ES version is available. Content is displayed in original English for accuracy.
Advertisement
Advertisement
⚡ Community Insights
Discussion Sentiment
46% Positive
Analyzed from 3384 words in the discussion.
Trending Topics
#cables#ground#cable#monitor#water#chair#static#outlet#problem#hdmi
Discussion (71 Comments)Read Original on HackerNews
These devices are built to the edge of performance margins. Throw in some high voltage transients and things flip out. I think displays are particularly susceptible due to long cables, large surface area and unavoidable shape of common displays: they're effectively patch antennas.
The furniture static case is amusing: I imagine some foam cushions can cause millions of tiny static discharges in parallel when they expand. This will flow through the metal stand to "ground" and probably make a VHF range RFI spike (based on the size of a typical chair frame.) Common 24-27" display panel geometry just happens to be in the same neighborhood...
Adding ferrites to cables, as I see suggested several times, might help. You could also get unlucky and make it worse: choke a line to just the right length an it becomes a better inductor/antenna. Electricity is fun.
To be clear, neither of these things are intended to address EMI/RFI. Most consumer-grade UPSes directly pass the AC power through when not on battery, so any noise on the lines will also pass through pretty much untouched.
Surge suppressors are just MOVs (metal oxide varistors) and a circuit breaker. If the line voltage rises too high, the MOVs try to shunt the voltage. But if the EMI's peak voltage is below the MOV's trigger threshold, it will do nothing and the EMI will pass straight through.
to add:
> In my particular case, my chair wheels are made out of plastic (non-conductive), so my solution was to “ground” my chair by adding a metallic chain from the chair to my room floor. I got the idea from reddit.
The chain grounded chair is used all the time in ESD rooms. The floors in these rooms use semi-conductive flooring which is tied to a ground rod. The chair is grounded to the floor which is in turn ground bonded to earth.
On a side note, distilled water is highly recommended with ultrasonic humidifiers. Heat-based devices evaporate solely the water and leave mineral deposits behind. Ultrasonics create tiny droplets _along with the dissolved minerals_. Hard tap water or mineralized drinking water will coat your work area in chalk-like dust.
The air outside is extremely dry (<5% relative humidity once heated to indoor temperatures), and the air is quickly replaced by the ventilation. I have anecdotely heard that in the US they have much lower requirements of rate of air replacement than here in Sweden though, so maybe that could work there, but then you would also have stale air, which doesn't sound great.
Also, y'know, your lungs. Deep inside your lungs.
Running tap water in an ultrasonic humidifer's going to spike the particulate pollution (PM1/2.5/10) throughout your entire house by hundreds of ug/m^3. And it seems that children are particularly prone to inhaling this stuff and having it deposited in their lungs (~2x more particles and ~3.5x more mass).
They really shouldn't be used with anything except distilled water. The things should come with a continuity tester that disables them if the water's conductive or something.
https://external-preview.redd.it/XKDz1OYC3I7A8BPUHzbzNPiysM-...
It’s pretty clear that most modern standards (HDMI, DisplayPort, thunderbolt, etc) are so close to their physical limits that there’s no more room for errors.
And yeah, we're pretty close to the limits. We always have been, though: At all points on the timeline of digital electronics, we've been pushing speeds to be as fast as we can manage today. But tomorrow (and the next day, and the day after that), we'll solve more of the problems and yet-again make it even faster.
Which brings us back to...ferrite beads, and problems.
I got introduced to Monoprice back when HDMI was still new and somewhat finicky, when stores like Best Buy were fond of selling $180 HDMI cables (and even Wal-Mart wanted something like $60). In that crazy world, Monoprice was the place to buy inexpensive cables that worked.
And it was clear that HDMI was the future, so I placed an order for a half-dozen or so different-colored HDMI cables with ferrites pre-installed near each end.
They showed up, and... they barely worked. They were glitchy, touchy, and intermittent. I was frustrated, and I felt like I'd made a poor decision that cost me money instead of saved me money. In fact, I was rather pissed off by all of this.
With nothing to lose, I used a knife to cut away the plastic overmolding on the ferrites on one of the cables that was being particularly problematic. And then I smashed those ferrites with a hammer.
With the ferrites thus-removed, the cable immediately began working perfectly. It was glitch-free. I couldn't get it to misbehave even if I tried. I repeated this with all of the other cables from that order and they all started working perfectly, too.
---
So, ferrites. Their presence adds a little bit of series-mode inductance. And that's something that can be useful. It slows down the edge of things like transient voltage spikes. And since the spikes are transient, slowing their rise-time in this way reduces their bandwidth and peak amplitude. Adding a snap-on ferrite bead can be enough to turn a problematic data bus into a well-behaved data bus.
But! They're just dumb hunks of minerals. They're indiscriminate. They can't distinguish betwixt the bad signals and the good signals -- everything is affected. So while ferrites can be useful in a fight against unwanted noise, they can also be destructive of the signals that we're trying to use.
They're good to have available, but they're also not necessarily something that someone should go forth and attach to every cable they find. If there's no problem that needs solved, then there's no solving to be done.
(These I days I make it a point to actively avoid buying cables that have ferrites pre-installed, both professionally and at home. But I've got a stash of snap-on ferrites in the top drawer of the toolbox just in case; it's good to have options.)
https://www.youtube.com/watch?v=r-V_Z3bD_PA
https://youtube.com/shorts/R0OhD2Bc6FY
Now I'm wondering if I should ground my chair to the shelf my PC is sitting on.
As pretty obvious evidence this is static related, it only happens in the winter.
My guess is that, like OP, we're both getting interference in the our DP connections, and that that interference is in our cases causing the GPUs to crash.
Haven't had a chance to try ferrite cores yet but that was going to be my first test.
Curious what system specs you have in case we have overlap in anything that could isolate the issue. Mine: https://pcpartpicker.com/list/Xpdb8Z
other things not encapsulated in the parts list: My PC sits on the bottom shelf of a foodservice-style wire shelving rack. My motherboard's I/O shield is integrated and wasn't a perfect fit into the case.
The DP cable is, probably, in-spec. I usually buy Cable Matters or BlueRigger.
I do wonder if ferrite cores would help.
edit: The only similarity I notice in our builds is lower-end ASRock b650 motherboards.
Since then I got a 4K display, and it likes to drop out in thunderstorms. I switched to a better DP to HDMI adapter, and the chunky original Samsung cable. I'm waiting for the next storm to see if it helps.
I found a similar issue with nearly all of my cheap USB cables, which I started looking into when I realized only some of them would work right with my camera or Arduino. Out of ~30 cables perhaps 14-16 of them had no shielding at all. I cut open five “shielded” ones and two of them had a thin wire connecting the shields, just to fool people casually testing them. It’s a real crap industry.
I found out this year that my reception is horrendous if those lights are on. The coax cable which goes from the attic to the TV downstairs runs right above the lights and the LED driver / PWM noise must couple with the coaxial cable. I'm not running anything fancy like RG6 since I just re-used whatever cable was already run up there from the 70s.
I'll definitely try some of the tricks from this article.
I suspected static electricity. The solution was a thin cotton pillow on the seat. Problem gone.
Of course I live in the Phoenix metro area, and it's dry all year round. So I began taking some extraordinary measures. Of course, I had a little "repair kit" containing one of those basic ESD wrist straps. I used that. I would also do all card swaps in the bathroom without a mat. I would also, you know, remove any garments that could possibly generate ESD in any way! And, I invested in the heavy-duty, rugged type of sdcards like you'd find in a GoPro on a skydiver's kit.
These days there isn't anything using sdcards anymore. My Pixel 8 Pro doesn't even have a slot for it. I'm thankful, because USB thumb drives are more resistant to this stuff, and I make use of those very sparingly, as it's mostly about cloud storage now.
I don't think my DP cables have any ferrite chokes built in...
It's not painful, but feels like pop rocks in my ear canal.
Laptop plus external monitor is an interesting case. The monitor should be grounded via its power plug, but the laptop's ground may be floating. Not sure about the grounding path for Apple laptops. Attempts to find info on laptop grounding online are returning AI slop. If everything is running off 2-prong plug external power supplies, there may not be any grounding.
Get the room humidity above 40% and most static effects will disappear. The water in the air grounds them out. That's often the easiest solution.
There are electrostatic field meters. Halfway decent ones start around US$150. I used to have a surplus store field detector on my desk when assembling electronics. It would squawk if the field level got high. Wearing a wrist strap would shut it up. With a meter, you stop guessing. This isn't mysterious, just something that needs instrumentation to chase down.
The chair chain is good, but make sure that the shiny enameled chain and the floors are actually conductive.
One of the things that people (well, idiots -- but idiots are also people) discover when replacing an old 2-prong outlet with a new 3-prong outlet from the big box store is that they've only got 2 wires to work with. There is no ground conductor is present.
So they do the wrong thing the wrong way, and connect the ground screw on the new outlet to the neutral wire. This satisfies them ("all of the outlet parts are wired up!"), and sometimes they even think about it hard enough to justify it as being Good Enough ("ground and neutral are connected together back at the panel anyway, so it doesn't matter!").
That's bad, mmkay? Nobody should do this. Ever. It is unsafe. But sometimes people do it anyway. It's a real problem that exists in the real world.
This problem is made worse because an outlet tester won't detect this fault -- at all.
And the badness doesn't stop there, but instead compounds: The tester doesn't just fail to detect the fault. Instead, the tester will (must) cheerfully report this condition as being perfectly cromulent and safe. That false assurance is problematic in and of itself.
So, yeah: Everyone should have an outlet tester. But everyone should also be aware that they aren't idiot-proof -- their results can be poisoned by idiots from the past.
---
Anyway, code. NEC 406.4(D)(2) allows replacing an ungrounded outlet with a GFCI outlet (with the ground screw disconnected and doing nothing at all). The outlet must be marked (that's why GFCI outlets include a sheet of stickers in the box that say "NO EQUIPMENT GROUND"), but it's code-compliant to do this.
So even if one pushes a landlord about an ungrounded 3-prong outlet on the wall, that doesn't mean that they're going to send someone out and tear into things to install a ground wire. It might instead mean that they put a GFCI outlet in, put a sticker on it, and call that good enough.
And, safety-wise: A GFCI used in this way actually is good enough.
But even though safe, it doesn't help at all with EMI/RFI/static issues and electronics, which the landlord doesn't have to care about. That part isn't their problem. :)
(406.4 also allows replacing existing 2-prong outlets with new 2-prong outlets, which are still being made in factories every day.)
And yes, I know about GFCIs with no ground and the warning. Had one of those once.
Still, an outlet tester will find the common case. I suspect that the situation here is that everything is on 2-wire external power supplies and there is no path to ground anywhere. But by plugging in an external monitor, the user created more external EMI exposure, by bringing his floating ground out of the laptop and monitor. Standalone laptops are tested for EMI compatibility (emissions in the US, emissions and sensitivity in the EU), but that doesn't cover being cabled up to random external devices.
I would not be surprised that touching such monitor will electrocute you.
If one has a medium-sized chunk of money to burn, one could try fiber optic cabling. I've personally had -AFAICT- perfect results from Monoprice's "SlimRun AV" fiber DisplayPort cables, and Nippon Labs' fiber HDMI cables. [0] I expect that Monoprice's fiber HDMI cables and Nippon Labs' fiber DisplayPort cables are also fine, but I've never used those, so I cannot comment.
For folks concerned about "dreadfully fragile" fiber optic cables, I do know that the Monoprice cables are durable... a vigorous misadventure caused me to torque the hell out of the monitor-side connector. The connector bent, forcing the case split a bit at the seam. After some counter-bending of the connector and pushing its case back mostly closed, the cable works fine. Given the outward similarity in build quality, I expect that the Nippon Labs cable I have is at least as durable.
[0] Both families of cables drive my "4k" HDR monitor at 60Hz without lossy compression.