India follows EU's example in requiring USB-C charging for smart devices India is on a path to require USB-C charging ports in almost all smart devices following actions taken by an inter-ministerial task force. Rohit Kumar Singh, Secretary of the Department of Consumer Affairs, said the move is "in the interest of consumer welfare and prevention of avoidable e-waste." The broad consensus in the …
Look how long it took for the EU to come to agreement about USB-C - when they started talking standardization they wanted to standardize on mini or micro USB.
How many years after the introduction of USB-C would it have taken for them to permit OEMs to change ports? Would they have done so yet, or would extensions for higher speeds, power delivery etc. been delivered for micro-USB and since everyone would be using that because its the "standard" would USB-C be used anywhere? Or would it be like USB-B, a footnote to history used only in odd places like printers?
Face it, once USB-C is written into the standards there all extensions will come with that port. Even if compromises have to be made, there will be way too much commercial mass behind USB-C for a new USB port type to ever become widely used enough that there's a serious effort to support USB-C in the standards. Heck, the USB standards forum might decide it isn't even worth their time to define a new port, because they would know how difficult it will be to supplant USB-C from its perch.
So you better like USB-C, it is going to remain the standard for the rest of our lives.
The EU legislation has been specifically written to allow for future changes in technology. The Indian legislation is very likely to be a cut'n'paste with localisations. Future US legislation is highly likely to be butchered by special interest groups and end up optional but set in stone so people can point to it as unworkable and probably not even be ratified until it's out of date anyway.
The maximum charging rates and data transfer rates are already beyond ridiculous using USB-C (depending on what's being transferred).
I would suggest if higher power requirements are needed, that any cable and connector of that size will never be adequate, and such a connector would not be suitable for small devices anyway.
My biggest issue is as always users, every usb revision the smaller connectors fail in some way thanks to users.
With USB-C users manage to plug in the USB-C connector at an angle (They manage to bend the casing and the contacts make connection with multiple pins, so we have had numerous dead USB-C ports on laptops and fried USB-C docking stations).
Indeed, one of the notable design features of the Nintendo switch is no protection on the rx/tx lines. With only 0.3 mm separation between the 4 USB-C VCC and rx/tx line conductors, any guesses as to what one of the most common failure modes is?
Icon for the state of the switch's APU during a frantic game session with the USB plugged in.
Yeah, mechanically, all the small USB connectors are rubbish. USB-C might be a bit better for longevity than micro-USB, but I have a Thinkpad with a USB-C charging cable and it's prone to falling out. Barrel connectors are much better from a mechanical point of view.
On the other hand, I'd love it if these laws had the side effect of outlawing Dell's utterly broken power-supply DRM. I've had that fail on two work laptops in a row. And it fails on the motherboard, so swapping chargers doesn't help. It's a notoriously crap implementation.
There is no rule against providing two alternative power connectors. I have old HiFi amps that accept three or four.
Moreover this kind of enforcement puts pressure on USB-D or whatever to be backwards compatible with USB-C, which can only be a good thing until a far better connector creates a new standard.
... except when the charger is able to deliver up to however many volts, defaulting to the highest output, or is even a fixed voltage output for a specific device but you have a different device which requires a fixed low voltage with no voltage negotiation so it blows its controllers off.
It's not good enough having a standard physical connector if the device/psu is not *always* defaulting to low power and negotiating *upwards* if it's connected to an "intelligent" device. (I believe a simply example of this is a Raspberry Pi USCB-C power connector ... be careful what you plug into it)
"I've not encountered any USB-C chargers which don't do the voltage negotiation"
Until the controller goes phut. I'll bet there's a failure mode that will apply excess voiltage to the connected device. Negotiated dynamic adjustment of critical things like supply voltage requre the supplied device to be resilient against all possible failure modes, and that's expensive to implement. I wonder how many actually are.
If the switch-mode PSU is designed by an idiot and built to the lowest possible cost then yes, it will fail in such ways.
Such a device also won't comply with the electrical safety standards, so the importer will be found liable for the damage.
Now if you'd been complaining that the safety standards for CE (and UKCA) marking aren't enforced, I'd agree with you.
Perhaps making Amazon directly liable for things sold on its website would concentrate minds.
Why would you assume lawmakers would not consider those types of thing? The EU law does not just specify the physical connector. It actually specifies that chargers and devices should follow IEC 62680-1-2:2021, more commonly known as USB Power Delivery, which deals with all those issues. So should anything go wrong there this means either the charger or the device is not in compliance.
I'm assuming India will do the same, they can basically just copy the EU rules. Making sure those are the same will make life easier for manufactures as well.
This applies to radio devices which are subject to certification before they are allowed to be sold anyway. So it should not be to hard to add a check for this into the process. But even if you don't, you can just wait for consumer complaints and act upon those. If you bring in thousands of devices you want to sell, but they get taken of the shelf before you sold half of them and get slapped with a fine, you'll learn your lesson pretty quickly.
On top of that, in the EU consumers will have a valid warranty claim if it turns out they bought a device which does not comply.
Sadly those things often come bundled, rather than as the main purchase.... and thus get used around the place, possibly overloading them etc. I have 2 on my desk right now that Big Clive would poke fun at... the clearance between 240v and 5v is around 0.2mm, and the cases just pull open when removing the plug from the socket
The plan is to stop bundling chargers/PSUs with devices. The upside is you have fewer chargers sitting around and more or less any you do have will work with any device. The downside is that many people buying a new charger/PSU will buy based on price and some will be buying "fake" chargers from Ali Baba et al for £2 including shipping from China.
Right, so we’ve gone from “bundle an ultra-simple OEM charger, just caps and an inductor worth 50p in every box, free at POS to customer”
To “Complicate the standard charger to a 700 page spec, now you must pay £20 minimum for a charger. If you’re only paying £5 at best it won’t work you throw it away immediately, medium you’ve just trashed your £1000 phone, at worst it explodes in your face and you will now be costing the NHS £80k to operate on your eye which might or might not save your eyesight”
Yeah, I’m not immediately seeing how that benefited the environment….
Put another way, how many 50p cap-and-inductor wallwarts would you have to landfill to exceed the cost of a single Grenfell tower that goes up in flames because *one* person bought a dodgy plug for a device that only needs 1A but the plug is *capable* of 30A and erroneously supplies it?
Rubbish. *All* chargers were ultra-simple until both Lightning and USB-PD came along.
What you mean, but have failed to think through, is that the cheap generic substitutes you could buy off Amazon, hadn’t had any safety checks done.
But the *OEM* ones shipped in the box by Nokia etc…..show me a single case where a branded Nokia charger caused a house fire? There aren’t. Because Nokia weren’t stupid enough to cut costs on something that only cost 50p, when their whole brand depended on it.
The problem is that EU and India are now forcing the market away from OEM branded goods, where the manufacturer risks a $1000 product reputation on a 50p add-on…..into the hands of generic manufacturers solely competing on price for a few quid.
This ends exactly the way anybody would expect who has been on the planet a few years, which is *a hell of a lot of house fires*.
Those old Nokia chargers were not ultra-simple. They were quite complex devices, partly because SMPs were not the off-the-shelf "just add caps and a transformer" chipset they are now.
These days, every device manufacturer buys one of the standard chipsets. Or an entire off-the-shelf unit and puts their sticker on it, usually just hoping that it complies.
Yes, Apple and Samsung do that too.
That's what I meant.
No, sorry that’s a very naive view, as if the load being driven were simply a passive resistor to which you are applying volts, which it absolutely isn’t.
Firstly, people use the word “charger” incorrectly. The wall wart thingy is *not* a charger. The charger for the phone battery, which actually sets the volts and amps driven towards the battery, is internal to the phone. The “USB-C charger” wallwart is actually a power-supply for the internal charger.
The internal charger may well be able to sink 5V, 5A instantaneously, from its power supply, ie have a 1 ohm input impedance. The system end-to-end relies on the internal phone charger sensing both temperature, and the battery charge parameters, and signalling request for duty-cycle back to the USB-C wallwart. A battery charging at 10 mA might be getting only 1% duty-cycle to the charger, smoothed out by the charger. If the USB-C wallwart isn’t functioning correctly and ignores signalling, it will push 5V to the 1ohm impedance, thus sinking 5A to the internal phone charger all day long. The internal battery charger will function correctly, and limit what’s going to the battery, if/when it all starts going overtemp.
However, you’ve still got 5V external driving into a 1 ohm load in the internal charger, and that is going to continue dissipating 25W within the charger circuitry irrespective of temperature. Eventually it will melt the chip off the board. Sorry.
There is no way round the fact that the world has moved from passive electric chargers, which *physically can’t* drive more than the boilerplate number and therefore pose no safety risk, to ones which *physically can and do* drive 100x safe thermal load instantaneously, relying on complex signalling managed in *firmware* to stop the flames by duty-cycling
That's not how any of that works. Please step away from your keyboard and do some real research before posting such scare stories.
The phone will consume whatever power it is designed to consume at most. It will draw less under most circumstances, but never, ever more.
When it stops charging the internal battery, the power consumption drops. They are non-linear loads.
It literally does not matter at all if the wallwart is capable of providing more. The phone physically cannot draw any more than it wants to. This is basic physics!
Note that it does matter if the wallwart cannot provide enough, as the supply voltage will drop when the phone tries to consume (eg) 3A from a 1A (nominal) supply. In that situation the phone will either draw less power (due to physics) or if the voltage drops too far it will shut down the charging circuit entirely until the supply voltage recovers, then start again, then shut down... Which is bad for its long-term health, of course.
To prove this, plug a small electronic load into a power socket and note that it does not explode.
The socket on the wall in the UK/EU is capable of providing over 3000W. Thus all 230V nominal loads smaller than that are supplied the power they need.
I've upvoted you, but have a slight correction, a SERIOUSLY crappy PSU combined with poor input tolerances in the charged device, could result in say 230v ending up on a port expecting 5v max, at that point, all that smart electronics DOES become a resistive load. And a successful incandescent lamp.
You simply aren’t correct.
“When it stops charging the battery, the power consumption drops”.
Yes, because the charger internal to the phone (call it CIP) requests a lower duty-cycle from the wallwart.
So long as the system operates correctly, the efficiency is maximised when only a small fraction of power is dissipated in the CIP. But that’s not *physics* it’s protocol design.
Take the older scenario of wall warts, pre-dynamic power, to understand this better. The wallwart chucks out 5V, 1A max. The internal charger has an input impedance of 5 ohms. While the battery is charging, almost all of the 5W is going into the battery, with maybe 10% being dropped in the internal charger. Once the battery is full, it stops charging….but the internal charger still has an input impedance of 5ohms. It does *not* depend on the output load. 5W is now being dissipated in the internal charger. You can *measure* this. It isn’t efficient or smart, but the phone is thermally sized to dissipate 5W indefinitely.
Now slightly different, attach a phone with input impedance 10ohms. Sure enough, the power drawn by the phone drops - the wallwart doesn’t “force” in more current. But the important parameter is the phone input impedance, which is *fixed*, and the thermal design of the phone that supports that implied power dissipation.
The dynamic power scenario is way different. The impedances have been dropped hugely…..1ohm down to even 0.2ohms is not unheard of. The theory is the same, the phone can charge super fast while all that power is going into the battery. But the phone is no longer thermally sized to take that indefinitely. It can and will overheat. It is protected by duty-cycle protocol.
This is all totally different to your “240V AC plug can supply 10A but won’t”. In that case, all it means is that the input impedance to the device is fixed by components, and might be say 100ohms at 240V, or 1 ohm at 5V. But it’s fixed by components, and thermally sized to dissipate at that rate.
Seriously, stop digging. You're in a really deep hole and making yourself look really foolish.
When the charging stops, the phone's (effective) input impedance changes radically and it draws far less.
I suggest you actually measure the current draw.
Yours, an actual electronic engineer who has designed these things - at both this scale and far larger.
For what I can tell, we agree on the measurable values of input impedance and current draw, in this rather simple operating mode of the myriad supported modes. We disagree on emojis like “basic physics” and the philosophy of “what does it mean for a phone to want something, and in which PCB component is that desire located”.
Also, without wishing to delve too far into your misunderstandings:
Given that you believe (*I agree correctly*), that the input impedance of the phone as a whole drops when it no longer requires battery charging current. And that the phone effectively clamps its input current down; such that the external Source does not “drive” more current in than “whatever internal component controls that input impedance. Why have USB-C PD put into the spec digital signalling negotiation of Current Capability between Source and Sink? Again, given that is a hundred pages of effort, and indeed most of the overall spec and implementation complexity?
Because you are 100% correct. For the one very simple case you have mentioned, a very simple solution is just that the Source always drives its capability, or more exactly clamps at it, while the Sink just sets its input impedance which clamps the *actual* current draw. And Bobs your uncle. No negotiation required, no signalling digital or otherwise. Not just does it work, that’s the way it’s always been done for wallwarts until now.
So, why did they specify it, given the cost? Do you think there might be any use-cases you might not have considered? I’ll leave it there….if you are a curious fellow you might pull on the string. Or you’ll flame me…I’m genuinely trying to help.
I believe a simply example of this is a Raspberry Pi USCB-C power connector ... be careful what you plug into it
We're all entitled to our beliefs, but the standard RPi USB-C wall wart pushes out 5V @ 3A, literally the lowest power that a USB-C charger could kick out. Its also entirely in spec, it never uses a higher voltage than that.
I think you are mis-remembering that early RPi with USB-C had out of spec USB-C implementation that misidentified itself to more powerful chargers and cables, and had the opposite effect - the chargers saw the RPi as an audio device, and refused to send power to it.
One important detail:
It can supply up to 3A @ 5V. It doesn't push it out.
If the Pi is idling, it'll be supplying a few mA as that's all the Pi draws.
The main reason for wanting a high capacity PSU on your Pi is to ensure the voltage doesn't drop below that needed for your USB mouse, keyboard etc when the Pi is doing a lot of work.
Ummm…you’re not going to like this…
One of the umpitty-billion features supported by USB-C is negotiated functionality dependent on country-code flashed into the device.
Same USBC plug, plugged into a Samsung S9 sold in France, absolutely can do something slightly different when plugged into a Samsung S9 sold in Germany. It’s in the spec.
Among the things it can do is “functions determined by National Authorities”
Look, in case you think I’m making this stuff up, have a read of this thread from some poor bastard tasked with implementing this spec
https://mobile.twitter.com/whitequark/status/1035767108846673921
“Cursed”, “Horrifying”, and “ what the actual fuck. the USB PD negotiation protocol was made by someone who wished they could work on Ethernet instead. because this is like an Ethernet-lite. who came up with this shit”
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
