ..."
A disc with a diameter of almost a metre!
Thanks for that link - perfect for a Friday!
Not entering the world of work until 97, and then IT sales in 2000 for Sema, I missed all of these contraptions (not that you'd want salesmen anywhere near any of it anyway). Would periodically trip over water-cooling pipes for the ICL mainframes, but by 2000 much of the stuff you chaps described had been consigned to the skip.
Nonetheless, my cautionary tale. Before turning to the dark side I trained as an aircraft tech. As was the way in our college, a small group of us had an evening project to tinker with; in our case an old water cooled inline four pot engine off something or other. We'd loving it stripped it, blueprinted it, and put it back together.
One Thursday evening came the big moment - the first run up. We'd modified a cradle for it, plumbed in the cooling, etc. Before firing it up, we nipped out to get MacDonalds. The moment arrived, and half a dozen spotty youths stand around a pristine engine in the middle of a maintenance hangar, munching down a variety of MSG-based 'food'.
Fire it up, and there's immediately a problem - the wiring loom smoulders, then melts, then ignites. Someone has mucked up the ignition wiring. Some has mucked up a bit more as the short seems to have caused a permanent live ignition and we can't kill the engine. Finally, someone has mucked up completely and moved the bloody fire extinguisher.
Necessity is the mother of the invention; inspired move by one of my peers reveals that a well chucked MacDonalds extra large strawberry milkshake is an excellent fire extinguisher, and also suitable for choking an engine into stalling out too.
However, it should also be noted that upon striking a cooling fan spinning at approx. 900rpm, the same milkshake also exhibits simply unmatched 'fling' characteristics, coupled with remarkable 'stick' qualities.
It was amazing how much aircraft hangar internals you can coat with just 750ml of gloopy sugary milk and a 900rpm fan.
During training I had to repair the PCB's of the 4 computers at work, 1978-80 room filling machines based on discrete transistors. These machines had magnetic 'drum' stores of a few hundred Kb, wire video recorders were bad enough to be operated remotely in securely locked rooms, flailing wire had shredded recording engineers, but the drum stores with a row of heads that were driven up and down the drum were considered safe. Until one day the top bearing of the head translation shaft failed and the heads were jammed against the spinning cast iron cored drum the size of a large dustbin, which failed explosively, the shrapnel passed through both computers in the room, smashing lots of boards on the way through, thankfully the only other two, bought as scrap from Farnborough, enabled them to be repaired. The concrete walls bear the evidence of the impact to this day. From that day on no-one was allowed in the room if the machines were on, when the first 5 Mb 'Winchester' drives arrived it was a revolution!
Well since I have spent most of my career working on CNC equipment lathes, machining centers mostly. There are quite a few stories one time we were troubleshooting and intermittent spindle problem on a 30 hp lathe. Well it stopped an there was nothing to do but go in the cabinet with lots of kit so you could not move easily and a multi v belt pulley (the better to cut fingers off). Found the problem shortly a micro switch was out of adjustment. All of a sudden the switch clicked it and WHOOSH motor starts at full speed. I scramble out as the operator is doing emergency stop. We both look at each other and he asks "Are you ok?" I respond with "Yes I think so" as I'm counting fingers. Thankfully things are a lot more safe these days. Lost a few friends from the Good Old Days.
"Well it stopped an there was nothing to do but go in the cabinet with lots of kit so you could not move easily and a multi v belt pulley"
Power off and lock the breaker. You can even get 'em for MCBs
https://uk.rs-online.com/web/c/safety-security-esd-control-clean-room/keys-safes-locks-lockouts/lockouts/
Two stories...
1: Overkill
Working in a machine room with a PDP11 and a MV8000. The dog had 2 x 250MB top load style removable packs and a 500MB? Winchester for the OS (AOS/VS?). The removable packs were user data. One fine spring day a strong smell not unlike burning insulation pervades the machine room and the dogs console barfing up errors; Disk errors. Shut down the machine and call an engineer. Turns out one of the packs had head-crashed. Engineers dismantles the entire thing including the very large permanent magnet now sitting on the machine room floor. Then starts on the second drive. apparently the dust is so fine that there was a good chance that the other drive would have sucked some up.
Now he placed the two head magnet assemblies at either end of the machine room stopped by a tile lifter each. I thought this was overkill until a little while later there was a very odd noise from the machine room. A low pitched scraping noise that very quickly rose in pitch followed by a very large ***WANG***
One of the tile lifters had come unstuck and the two magnets were now fused together permanently.
2: ROTM
Got a ticket from the hell desk. A secretary had complained that her PC was trying to kill her. A bunch of us decided to visit said homicidal PC. We asked the secretary to demonstrate the fault. She refused, handed us a cdrom disk and fled the room. The PC was one of those mid tower machines sitting on her desk. Head height for her and about neck height for me.
Upon loading said disk into the drive. running though some quick checks all seemed well; Until I ejected the cdrom. The tray opened and at first I didn't notice anything wrong. It was at that point the disk, still pinning at a high rate of knots did a very neat VTOL takeoff from the now open tray, wobbled a bit, nicked the PC case and promptly zipped off past my neck I felt the draft as it passed. It then ricochet around the cubical a few times (still at neck height) until it came to rest.
Replaced the drive, but the secretary in question never quite trusted that machine gain. Next time I saw that machine the cdrom drive was facing away from her, towards a nearby wall :)
If you have worked with this stuff (especially punches) that are high speed, they had lots of pent-up energy. You took off ALL your rings, and ties if you ever worked with this stuff. Most were belt driven which increased the likelihood of getting stuff caught in the "works".
Be careful out there!
Many years ago, I worked for the local power company as a distribution engineer. At one point, we had to take a local substation out of service for maintenance. In order to feed local circuits, we ordered a truck-mounted temporary substation to be connected to the circuits. I decided to stop by and watch as the temporary station was switched on line.
The truck was delivered and the techs dutifully checked jumpers so that the transformer was properly set to step down 66 kV to 12.5 kV. Unfortunately, the input to the transformer was protected by stacks of three 25 kV surge arresters, connected in series. The previous use of this portable unit was to step 12.5 kV down to 4 kV. And part of that installation had been to bypass two of the three arresters on each phase. Leaving only one rated at 25 kV.
The lineman who was assigned the duty of closing the primary side switch opened a panel on the side of the truck, which swung upwards. He was a pretty tall guy and had to duck down under the opened panel to reach the switch handle. And he made certain that his displeasure with this setup was heard. Until he threw the switch.
A fireball erupted out the top of the temporary sub as the arresters failed. Surge aresters are designed to bypass relatively brief lightning strikes with lower levels of total energy. Not to bypass a continuous 66 kV feed. The guy who threw the switch was (fortunately) protected by that nuisance of a steel panel over the top of his head. The fault was cleared a few seconds later by a breaker in the transmission yard feeding that circuit some miles away.
After the power had evidently been cut, the crew reversed their hasty retreat and slowly moved toward the substation with the idea of putting out the fires. Until the foreman yelled, "Wait for the reclose!"
Transmission line circuit breakers are often programmed to attempt to re-energize a line in the event that the fault was a tree branch bumping the line. But in this case, we were treated to a second fireball and more porcelain fragments.
55 Years ago I worked between school and University in an engineer's design office and was told to buy my own scale rule and a holder for razor blades. The holder grips the blade down one side and has a hinged cover to secure the other working side of the blade when stored. I still have it. I was never happy with wrapping draughting tape down one side and keeping it tucked into a slit on the backing sheet.
And since people are going down memory lane.....
Razor blades were used to remove ink from the tracing paper on ink drawings.. The characteristic sound from the draughting section was the scrape of the razor blades as drawings were updated.After scraping with the razor blade, the paper would be rubbed with a hard fibre rubber to restore the paper surface which would otherwise have soaked up the ink like blotting paper. It was also possible to use a mains powered eraser instead but had the risk that it would lift the paper and twist it into a crumpled mess in less than a second.
Nowadays when drawings are created in CAD and even BIM, razor blades are deprecated as they permanently damage the computer screens :-)
I once worked on a computer impact printer that printed on green bar paper 132 columns wide. The printing mechanism was a bank of 132 electrically actuated hammers. The hammers impacted a the paper against an inked ribbon onto the printer font. The font characters were cast on a bar about 40 inches long. This bar rode in bearings linked to a powerful electric motor that made the bar reciprocate back and forth very rapidly so the printer could print hundreds of lines a minute. There was an access door on the side of the printer so the bar could be removed for maintenance. Occasionally the link would break and if the timing was right it would strike the access door so hard it would fly open. The story was going around that someone left the access door open, the link broke and the print bar embedded itself in the side of the mainframe computer nearby.
Stories of 14 inch spinning aluminium discs careering across computer centres were common back in the day. Usually seized bearings shearing off shafts and other things. I've heard them often enough to think they are largely apocryphal, but who knowns.
The exchangeable packs I was familiar with rotated at 3,200 rpm. Most certainly they packed a lot of rotating kinetic energy when spun up.
Only 50Mb? All mine, bought in 1980, were 700Mb - about the same as a Data CD would be a few years later.
On top of a 2 horsepower spin-dryer.
I'm astonished at 'shards' - what on earth were the shattered platters made from? I'm remembering thick aluminium plate!
My staff did did a nice line in Gift Clocks.
Whenever a head-crash rendered a disc-pack useless, we'd dismantle it and fit each artistically-scarred brown disc with a cheap quartz movement. Wish I'd kept one :-(
Nostalgia isn't what it used to be.
David
I had something similar in the mid 80's with reversed wires though much less dangerous, i was on site to assist to shut down and then restart a system should there be problems, after the electricity board had done some work outside which necessitated shutting off all electricity to the building.
When power came back on, the machine looked good, lights came on, but it wouldn't boot from disk (similar type of 50Mb multi platter disks) even though it was spinning.
After some investigation it seemed the electricity board had swapped two of the three phases outside the building somewhere since the disks were spinning backwards ! Fix was to reverse the wiring inside the disk cabinets !
Back in the 1970's our computer room became a dangerous place. People (including myself) were tripping on the tiles that covered the raised floor. Nothing like this has ever happened before, and we didn't have any idea why. New/old it didn't matter. One Sunday morning a fellow sysprog was called by the VP of the data center to get his camera down to the DC and take pictures. When the sysprog got down to the data center, he found a couple of tiles ripped off of the floor and thrown all over the place. One, unfortunately, hit a few switch's for disk drives. It's been a long time, so I don't remember the brand, but it was a plug compatible switch for IBM's 2914. He took the pictures (back then the data center was closed on Sunday's). The next day IBM showed up in the VP's office and pleaded guilty to doing the damages. His apology included the fact that he kept tripping on the floor because of the tiles. There was not any major damage done, but I think I heard the number of 750. USD. The VP went through the roof when he heard this has been going on for weeks. He got the head of the Data Center up in his office and demanded why this wasn't this brought to his attention and what was causing the problem? The DC manager didn't know either. That afternoon one of the operators noticed that the cleaning clue was using a dry mop to clean the computer room floor and he had the idea that might be the cause of the issue. The VP called the VP that was responsible for the cleaning of the computer room floor. The people who manufactured the cleaner were brought into the fray, and they denied responsibility saying it was used nationwide and no one else has had the issue. The DC manager was not taking no for an answer as he didn't want to get into any more trouble, so he asked the provider to run tests as it was the only possible candidate. Meanwhile, the cleaning people were ordered to change to the previous cleaner. AÏter 2 weeks the manufacturer came back and said "oops" it was the cleaner. The VP said to replace all the flooring or else you will have a lawsuit. The manufacturer didn't have much choice and if I remember the cost turned to be a lot of money if memory serves me it was 100,000 USD (I could be wrong). We got all new flooring (not that it was old, to begin with). The manufacturer also paid for cleaning supplies for two years. I don't know if there was a recall or not and if it happened at any other DC but we were happy, IBM was happy.
...in 1980 I worked at a bank with a computer operator on a Burroughs 3800 who managed to crash every production disk pack in the shop. We used a dozen of the Burroughs type 235 (B 9484-8) packs each with 10 platters, 2 packs per cabinet, ~300 MB capacity per cabinet.
Our B3800 had failed, massive power supply failures i think, and we relocated to the regional support center (for us it was Atlanta, GA) for disaster recovery.
Said operator mounted all the production packs (having just come off an unheated aircraft cargo hold) into the climate-controlled drives of the DR center..... without allowing time for them to acclimate/warm up. And hit the LOAD buttons.
Within a few seconds every pack crashed. How he didn't lose his job is still a mystery. Fortunately, we had backup tapes as well. And after ALLOWING THEM TO ACCLIMATE, we were able to reload from tape onto new packs ($$$) from the DR center and proceed with the recovery. Burroughs sold a dozen new disk packs that weekend.
A crazy sleepless expensive weekend......
I remember this exact kind of story if somebody inserted a damaged CD on a CD drive.
Nobody in their sane mind would try to read a CD if the core plastic (that shapes the central hole) was damaged.
CDs rotate at over 500 rpm, but at a fraction of the energy.
However, CD readers are covered with steel boxes for a reason.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
