Consumers shouldn’t need to worry about the specific output voltage; the presence of a USB-A connector should indicate that it abides by the specification.
The hacker in me hates the idea of enforcing something like this, but poor interoperability is such a pain that it would be nice to have stronger guarantees.
Many applicances would break, or perhaps even catch fire.
If there were many different USB-A devices and people were used to checking the voltage output before plugging, that might not be needed, but since probably none of us check for a USB-A port to see if it outputs 7.5V, it better be mandatory to be certified.
Though, maybe it is, but still those cheap Chinese adapters slip away.
We have something like that here in Brazil. On cities which use 127V as the standard voltage (some cities use 220V), you can get 220V by using a pair of phases instead of a phase and the neutral, and it's common to find a few standard-looking sockets which are 220V instead of 127V. They might be colored red, they might have a yellow sticker saying "220V" next to it, or they might be completely unmarked. If your device is 127V only, or it has a manual 127V/220V switch which is on the 127V position, and you plug it into one of these 220V sockets, it will be damaged.
(The standard we use for consumer AC power plugs and sockets, NBR 14136, does not make a distinction between voltages; the same plugs and sockets are used for both 127V and 220V.)
Overvoltage is now routine in places where there's lots of PV installed in homes and not enough load nearby.
If you live in such a place, you have a deep appreciation of varistors and should have either good insurance or a little bit of soldering skills... preferably both.
Every once and a while an electrician wires up standard US outlets to 480v 3-phase instead of 208v 3-phase, so you end up with an outlet designed for 120v that puts out 277v.
Sometimes noone notices because the switching power supplies plugged into them are just fine with this.
Edit: I am not advising anyone to deliberately do stupid things, simply mentioning that from my experience the builtin margin is fairly large.
It wasn't mandatory but this is one of the main reasons Underwriter's Laboratory was started. The market needed a certification to say "yes, this things does what it's says it does and in a safe manner".
My dad owned a retail store in the 1970s and he distinctly remember being told "Lots of retailers won't carry a brand/item unless it has the 'UL' sticker on it."
The device is specifying the output voltage, which isn't in spec for the connector, but it is (apparently) accurate.
[0]https://www.kimuagroup.com/news/differences-between-ce-and-c...
I don't think the CE mark protects consumers the way it could decades ago, with international imports taking a few seconds and free postage to boot.
https://donglec.com/blogs/journal/are-third-party-chargers-h...
It's a self-certification. You can decide how much that's actually worth for non-EU vendors.
Nintendo comes to mind; charging the Switch (the earlier models IIRC) with a third-party adapter was a lottery, and the losing prize was a fried console.
They've been guilty of this practice for as long as they've been making electronics - I think it was the NES or the SNES that used a standard barrel jack and voltage, but reversed the polarity, with no circuit to protect the console.
Also - using Ethernet ports/cables for serial console access is not even evil, it's just stupid. Granted the switches/routers that do this are not exactly SOHO equipment, but you need to let students in a lab near that, and it's just asking for trouble.
Is there actually a standard about that? IIRC, recently things have drifted towards "center positive" as customary, but most adapters and devices have the little diagram about how they're wired for a reason.
The problem you're thinking of was caused by a Nyko third-party docking station with a faulty hardware implementation of USB-PD. While Nintendo's software implementation of USB-PD was flawed in some ways, it didn't cause this problem.
> I think it was the NES or the SNES that used a standard barrel jack and voltage, but reversed the polarity, with no circuit to protect the console
There is, sadly, no "standard" polarity for barrel power jacks. Center-positive and center-negative are both fairly common.
It seems like somehow we're used to the idea that this particular plug will only work with the device that came with it, but we assume that USB-A is USB-A. I know I do.
This does make me wonder about the USB adapter I got with my dash cam. The documentation says all over the place to pretty please with sugar on top use our adapter. I'll have to stick a meter on it. [Edit: 5.1v.]
As long as you are free to modify the hardware or purchase specific non-conforming hardware I see no reason why this would be painful for a hacker.
For pure consumer products (whatever that means) safety standards are quite reasonable. There are already legal requirements in the USA that plugs have certain shapes controlling orientation, grounding, and expected voltage, as well as building code so that people don’t trip over long power cables. Food safety means you can pick up something at the shop and have some faith it won’t kill you.
This reduces cognitive overhead and doesn’t require you to be an expert in every domain.
It also doesn’t stop you from hacking your own stuff (mostly — please don’t mix ammonia and bleach!) and easily provides you a bunch of dependable items on which you can build your hacks. So you can (briefly!) supply 100 amps of 240V AC over USB A at home if you want, you just can’t sell it. You can make yoghurt at home and store it on the counter in the sun if you want, no problem.
> The hacker in me hates the idea
If the hacker in you really hates this, then why insist on enforcing via means that restrict both vendor and user freedom (IP) instead of regulatory enforcement that exclusively targets vendors & leaves hackers free to hack?
If I develop a new connector, how do I get the regulator to start enforcing that everyone who uses it is following my standard?
What happens when I find a way to extend the specification in a safe, backwards compatible manner, but this conflicts with the standard as enforced by the regulator?
I’m not saying that these are insurmountable problems, but I feel like the regulator should focus on issues such as safety, and interoperability should be enforced by the company/consortium that develops the standard in the first place.
At least adhere to the spec to such a level that it won't destroy other hardware.
I think your initial instinct is right.
Why? I'm not sure what being a hacker has to do with it, but this sort of thing would benefit everybody and harm nobody except for the negligent.
Isn't there already some law somewhere against things that are so obviously not what a consumer would expect?
This may violate the USB spec and burn out phones, but it's not necessarily undeserving of the CE mark.
Even if the connector itself were patented?
I sent Lidl a courteous and long mail about this and they didn't really care other than "please open a service ticket with your local Lidl", which I didn't care to do since what would they do?
How about urgent and short?
e.g. One day a kid will charge their toy over night with your 13v usb-c charger and they'll burn their house down.
“and this email notification demonstrates you are aware of the hazard which will be relevant in lawsuits.”
"The takeaway of this is that given the plug looks exactly like a standard USB-C phone charger's plug, it's easy for a novice user (say, a child whose phone charger's wire has been bit through by the family dog) to attempt to charge a device not rated for 13 V using this charger, which is quite likely to damage the device."
Newer revisions of the tool don't seem to have this flaw though. They only come with a USB-C cable and no charger in the box.
Also thanks for reminding me to put a big sticker on the charger with a warning!
Those common charger EU rules pushing USB-C via the Radio Equipment Directive are actually only for the device, and even a specific list of types. But in briefing document [2] (made for RED change last year) mentions:
> a separate initiative on the eco-design for the external power supply (EPS) would harmonise the EPS-side receptacle and the communication protocol.
So "EPS" is what we are looking for and they use some eco laws to regulate them. [1] gives an overview and links COMMISSION REGULATION (EU) 2019/1782. And again, it only applies to a specific set of devices, with power tools not being listed. Doubt those fall under "Other toys, leisure and sports equipment"^^
Back to that EPS initiative... from my understanding, it [3] seems to be still "Planned for Third quarter 2023". Feedback period ended in May 2022, ofc.
The department seems to have vacant spot [4] for an expert, btw :)
[1] https://commission.europa.eu/energy-climate-change-environme... [2] https://www.europarl.europa.eu/RegData/etudes/BRIE/2021/6988... [3] https://ec.europa.eu/info/law/better-regulation/have-your-sa... [4] https://www.cnr.it/sites/default/files/public/media/internaz...
Probably a side effect when USB-PD was introduced and people neglected the facts that there needs to be an elaborate handshake for anything other than 5V.
Previously we always needed to check non-USB chargers for their voltage and there were connectors and adapters for different sizes. But now you need to be more careful with USB as well.
People don't like to read. Short & concise wins the day here.
For example, the handwarmers I'm using right now are USB-A at 12 volts, but have a dial to adjust the voltage up and down between 0 volts and 12 volts to adjust the temperature (I think it's actually PWM on the 12v power lines, but haven't actually checked).
5.5V or 6V is also common because it lets you use thin cheap wires which remove about a volt and still can fast charge phones (which tend to be picky and will drop back to 500mA charging if the voltage droops).
Big manufacturers also use non-5V over USB sometimes, but you need to send a special message to enable higher voltages, so it's safe to plug them into anything. Still troublesome if you're using a USB Y-splitter or multi-headeded cable.
I wish I would have wrote the model down posterity, but at one point I worked a trading company and a number of the users had ISDN at home. There were two 'different' modems, and I say different because only the model number on them notified you of it.
One had a plug that output DC power. One had a plug that output AC power.
They were the same size wall wart. Same size barrel. But if you plugged the AC adapter in to the DC modem it would make a noise like you curled the modem up into a ball and let the magic smoke out.
The only thing that is truly standard is the polarity of power on a USB-A plug.
It sound like the manufacturers just see USB-A as a convenient and cheap connector for delivering power.
By the way, are you Finnish? I've never seen an 'O with tilde' character before.
Cats instinctively love & trust moving water as stagnant water can accumulate bacteria & disease.
So a cat fountain sounds like a gimmick or novelty but it’s a great idea.
It’s cute and adorable (the fountain is a bit silly looking), and I have to assume it’s good for him.
And to guard against power outages, I got him a separate fountain with an open top, so in a pinch at least he had some standing water.
So the cats have their own cat towers and a water fountain. The first day with the fountain, the cats were so chill. I wonder if running-water sound soothes them.
To be on the safe side, one would actually have to check the voltage. Both the output voltage and whether these are really 0V and 5V to ground. The output frequency spectrum should also be checked.
There are much worse things than 7V that a bad adapter can output.
One of the scariest failure modes I’ve recently learned about is a supply that still provides 5V between both pins – but 225V and 230V to ground, respectively…
When I raise and lower the desk, the lamp lights up. That might not be the power port, it may be the "induction" power button on the lamp's chord getting interference from the desk motors. I'm living the tinderbox life.
not sure I follow you. can you clarify what you mean by both 225v and 230v to ground? You are measuring the potential between ground and USB output?
I remember showering in Costa Rica at a “hotel” and the shower head was one of those that heats the water (with an electric heater) and it was shocking me the whole time I used it. Probably shouldn’t have used it
Apparently those are common in central and South America.
And yes it's because of the metal case. It's not the actual mains you're connected to though, just some induction on the PSU shielding.
That leads to the problem child which is China and its lacking morals. Chinese companies will just slap the CE on anything, because the EU say that must and Europeans seems to enjoy seeing it. Sometimes, they will slap on a different CE marking, meaning "China Export". You can tell the difference by tracing the C to a full circle, if that hits the E exactly, its CE as in "conformité européenne", if it doesn't, it's China Export.
If you're a fly by night USB charger manufacturer, you just invent a new brand for next month, so you won't get the fines or be banned.
Never did I thank that would be some kind of "this is a great thing" approval stamp.
However, it never occurred to me that a USB-A charging brick would be able to give 7.5V, or even 9V or 12V as others have commented. That might completely fry some devices...
EDIT: I may be wrong. Apparently this is now allowed by the Quick Charge spec. See comments below.
Cats.
I'd also be on the lookout for the electrical hum mentioned in TFA; that would drive me crazy, and I can't speak to how it would make any kitty wary
Also, and this may go without saying, but watch out for the fountain(s) being made out of materials which are hard to clean
That said, what an awful USB port to come across. A little reading wouldn't have gone amiss though. My natural reaction would have been to assume it was an output, not an input.
And how would a loose cable cause a voltage spike?
Zener diodes can temporarily protect an overvoltage too but you will need something to take all that current that is now going through it, making it pretty hot and likely to fry something
You can pair something like this with a positive temperature coefficient thermistor to act as a fuse (as it gets hotter the resistance goes up).
These circuits are tricky to tune right, take up space and can add significant cost to a product. So with a drive in consumer products to lower cost smaller things you may not see them everywhere.
> how would a loose cable cause a voltage spike?
From the top of my flu-y head, connecting and disconnecting an inductive load (like a motor) can cause spikes. However in USB land, that's pretty rare.
dc-dc step up devices use an oscillator to charge and discharge an inductor to achieve a certain voltage increase. They _could_ be noisy when starting up, or have transient load.
From that time I tend to use the provided one, but I never noticed it provides a different voltage. Thanks for the warning
3rd party ones on the other hand....
Also, I'd love to see ones with 65w charging so I can charge my laptop and still have an outlet to use!
There are a few no-name brands with 65w, but I don’t trust them.
Just note, they are much larger than normal outlets and can be a squeeze to fit in a normal box.
So another self-made problem then.
Stop trying to 'improve' life by more & more tech-based BS.
Turns out that port is intended for connecting a receipt printer, and clearly not in any way a standard USB-C.
There is actually a micro usb port elsewhere on the device that can be used for debugging. Lesson learned the hard way.
I was selling cables and adapters on ebay at the time and we had a package with both types. Easier to sell people both versions than deal with "oops i needed the other one."
(I may well have mis-named the parties; apologies if so)
Thanks for the PSA, I'll be sure to get that adapter off my kitchen counter right away!
Simple small screen or just led lights.
Maybe even a "kill-a-watt" for usb power cube/banks.
- A pass-thru USB power meter. Measures volts, amps, wattage (simple arithmetic), and watt-hours
- a USB load device. You can configure the voltage and amps to draw. Used in conjunction with the above, you can see just how much a USB supply can supply
- a watch cell powered cable tester. Not essential for your use-case, but great for working out whether a cable is at fault or not.
None of this directly measures quality, but usually you can infer that from amps, and from negotiation.
But the charger was always giving 12 or even 20V (i don't remember exactly, I don't have access to this device now).
A real trap.
Latest USB PD supports up to 48V and up to 5A, resulting in 240W
That’s also a reason why not all fast chargers work well for all products – if the required voltage + amperage combination is not supported then the closest supported lower combination will be used instead.
That said, 7.5 volt is not a supported voltage in USB PD at least and there seems to be no handshake that guards it in this case
1. just accept the 7.5V and charge happily.
2. trip overvoltage protection and refuse the charger.
That doesn't make this okay, it just means it isn't catastrophic.
Note that I said consumer device. If I plugged one of my microcontroller dev kits into this thing, the magic smoke would be released pretty quickly.
Not just USB-C... even early generation QuickCharge could (https://en.wikipedia.org/wiki/Quick_Charge).
