Had the same issue a while back, had a hell of a time figuring out why one of my screens would flicker or shut off momentarily whenever my girlfriend sat down at her desk (which is next to mine). Even initially figuring out that it was the act of her sitting down that caused it took some time, with a lot of jokes about her telepathically messing with my setup in the meantime.
Turns out that the gas piston in her chair (not an IKEA chair in this case) has a bit of "give" to cushion oneself when sitting down, and that compression caused some kind of electromagnetic pulse (I assume?) strong enough to mess with the monitor.
I do wonder if it's perhaps bad for the monitor's lifespan, but it only affected my cheapest one, and with the cause found I can live with it.
There is a noise filter that you can clip onto the power line or other cables to help prevent radio transmissions from going into the line or other sources of noise.
A friend of mine bought IEMs with a microphone (Moondrop Chuu) and it would pick up radio frequencies that you could hear through a voice call. These didn't end up fixing it, I suspected it was bad grounding but we never ended up solving what the root cause was.
I had exactly this issue with a chair gas piston and a large tablet display (drawing tablet), which is presumably more sensitive to EM/ES interference than ordinary screens since it's large, not battery-powered, and has both touch and stylus support.
I have this problem too! I have been trained to roll away from my desk before I stand up. I assumed it was static, and my hat's off to you who tracked this down.
Remotely related, my screen makes a noise, somewhere between static and a whine, while displaying certain text files. Minimizing or closing the editor stops the noise, getting the text back on display restores the noise.
I too have one screen that makes a noise, akin to coil whine, only when a certain spreadsheet is open and the window expanded past a threshold of the screen's real estate.
Naturally, its the speadsheet that I use most often in my duties.
Perhaps a silly question, but does turning it off and back on again also resolve the problem?
I can occasionally hear such noises from monitors, and have always thought it was some kind of interaction with the phase of the AC power and some kind of internal physics of the monitor. Generally, turning it off and back on again fixes it for me.
I don't honestly think turning it off and back on would do anything, either. I also don't have one of those chairs, nor do I want to buy one and use it to test the theory. But, turning it off and back on again is a simple and easy thing to do that should be reasonably safe for the equipment. There is, after all, a reason why power cycling equipment is often a first step to diagnosing and/or fixing weird problems. :)
When we were teenagers, my friend used to call it "waving a dead chicken". He coined this term to describe the way he would resurrect dead inkjet printers that even I gave up on - by disassembling and reassembling them until they started to work again, while being perfectly open that he has no idea how it could fix the problem, just that in practice it often did.
While this was just a funny term and pretty absurd approach for fixing things (even though it worked!), I took away from it an important realization: the scope of possible causes of a weird, randomly-occuring problem is much larger than I'd normally assume. Over the years, I learned to identify some "outside context" things for computers - ESD, thermals, UV exposure, RF interference, voltage spikes in power lines, devices being almost but not quite connected. Because of that, when in a bind, "waving a dead chicken" may just be called for - in forms of e.g. percussive maintenance (hitting the thing with a wrench), moving things around, switching cables, disassembling, etc.
I realize now I have never thought to keep track of a sample which triggers the phenomenon, and I can't really find time to probe for one right now.
I'm not entirely certain, but as I recall it switching it off and back on again is not effective against this issue. I've tried, no doubt, but the only thing noted which seems to have an effect (while the screen is active) is removing the text from display.
sometimes when manipulating 3D CAD models, I can get a similar noise from my graphics card / monitor at certain orientations of the model sweeping back and forth as I rotate it
I have a feeling that grey #A5A5A5 is row-hammering something
Same here; my PC GPU would make a high-pitched whine in certain conditions, most often encountered in a CAD program when I was rotating things around. I assumed it's the dark-grey coordinate system / grid on lighter-grey background making some kind of digital equivalent of Moiré patterns on the traces in the GPU card, that happen to generate an audible frequency.
Similar story, my computer makes a small whine when I move the mouse cursor. Not the screen or speakers, the actual PC. With both wired and wireless mice.
Spreadsheets are what do it for me. Haven't noticed it on anything else. Only whole-screen spreadsheets. Opening the start menu (covers maybe 1/5th of the screen width and half the height, so 10% overall I'd guess) is already enough to break the pattern. At 110% or 90% zoom it does not happen at all, it needs to be the default zoom level. It's also noticeably less if there are colored cells.
You can dial the screen's volume by making the libreoffice window partially transparent (I have that bound to scrolling on the title bar). This is on an Acer 1920x1080 (~23"?) screen from around 2009.
Indeed. It's the frequency of pixel changes causing ripples in voltage rails, subsequently causing inductors in switching voltage regulators to physically resonate at audible frequencies.
There's specific test images you can find online, designed to maximally stress voltage rails in LCDs. Lower end monitors can actually get enough voltage ripple that the image quality visibly degrades.
Is this a Samsung G-series (G7/G9)? I returned a G7 that would make a high pitched noise and also dim the entire screen when certain patterns displayed. It also did this very strange 'interlacing' behavior which was apparently a known problem with Samsung monitors. Maybe something similar for you.
In my case it's an LG monitor from the early 2010s (IPS). Now that you mention it, there might be a slight flicker and dimming effect on mine as well when it happens, but if so, not very salient compared to the noise.
Wow, I can't believe this. I have a Markus Ikea chair and a dual screen setup with a USB c hub connecting them all.
The screen likes to turn off all of the time and it drives me nuts. A good smack quickly fixes it. I never considered the chair. Mind blown with static discharge.
You might want to look into some extra shielding for your mind. I hear having it blown by static discharge too many times can cause brain damage. ;)
In all seriousness, yeah, this is crazy! IMO, IKEA should either recall and fix, modify the design of, or stop selling these chairs. ESD can seriously damage equipment, and I could easily see there being cumulative effects from something like this.
I don't think it's classical high-voltage ESD reaching the monitor; it's RFI generating enough voltage to mess with the HDMI signal. Voltages generated by RFI will be relatively low compared to direct ESD.
I've got the exact same problem and now I'm nearly positive (ha ha) that it's being caused by my mesh-backed AmazonBasics chair. Truly one of the most personally useful posts I've ever seen on HN.
Had the same issue with my Secretlab. The softweave fabric makes static like crazy an when getting up the main body comes in contact with the roller base which is shaped like a star thus working as an antenna. Big ESD pulse, all Display port screens flicker. Solved by using a towel where I seat.
> Fast drying, anti-static coating eliminates static charge and reduces triboelectric generation from flexible surfaces.
The page unfortunately doesn't explain what this is or how it works. Is it a conductive coating so that it can dissipate the electricity into something else, or does it form a shielding layer instead so that it won't zap stuff?
I don't know which device(s) are at fault but when I get off my (non-Ikea) chair[0], my keyboard sometimes stops working. It's easy to resolve by double-tapping the 'switch input' button on the USB switch, although sometimes the keyboard doesn't work until I click my mouse (or maybe vice versa). Anyway, I've gotten used to it and don't have any motivation to diagnose the issue further.
Wow. Maybe this is what's been happening with my screen. When I stand up, it resets. I figured it was ESD but had no idea from what or how. I have a secretlab chair though.
I also have a Secretlab but I have an ESD mat on my work surface which is grounded with a clip at the corner, and I tend to put a hand on it to stabilize myself as I stand up, which has the side effect of dissipating any charges as they're generated.
I also do electronic assembly work at this desk so it's sort of a no-brainer to have the mat, I just had no idea it was also saving me from other weirdness!
I once had the problem that running make with too many parallel jobs (-j) would change my keyboard layout.
The machine was some laptop mainboard glued to the backside of my monitor, and the USB socket came out at the top of the mainboard. On its way down, the USB cable for the keyboard passed across the whole mainboard. On high load, the mainboard created enough interference to cause the connection to reset, re-hotplugging my keyboard, so the previous setxkbmap call was not effective anymore and i was back to the standard US qwerty layout.
I love when software and the real world have entirely unexpected interactions. And by love, I usually mean hate, especially when I'm the one that has to debug them.
These sorts of things make for good war stories. I find a trick to improve my attitude while dealing with them is to remind myself "this will be a good story at least, and I'll be glad it happened once I'm regaling others with it".
my favorite story on this is a town in north east usa changed it's traffic lights from energy wasting old fashion lights to new fancy low energy lights. The unexpected result: in winter time these lights did not melt the snow away and the lights were opaque with white snow.
I remember a story about a car that consistently refused to start after a family went to the store to buy vanilla ice cream but it worked fine when they bought chocolate ice cream. (My apologies for any inaccuracies in my recollection of it).
They had a manufacturer engineer verify it happened even.
Turned out to be an overheating issue. The chocolate ice cream was at the back of the store while the vanilla ice cream was at the front, effectively changing how much time the car had to cool down properly before attempting to start it again.
I currently have an RPi4 mounted to the back of my TV for easy Kodi streaming, but I never considered using the business end of a laptop. I'm sure I have a decade old laptop with a broken screen sitting at the bottom of a shelf somewhere...
I used to think this way, but then HN scared me into thinking that all this 5+ year hardware is much less power-efficient, so now I just... avoid the issue entirely, and don't do anything requiring hardware, old or new.
You may find it helpful to check how much power things actually use and what power actually costs where you are. I used to obsess over energy efficiency until I realized that all my machines were laptops that maxed out at like 40W running full bore and I'd spent hours saving like $0.10.
I invested in a frame.work laptop. If I ever purchase a new mainboard, I'll get this cooler for my old mainboard and use it as a server or mount it to a TV:
The fact that traditional parts makers like Cooler Master are putting out Framework cases and such bodes really well, I reckon. There are few things I've bought that produce the opposite of buyer's remorse, and my Framework is at the top of the list.
As much as it would interest me to learn that Cooler Master designed and created a product out of sheer enthusiasm for a small third-party vendor, it much more likely came about through traditional means.
Not long ago, Framework engaged in marketing and promotion campaigns to demonstrate aftermarket uses for their mainboards via direct outreach to popular creators on YouTube and Reddit. I spoke with one of the participants who confirmed that they were sent the mainboard and commissioned by Framework for their project.
Doesn't make any of it less cool, but there's no "secret cult of Framework" driving things, just smart business strategy.
Well yeah, didn't mean to imply there to be a secret cult of Framework (though if there was one I'd happily join it). More that it bodes well to see companies like Cooler Master giving Framework the time of day when it comes to these sorts of business deals.
Had a similar thing happen with "fake" (passive) PoE working fine but some types of network activity would cause the CPUs in the network devices to work harder, leading to voltage sag which would sometimes cause the remote side of the link to reboot or hang. The problem went away with a separate power supply for the local and remote side.
I once had to rip VHS tapes from family to digital. I had a VHS deck and a spare laptop with a little RCA plug to USB dongle to ingest content. My first few test rips were awful quality, full of analog noise and weird banding and just unexplainable signal degradation. I couldn't understand, because when I was just playing with the dongle the signal was great.
Eventually it dawned on me: I sat the laptop, right on top of the VHS deck while running the rips. The VHS head ended up directly under the CPU and HDD, such that, CPU and hard drive activity were interfering with the tape reading! I moved the laptop off the VHS deck and everything worked just fine.
So, a DIY setup. I guess that when you are making a production model (ex: iMac), this is the kind of thing you have to test for.
We had an interference problem at work once, in a VME rack, an I/O board with a particularly large coil was messing with the CPU board, solved by moving the offending board in another slot. The effect was mostly random crashes and reboot though, nothing fancy like a keyboard layout change.
Oh my goodness. I think this might be happening to me. I’ve had a semi-frequent issue where my monitor will suddenly power off and the only way I can get it to wake up is by unplugging replugging the Thunderbolt cable.
Wow. I had the same issue in my previous apartment (where I was using the Marcus chair), but I could never figure it out. I just always assumed its a faulty monitor.
Thank you for the link! I only knew the referenced folklore page, but never saw a picture and the following exchange. Was just about to post this story as well :)
Wow, so I wasn't crazy - exactly the same thing used to happen to me when I got up from my Markus and the screen would turn off for a second. My hypothesis was a loose cable which I bumped slightly, but not touching the table when getting up didn't work. Not sure why, but it eventually went away.
One of my first jobs 30+ years ago was ESD testing automotive and consumer electronics. I would spend a week with a discharge gun methodically running different discharge energies and waveforms both directly onto the devices as well as onto radiators at various distances from the device under test, as well as any cabling/harnessing that attached to the device.
Then the design team would figure out the reason for any resets or operational anomalies (or damaged components) and put whatever additional suppression was needed. Sometimes this required rerouting of traces to reduce coupling or redesign of the ground plane. It's a tricky business and expensive if you want to do it right. I suspect that your average $120 display does not see this kind of testing.
Cars are a terrible ESD and EMC environment and safety critical at the same time.
Computer equipment usually does pretty well, especially after installation (when there's good ground path everywhere and it's in a chassis). Between ferrite beads and TVS, there's a pretty good amount of protection, and I've seen pretty few ESD anomalies-- a reset or two when touching connectors long ago is about the extent of it.
RFI and EMC is another matter. When I was operating a 100W 30MHz transmitter indoors briefly, the computers around did all kinds of wacky stuff. Flashing screens, random mouse clicks (from wired mice), transmitting packetstorms on their own, etc.
As a kid, someone gave me a toy keyboard. Checking Google, it looks nearly identical to a Casio PT-1, but was probably a clone or slightly different model.
9 year-old me was delighted to discover that it would start playing on its own when it was near the plasma globe I had bought at a science museum gift shop. I couldn't explain it, but eventually came to a vague understanding.
It was semi-random, mashing together short, distorted sequences from the song bank stored in memory. Being almost recognizable made it more haunting.
I remember bringing this out one night during a sleepover and we all got kind of spooked. Fun times!
I am pretty sure I had the same keyboard and plasma globe!!!! IIRC, you could also get the keyboard to play when you set it on top of the SCSI external CDROM drive that came with my mac 2400c.
I had that keyboard and we sat listening to the spooky random music once, i did not know you could reproduce it with a lava lamp! I loved it it was even playing notes out of the register so low that the speaker could not reproduce. Pure magic
My dad's car had velour interior. Anytime the humidity was below 50%, expect the car to zap you and to be zapped touching door handles. I assume there are/were numerous gas station fuel fires from cars with velour upholstery.
One stupid little thing I realized after having sometimes been consistently zapped by mybcar door and other times not getting zapped in the same weather and humidity:
If I open the door and grasp for the (metal) edge of it before I start sliding out of my seat I will be grounded and consequently not build up charge as I leave the seat.
One of the worst shocks I've received in my life was not accidentally touching one of the prongs of a half-inserted 120v mains plug, but pulling one of my fleece blankets off of the other on my bed, and then getting my shoed foot within 6 inches of my metal bed frame. I almost fell over, grabbed the upper part of the frame with my hand, and received yet another painful shock.
I’ve had more than my fair share of 120v shocks, but the worst shock(s), by far, happened when blowing cellulose insulation into my attic. The 75’ (100’?) hose was a little unwieldy and I wanted more control in filling areas near eaves (small attic, lots of crawling), so I taped a 3’ section of 2” pvc pipe to the end of the hose. Holy hell… continuously, the static would build for a few minutes until it’d discharge through my gloves. First time it happened I pulled my glove off thinking it left a burn. There wasn’t really much I could do the discharge the static when I wasn’t near a vent or flue.
We had some carpet at my old workplace that would always charge you up and get you zapped. Whenever you touched anything metal. So the old and wise folks there immediately told you to grab your key when you get up and then use it to touch something metal, so the spark is between your key and the metal object and you don't feel anything.
I am a static electricity "magnet". I keep getting shocked a lot, painfully, in environments where no one else is, which is kind of bad for my mental health if I focus too much on it. I've learned all kinds of methods to cope.
For example, we have a few bar stools at home; sitting on one always primes me for getting shocked after getting off it. I figured a few rules, such as never wearing anything isolating on my feet, so I can dissipate charges by keeping one foot on the metal part of the chair; or, failing that, I make sure there's always a metal object in grasping range, which I can use to later discharge in a less painful way.
(Pro tip: don't do the metal discharge thing with the hand you wear rings on, and hold the metal item so it doesn't touch the underside of your fingers - getting a shock through a nerve isn't pleasant at all.)
Another thing: I always keep a metal coat hanger around the bedroom, so that whenever I have to deal with blankets, I can keep "swiping" them to collect charge and then transfer them away by touching something grounded with the coat hanger.
Also: I always have my keys on me when away, in an easily-accessible place, specifically so I always have a metal object I can use to offload static charge in a pain-free way.
Also: over the years I kind of habituated all kinds of subtle behaviors designed to keep me safe from getting shocked by my wife or kids. Basically, if I feel one of them just got charged (e.g. via the blankets or the bar stool mentioned before), or I haven't kept track of their recent movements in a static-rich environment, if there's a need or chance of any kind of physical contact, I instinctively first touch using my elbow or some other pain-minimizing way, just to equalize charges with them. My wife sometimes notices when I do it to her, but fortunately, she is quite understanding.
Yup, explicit TVS or just diodes doing clamping or whatever. There's a fair bit of protection circuitry around. Often there's series resistors which help a lot, too.
Not to mention that all modern CMOS chips have protection structures on the input (though not really as good as you'd like for something that is randomly handled by charged up humans).
> Cars are a terrible ESD and EMC environment and safety critical at the same time.
Had a bad incident a few years ago where EMI created by the windshield wiper motors on a large vehicle was causing voltage dips and spikes for our product. We had protection circuitry so that if the incoming voltage was too low, it would shut down our system cleanly.
Saitek (now logitech) sells flight simulator peripherals in the form of a throttle box and separate flight stick. They use two different USB cables and plugs instead of connecting one to the other, and they are technically two different devices. If you connect them to the same USB hub or to a USB system with not enough power, they will seem to work just fine, but will send a completely random button press or stick movement every so often. Changing ports will fix this.
Another logitech one: I have a g27 steering wheel/pedal set/shifter combo. The pedals plug into the wheel, which also connects to a power supply and USB cable to connect to the computer. If you plug the steering wheel into the power and DON'T plug in the USB cable, metal parts of the pedals "leak" current, and you can feel a painful sensation if you touch the metal with exposed skin.
Personally I remember having a pickup truck with a velour seat. During winter I would slide off the seat at a gas station and the first thing I would tough would be the gas pump handle - ZAAAP. Not what you want at a gas station!
That’s why in Europe the gas nozzle doesn’t have the little lock - It requires that you keep pressing it. So that you avoid going back to the car, charging static, coming back to the nozzle and having a spark.
This reminds me of the 1st electronics company I worked at. We didn't have any ESD equipment, so I built a circuit which took power from a wall socket and pumped the voltage up to 2kV using diodes and caps, and sent that through a simple human body mode resistor/cap (https://en.wikipedia.org/wiki/Human-body_model) to do ESD testing. It worked well enough for the simple testing that we did, and was successful at finding weaknesses of our PCB layout for static protection.
That's just bad discipline on your part. Just because the options are there doesn't mean you have to constantly look at your chair every five minutes.
My company is building a chair that doesn't compromise on features for when you need them, but elegantly hides them away when not in use so you won't be tempted every second of the day. We have room in our A round if you're interested.
When I was a kid we had to return my Commodore 64 for repair several times, only to be told that once they got to it, after it had been waiting at the store for a few days, there were no problems with it.
Turned to be a result of storing the C64 on a bench under the "large" 26" CRT... When it was kept away from it for a while and had a chance to discharge, everything was ok. But after a while near the TV, it started "typing" gibberish of its own accord.
You folks are all lucky. I have the same chair, but the ESD problem didn't manifest as screen flicker - it manifested as my work laptop bluescreening about a minute after me getting up from the chair. I wouldn't have guessed the cause of not for some random HN comment some half a year ago. The solution for now is that I don't use external screens with my laptop. One of these days I'll find a better-isolated display cable.
I have another version of this problem. Almost every time I get out of my chair, my Schiit Modi DAC disappears for a moment and my music stops, then starts again a few seconds later as the DAC comes back online.
I spent a fair amount of effort trying to figure that one out. Thought it was loose cables, something, but no. I don't have a Markus chair, but my Steelcase Gesture is still capable of making a pretty good amount of static electricity. Once in a while when I stand up if my boom microphone is too close I'll shock myself on it, and most of the time I have to reboot things after that. Haven't permanently killed anything yet, thankfully.
I have an HP OMEN laptop, W11 and RTX and a 165hz refresh rate.
When I plug the machine in, the audio will get all static-y from the built in speakers. It seems to especially happen when I am moving my mouse over a video thats playing - and the playback of videos stutters when pulling the cord in or out of the machine.
Battery life on this machine sucks though... but other than that, its fantastic...
I had a job in university testing breakdown voltages in various flavors of Kapton. IIRC the idea was to see how suitable they would be for spacecraft construction where they would be exposed to high voltages from solar wind. We could always tell when a test run was complete because the EM interference from >10 kV suddenly shorting to ground caused our computers to freeze. (Thankfully they still recorded the data we needed.)
213 comments
[ 3.0 ms ] story [ 277 ms ] threadTurns out that the gas piston in her chair (not an IKEA chair in this case) has a bit of "give" to cushion oneself when sitting down, and that compression caused some kind of electromagnetic pulse (I assume?) strong enough to mess with the monitor.
I do wonder if it's perhaps bad for the monitor's lifespan, but it only affected my cheapest one, and with the cause found I can live with it.
Maybe one of the cables is acting as an antenna.
https://www.amazon.com/VSKEY-Anti-interference-Telephones-Eq...
Incredible read on that topic: https://www.ranecommercial.com/kb_article.php?article=2127
Steelcase Criterion / Dell S2716DG
I suppose it's some sort of resonance phenomenon.
Naturally, its the speadsheet that I use most often in my duties.
I can occasionally hear such noises from monitors, and have always thought it was some kind of interaction with the phase of the AC power and some kind of internal physics of the monitor. Generally, turning it off and back on again fixes it for me.
But as this article describes, you never know!
I don't honestly think turning it off and back on would do anything, either. I also don't have one of those chairs, nor do I want to buy one and use it to test the theory. But, turning it off and back on again is a simple and easy thing to do that should be reasonably safe for the equipment. There is, after all, a reason why power cycling equipment is often a first step to diagnosing and/or fixing weird problems. :)
When we were teenagers, my friend used to call it "waving a dead chicken". He coined this term to describe the way he would resurrect dead inkjet printers that even I gave up on - by disassembling and reassembling them until they started to work again, while being perfectly open that he has no idea how it could fix the problem, just that in practice it often did.
While this was just a funny term and pretty absurd approach for fixing things (even though it worked!), I took away from it an important realization: the scope of possible causes of a weird, randomly-occuring problem is much larger than I'd normally assume. Over the years, I learned to identify some "outside context" things for computers - ESD, thermals, UV exposure, RF interference, voltage spikes in power lines, devices being almost but not quite connected. Because of that, when in a bind, "waving a dead chicken" may just be called for - in forms of e.g. percussive maintenance (hitting the thing with a wrench), moving things around, switching cables, disassembling, etc.
I'm not entirely certain, but as I recall it switching it off and back on again is not effective against this issue. I've tried, no doubt, but the only thing noted which seems to have an effect (while the screen is active) is removing the text from display.
I have a feeling that grey #A5A5A5 is row-hammering something
You can dial the screen's volume by making the libreoffice window partially transparent (I have that bound to scrolling on the title bar). This is on an Acer 1920x1080 (~23"?) screen from around 2009.
There's specific test images you can find online, designed to maximally stress voltage rails in LCDs. Lower end monitors can actually get enough voltage ripple that the image quality visibly degrades.
Not sure if it is the fabric or the piston, but definitely suspected static since I got the chair.
The screen does not seem to age prematurely because of this.
You might want to look into some extra shielding for your mind. I hear having it blown by static discharge too many times can cause brain damage. ;)
In all seriousness, yeah, this is crazy! IMO, IKEA should either recall and fix, modify the design of, or stop selling these chairs. ESD can seriously damage equipment, and I could easily see there being cumulative effects from something like this.
The page unfortunately doesn't explain what this is or how it works. Is it a conductive coating so that it can dissipate the electricity into something else, or does it form a shielding layer instead so that it won't zap stuff?
My connection:
WASM CODE V3 keyboard -> UGreen USB 3.0 switch -> desktop computer
I don't know which device(s) are at fault but when I get off my (non-Ikea) chair[0], my keyboard sometimes stops working. It's easy to resolve by double-tapping the 'switch input' button on the USB switch, although sometimes the keyboard doesn't work until I click my mouse (or maybe vice versa). Anyway, I've gotten used to it and don't have any motivation to diagnose the issue further.
[0] https://www.haworth.com/na/en/products/stools/very-0.html
I also do electronic assembly work at this desk so it's sort of a no-brainer to have the mat, I just had no idea it was also saving me from other weirdness!
https://emcesd.com/pdf/eos93.pdf
The machine was some laptop mainboard glued to the backside of my monitor, and the USB socket came out at the top of the mainboard. On its way down, the USB cable for the keyboard passed across the whole mainboard. On high load, the mainboard created enough interference to cause the connection to reset, re-hotplugging my keyboard, so the previous setxkbmap call was not effective anymore and i was back to the standard US qwerty layout.
I love when software and the real world have entirely unexpected interactions. And by love, I usually mean hate, especially when I'm the one that has to debug them.
As an example: many backpacking trip stories are type 2 fun.
Turned out to be an overheating issue. The chocolate ice cream was at the back of the store while the vanilla ice cream was at the front, effectively changing how much time the car had to cool down properly before attempting to start it again.
https://www.coolermaster.com/catalog/cases/mainboard-case/fr...
Not long ago, Framework engaged in marketing and promotion campaigns to demonstrate aftermarket uses for their mainboards via direct outreach to popular creators on YouTube and Reddit. I spoke with one of the participants who confirmed that they were sent the mainboard and commissioned by Framework for their project.
Doesn't make any of it less cool, but there's no "secret cult of Framework" driving things, just smart business strategy.
My wifi is on the same breaker as the outdoor outlets so any issues with water intrusion affects the wifi.
Eventually it dawned on me: I sat the laptop, right on top of the VHS deck while running the rips. The VHS head ended up directly under the CPU and HDD, such that, CPU and hard drive activity were interfering with the tape reading! I moved the laptop off the VHS deck and everything worked just fine.
It's such a Rube Goldberg kind of error, I love it.
We had an interference problem at work once, in a VME rack, an I/O board with a particularly large coil was messing with the CPU board, solved by moving the offending board in another slot. The effect was mostly random crashes and reboot though, nothing fancy like a keyboard layout change.
Time to run some experiments
https://github.com/jwilk/zygolophodon
Then the design team would figure out the reason for any resets or operational anomalies (or damaged components) and put whatever additional suppression was needed. Sometimes this required rerouting of traces to reduce coupling or redesign of the ground plane. It's a tricky business and expensive if you want to do it right. I suspect that your average $120 display does not see this kind of testing.
Computer equipment usually does pretty well, especially after installation (when there's good ground path everywhere and it's in a chassis). Between ferrite beads and TVS, there's a pretty good amount of protection, and I've seen pretty few ESD anomalies-- a reset or two when touching connectors long ago is about the extent of it.
RFI and EMC is another matter. When I was operating a 100W 30MHz transmitter indoors briefly, the computers around did all kinds of wacky stuff. Flashing screens, random mouse clicks (from wired mice), transmitting packetstorms on their own, etc.
9 year-old me was delighted to discover that it would start playing on its own when it was near the plasma globe I had bought at a science museum gift shop. I couldn't explain it, but eventually came to a vague understanding.
It was semi-random, mashing together short, distorted sequences from the song bank stored in memory. Being almost recognizable made it more haunting.
I remember bringing this out one night during a sleepover and we all got kind of spooked. Fun times!
If I open the door and grasp for the (metal) edge of it before I start sliding out of my seat I will be grounded and consequently not build up charge as I leave the seat.
For example, we have a few bar stools at home; sitting on one always primes me for getting shocked after getting off it. I figured a few rules, such as never wearing anything isolating on my feet, so I can dissipate charges by keeping one foot on the metal part of the chair; or, failing that, I make sure there's always a metal object in grasping range, which I can use to later discharge in a less painful way.
(Pro tip: don't do the metal discharge thing with the hand you wear rings on, and hold the metal item so it doesn't touch the underside of your fingers - getting a shock through a nerve isn't pleasant at all.)
Another thing: I always keep a metal coat hanger around the bedroom, so that whenever I have to deal with blankets, I can keep "swiping" them to collect charge and then transfer them away by touching something grounded with the coat hanger.
Also: I always have my keys on me when away, in an easily-accessible place, specifically so I always have a metal object I can use to offload static charge in a pain-free way.
Also: over the years I kind of habituated all kinds of subtle behaviors designed to keep me safe from getting shocked by my wife or kids. Basically, if I feel one of them just got charged (e.g. via the blankets or the bar stool mentioned before), or I haven't kept track of their recent movements in a static-rich environment, if there's a need or chance of any kind of physical contact, I instinctively first touch using my elbow or some other pain-minimizing way, just to equalize charges with them. My wife sometimes notices when I do it to her, but fortunately, she is quite understanding.
https://en.m.wikipedia.org/wiki/Transient-voltage-suppressio...
Not to mention that all modern CMOS chips have protection structures on the input (though not really as good as you'd like for something that is randomly handled by charged up humans).
Ehm, can it still be considered "indoors" if it's 100W at 30MHz?
Had a bad incident a few years ago where EMI created by the windshield wiper motors on a large vehicle was causing voltage dips and spikes for our product. We had protection circuitry so that if the incoming voltage was too low, it would shut down our system cleanly.
Another logitech one: I have a g27 steering wheel/pedal set/shifter combo. The pedals plug into the wheel, which also connects to a power supply and USB cable to connect to the computer. If you plug the steering wheel into the power and DON'T plug in the USB cable, metal parts of the pedals "leak" current, and you can feel a painful sensation if you touch the metal with exposed skin.
Personally I remember having a pickup truck with a velour seat. During winter I would slide off the seat at a gas station and the first thing I would tough would be the gas pump handle - ZAAAP. Not what you want at a gas station!
To give people some idea of what they're clicking
The manufacturer does not publish a kit of approved USB devices on its website.
My company is building a chair that doesn't compromise on features for when you need them, but elegantly hides them away when not in use so you won't be tempted every second of the day. We have room in our A round if you're interested.
Turned to be a result of storing the C64 on a bench under the "large" 26" CRT... When it was kept away from it for a while and had a chance to discharge, everything was ok. But after a while near the TV, it started "typing" gibberish of its own accord.
and their USB cable that comes with them has a ferrite node.
https://www.amazon.com/AOC-E1659FWU-16IN-Powered-Monitor/dp/...
I spent a fair amount of effort trying to figure that one out. Thought it was loose cables, something, but no. I don't have a Markus chair, but my Steelcase Gesture is still capable of making a pretty good amount of static electricity. Once in a while when I stand up if my boom microphone is too close I'll shock myself on it, and most of the time I have to reboot things after that. Haven't permanently killed anything yet, thankfully.
When I plug the machine in, the audio will get all static-y from the built in speakers. It seems to especially happen when I am moving my mouse over a video thats playing - and the playback of videos stutters when pulling the cord in or out of the machine.
Battery life on this machine sucks though... but other than that, its fantastic...
I submitted the link here, it’s interesting in its own right: https://news.ycombinator.com/item?id=35744187