Remix.run Logo
USB Power Delivery: Plugging into the Benefits(aptiv.com)
23 points by mooreds 3 days ago | 37 comments
edent 2 hours ago | parent | next [-]

I've basically stopped buying any portable electronics unless they take USB-C.

Currently travelling with a laptop, watch, toothbrush, eReader, camera, bug-bite treater, and phone - all charging from the same power brick.

I'm guaranteed of getting a replacement cable / charger wherever I am in the world if I need it.

The only slight snag is some cheaper itema refuse to use PD and insist on plain 5V/2A - buy most decent travel chargers have NON-PD ports.

Amusingly, most of the buses I've taken recently also have USB-C ports on them for ad hoc charging. Perhaps one day EVs will use USB-PD-Max rather than CCS :-)

blacksmith_tb an hour ago | parent | next [-]

I'm hoping we'll see most e-bikes at least use 240W usb-c pd charging (I figure I have about a decade until I will wish I had some assist and buy one, so probably by then, they'll have gotten there...)

I also have assorted products that won't charge c-to-c (some from respectable manufacturers even, like Philips), but I see you can get little adapters with 5.1K resistor you plug into said crappy devices to cover that, I will have to try some out.

ElijahLynn 2 hours ago | parent | prev | next [-]

Same!

I've also returned a few USB devices that ship with a USB-A to USB-C cable and ONLY charge in that mode, they also MUST charge with USB-C PD.

The two so far were a therapy light and some Zippo hand warmers. Like, who in the hell would design a device that has a USB-C port on it where only a fraction of chargers will work on it. It feels even worse than proprietary charges, because you see a USB-C port on it and think, oh I have a plug that fits it, and then it doesn't F**ing work. Idiot engineering/product teams, making the world suck with their falsely advertised USB-C ports. If anyone of you are on a team that ever makes this decision, just know that it is a stupid decision, and jump ship when you can.

myself248 2 hours ago | parent | next [-]

The thing is, making a 5v-only device PD-compliant is literally one resistor. It costs well under a penny.

It's pure ignorance, not a decision, but the lack of one. Lack of caring, lack of having an actual engineer involved, just slapping an oval-shaped port into a product where a trapezoidal port had been, and blindly thinking that magically makes it spec-compliant.

Or not thinking about the spec at all.

I return these devices too. Lots of them. My e-commerce returns over the last year are probably 50% PD non-compliance, 50% all other defects combined.

helterskelter an hour ago | parent | next [-]

There's an otherwise decent shortwave radio out there that was originally charged with a micro-usb, then they released a "new" USB-C model...except it will only charge with a 5V brick because they literally just swapped out the ports. Really annoying.

dabluecaboose 16 minutes ago | parent | next [-]

Oh man, please tell me it wasn't the CC GP-7. I have the micro version and have been hemhawwing about updating it.

dotancohen 44 minutes ago | parent | prev [-]

I'd imagine that a significant portion of the shortwave radio community is capable of soldering in the two resistors.

exmadscientist an hour ago | parent | prev | next [-]

It's two resistors, actually. But they cost $0.0003 each (that's 0.03¢, or just around 3,333 of them for US$1) from distributors. Though there appears to be a bit of a stock crunch right now.

So... yeah.

The bigger issue is not really the parts cost, it's the fact that it adds an extra part to the design that has to be purchased and tracked and assembled and blah blah blah. This is the real reason it often gets left off on the bottom-of-the-barrel products. Many times there is no other use for a 5.1kΩ resistor. And it might not even fit well at the cheap sizes (0603 or 0402), and going down to 0201-capable assembly factory flow just for these two resistors is not going to happen.

dotancohen 39 minutes ago | parent [-]

These companies are not manufacturing the device PCBAs, that is done by dedicated companies such as Flex. The PCBA manufacturing companies have warehouses of different resistors, and 5.1kΩ is extremely common. In fact, most PCB resistor values are quite flexible, to save on SKUs (in practice, to save on loading another carrier on the PnP machine) often if a specific resistor needs a specific value then all (or most of) the other resistors will use that value.

exmadscientist 9 minutes ago | parent [-]

I was speaking a little more towards the AliExpress end of things, which is a sadly high proportion of the devices out there. For the midsize CMs and up, you're right, they've got piles and piles of stuff and don't charge by the reel loaded.

5.1k is a surprising resistor value, a lot of modern designs don't really have anything else in that area. I'm often not able to combine anything with it when I'm cost reducing. 4.7k, sure, but there aren't a lot of those either... 2.2k is just not close enough a lot of the time (or ends up as 1k), and same for 10k. So, sadly, it often does stand alone.

megous an hour ago | parent | prev [-]

Only if the device's consumption is < 2.5W, which is what a USB 2.0 computer USB-A's data port limit is. Anything above that, compliance gets a bit more involved and complicated.

simoncion an hour ago | parent | prev [-]

> I've also returned a few USB devices that ship with a USB-A to USB-C cable and ONLY charge in that mode...

By "that mode", do you mean "1.5A @ 5V" permitted by BC, or do you mean "3A @ 20V" permitted by non-type-C PD?

> Like, who in the hell would design a device that has a USB-C port on it where only a fraction of chargers will work on it.

Who in the hell would design a charger that can do Type-C PD but can't do either pre-Type-C PD or BC? Does the charger in question also shit the bed when a USB 1.0 device attempts to draw 100mA @ 5V? I hope not! Were it me, I'd return that crappy thing for a refund.

duskwuff an hour ago | parent [-]

> By "that mode", do you mean "1.5A @ 5V" permitted by BC

Neither - OP means devices with missing CC resistors which will fail to charge with a compliant PD source. (The A-to-C cable works because it provides 5V Vbus unconditionally.)

exmadscientist an hour ago | parent [-]

The A-to-C cable often does not work because the resistors are supposed to be in there.

So if you are having complete charge failures, try a different cable.

megous an hour ago | parent [-]

A-C cable assembly always works, CC signal is connected within the cable to Vbus via 56kOhm resistor, but that's only relevant to the downstream port, not to the upstream USB-A power sourcing port which does not have access to the CC signal. Upstream port provides power unconditionally within some limits depending on port type (CDP/DCP/USB3.0/2.0 data port/...).

exmadscientist 44 minutes ago | parent [-]

That's how it's supposed to work, yeah.

But there is some trash out there in the world. A lot of it, actually.

Some naughty cables work with some naughty chargers work with some naughty devices. Postel's Law in action, I guess?

Usually the best place to fix it is by getting rid of the bad cables. Usually.

mcsniff an hour ago | parent | prev | next [-]

I'll bite. What type of watch do you have that has a direct USB-C port on it?

margalabargala 13 minutes ago | parent [-]

They may be talking about something like the OnePlus watch, which does not have a usb-C port on it, but the charger device for the watch takes usb-C.

ianburrell an hour ago | parent | prev | next [-]

I think you are confusing the devices with USB-C that require USB-A, and devices that charge the standard USB-C 5V/3A/15W. The USB-A ones cheaped out in including the resistors that signal legacy USB mode, they work with the ones in the cable or adapter.

Lots of people assume that USB-C always uses USB-PD, but the basic signalling is done with resistors. Lots of devices only need 15W, and it is better than USB-A charging. If you want faster charging, buy more powerful chargers.

m463 an hour ago | parent | prev | next [-]

For travel I have a bunch of cables with adapters on the end (choose usb-c, lightning, micro-usb). Can use usb-c, but have the ability to use the others.

It has helped out in a bunch of unexpected situations (usually someone else's device)

MrBuddyCasino an hour ago | parent | prev [-]

You can always use a „PD Decoy“ if the voltage is USB compatible. A 5 / 9 / 12 / 15 / 20 Volt barrel plug is trivially USB powerable.

Animats an hour ago | parent | prev | next [-]

240 watts over a USB-C connector? What next, USB toasters and coffee pots?

dotancohen an hour ago | parent | next [-]

That's only a single ampre at standard European mains voltage. It's still a lot of power for those tiny connectors and insulation, but an order of magnitude insufficient for those appliances.

I bet that cable gets plenty hot at 200+ watts.

Animats 15 minutes ago | parent [-]

USB-C is limited to 48 VDC. Above about 50V, electrical safety codes apply.

m463 an hour ago | parent | prev | next [-]

I remember scoffing years ago at USB powered coffee warmers. Maybe the situation has changed.

ksec 31 minutes ago | parent | prev [-]

I mean my electric Fan and standing lamp are both powered by USB-C.

There are things that shouldn't be powered by USB-C. But there are plenty of sub 100W consumer electronics devices that really should be USB-C. I waited years before Panasonic released their lamdash shavers using USB-C.

hazkoulia 2 hours ago | parent | prev | next [-]

I'm looking forward to USB-C PD small format factor PC's. A decent amount of room in the PC cases is taken up by the power supply. And if USB-C could somehow provide a range of voltages to the motherboard, SFFPC's could be downsized even more

dotancohen 34 minutes ago | parent | next [-]

The issue is that a USB charger is not a USB power supply. A charger does its best, but makes no guarantee of constant power delivery or duty cycle. The power supply absolutely must provide its rated output at 100% duty cycle.

exmadscientist an hour ago | parent | prev | next [-]

> And if USB-C could somehow provide a range of voltages to the motherboard, SFFPC's could be downsized even more

You reeeeeeally don't want to do that. Cable inductance is a big deal, among other issues. You want the main DC-DC regulators on the board, usually right at the load, for the main loads. Most of the PSU bulk is for dealing with mains itself: handling 50/60Hz or mains isolation is just physically large. Getting in secondary 20V DC (or so) from a single connector and then regulating it down on board is pretty much the ideal solution.

(I can't even begin to comprehend the horrors of a USB-PD negotiation involving multiple voltages. It's already the worst standard I've ever had to deal with.* Don't make it worse!)

(* Not hyperbole, it is truly, truly awful. At least things like 60601 are bad because, you know, they're covering lots of stuff like lifesaving medical devices. USB-PD is... holy hell, it is just bad.)

omh 2 hours ago | parent | prev [-]

There are some SFF PCs that can take USB-C power.

Lenovo have some,but sometimes require adapter cards. And a few of the Chinese N150 units will take PD power

It's great for hot swapping and more portable than a laptop.

smallmancontrov an hour ago | parent | prev [-]

Speaking of which, does anyone know a line of PD Decoy modules to convert barrel jacks to USB-C without the atrocious behavior of "oh, the charger doesn't have 12V, here's 9V have fun!" that the early ones all did? Ideally I'd like a little red light to come on or something, but I'd settle for not silently browning out the device.

s0rce 19 minutes ago | parent | next [-]

There are things like this: https://www.adafruit.com/product/5501 haven't used it, trying to find one for my mini pc, mostly so I can use a usb c power bank as a tiny DC ups.

smallmancontrov 17 minutes ago | parent [-]

The description explains (in bold, multiple times) that it has the awful behavior I am looking to avoid.

miladyincontrol an hour ago | parent | prev | next [-]

iirc theres ones that do. However I dont recall there being any clean fix to the amperage constraint issues. Especially when a lot of usb-c chargers will vary output as they heat up with usage.

Which is kinda part the issue, usb-c charging bricks, they aren't usb-c power supplies, there is no expectation of sustained output capacity. Thankfully at least some the multiport ones have renegotiation more or less solved cleanly rather than what is essentially rebooting the PD controller.

smallmancontrov 18 minutes ago | parent | next [-]

I mean, so long as it cuts off and throws a red light I'll figure it out, especially if the brick got toasty, and then I can find a new brick. My problem is that "best effort" seems to be almost universal behavior among PD triggers and it multiplies the failure modes.

m463 an hour ago | parent | prev [-]

sounds like someone needs to create a DC-DC charger

dotancohen 33 minutes ago | parent | prev [-]

Actually, that was 5v, not 9!