Drama
Wa-ter drama!
The weekend is approaching, dear readers, but before we get there, our weekly column of tech support drama beckons. And this week's On Call is certainly dramatic as "Brandon" tells us about a tense phone call with a user who was, quite literally, in at the deep end. "It was a dark and stormy night," Brandon began, with just a …
Obviously a stock photo, but I wonder where from? The sockets are British standard style, but the switches are certainly not common here, and the plug is completely the wrong shape, wouldn't accomodate the fuse etc.
The switch shape is similar to Australia, though the sockets are wrong. Perhaps Singapore?
I suspect that UK BS1363 sockets may have been adopted but not ring circuits. If you were using UK sockets on normal radial circuits, the fuse wouldn't be any more necessary than it is on Australian, Continental European or North American plugs.
There's a strong argument against using ring circuits, but the UK plug and socket system is pretty good. In Irish residential wiring for example, rings are permitted but most wiring is like continental Europe, with 16 or 20 amp radials feeding sockets. Although, like the UK plug top fuses are mandate if your using those plugs.
It's possible that perhaps Malaysia simply didn't adopt rings at all and thus allows plugs that fit those sockets, but without fuses. Just because they're BS style sockets doesn't mean the whole standard was adopted.
The only thing that makes those fuses necessary is that UK wiring has 32 amp rings connected to those sockets. The fuse is necessary to protect the flex in a fault.
@Six - "I suspect that UK BS1363 sockets may have been adopted but not ring circuits."
HK uses ring circuits. The "intended for use on radials" doesn't make sense: no sane wiring authority would adopt pin sizes that matched BS1363, but chang the plug and cable sizes. The body of the plug looks too small, so wouldn't meet the requirements for preventing small fingers underneath, and the flex looks smaller than 13A rated. Why have 13A-rated pins in a lower-rated circuit standard?
You'd be surprised! Standards fork.
Plenty of countries adopted bits of German DIN standard sockets but not the whole system.
In the Irish context the BS1363 plugs and sockets are absolutely enforced as per the UK, but codified in law as IS 401 (plug) and IS411 (socket). The wiring isn't the same though. For example rings are not recommended for use in kitchens for heavy appliances. There's a concern that having all of the load at one point on a ring creates a potential fire hazard. A ring makes a lot of assumptions about loads being relatively evenly spread and if you concentrate all the loads in the kitchen, which might be close to one end of the ring, it can run a tad warm.
The usual topology here tends to be a 20 amp radial to ever major room, with some shared between smaller rooms / corridors and several radials to the kitchen. They to mad on kitchens - we've about 10 double sockets and most rooms have at least 2 or 3 doubles. There's certainly no skimping on them anyway and this house was built back in 1979.
Rings are allowed but electricians tended to remain fairly conservative and stuck with using more continental style radials.
I recently had an extra double socket installed for a washing machine and dryer moved to a laundry closet we decided to put into a hall and the electrician ran a dedicated 20amp circuit in an RCBO serving one socket only rather than connect to any existing radial, as apparently it's considered safer practice for high load appliances.
Back before the 1960s Ireland used to use Schuko (German style 16amp plugs and sockets). They remain standardised as IS180 with a reference to the most up-to-date CEE 7 standard sheets but they're not normally installed and appliances are required to be sold with
BS1363 plugs.
The only place you'll find Schuko here is is as an extra socket or two in a lot of hotels rooms. The reason for this is a lot of hotels have issues with serious damage to sockets caused by people jamming in 16amp European plugs (with fatter pins than the 2.5amp flat plugs found on mobile chargers etc) and wrecking the contacts on the BS style sockets leaving them loose / dangerous. So it just makes more sense to provide one or two Schuko sockets. The only difference is that you can't legally install sockets without shutters, so they'll be shuttered versions of German sockets.
A lot of our underlying wiring systems though are more like Northern European continental systems - radials and old systems used Diazed / Neozed bottle shaped cartridge fuses defined in DIN standards. We still use a single Neozed combined fuse switch as the main over-current protection on every consumer unit.
There was also much earlier adoption of RCDs on sockets, water heaters and certain fixed as mandatory here in the late 1970s. Although the UK went from not requiring them at all to having them on every circuit, leap frogging us. Irish regs still continued to exempt indoor lighting circuits (other than bathrooms). Thats likely to change this year though.
So even though the plugs and sockets are BS1316 the wiring standards behind them are not identical to UK regs.
"A ring makes a lot of assumptions about loads being relatively evenly spread and if you concentrate all the loads in the kitchen, which might be close to one end of the ring, it can run a tad warm."
The ring has an end? I thought the point of a ring was that it was a complete loop, ending where it started. (North American here, no UK electrical standard knowledge)
A ring circuit is where you take two cables leading out of the fuebox and connect them in the middle, so:
fusebox ---------- appliance ----- appliance ---- appliance ----- back to fusebox
The point being that it makes 2.5mm wire ( standard for sockets ) go much further. It's easier to handle than 6mm and costs less.
It was introduced after the war due to copper shortages.
Further to that, generally in the UK only the sockets circuit is a ring. Usually either one ring per floor or one ring per house. Mine is wired as one ring per house although I'm thinking about splitting off the upstairs onto its own ring at some point.
Everything else is generally radial - lighting, electric hobs, ovens, etc.
They get sold online from China
I'm trying to find these totally illegal things you can plug into a UK socket to make it usable for the two pin Euro plug. I can, of course, just use a screwdriver to push down the earth pin to unlock the socket (leaves the child safety operational which is likely what made those plastic things illegal), but I can take one of these plastic things through airport security whereas a screwdriver is now seen as a murderous tool that is far more dangerous than the duty free whiskey bottle that I only have to smash to give me a lethally sharp weapon..
Ever tried to smash a whisky bottle?
First you have to find somewhere on the plane hard enough to offer the promise of broken glass and sturdy enough not to offer the chance of broken aeroplane fixtures.
Then the bottle will need a few good whacks to break, which it will, usually uselessly short at the neck.
So now you have a very short Glasgow Pigsticker and the sky marshal* has had plenty of time to get into position for a decent shot.
If your whisky is of the hi-test variety, you might be better off using it to set fire to the seats as a distraction while you retrieve the knife you had an accomplice on the cleaning crew hide on the plane for you.
But your point about screwdrivers is well-taken.
* - Sky Marshals are like Space Force, but real.
"First you have to find somewhere on the plane hard enough to offer the promise of broken glass and sturdy enough not to offer the chance of broken aeroplane fixtures."
How about the head of that mewling quim sitting next to you and spoiling your flight with that mewling?
And yes, setting fire to it seems logical, but doesn't work as well in airplane conditions (reduced oxygen).
Then the bottle will need a few good whacks to break, which it will, usually uselessly short at the neck.
That's why you should twist the bottle as you hit it against the sharp edge. Didn't they teach you that at school?
I'm not saying I went to a rough school, but the teachers used to break out the riot gear whenever they planned a raid on the smokers in the loos. And one chemistry lesson came to an abrupt end when the Bomb Squad rolled in.
"And one chemistry lesson came to an abrupt end when the Bomb Squad rolled in"
Bah, you've never had the Fire Brigade turn up (unannounced) in hazard gear to collect a Strontium 90 ** sample, have you ?
(** EDIT: Someone told them it was a -90 sample, but it was actually a lump of the naturally-occurring stuff that ISN'T terribly radioactive)
"in hazard gear to collect a Strontium 90 ** sample,"
No, we had people in white hazmat suits turn up to collect spilt mercury from a lab thermometer (1990) which was ridiculous as just a few years before (1986) we could stick our fingers in the stuff and ping it around the desk. Mercury is weird.
My school had a chemistry teacher who used to play a trick on next year's intake, visiting from their primary schools. He would place a beaker of concentrated ammonia and a beaker of concentrated hydrochloric acid in each of a pair of old boots; place the boots next to each other behind the front desk, resulting in a nice plume of "smoke" where the two invisible vapours reacted; and strew some broken glass around. Thus creating an illusion of somebody having blown themselves to bits, leaving only the canonical pair of smouldering boots. Then he would hide in the prep room, and watch the looks on the little darlings' faces .....
He would place a beaker of concentrated ammonia and a beaker of concentrated hydrochloric acid in each of a pair of old boots; place the boots next to each other behind the front desk, resulting in a nice plume of "smoke" where the two invisible vapours reacted
Love it. By the way, the fun bit about that smoke is that it is actually edible if you crystallise it out.
I had 2 in my school end up in chokey on attempted murder charges about a week after they were asked to leave.
Prior to that one of them was in my chemistry class and revelled in causing disruption and damage.
My German class (mixed ability teaching putting the more studious kids in with hellians - funnily enough the kids from the "better" postcodes were exempted from this social experiment) went through 5 teachers in 2 years, one had a nervous breakdown due to some of the kids in the class, I lost interest in learning languages....shame as I was actually quite good until I got sick of the disruption and hassle for actually doing the work......
Your chemistry classes sound positively educational.
We only had the soft thud of bodies hitting the floor after inhaling acetylene. It was a gas so it must get you high right.
Note: given those involved, any reduced brain function due to a lack of oxygen was likely a pre-existing condition...
Our fun where I almost crapped my pants was when the stupid guy that was held back 2 grades decided to instead of breaking off a piece dropped the baseball hunk of potassium into the beaker of water. The explosion was really load but the smoke quickly cleared since a couple of plywood sized glass windows were not there anymore but in small pieces in the parking lot outside the lab. That will wake you up in a boring lab class real quick.
Next time you fly take a look a the side of your seat. Extending under all the applied soft furnishings are sturdy, strutted metal legs. Wishing a suitable surface to break a bottle in a plane (not that I would ever do such a thing) I would use that.
The thing about the bottles is not so much their lethality as sharp objects but their utility as blunt force clubs. If the bottle should break to sharp lethality on someone's head that would be all to the good to a miscreant would it not?
What everyone has overlooked is that it's unlikely you'll get that bottle into the passenger cabin of the aircraft and in the US the amount of fluid not the alcohol itself will have them confiscate or make you put it in your stowed luggage. So this is an unrealistic scenario.
The adaptors that you can buy on the High Street for a couple of quid don't usually have the earth contacts at the edges for Schuko plugs. They will accept the line and neutral pins of the plug, but leave the appliance dangerously un-earthed.
(Earthing might be less important if you have an RCD-protected supply, but I still trust a wire fuse more than I trust a chunk of complicated electronics.)
Argos' 3-for-a-tenner travel adapters are earthed, not that it'll help in this situation.
An unearthed Euro plug (CEE 7/16 or 7/17) will fit an earthed CEE 7/3 (Schuko ) or CEE 7/5 (French) socket, as they should, and there won't be any problem fitting them in an earthed BS1363-to-Euro adapter.
"(Earthing might be less important if you have an RCD-protected supply, but I still trust a wire fuse more than I trust a chunk of complicated electronics.)"
They actually do different things. An RCD will trip when there's faults which will leave a fuse unblown. Conversely, a earthing/grounding, will protect against conditions an RCD doesn't detect. (I'm not at all happy with my TV (quite new) which feels live whenever it is turned on at the wall. It's not going to injure anyone, but I do wish the metal on the surround was grounded.)
The idea of an earthed metal enclosure is, if a live wire comes adrift inside the appliance and then touches the metal case, the fuse will blow.
If you insert a Schuko (earth connection by means of contacts at the edge of the plug) or French / Czech (earth connection by means of a pin protruding from the socket) plug into a non-earthed socket, the metal case of the appliance will not be earthed. Now if a live wire comes adrift inside the appliance and touches the metal case, it will become live.
An RCD measures the difference between current flowing out the line and returning up the neutral -- which should be equal. If they are different, either Kirchoff was mistaken (which is actually not impossible, since verifying either of Kirchoff's Laws or Ohm's Law requires the use of test instruments which rely for their operation on those laws being true in the first place) or some electrons have unilaterally decided to take a different route back to the supply, possibly via you. So the RCD disconnects the supply.
Either a fuse or an RCD will offer protection against a live wire touching an earthed metal housing. (Other faults are available.)
(Earthing might be less important if you have an RCD-protected supply, but I still trust a wire fuse more than I trust a chunk of complicated electronics.)
An RCD breaker is not at all a bunch of complex electronics. It's a set of coils, one per phase and one for neutral, arranged on a common core so that the fields induced by the passing currents cancel out if there's no residual leakage. If a ground fault happens, the fields don't cancel any more and the resulting magnetism pulls away a latch, releasing a spring that pushes open the breaker contacts.
An electrician friend of mine pulled out a piece of test equipment to show me how bad the cheap and nasty RCDs I'd purchased were.
A moment later he was telling me he was very impressed with both the low response current of the RCD (well below what it takes to blow a fuse, but also well below what it takes to kill you - you can be smouldering ashes with a fuse not even reaching 1% of it's required failure current) and with the quickness of it tripping. You almost won't notice you've zapped yourself but the power will be safely shut off.
Fuses are relatively slow to respond and have a very high trip current verses what it takes to kill you (at 240V).
Understand about prefering simple vs complex, but sometimes the complex is the better option :)
Just take a UK shaver adaptor. They're not classified as lethal weapons, cheap, and also have an actual fuse in them. Altho they're rated for 1amp max, the fuseholder will handle a 2.5 amp 20mm HRC fuse no problem , and in common with most BS1363 kit, will handle several times its rated capacity continuously.
2.5amp being the rated capacity of a europlug.
Unless you mean Shucko or French/Czech standard plugs, which are usually earthed. Adaptors are available for them too :)
Thanks for the tip about searching for images displayed in Chrome.
I note ElReg's article is on the first google page! ElReg must be crawled at least hourly.
On more than one occasion I've googled a subject mention in an ElReg article, and the ElReg article itself was top of the results!
See? That's why I fit right in with El Reg's readership.
Forget the possibility of a fried user: what's the provenance of the sockets in the pic?
Hahahaha - thank God I learn from experience and had put my coffee down before reading the comments.
/wipes tears from eyes laughing.
Back in the 90's I used to support a theatre. Among other events, snooker was hosted there.
This was back in the day of Netware and coax :)
The "IT Manager" - i.e. the lady that had been elected to look after everything was a nice person and had picked up, quite well, the usual checks to do before phoning in, such as making sure all the BNC adaptors were connected, and that the terminating resistor was present.
So when she did call, you tended to know it wasn't going to be anything immediately obvious or basic.
She called in and it was my shift on the hell desk, so I took the call.
Pleasantries out of the way, and a bit of troubleshooting later, I asked her to tell me what was on the screen on the server. Since it was in the basement, it meant a small trek, so she hung up and said she'd call back when she'd noted it down.
A few minutes later she called back to say "Don't worry. Found the problem - the basement has flooded and the server is off...."
Ok then...
Once had a user's pre-unibody Macbook fall out of their bag and into the pond in front of our building. Took the facilities guy 30 minutes to fish it out. We put the now affectionately named "Deep Six" in the boiler room and promptly forgot about it. A month later, we remembered Deep Six and retrieved it from the warm and dry boiler room, and the damn thing powered up! The display had some foggy patches but it worked.
Explosion as it is the warmest icon ;-}
Apple MacBooks of unibody construction are sturdier than they look.
I once was asked to check one that was left open (i.e. “standard usage” open) overnight on a reception desk.
Where’s the catch you may ask ?
Well, above the desk was a rather large painting enclosed by a solid frame.
Overnight the hanging mechanism gave way and the MacBook cushioned the work of gravity.
Diagnosis ?
Other than a couple of minor cosmetic items (a rather large dent at the top of the metal bezel on top of the monitor, and a bit of a scratch to the metalwork around the keyboard) the MacBook was operating perfectly, no sound of rattling.
"Just cleaning an unibody with a wet bit of paper towel will damage the keyboard. There is no protection to moisture at all in Macbook Pro's."
Really? That's absolutely appalling design, especially considering the price of, supposedly quality, Mac gear.
Sadly, this wouldn't surprise me at all for the Apple of today. :-(
There is no protection to moisture at all in Macbook Pro's.
Pretty sure that's the same for Dell, HP, Toshiba and any other brands out there. Have seen one "Excell" laptop (no one imported them into NZ so rare here) that had great protection between the keyboard recess and the rest of the unit, but aside from that one the hundreds (thousands?) of laptops I've had open have not been even remotely spill resistant, though some have survived better than others.
Also not seen any with even a token coating to help with preventing corrosion, even seen many that rust in high humid environments.
Don't like Apple, but their lack of spill resistance isn't unique to them - pretty sure they copied that from every one else as well! :)
I can beat that!
I had a call to help the friend of a client.
He had been knocked off his motorbike on the A40 by a hit and run and left for dead!
He suffered awful injuries and when I met him he was still in a bit of a state but back home from hospital.
The reason for my attendance was his MacBook Pro. It was in his backpack at the time and by the time it was recovered had been run over several times.
He had been on his way to pitch for a design job and his pitch was in the Mac. Unfortunately the Mac was bent into the shape of a taco. Seams were split apart and one the hinges was bust. It was still closed and in one piece but completely out of shape.
My task was to extract (if possible) the HD and if successful, try to extract his pitch and the other work on it.
I prised the lid open and as I forced it into its very unnatural new hyperbolic shape with its one hinge, some key-tops fell out and the shattered screen lit up and seconds later his desktop appeared. It had been asleep for over a month in its fortune cookie state.
I don't know which one of us was more surprised. Looking at the Mac I wasn't st all confident that the HD would be intact.
I only wish I had had a phone with a camera at the time so I could document it.
(The delay didn't stop him getting the gig though)
Colleague at a conference fell into the harbour in Amsterdam one evening. His macbook was in his backpack, so of course got well-soaked in the brine.
Undeterred, he carefully cleaned and dried it, whereafter it worked just fine. Except that the backlight to the screen was dead. If he ever got that fixed I never heard about it, but he still had all his data, and a machine that worked fine if accessed over the 'net or plugged into an external monitor. Or even on its own under the right light and with keen eyesight.
Somewhat similar events made me look like a hero.
My dear wife likes a cuppa in bed before breakfast. She dozed off while holding said full cuppa, spilling it on the MacBook Pro she was using at the time. Result: a drowned & totally dead machine. Extremely expensive to fix, so she migrated to an old iPad.
I kept the dead MBP. About a year later, I decided to get rid of it and wanted to scrub its internal drive before I did. I plugged in the charger in the faint hope it would power up and let me clean the drive without having to disassemble the machine. To my surprise, it booted up. A year of drying out did the trick.
So my wife thought me a techie god when I brought the resurrected laptop to her. She nicknamed it Lazarus.
”I kept the dead MBP. About a year later, I decided to get rid of it and wanted to scrub its internal drive before I did. I plugged in the charger in the faint hope it would power up and let me clean the drive without having to disassemble the machine. To my surprise, it booted up. A year of drying out did the trick.”
My father’s iPhone 5s did scupa diving and got soaking wet while doing it. Careful drying, a week or so, did the trick and the phone was back up and runnig. Except WLAN and Bluetooth, they were gone. When WLAN started to work in a month, I was a bit surprised. Then Bluetooth came happily back in 6 or 7 months after the incident, it was quite a bit more surprising, and it still works to this day. (The poor phone took a lot of other beating too, including bent frame and forcing it back into shape. It’s a story for another time, though, for not to get too much off topic.)
Yeah, a watertight case. Too bad it wasn’t that watertight... maybe I misread the quality assurance document that came with the case and they tested that water stays inside.
I have to admit that I may have overstated the incident a bit. The phone didn’t dive too deep and it was fresh water in a lake. Still, quite impressive that it survived, don’t you think?
Years ago, I had a spare SPARC, and being a curious fellow, I decided to deploy Snort on it and leave it running as an IDS. It never detected a single thing (not surprising as I didn't have clearance to run it, and thus it was "internal only" - and on a switched network of course).
Except for one day when it did. It detected water. A pipe above it burst and electrocuted it.
If you have one of the most modern ones, of course, your computer can function in one and a half metres of water for up to thirty minutes.
From an accidental practical test, though, you have to shake it a bit to get water out of the microphones and speaker before phoning to tell your wife you fell in the river.
I wonder how long it will be before somebody offers servers with IP68 rating? There are places in the world where it might make sense.
There are places in the world where it might make sense.
Certainly. Elbonia comes to mind. Besides, I leave it to the reader to decide if the following made sense.
In one of my former lives, more than twenty years ago, we had "laptops" in the army. Those beasts weighed half a ton, were rugged, NEMP and waterproofed (don't remember if IP67 or 68). Nearly undestroyable and yes, we tried a tiny bit. So far for the impressive housing. Connections were mains (internal power supply) and serial port (RS-232 proprietary connector). Inside was a stupid terminal. So far, so useless. But it gets even better: costs per piece, I don't remember exactly, but it was well in the five figures (might have been around £25k in the mid 90s).
@EvilAuditor
Military stuff always comes at eye-watering mark-ups even where it doesn't make sense. NHS paying more for aspirin than consumers. Its just routing government gouging, because its nobody's job to say if its reasonable, only that a "process" has been followed.
In civvy life you just have to breath a keyword like "Wedding", or "new baby" for example for everything to (at least) double in price from identical items not sold under that heading. So its just as bad elsewhere.
@Giovani Tapini
I totally agree with you. And while many "price tags" in military terms are indeed eye-watering, more often than not it is -more or less- justified. Where my understanding comes to a full stop though is, e.g., when we finance an improvement programme for ageing vehicles which costs considerably more than buying a whole new fleet of the same, modernised items. Or developing portable terminals for silly money with no use at all*.
* what the fuck did those terminals connect with? Sure, we could connect them, one-to-one, over an encrypted data link. Just like SMS - slow messaging service. They were nothing more than telex with a screen. And no one, really no one ever used them.
I believe in some instances a large part of the cost of mil spec equipment is that the company involved has to employ people to deal with the MoD's* tendering process. Obviously for a laptop it's going to jack the price up a lot, whereas for a warship it'll be a rounding error, because why would you have a different tendering process?
*Fairly sure it's Treasury mandated so the NHS etc. will have similar issues.
"In civvy life you just have to breath a keyword like "Wedding", or "new baby" for example"
Or "boat", "marine", "watercraft"...
An entire industry where people pay big $€£₴лв₪¥﷼؋ for the pleasure of a single stainless steel fastener blessed by BRP, Mercury, or similar.
Or anything with RGB LED accent lighting and a fancy heatsink marketed towards "power gamers" by the tech world.
Or "boat", "marine", "watercraft"
Not something I was aware of, but good to know. Funnily, since many people (in sailing, at least) want as much boat for their quid as possible, they end up with rather poor quality (probably overpriced, too) and constantly tinkering, repairing and cursing their vessel.
George Buehler, one of the gods of wooden boat builders, distinguishes little big boats and big little boats. You are describing big little boats (i.e. overstretched with scantlings and rigging dimensions intended for smaller boats). Buehler's preference is for little big boats, i.e. you make your money go further by building a smaller version of a big boat but with the scantlings of the big boat.
His point is that if you want a little big boat you have more or less to make it yourself because that isn't where the mug spends his dollars. But when you hit a rock you live to tell people about it.
I just discovered a site that rates valves (I'm thinking of making a valve preamp purely for amusement value). Audiophiles queue up to claim that one maker's small signal triode or pentode creates better stereo separation and fuller bass than another. I want to know how the inferior ones magically communicate with one another and affect the output transformer and speaker. If bottled, it could be very useful.
Why bother with mass data leaks? Identifying these people would create a very valuable list of exceedingly gullible people with more money than sense. And I suspect some of them actually use their real names. Unless they are the peddlers of the valves that have been treated with liquid nitrogen to "affect the molecular structure of the cathode".
Things fail on boats, generally speaking you fix them while the boat continues to float quite happily.
Things fail on aeroplanes, generally speaking the aeroplane doesn't just sit in the air waiting for you to finish fixing it.
I think aeroplane things being expensive has a bit more justification.
You go down to your local PC dealer of choice and get a top of the range PC - £2,000 to £3,000. Then you molly-coddle it in a nice warm office.
Now take the milaero version:
To start with, no fans. Then add that it must work in a Siberian winter and a Saudi summer. (-40 deg C to +85 deg C)
Add to that, it will be working in a tank, fighter jet or helicopter - shock, Gforce and shaking.
It might also be subjected to condensation as the vehicle warms up, or salty water if it is in a small jet boat.
And it must have the latest processor, graphics, etc. that you have in your desktop PC.
I think that adding a zero to the price is actually quite reasonable!
Disclaimer: I have never used military equipment. I'm not stating things, I am asking them.
All those requirements about what the machine must withstand make a lot of sense for why the prices are higher. However, I have a couple of questions. First, do they require that a device be capable of use in all of those situations simultaneously? For example, I would have assumed that machines intended to be installed in airplanes would be distinct from other classes of devices, as a ground or tank based machine would not need to sustain the G forces, the weird acceleration, or the recoil of firing plane-mounted guns. Second, do military computers really have the same specifications of top of the line machines? I don't know about the military models, but every time I have looked at or been requested to find computers for difficult environmental conditions, the options have been the following:
1. Machines with very old specs, usually something designed for windows XP or some old version of android.
2. Devices with somewhat modern specs, but with little access to the system. You stick with the OS installed on it and just write your application over that. This can be problematic when the device is running some restrictive OS like android (there are many of those).
3. Devices that seem to have a modern processor, but are very heavy, power intensive, and run very hot. These would often be unacceptable for military use per the specifications above because they use fans to dissipate the heat and, as far as I'm concerned, seem to be of somewhat dubious build quality.
I would assume that most military machines are running a slower system that is capable of running the program required, but does not have a ton of extra speed or graphics capability. From the original post, the machines described were sufficiently behind the machines of the time in terms of computing capability. From your experience, is most military hardware more advanced or more restrictively specced than I had assumed?
I can tell you - from observation, not experience - that the Panasonic Toughbooks we used are not fist-to-screen resistant.
I can also tell you - from direct experience - that the COTS computer we used for a handy but non-essential function is not resistant to a bucket of water being accidentally dumped on it.
Having managed several projects deploying CF19 toughbooks I came to the conclusion that 'prepping' the laptops with a 'standard build' is what causes the issues. What really wound me up having spent 8 weeks working with the deployment team aand Pansonioc engineers to come up with the best possible Panasonic build (including new aircard drivers developed for us in japan) was when the internal ITteam decided to remove all the 'Panasonic bloatware' and use the default windows drivers, cue all the problems with connectivity screen rotation etc we had spent 2 months resolving re-appearing. It took another 6 weeks to get authorisation to have a special Panasonic build approved even though we were rolling out 1500 of the them. A couple of years later working on an unrelated project at another customer I found that their own (very small) fleet of toughbooks would not work in tablet mode and the screen didn't rotate, again because the Panasonic drivers and utilities had been de-installed to 'clean up' the build. There is a reason toughbooks cost so muych and only part of it is the hardware.
So-called "ruggedized" hardware is usually way behind hi-tech or recent. Part of it the chip makers aren't making the hardware a priority. Plus, it may have been hi-tech when specified but by the time it's deployed it's damn near obsolete. If you can get access to the specs for the hardware, you'll see the differences in components compared to civilian components. A big cost adder is "testing"... as some have pointed out.. 100% testing as opposed to civilian being maybe 1 out of 10 without the harsh environmental specs of the military.
Not defending the prices or anything else, just explain what I've seen in the industry.
One of the big problems with testing military components is that the test process may damage them. Back in the day when things like transistors and ICs came in metal and ceramic boxes there was a test called PIND - particle impact noise detection - which involved shaking the shit out of the things while listening with sensitive microphones for any rattles. Of course there was a risk that the shaking could loosen the bond between wire and pad. Then, how do you do ESD testing while being sure that the last ESD zap didn't cause any damage that would cause early failure?
I remember this because of a scandal in which National Semiconductor were caught faking test records on an industrial scale for military testing that had not been carried out. IIRC, the affected batches were more reliable in service than the tested ones.
The thing is, what is the AQL? For boring old JAN stuff sampling is OK because the usual side effect of a defect is that something may not go bang or a radar module might have to be replaced.
With JANTXV the theoretical AQL is zero defect and you need to inspect and test everything.
One of my recurrent dreams is the time I had to go to a certain large French company and explain to them that if they were only prepared to pay for 4% defect rate, it was no good complaining because about one percent of the units failed in the field. Though the engineers took me out to lunch because they were so pleased I had made their manager look an idiot in front of his boss.
"without the harsh environmental specs of the military." Environmental tested COTS ICs for over a decade. Difference between commercial and military stressing per MILSTD 883 was that commercial ICs procedures were stricter than the requirements for military. example: where 883 required, say, 8 shocks, commercial procedures required 10-12 shocks.
This post has been deleted by its author
“Paying 10x for a ruggedized computer is painful, wince-inducing, but justifiable. It's when the MoD are getting charged 100 pounds for a six pack of toilet roll that I get grumpy.”
When you’re in a hostile environment putting your life on the line, wouldn’t you prefer to know that a little more has been spent preventing unnecessary roughing while maintaining adequate strength to avoid digital penetration?
”When you’re in a hostile environment putting your life on the line, wouldn’t you prefer to know that a little more has been spent preventing unnecessary roughing while maintaining adequate strength to avoid digital penetration?”
Not to mention per unit stress tests and all the paperwork involved in quality control. Absolutely worth the price.
If you have to pay for the tests to simulate "on the flight deck of an aircraft carrier at sea" you will know where some of the money goes.
And the design briefing begins (or used to) "remember the chaps using this may be upside down in a ditch with people shooting at them" - which is why you have too be able to identify the controls by feel. Pig ugly is a design requirement.
"Military stuff always comes at eye-watering mark-ups ". Well if it's like other US government stuff. Part of the price is stockpiling enough spare parts for 10 or more years or so you can fix it without redesigning the item or requalifying a new part.
At least that was Univac's response when justifying the high cost of spare parts. Yes, diodes don't cost $$$ normally but we have to warehouse 100's of them and all the other parts (as per the contract the government wrote) so they can be fixed for umpteen years.
A relative of mine tells a story of getting parts for the military:
A particular bolt, with fairly common, reasonable specifications, was commonly found to not fit. They carefully measured the bolts - and found they were slightly out of round. It was being manufactured specifically for the military, with 100% inspection. So, for future orders, they tightened the tolerances (for a significant markup) - no improvement. Finally, they went and visited the site.
They examined the manufacturing equipment, pulled a couple bolts straight off the line, measured them, and found them to be well within spec. Grabbed a couple right before QA, checked them - well within spec. Watched QA pick up each bolt, carefully measure it, verify it was within spec - then grab a large hammer and a stamp to mark the part as "inspected"...
Next version of the spec expressly prohibited stamping the parts as inspected, and loosened the tolerances again. No further problems.
"QA pick up each bolt, carefully measure it, verify it was within spec - then grab a large hammer and a stamp to mark the part as "inspected"..."
Sounds just like the place I worked where the shipping department added corporate asset tags to precision weights.. (before my time, though, so who knows the accuracy)
You forgot the paperwork establishing the source of each item, the names of all that looked at the item and the herd of clerical types exchanging memoranda (thought I use a little pedantry here). Each time a pol pushes the 'by cutting waste and fraud' meme a new set of forms and certifications are born and they with supporting staff must supported on either the existing budget putting project completion further out or of obviously a bigger budget to maintain the project. Parkinson missed 'cutting waste and fraud' as a source of bureaucratic growth.
Then 'if we don't spend all the budget, we'll get less money next year' meme which is so popular in the military. End of the fiscal year more gold plating takes place and thereby blowing to hell the idea of saving based on zero budget costing.
Its just routing government gouging, because its nobody's job to say if its reasonable, only that a "process" has been followed.
Part of the increased price is CYA paperwork, plus some extra testing and certification.. etc. etc. etc. The end result is everyone on the board of the aspirin company gets a new Porsche or Ferrari. So, not "gouging" but "profit" for hard work on their part.
"Military stuff always comes at eye-watering mark-ups even where it doesn't make sense. "
Apart from where "toughness" is required, thus ramping up the price, warranty repairs can be a bit difficult. If any part of a PC is or has ever been used for "secret" work, then any part with storage can't be removed from site and returned for warranty replacement. And these days, apart from the physical case, pretty much every sub-system has programmable firmware chips or some sort of storage on it. All those items could potentially store data for exfiltration, even if just credentials. That usually means on-site repairs to IT kit cost more because no one in the supply chain can pass the faulty bits down the line to the manufacturer for replacement. This is all factored into the sale price.
Yes, I have had dealing with the MoD.
"because its nobody's job to say if its reasonable"
Maybe this is the case where stuff is centrally sourced. If you're running a lab. in the Civil Service and you have an annual budget you do - or at least I did - take good care to get value for money. For instance discovering that one of our more expensive biochemicals had a much cheaper source than the usual supplier and that a useful test strip worked perfectly well for our purposes when bought cheap beyond its use-by date as a medical diagnostic.
“Any evidence for the NHS paying more for aspirin than consumers?“
From a quick google search, it looks like it may have been paracetamol/ibuprofen rather than aspirin:
https://www.dailymail.co.uk/news/article-3711703/NHS-spends-87m-paracetamol-20-TIMES-high-street-cost-Doctors-wrote-22-9m-prescriptions-year-average-cost-3-83-each.html
There’s a lot of details missing to judge if it’s a fair reflection of costs (ie comparative strengths) and having read the article I’m none the wiser as to whether the mail is most concerned about the per item cost or the NHS covering over the counter items.
My suspicion is that this is higher strength than available in supermarkets (ie. prescription only) and it is prescription only medication to reduce the chances of overdoses. And there is likely to be another article on the Mail calling for increases in dose restrictions to prevent overdoses after some one died from overdosing on supermarket painkillers.
My mother gets prescription Paracetamol, it's the same dose as regular (500mg/caplet) but is over the 32 allowable in otc or off the shelf paracetamol, typically boxes of 100 caplets. She often forgets to take them, or deliberately doesn't take them so if she gets a headache she's got something to use.
The end result is there's enough paracetamol in the house to stock a couple of pharmacies for a month.
"From what I understand there will soon be a lot more places around the world, mainly coastlines and large capital cities that will need more than IP68."
The relationship between land and sea is never stable unless you have a short term view. Raised beaches (Antrim coast), peat buried under the beach (Portrush), major port disappeared beneath the sea (Dunwich), major port now a couple of miles from the sea (Rye), ancient city sinking into the sea because its groundwater has been pumped out (Venice). If you think land/sea changes won't happen or if you think you can prevent them you're going to be disappointed.
"I will never forget this call," Brandon told El Reg. "It taught me to always ask for more information before blurting out the seemingly obvious answer to what seems like an inane question."
I dont have time for that , by the time the user is half way through babbling out irrelevant information , i will have read between the lines , know the issue , figured out the hostname of the users pc , connected to it and fixed it*
Then , out of politeness, rather than just hang up, i attempt to interrupt the so far continuous stream of waffle to inform them the icon is now on their desk top , or that print queue is now set up , or whatever ...
*if its an easy one obvs.
"Apparently, the cooling system had burst a pipe and the data centre was filling up with water."
The number of levels of fail that goes with that event is rather telling. - Starting with "Floor drains? What floor drains"
Not too many years ago I walked into a new building being completed to check out our new dedicated data store - upon looking up, I discovered that the room where I'd specified "Absolutely no water allowed in this room, must be dry at all times" now had every water pipe in the building passing through its ceiling cavity, complete with inspection joints (which are more likely to leak than any other point)
The builders hadn't bothered talking to anyone when they encountered a problem with the design and just bodged it.
The other classic - mentioned in the story - is putting computers in the basement "because they don't need light"
As someone mentioned in the Kwik-Fit story, companies simply don't realise how much they depend on their computer systems - until they don't work.
The realisation usually dawns on management when you ask them to assess how they'd cope without computers and/or total loss of data.
The number of levels of fail that goes with that event is rather telling. - Starting with "Floor drains? What floor drains"
Now I'm curious. Having been in many basement datacentres (and IT depts), I've never really thought much about drainage specs. And as you say, there also tend to be hot & cold water pipes running through them, or near them. Or in the case of my first DC, big chillers to feed water into the IBM3090. That was on the 1st floor though, and leak detection/drainage would've been to the ops people working on the floor below.
But DCs often seem to be fairly well sealed up (or should be), so curious what standards there are to deal with drainage. Which being basements, presumably might also need sump pumping. Or a note for the Ops people to let their telcos know that ducts may now be drains. And occasionally the telco may be the source of the flood. Like when an outside plant guy happily drilled a hole through the wall of a new customer.. and also drilled through their main hot water pipe. Oops.
I can't believe that was 27 years ago, it feels like yesterday. I was in Chicago that weekend and the Loop looked like something from a Twilight Zone episode, it was completely deserted and from the street level there was no sign of what had gone wrong. For those who don't know, here's what happened. Chicago has a vast underground network of coal tunnels that were built in the 19th century. These tunnels connect to almost every major building in the Loop, originally they were built to transport coal but every new utility, from phones to fibre, uses them. Unfortunately the tunnel system runs under the Chicago River and one day in 1992 a contractor was driving a piling into the riverbed punched a hole into the tunnel which promptly caused the river to drain into the basements of every building in the Loop, it was months before the mess was cleaned up. If anyone had a data center in their basement, and I bet there were a lot of them, they would have been completely destroyed.
I can't find it (Circa 2004 on El Reg), but I recall BT scrapped a whole new server data room because the sewer pipe that ran parallel to the room burst & filled it.
Nobody wished to repair\evaluate the equipment in the sewer filled room.
" I ain't touching that it's full of shit!"
I once encountered a factory building that flooded regularly. It turned out there was a small and rather inaccessible ditch behind it which had become blocked. For some unaccountable reason, the support I-beams adjacent to the ditch were mounted in such a way that water could enter over the top of the end plates and up the sides of the beams. Presumably the builder had thought that this would improve the building drainage if there was a leak. But in fact the last thing you want is something like oil draining into external ditches from a factory leak.
The fix was half an hour of a workman clearing the dead branches from the ditch and a pot of quick setting cement to block the holes, a tiny fraction of a percent of the production time losses due to flooding.
Perhaps you've heard about the Holy Cross football team of a few decades ago? As there were no backflow preventers, a drain pipe allowed standing water to be sucked into the cold water system. Result was that the majority of the football team were stricken by long term intestinal complaints and the season was cancelled.
'Nother tale: Once upon a time in a town not far away, there was a local who scorned the necessity for requiring licensed professionals and decided to (loudly, that being his habit) show that after all anyone of reasonable intelligence could do a similar job. He decided to do some plumbing of his house and then have the job professionally inspected to show that his work was on par with that of a professional. (Sneer sneer) The pro did the inspection and issued a large "Condemned" tag, commenting that the guy should tear out the job and start over before he poisoned his family. I do recall a few of the issues: no backflow vales, no lead free solder, several other deficiencies.
Not a datacentre, but I do know someone who had a cellar built in a house in London. It was on a hill, what problem could there be?
The cheapo Bulgarian builders, however, did not want to (or did not know how to) fit a floor drain that would emerge down the slope. Instead, they suggested a sump pump which would pump water up to the main drain. Which went through the top of the corner of the cellar. The sump was built, the pump installed, and the piping went behind the wall lining. I did ask what would happen if a leak coincided with a power cut while nobody was in the house but this was dismissed as unlikely.
The inevitable happened; the washing machine decided to leak and the cellar filled up with water while the pump was running quite happily. Because the cheap plumber had forgotten to connect the pump outlet to the overhead drain. The pump was sending the water into the space between the lining and the walls, from where it poured out of a ventilator.
I'm not suggesting Bulgarian builders are worse than any others; I'm suggesting that blindness to the possible effects of water leaks and their sheer destructiveness is remarkably common.
pipework loose fitted
Had almost the same thing at a house I where I was doing some electrical work. They told me that when they'd had an en-suite fitted in the front bedroom a few months earlier, as the drain was at the rear of the house a macerator was needed. The plumber ran the waste pipe under the floorboards (the joists were going the "right" way), out through the wall and then down to the drain.
Except that the twit had used push-fit. The elbow shot off after a couple of pumps and "stuff" fountained all over the very small courtyard garden. At least it hadn't come apart under the floor!
M.
In my old house we had a little Hepworth H20 push-fit plumbing. Somebody decided it'd be fine from plastic to chrome-plated pipe.
I was awoken to panic as water started leaking through the ceiling.
I was in the middle of selling the house as well, and had the Surveyor coming round that day - I still don't know how I managed to clear that up. Put compression fittings on from there and it was fine. Luckily it was the nice clean supply side, rather than waste. Ewww.
I had a gig near Gatwick and stayed in a new hotel. So new they hadn't finished the snagging. I went into the bathroom of the first room I was put in and discovered water coming through the bath overflow. Nothing through the main drain, just the overflow. I don't know how they managed that.
That's a very shitty thing to have happen. Bet your friend had a crap day when he returned.
It's also why you ALWAYS either leave something clearly unfinished or finish it completely. Or at the very least tape the bog lid shut with a very big notice on top that the pipes are not connected.
The builders that built my house decided that the best way to attach the shower to the wall was the grip-ring of a push-fit fitting and rely on the water pipe to hold it in place. No screws needed.
Surprisingly it lasted about 3 years before it jumped off in the middle of the night and sprayed water everywhere.
Girlfriend panicked. I calmly turned the water supply off and went back to sleep.
5 minutes work in the morning with a power drill and a couple of screws sorted it properly. And lots of cursing of the lazy bastard builders for not doing that.
putting computers in the basement
ISTR (though I can't find the story now) that the National Archives - also on the banks of the Thames - quite deliberately put their servers in the basement because the data is backed-up, computers can be replaced if the place floods and ancient documents most certainly can not.
M.
Working in a hospital that has been on the same grounds since WW2 it was probably expected that the old air-raid shelter under one part was used for cables & pipework. Unfortunately some bright spark decided a 6U Cab could be mounted on the wall (about 5ft up) as "All the cables go through there".
This only became an issue for me after a few days of constant rain when I had to roll up my trousers and wade through ankle deep water to patch in a new port. The estates department took 2 days to pump it all out. Luckily no IT equipment was harmed.
The Cab stayed in that location for a few years & yes it flooded again but by that time I was senior enough to send a minion.
I got to see inside the comms room of a now-defunct ISP many years ago. The comms room had a window to the rear of the building which was clearly left open most of the time for ventilation purposes.
Unfortunately pidgeons had made use of the open window for nesting purposes at some point ...
Not quite water.
One place I worked at had a basement canteen on split levels.
All well and good until the fuel tanks for the diesel generators sprung a leak and flooded the lower part of the canteen in 3 feet of diesel.
There was a background smell of diesel for months and even after redecorating could still see the depth line the diesel made.
I did find a 100mm pipe in the subfloor of one datacentre I worked in. Nobody, not even the facility manager could tell me if they were for cable runs or drains.. Nothing ever came out of it though (at least not while I worked there)
More in line with the question of colleagues in hazardous situations... One colleague, was questioned at gunpoint by security while working out of hours in a foreign branch. The implementation PM had apparently failed to inform security that he was supposed to be there. The service desk had to find us to confirm he was legit!
A second colleague was stopped by the police after scaling the car park gate which was locked after a weekend callout. He had been given all the building keys, but no one thought to provide the gate padlock key! Someone had reported him, which, to be fair, was probably the right thing to do...
and people think IT is a boring office job..
> One colleague, was questioned at gunpoint by security
At uni, some students (AKA cheap labor) were running network cables.
We discovered the hard way that the robotics lab had silent ceiling sensors to protect the very expensive robots, and these went straight to the uni police department.
Not quite the same, but in the "work hazards" story space...
At one place, one of the big industrial UPS units - one of the ones large enough to climb into - to needed some work doing on it. In situ. While still plugged in.
Oddly, it was very hard to find anyone willing to take the job on. Eventually, someone was found - an ex-military lady engineer.
We ended up rigging a hook on a wooden broom handle, so we could pull her back out if anything went wrong. Though thankfully, she came out unscathed, and with a healthy hazard-payment top-up on the invoice, to boot!
If someone higher-up approved that they deserve sacking. No way in hell that's a safe thing to do. Just reading this I can't help but think: "The lot of them deserve to get sacked for doing that"
No amount of hazard pay is worth putting your life on the line. Not to mention the amount of power in one of those large UPSes means that if something did go wrong, the hook would probably not do you much good because there would either be nothing left to bull back or you wouldn't be able to get close enough. People seriously underestimate the power in an arc-flash accident.
It certainly would be the kind of thing we're I'd refuse to even be a part of it if doing it "live" is the only option. Because if something DOES go wrong, "death by criminal negligence" is actually something that can be made to stick, depending on local laws.
Not sure what the voltage would be on a UPS for servers, but have to agree with that comment.
Where I worked we built proper high voltage convertors. Everyone, including us softies, had to have safety training which usually included being force-fed some rather distressing photos and watching a video from an external security camera of some men working on a cubicle (with an open door). There was a flash followed by fleeing people, some obviously on fire.
We were then asked how many people were involved, the common answer being three as that was the count of fleeing people.
The fourth person was not seen as they had to be cleaned out in (fried) bits.
> If someone higher-up approved that they deserve sacking. No way in hell that's a safe thing to do. Just reading this I can't help but think: "The lot of them deserve to get sacked for doing that"
Hell YEAH! Juice, what the HELL were you THINKING!?
If you HAD actually TOUCHED her with that pole, never mind your job, your whole CAREER would have been #METOO'd!
If there was a proper risk assessment and work plan it might be acceptable depending on the risk of switching off. (I suspect there wasn't though). But if she was ex-military she might be up to doing that for herself. Lots of boaters, after all, are quite happy to work on their electrical systems - basically big UPSes - while the engine is running and accidents are rare. Though I did hear of someone who lost a lot of hair after failing to understand that when the batteries start bubbling a lot, that's a rather inflammable hydrogen/oxygen mixture coming off.
A few years ago our local pylons were rewired. Watching those guys up there feeding the new cables through and attaching the new insulators was interesting, but not something I would do even for the promise of huge amounts of money. Apparently they often do work on live installations using tools on long glass fibre poles.
I may have seen a more dangerous method of dealing with high-tension wires. Helicopter full of people in special foil/metal coated suits. Conductive pole to attach helicopter to live wire so every thing is at the same voltage potential. Fly next to wire, attach pole, Fix wire, detach pole, fly away.
Great for remote locations, and can use helicopter to detect problems to fix.
Where I worked a large high voltage transformer that powered the building blew up early one morning. This was right near the guard station at the entrance of the building. On questioning the guard said there was a hell of a noise but didn't think to report it!
When the replacement transformer was installed a workman attached the wires live to the new transformer. Two other workmen held 2x4s which the workman reached over to do the final connection. They braced themselves - ready to pull hard. Their job was to pull the first workman off of the wires if anything went wrong. All three workmen took their job very seriously.
This is a job I would never want.
We have two offices plus server room in the basement. The one office have a toilet.
You can see where it's going.
Yep. The toilet's water inlet to the cistern sprung a leak and flooded the entire basement. Luckily the building was built on a hillside, so each office have a large sliding door to the west where the water could exit.
Because we had a flooded server room a year ago, recovery this time was quicker. All servers + UPS'es was shut down safely, which allowed the cleanup detail to mop up excess water without risk of electrocution, and when everything was clean and dry, the UPS'es and servers was started up again. All in all loss of an hour's productivity due to offline servers.
I had a hamburger (or was it hamberder) restaurant manager turning on a CRT in front of me to show it was dead.
I was there to assess the damages from a badly leaking roof and heavy rains during the night. Water was still pouring almost continuously on the screen, from the ceiling, as the guy pushed the button !
I stepped back 2 meters at that moment. I was the one in shock.
This story gave me a flashback of working at a London-based office of a well known American investment bank who later merged with a large American insurance company.
We unexectedly lost access to a number of systems one afternoon. Word came through that the European operations centre had gone offline due to flooding. More details surfaced later on that an air-con engineer working on site had accidentally stood on a high pressure coolant pipe, causing it to rupture, at which point coolant began jetting directly into the hall where our equipment was housed. Rumour has it that said engineer simply fled the site...
We didn't get our systems back online until several days later, being non-production instances which fell way down the list of priorites for such an incident.
our old server room (in a laboratory) was underneath a lab with a milli-q water purifier in it that quite regularly over flowed. Anyway we needed to build a new datacentre and the perfect location was a room next to the old one as all the services were already near by, ducts for power, routes to the chillers on the roof, networking, right next to the lift, etc. Anyway our total halfwit of a commercial director at the time decided NOo a far better location would be to build the DC in an old store room in an extension to the main building, no where near any services which meant a right old job of pulling a new electricity supply up from the basement as well as sorting networking out, etc. All this so he could turn the existing DC in to fancy meeting rooms. But the real killer was it was RIGHT NEXT DOOR to a massive refrigerated mesocosm labwhich had something like 10 tanks each holding several 1000's l of salt water! what could possibly go wrong! Anyway 5 out of 6 DC design companies said ermmmmm I really wouldn't put it there. But he was determined and if it wasn't for planning permission being hard to get as we would have had to have a noise survey for the chiller units to be installed there (as it was opposite residential areas) it would have been built! To say the guy was a total utter feckwit is an understatement. He was also a bullying c*nt but that's a whole other story!
Yes, drains can become fountains. What is more terrible is when toilets become fountains. Yup, had one of those in my two story house. The lowest toilet in the house gladly "fountained" and you can assume the rest. Had to get the drain cleaner guy out and snake the drain to the street (75 feet of it!) to unclog the mess. Before I re-assembled things, I took things outside and hosed the toilet down (inside and out), then hosed down the half bath where the subject toilet was supposed to be.
It wasn't a pleasant experience.
Or as the saying goes: When you are up to your @$$ in alligators, you forget that the original task was to drain the swamp. Priorities, Priorities!
Computer rooms tend to be fed with with either 230V or 400V, both of which come under "low voltage" (those low voltage lights in BnPoo are actually extra low voltage) Standing in water though....more than enough to kill you, 25 volts when wet will penetrate your skin.
High voltage is actually well in excess of tens of thousand volts heading towards 250,000 volts and that can jump air gaps
"High voltage is actually well in excess of tens of thousand volts heading towards 250,000 volts and that can jump air gaps"
Actually no. Not according to the IEC, to which I was once an unworthy national representative.
SELV - separate extra low voltage - not connected in any way to mains - up to 60V and regarded as safe. As used in old style POTS exchanges, mobility scooters, lawnmowers and many small electric scooters. The 60V is supposed to be the voltage you get when charging the nominal 48V standby batteries of a telephone exchange.
ELV - extra low voltage - less than 50VAC or 120VDC
LV - low voltage - up to 1500VDC or 1000VAC. The peak voltages between conductors and earth are restricted to 600VAC/900VDC by the British Standard.
HV - high voltage - anything over 1000VAC/1500VDC.
This has the virtues of being simple and consistent (and really being a fairly good guide to the associated risks) and the downside of confusing the general public who think the AC mains is "high voltage".
The US, of course, has its own standards, or lack of them. NEC doesn't even define high voltage. NEMA regards high voltage as cross country pylon levels and is out of step with everybody else in the world, just about.
The auto industry adds to the fun by calling what the IEC calls low voltage, high voltage.
The point of all of this is that there's a reason people need to be certified to work on certain system types - the terminology being so confused, it's essential to know it in context.
Incidentally, as a student at a certain university in the fens, I remember that the lab benches could be supplied with DC to rather readily accessible screw terminals, with voltages up to at least 200. When I asked the man who set up your bench voltages for you if people often shocked themselves with the DC, I got the reply I should have expected - "Not usually more than once".
I'm with the public here - anything strong enough to kill you should be named as such. "low voltage" sounds safe to those who don't know the definition.
SELV sounds like the only range I actually want to poke around with. If I were to do the naming, that would be "low voltage" and everything else would be "high voltage" with modifiers "super high" etc.
I used to do support for an accounts software company. Visited one customer, an engineering supplies firm, and all the computers on the ground floor that would normally sit on the floor under the desks had been placed on little home made wooden plinths.
Apparently, between a neighbouring set of playing fields and them was a car park, which had recently been resurfaced. First decent downpour and they discovered that the car park was a lot less absorbant than before, and all the water running of the playing fields was now getting across the car park and pouring through the offices and warehouse area..
After a couple of episodes, their solution was just to raise everything up by six inches, above the water level, until The Powers That Be could come up with a permanent solution..
"After a couple of episodes, their solution was just to raise everything up by six inches, above the water level, until The Powers That Be could come up with a permanent solution.."
If the floor is carpeted, then that's a good dust prevention measure too. The air intakes are usually at the bottom of the PC, the rubber feet designed to lift the case above the surface tend to sink into carpet and every time someone walks by or the user shuffles their feet, dust is raised from the carpet and sucked into the PC. 6" seems to be the magic number for a quite dramatic reduction in the amount of dust sucked into a PC.
I mean at 110 or 230/400 Volts, the main danger is actually caused by the hydrogen created by electrolysis. Since electricity will favour the shortest and easiest path, and power circuits will typically have all their conductors close to eachother. Therefore the flow of current will be localized. Additionally while your body conducts electricity about as well as water, your skin is somewhat more of an insulator.
What's actually more important is to check all connectors after such an event, as corrosion can be a real problem.
Additionally while your body conducts electricity about as well as water, your skin is somewhat more of an insulator.
You'd think that. But there have been one or two incidents of hair dryers and the like being dropped into bathtubs, with fatal results.
Nowadays we're supposed to have earth leakage circuit breakers (built in to RCBOs) to reduce the danger of death.
Hair dryers are weird - I've never understood why it's allowed to have live conductors having air blown over them in a forwards direction. All it takes is a partial burnout and a bit of red hot wire shoots out and can set fire to hair.
I once had an enquiry from someone at Trading Standards after someone had this happen and burnt themselves quite badly. Why hadn't the fuse blown? Because the cause of the problem was the element going open circuit, so the fusing was totally irrelevant.
Ha ha. Reminds me of working for a controls company, doing real-time control by computer. One Control Room was in a basement which got flooded. Of course, everything went down and we were called. They got the water drained off from the control room. As we went underneath the desk-level CRTs and operator keyboards to see what damage there was to the kit underneath, we took the panels off. The first panel exploded with water behind it when sufficient screws were loosened. All the panels were still full of water. The Site Report following this incident suggested in the Lessons Learned section that in future all control panels should be fitted with drain holes !! :)
I guess the machines were written off?
Also things you shouldn't get water in: Anything with ribbon cables.
Turns out that they have a bad habit of corroding over time and it may work for a while then fail badly.
This is why laptops, camcorders and cameras typically get condemned if there is the slightest chance they got wet,
although some folks still use them for parts because the SMD chips are normally fine if it was dropped in water
unpowered.
Got some very nice EVF units this way.
Worked on any number of data centres back in the nineties and never came across one that was water cooled. Indeed you would always be careful to ensure that any cooling plant was outside the data center itself so that hot air was drawn out of the computer hall and cool air blown back in through ducting. Generally speaking cooling plant has condensate to drain. Even then you had moisture sensors in various locations below and above floor level. While it would be preferable to run power overhead in a data centre I came across an awful lot where power was run below the floor.
So in short that story puzzles me. Even if it used water in the cooling why were the pipes in the computer hall itself? Why were there no moisture alarms screaming the place down way before it got to that stage? And how had water do enough to wade through not already caused an almighty short? Oh and if there was a water cooling system why was there such a large volume of water involved.
Maybe because of the little incident I describe at Digital Equipment Corporation in the mid-70s? In fact, the pipes were running along an outside wall of the computer center. How the forklift operator managed to punch through a cooling pipe and presumably through the wall of the computer center in a way that very precisely funneled the water in under the raised floor is unknown. I'd have assumed it would take some very precise engineering and construction to achieve that result.
That sends me back to the mid-70s. I was working for Digital Equipment Corporation in Maynard, MA. The company had just built out a new computer center and started moving equipment in. Suddenly, our DECSystem-10 hosting the main engineering databases shut down. Calls to the new computer center weren't being answered. After about 30 minutes we got word that it was going to be off-line for a couple of days, as were all the other systems in the new center.
A week later I ran into one of the data center operators I knew who had been on duty that day. He was sporting a cast. A fork-lift had run into one of the (2) cold water pipes for the center air conditioning and somehow managed to do it so that the water poured in to the space under the raised floor. He looked across the floor to the still empty side of the room and saw air driven water spouts start to march across the room towards the newly installed equipment. The cast on his wrist was from hitting the EPO switch at a dead run so hard he broke his wrist. Fortunately the EPO also killed the air conditioning fans and that slowed the water flow just enough that they were able to cut off water a bit before it reached the machines. It took a couple of weeks to get everything fully dried out.
I've walked around an office in 50mm / 2" of water. Even though it was drinking water from a filter, the crap in the carpet made it very non-clean and therefore probably quite good at conducting.
Walking into one room there was a SFF desktop on its side on the floor, madly paddling water with its fan . So I reach down and turn it off at the wall switch, which is well out of the water.
BEEP BEEP BEEP! Unbeknownst to me, there was a small APC UPS on the other side of the desk, mostly submerged in the brown water. Being a standby UPS the batteries would have been in the bottom, and all the boards across the top of the unit. And it was passively vented by slots to let heat out and water in.
Thinking through all this, I took two very slow gentle steps toward soasto not make waves, and carefully held down the UPS's power button for what felt like several minutes, but was probably only 5 seconds.
PCs went off, and all was well. I should have just switched off the building mains - know for next time.
Beer, cos that's brown water too.
"Thinking through all this, I took two very slow gentle steps toward soasto not make waves, and carefully held down the UPS's power button for what felt like several minutes, but was probably only 5 seconds.
PCs went off, and all was well. I should have just switched off the building mains - know for next time."
We had a flood thanks to developers on the industrial estate cutting through a water main when demolishing an old building. The Disaster plan procedures were followed and the mains power was shut and padlocked off.
Unfortunately the disaster plan procedures were written before the server room got a containerised back up generator that automatically powers up and keeps the servers alive when the mains power goes off.
Thankfully the resulting voltage tango was non fatal.
Lessons learned. Robust procedures etc etc etc.
According to the expert at H&SE that I used to work with, there is NO safe level of current through the body. Fibrillation has been caused by currents of the order of microamps. If you are designing human usable equipment you should design for zero risk.
Farmers of course get an exemption, but people are regularly killed by electric fences.
Can be 5mA but it varies depending on the duration and frequency.
Irony: 50-60 Hz is most dangerous because the vulnerable period is during the QRS complex and this happens quite often .
Read somewhere that to *intentionally* stop the heart eg during certain bypass procedures they use low current 40 Hz using paddles and a
much larger DC shock to restart it later but they also use weak KCl because it does less damage.