PSA: The Steam Controller’s magnetic charger can be a fire hazard

Keep the charging puck’s exposed pins far away from anything metal.

See full article...

 

[motytrah]

For everyone else, let this be a reminder to keep your Steam Controller puck away from any metallic or magnetically sensitive objects.

I dont understand how this is the advice and not a mass recall and a redesign to avoid this. How many houses will have to burn down or people die before this is addressed?

It's something I'd expect from cheap no name brand stuff, but would hope a major company would have short circuit protection.

 

[Dark Jaguar]

We have a dinosaur at home.

View attachment 135553

I see you've spared no expense.

See this is why genetics can be this double edged sword when it comes to definining lineages. Sure, what the genes say is above repute, and we know the family trees far better than ever before thanks to it. We know, genetically, birds are dinosaurs.

But... culturally, we defined that word morphologically, and birds just ain't dinos in THAT critical sense. That's the tricky part, because we have two entirely valid definitions of the word- and you have to go and be CLEVER about it!

 

[bretayn]

Scrolled back and read that. Agree that it looks like Valve did the right thing and something else is going on here.

I'm not an electrical engineer, just a tinkerer. Is it possible that the Milanese Loop (the Apple Watch band shown in the article) is setting up unusual current flow that a simple connection (a scissor blade) would not? The watch band is either made of stainless steel or titanium, and is magnetic. (At least I assume that, because the clasp is magnetic, and can clasp at any point along the band.)

Edit: I was wrong, it's a Pixel watch, not an Apple watch. I'm not sure about the materials in play here.

 

[antibolo]

Oh, now I'm triggered. My house has upside-down outlets to indicate half the outlet is controlled by a wall switch. I freaking hate these outlets because I use flat-plug power strips (where the wire is supposed to go straight down flat against the wall), but on those outlets the wire goes straight up putting tons of strain on the plug.

Our modern understanding is that putting the ground up (ie. "upside down" from how it was traditionally intended) is the safer way, because if something were to fall on a partially disconnected plug, it would touch the ground rather than the live blades.

It's really those flat-plugs that are wrongly designed, and deserve the hate.

 

[Quasarsaurus]

I've looked through this thread several times and ctrl-F'd for coonwhiz, I'm not seeing this comment/link that is being referred to.

Edit: Is this referring to @Quasarsaurus comment? Was there a name change?

Yeah, I changed my name. Been meaning to do it for a while, just picked today for w/e reason.

 

[Dark Jaguar]

Our modern understanding is that putting the ground up (ie. "upside down" from how it was traditionally intended) is the safer way, because if something were to fall on a partially disconnected plug, it would touch the ground rather than the live blades.

It's really those flat-plugs that are wrongly designed, and deserve the hate.

I think the better fix, in the long run, is to adopt a European standard and run protective shielding halfway down the electrical contacts. This would prevent any dangerous risk of an errant tool or finger brushing against the hot lead or shorting across two. Since the previous standard allowed the "grip" of the plug to be at an arbitrary point along the internal wall though, applying that shielding would make it incompatible with any plug who's contact grips are closer to the front. There'd be some "growing pains" I suspect, but it's the only way to fully protect. I suppose having the ground up makes it less likely to contact live, but it'll still happen. "Less likely" is still an improvement, true, but yes, in practice most devices we're plugging in kind of "expect" the typical "face shaped" orientation, and that includes my old "wall wart" AC/DC adapters. Hang a few of THOSE upside down and it'll cause even more risk with the weight tugging it right out of alignment.

It's complicated, but Europe already engineered a fix. We just need to adapt it into our plug standard. Maybe requiring the contact point to be right where those HOLES everyone puts on our plugs go. If plugs are by and large already made with that in mind, problem solved!

You know, when I was growing up, I was in a number of houses that were built before ground pins became a requirement. We had a number of little adapters we had to use to plug a 3 pin device into a 2 pin outlet. Should there be a number of outlets "gripping" the plugs earlier, and someone buys a new device with that halfway down shielding in place, they'd just also have to get an adapter to let them use it during the transitional period.

 

[jonbob_newcastle]

T

Oh, now I'm triggered. My house has upside-down outlets to indicate half the outlet is controlled by a wall switch. I freaking hate these outlets because I use flat-plug power strips (where the wire is supposed to go straight down flat against the wall), but on those outlets the wire goes straight up putting tons of strain on the plug.

They’re quite rare these days but some houses here have different sockets for lights that are controlled that way. They’re like miniature Type Gs with round pins.

All our plugs have the wire coming out at the bottom. It means you can have something plugged in behind e.g. a book case without needing to have a gap between the furniture and the wall.

What ground loop? The Pixel charger uses inductive charging.

Devices without a ground connection (or a bad one) can sometimes leak a very small amount of current. This is noticeable if they have a metal case; you’ll feel a tingle if you brush your hand against them.

This also means there will be a voltage between the case and ground. This can be a huge problem for audio equipment as it interferes with the audio signal. It’s not dangerous per se, but it’s not inconceivable that there might be some arcing if, say, there’s a round pin with a very small point of contact. This article explains it better.

 

[Quasarsaurus]

What ground loop? The Pixel charger uses inductive charging.

The Pixel 3 watch, which the original user reported to be using does not use inductive charging. Here's the charger, there's 4 metal pins.
https://store.google.com/product/watch_3_charger

 

[fe3a8b63]

By the by, I’ve often wondered why American sockets aren’t installed upside down. Having the earth pin at the top would stop the rare instances of something conductive falling onto a half-inserted plug and shorting live and neutral.

They aren't because earth pin is longer and automatically connects first and breaks last. Hospitals, businesses, etc using North American NEMA connectors sometimes have the sockets "upside down". No the electrical code doesn't specify the orientation, despite common belief.

The earth pin won't save you in a way that a real GFCI/ residual-current circuit breaker (RCCB) will.

You may want a arc-fault circuit interrupter (AFCI)/ arc-fault detection device (AFDD) instead or in a combo with a GFCI/RCCB

Oh, now I'm triggered. My house has upside-down outlets to indicate half the outlet is controlled by a wall switch. I freaking hate these outlets because I use flat-plug power strips (where the wire is supposed to go straight down flat against the wall), but on those outlets the wire goes straight up putting tons of strain on the plug.

The upside down thing is a suggestion and isnt part of the code. I do wish less strain would be placed on the conducting wires tbh.

 

[stormcrash]

T

They’re quite rare these days but some houses here have different sockets for lights that are controlled that way. They’re like miniature Type Gs with round pins.

All our plugs have the wire coming out at the bottom. It means you can have something plugged in behind e.g. a book case without needing to have a gap between the furniture and the wall.

Devices without a ground connection (or a bad one) can sometimes leak a very small amount of current. This is noticeable if they have a metal case; you’ll feel a tingle if you brush your hand against them.

This also means there will be a voltage between the case and ground. This can be a huge problem for audio equipment as it interferes with the audio signal. It’s not dangerous per se, but it’s not inconceivable that there might be some arcing if, say, there’s a round pin with a very small point of contact. This article explains it better.

We have 90 degree plugs too, in fact we have the choice (or at least manufacturers do). An appliance like say a vacuum cleaner has no need of a 90 degree plug since that would actually risk more strain on the cord if pulled while vacuuming.

I will caveat though by admitting that most products chose the straight out plug because it's cheaper and familiar, but I have AC adapters and power strips with 90 degree plugs, and they've long been standard on appliances like fridges and clothes washers where the cord is meant to be tucked behind the unit with minimal space.

I think a lot of it is that for normal consumer applications the more familiar form factor was established when basically all we were plugging in were either lamps/radios, or appliances that were meant to be used and then stowed rather than perpetually hooked up, and overall consumers still find that plug more familiar and intuitive for plugging and unplugging than the 90 degree style

 

[stormcrash]

They aren't because earth pin is longer and automatically connects first and breaks last. Hospitals, businesses, etc using North American NEMA connectors sometimes have the sockets "upside down". No the electrical code doesn't specify the orientation, despite common belief.

The earth pin won't save you in a way that a real GFCI/ residual-current circuit breaker (RCCB) will.

You may want a arc-fault circuit interrupter (AFCI)/ arc-fault detection device (AFDD) instead or in a combo with a GFCI/RCCB

The upside down thing is a suggestion and isnt part of the code. I do wish less strain would be placed on the conducting wires tbh.

I would love to see electrical code updated to include smarter breakers, though of course then people will complain about having to go to the panel to reset vs a gfci outlet chain

 

[jonbob_newcastle]

Sort of off topic but I just remembered that I use rechargeable li-ion AAs in my Xbox controllers. Be careful if you use these in devices that draw a continuous current as the cells often skimp on overcurrent and temperature protection. I put some into a camping lantern when we had a power cut and they were very hot after a couple of hours. I only noticed because it started to flicker and I went to investigate.

 

[Dark Jaguar]

I would love to see electrical code updated to include smarter breakers, though of course then people will complain about having to go to the panel to reset vs a gfci outlet chain

I'm already going to a panel as it is. It seems to me using the breaker right at the outlet that done did it would be more convenient and I'd welcome it.

Something simply MUST be done about the "hair dryer problem" though. Those things blow and half the house goes out. Maybe setting that on it's own isolate circuit so if it goes out, THAT outlet is the only one that trips is the best bet.

 

[Autapomorphy]

By the by, I’ve often wondered why American sockets aren’t installed upside down.

There really isn't any concept of upside-down in the electrical code. Receptacles can go in whatever orientation is desired. They're often horizontal when installed above countertops, and there's no concept of "up" when installed on the floor or ceiling. If you wanted to be, um, "bold" and "creative" then you could install receptacles in your house diagonally.

Because that failure mode basically never happens. If the pins are exposed enough for that to matter then the plug has basically fallen out of the outlet at that point.

I wouldn't dismiss it as something that never happens. It's happened to me while using a tape measure that slid down the wall and shorted the hot and neutral prongs together. Fortunately, metal tape resulted in enough current to trip the circuit breaker almost instantly. I ended up replacing that receptacle because it had a bit of flash charring, and I was sure to put the replacement ground up.

The good news is that modern residential electrical code now requires arc-fault circuit interrupters on most receptacles. Those increase the likelihood of power being cut off if something shorts those pins that's not conductive enough to trip overcurrent protection.

 

[FranzJoseph]

I've looked through this thread several times and ctrl-F'd for coonwhiz, I'm not seeing this comment/link that is being referred to.

Edit: Is this referring to @Quasarsaurus comment? Was there a name change?

Yes, sorry. My mistake.

 

[Xepherys]

In theory this shouldn’t be much different than Apple’s longtime laptop MagSafe charging connector. Can anyone chime in on whether that’s less of a fire risk, and if so, how?

Apple MagSafe has a handshake procedure that occurs before power flows. I would imagine that these do, too, but this one had a faulty controller? Or they somehow failed basic EE concepts during the design process?

 

[kolepard]

Is there a good reason you couldn't use inductive charging like on most phones and many watches? I'm not an EE and I know it'd be less efficient, but with Qi2 charging at up to 25W it doesn't seem like it would be a problem with time to charge. Is it more a matter of the space needed? Even the Apple Watch can charge at up to 20W from what I read.

I'm not trying to troll here; I amm genuinely curious as it seems the inductive charging would avoid the shorting issue altogether.

Kevin

 

[jonbob_newcastle]

Something simply MUST be done about the "hair dryer problem" though. Those things blow and half the house goes out. Maybe setting that on it's own isolate circuit so if it goes out, THAT outlet is the only one that trips is the best bet.

You can solve that by not putting electrical outlets in bathrooms. The only ones we have are for shavers and electric toothbrushes. They’re limited to 110V, have an internal fuse, a different plug and they’re galvanically isolated so there’s no risk of electric shock unless you actually set out to electrocute yourself.

We also have GFCIs (RCDs in our parlance) in the breaker box (consumer unit) so every outlet is protected. Some circuits also have their own dedicated RCD, e.g. the one for an electric cooker. They might have a slightly higher trip current but they’re still adequate. I was under the impression that this was standard in the States now too.

There was a case of someone being electrocuted in their bath but that was because they ran an extension cord from another room so they could charge their phone while they used it. In the bath.
At the risk of sounding callous, there really is no helping some people.

Is there a good reason you couldn't use inductive charging like on most phones and many watches?

Short answer: if it’s the pad type of charger you can’t use the device while it’s charging. If it’s the kind that attaches with a magnet there’s usually no pressing reason to make it wireless in the first place.

 

[Randomizer]

I would love to see electrical code updated to include smarter breakers, though of course then people will complain about having to go to the panel to reset vs a gfci outlet chain

It's not just that - it's that AFCI/GFCI combo breakers breakers are very expensive ($70ea, as opposed to basic 20A thermal breakers going for $7) and also have issues with nuisance tripping, especially when using anything with brushed motors. Older vacuum cleaners are especially are terrible with these.
GFCI breakers without Arc-Fault detection do absolutely nothing for arcing and may not trip if you briefly short neutral and hot. They trip when there is a current difference between neutral and hot, which is why they are strongly recommend wherever there are outlets without ground, and they have to be labeled "Not Grounded".

 

[Autapomorphy]

Something simply MUST be done about the "hair dryer problem" though.

I still haven't figured out how a lot of hair dryers can legally be sold. I see some that are labeled at 1875 watts, which is more than 15 amps at 120V. How can such a device be sold with a 150-amp plug instead of the 20-amp plug?

Of course, that gets into one of my pet peeves about residential wiring in the US. Everything in a modern house is on 20-amp breakers and 12-gauge wire, so why are all the receptacles 15-amp instead of the ones that accept 15- and 20-amp plugs? I ended up having to replace some of the receptacles in my house because the 20-amp capable PDU in my server rack needs a 20-amp receptacle.

 

[jonbob_newcastle]

I still haven't figured out how a lot of hair dryers can legally be sold. I see some that are labeled at 1875 watts, which is more than 15 amps at 120V.

Try 230V@13A. I might be smug if I had enough hair to need a hairdryer.

 

[Randomizer]

You can solve that by not putting electrical outlets in bathrooms. The only ones we have are for shavers and electric toothbrushes. They’re limited to 110V, have an internal fuse, a different plug and they’re galvanically isolated so there’s no risk of electric shock unless you actually set out to electrocute yourself.
...

The US is all on 120V, and outlets in bathrooms have been required to have GFCI outlets for at least 30 years now, AND every hair dryer in the US must have a GFI circuit in the plug as well (which I don't like - they are big, bulky things). It's practically impossible to electrocute yourself with a hair dryer in the US.

Breakers tripping due to overloads - if the house wiring is so shoddy (or so old) that a single hair dryer trips "half the house", that's 100% an issue with the house. This is NOT AT ALL typical in the US.

 

[jhesse]

I still haven't figured out how a lot of hair dryers can legally be sold. I see some that are labeled at 1875 watts, which is more than 15 amps at 120V. How can such a device be sold with a 150-amp plug instead of the 20-amp plug?

Because the North American household voltage is 120 volts ± 6%, and marketing went with using 125 volts in the wattage calculation. So, it might be true... someplace. Just as long as it doesn't draw more than 15 amps.

(Last I checked, I measured 118 volts at the outlet. That hairdryer would suck down about 14.2 amps for 1675 watts.)

 

[Randomizer]

I still haven't figured out how a lot of hair dryers can legally be sold. I see some that are labeled at 1875 watts, which is more than 15 amps at 120V. How can such a device be sold with a 150-amp plug instead of the 20-amp plug?
...

Although actual voltage is nominally ~120V, outlets (and the NEMA code) are rated for 125V, I think to account for possible regional or temporary voltage variations. 125V x 15A = 1875W.

 

[habilain]

I always assumed these types of things had some sort of smart logic that only sent any measurable power through after confirming it was properly seated. What happens if I put a lightning cable in my mouth? Does it send 15w to me or confirm Im an iphone?

Lightning is basically the same as USB charging i.e. it starts at 5V and then it can use the USB PD handshake to raise to a higher voltage. I'm pretty certain this is separate from Apple's authentication chips for Lightning cables as well.

So if you put a Lightning or USB cable into your mouth and it's plugged in to a modern charger, it'll probably try to apply 5V/0.5A to your mouth (i.e. 2.5W). This will fail, as 5V won't overcome your electrical resistance. I assume you are not capable of answering the USB PD handshake, so the voltage will not be increased.

 

[Randomizer]

Is there a good reason you couldn't use inductive charging like on most phones and many watches? I'm not an EE and I know it'd be less efficient, but with Qi2 charging at up to 25W it doesn't seem like it would be a problem with time to charge. Is it more a matter of the space needed? Even the Apple Watch can charge at up to 20W from what I read.

I'm not trying to troll here; I am genuinely curious as it seems the inductive charging would avoid the shorting issue altogether.

Kevin

The inductive coil adds a little bit of thickness and weight, and they tend to get very warm when charging. There are advantages and disadvantages to both approaches.
Is there a good reason no one takes the approach to charge via a single pin and then use a resonance frequency to couple the device and charger? It would be similar to wireless charging but without large coils and wasting energy on pumping EM fields, and you only need to align on a single pin, so orientation wouldn't matter.

 

[sd70mac]

I always assumed these types of things had some sort of smart logic that only sent any measurable power through after confirming it was properly seated. What happens if I put a lightning cable in my mouth? Does it send 15w to me or confirm Im an iphone?

I am pretty sure a Lightning cable or the power source confirms that it is plugged into a device before blasting 15+W down the cable.

 

[markgo]

I assume you are not capable of answering the USB PD handshake, so the voltage will not be increased.

I can’t answer a PD handshake, but I can whistle a 300 baud modem answer tone, if that’s any help.

(God knows why my friends and I spent so much time practicing that useless skill).

 

[habilain]

I can’t answer a PD handshake, but I can whistle a 300 baud modem answer tone, if that’s any help.

(God knows why my friends and I spent so much time practicing that useless skill).

Maybe that'll get you the 48V used on the old telecom network?

 

[knighttime]

So if you put a Lightning or USB cable into your mouth and it's plugged in to a modern charger, it'll probably try to apply 5V/0.5A to your mouth (i.e. 2.5W). This will fail, as 5V won't overcome your electrical resistance. I assume you are not capable of answering the USB PD handshake, so the voltage will not be increased.

One caveat would be you may get shocked if you started a circuit that didn't involve the charger (like what is speculated about the watch and Steam charger in these comments). Also if you're unlucky enough to do that during a power surge, then you're electrocuted.

An example would be if your (not dry) foot was touching the unpainted part of a PC case (grounded) and the USB cable was attached to a cheap charger that was not grounded, maybe even from a different circuit in the building. A circuit could form with the current going through the cable/connector's shielding, to through your mouth, and to the other plugged in object your foot was touching.

I checked one of my cheap USB chargers without the ground, and my multimeter shows a few volts AC at 60 Hz between the shield of a USB connector attached to said charger and the ground of my surge protector.

 

[approximate_pi]

The metal pins in the MagSafe charger are recessed, so they present less of a short circuit hazard.

The charging/data (for kb input) pogo pins on ipad keyboard cases though arent, without this problem.

I'd also speculate that the Apple device has short protection circuitry that the Steam device clearly lacks.

This is likely the answer