
There is no problem
that can't be solved by a sizable application of high explosives.
"Work or I'll pack you full of c4 and turn you into atomic dust... Oh look, it's working now. Good. Pub time!"
Friday is here and while we cannot promise a break in the weather, we can bring forth another recollection from those brave souls staffing the helpdesk courtesy of The Register's On Call mailbag. Today's tale comes from a person the Regomiser has elected to call "John" and rolls back the decades to when DOS and dot-matrix …
I prefer the slightly more tactile approach.
That is to say, approaching a recalcitrant machine with a well selected hammer and giving it a a touch of percussive maintenance.
It's important to have a tool kit that will permit incremental levels of maintenance with the largest hammer fully visible to the offending machine.
When I was in the sixth form, I was known as the go-to person for solving car problems. One day, one of the teachers asked me to find out why her A40 (pre Pininfarina) made a clonking noise whenever she turned left. I asked her if she had had any work done recently, and she said that she had had a new battery fitted by the local garage the previous day. I opened the bonnet, and there, resting on the steering drag link, was a seven pound lump hammer. It would remian upright until a left turn overbalanced it, whereupon it would tip over to the right and the handle would knock against the inner wing, returning to its stable upright position when the steering was centred. I removed the hammer and the noise ceased. The teacher complained to the garage, who had the audacity to ask for their lump hammer back. I still have that hammer, it is used to threaten recalcitrant equipment both in the garage and elsewhere.
<Pedant Alert!!> Ammonium nitrate isn't a high explosive (fortunately). If it was, then the pressure wave would have been supersonic (a "shock wave" and many more people would have been killed.[1] </Pedant Alert!!>
[1] See e.g. https://www.wired.com/story/tragic-physics-deadly-explosion-beirut/
https://arxiv.org/abs/2009.05674
Explosion analysis from images: Trinity and Beirut
Jorge S. Diaz
Images of an explosion can be used to study some of its physical properties. After reviewing and clarifying the key aspects of the method originally developed to study the first nuclear explosion, the analysis of the data is discussed in connection to undergraduate laboratory experiences. Following the exposition of the procedure for the Trinity explosion, the method is applied to the Beirut explosion of August 2020 by using the frames of many videos posted online and producing remarkably accurate results. The estimate for the energy yield of the Beirut explosion is found to be 4.19+0.19−0.19 TJ or 1.00+0.05−0.05 kt of TNT equivalent. A basic modeling of the pressure wave indicates the temporary supersonic speed of the blast.
I ended up fixing a printer with a very similar problem, but not such that there was any kind of fix the manufacturer could apply. In my case, it was the pre-printed stationary they were using. Those old printers often had a very small buffer sometimes barely more than a line of text. Long story short, the pre-printed stationary had a thick dark separator line and the printer paused waiting for the buffer to fill just as the dark line passed over the optical sensor which then reported "Paper Out". The solution? A carefully applied dab of Tippex showing the operator precisely where the paper needed to be lined up so the pause didn't happen directly over the sensor. It was, literally, the difference of one click of the platen roller knob, hence the intermittency of the issue and why previous engineers had not found the cause.
I remember a customer who had a problem with the software I supported going on for months. Every morning without fail, at Market Open, a link would stop and restart. We said it looked like "application code". We could see nothing wrong with our code. It got political and I was sent on site. I turned up Monday morning, and we all stood around in the ops centre waiting for it to fail - and it worked without any problems. There were comments like "the code knows you are here".
I spent the day doing a health check, and said I would come back the next day.
I went back to the hotel and saw one of my colleagues, and so had a beer and catch up. He had spent the weekend assisting in a major ugprade to another product at the same site, putting on a year's worth of fixes and restarting all 100 transaction servers.
Next morning in the ops centre it worked perfectly. Then the light dawned on me. The problem was with the "transaction server" which they had fixed with the big upgrade. The problem they were having was the application would have a problem, close the connection to my product and restart. My product reported the break in the connection and so got blamed. It was a case of "shoot the messenger".
The customer took his team and us out to an expensive dinner to say thank you.
Once had an interactive screen with sensors around the bezel, and during the fall months, the sun would come in the window just right, and "blind" the sensor and all kinds of interesting things happened to the interactivity. Small cardboard triangle taped over the corner of the bezel was the quick and dirty fix.
Used to be a problem when repairing VCRs (remember them?) as well. The tape beginning/end sensors were triggered by some clear leader tape at each end of the magnetic stuff. I've heard of bench technicians opening up a unit to repair a minor fault, and having it stubbornly refuse to operate at all, eventually traced to sunlight on a bright corner of the bench falling on the tape sensor once the cover was off.
Broadcast video cassettes - Sony Umatics, mostly - used dozens of optical sensors around the lacing path. When some poor neophyte was first exposed to a broken unit, it was common to 'help him get some light on the subject' with an anglepoise lamp strategically placed. The poor little Umatic would generally decide to have some very strange behaviour, often (but not always) completely unrelated to the original fault.
30 years ago on some of the early video recorders with ir remote control used to malfunction when the sun hit the ir receiver. The solution was to fit a grill to stop the sun swamping the ir sensor.
Something the TV designers had already figured out (they had a grill or a cowl) but I suspect the video designers had never worked on a TV.
I had a primary school where mice kept acting up, but the teachers couldn't work out why it worked for some people (them) and not others (pupils).
I traced it down to the mice casing allowing IR to pass though it when the sun was in the right (wrong?) place. The kids hands were too small to cover the entire mouse body, but the teacher's did.
Coating the inside of the shell with paint fixed the issue till the mice were eventually replaced with the new fangled optical mice.
A friend had a problem where the screen would suddenly start scrolling when she was away from her desk. It would stop and start randomly. On a sunny and windy day there was a pattern of moving shadows from the tree branches outside the window. Her recently acquired optical mouse was sensitive to the ripples of sunlight.
Outside chance, was it this 1983 Hewlett-Packard job? https://en.wikipedia.org/wiki/HP-150
We had one of those donated around 1989 (?) (we were a non profit or something). MS-DOS yes, compatible disks no. I set it up for word processing (maybe WordStar and may even have used the touchscreen) but had to use I think Kermit on RS232 to move documents to somewhere more useful. Or maybe a program that simply captured bytes from there - OK that would be just "copy filename.doc aux" I suppose. Or maybe it was... print the document, to aux, but the "printer" was my own PC capturing it. Some such thing.
On a similar note, Electronics Australia related a story in the 1980s about an office security system which would go off for no apparent reason like clockwork during spring/autumn on Saturdays/Sundays at *somungodlyhourofthemorning*
It was traced to the morning sun reflecting off the windscreen of the local milkman's van as he went on his round - 2 hours later on weekends than on weekdays - and hittiing a PIR facing a window. The window was south facing.
This was passed on as a lesson that one should never assume the sun (or other bright things) can't shine in windows not facing the equator
Back in the day I worked on a lot of T-carrier stuff. I can't tell you how many times an owner/client ranted about a shiny new (fractional) T1/E1 link being down, how the equipment was shit, the field guys were incompetent, and how pretty much everybody involved with the installation should be taken out behind the barn & horsewhipped. Most of the time, it was an incorrectly set loopback switch on the new node. Seems bosses in general can't resist flipping switches ... and can't read blinkenlights.
Sometimes I'd casually reached out and toggle the loopback switch, thus fixing the link and painting the boss's face an interesting shade of red when I presented him with the bill reading nothing more than[0] "Call out. Flipped loopback switch. $1,000" on an official invoice.
But once in a while, after inspecting the node, I'd stand aside & motion the boss through the door before me. While he had his back to me, I'd flip the switch ... and we'd go off to his office for a chat about fixing the obviously broken machine. I'd let him rant on for several minutes, around 20 was the record, but always ending up with "so what are you going to do about it, then?". To which I would quietly reply "Oh, I've already fixed it. We'll invoice you for the call out". Sometimes the resulting sputtering reached epic proportions ...
[0] The rest of the time it was a cable that had fallen out of the CSU/DSU because it hadn't been screwed down properly. We always took the blame for that, even if it was their guys bolting stuff together. We've all done it, we're only human, I'll take the blame, no charge ... sometimes it's handy to have a friendly couple of faces in a client's datacenter who probably won't ever try to throw you under a bus.
@Steve
Thanks for the memories.
Ventura Publisher, Gem Plus (?) and Corel Draw and awe from the rest of the office over the high spec PC, massive CRT and high end video card that drove it all, plus a nice HP laser printer.
Spent a chunk of my apprenticeship running one of those before getting back to proper engineering, but at least the datasheets were accurate when they came out
I had VAXstations & X-terminals on my desk in the 80's.
I had a Sun workstation and a Symbolics Lisp Machine on my desk in the mid/late 80s. I also briefly had an HP Smalltalk-80 machine(*), but it was a prototype and failed almost immediately. They took it back promising to return it in a few days, but then HP management changed their mind and "it never happened, I must have been dreaming it". Shame, because I probably could have taken it home otherwise.
(*) Based on a NatSemi 32032, which was unusual.
At that time I was at Uni and all of my housemates had either Atari STs or Amigas. Not for me, I had a green screen dumb terminal connected to one of those old Rair Black Box computers (Think mine was badged as BT Merlin) running MP/M, still, I was the only one in the house with a wide carriage printer and 'professional' software (Wordstar, SuperCalc and DB2).
I still managed o get my reports/assignments submitted on time and since I'd spent some time working in industry before I started Uni I managed to get them looking a lot better than many other submissions.
Having that old Rair Black box stood me in good stead for sorting out all sorts of Serial Comms issues, and for a while (only threw them out last year) I had an extensive range of RS232 cable adaptors.
Also GEM (1984) as used in - among other machines - the Atari ST and some Amstrad PCs (though it wasn't the primary interface in those)...
Also Amiga OS (1985)...
Also "Arthur", a.k.a. RiscOS 1.0 (1987).
All these machines could be - indeed were - used in offices, particularly some of the Amstrad machines. All predated 1988 :-)
M.
"All these machines could be - indeed were - used in offices, particularly some of the Amstrad machines. All predated 1988 :-)"
Although you are correct, my experience as a field engineer back than was that most offices were running MS-DOS apps with the occasional old CP/M computer still around. The only places I saw a GUI type thing running were those rare places running Pagemaker with the Windows Runtime, not full Windows, or Ventura with minimal Gem. Almost no one was actually using a full GUI in day to day use because all the apps were not GUI based apps.
When I started my first job in 1993, I was given a PC runnings Windows 3.1 but all but one of the programs I had to use were DOS programs that ran in a DOS box, and several of them would not co-operate with others. So while I had a "GUI" I was still basically running single-tasking DOS.
Conversely, the people using DOS were looking bemusedly at the people running Macs, Amigas and Acorns, wondering how they got any work done with no software to speak of.
And those of us with clues were running BSD (or perhaps Coherent), wondering why the rest of the world was putting up with illogically thought out, intentionally hamstrung systems that basically didn't work.
So confusing, I have a VAX 4000 (vacuum) in the garage.
Different VAX, but still makes me laugh.
Talking about vacuums mention the computers.
Talk about computers mention the Vacuums.
And even some same model numbers.
And I live near a hotbed of vacuum manufacturers (OK two).
http://www.catb.org/~esr/jargon/html/magic-story.html
Though in that case they never got a completely satisfactory root cause for why flipping the switch crashed the machine and could only speculate as to why a switch that all logic and reason says shouldn't be able to do anything would crash the machine whenever it was flipped.
Reminds me of an AI koan ...
A novice was trying to fix a broken Lisp machine by turning the power off and on.
Knight, seeing what the student was doing, spoke sternly: “You cannot fix a machine by just power-cycling it with no understanding of what is going wrong.”
Knight turned the machine off and on.
The machine worked.
I've actually done that and at about the end of the 80s. We had a hulking elderly printer - might even have been a daisywheel, it was a long time ago and my memory is hazy. It had an iffy power supply that took longer to come up to voltage that the control electronics allowed it and if you turned it on from cold it would always come up with an error condition. If however you turned it off and then back on again with exactly the right timing the capacitors in the PSU would retain the right amount of charge that it would come up and then work fine from then on. I had the knack and more than once I came across someone else trying and failing to get this thing to work by repeatedly turning it off and on and then I'd do it and up it came first try.
Two Windows machine failed on our site (Out of our control\scope).
Trouble was they were the access card control computers for all the secure gates on site (& legacy with no known documentation).
The issue\fix was they required handshaking to each other & it was required that one of them be fired up first before the second so the handshaking could commence. Turning the wrong one on first or at the same time caused both machines to hang mid boot while they both tried to handshake.
It comes from movies where rapidly jiggling the switch hook and shouting is supposed to fix the broken connection. Sometimes worked as an attention signal to a human operator, but did nothing useful once state machines took over the job.
Might help with cranky fluorescent lights where breaking & making the circuit at the correct point in the AC power cycle would give an extra voltage kick to the transformer, exactly the opposite of what is needed in a computer.
"It comes from movies where rapidly jiggling the switch hook and shouting is supposed to fix the broken connection. "
Hollywood and the TV industry have a lot to answer for! In our industry, at least, it used to be jammed printers because people would "RIIIIIIIIIP" the paper off just like in films. Nowadays, it's broken laptop screens because people slam them down hard, just like on TV. It's lack of training or even just basic instruction, so people think they know what they are doing because some actor did it telly.
"It's lack of training or even just basic instruction, [...]"
I have seen a mangled laptop where the man had apparently lost his temper with Windows.
Teenagers have been known to leave a closed laptop on their bedroom floor - and someone has stepped on it in the night.
Apparently many american cops hold their handgun wrong inside buildings because they hold it straight up in the air next to their face the same way that they do in a movie, which is slower that the correct way of holding it pointing forward with minimal adjustment to shoot if needed.
In a movie they hold it like that because of the close up camera angle to get the actor's face and gun in the same shot, it has just been adopted in real life due to familiarity.
"It comes from movies where rapidly jiggling the switch hook and shouting is supposed to fix the broken connection."
In the days of manual exchanges it would cause an attention light to flash on the operator console (and there was operator supervision of long distance calls even after strowger switches came along, but that meme is from "number please" days - something I encountered as late as the mid 1980s and was involved in updating to automatic exchanges plus getting rid of party lines)
As for flourescents: If you find yourself doing that, you have a non-electronic ballast. Change the bloody starter as it's dieing and check the ends of the lamp for discolouration (reddish ==ok, blackish == change the lamp as its on its way out. Always change the starter when you change a lamp assuming a non-electronic ballast)
"It had an iffy power supply that took longer to come up to voltage that the control electronics allowed"
I hate these kinds of kludges. The correct fix is to sort out the PSU
If necessary add a delay relay to apply power after the power rail stabilises
In the 1980's I had a Xerox dorado computer run remotely. The boot sequence took a minute or two. One morning the PHB walks in after I had pushed the boot button and the screen went crazy (as it always did during that part of the boot cycle). I took a look at it, muttered a swear word, snapped my fingers exactly at the right time, and the screen cleared and everything looked normal. The PHB knew he had been had but couldn't figure out how I did it.
We've got an office like that, though the outside walls are almost entirely glass panels with blinds inside that a) aren't quite blackout and b) have inexplicable gaps between them, meaning that for any given member of staff at some point in the day the sun comes through the window, hits their eyes and turns the member of staff off.
I've been a few places like that and a telltale sign is pieces of paper taped to the windows seemingly at random combined with oddly totemic arrangements of objects atop monitors.
There were buildings designed round about the 1960s that sounded and looked good on paper.
Gorgeous impressions drawings, Bright and airy....natural sunlight blah blah blah.
Plenty of schools and offices fitted this design. But the reality was rather different. They tended to be rectangular so that there was always some section that faced into the sun at some point in the day. Which meant that a) a significant number of the people in there could see almost nothing at all for a significant part of the day on any but the cloudiest days. And b) in some areas it got so hot that the mercury in the thermometer started to complain about being made to work in such inhumane conditions. Quite literally the desk tops would be too hot to touch.
Of course some of these buildings had blinds fitted to help. Some even had them designed in by build time. No one ever designed an annual budget to fit though. So after a couple of years they would need an expensive repair, but there would be no money to pay for them.
"They tended to be rectangular so that there was always some section that faced into the sun at some point in the day."
Our office in Africa consisted of two rows of offices on opposite sides of the building - facing North and South respectively. The rooms were air-conditioned - but with one global thermostat. Our office was on the wrong side - so we would freeze for half the day - then swelter for the other half. The people on the thermostat side had no problem.
We had similar in one site. A fully fitted office full of ladies booking customer service visits was connected to the same thermostat as the warehouse office which was by the warehouse door in a large metal uninsulated shed.
During the summer the service office was freezing because the guys in the warehouse were trying to get their office below 34 degrees. In the winter the warehouse boys were freezing.
I was in an office between the two so got to hear all the arguments.
Pretty much everywhere I've ever worked there has been a long standing feud between two halves of the population that I can't mention without being accused of being sexist ... One side always says it's too hot, the other side always says it's too cold. Facilities says "set it all to 72F, that's what the HVAC is optimized for" ... and so we listen to pretty much everyone bitch about the temperature.
Until one place I worked at upgraded the AC, and all the controls that go along with it (had to do with a couple of new clean rooms). Naturally, the folks installing all the new gear left all the old thermostats in place. They were no longer connected, so why worry about them. A friend of mine noticed one of the secretaries would inevitably turn one of these controls up, and then keep an eye on it from her desk. Within an hour, one of the engineers would stroll by & turn it down again. Then she'd turn it back up, and so on ... This dance went on all day.
So we hatched up a Plan ... with the Boss's permission, we installed unconnected thermostats quite near both the secretary and the engineer ... and removed the one they were "fighting" over. Now both could happily set "their" temperature to whatever they wanted. It worked. Both were happy, and both commented how comfy the office was with the "new, improved" controls. People in their circle of friends made similar comments. The complaining about the temperature stopped, virtually overnight.
That would have been the end of it, except ever since then I've installed faux thermostats for 'special" people. It has never failed to shut them up about the office temperature. However, be warned ... that type can always find something else to bitch about. Don't say I didn't warn you.
Our office AC wasn't subject to some people altering a thermostat. They merely opened a window near to them - and that made our area get much hotter.
After a refurbishment the windows no longer opened. Unfortunately it was decided to set up a large network equipment test lab next to us - in an open plan office. That disrupted the temperature in our area - and all the previous personal electric fans had been confiscated and junked. The scream of the equipments' blade fans reflected off the low ceiling - without any apparent amelioration by the low cubicle screening round that area.
"Facilities says "set it all to 72F, that's what the HVAC is optimized for" ... and so we listen to pretty much everyone bitch about the temperature."
The trick is almost always to put a few mechanical thermostats up not connected to anything and let them adjust to their heart's content.
Placebo effect is a wonderful thing
It used to be that windows facing east/west/sunwards would have eaves on them so that summer sun would be shielded from shining in and overheating things. Architects would always handle this as part of the design
Then along came AC and brute force solutions. We're still undoing the damage
I got talking to one of the phone guys about weird faults and he recounted the tale of the company that kept reporting that random phone lines were going down in the afternoon but by the time he'd been called out and arrived the next morning all was fine. He tried swapping everything he could think of but the faults were random, so things might be fine for a while and he though he had fixed it, but he would receive a call a few days later to say the problem was back.
In the end he decided to camp out so he would get to see the problem as it started to occur. After a week or so he figured it out. It only happened late on sunny afternoons, just as the sun blazed on to some industrial-type shelving that had been attached to the back wall of the switch room and OVER THE TOP of the phone cabling... the large metal sheets expanded in the heat and were crushing the cables
Back in the day when a sheet feeder was a clip-on-the-top option for Diablo daisywheel printers, we had a customer in Phoenix, Arizona with 10 of them. Around 3:30 pm, one printer would start printing on the platen, without loading any paper, almost every day, until maybe 4 pm. We showed up several mornings trying to catch it in the act, not a single mis-feed. Still, most afternoons we would get the service call. We replaced the feeder, the printer, swapped around the printers, to no avail. Turned out the sensor for paper being under the platen was washed out by the sunlight through the office window, on just that one printer, which would only last about 30 minutes. Moving the printer about 2 feet over took care of that one.
A network node processor in Manchester, UK would occasionally go offline - then come back on again before an engineer arrived. It only happened on summer afternoons - but not every day. The node was in a room with large windows which received intermittent afternoon sun. The node had removable cabinet doors - which maintenance engineers occasionally forgot to replace. So - then the circuit boards would get warmed by the intermittent sun. One board had a faulty IC on which a crack would expand to break the pin connection until it cooled down again later.
when i was a child, (early 90s) my dad used to sell computers, and give support to his customers.
One day a tower came in that had "random freezes". We're talking a full tower case (about 1m tall) fully loaded with a P1 166MHz.
My dad startet running the usual series of troubleshooting tests, and the bloody machine seemed to work flawlessly.
Being the before-pf-youngster that i was, i crawled under the table and peeked inside the beast - we had removed the side panel.
EUREKA: the tiny cpu fan was not spinning!
So when the customer hit the machine with a serious load, the cpu would just thermal throttle and halt everything, and then when we tried to troubleshoot, we didn't do much more than open a couple of programs and look at settings.
Now, 25 years after that, i work at the IT helpdesk of a medium-large company........ Coincidence?
Ah, the days before a machine needed an active fan even for idling.
Intel were worried about cooling a CPU that drew 15W at full chat. These days a desktop PC will draw anything between about 65W and 135W and will pretty much fall over instantly if the heat sink is dislodged (stopped fans will take them a little longer).
I had a tower with a big heatsink and big fan. The fan was running fine but the damn thing was still overheating. All looked good - shiny fan blades seemingly whizzing round but holding a bit of tissue nearby indicated little airflow. Unscrewed the fan to find the heatsink was chock full of fine yellow dust which turned out to be from the foam in the comfy old chair a couple of feet away. In the right light you could see a little puff of the dust coming out from the cushion as you slumped into it heading directly into the air intake on the tower. For some reason the stuff only liked to stick on to aluminium which it did with relish. Took about two hours to scrape it all out as I hadn't got any thermal paste to hand and didnt want to pressure wash it in situ! Turned the cushion round so the leak went into the chair body and all was well,
Early CPUs just cooked themselves until they broke. The next improvement on this was there was a thermal cut off where the CPU shutdown itself before it cooked itself. The following improvement was the CPU would slow itself down to try and prevent the heat getting to the point that it needed to shut itself down.
These needed the advancements in CPU control and thermal monitors that later CPUs had. Earlier CPUs were controlled exclusively by the motherboard's timing circuits.
"And 25 years later, the fans are still crap."
In the 1960s mainframes consisted of many tall cabinets. An array of small exhaust fans were mounted in a tray to vent out of a cabinet's mesh top.
When a fan failed - its replacement was labelled with the date. On a new prototype the fans were failing so quickly that the time was added to the replacement's label.
Rather than wait for an over-temperature warning - the engineers put strips of paper on the mesh above each fan's exhaust point. A large metal nut was used as the weight to hold down one end of each strip. These Medusa-like flickering vertical strips gave a quick check for when a fan had stopped.
A friend had a new-ish desktop PC that was rebooting itself intermittently in the middle of her word processing. She brought it to me - and reproduced the problem. When I burnt my hand on the cpu heatsink I realised that the fan was stalled by a badly fitted disk cable. The fan was dead - but jury rigging a larger one from the spares box fixed the problem.
maybe 10 years'ish ago I was working for a lab, our WAN connection was via a radio link to a building at a university, as the crow flies maybe 1km away. We suffered a very unreliable connection during a summer period, the link could go down for hours. Never got to the bottom of what was going on but the uni was building a new large tall copper clad building at the time that was in the Los from our site to the building where the B end was, was its omething to do with the hot sun and the copper cladding? or maybe tower cranes jobs etc getting in the way? All very odd and annoying at the time, both A and B end dishes were swapped out etc
A 64kbs downlink in the early days of satellite comms to remote studios: many but not all afternoons the link would die in the Delhi office. Spent several days out there testing and trying... and eventually discovered a vulture liked to roost on the satellite dish LNB support arm, pushing it out of the focal point.
That reminds me of the story of a guy who had a problem with his car. He apparently reported the problem that when he went to his local corner shop to buy ice cream, if he bought vanilla his car wouldn't start when he went back to it, but if he bought chocolate it started fine. It turned out that vanilla was at the front of the shop by the till, and other flavours were at the back, and the car would flood something if it was started within a short time from switching off, where it would be fine if given a minute or two longer to drain.
That reminds me of a Kia car that I ended up with for a while that would refuse to start (seemingly at random).
weather made no difference, time of day... no difference.
if it started it would be perfect, if it didn't start you could spend the rest of the day trying and failing to get it started, only to have it fire up perfectly the next morning.
Eventually it was worked out that if you even touched the throttle before turning the key it would flood and any attempt after that would only make it worse. Leaving it alone for around 3 hours let the excess fuel evaporate off and then it would start.
(Posting anonymously because of implications)
A friend of a friend worked at the local [Car_Maker_Which_Is_THGTG's_Prefect's_Fist_Name] manufacturing plant as a technician. One day, a customer comes in with an issue on his almost new Ec* Sp*rt which any of the service centers he visited was able to diagnose/solve. Whenever he passed 120 km/h, the stereo would up the volume by itself until it was maxed out. Nobody at the service centers believed the customer. Then the customer begged this friend of a friend to take a ride with him down the highway, and upon hitting 120 km/h voilà! the stereo went all out.
The friend-of-a-friend technician asked the customer to leave the car at the plant and promised to call back once they had a solution. After a month of diagnosing, conference calls with the Mothership at the US, and general candle lighting and praying with no results, the friend-of-a-friend tech decided to seach at the lowest level. This led to the customer getting really agry about not getting a call back and visiting the plant. Then he entered the storage deposit where security escorted him to, only to find his car completely torn down to every single piece. All was stripped from the body: transmission, engine, suspension, brakes, control arms, upholstery, even the cables.
Naturally the customer had a nervous breakdown. The company ended up reimbursing him the full price, and then they grabbed all the pieces of the car, put it in a container and shipped it back to the Mothership for further investigation.
Source: I got the story from the friend-of-a-friend technician's mouth maybe 8 years ago, when exchanging stories about tech pieces and spooky action at distance over a few beers.
As far as I'm aware, this is a common feature of factory fitted car audio systems. In VWs (Blaupunkt?) it is/was called GALA, and could be adjusted or disabled. I agree it shouldn't continue to crank the volume up to maximum, though it seems odd that the service centre would completely strip down the vehicle. Presumably they tried simply replacing the stereo?
Also same in GM (vauxhall) cars with Phillips stereos from the turn of the millennium. The service manuals even allow you to set all the parameters for that, default power on volume, behaviours, what signal strengths to count as a 'valid signal' when using t'wireless, etc... Work of art they were
The worst key I ever had the misfortune to drive belonged to the people l worked for.
It was a new (at the time) Renault Megane - the model that looked to be designed by people who didn’t know what the others were doing.
Amongst the bad faults was the fact that if you were driving with the radio on and a traffic alert came through the dashboard shut down until the alert finished, it also had some very odd brakes, and a lot more.
The worst bit was I dropped the car off with a list of issues. The next week I saw the person who I had handed the keys to, the next driver apparently looked at the list and said fine and took it out!
Who the fuck invented traffic alerts? I used to drive down from Herts to the west country to go diving many weekends, 6:30 on radio 4 there was normally some pretty good comedy. It would invariably be interrupted by a traffic alert from some pissant station two hundred miles away and of no relevance and I'd manually switch back to R4 to hear the studio audience dying of hysterics and then miss the rest of the program arguing with the missus about turning off the traffic alerts in her car as I've never heard a relevant one. Why the fuck I'd want to hear about a Norfolk tractor accident in Dorset is beyond me,
Or switch off "TA" altogether. Again, should be standard in any RDS radio.
The way it works is that when you have selected the TA function the radio searches for a broadcast which has the "TP" (Traffic Programme) flag set. Very often this will happen as soon as you start the car, so it will find a station reasonably local to your starting point, but that depends on the way radio works. In Cardiff, for example, you could easily pick up a station in North Somerset, which is a bit pointless.
The really tricky bit is what happens next, and this depends on how the radio stations are set up.
In the bad old days, independent stations often put "TP" up, whether or not they had any traffic bulletin planned. On top of that they don't implement the "AF" (Alternative Frequency) list properly. Most ILR stations only have one or two transmitters, so often (maybe things are changing in these days of massively syndicated stations, but this is how it used to be) the AF will only list one alternative frequency - the station's other transmitter (if it has one).
Thus as you drive away from your start point the radio will keep listening to that first radio station as long as it is able to decode RDS with any degree of accuracy and only when the signal fails altogether will it search for another station with TP set.
BBC Local radio used to work differently. The AF list will include adjacent transmitters of other Local stations, but some radios won't retune because the PN (Programme Name) is different.
So you are listening to Radio 4, which doesn't usually broadcast TP, and the radio is "keeping an ear on" some other, often quite random station, which does have TP set. Good radios will actually have a second tuner for this job, but not-so-good radios used to dip out every now and then, for a fraction of a second, retuning briefly to the other station. This second radio (or lack of) is also the way the radio determines when it is time to swap to another frequency in the AF list.
When the traffic bulletin is about to be broadcast, and often triggered by the jingle, the transmitter of the TP station will also set TA (Traffic Announcement). At this point, whatever else you are doing is interrupted while the main radio retunes.
ILR would then keep TA active until the end of the sponsor's jingle at least, and often until a little while later, maybe until the presenter had made a link or an ad break had played. Once TA is cleared, the radio will retune to the original programme.
It all sounded very useful when RDS was invented, but it does tend to get abused... which is why any decent radio will be able to turn such functions off.
Hwyl!
M.
Part of the issue here is that car radios have these buttons with things like TP and AF on them that hardly anyone ever uses, so they don't know what they do, having at best long forgotten and at worst, never actually known.
Then one day one of them will get accidentally pressed. Without you knowing.
And suddenly your radio is doing weird stuff. But it's never clear what's caused this. The display, if you have one, doesn't seem to flash up a message saying "TP activated" or whatever. At least not in any of the cars I've driven over the last decade or so.
I've always assumed that this system was designed by a jobbing GUI designer during a slow period.
Good point. The Phillips radios in our Renaults had "TP" and other flags on the LCD display, so if you knew what they meant it was obvious what was going on. Our Citroën, on the other hand, doesn't have anything obvious on its multifunction LCD monitor and on top of that it will usually (but not always - haven't quite worked that out) auto re-tune to DAB if the FM station has a DAB simulcast, and I don't think TA etc. works in quite the same way on DAB.
M.
Bad ground.
You know how in networking, weird shit is always DNS? In automotive electronics it's always bad ground.
The tough part is that lots of the intentional signals (like CAN) and unwanted noise (like that radiated from spark plugs or fuel injectors) doesn't always see a nice shiny wire as an easy ground path.
Years ago my car was playing up ( maybe late 80s) I called RAC one day, on the way to work because it had started to run so unevenly.. And mentioned to the RAC guy that it had to be electrical because of the clicking I could hear on the car radio. It got faster as the engine got faster. He just dismissed this out of hand, "Nothing to do with the radio". Really patronising. Of course, that's not what I'd said. I did not say it was caused by the radio.. He couldn't find the fault and I got towed home.
Where I checked the electrics, just in a simple way, I don't know much. I think I bought an inexpensive device that helped me to detect it. A break inside ( if I remember correctly ) a carbon HT lead ( if that's a thing). Whatever it was, it took me about 5 minutes to fix.
In a similar vein I had a Renault 20 that lost its brakes. RAC came, said there was nothing wrong. Car seemed to be working perfectly. Drove home, somewhat nervously, and it was indeed OK. Stayed OK until the next time I drove in traffic. And the brakes went again. That time I waited a bit before I called the RAC. And found that the brakes were working again after about 10 minutes. I took it to my garage, immediately. Where they heard what I had to tell them, muttered "Discs breaks binding" and had it fixed by the next day.
I think 30+ years later the RAC have improved. I hope so.
I used to commute into central London by motorbike. One fine day on my way into work I filtered to the front of the queue at a roundabout, whereupon the bike died. Wouldn't start. Wouldn't even turn over. So after a few minutes of head-scratching I called my breakdown service (I can't remember which one), who showed up about an hour later. The nice man also fiddled and jiggled things, and also couldn't get a peep out of it. And he didn't have the kit to transport a bike, so I had to wait another half hour for someone who did. Breakdown Service Man Number Two also had a go at getting it going, nothing doing, loaded up the bike and took me to a local garage of my choice. Where the mechanic took one look at my bike, turned the kill switch back on and it fired up first go. Looking back, as I got to the roundabout I had filtered past a van and knocked the kill switch on its wing mirror!
The only reason I don't feel more embarrassed about this is that I feel I was in good company :-)
Had a VW Polo that had weird intermittant electrical faults. Every few months, the electric windows wouldn't open. On the same time scale, the radio would just cut out and display "LEARN". Towards the end it would start misfiring and cut out.
All these faults were (temporarily) fixed by the classic "switching it off and on again"
Funny you should say that. I used to run a support department for the Southern African SCO Unix distributors in about 1990. We only provided support to resellers except in exceptional circumstances, usually when some large corporate client was ready to throw out the whole setup.
I remember being flown to Lilongwe in Malawi and then catching a bus to Blantyre to go to the Mercedes Benz agents, where they were having unexplained freezes of the fancy IBM server hardware. The software application guys were blaming the IBM boys, and the IBM hardware geeks were blaming the software blokes. Eventually they decided to gang up and blame the operating system. Now, as operating systems go, despite whatever brain dead antics the corporation got up to later in the nineties, SCO UNIX was solid as a rock. It never froze unless there was some sort of hardware, or should I say physical, issue.
I was there for a few minutes when I noticed a long RS232C cable running along the wall and out into the distance. I asked them where it led, and they told me it was attached to an HP serial printer/plotter at the far end of the workshop. I then asked them if they were using three phase power, which they were, and then I asked them if there was a common ground across the phases. You see, I had come across this type of situation before. There was a fairly hefty voltage coming up the ground wire of the serial device, which was on a different phase to the server, thus causing the server to fail.
My advice to people in those types of situations was: "Call in an electrician, but not the one who installed the setup originally."
Both the software guys and the IBM guys were suitably convinced that it was someone else's problem, but not mine.
I had a Rover - a 214 with K-series engine which would cut out if you went around a left-hand bend with less than a quarter of tank of petrol. Friendly mechanic who managed to solve every other problem with every other car I have ever owned completely failed with that one - though he initially thought it would be something very simple in the tank - so the only solution was to keep the tank topped up at least to half way. If you happened to be "caught short" you dipped the clutch, coasted around the bend and as soon as the road straightened, lifted the clutch, whereupon it would bump-start and carry on as if nothing had happened.
It was only a real problem if you happened to need to go around a left-hand bend and uphill at the same time, and was quite a good conversation starter.
That car was two years old when I bought it with 97,000 miles on the clock, and I thought it had done well when I got rid of it a few years later with 198,000ish miles. My current car, a Renault Modus with 1.5l dCi engine has just crossed 198,000 miles in nine years and still regularly makes 65mpg or more.
I bet now that I've boasted about that it'll die next week.
M.
"engine which would cut out if you went around a left-hand bend" - I had a Suzuki, where the gearbox would shift down two gears when doing a 90 degree left hand, causing the engine to overrev, hit the limit at 7500 rpm and then cut out, causing a lot of noise, a jerk and a spluttering stop - with any pedestians in the vicinity running away as fast as they can.
IIRC VW had customer complaints - possibly even accidents - where the car had accelerated when they were braking in an emergency.
It only happened when they had a front seat passenger who was also a driver. In the sudden situation the passenger would instinctively jam their "braking" foot against the footwell wall. The metal would flex enough to distort a cable linked to the accelerator.
I was hiring a jeep in the Caribbean while suffering from jetlag. It was an automatic (?) and I started the thing and attempted to manually put it into gear which involved my left hand reaching for the gear lever and it being LHD opening the door while simultaneously trying to push down the non-existent clutch unbalancing myself and falling out onto the floor next to the nice lady who had just rented me her lovely new car!
Somewhat more esoteric, and much more expensive, was a chap I know who rallied a Lancia Stratos quite a long time after it's heyday. It was a wonderful thing to see (and hear), but he failed to finish events due to mechanical failure more often than he got to the final time control. The failure was almost always the gearbox, and happened on right-hand bends (I think - it's been a while!) Eventually, the owner and his service chief went to Italy to track down a long-retired drive-train engineer from the Stratos factory team. The conversation went something like, "What gear are you in when it happens?"
"Hmmmmm, changing from third into second. Why? " (again, the specifics might be different).
"Ah, yes, that has always been the case with the Stratos. [Long discussion about why it happens, which I can't remember.]"
"Really? Why isn't it documented?"
"Because it kept people who bought them coming back for new gearboxes."
"OK, so what did your team do to cure it?"
"We told our drivers not to change from third to second in the middle of a right-hand bend..."
Back in the day I had this Audi. While driving on a busy motorway in the outside lane I lost all power, the engine just stopped i managed to steer it on momentun alone to the 'breakdown' lane without an accident. Opening the engine bay I eventually found the distibutor cap with all 4 HT cables was detatched. Looking at the caps interior I found one lug had been smashed. Called the AA. he fixed it my tapping a screw into where the lug had been - fixed. Except the cap still on occasin detatched itself but at least reseating got it working, the screw remained. No one had any idea why it was doing this. Solution - one of the early electronic contactless ignition replacements. it was one of the most irritating intermittent faults I had with that Audi, but not the last.
I got that in a Polo entering a petrol station. The entrance road when down a steep dip, to traffic lights, then up a hill. Because the light was on red, I stopped, and heard the engine die. I then couldn't get up the 100m hill into the petrol station. A couple of other drivers got out and helped me push it to the pumps.
Draining and flushing was (IIRC - it was many years ago now) one of the first things the bloke did, as was checking the baffles, looking for blockages, all that sort of stuff. From memory he said that the sending end had several inlets, intended to mitigate exactly this sort of problem. I learned to live with it and it never really caused me any problems, other than not running the tank quite as low as I might otherwise have done.
That car had a distributor, but electronic timing. The spring on the rotor arm rusted away and I conked out about 180 miles from home late on (again, if I remember correctly) a Saturday evening. An acquaintance correctly diagnosed the problem, and it was fixed with the spring from a ballpoint pen. Fixed enough to get me home and to Halfords on Monday for a new distributor cap.
Had the same problem a year or so later, but as the cap was something like ten quid, it didn't seem to matter quite so much!
M.
Dunno - but as far as I remember I never did have a new gasket. I did have a new water pump at one point - the oil was beginning to show emulsification and I assumed it would be the head gasket, but the mechanic spotted the water pump.
I suppose he might have done the gasket anyway at that point, particularly if he'd taken the head off to check. I can't remember. Sold that car some 19 years ago and it was still running fine.
M.
Had a co-worker tell me the tale of a rogue temperature sensor. Said sensor controlled the temperature for an office space. Early in the morning, everything would be fine. Midmorning, employees would start calling complaining that the room was cold. By mid-afternoon, it would return to normal. Sensor reported the temperature was right on setpoint the whole time. A tech dispatched to the office (in the afternoon) verified that the sensor was reading accurately. This went on for quite some time, until my co-worker got fed up with it and went to investigate.
He sat and watched the area one morning. Nobody got anywhere near the sensor (thoughtfully mounted in front of a window), but when the sun got high enough to clear the trees, it shone directly on the sensor, heating it up, which turned on the AC so that the sensor reading stayed on setpoint. Once the sun was high enough, the building itself shaded the sensor, so it went back to normal.
One small piece of cardboard taped to the window later, and all was well.
No, it was one of the Pet Shop Boys and a Doctor Who, the latter took the former's name actually, awfully modern :-)
...we should have a "dinosaur" icon, not aggressive, brontosaurus maybe, for old person stories like "Our first company cellphone needed the floor reinforced under it" and "I remember when variance from heteronormativity constituted a joke in itself and I don't mind admitting it".
Well, there is IGMC.
Back in the early '80s I worked at a Polytechnic, and was responsible for spending a chunk of the contingency fund at the end of one financial year, with the aim of producing a teaching room for what was called "Computer Appreciation". The purpose was to have lots of different types of hardware available so people who did not know anything about computers could see what they could do. The systems were BBC micros, of course.
One of the devices we found was a robot arm with 6 axis's of movement that was very functional for the price. To keep the costs down, they used normal rotary motors, with shaft encoders made up of IR transceiver devices, with a 4 quadrant rotating reflector on each shaft to 'bounce' the IR back from the transmitter to the receiver. It was all very elegant, and worked really well.
That is, until I tried using one in direct sunlight one day. Unfortunately the design did not have light covers on the encoders. As soon as I issued the return to home command to the controller to calibrated the position. It started all the motors (something quite interesting, because all the motor channels could run simultaneously, unlike most of the other non-industrial teaching robots), and proceeded to pull itself off the desk as all of the moving parts moved to the physical end-stops, causing the arm to contort in a way it was never really designed for.
We worked out that the bright sunlight swamped the IR detectors, so even though the motors were running, it could not detect the movement of the rotary shafts. I suspect that there was also a bug in the controller software that needed to see the shaft rotate before it would look for them to stop to detect the end stops, but I never knew that for certain.
The fall damaged some parts, so we went back to the manufacturer, and asked whether they could provide spares. We explained how the problem happened, and they said that they would see about designing some light tight covers, but although we got the parts to repair the robot (they came in kit form anyway, so replacing some of the parts was not a problem), we never saw any covers from them.
Shame really. They were the best educational robot arms I ever saw, so long as you used them out of direct sunlight. It's a shame that some of the HND students never saw fit to do their project with the kit available in the lab, as we had hoped. I often wonder what happened to all of the 'toys' when the lab. was replaced ( I left before the lab was dismantled), but I guess that some of the lecturers with their own BBC micros gave them a home.
Way back when, the project I was on had a support team over in Wales. Nice bunch of people, too.
There was one server which occasionally needed to be stopped and restarted. And this was all documented and (IIRC) fairly simple.
But for some reason, this process never worked for the Welsh lads. They'd try it, it'd fail, and then they'd call me out. And when I did it, it worked perfectly.
This happened a few times, until eventually they discovered that the restart would work perfectly if they were talking to me on the phone when they pressed the button...
I'm sure there's some boringly scientific reason for this, but I just took it as evidence of my Alpha-BOFH status. And charged accordingly for the call outs :D
EIV but by phone. There is a condition called Engineer In Vicinity whereby a machine that fails regularly will function perfectly when the engineer with eye-watering charges is in attendance. I had a Superbrain II that failed to boot from disk or exhibited some disk problem. The engineer was called and it booted flawlessly about 30 times and flawlessly passed a multitude of tests, Forms were signed and the machine functioned flawlessly until (in the days before mobile phones) the engineer drove past the security for her 3 hour drive home only to be called back to repeat the process the next day.
I remember a chemical balance that changed its reading as you leaned over it. It was one of these https://oertling.com/balances/top-pan/tp-series.php used to weigh out small amounts of expensive reagents. The reading would change as you leaned over it which made using it a tad tricky.
It was on a bench on the first floor of a building put together by some 'orrible '60s steel-framed pre-fab technique. The concrete floor slabs were supported on a triangulated mesh of steel tubes sufficiently flexible to move very slightly in response to the movements of whoever was standing on them.
We didn't have any problems with the replacement. It was replaced after the building was destroyed in a fire and the replacement building was a nice solid reinforced concrete job.
Super-high-accuracy balances are notorious for vibration or level issues changing the reading. I remember performing a test on one, and discovering that I had to have the room door closed AND nobody entering/exiting the next room in order to get stable readings. Balance room didn't have HVAC - on purpose, as the draft from it would have thrown off the balance, even with the draft shield.
we had a high temperature thermo-microbalance at work, you could weigh the sample(1), 0.897654321g then gently warm it up to 2000 K whilst recording the weight. This often changed in a nice predictable manner, especially when we flushed with argon to measure the delta-W without oxidation.
It was in a fume hood, on a massive marble table, sitting on a laminate sheet of cork, rubber, copper.
The four legs ended up in buckets of sand, supported on lead blocks etc etc. Overengineering!
This (Cahn) system didn't change weight when you went near it, tho' the older (Sartorious) was a bit iffy.
At one point, seeing as how an experiment could take a weekend, our Compaq 386 with 4 meg ram was doing a great job on data recording & we had a back up chart record. We started to notice big glitches on the normally smooth recorded data curves. Some days would record perfectly, other times had a few, or many big fuzzy incidents. It was resolved as we noticed a stiff brise one day, it was normally breezy, being near the coast....an isolated bit of the coast, ideal sort of place where you might put a research reactor. The brise was followed by a few more briseances, and we noticed that the next field was filled with soldiers playing games, with bombs. We negotiated a truce, and got the army to send us a fax about a day before blowing things up, such that we wouldn't run an experiment that day/weekend - and everything was fine, after that.
(1) superalloys, car-exhaust catalyst support materials, fusion reactor first-wall materials, carbon fibrous matrix...
I worked on a government IT contract once where we spent part of a winter uncomfortably cool. We pushed the thermostat to what should have been greenhouse temperatures, but it didn't help. Then one day somebody noticed that we had a LaserJet parked immediately beneath the thermostat. We moved it, and the thermostat became much more responsive.
No, just percussive maintenance.
"Back in the day", as they say, we had several large metal cased Microvitec monitors. For those not familiar with them they were literally cubic. There was a known fault where occasionally a connection would get too near the top of the case causing instability in the image. The common fix was to give the case a "tap" which usually worked, or of course...
You'd carefully explain to the user that the following process required special skills and should only be attempted by an IT professional. At which point you'd bang both sides of the monitor with your fists, usually causing the user and colleagues to just about jump out of their chairs. The only skill was in getting the maximum effect while not leaving any tell-take dents in the casing.
I was trying to recover data from somebody's stalled external hard drive, one of those things the size of a shoe box. After failing to get it working I asked: If you're prepared to treat it as a lost cause, there's one more thing I can try. "Go ahead". I lifted the whole thing 12 inches off the desk and dropped it. It started up, and we very quickly got close to 98% of the data off it.
I remember a couple of internal HDD's in the early 00's giving me the click of death and seizing. Stuffed in a ziploc and put in the freezer for an hour or three, yanked out and connected immediately to quickly spin up and recover data. The strange looks I got when putting the drive in the ice box...
I did this last year to retrieve data from a failed Acorn Risc PC hard disc. In this instance I used a USB to IDE adapter, so ran the disc in the freezer! The image was then transferred to an SD card, and with a suitable adapter the Risc PC now uses that as its hard disc - and it now runs faster than ever.
While most drives back in the day were susceptible to this in one way or another, some drives were more susceptible than others. Look up stiction. Some could be coerced into starting with a quick smack to the case at just the right place in the startup cycle.
The advice was always "It;s running NOW? Good. Whatever you do, do not turn it off until you have backed up everything important on that drive. Then replace it. It is no longer trustworthy."
Yep, in the past I came across some computer administrators that had the slightly daft, but understandable notion, that PCs should never be turned off, never cold restarted either if possible - at worst a soft reset. All because they had failures on disks when going cold and restarting and also SMART messages notifying them of non-critical failures.
The proper solution, of course, would be to replace the faulty disks as they had already failed, but instead they insisted on keeping them limping along chewing through power (nothing could be spun down just in case) and waiting for them to spontaneously fail in the near future instead. There's a balance to be had in the cost of HDDs vs power but most of these places had this rather wrong.
Back in the day (1960s), we kept all the computer gear in the Glass House at a specific temperature for the simple reason that Core Memory liked it that way[0]. This meant leaving drives powered on. Drive hardware was pretty flaky back then, and the drives would sometimes fail on the rare re-start. As a result, the myth came about that all drives failed on restart. Which carried over to the world of PEE CEEs in the '80s.
Intelligent drive monitoring started with IBM's AS/400 line, with what IBM called "Predictive Failure Analysis" tools. These came to be known as "PFAAARGH!", for somewhat obvious reasons. This was in the early '90s. SMART drives came from an industry consortium in the mid '90s, with the first units in the market by late '95 or thereabouts.
The biggest change I witnessed with the advent of SMART technology was that it gave Management yet another excuse to cut proper backups out of the budget ... "If the drives say they aren't about to fail, why do we need to spend gobs of money on backup?" was a question I heard all too often.
[0] Little known fact: Most of the power consumed by Core Memory was used to heat it up, in order to keep it within the optimal working temperature.
Users / other non network types always think fixing network issues is done kind of magic.
1 guy the other week was befuddled as to why device up addresses change.
Others are mystified by subnetting.
ACL’s and fire walling blows people minds.
Load balancing and amending traffic on the fly is like McGyver on steroids.
Routing and switching are best left unexplained as.....,,,
Anon because I don’t want to blow your mind too :)
As a wee nipper on my Holls, I was spending as much pocket money as i dared on the Time Crisis arcade machine. (light Gun pew pew - with recoil!)
I had gotten further in the game than ever before, but this arcade machine was near a window, and during sunset, the sun was at such an angle that it blinded the sensor along the left edge of the screen. That didn't matter during most of the game, except near the end, where any bad guys along that edge became unkillable.
The only thing you could do was use a grenade as an AoE weapon. So I used the one I had, then died in the next section having no more grenades. Put more money in, used my single (new) grenade again. Repeat until I ran out of money;.
*shakes childhood fist at the sun*
My favorite was back when CRT's were still actively used and we had this one customer who complained that their monitor would have squiggly and wavy lines on it randomly. We eventually stayed onsite the entire day and got to see it happen...we replaced the monitor and the next day, same problem.
I went onsite again and was trying to figure out why could possibly be going wrong when it started doing it again. After a few minutes I realized I could hear a humming noise and after tracking the noise to the opposite wall I asked, what's in that room?
Oh, that's our printing machine for T-Shirts. Apparently the damn thing covered the entire wall on the other side and was nothing but some big electrical magnet (600V power). Everytime they switched it on, the monitor went beserk but no one made the connection there.
Besides telling them to move the system out of the room (surprised the HDD never scrambled) I told them maybe it wasn't such a good idea for people to be in the room either..unless they got some lead drapes for that wall...
Every time he came near his laptop keyboard it would shut down the screen. He was baffled.
Till I pointed out his 'chakra balancing' copper bracelet also had magnets on it, and he was tripping the magnetic sensor into thinking the lid was shut.
Remove braclet. Problem solved.
His chakras however were misaligned.
I used to keep a magic wand and crystal ball in my desk drawer for when people expected me to perform miracles (I had a reputation for them)..
Once I had a call from one of the users. He proceeded to describe at length some problem (I can't remember what it was now). At the end of the call he said "I'm not sure if this is even possible. Do you think you can do it?". I replied "Yes, I can.". He said "Oh good! When do you think you can get it done by?". I replied "Five minutes ago."
He took so long and went into so much detail explaining what turned out to be a fairly trivial change, that I did it halfway through the phone call.
During the war our early warning, such as it was, monitored the German Luftwaffe in France gathering for a raid so they would scramble to meet them.
Several times they would scramble and they raid would strangely disappear before they could intercept.
After a while they noticed that this happened in the early morning but not at any other times.
Eventually it was realised it wasn't a raid at all, but the Sun rising and so radio astronomy was born.
If only the Germans had known they could have caught us with our pants down by launching at the right time.
Reminds me of this quote from Paul Fellows about an early Acorn Archimedes:
... we had one of the machines that we just could not get this thing to boot reliably. You could boot it, turn it off, reboot it and sometimes it would work and sometimes it wouldn't. It turned out, it would boot fine if you left it long enough, but if you didn't turn it off for very long then it didn't reset properly, and this was because the fan on the board was still spinning and the back EMF on the fan was enough power to keep the ARM running. And that's why you've got an ARM in your phone today, it would take no power to keep a 3 micron ARM with 25000 transistors would run for 30 seconds off the energy stored in the fan.
They cheaped out by not using a standard external reset? How unlike Acorn.
Sounds like a variation on the story of powering up the first ARM test silicon. It ran just fine, which everyone was pleased about because that rarely happens with the first iteration, then someone noticed they hadn't connected up the power supply and it was running on stray power coming in via the I/O pins or somesuch.
And at the opposite end of the scale, computer fans are quite good for demonstrating why perpetual motion is impossible. After demonstrating that I could light an LED by connecting it to the leads of a computer fan and blowing into the fan, I then demonstrated that there was quite a lot of difference in the power required to run an LED and the "generator" fan's twin, but connecting the two and asking the class to volunteer to blow.
A can of compressed air usually did the trick, but the second fan would run slowly enough to make the point.
M.
An unattended remote job entry card reader & printer was intermittently losing its connection - but was always ok when we were in the room. One day we dashed to the terminal room in time to see a cleaner flicking his feather duster over everything. The modem connection had a free-floating plug/socket that had a dry joint on a wire. After being disturbed by the duster - it always rolled back into the "connected" position.
I worked with an office manager who was - shall we say - a bit odd in some aspects. She was not the most patient of people, and at the time was expecting the End of the Word - really. Her pastor had foretold just when it would occur. She also had a spouse who was a professional bowler on a long losing streak. There were other personal issues as well. Every so often, she give me a yell, literally, and I would go to her desk and find that her computer was frozen. She could never reboot because she would lose work. This was in the early days of Windows and that might have had something to do with it. At that time we referred to the compiled code running on the systems "programs" - not "applications" or "apps," BTW. I believe term appeared a little later with the advent of java. Any way, I would walk in, shoo her away from her desk, sit down and "magically" the machine would behave flawlessly - with an innocent cooperativeness that had her gritting her teeth. The best hypothesis I ever came up with is that she had some form of buffer problem that cleared in a fairly brief interval, she did all the graphics and page layout for reports we generated, and her machine was limited to 640K. Either that, or our boss was routing pr0n through her system to his. My "fix," such as it was, was to tell her that electrons were sulky beasts and that they didn't like people being impatient with them. What she needed to was to sit back, clear her mind and breath. Oddly enough she followed the advice and it seemed to work.
"programs" - not "applications" or "apps," BTW. I believe term appeared a little later with the advent of java.
Nah. Where I worked, we had standardised directories off the root on our MS-DOS HDDs. c:\apps was where Wordstar and Supercalc and similar "apps" lived. there was also c:\tools where more specialised stuff like Norton Tools, 4DOS and the like lived.
"programs" - not "applications" or "apps," BTW. I believe term appeared a little later with the advent of java
Well, Acorn was using the term "applications" with the Archimedes long before 1995, which is the date that Wikipedia gives me as the inception date of Java. Not sure whether they started with Arthur (1987) or with RiscOS 2 (1989) and I'm certain Acorn weren't the first to use the term in that manner.
In Acorn's case it signified a slight difference from the "programs" of old, which tended to be a single lump of executable code, called directly by the user. Acorn's Applications were actually directories (folders) containing not only the main executable but also any support files required, other than shared components. A double-click on this special directory (signified by a name starting with "!") usually caused the OS to search for Acorn's equivalent of a batch file named "!Run" in the folder and use the instructions contained therein to launch the application.
I've just realised how similar the basic concept is to the systems such as Flatpak or 0install... oh, I see Wikipedia has an article.
M.
But when my department started getting optical mice instead of rollers, I had a coworker complaining about random mouse movement. After checking drivers and connections, I noticed his mouse pad had a lenticular surface that was great traction for a physical mouse, but reflected the optical beam at random angles. One new mouse pad solved the problem.
This reminds me of my old tech support job when dialup was the way you got online from home. I was befuddled by a user that could not connect, and nothing I tried over the phone worked. So I decided a house call was in order and drove the 50 miles to the customers house on my own time just because I needed the challenge.
Upon arrival, I connected my laptop to their phone line and attempted to connect. Whilst the loud negotiation tones rang from my speakers, I noticed the sound cut out once every second. Having grown up on a farm, this sounded suspiciously familiar. I picked up on the phone line and heard an even more familiar ticking sound that told me exactly what the problem was. I asked the customer if they had recently installed an electric fence, which they confirmed. After unplugging said fence, Internet connectivity was promptly achieved.
Go figure.
Back in the early 80's I worked in Saudi commissioning Phillips PRX/A 205 telephone exchanges. One exchange (RMR2 - Riyadh Mecca Road 2) used to randomly reboot, this was a very rare occurrence so a lot of resource was placed on solving it. Long story short: the outgoing trunks all went via a microwave hop to RIPX, an Ericsson AXE10 exchange. These were newer exchanges compared to the SPC analogue 205 so were probe to restarting individual processors when in trouble. These restarts could take a couple of minutes or so anyone dialling outside RMR2 got routed to a voice announcement telling them "All trunks busy", this announcement could only take 10 simultaneous connections so when the 11th person was routed to it the exchange crashed..... "Cause and Effect! :-)
A former boss had a Surface that mysteriously started registering random pen touches. When he brought it to me, it seemed fine, which he attributed to my healing presence. As soon as he was back at his desk, it starts going crazy again. While trying to isolate the problem, I was annoyed by his banker's desk lamp, which was spitting out a 'white' light with a CRI in the low teens and making my hands appear diseased. As soon as I turned it off, the problem solved itself.
It turned out that the final incandescent bulb from the drawerful he'd lovingly hoarded had finally burnt out and the cleaner had helpfully replaced it with the jankiest, EMI spewing CFTs that Eastern Europe could carefully manufacture. Replacing it with a nice high CRI LED bulb sorted his Surface issues and also helped bathe his desk in a warm glow that didn't give skin the appearance of a terminal illness.
I had a coworker confused as the laptop he was working on kept going completely dark--screen, power light, and everything, then it would mysteriously wake up and resume exactly where it left off. It was not crashing. I had experienced this before though and realized it was because he had it stacked on top of another machine and the speaker magnets from the machine on the bottom were tripping the 'lid closed' sensor, which is magnetic on Dells.
I had a number of powerful neodymium hard drive magnets so I concealed one in my hand and proceeded to tell him that it was his body's electric field that was causing the problem, and that it happened to me all the time. I then proceeded to wave my hands over 2 more machines causing them to blink off and on, much to his astonishment as he couldn't get the same thing to happen. It was amusing to see him waving his hands about though. After letting him stew and scratch his head for a while I explained what was really happening..
Back in the Good Old Days of CRT monitors, one of our desk jockeys had a problem with a monitor: it was getting what appeared to be 60Hz interference. I couldn't work on it during the day so after several late afternoons of trying a number of things (including swapping CRTs a couple times, checking for nearby fans and so on) I finally hit upon the issue. It seems his desk was directly against the wall that contained the breaker box where all the wiring came through for half the floor. Access to the breaker box was in the hallway on the opposite side of the wall, so it wasn't obvious from inside the office. His desk was perpendicular to the wall and the CRT was on the end closest to the wall, so it was getting full benefit of some humongous amount of AC flowing mere inches away.
I moved some of his tchotchkes aside and placed the CRT on the other end of the desk, ran a quick test, and left a note to "try it now". Next morning I explained what was going on, and he happily said that he'd just rearrange his desk.
It was that or re-wire the entire factory, and not all users are unreasonable. :) Problem solved.
A friend of mine had a similar adventure with an optical disks jukebox.
The customer had bought a display model that came with a plexiglas (or glass) back panel.
Since it was installed in front of a window, the sensors inside would get tripped by a summer afternoon sun.
Fix : an 180° rotation which made for much easier viewing of the robotic arm movements but made it slightly difficult to use the mailslot. :-)