Poor On-Call this week
Not just because the wrath to the user was not forthcoming, but because the sub0heading kind of gave it away.
You know, a bit like asking if someone's heard a joke by describing the punchline.
BTW, County Durham?
Thundering IT incompetence by government is hardly new. While the antics of the present UK administration may have gone beyond satire, round out your week with an On Call reminder that things never really change. Register reader "Bernard" (for that is what the Regomiser has called him) furnished us with this tale from the …
This post has been deleted by its author
Me Grandad used outwith. He learned it from his dad (a native of Northern Finland), who learned it from his first wife (a Plains Indian, apparently Otoe), who learned it from her first husband (a Yorkshireman). I re-ran across it as a teenager in Yorkshire, and still use it occasionally. So I disagree, outwith is indeed called outwith outwith Scotland.
Good to hear it, jake!
(But how many of your or your grandad's acquaintances would also be familiar with the word? We are quite happy to lend it to others, nevertheless.)
On this side of the pond, it's always highly amusing to see the braincrash expression on someone's face when you use the perfectly cromulent word "outwith" within a conversation, and they (being a southern sort) suddenly adopt a "huhsaywhat" look!
Yes, Englishes are a funny language(s).
(The same can probably also be said for most other languages, which have words commonly used in some regions or countries in the sprachraum, but not in others.)
You do have to wonder if sometimes it's just bloody-mindedness though. Take the Yorkshire habit of using "while" in the sense that most of us would use "until". That confused me the first few times I heard it. How do Yorkshire coders get on with pre-tested conditional loops?
In England, at least, the national broadcasters tend to cut down on the strongest dialects, though regional accents have become more acceptable in recent years. Strong dialects seem to be confined mainly to local radio.
In Wales, in Welsh, where there are vast dialect differences not only between North and South but between East and West and even between one end of a valley and the other, the policy seems to be to put all of these on the TV or radio at once and force you to cope. This isn't a bad thing, though for a "learner" it can get a little confusing, and it's not uncommon to hear people mixing their dialects in various exciting ways.
Welsh is particularly good at the "bloody-minded" dialect differences. The obvious example is that in most of the South, the word for "now" - as in "it's happening now, this minute" is "nawr". Probably related, some kind of transliteration of the English because Welsh people aren't renowned for doing things the moment you ask them to and probably didn't have a word for it until relatively recently, hence the phrase "I'll do it now in a minute".
In most of the North, the equivalent word is "rwan". Note how it is exactly the same word, spelled backwards.
I had a teacher at school with a very peculiar brand of Welsh from a very specific, small, geographic part of the southern valleys. Apart from very "Cehrdiffian" vowels one of its peculiarities was the way it described money. In most of Wales, someone looking for a 50p piece to put into a vending machine would ask for a "darn hanner cant" - a "half a hundred piece". Asking for 50p to pay for something they would ask for "hanner can ceiniog" - "half a hundred pennies".
Obviously this teacher was stuck in a pre-1972 world because her word for 50p was "darn chweigain" which translates as "six twenties", and in "old money" for those to young or too distant to know, a pound consisted of 240 pennies, so six twenties (120) was, indeed, half a pound.
And to tie up my Saturday morning ramble, Welsh still maintains two counting systems. For most purposes you use decimal counting these days which works in a very similar linguistic way to counting in English but for certain uses - times and dates being one - it is more "proper" to use the old counting method, mainly for numbers under fifty, though there are lots of exceptions.
The old method has different language for the "teen" numbers and counting beyond that is based (sort of) on counting in twenties, with the Roman-esque (informal?) variant in some areas of using subtractive counting for "9" - "deugain namyn un" translates as "two twenties less one", for example, which is quite a long winded way of saying "thirty nine" but has fewer syllables than the alternative old-counting phrase of "pedwar ar bumtheg ar ugain", or "four on fifteen on twenty"!
I'm told that hill shepherds in Cumbria use a similar counting method too.
M.
If you want real dialect weirdness, try the Kingdom of Fife. Not a county, a Kingdom. Goodness knows when Fife last had a king. Anyways, Fife is so full of nutty phrases, even their Scots neighbours are puzzled, so you can imagine my difficulties as an Englishman. A woman asked me "are ye courting this weather?" I eventually found out she was asking if I had a girlfriend. Perhaps she was trying to chat me up.
I used to date a girl from Fife. Once, I met her grandmother- a woman so old she probably remembered Fife's last king. And she had a proper old-school Fife accent.
In the hour or so I was there, with those two nattering away constantly, the only words I understood were 'hello' and 'cheery-bye'.
In Cumbria it's something like yan, tan, tether, mether. It's basically "Cumbric" (used up until 13th centuary IIRC) which is related to Welsh as both are Brythonic languages. Pre Angles/Saxons/Jutes all of Britain spoke Brythonic (plus a bit of Gaelic from over the water) and where I live there are still names of hills and rivers with a Brythonic origin. For Example the River Tame ... where Tame essentially means river.
As an aside, where I live we are at maximum a couple of miles from Derbyshire, Cheshire, Lancs & Yorks which throws up some interesting use of mixed dialect and accents.
A free virtual beer for anyone who can tell me the difference between an "Entry" & a "Ginnel".
Ginnel is what others refer to as an alleyway. It's interesting that you've brought up that particular word because we were having a discussion about the regional dialects and their different names for things. Ginnel/snicket/alleyway was one of the terminologies mentioned. Entry however is new to me, although I suspect it's related. :-)
The way I learned it, an entry is a passageway (often covered) into a "yard", kind of a small private(ish) cul-de-sac with several buildings/dwellings all facing into it. A ginnel is a short footpath which connects two streets.
Presumably a ginnel passing through a yard would be comprised of two entries and the yard? Or is that a "close"? I haven't even thought about this stuff in decades :-)
Up our end, the entry runs between two terraced houses and leads to the back door (hence, I think, entry), and is only used by the residents of the two houses in question, and callers, whereas the ginnel is what others might call an alleyway, a path running at the back of a row or two of terraced houses, allowing access to the back garden, or a short cut to the next road.
Funny how "up our end" translates exactly to "down our way" over most of the English speaking world. I've also heard "down our end" and "up our way" ...
Here in California, an alley(way) is an often unnamed secondary street between two rows of buildings on named streets. See this map, which has a bunch of unnamed alleys running north-south. Note the rather unusual named alley "South Harold" between South Harold Street and Park Street. In the above map, the alleys were originally were unpaved, but most are paved these days (the alley that would be called North Harold (but isn't) is unpaved, while South Harold is paved).
Alleys here were built as back entrances, where you put away the horses, for deliveries, garbage pickup, etc. Later, they were dug up for municipal plumbing. Today, the lots have been split (no need for horse barns and paddocks, much less the family hogs & chickens and veggie garden), and houses have been built in the back making the alleys into actual streets.
An alley is pretty much the same concept as a backstreet, as in this map of Harrogate. See the unnamed street between Dragon Avenue and Dragon Parade. (Yes, kiddies, that's Mornington Crescent to the North East. According to the Acadian Rules of '23 (as I read them), this mention-in-passing in no way indicates the end of the road for this round.)
To say nothing of the rivers Avon, Ouse and Esk.
Or Pendle Hill... "Pen" means hill. Pen Hill became Pen 'il, became Pendle ... Which became Pendle Hill, because, well, because it's British. So basically, we now have the rather imaginatively named Hill Hill Hill.
Somewhat sadly, Torpenhow Hill doesn't actually exist.
Shortish version...
The HP printer had been down for days, screams to HP, who screamed at our service department while waiting for the power supply to turn up before we get scheduled to go to site.
This is a two assembly replacement, requiring a fairly awkward tear down. One part finally turned up, the other being shipped by plane to be collected that evening.
The order from on high "Go to site" & the first thing I did was plug my DVM into the power cable & then the wall power socket itself, check circuit breakers, you'll have to get a sparky in & by way of demonstrating, used a extension lead they had kicking around to plug the thing into a nearby office, where it sprung into life.
"plug my DVM into the power cable & then the wall power socket itself"
Even that doesn't always work. A while ago the guys in the workshop complained that the electric water heater didn't work and could I come and remove it for return under warranty. I started by checking that it was plugged in and used a two-pole tester to confirm that the socket outlet was live.
Disconnected the heater from the water supply, awkward due to the narrow kitchen cabinet. Removed the unit, capped the pipes, etc. Decided to be a good handyman and use the vac to clean the cabinet. Plug the vac into that outlet - doesn't work. Odd, tester shows it's live.
Trip the circuit breaker, take the cover off the outlet and discover that a builder had decided to rerun my cabling but didn't reconnect it properly. One wire was not fixed in the terminal but just pressing against it. High resistance: the outlet will light up a tester, but can't provide enough current to run the heater or the vac. Needless to say the innards of the outlet were rather scorched. Replaced outlet, reconnected water heater. Told them not to work with the contractor again.
Sometimes its the other way around too, thinking back on Mr D. Edmunds* tale to his students of a dead TV that he couldn't find a fault with & testing the socket which was live with the expected 240V.
It was eventually fixed with a piece of sandpaper to clean up the power socket connections (& possibly the plug) as the brass had got coated in something over time the only way a Voltage was registered was the probes breaking through the surface muck acting as a insulator to make contact with the conductors themselves
*Not the musician with Nick Lowe\Rockpile, but rather my Analogue Electronics C&G 224 tutor & not the one who blew up anything electronic due to static either.
One quite common brand of budget UK socket (LAP) has a failure mode where it will test ok with probes, and actually work fine if the plug isn't all the way in, but will cut off the current (usually on the neutral for some reason) when the plug is fully inserted. Have had a few of them lately
That's interesting - especially as LAP is the cheap-brand sold by the huge outlets B&Q and Screwfix.
Rebuilding our house I managed to squeeze the few extra pennies into the budget required to specify MK fittings, rather than "cheapest available". By this count I'm glad I did as otherwise I'd have been buying LAP or BG.
Nevertheless we had saved a few of the newer fittings from the old house and I had a few of those fail due to high resistance contacts. Probably just dirt and dust on the contacts from having been stored in a garden shed for eighteen months, but not worth the hassle of dismantling them and getting the fine emery paper out. Fortunately I had spotted them on my own testing (I used to be Part-P registered and still have the kit, albeit out of calibration) before the "real" electrician turned up and caused problems.
M.
Maybe not, but their low-end stuff isn't the price it used to be, either. The very fact that I could afford to use MK instead of LAP or BG was something of a revelation, and although build quality may not be as rock solid as it once was, thought has still gone into the design. On the whole the MK kit I've fitted has had better-placed, more robust terminals than the cheaper brands, switches which "click" in a reassuring way and such "niceties" as terminal screws that are "backed off" ready for the cable and accessible from one direction with a screwdriver, rather than some from the top, one from the side and one from the bottom as is often the problem with things like Switched Fuse units.
And it's still difficult to get anything else which quite matches the Grid Plus modular range for flexibility. Expensive, but a life-saver when you absolutely must have a three-gang light switch composed of two two-ways and an intermediate :-)
M.
I do have a few Schneider items - perfectly serviceable - but I'm looking for some specific Legrand parts at the moment (Electrak sockets and plugs), but only CEF and RS seem to stock them, neither of which is particularly value-for-money :-(
M.
HinH: I've just discovered exactly a similar issue.
A bit ago my kitchen fluorescent light stopped coming on. It would occasionally flash, but fail to start. Changed the starter, changed the tube, still doesn't work. Ok, need to change the fitting.
Yesterday, got the step ladder out. Decided it would be a good idea to test the wiring before the full fiddly of taking the fitting out, so disconnected the wires and put into a chockblock with a lamp connected. Turn on wall switch. Lamp flashes and goes out. Odd.... Wires lead and plug to existing fitting and plug into wall socket. Light fitting works perfectly. Odd...
Remove wall switch. One of the wires sprongs out. Ah ha!
The important information is that the test lamp is a fluorescent bulb. With the loose connection the switch was able to arc over for the high current to kick start the lamp, but the low current to keep it running wouldn't arc over. Which also explains the tube's symptoms.
So, today's job is to turn the power to the lights off while it's daylight and replace the wall switch.
Indeed. Miscreants at work in our esteemed organ (AKA The Reg)
Maybe now is the time to revolt and launch an uprising against bad punchlines and janitorial incompetence. Scooby Doo and Shaggy could unmask a janitor at twenty paces.
On the other hand, what would we do without El Reg, it brightens my day, it's the only thing that does!
Toodle Pip,
Mark.
I had moved to server support by the time I got a gig at a London Borough. The best story was a councillor who also had a title. They were very pushy and demanding. They also had a habit of losing phones before graduating to crackberries which they were equally adept at losing. The end came with a daily fail moment while one over the eight at the wheel of a car.
"Quite a few of them of them are total halfwits"
See also: (supposedly smart, well book-smart, at least) academics. The more letters they have after their name, the less clue they have (and often patience in similar proportion).
(I should add that there are always some decent exceptions to that rule.)
This was all too common "in the day". So much so I almost didn't bother to read the article.
Running support for a small software/hardware house I learnt very quickly that you don't just ask them to check the cable is plugged in correctly, you get them to unplug it and then plug it back in firmly. One user who should have been intelligent enough to know better didn't heed the instruction but confirmed they'd done so. Cue call to 3rd party hardware support, and the engineer finds the plug into the PC hanging half-out of the socket. I don't think the user's immediate boss, a senior consultant for a large insurance company, was very impressed with the near £100 bill (equivalent to around £300 today as it was 30 years ago) for the unnecessary call out.
Another power supply one at the time involved an IBM PS/2 Model 50. They had a known fault that the PSUs would fail if they were switched* off and on again too quickly or if there was a brief interruption in power. Call from user to say she'd gone for coffee, come back to her desk and the computer was dead. Interrogation over the phone didn't elicit an admission of turning-off-and-on or a power failure but again it was a call to hardware support, IBM this time, and the engineer arrives with the replacement PSU**. When the user's boss saw the engineer he confessed, while she was at coffee he'd knocked the cable out of her computer and shoved it back in hoping no one would notice. Fortunately for them IBM didn't charge for the callout.
*I do mean switched, as in a big toggle switch that went clunk!
**Engineers maintaining PS/2s invariably carried spares of these.
I spent quite a lot of time in the mid 90's digging around PS/2 machines (running OS/2) and fortunately quite a lot of time a few years later replacing them with Compaq Deskpros running NT4
They were solidly built and had nice keyboards but thats about all the good stuff that I could say about them.
"Another power supply one at the time involved an IBM PS/2 Model 50. They had a known fault that the PSUs would fail if they were switched* off and on again too quickly or if there was a brief interruption in power."
My first part time job I got roped into helping admin the till system, including performing the weekly failover test. Now I know why the instruction to wait 5 minutes before powering back on was in bold and underlined!
God no! I tried this with my Mum (I'm US, she was from Huddersfield orig... go Terriers!) and it never worked right. The problem is that the programmable remote hasn't a damn clue what the status of the devices actually is. Because for some inane reason the engineers that be decided to make the code to turn on/off be a toggle, not seperate commands. Probably to save a button.
So I've run into this at home and I built a small widget that sits behind the TV waiting for it to go on, and when it goes, it watches the sound bar via a USB port to see if it has power. If not, it sends an IR command to toggle the damn soundbar on, all while watching the power state of the USB port on the sound bar. Magic!
It's amazing what you can do when you have actual sense of the remote system. Those stupid programmable remotes don't, so they never work reliably. Esp for Mums who don't think like engineers. They just want it to work!
Because for some inane reason the engineers that be decided to make the code to turn on/off be a toggle, not seperate commands. Probably to save a button.
Yes, even worse if same applies to input selection..
Even if they want to save a button, that is no excuse to at least have, and recognize, discrete codes so that with a decent programmable remote (like old Marantz for example, you can code it to use discrete codes even if supplied remote doesn't have buttons for it).
Oh god yes. I completely gave up on using any kind of programmable remote because it was impossible to have a sequence that would get to a guaranteed state (e.g. TV and sound bar on).
As someone else has said they could at least have a code for power on, even if the bog standard remote only sends a toggle.
The latest Sky Q remotes have an equally annoying stupidity. The old (original and Sky+) remotes could control the TV as well as the Sky box using the "TV" button. So TV then Power would toggle the power state of the TV whereas Sky then Power would toggle the state of the Sky Box. OK, it was slightly annoying to have to press two buttons (since just pressing power would affect either the TV or Sky depending on what you last did), but at least you could get a guaranteed result each time.
The new Sky Q remotes don't have a "TV" button. Instead, if you want to toggle the Sky power state you just press the Power button but if you want to toggle the TV state you must press and hold the power button for a few seconds and it will eventually toggle the TV state. Leaving aside the fact that this action is non intuitive when you first see the remote (it was some time before I found out about it), it is really annoying even when you do know what it does.
The problem is how long do you have to push the button for? Too short and you will accidentally toggle the Sky state. To long and you stand there like an idiot getting cramp in your finger.
Part of the problem is that different TVs behave differently. Is the TV you are looking at one that has a stand by LED that goes out when it is powering on or is it one that has no LED when off and has one that comes lights up when powered on? I've got 4 TVs in my house and they all have different behaviour and I can never remember which is which. Sometimes you just have to sit there with your finger on the remote until the TV gets through its incredibly long power up sequence and shows a picture.
Of course, you might think that this saves both the space for and cost of a button on the remote. Only it doesn't. That funky "Sky" logo at the top of the remote? Yes, that's a button as well(*) and it has exactly the same effect as the "Home" button. The cost and space would have been better used on a dedicated TV button.
Do the designers of these things ever actually run their designs past real users?
(*) It was a *long* time before I realised that.
My old Denon AV amp has a remote with a separate ON and an OFF button, only one I've ever seen.
If you are programming a universal remote, and get royally pissed off at trying to guess the power status of a device because it just has a toggle switch, then check the codes that are not on the remote. There is often an unlisted, individual power ON and power OFF command.
the most annoying thing with the SkyQ remote is the different behaviours of the up\down channel and the up\down cursor arrows. What total FECKWIT thought it was a good idea for the up\down ch to work in REVERSE to the up\down cursor buttons you use to scroll though the EPG
Or be like my better half who does want to understand that when she presses the power button, it sends a signal to each device, one at a time, aux, tv, then cable, and if she drops the remote after the tv shuts off, the cable box will remain on... She refuses to listen when I tell her that she has to keep the remote pointed at the devices UNTIL THE LIGHTS STOP BLINKING ON THE REMOTE...and its been almost 10 years...
I still have old Marantz RC2000-MkII and RC5000. Admittedly not used either for long time. VM/Sky remotes usually can control enough basic functions on TV so that the real TV remote can stay in the drawer. As for hifi, many receivers have learning remotes.
I did look at the Harmony line, but the apparent requirement to be programmed via mothership over the net put me off. I can live with Setop box + receiver remote combination so didn't bother investigating further into possibility of programming them offline.
Oh yes. I've been sworn at about that. Me "Is the computer turned on at the plug socket". Salesperson "Of course if f***ing is, do you think I'm stupid?" Hmm, better not comment. Wander across to his desk, machine is humming happily away. Me "Is the monitor turned on?" "What's a monitor?" "The TV thing". Again, "Of course if f***ing is." "Is it plugged in?" (Having already looked under desk and seeing that it's not). Silence...
Wrong approach - ask them to switch it off, unplug it, blow on the plug, then plug it back in and switch on in case 'there's a loose wire' or 'sometimes that helps' or 'there's dust in the connections'.
Then they can notice that it's not switched on and fix it without looking like a right numpty.
Your generosity of spirit does you credit. It is all to easy to take the piss, but it is not always the wisest policy.
Having said that, at my work, call-outs are paid for by the customer if it is their fault, and by us if it is our fault, so you actually have to show the customer was an idiot.
Apropos of blinkenlights - back in the days before LEDs were common mains power indication was almost always by neon lamp. Which had an odd property... as they got old, they decided they weren't going to ionise the neon properly without the help of a little daylight (I think they used tiny amounts of tritium to trigger the ionisation).
The net result: lamps that wouldn't work in the dark... you'd walk into a dark equipment room, and the power indicators would be out, or flickering, or in extreme cases actually working properly. Turn the lights on and the neons would all spring to life and sit there looking as if butter wouldn't melt...
This post has been deleted by its author
Re: My favourite: "(I think they used tiny amounts of tritium to trigger the ionisation)"
No, it's the photoelectric effect -- first proposed by Einstein IIRC! The ionisation that produces a neon discharge needs an electron to initiate the avalanche. As the electrodes degrade over time the voltage needed to spontaneously emit an electron increases until none are forthcoming. Shine a light on the electrodes and it stimulates electron emission, the neon springs to life, and self-emits enough photons to keep the light continuing.
I suspect it's both. Certainly I have read datasheets - the GEC book of neons springs to mind, but I'm in the throes of a transcontinental house move and the books are all in storage - which discuss the presence of radioactives to trigger the initial ionisation and maintain a constant operating voltage.
In the days of my youth I played with neons as logic elements, just to see what would happen - oscillators, ring counters and the like. I found them inconsistent and gave it up as a bad job, but it might merely have been old and tired neons. A fine book on the subject can be found here: http://www.tiffe.de/roehren/neon.pdf
You also have to suspect that neon, being the next monatomic inert gas up from helium, shares that gas's fondness for leaking out of containers that can contain more mundane gases such as nitrogen etc. Anything will leak, given long enough, even glass.
Coming soon: since LED technology has advanced to the point where glowing filaments can be created for ordinary light bulbs, I think we can look forward to a day when LED filament Nixie tubes make an appearance for that real retro look.
I looked up noble gases on Wikipedia. Let us assume that atomic radius is a measure of how easy it is for a gas to leak out of a container. I was surprised to find that neon, at radius 38, is not much bigger than helium, at radius 31. This is despite neon being much "bigger" in terms of atomic number: 10 versus 2. Argon, the next noble gas after neon, has atomic radius 71, which is significantly bigger.
I think they used tiny amounts of tritium to trigger the ionisation
I used to make neon sign stuff, and we didn't need radioactives to make them light up, even though the electrodes were a number of feet apart and around several corners, unlike a NE-2 where the electrodes are a couple of mm apart. We didn't need heated electrodes like in a fluorescent tube, which is just an argon flavor of "neon" lamp with a coating that glows, to get them started, either. Though we did introduce a tiny bit of mercury before sealing them up, and use ballasts that provided anywhere from 6KV to 40KV,
I was working on a 200 page Word document on a Mac Plus in Word (end of the 80s). I was going through marking all the words to be indexed... I'd done over 100 pages, when whumpf! The screen went black.
The cleaner was over by the wall and had yanked out the plug, so they could plug-in the vacuum cleaner, GRRRR!
That was when I learnt to hit save every couple of minutes. It has saved me a lot of misery since.
Socket switches seem like a weird idea. I can understand if the socket is controlled from the room entrance, so you can put a lamp there that you can switch on when you come in, but what's the point to put it on the socket? If the appliance itself doesn't have a switch, just unplug it...
I believe this sort of thing is inevitable when you are the first (tribe/group/unit/army/nation...) to introduce something without having the benefit of someone else already having tried it for you. Those typical British separate faucets for hot and cold water made perfect sense those hundred years ago when Britain introduced rules for going about with public water mains. Today they are a relic of a industrious past, but you can't quite get rid of them. Same with those funny German EC cards and many other examples. I, for one, find the socket switches a good idea. My certified electrician apparently does not, he resisted vigorously when I instructed him I wanted my sockets for kitchen appliances switchable (I reside in the metric part of Europe where these are not mandatory).
And your certified electrician was right, it is another (unnecessary) moving part which can fail. If it fails in the off position, it is an annoyance. If it fails in the on position but shows off, it can be dangerous. In continental Europe sockets are expected to be live unless the power to the socket and the group it belongs to was cut, preferably with a lock on the central switch for the group.
> Socket switch failure coming in 3 ... 2 ... 1
Coming, ready or not!!
I've had several extension leads with individual switched sockets fail on.
The only time I've had a wall socket fail was when the switch jammed and wouldn't move at all. Turned out the bloke doing the tiling had let a big blob of plaster drop through the cable cutout and it had set solid in the socket.
It took me ages to find a double pole switch over intermediate MK switch, the 1960s toggle, not the flat style. Just so I could match it to the rest of the house. Embrittlement? No sign of it.
The only switch failures I've seen are on the triple sockets to fit double boxes. Can anyone recommend a good make?
As an electrician, word of warning. Specify MK Logic plus accessories. NOT MK essentials. They're made in china to no standards whatsoever. The mighty MultyKontact (no really that was their name) have fallen :( They don't even make their own distribution boards anymore. But MK Logic plus is still ok.
The rot only started a few years back so any older MK stuff is likely to outlast all of us and the cockroaches.
I'm sure I've mentioned this before, but MK used to run a full page ad in one of the electric trade magazines with the wording "you can tell a good electrician by the mess they make". The picture was of a load of discarded packaging of MK products.
i've got a few in my house. The sockets are decent brushed metal ones and cost a reasonable amount per socket. Sometime the switch won't turn off, the switch just rocks but doesn't fully switch from on to off, I've had it happen on a 2 or 3 different sockets and light switches, same brand, and they aren't that old as the house was renovated 9 years ago! If you give the switch a short sharp smack with something that seems to fix it, doesn't need to be a hard smack just a light tap can do it!
That's typically dust in the switch that causes it. The switch doesn't QUITE make, and the high resistance contact heats the moving contact. That contact is pressed on by a tiny nylon widget in the switch rocker (via a spring). The nylon widget melts slightly and sticks to the metal. Bad cases can cause the switch to completely jam (usually open). Slight cases will cause the symptoms you describe
I love this attitude of "if it's never happened to me then it doesn't exist"!!
I assume you were never in a plane crash either so they don't happen and any such reports are #fakenews ha ha.
Just for your information I had a faulty socket switch only 2-3 months ago on a brand new unit bought locally from a reputable supplier.
I'm nearly 65, have lived with switched 13A sockets every day of my life and have worked in schools' IT support, where sockets have to put up with a lot. I have never, ever, known the switch of a socket to fail in either direction except for one case, where a jammed switch apperared to be, and was, off. Now those poxy 3.19A fuses that some multiway extension leads have is another matter...
Are you also opposed to light switches? After all, you can just remove the bulb.
Remove the bulb? Stuff and nonsense. Just use sunglasses or a blindfold.
Coincidently I had a switch fail about a week ago. But it was a cooker isolator switch not a socket and it failed 'off' when switched off for the first time in probably 10 years.. Cue massive hissy fit as I was denied my Saturday morning croissant.
This post has been deleted by its author
Now that is an issue. European and other countries use two pin connectors that are in most, although not all, cases reversable. That means that if, say you had a bedside light with a switch on the wire and a faulty bulb. You switch off the light and by doing so may make a 50% choice that you have switched off the supply. Then your fingers touch the contacts in the socket. hmmmm!
I firmly believe that we have the best plug and socket system in the world. I also laugh when I hear North Americans talk about our dangerous 230v against their safe 115v. Most disasters in domestic electrical problems are caused by fire. Remember that it is I squared R that matters in the heat produced.
"Then your fingers touch the contacts in the socket."
No they don't. The sockets are normally fully plastic (or ceramic in older or v high power models) on the outside - so normally no live parts you can touch. Live parts are only potentially exposed if you unscrew the lamp - the assumption is that you would unplug the luminaire first.
Are you also opposed to light switches? After all, you can just remove the bulb.
Similarly, I can turn my toaster off by not making toast.
Switched sockets are a concession to the people who believe that electricity can leak out of live sockets. And who unplug things anyway.
Seems like the reason some don't have switches is "there's always one on the outlet, isn't there?"
Being in the US, I think the only two wall-powered things I own that don't have a switch (or wall-wart) are a soldering iron and a really old air compressor. (I did put a dead front plug on the air compressor, to replace the one **with live screws sticking out**.)
"I think the only two wall-powered things I own that don't have a switch (or wall-wart) are a soldering iron"
Throw that antique wood-burner away and get a Weller before you burn your house down.
"and a really old air compressor"
You can purchase and easily retrofit a modern pressure switch assembly with integral main power switch to most ancient electric air compressors. I've seen 'em for under $35 ...but if you plan on keeping the thing, and using it regularly, shell out two or three times that for a quality unit.
throw that antique wood-burner away and get a Weller before you burn your house down.
Nah. Soldering irons (which is what I'd call them when they weigh a pound or so and are a foot long, too big with too much heat for electronic use) don't have switches. Nor do soldering pencils (which is what I'd call the small low-wattage ones similar to a wood-burner), unless maybe they connect to a temperature controller that has a switch. The only real danger is forgetting they're still hot and getting burned. Soldering guns like the Weller or Wen have trigger switches.
My toaster is an old model that doesn't have a switch.
For this, in Schuko-land, there are plugs with an integrated switch. I've recently fitted one of those, with an illuminated rocker. to an immersion heater. Others present in this house are fitted to work lights and such.
Ahem.... Bulbs?!!!! They're called lamps. Bulbs go in your garden. ..... Old electrician's joke. I'll get my coat.
But before I do, UK 13A sockets have switches on for added safety. We like to take it seriously as it's a lethal force ;) They also have integral live and neutral pin/cocket guards, lifted out of the way when the longer earth (ground) pin goes in. Again for added safety. AND you can even buy plastic covers for them so little ones can't try to stick things in the holes.
At least we try. Now where's my coat?
I'm nearly 65, have lived with switched 13A sockets every day of my life
There was a time when the different amperage circuits in a house had different plugs, and that time persisted into the early 60's, so if you remember nothing of that, you must have lived in an ultra modern bijou maisonette. The relevant BS for the modern sockets originated in 1947. Those of us who lived in pre-war or even older housing were privileged to watch a qualified electrician (my Dad) do the rewiring.
BS546 roundpin plugs/sockets are still legal and available. 15A is often used in theatre wiring (having a fuse in the plug is a nightmare if a failing par can takes it out high up in the lighting rigging) and you'll often see 5 or 2 amp ones in restaurants/pubs for the 'mood lighting' table lamps, to prevent customers unplugging them to charge their mobile device.
Yes, I've often wondered about that because surely the only thing which is likely to take out the fuse in a traditional light is the lamp failing, so if you have to get the blasted can down to change the lamp anyway, it's no more difficult to do the fuse at the same time.
I would put a word in here for lamps with DMX dimmers built in, but I've had problems with a couple. One set of (early) RGB PARs had 3x750W lamps behind dichroic filters, each with a local 20mm fuse. Generally speaking the original fuses were glass jobbies which tended to explode whenever a lamp went causing all sorts of havoc inside the device. I retrofitted ceramic ones until we could afford to replace with LED lamps.
I also have some LED spots. Not quite as good as our Halogen Source Fours but reasonable. The main problem is that they seem to "forget" their settings on a regular basis, requiring a visit to the back panel (in my case using a "basket-on-a-stick" access platform), which is tedious to say the least. The thing drops out of DMX mode and instead of just going dark, goes into "auto fade" mode where it fades from dim to bright and back again over the course of about 15 seconds. Bloomin' annoying.
M.
In my "equipment rack" (i.e. stuff on some shelves in the spare bedroom) I've got stuff that's on all the time and stuff that I only turn on when needed. The socket strips I use for this have switched sockets. If they didn't, and I had to plug and unplug kit as required then there'd be an unholy tangle of mains leads and a lot of angry shouting when I needed to plug the A3 printer in.
As for the failing switches - in my half century of life on the planet I've never come across a failed socket switch. My mum turns off every socket in the house every night except for the bedside light and she's never experienced a failed socket to my knowledge. She experiences a very annoyed son every time he visits because when he puts the toaster on in the morning it seems to be working (it's a Dualit, so the mechanical timer whirrs) but it's turned off at the wall. She says that if I went home more often I'd remember.
Yes, he explained it to me in similar terms (not only are the sockets expected to be live at all times, there was, too, no switch available for this purpose, as no manufacturer would test and sell a switch for a type of operation which is not permitted by regulation). He further explained the importance of standard and as-expected behaviour of home electrical appliances. We finally managed to arrive at a legal, workable and overall sensible solution, but it was no easy feat. I wish I had the option to just put some switch-equipped sockets - of course, only where I want them, keeping the rest as it is. Coming to think of it, instead of a kitchen isle I wish I had built a British isle. See icon.
My employer has expressly banned contactless testers, and has specific instructions on verifying the voltmeter is functioning properly (check a known-good outlet!) before performing the test.
And verify again after testing negative to see it didn't fail during the test. No, I am not paranoid (at least about this), I've been shocked too often already by supposedly dead contacts/wires. I've never been shocked (yet) by wires I knew to be live, even though I sometimes work with those when forced by circumstances.
"No, I am not paranoid (at least about this), I've been shocked too often already by supposedly dead contacts/wires."
Hear, hear! I not only check for a voltage between the live and neutral wires, but also check each of those wires against earth and an earthed object such as the wall (just in case the earth connection is interrupted).
The first time that saved me was when I went to replace a socket outlet at a friend's place. No voltage between live and neutral, no voltage between live and earth, but full mains voltage between the supposed neutral and earth. Turned out some !#@! had wired the consumer unit (old, with single pole breakers) incorrectly, with live connected to the neutral house wiring, and neutral to the live wires. So had to start by sorting that out.
The second time was when a circuit at home failed. Breaker switched on, but no voltage between live and neutral. Then decided to measure both live and neutral against earth, and found them both to be live - the breaker in the rather old consumer had gone bad in a way it ain't supposed to. That's when I decided it was time to replace the whole consumer unit (which here in NL you can do yourself, but in the UK would be a job for a qualified electrician (who never make mistakes, oh no they don't)).
It's almost that time of the week again -->
This post has been deleted by its author
"but in the UK would be a job for a qualified electrician (who never make mistakes, oh no they don't))."
When I installed storage heaters in the old cottage we bought, I had to get an electrician from the power company to check my work, certify it, and then connect it to the incomer. First electrician comes along and says "You must have a separate Earth Leakage Trip (ELT) for the heater circuit, as well as the original one for the rest of the house", and leaves. I obtain and fit said ELT. Second electrician turns up, OKs my work, but insists that the Protective Multiple Earth (PME) must be connected directly to the earth rod outside the house. He will not listen to my plea that such a connection would render both ELTs inoperative, as any fault current would bypass them down the thick earth cable. He waits while I run an extra earth wire, then presses the yellow test button to prove that the new ELT is working. He then signs it off and leaves. I then whipped out the wire cutters and removed the offending earth wire, before checking both ELTs were working by connecting a resistor from Live to earth in a spare plug to inject 30mA down each earth wire.
I had a qualified electrician install a new fuse box with half on an RCD.
Whenever anything was plugged into the RCD side it would trip out, the electrician blamed leakage in the house wiring, as it was old I took at face value, and he moved most of the circuits over to the non-RCD side.
Many years later I needed to install something that required RCD, so this really started bugging me. I put the new circuit on the RCD side and it tripped out immediately on first load. I then purchased a megger and checked the other circuits out, OK there was leakage but only microamps @240V. Something fundamental was wrong here.
After a bit of head scratching and inspection of the fuse box I found the problem: the original qualified electrician had installed the sense wires for the RCD the wrong way around - so it would trip on any load at all! Sense wires back in the right way, all was well, and mains circuits could now be RCD'ed.
Lesson leaned: a degree in Electronic Engineering trumps a HNC in Electrical Engineering...
As an ex-qualified electrician (gave up my Part P when I got a salaried job) I was recently called out to look at the wiring in a friend-of-a-friend's house, where the thing had suddenly started tripping.
Said f-o-a-f has been in the house maybe two years and had a full survey done before purchase. You know, the sort that usually flags up any little thing as a potential problem but doesn't offer any solutions other than "consult a qualified <name of trade> person". The sort which often causes the mortgage company to withhold silly amounts of the mortgage until you get things sorted (my wife once had £1,000 withheld - around 1998 - until she had fixed the guttering on the garage, which was at the end of a 60ft garden. There was no drainage, so she bought a length of gutter, a downpipe and a water butt, which cheerfully overflowed regularly in a not dissimilar way to the previous situation, but that was sufficient for the surveyor to sign it off).
Anyway, back to the plot, there was (I am told) absolutely nothing flagged up on said survey before f-o-a-f signed on the dotted line.
First thing I spot in the "refurbished" kitchen is a socket held on by one loose M3.5 and a woodscrew, which was itself not really doing anything. No grommets either. F-o-a-f says "I never use that socket".
Cooker turns out to be wired in Aluminium, so that dates it. PVC is brittle and falling off.
Breaker labelled "shower" is actually the cooker.
"Cooker" was (presumably) the shower, but the shower has recently been removed (thank goodness) because it was on the non-RCD side of the board.
Oh, did I mention that this is a house with an earth rod? TT is still very common around here.
There were several other minor issues but on to the biggie.
Yup. A split board with earth rod and several circuits not RCD-protected at all. Obvious the moment I clapped eyes on the board, so whoever it was did the "electrical" part of the buyer's survey obviously didn't even look at it.
F-o-a-f really isn't in a position to have the place rewired at the moment and follow-up by a "real" electrician (from the pub) sorted the original tripping (probably just damp) without mentioning anything else. Both friend and friend-of-a-friend are worriers and I have tried to point out that the thing needs a proper "seeing-to" without scaremongering, but now that everything's working again I suspect nothing's going to be done...
M.
I thought the most recent ELCBs are the current sensing type (RCCD) which monitor the difference between current in the live wire and the neutral wire and don't care what the earth is doing. Any measured difference in current is assumed to represent a leakage to earth.
I guess from your description that you were using voltage sensing ELCBs which must monitor the earth connection. Are they still used? Perhaps someone with more knowledge can set me straight.
ELCBs were common in the UK up until the 1970s or 80s when RCDs became easily available and I was still seeing a fair number of them in the mid 2000s when I was working as a domestic electrician as PME has only recently come to large parts around here and lots of houses are still on TT (earth via a local rod - a short circuit to earth is unlikely to cause enough current to flow to blow the fuse, so additional protection is needed). Most of them were Voltage-operated and nearly all of them were fitted incorrectly.
A Voltage-operated device uses two earth rods which must be placed some distance apart. One rod is connected to the protective earth in the usual way, all the extraneous conductive parts and one terminal of the ELCB. The other is only connected to the ELCB, which detects an earth fault by sensing a difference in Voltage between its two earth connections.
PME (or to give it its technical name, TN-C-S) where the installation's earth is connected directly to the incoming neutral at the service cutout has the major disadvantage that a fault on the neutral before the installation can leave the installation "live" but not working and (crucially) with no effective earth - and anything connected to the protective conductor probably at mains voltage.
For this reason, rather than just being earthed at the substation, the neutral of a PME system is earthed at multiple points along its journey to your cutout and there is an argument that if the distance between the last earthing point and your cutout is sufficient, or if the cables are vulnerable in any way, then a local earth should be provided - but in the event of a fault which disconnects the external neutral, your earth becomes the neutral and unless your earth is very good this will cause all sorts of dangers. Fifty years ago, PME was still relatively rare and it's possible that the engineer didn't fully understand things.
PME these days is rarely offered where the cables are separate (as used to be common with overhead cables) and almost exclusively uses co-axial cable where it's almost impossible (without doing it deliberately) to break the neutral without also breaking the live and probably shorting the two together in the process thus taking out the fuse at the transformer.
M.
Had to replace the main oven element in our cooker last week. First, turn off power at wall switch, then off at main breaker, and finally test with field tester. Reach inside oven to remove four screws (to take out back panel), one screw to loosen element and gain access to power leads. Test again with multi before touching leads. Total job takes about 15 minutes from switching off to back on.
SWMBO is brilliant cook and baker, so our cooker gets well used. 15 years old and only second time I’ve had to do a repair (and it was the same element about 5 years ago). Previous one was just £8 off *bay - maybe why it only lasted 5 years. Replacement only £12 this time (and labelled as OEM - might even be true) so not really a hardship.
You're lucky
First failure - Christmas morning, no less. SWMBO had just put dinner in the oven. We always have as guests her sister & husband who live a few hundred yards away so the whole lot was quickly whisked away to their oven while I went online to order two replacements.
A few years later the next element went in a shower of sparks in the run up to Christmas. Just as well I had the spare; fit it and order another.
Another year the fan seized up just before Christmas.
Sigh. This year is going to be quiet....
Wow, that was a flashback I want to forget. Oven fails on the day before Christmas eve. Local "trade" people wanted £200 just for the call out with no guarantee of fixing it. Managed to diagnose a fan and sensor failure myself, and source a part from a place that stayed open on Christmas eve so I could pick it up. Electrical, gas and water "engineers" - tossers the lot of them, learn how to do it yourself. And don't believe the Gas Safe, and other bullshit that claims you cannot work on your own appliances.
Standard safe isolation/testing procedure. Good to see someone actually using it :)
I keep a proving unit (basically a HV inverter with built in batteries) in my toolbag along with 3 different 2 wire testers (multimeter, modern LED one, and an old Drummond filament test lamp.) The latter lamp is useful for quickly identifying a circuit, run from L to E and go down and see what RCBO has tripped...
"verifying the voltmeter is functioning properly (check a known-good outlet!) before performing the test."
If I'd have done that as a young electrician I wouldn't have blown the arse out of a perfectly good AVO when I selected A instead of V and tested across two 440V 3 phase fuses. I still have them varnished in a box somewhere. They were quite impressive, as was the flash!
Over here in the land of the treasonous colonials they have the top and bottom socket in an outlet fed from different circuits. So you blow a fuse, carefully check that the socket is dead (using a known good lamp) take the cover off and get sock from the other pair of still live wires.
Fortunately their electricity is, like their beer, pitifully weak.
"Over here in the land of the treasonous colonials they have the top and bottom socket in an outlet fed from different circuits. So you blow a fuse, carefully check that the socket is dead (using a known good lamp) take the cover off and get sock from the other pair of still live wires."
That can't be to code, can it?
Should be 1 outlet = 1 circuit, surely?
"That can't be to code, can it?"
It is code in some (not all!) places this side of the pond. The socket pair is supposed to be installed upside down to indicate there is something "non standard" about it. Often one of the pair is always on, for a TV or the like, and the other is connected to a switch next to a door, for a lamp. They are always supposed to be on the same leg, even if on separate breakers, so there is no chance of managing 240V at that point no matter how much Murphy would like to make it so.
For the record, none of the sockets on this property are wired that way. It sounds dodgy to me, and I want nothing to do with it.
In the very rare case where a device fails and does cause a fire being able to turn off the current at the wall is very useful.
Back in the late 80's we had a number of small fires caused by people leaving paper on top of monitors, eventually the monitor would overheat and catch fire which would then ignite the paper above.
Users would be advised to turn off the socket and monitor the situation. the fire usually went out. In this particular case the sockets were mounted above the desk and she would have had to put her hand through the flames to turn it off.
She was told to evacuate the building and call the fire brigade.
The news spread through the organisation and it finally seemed to get through to users that storing papers on top of monitors was a very bad idea. It was also helped that the design department for a local engineering company was also actually burnt down after a similar incident. Quite why early CRT monitor manufacturers made the top of the monitor so inviting to be used as a shelf is beyond me.
Having read some of the other posts i can see some advantages for convenience of a switch,
..but for me , if the monitor is ON FIRE , i would consider it one of those "go the extra mile" situations and unplug it, whether theres a handy switch or not.
When my sister was a student operator on an IBM 1620 at San Diego State College, they were told that if smoke was coming out of the system, hit the normal power switch (which did a phased shutdown). If there were flames coming out, *then* use the emergency power switch (which cut power immediately) as there wasn't going to be anything to save anyway.
Having been in exactly this situation, I can report that my first response was the 'off' switch followed by unplugging. There were multiple sockets with things plugged in and it was quicker to switch everything off than pulling all plugs. (Didn't even contemplate trying to figure out which plug was the correct one.)
I built a shelf that sat on top of the--solid--bezel of a monitor, but had extensions holding it up off the rear part of the case where the ventilation holes were. Had nothing to do with trying to store papers there. It was because of the cat that liked to lie down on top of the (warm) monitor with his claws hooked over the edge of the bezel.
I thought the heart being on the left was a myth?
In most people (and other vertebrate animals for that matter), it is slightly left of center, but it isn't as far to the left as normally shown in movies (i.e. heart target for sniper). And there are exceptions (in humans as well as other vertebrate animals) where the heart is slightly right of center. If the intestines and abdominal organs are also mirrored (stomach, spleen and pancreas right, liver left), there normally are no problems, but in the rare cases either the heart (and lungs) are mirrored or the abdominal contents are mirrored, there usually are.
I remember being taught how to follow a wall\navigating to an exit* during an oil rig fire by using the back of the hand against it, just in case you found yourself encountering exposed wires as the hand will clench onto them if you use the palm.
My next course was on how to put big fires out on a oil rig.
*Smoke filled room, visibility near zero & wearing full BA.
Icon - Big heavy firefighters coat.
Been there, done that. It was the exercise crawling through the smoke (real fire, not “disco smoke”) without BA that really got me - reeked of wood smoke and coughing the rest of the day. But always back of hand first. Get a kick this way and you’ll appreciate why it’s taught.
Another reason to use the back of you hand in a fire, is if touch some hot metal and you use the palm of your hand, you can lose the use of that hand from being badly burned. If you use the back of your hand, you can still use your hand as pretty much normal, till you can out and get medical attention.
I thought the heart being on the left was a myth?
With most people it's on the left, but one classmate from secondary school was exempted from military service when the medical examination found[0] he was mirrored, with his liver and spleen being on the right as well. Can't remember if he was left-handed, although that would fully make sense, in his case.
[0] of course he, and his GP doctor knew about this, but the army medicos ignored the supplied documentation to that effect.
I think Spike Milligan's war memoirs included the story where the soldiers discover they've been given defective soap because they are ordered into the sea to wash with it and they can't. All the soap that they can lay hands on has the same fault and it is all ordered returned to the depot.
Because "common soap" made with sodium doesn't work in salt water. You can get soap with potassium, though.
https://en.wikipedia.org/wiki/Saltwater_soap
"the first (tribe/group/unit/army/nation...) to introduce something without having the benefit of someone else already having tried it for you."
The punchline to the joke of course being that the Brits weren't actually first, did actually have the benefit of learning from the efforts of others, had two attempts at it and still managed a mediocre result no better than other's first efforts at standardisation.
The schuko plugs used today across most of Europe with no real issues were introduced in the early thirties (after half a century of electrification), while the huge clunky british three-square-pin things with the overcomplications of fuses, switches etc. were standardised in 1947 as a complete clean-sweep modernisation of the round-pin arrangement they previously standardised on between the wars which in turn had codified the various manufacturer-specific type that had been around since the 1880s.
The current US plugs and sockets were also defined in the 1940s, and they similarly felt no need for switches, in-plug fuses etc. even in the 250V versions.
Quote: "The punchline to the joke of course being that the Brits weren't actually first"
Erm, yes they were, the first public electricity supply in the World was in 1881 in Godalming, a small town in England, which used a water wheel.
The 2nd public electricity supply, (and the first large scale implementation in the World), was at Holborn Viaduct in London in Jan 1882, which was also the Worlds first coal power plant. Granted this was built by Edison Electric Light Company. As in Thomas Edison, who was of course from the US.
He used this London installation as a proof-of-concept for the worlds 2nd (and larger) coal power plant, which he built in New York in September 1882.
The UK plug was designed from the start to be safe - plastic shutters that cover the live and neutral that only open when the earth pin is inserted, the live and neutral pins have insulation on them so even when the plug is half out you're not touching bare metal, the cable is at right angles so you can't plug the plug out by the cable and if you do pull the cable, the wires inside the plug will disconnect from the live terminal first.
While the Europeans have a recessed socket, it is still possible to poke inside the holes.
By comparison, the US version was designed by the cheapest bidder!
"plastic shutters that cover the live and neutral that only open when the earth pin is inserted"
I think modern Schuko sockets also have shutters, which will only open if both pins are inserted. They are not operated by the rim earth contact as the flat Europlugs for double-insulated equipment don't have an earth contact.
Unfortunately, the UK design choice means that plugs for double-insulated equipment are just as bulky (and unpleasant to step on in the dark) as plugs for earthed equipment.
Some extension cords have shutters. If you have kids then you can buy little plastic inserts to cover the plughole, they open by twisting. Dutch sockets are wired with 2,5 mm2 copper, wires must be in a tube in the wall. The UK method of plastering wires directly in the wall makes us cringe.
No its to protect the house from the flex to the plugged in device. The socket is 13A but the ringmain is fused at 30A or MCBed at 32A. If your low powered device with say 1mm2 flex develops a fault and it isn't protected by a 3A plugtop fuse, then the flex gets hot enough to burn. Likewise all other flex sizes should have the correct size fuse up to 13A. Since the ringmain itself has 2 paths back to the distribution point the house wiring doesn't even get warm unless you are doing something (hopefully) unintentended like simultaneously drawing 3KW from 2 other sockets on the same ring at the same time as your device has a fault. Even then it should only get warm not hot enough to start a fire.
MeDearOldMum was horrified when she found out[0] that the hot water side of the first house we lived in in England was connected to an open cistern in the attic. The attic was used as roosting space by some local pigeons! Needless to say, we didn't stay there for very long, maybe a week total.
I still can't believe how many of you lot still live in places with who knows what sewage and dead rodents/birds floating around in your household hot water (search YouTube if you don't believe me) ... and you shower & do dishes in that?
[0] I was a curious kid, there was a ladder in the garage ...
1. You cover the header tank.
2. The combi-boiler has largely made this obsolete, for better or worse.
The tank may well still be there. It may have been put into the loft before the woodwork for the ceiling was completed and there's now no way to remove it. If, like me, you live on a hill close to the limits of the head of water in the mains* you realise that that local head of water would probably improve the shower performance if it were still available.
Then you have to take into account all the Canada geese on the reservoir, townies coming over in summer when the reservoir levels are low and letting their dogs crap below high water mark and similar.
* Any time they try to jack up the pressure at the PRV it causes bursts of the mains lower down.
"Then you have to take into account all the Canada geese on the reservoir, townies coming over in summer when the reservoir levels are low and letting their dogs crap below high water mark and similar."
Reservoir water doesn't go directly into the water main. It goes through a treatment plant first. We don't store potable water in open reservoirs in the UK
Local water treatment plant built in the 1950s (should have been listed as a beautiful example of what mid-C20th municipal architecture could achieve but now sadly demolished) had two wings each with a set of sand filters, one in use whilst the other was being backwashed. There was a double height central hall which would be called an atrium nowadays. In it was a sink with a set of taps, each connected to one of the filters. With each tap turned on the effect of each filter could be seen as the successive streams of water got clearer. As I lived up the road between the reservoir and the filter works I was quite familiar with it.
I think I've got a cold water tank in the loft for the geographical reason that you mention, and a boiler possibly supplied from that. House built in Scotland around 1994, I believe.
In the Highland glens, the water tends to come direct from the territory uphill. From time to time, someone tests it for you in case there is a dead sheep upstream or something.
In the year of coronavirus, we should be aware that we breathe or drink all sorts of potentially offensive adulterations all of the time, it is more a matter of the quantity. A very little can be extremely dangerous, but less than that is not particularly dangerous.
hot water side of the first house we lived in in England was connected to an open cistern in the attic
Allegedly the fault of Napoleon. During the napoleonic wars there was concern that the dastardly French might be sent to sneak in and poison the water supplies, so all houses were required to maintain a store of water. Over time that got incorporated into plumbing codes.
I don't know the actual reason but it makes it much safer to plug the iron in when it doesn't have a plug on it, just an inch or so of bare wire at the end of the cable. You use a plastic tail comb in the earth socket to pop the shutters then shove the wires in the live and neutral sockets. Pull out the comb and the shutters hold the wires in place. That's what my mum used to do when I was a kid. I've no idea why the iron didn't have a plug on it. The hairdryer had a plug which plugged into a ceiling light socket.
I remember my father's trick to be to put the bare wires over the relevant hole and plugging something else in on top to release the earth and hold the wires in place. Mainly done when he was repairing/testing some electronical type device and couldn't be bothered to wire a plug.This was back in the days when new stuff often didn't come with a plug attached.
The scary bit, though not to me at the time as it was just 'normal', was that he was fully qualified in electrical/electronic engineering, and worked at the local nuclear power station. Mind you he did have a reputation locally as being able to fix anything electrical or electronic. You never asked how, or looked to see what he had done, just be safe in the knowledge it now did what was expected of it.
Have to be careful with that though. In my yoof I worked in a small TV/Computer repair shop, and we used to regularly do that with kit on display. Until the one day the manager did it and hadn't properly ensured there were no stray strands of copper floating about.
Cue instant flashover and fusing of all the electrics in the storefront (and a very loud bang and shocked manager)
Indeed, the lower US voltage is not only more dangerous because of the higher current, which means more heating in conductors, so they need to be accordingly heftier to avoid failure, but also means transmission losses are greater. There's a reason that overhead power cables are high voltage AC, because the transmission losses are proportional to the current, and thus inversely proportional to the voltage.
Er..."Lower voltage means higher current" is nonsense I'm afraid. Stick your finger in a 120V socket and you'll get Y milliamps flowing to earth. Stick same finger in a 240V socket and you'll get Y x 2 milliamps flowing to earth.
Lower voltage means you'll need to pass a higher current to get the same power output. You'll get that with a lower load resistance. Therefore you'll need larger, more expensive conductors in your 120V installation. A 3KW kettle will pull 12.5A on 240V (hence a UK socket is rated at 13A), whereas it'll pull 25A on 120V, requiring larger conductors and also greater transmission losses as heat is I²R. Hence why power is transmitted across country at such high voltages, to keep the current and therefore losses AND cost of infrastructure down.
My UK 3kW kettle (230V, 13A) is a different beast to a 3kW US kettle.
I'm assuming that left-pondians do have high-amperage kettles: I can't imagine them being happy to wait twice as long to boil water, though maybe they do, and that's why they're culturally unable to make tea properly.
"I can't imagine them being happy to wait twice as long to boil water, though maybe they do, and that's why they're culturally unable to make tea properly."
Unlike you lot, we don't have to start with cold water. We have actual hot water coming out of the hot tap, not hot sewage. So in actual fact, our tap water boils a trifle quicker, despite the lower power of a typical kettle.
"augmented with bits of sacrificial anode"
The chemistry of my water coupled with the short length of time that my hot water spends in the system makes this a non-issue.
"and possibly legionella."
My hot water is set to "adult", not "nanny state". Legionella can't survive temps over 70F (taps that are accessible to children and visitors from the UK have tempering valves that are set for the duration of their stay).
"you must use freshly boiled water"
That's what "they" say (whoever "they" are). But have you tried a side-by-side tasting for yourself? I have. Tea made with water from my hot tap vs tea made with water from my cold tap. Both brought fully to the boil, immediately poured over the tea in pre-heated identical pots, tea measured out by weight, not volume ... Out of a dozen tasters (half Brit, half Yank, all tea drinkers), none found a discernible difference in flavo(u)r.
So why does the kettle pull more current on a lower voltage?
It's designed for a given power rating. To get 2kw on 110 volts it has to draw more current than a 2kw kettle for 230v.
Plug your US 110v kettle into a UK socket & it'll draw 4x the current, and make your tea very quickly (but nor for very long).
And like others, what I've heard and you may have seen is that U.S. households do not go in for the "electric kettle" as such, anyway. A kettle on a stove, maybe. And after certain historic troublemaking in that part of the world, the hot cup of tea is not their thing, either.
@P .V. Jeltz (nice username, how is the poetry coming along)
A kettle is designed to run at a particular voltage, if a 120V kettle is to have the same power as a 240V kettle it must be designed to present a quarter the resistance and draw twice the current.
If you plug a kettle designed for 240V into 120V it will draw half the current and have a quarter of the power because P = I x V and your brew up will be delayed, if you plug a 120V kettle into 240 V please warn me first...
PS I've always wondered why we use "I" for current.
Hmmm. Really to be consistent, current should use the letter A(mpere). V(olt) is for potential difference. W(att) for power. Resistance should be O(hm), but the Standards Committee no doubt interpreted the suggestion of Oh-um as indecisiveness, so used R instead. And as Jake said Coulomb is for charge.
Because I is used for current, the knock-on effect is that electrical engineers use the letter j instead of i when dealing with complex numbers.
"A 3KW kettle will pull 12.5A on 240V (hence a UK socket is rated at 13A), whereas it'll pull 25A on 120V", "
If that was the case then you could not plug a kettle into a 15 amp circuit and expect it to boil. House circuits are typically 15 amps. Higher amperage circuits are reserved for certain appliances such as electric stoves (cookers) and clothes dryers which require different outlets.
Here in the US, household lighting circuits are typically 15A. Kitchen, garage and bathroom circuits are typically 20A. Specialty circuits can be anything from 120V 5A (quite rare these days) to 240V 50A (common). 240V is achieved by combining two different legs of 120V. Main panels on new construction are typically 200A or more, older houses run around 125A, some very old homes are still stuck at 60A or thereabouts.
Higher power capability, including 3-phase, is also available if you need it, but it might cost more than it's worth to get it run to your home. When I first started restoring big iron, it was much cheaper to rent a small commercial space for a couple years for the power than to have it run to my house.
The above is obviously just a rough overview. Variations exist from state to state, county to county and even city to city in some cases. As always, check your local code before ordering parts.
This post has been deleted by its author
> That's nonsense, I'm afraid. Lower voltage means higher current, which is worse.
It is true that current is lethal, rather than voltage. However, a higher voltage overcomes skin resistance more easily, so is more dangerous if there is sufficient current capability, which is just a few amps, if I recall. Put it this way, I would not trust a fuse to protect me from electric shock.
I was always told it's because 240V will kill you easier than 120V.
I am also reminded of a certain American inventor* electrocuting an elephant with AC to "prove" it was more dangerous than DC.
*I use the word "inventor" advisedly here, as the individual in question appears to have made most of his money by patenting other people's inventions.
While Edison did round up animals and electrocute them to "show" how much more dangerous AC supposedly was than DC, he may or may not have been involved in the elephant incident.
https://science.howstuffworks.com/innovation/science-questions/why-did-thomas-edison-electrocute-elephant.htm
When my father was working in the US, developing equipment for a medical research team at the University of Louisville, he caused consternation, nay, panic, amongst several of the research staff (who all had Doctorates), by checking if the supply was live by touching two fingers to the ends of the wires. "It's only 110 Volts" he exclaimed, but the researchers insisted he go down to A&E for a checkup as he had just had "a near death experience".
Screwdriver/testers that I trust[0] to work as expected are few and far between, and aren't ever purchased at one's common hardware store. Apart from that, the screws involved are almost always PZ/SL now, and the common size 2 slot blade that those screwdriver/testers offer tend to not fit those very well.
[0] and still test to verify they do.
It's the volts and the amps and the frequency and the duration that kill you.
The only thing you have to remember is that mains electricity anywhere on the planet has enough power to kill a human and does not care about the concept of nationality. Treat it with respect and understanding or die.
"I don't know the actual reason but it makes it much safer to plug the iron in when it doesn't have a plug on it, just an inch or so of bare wire at the end of the cable. You use a plastic tail comb in the earth socket to pop the shutters then shove the wires in the live and neutral sockets. Pull out the comb and the shutters hold the wires in place."
Anyone else here getting flashbacks to the old Public Information Films, especially the one showing the guy using an electric drill "plugged" in using matchsticks and subsequently getting electrocuted as the earth wire pulls loose and contacts the live?
The British kept this up a bit longer than most other countries.
The older UK BS546 (round pin) sockets usually had switches, especially for the 15A variant, but the newer UK BS1363 socket outlet dates from the late 1940s and originally didn't specify a switch. The switch was added back to BS1363 in the late 1960s.
I'd guess that switches were originally required as being safer for DC supplies to avoid drawing an arc and burning contacts if the plug was withdrawn while the appliance was switched on, which is less of a problem for AC. Adding them back to BS1363 may have been a safety measure, since many devices available from the 60s onward didn't have a switch on the mains side of the supply.
That "switch the socket off before unplugging" came about because people could get their fingers behind the plug body and make contact with the live prong when trying to pull out a stubborn plug.
These days most live and neutral prongs are partially insulated but not all are. It's still good practice.
One place where I worked (decades ago) the head sysadmin said that the fire safety officer (from the fire brigade) told him that there were more fires caused by worn-out sockets from people unplugging stuff for safety, than from things left plugged in when not needed.
I find that hard to believe, but that line of thinking could lead to providing a switch on the socket.
My wife unplugs the LCD TV - because of that bloody 70s TV ad
<sings> the TV set should be unplugged last thing every night.....
Which means 5mins of it booting and trying to get updates everytime you want to watch it.
She also unplugs it during thunderstorms - we don't have an aerial, we don't even have cable - how she thinks lightning is going to go over the wifi while watching Netflix I have no idea.
What;s really worrying is that she's not an idiot - she's a doctor !
Had a modem commit suicide, thanks to a lightening strike in the nearby pub car park (Not sure what it hit).
I saw a strike hit a pylon on the A13 going into London, quite a few sparks the the supporting metalwork above the insulator glowed red for quite a few seconds until it cooled down.
We had the Vermin media TV box die (with an impressive pop) due to a nearby lightning strike last year. I'm assuming the lightning induced a current in the coaxial cable supplying it. Curiously, both the TV box and router are connected to the same cable (via a splitter which I'm assuming is just a HF filter of some sort). The router didn't even reset, but the TV box was well and truly cooked; I'm assuming it was an electrolytic cap exploding that destroyed it, it did rattle quite nicely afterwards when shaken!
We never worked out what the lightning actually hit, but it was very close because the flash and (very loud) bang were simultaneous.
The call to the inevitably offshored call centre operative involved explaining that, no, there weren't any lights on it, and no, there wasn't an error code on my TV, because, as I repeatedly explained, it had been STRUCK BY LIGHTNING.
This all happened the day before we were going away for a few weeks, and to be fair to VM, the replacement box did get delivered successfully to a neighbour well before we got back.
We never worked out what the lightning actually hit, but it was very close because the flash and (very loud) bang were simultaneous.
A nearby lightning strike killed my POTS modem[0] and a soundcard. There had been some distant strikes with the rumbling delay at about 15 seconds, so 5-ish km. Then nothing at all for maybe ten minutes, an utterly deafening bang as a strike hit a house right across the street, and nothing more.
Street lights out, POTS out, cable TV out, RCD tripped. Modem was just dead with no visible damage, the sound card PCB was twice as thick around the sockets compared to the rest of the board due to what must have been a ground loop in the external cables.
[0] not even cable internet back then where I lived, so mid 1990s. ISDN a couple of months later, then cable internet, then ADSL
When we lived in the cottage I referred to in my post above, our electricity supply was from a pole transformer across the field. It fed quite a few houses in the village, and nearly all of them had storage heaters, so the voltage went up and down according to the loads imposed on the mains. One stormy evening, we were watching the telly when there was a blinding flash, accompanied by a very loud bang, and everything went dark. It stayed dark all night, and in the morning we could see the charred remains of the transformer dangling from the pole, surrounded by a mare's nest of frayed wiring. Most of the village was without power for nearly a week until a new transformer and cables were installed.
a mate and I were out shooting a few years ago when a thunder storm rolled in, before we could make it out of the middle of the field a MASSIVE bang and a FLASH that was like someone letting off a camera flash about a foot from your face! Must have been really close to us, an added smell of excrement as we both shit ourselves!
I saw the old-style phone of our neighbours' after a non-strike had induced enough current to warm it up a bit. The perspex dial looked as if it had been boiled - it was full of bubbles. Both our electricity and phone supply are underground which gives us some protection. Sometimes I'm still inclined to unplug electronics from the mains if there's thunder about.
We learn the fearful warning stuff in our childhood. Possibly even then obsolete or just plain mythical. And it can get a deep hold that outlives reason.
Things like you must/must not ; put away the cutlery in a thunderstorm, watch TV in the thunderstorm, empty the kettle every night ( apparently it would corrode), not wash hair during a period ( I could never ascertain from the g/f or her sister what the consequences might be). And so on.
> What's really worrying is that she's not an idiot - she's a doctor !
A cancer doctor that I know explained the radiation treatment: "It is like X-rays, but the photons go much faster". Just because you have expertise in one field, does not stop you being completely wrong in other fields. In fact, only intelligent people can excel at total silliness. But don't ask me, I am an engineer, not a philosopher.
To be fair, it's nearly impossible to find anyone qualified to teach critical thinking. Of the myriad teachers I've had, those who actually said the words 'critical thinking' were frequently those least qualified to discuss it.
It's quite a shame that so much education is so narrowly focused. Having a broad base to start from, even if it isn't in-depth to any serious degree, makes application of what you're specifically studying that much better. Including little things like troubleshooting a computer that appears completely dead, per the article.
it's not that the photons are faster, but there are more of them... or something.
The best analogy would be not "faster" but "bigger", as in each photon has more energy packed into it. It actually has a higher frequency, but since you're talking about photons as particles, that inevitably leads to having to answer questions about how a particle has a frequency, which leads down the rabbit hole of EM radiation actually being neither particles or waves in the classical sense, and whoops you just said radiation and now your patient is back to being scared of the radiotherapy...
If you're trying to explain how radiotherapy is different to an x-ray, how would you go about explaining it to the layperson? Technically, the photons used are more energetic, and are more tightly focussed than an X-ray but to Joe Public you might as well be talking Chinese.
We have switched sockets in Australia too, and the standard goes back to the 1930s.
They are useful for
• switching off a device when not in use to avoid "vampire devices" (devices which, while apparently dead, suck a small amount of power throughout the night). Collectively, those devices consume a significant amount of power across a whole country.
• isolating at least the active line when not in use from power surges caused by lightning. It doesn't prevent neutral strikes. (Only the active is switched in Australian power points, IIRC.) This is becoming less important as new equipment detects and responds to strikes more quickly... but plenty of people have home cinema setups worth many thousands that is actually only used a few hours a week.
• reducing aging of the power supply in devices. My Region A Blu-Ray player is not easily replaceable.
• power-cycling a troublesome device, most recently for me a printer.
You get the idea, I hope.
I *have* had one fail. It was a brand new double socket just installed to replace a single, and it failed with the contacts closed but the rocker movable between on and off. When changing a circular saw blade, I don't rely on the switch, but pull the plug right out and place it in my field of view!
"If the appliance itself doesn't have a switch, just unplug it..."
I've always assumed that the function of the plug/socket interface was simply to provide a connection with suitable capability to withstand external forces and that the switch was designed to cope with issues such as arcing on breaking the circuit. Two different designs of connector to deal with two different aspects of connecting an appliance to the supply.
I have actually tripped a circuit breaker by plugging something into a socket at an awkward angle. Contact was rapidly made and broken 2-3 times and the trip switch went. That was the first time I figured switches had a purpose.
Ruined a few people's mornings until the caretaker arrived to open the breaker box.
Switches on faceplates are purely for convenience. The British standard permits unswitched sockets, and they're reasonably common.
I can't say I find a switch an essential feature, but it's handy sometimes. It's usually a lot easier to flip a switch down the back of the sofa (say) than to unplug.
"I can't say I find a switch an essential feature, but it's handy sometimes. It's usually a lot easier to flip a switch down the back of the sofa (say) than to unplug."
Which leads to...just why ARE standard wall power sockets so close to the floor? They always seem to be the about the same low height, where ever you go. Is there some sort of regulation? If so, what's the reasoning behind it? Clearly they can be placed higher, eg kitchens along the work bench, or near dressing tables and similar in hotels.
It's saving money. Socket near floor means shorter cable runs, less chasing. They were often surface mounted to skirting boards too.
Same reason too many 1970s schools have the only socket right by the f***ing classroom door where it's of no use to anyone except possibly the cleaner.
Also, to be fair, the aesthetic of the period was to have the things down out of sight..
In the days before the plugs had insulated bits on them they kinda made sense as a safety measure*, and many older electric things in the UK did not have their own switches IIRC. Electric hair tongs for a start.
I never really thought about the why of it. Some sockets were switched and some weren't. In my parents' house we had five and fifteen amp wiring, so if the previous owner had gotten excited about the new "European pattern**" 13 amp plugs there was also the excitement of a fire risk.
*Of course, if you left them switched on all the time that was bollocks.
** As some called them at the time.
Some homicidal manufacturers make moulded 13A plugs without the ridge that enables you to get a grip on them. When these are inserted into tight sockets it's almost impossible to get them out without curling your fingers around the plug face in disconcerting proximity to the live pin. In this circumstance, a switch is welcome.
There are still plenty of non-fixed, internally fused, screw wired plugs around, that are made in two sections that come apart to fit wires or fuses. One day you'll come to unplug something and the plug will feel sort of loose (or fall apart even). That's when flicking the switch on the socket seems like a very good idea.
My grandmother (who was brought up when gas was available in Burnley, but not electricity) didn't agree. She thought that just like you need a gas tap on every outlet, you also need an electricity tap. When the plug-in electric clock (a present to my grandfather from grateful parishioners) finally broke and was thrown away, she saw that there was no switch on its specially installed round 2-pin socket over the mantelpiece. (This was years before 13 amp sockets were installed.) So she sealed off the escaping electricity by Selotaping an aluminium milk bottle top over the socket. (This was years before home milk deliveries stopped.)
Fortunately the short circuit that readers are no doubt hoping for did not occur, and she survived for many more years. The next time we visited, my father ripped off the aluminium, but I'm pretty sure Grandma insisted on a stout cardboard replacement. You don't want electricity leaking into the room.
My great Grandmother, who was probably of a similar vintage to your grandmother, would not stand for electric cables to be twisted or otherwise be kinked "in case it slows or reduces the flow of the electrics and then things stop working". Again, related to gas appliances, such as the gas powered iron she used to use that was "plugged" into the outlet of a gas lamp via a rubber hose.
Homor writer James Thurber reported, possibly tongue in cheek, that his grandmother "lived the latter years of her life in the horrible suspicion that electricity was dripping invisibly all over the house. It leaked, she contended, out of empty sockets if the wall switch had been left on." Including lampholders apparently, but who sits about with empty lampholders? Somebody who doesn't want to have all of their lights turned on and costing money, I suppose. Or possibly someone who plugs other appliances into a lamp socket, which is out of fashion now. The episode in "My Life and Hard Times" has her screwing-in bulbs as a check for leaking electricity: I would suppose too that an appliance outlet was different from a lampholder even then, but Thurber was born in Ohio in 1894, so at the time he's talking about, electrical arrangements may have been pretty wacky. Bakelite appeared in 1908; what did they have before that, just rubber?
Before Bakelite, there was porcelain. In fact into the 1960s I remember a certain type of electric kettle, called a jug, made of porcelain, although they did have a Bakelite lid, which was very common in Australia. (They were yellow with a vaguely art nouveau design, so might have been in production since the 1920s. The elements were replaceable so the jugs lasted forever.)
Inside the walls, wiring was rubber and/or cotton. I've seen a video online of a very old installation in an American house where *uninsulated* rods were run between porcelain standoffs in the loft. At least the rats won't be eating that.
I've seen a video online of a very old installation in an American house where *uninsulated* rods were run between porcelain standoffs in the loft.
The uninsulated rods (as opposed to wires) are probably actually mineral insulated metal sheathed cable (MIMS) sometimes referred to as "pyro". The outer metal sheathing does not carry current. An electric range or oven uses similar technology in its heating elements.
Which reminds me...
On some underground/sub-surface sections of London Underground you will see, at window height, two bare wires running alongside the track, mounted on porcelain stand-offs. (Rule #1: If there's wires on porcelain mounts, be careful!). These have the dual purpose of being able to strap a telephone onto them to enable a train operator to communicate with the control room; and, when pinched together, to indirectly knock off the traction power. Though the voltage between them is not too big, because a relay (coil) is in circuit one could get quite a belt (recoil lol) off the wires if one were not careful. The relay operates as a contactor which is capable of switching 600v DC (traction voltage). Once triggered, the power should remain down (the train crew might have to walk the passengers along the track), though there are other ways to ensure traction remains discharged.
Well, electricity can leak from empty plugs, sort of. Here in Thailand, I have often seen the wall plugs being installed on the neutral wire instead of the live (presumably in case the switch get destroyed and you touch the inside, it would be touching the neutral?).
As a result, fluorescent lamps would never go completely dark but would keep a constant glow due to the live being present all the time.
Things may have changed since a proper electric panel, including some sort of fault detector is now compulsory for any newly built house.
...of a particular council.
It all started when I did a migration from NT4 domains to 2003 Active Directory. 2003 AD was new and shiny so it gives an idea of the timeframe.
Having done the migration fine, it was time to upgrade their Citrix farm. Half a dozen new ProLiant DL360's and a pair of Gigabit Cisco switches.
Again it all went fairly smoothly beyond a bit of application wrangling and the customer was happy and off I trotted.
A few weeks later I was asked to speak to them because "everything was suddenly slow...logons could take 3/4 of an hour" etc etc.
I did some remote troubleshooting and yes indeed it was dog slow.
So I drove to them. This wasn't a little jaunt but a 6 hour drive.
When I got there, I noticed straight away that there was a new network cable suck out of the top of the server rack door (which itself was left ajar), ran across to the server rack opposite and into the guts of said rack.
Traicing the cable and I came across a 3Com (who were pretty defunct by then already) 10Mbps hub. A. Hub. 10Mpbs half duplex at that.
And of course that cable was in the uplink of the Cisco switch pair and then off to the rest of the network. No wonder the thing ran so badly! I was not best pleased with their network manager.
Or the time an application on one server stopped working. Again, did the usual troubleshooting - nothing obvious. Nothing had changed (right...) but when I drove down again and looked at the server in question there was an MS VB6 CD in the drive. Someone had installed an earlier version of the full blown product (whilst live, with users on) that was older than the needed runtimes for tha application.
Oh yes. Happy days.
I recall when I worked at the local college (now a University of course) in the early days of co-ax Ethernet. Operations couldn't figure why performance across the campus varied so much until they found the one very slow* speed, and overloaded, device that was bridging two subnets.
*Memory has faded as to just what the speeds were.
Re : "10Mpbs half duplex at that".
Now, I'm trying not to be pedantic, but failing so I have to point out that all hubs are half-duplex by their very nature.
I still keep one around as it is occasionally useful when having to troubleshoot a network issue and the blasted Sysadmin has given you the wrong password to the router and then gone un-contactable - Instant monitor/mirror port!
If I like the site users*/sysadmin, I'll even plug the 100Mbit hub into the uplink port, not the 10Mbit one.
*Liking users is usually professionally frowned upon.
"...Now, I'm trying not to be pedantic, but failing so I have to point out that all hubs are half-duplex by their very nature..."
Well silly me forgetting that minor detail almost 20 years down the line. But thanks for correcting a minor error in a little anecdote. <rolls eyes>
I had that one from the other end a couple of weeks ago.
I had reported a problem with a set top box (STB) for the TV. The problem itself was solved, but I received a phone call from support about it. "Yesterday evening it worked, but now it doesn't ... We have a power failure". Laughter on the other end and "I didn't see that one coming".
Not really like that and way off topic, but in the blissful oblivion department I have a royal flush in spades.
9/11, NYC. About three miles from WTC with direct line of sight. Both towers still up, but burning and obviously not long for the world. I sit in shock next to a phone, which rings.
It is head office in Albany.
"No-one is picking up for our weekly staff meeting!"
"We're all a bit distracted just now. I suggest you reschedule."
"But we HAVE to have our staff meeting!"
"No we bloody don't, not today, not now. Go and find someone with a bloody clue to tell you why" and I hung up.
True story, shortened, cleaned up a bit.
But the problem was that we kept loosing data! Machine was working fine, the days statistics printed off, but next morning they were missing.
After several days of the same problem I went on site to monitor things. A bit longer run to get to site. Daily stats run off everything fine. The machine was left on so it was still accessible remotely to allow managers to check anything they needed, and shut down later to restart in the morning.
7PM a hand appears around the corner, pulls out the plug and plugs in another plug - without even looking - then the vacuum starts ...
The real irritation here was not so much the hand of god, but the fact that while 'write to disk' was activated on the machine it turned out it only happened when the program shut down! Fixing that fault help a solve other site problems where electricity was an equally mystic provision.
When I first obtained (used) my current riding lawnmower, it wouldn't turn off. I DID pull the plug lead the first few times. That being right painful, though, I managed to dig up a wiring diagram, and discovered that the engine side of the magneto's transformer (the lower-voltage side, not the spark plug side) was grounded to earth through the ignition when the key was in the off position. Or would have been if that wire wasn't unplugged. Plugged it back in, and it worked perfectly.
Many moons ago, my dad had the electrical maintenance contract for a local software company. One day he gets a call from the head of facilities.
"There's a big board meeting going on and the room's in darkness, can you get here ASAP and look into it?"
My dad drops his current task and hies himself to the software company. He gets escorted to the boardroom where his first diagnostic test is to flick the light switch, which had the effect of illuminating the room. My dad left shaking his head at the thought of all these people charged with running a company not thinking to check the light switch.
One of my common quips to IT is - If it plugs in the wall and doesn't make food it's IT equipment
In fairness the guys at IT and their managers seem to like to take on every electrical item. I think there is job insecurity that drives a lot of stuff.
I don't miss the screwdriver side of the work at all.
I used to work evenings doing staff support in a university. One of the computing lecturers phoned my office in a panic one night. He was in a different building, but on the same campus.
He was giving a lecture, and apparently none of the av system was working. I took him through a quick diagnostic process, but he couldn’t do much as there was no power.
Being slightly concerned that something serious had gone wrong, I packed up a laptop and portable projector so he could at least continue.
Went over there, braving the ice and snow. Not a long walk, but a little precarious due to the ice. Unfortunately at the time, the projectors we had were only portable in the loosest sense, so I was carrying some heavy equipment.
Got to the room , unloaded all the stuff I was carrying on the desk, went over to the av system, looked at the wall and flicked the power switch. The lecturer had been using that equipment for years, so he knew his way around it, but hadn’t apparently thought to turn it on.
In his defence, those systems weren’t usually turned off at the wall,
Anyhow, with the system working, I picked up the laptop and projector and traipsed off into the snow,
All the big wigs thought it was some pleb's job to switch on the lights......
Work in a hospital - was interrupted at lunch to go to the MDT (Multi-displinary-Something), a teleconference room where docs from different hospitals would discuss patient cases. Apparently the camera connected to the microscope wasn't working. So went down to have a look - even though it's nothing to do with IT, room full of consultants and found that the little mirror thing* to shine the light through the slide was flipped...
* I dunno what it's called
It's a bit alarming that nobody in a room-full of doctors could use a microscope properly. The trickier to spot version is that the trinocular* head is set to the binocular. Unless you notice the glint through the eye-pieces there's nothing obviously wrong.
* Pair of binocular eye-pieces plus straight through port for the camera.
Once in ye olde days of dumb terminals got a call "terminal isn't going", and went through all the questions about power and switches. User was adamant that the terminal was plugged in, switched on, data cable connected etc. And would I "please stop talking to her as if she was an idiot".
The VAX (yes, that long ago) still maintained the terminal was turned off. In desperation I asked "Is there a light on the front?" to be told there was.
When I got to the terminal I found of course that the wall switch was turned off.
What about the the light on the front? Of course there was a light on the front. I hadn't asked if that light was illuminated.
I had something like that once. My question (before the problem became obvious) was: What lights are showing on the front? Answer: none. Me: Is it plugged in? (Pause) "It is now." The big power dongle was under the desk and had gotten accidentally kicked loose.
After that, for any similar calls, what lights are on on the front and is it plugged in were much earlier in the dianostic questions.
I doubt the guy even looked at the wall socket. I've been to plenty of calls where the power is the issue, only to find the questions I asked on the phone or tickets were answered affirmative, but in reality they never looked at the light on the power strip, or realized the darkness in the hallway...
only to find the questions I asked on the phone or tickets were answered affirmative, but in reality they never looked
Arguably that's your own fault (not you, personally, but that of support centres). It's because we're all far too used to helldesks with pointless scripts full of irrelevant questions which they have to plough through before even considering what the reported problem actually is. We've been conditioned to say "yeah, already done that" to everything, without even looking.
Proposed answer:
"In order to verify the machine is fully, completely shut down, I need you to turn off the socket switch, unplug the power cable at BOTH ends, and wait 5 seconds. That will allow the computer's capacitors to fully drain, allowing a proper cold boot. Then plug both ends of the cable back in, turn on the socket switch, and then turn on the computer."
SOUNDS like a secret "this is the way to really do it" answer, but really you're just making sure it's plugged in and turned on.
Support center? Scripts? Never heard of them, we a shop of 4 admins supporting 1000+ no time for such frippery. We have a ticketing system, I think 7 users actually send tickets. The rest is by email, phone or as I like to call them "drive by shootings" as we are in transit to another user...
I think it's selective hearing -
Servicedesk - "Is it plugged in?"
User hears "Would you like an engineer to walk all the way across the hospital to check to see it's plugged in?"
User - "Yes"
99% of calls I get where the SD have put "User has checked cables", I can guarantee that's not plugged in
Not a local government, but an auspicious hotel.
The computer in question stood proud in their "IT room". It was about the height of a washing machine and half as wide. It was also often found loaded with papers, folders and coffee cups, being at a convenient height and having a central position. While running the entire hotel's IT, it also made a convenient side table.
The call came in that the computer had "died". Yes, it was plugged in. Yes, the mains was switched on. Yes, pressing the button on the front did nothing. No, there weren't any lights on.
Out went a hardware engineer, complete with an entire new box, just in case. On arrival, the first thing he did was to remove the pile of junk from off the top of the machine. To his huge mirth and the IT manager's embarrassment the machine sprang into life!
It turns out that the "kettle connector" at the rear was near the top of the case. Some of the papers that had been dumped on the unit had flopped over the back and pushed the connector just far enough out that the machine lost power. Naturally nobody claimed responsibility, though most of the documents were addressed to the IT manager. After ensuring that the connector was well and truly shoved as far in as possible, our intrepid engineer returned to base.
We didn't get any more calls from that site for quite some time.
I've had similar to that. Computer not working, blah blah, site visit. I pop the lid of the computer, check all the innards, power up, works fine, "play" for ten minutes, still working fine, put everything together, still working, pass back to user.
User calls back, still an issue. Back and forth. Eventually I spend the afternoon there, and at some point the user adjusted the monitor - which was stood on the computer case. A couple of minutes later the computer failed.
It turned out a hardware expansion inside was juuuust tall enough to almost touch the top of the case. Of course, whenever I went there I moved the monitor and opened the case. When the user was using it he subconciously fiddled with the monitor and eventually the weight would be in just the right place to hit the top of the expansion and knock it just enough out of its sockets to make it fall over.
1990s, Hour each way trip on site to a callout at a small business.
Speaker and microphone wires were swapped over... You know the sort, the one with coloured plugs that have to go into matching colour sockets. Quite why they needed sound on a office PC (long before video calling, even VoIP, became possible) and why it was so important I had to make a special trip, I never knew. I also don't know if they got charged for the callout. This was a new business being run from the converted garden outbuilding of the MD after his previous venture had gone bust, I'm assuming his business skills were on a par with his IT skills...
"long before video calling, even VoIP, became possible"
A company called NET was selling a VOIP and Video over the internet solution in the mid 1980s. IBM used it quite extensively in its own internal world-wide WAN, and resold it to others. Ebay has one of NET's QAVPs[0] of the era listed as I type ... Don't bother purchasing it unless you have the required back card with the 4 POTS ports.
[0] Quad Analogue Voice Port
Try a 6 hour round trip through summer traffic in the Lake District for this.
User - My Monitor is dead
Me : Have you checked the power?
User : Yes! The sparky has tested the socket and has verified the fuse in fine also.
Me : OK I will pop down with a new monitor
3 hours drive later I carry a 17" CRT up 4 flights of stairs
I am about to install new monitor and notice power socket is switched off - I flick the switch and monitor comes to life. I found out shortly after the user asked the sparky to check the socket and plug fuse which he did and verified both were OK, At no point did the user say to him this was because the monitor was not powering up and the sparky left everything the way he found it -plugged in and powered off!!!!!
I did learn after this call to ask a few more probing questions to users before arranging a site visit.
UK domestic power is on ring circuits connected to the distribution board at both ends so a break in the circuit still leaves the sockets powered. Each socket requires a switch and each appliance plug needs to be earthed and fused. There is a theoretical reduction in conductor size in the ring circuit compared to radial circuits.
Continental European circuits are on a radial basis with the switch if any sometimes co-located with the light switch. These circuits are protected at the distribution board and no fuse is required in the appliance plug. The polarity is not defined with either of the two pins being live or neutral. Earthing was something of a late comer to the continentals and varies across Europe.
Remember the pre-war Times Headline "Storm in Channel Continent isolated."
"UK domestic power is on ring circuits connected to the distribution board at both ends so a break in the circuit still leaves the sockets powered. Each socket requires a switch..."
No, you can buy sockets with no switch on them. All perfectly legal.
https://uk.rs-online.com/web/p/plug-sockets/7758071/
Unswitched 13A Sockets
BS1363 13 Amp unswitched single gang sockets in white
If he was working anywhere near where I lived as a child, I doubt that was the result of too much Colossal Cave. I've definitely been in a car that couldn't move while we waited for the herd of cows to make their way round the car.
And one epic story where a lorry was sent the back route between the 2 nearest towns via our village. Both ways involved a steep hill. However the back way involved, after coming down one hill, and crossing a narrow bridge over a millpond, a right angle bend into the other hill, with a single width road with high stone walls on both sides. About half way up there was another right angle bend. And no relief from the walls.
The lorry was there for a good long while.
Not much different to where I live. At least routing S/W mostly doesn't direct HGVs that way any more. Descending from the cross roads there's one relatively(!) wide stretch with a right-angle bend with a high wall on the right. A hundred yards or so below there's a right-right angle bend to the left and the road narrows. It was just round there that they mostly got wedged against the left hand retaining wall, probably on account of the fact that from an HVG cab they can see the 20 foot drop on the other side of the right hand wall. We used to get a few diversions whilst the local tow truck came to pull them out.
A couple of years ago I saw a tanker parked just past the first corner and still there when I came back an hour or so later. I realised he'd discovered his problem sooner than most but still couldn't back up unaided. A few hours later he'd gone and there were wheel-spin marks on the road.
...was getting sent to handle an irate conference guest who insisted that despite what they had been told, there was no internet socket in their room (These being the days before ubiquitous wifi.) The socket in question was in fact on the wall just above and behind the desk, right next to the power socket they had left their laptop charger plugged into.
Since the person in question wasn't there when I showed up, my response was to use yellow electrical tape to put an extremely large arrow on the wall pointing to it.
# Hello, IT
* My laptop can't get onto the internet, it's plugged into a network socket.
# Ok, can I have your machine ID.
* blah blah blah etc
# Hmmm. That's not in the network database. Can I have your user logon ID or your payroll number so I can check your user details.
* Oh, I don't work here.
Yep. Random stranger wandered into office, expected to be able to plug into corporate network and use it.
"checked that it actually worked?"
Problem : irate conference guest who insisted that despite what they had been told, there was no internet socket in their room.
Solution : Clearly identify location of actual internet socket to absent irate conference guest in a completely unmissable and unambiguous manner.
Effectiveness : Addressed 100% of the customer complaint, despite their absence.
Nobody said anything about the socket needing to actually work..that's a separate issue (Depending on how upset you or your co-workers were by the guest's attitude - the more irate/impolite they are with me or mine, the less I am inclined to help them out ).
This didn't happen to me, but to someone in the IT department of a company I was consulting for. The "problem" was with a PC at an industrial plant, about 300 km from where the IT department was located. Also, it was night shift, and nobody from the office area was at the plant - just a couple factory workers. Apparently, someone shut down the computer while leaving the previous shift.
The worker on shift did not know how to turn the PC on. So he picked up the phone and woke up someone in IT. The IT guy told him to push the power button on the PC. The worker could not find the power button.
This escalated into a bit of a shouting match, which culminated in the worker claiming that he wasn't qualified to turn on a PC, because the firm did not provide appropriate training.
In order to avoid losing an entire shift worth of work, the IT guy had to drive 300 km to the plant, push the power button on the PC, and then drive back home.
"In order to avoid losing an entire shift worth of work, the IT guy had to drive 300 km to the plant, push the power button on the PC, and then drive back home."
Yeah, course he did. /s
Exactly the same as if he'd had a genuine problem. Pretty poor IT support arrangement by the sound of it. Another one of the benefits of "centralisation" :/
Kind of - with a slight twist.
A colleague and I were requested to go and install the O/S on a new HP Superdome (so yes, some time ago). Those of you familiar with the hardware will know it takes a while to do a site install before you even get to power it on, let alone do an O/S install. So we asked several times if the hardware was ready for us to work on. We were emphatically told yes, its ready. Made sure we had email to prove it as we had to book a hotel (couple of days work, and the site was over 100 miles away).
We rock up to the DC, waltz into the machine room - to find the Superdome still in bits and the engineer still working on it (great guy, not his fault at all).
SO we called the boss and gave him the news - we got the rest of the day off, an overnight stay in the hotel, and a leisurely trip back the next day after a hearty breakfast.
Happened to me just last night. The daughter called, couldn't get the PC I'd put together for her running.
Being a bright lass, she swapped the display and PC power cords and realised her problem. No call-out necessary.
Well trained, she is. My pride and joy.
If you have two or more way switched circuits you get used to the idea that you can't rely on which way the toggle is to tell whether it's on or off.
After several years my granddaughter's bedroom light switch is still the wrong way up. It was like that when they moved in. I couldn't work out how to get the face plate off without prying, probably hard enough to break it. My BiL who's a sparky was to do some work there so I asked him to take a look. Still nothing. I suspect whoever fitted it originally had an "Oh, shit" moment when he realised what he'd done and he couldn't move it either.
I always have this when I go stateside for work, where their light switches are generally up for on and down for off. Of course completely reverse of what I'm used to as a right-pondian Brit. Always takes a day or two to mentally flip over for that one, especially when fumbling for said switch in the dark.
That makes sense for toggle switches, and was the usual standard in the UK when toggle switches were still around. But with rocker switches that only protrude out from the fascia by a couple of mm, it's not an issue. If you fall against one, it doesn't matter which way is on or off, odds are you'll flip it.
we had a number of Xerox Phaser wax printers in my old job, now anyone who has ever used these will know they don't like being just turned off you need to shut them down properly. Anyhoooo we had large DO NOT POWER OFF stickers on all of them, obviously not large enough for one particular know it all boffin who I caught just pulling the plug out as his job hadn't printed. "ere knob head what are you doing didn't you read the bastard big DO NOT POWER OFF sticker?", "oh didn't think that applied to me" , "TWAT!"
The Wyse 50 had its mains switch on a bezel on the front at keyboard level. The keyboard, when moved and makes contact with the bezel, would turn the terminal off.
Virtually every morning we would get a fault call where a cleaner had moved the keyboard and inadvertently turned the terminal off or people tidying the desk at the end of the evening did the same thing. Usually a quick conversation would have the terminal turned back on and everyone would be happy.
Not this so important solicitor who simply wouldn’t try what a lowly support person asked no matter how I reasoned with him and how long it would take to get there so four hours later I turn up in Aberdeen, having driven from Glasgow, and have to wait as he is with a client.
His staff are friendly, cup of coffee and biscuit supplied. Taking to them it was obvious they all thought the guy was the type of person who rhymed with Rick.
Anyway the time came and I got to walk into his office, turn on the terminal and walk out without saying a word. He complained to my boss about my conduct but as he thought his name was also Rick nothing came of it. Or was it Rick with a silent “p”.
Still happens with PCs. I solved that particular problem in one office by gaffa'ing two telephone directories together and plonking the PC on them, raising it up the crucial one inch. (Yep, even then they didn't make phone books like they used to do, had to use two of 'em.)
... of the person seeking IT help for an inoperative desktop in a building where the power had failed.
The call went on with the phone support person asking the user if they still had the box the PC was shipped in.
"Yes" was the answer.
"Well pack it up and take it back to the store. And tell them that you are too stupid to operate a computer." ... Click.
My favorite variation on the theme, as personally experienced by me:
The magnitude 6.9 Loma Prieta earthquake hit us on October 17th 1989, at 5:04 PM-ish Pacific time[0]. It was centered approximately 30 miles SSE of my home. PG&E power and Ma Bell landlines were out over almost all of the Bay Area. My acting boss called my DynaTAC at 5:10 PM & screamed that he would fire me if I didn't fix it immediately. I told him that he needn't have wasted money on the phone call, he could have just opened the window and bellowed. And then I hung up.
I have hated cellular telephones ever since ... not because of what they are, or what they can do, but rather for what they are actually (ab)used for.
[0] The so-called "Bay Bridge World Series Quake".
Used to work 1st/2nd line support for a company that made automatic warehouse systems where robots would go round the warehouse knowing where they were by reading a barcode or matrix of some sort. Middle of the night calls were along the vain of "You F**king robots broken down again". For which the reply would be "Take the poster advertising the football match off the wall". But would often still have to visit said company to walk up to wall and take poster down and suprise suprise the robot burst into life and a bill for a vast amount would be generated.
A support call, allegedly made in the USA to a service desk.
User: My computer isn't working
IT support: Is the screen on?
User: No
IT support: Is it plugged in?
User: Yes
Are you sure?
User: Well, it's difficult to see, the plug is under the desk and the lights are off
IT support; Could you look, please, and turn the lights on.
User: I can't turn the lights on there's a power outage
IT support: There's a POWER OUTAGE!
P A U S E
IT support: Do you still have the box the computer came in?
User: Yes.
IT support: OK, I need you to unplug your equipment and put it back in the box and take it back to the supplier you got it from.
User: Is it really that bad?
IT support: I'm afraid so.
User: What do I tell them at the suppler?
IT support: Just tell them you are way too stupid to own a computer <hangs up>
Had a user testing some new kit alongside his existing device. Called to say he was trying out the main app they used and was trying to load something but "only half the files are showing compared to the old kit" (they opened one file, but the software opened an additional 6 or 7 files). Checked several times that he was doing things exactly the same way on both kits, so when the answer was in the affirmative, I spoke with my boss, arranged a hire car then drove the 140-odd miles to his office.
On the walk from the lift he said he'd left both kits side-by-side and not touched them since we spoke, and as we walked through the door of his office, there they were and I immediately pointed-out that one was looking at the files in File Explorer whilst the other was looking at them via the File/Open dialog of the app...
2.5 hour drive for literally three minurtes to get from the car to his desk and then 5 seconds actual work. Followed by another 2.5 hour drive back in a worse mood than on the way up.
At <former company> we were working to deploy a global CDN for <customer>. We were building and configuring racks and shipping them out to sites with instructions for local hands. The instructions were very clear "this plug goes here--" with diagrams.
Rack arrives at data centre in New York. Customer comes round to our desks and says "New York should be live". A quick ping. NADA.
Customer says "must be your fault, you'd better get someone over there" (this is in the UK).
"OK, but if it's not our fault you pay".
Newest recruit (been with us six weeks) gets return ticket to JFK. Arrives, takes cab to site, walks up to the rack. Opens back door. Moves an RJ45 one socket to the left. Calls us. Ping. "Yep, it's working".
Cab back to JFK, flies home.
We did NOT pay the cost--.
I got a 1AM call from the CEO of a company I did consulting for. The home genset I had installed for him as a side project didn't work in a power failure. He was completely impervious to answering questions intelligently over the phone.
His company was a rather large account, so I told him I'd be there as soon as I could. (I was in Palo Alto, he was in Half Moon Bay.) It was raining (naturally), so instead of one of the bikes I drove the Taurus SHO. 40 minutes (ish) later, I arrived.
I discovered that said CEO was a) legless, and b) had managed to actually fire up the genset, but couldn't figure out the simple transfer switch. I had even labeled it "Genset/OFF/PG&E" ... He was flipping it between OFF and PG&E when I arrived. Somehow the concept of a three position switch eluded him, despite having been walked through the simple procedure not a month prior.
So a late-night mad dash in the rain to flip a switch for an idiot. I stopped doing personal favors for corporate clients after that, no matter how lucrative.
Client with half a dozen Apple Macs with the AAUI interfaces and connected via BNC in a token ring. On two previous occasions a connector had 'come undone' and on each visit to fix it was carefully pointed out what had happened when they 'lost their network'. The third time I insisted they checked every connection and they insisted that it was all ok . I walked in and before I'd even removed my coat I pointed to the connector on the desk... 'No No NO' they insisted they'd checked... I connected it and walked out in less than 5 minutes and charged the full call out fee.
Sadly, there is a tragic true story from India from June of this year about a family in the hospital room of their son who is on a ventilator. It's 41C outside, no AC on inside to prevent the spread of coronavirus, so they buy an air conditioner, plug out the ventilator without telling anyone, and plug in the AC. Inevitable sad result. Of course, they blame the medical staff. https://www.indiatimes.com/trending/social-relevance/family-unplugs-ventilator-covid-19-patient-dies-516000.html
I do pause at "Would not believe it if I had not read it." In the "urban legend" usual version* it's a cleaner who does it.
I'd expect any mains medical appliance to bleep when it goes onto battery, but there are possible reasons that that wouldn't be noticed or attended to. On the other hand, the COVID-19 intensive care ward is not going to welcome visitors. Not now anyway. And if this is a true "ventilator" then the patient is kept unconscious and not really worth visiting, but the term may also be applied to different types of oxygen supply machine.
I had the school admin server that went onto battery and shut down before making its overnight backup for a while (we didn't want it to shut down at all), because the "janitor" in role of watchman had got it into their head to save turning off lights one by one by just turning off the whole building.
* https://www.snopes.com/fact-check/polished-off/ "arranging for an electrician to fit an extra socket"
Ah yes after spending very little time in the UK over the last 3 decades I'd forgotten these quaint throwbacks to a bygone era. Forgotten even by the local denizens it would appear. "Why is the phone not charging...? Oh yes, socket switch." Charming bit of British heritage. Liken thatch. And Laura Ashley style furnishings.
Asking someone to check if a socket is on, or a cable is properly inserted, never works. Ask them to unplug and plug it back in instead.
Back in my tech support days in the mid 90's I once got a call from a user whose computer was stone dead. After she'd unplugged the cable and plugged it back in, both ends, it still didn't work, so I was preparing to send a repair tech out - which would take a couple of hours as she was in a very small site far from the nearest tech support centre.
Then I got the brilliant idea to ask her to swap sockets, plugging the desk lamp (which functioned) into the socket she'd had the computer in. Desk lamp stopped working. Computer worked in the other socket.
I saved the company a total of five-six hours' worth of travel, and the user had a productive day instead of sitting around waiting for the repair person.
In the early 90s I developed a hardware diagnostic for Macs which we sold off to a much larger company that did PC diagnostics. During the hand over one of their support people told me a story which is directly germane to this one. He received a call from a user who said that his PC wouldn't turn on. The PC had worked fine in the kitchen but it didn't work in the living room. He asked the usual questions, is it plugged in?, yes, is the cord in the back of the PC plugged in?, yes, do you have a switched outlet? he didn't know. So the support guy suggested that he plug a lamp into the outlet to see if it's live. The user wouldn't do it because if he unplugged his lamp it couldn't see anything. The support guy went around in circles with the user who refused to flip the wall switch or plug in a lamp. Finally the support guy asked what kind of carpeting did the user have in the living room?, shag carpeting. That's your problem, your PC isn't shag carpet compatible, take it back to the kitchen and it will work fine.
Not related to this week's On Call but maybe related to Hallloween:
"The printer won't stop screaming"
https://notalwaysright.com/using-old-technology-is-murder/215546/
But also:
https://www.reddit.com/r/talesfromtechsupport/comments/jbd2y1/why_wont_the_screaming_stop/
(exactly the same story 2 weeks earlier)
"like a peacock being murdered" but a peacock sounds like a human being so... wait, how do you know what a peacock being murdered sounds like...
The first site does have a pre review channel of "Unfiltered" stories, so despite the dating, they could have had it first.
Once had to drive from Yorkshire to Edinburgh to switch a server on. The machine in question was a low end Netfinity with no remote support card. When plugged in, it would briefly power on, flash the lights, spin the fans, then go off.
When we got the call, and knowing the server behavior when presented with power, I had asked in as many different ways as possible "have you pressed the on button?". I eventually got the inevitable question "Do you think I am stupid?", to which I could not give an honest answer.
So set off, drove to Scotland, walked in, pressed the on button. Fortunately, I had managed to slot in another job while I was there, so not a completely wasted trip. Plus, took a scenic route back (I have attempted to cross the border using every different road I can find on my many trips there) and had a nice leisurely drive through Kielder Forest.
In 2006 I worked for a shitty cabling/infrastructure company in Worcestershire (I can say it's shitty now because it doesn't exist ... because it was poorly run). I was employed as the IT lacky (with the title "Network Manager" I had no management responsibility given to me, but all of the blame). I'd inherited an undocumented shit-show, but due to my age it was the first undocumented shit-show I'd ever inherited, so I didn't know what to expect. Needless to say, IT, and me, got a really bad name as I tried to fix things, improve things, and generally make things better.
At 8:30am, as I was driving in to work, I had a call from the woman who works on reception. She told me, in pained tones, that her computer wasning working "again" (despite me having no recollection of the previous time it wasn't working). It wouldn't even switch on, she said. Flustered and embarrassed, I assured her that her desk was the first I would visit when I got to the office at 9am.
And, as I did, and I walked over to her desk, I could see the waist-level socket that her machine plugs into was switched off. So I switched it on, and said "try now" -- she dutifully apologised, and I didn't get that type of call again.
The bashing about everything else (that I hadn't set up, including home VPN config setups conflicting with the IT infrastructure so any VPN-connected user couldn't see the single server that they were to use for all files & on-prem email)...
It was a Burroughs device. I turned it on: nothing, checked the cable,plug, fuse all fine.
Called the engineer he arrived and turned it on immediately, no problem. What I didn't know (and it's an interesting design decision) was that there were two mains switches, the "normal" obvious one I'd been using for months on the front, a second one on the back that someone else must have found. I blame the cleaners...
I casually offered assistance to my next-door neighbour of some 18yrs... although known her a lot longer as ran around same streets as junior-school kids [always was a bit..erm.. o_O]
"Any problem with your computer.. let me know.."
I was abruptly informed...
"I don't use computer anymore.. I got a laptop"
I wish I was making this up for comedic effect.
The winding twisting roads of the south Wales valleys...
It seems to be a common thing - "a computer" is the thing that sits on or under your desk at work with a separate keyboard and monitor, "a laptop" is not "a computer". South Wales too, though slightly more coastal (workwise) these days.
Another common one - though I am beginning to wonder how these people are coping in lockdown - is "I do everything with my iPad" or even "my mobile". I know one or two at work who do not have "a computer" at home and really struggled in the first few weeks even just with email (OWA is not pretty on an iPad). We (local team with no corporate IT influence) scrabbled to hand out even our scrattiest "public access" laptops until corporate IT had sorted some Surface units, which haven't themselves been exactly plain sailing.
Basic setup on Surface sufficient for web in the first few weeks needs "sorting out" in order to connect to new remote services for long term WfH use. Cannot be done remotely via member of staff's home WiFi, needs to be on the work network so special arrangements have to be made to bring some of these machines in to the closed - except for a "security detail" (me or a colleague) - office. Cannot connect to work WiFi because locked down (and far too obvious for IT to consider an additional SSID with VLAN access), cannot connect to work wired network because...
...Surfaces have no network socket.
So muggins 'ere has to procure suitable network adapters. Adapters plural because nobody is quite sure who has what type of Surface so I don't know if I need one with a USB-A plug or one with a USB-C plug.
Back to "computers", we have at least one relative who ditched their computer and printer and camera in favour of just an iPad (and probably an iPhone, not certain). Granted their house isn't huge and it's freed up quite a lot of space in the study, but personally it gives me the creeps that this person now only keeps family photographs in the iCloud, except for the few she occasionally gets printed to hang on the wall or give to relatives, and her emails have turned from "letters" to something akin to slightly verbose text messages.
And how do such devices cope with the government websites necessary for claiming benefits etc., now that jobcentres are essentially out of bounds?
But I suspect this will become more and more common. We don't have a 'pad' device ourselves, but when it looked as if I would have several children trying to do schoolwork remotely at home and only one laptop to share, my first (and cheapest) thought was to buy a webcam for one of the computers and a £30 Bluetooth keyboard and mouse to pair with one of the better smartphones, just in case. Apart from screen size, a modern smartphone paired with a keyboard and a pointing device makes a passable laptop for emergency use. As it happens, the schools involved all have a "no cameras" policy when on live lessons.
They also have a "no bedrooms" policy, which we're managing, but I know others had difficulty sorting out enough separated work areas in smaller houses.
Rambling, sorry :-)
M.
A very similar story, whereby I saved the day. I was attending a musicology lecture at a prestigious UK university. The lecturer had his synth all plugged in but, alas, it would not switch on, so no sound could be produced. The duty engineer was called, gave the set up a look, said "I'll be back" & disappeared. 15 mins later, no engineer, no nothing. So I got up, go & looked carefully at the synth, followed the synth cable to the wall plug and...realised the wall plug is not switched on. I switched it on & voilà, the lecture could start. Nothing more was heard from the engineer throughout the 90 mins lecture.