USB Amperage?

[T3Kgamer]

Kind of a weird topic but I was planning making a device that can charge more than several usb devices at a time.
But I'm not entirely sure how that would work, say if I just pumped 15 amps through all the ports am I going to be blowing things up? Or do devices just draw what they need as with most wall bricks and PC power supplies?

[TechyBen]

AFAIK smart charging/fast charging checks for continuity along some of the data pint/ground pins. So some devices will not draw it if not detected as "fast" charging safe.

[T3Kgamer]

That's good to know I've seen on cheap power bricks they just short the data pins to each other.
As far as "fast charging safe" goes they're not going to draw as much as they can are they? that wouldn't exactly be good 
I know most ^cheap cables are barely rated for the 2amps that most wall adapters put out nowadays, i'd hate to be frying cables and phones

[TechyBen]

3 minutes ago, T3Kgamer said:

That's good to know I've seen on cheap power bricks they just short the data pins to each other.
As far as "fast charging safe" goes they're not going to draw as much as they can are they? that wouldn't exactly be good 
I know most cables are barely rated for the 2amps that most wall adapters put out nowadays, i'd hate to be frying cables and phones

I don't know. I know to some extent that is true. But you also need to regulate voltage well. When you say "planning on making", what is your experience? I've only done Raspi Pies and Arduinos.

 

IT needs to be DC, and nice strong diodes!

[FlappyBoobs]

500ma is the USB spec for 1 to 2. USB 3 allows 900ma draw.

 

However there is also a separate standard for charging via USB, if supported a USB 2.0 charging port can give up to 1.5A and a USB 3.0 port can give up to 5A, but you need to use a charging specced port for that to work, and most PCs do not have this as standard, it's more for dedicated charges than devices.

 

As for your idea, you can put in as many amps as you like and the device will draw what it needs, however you need to be careful, most devices will take what they need and nothing more, but some will just draw everything in which will over heat the battery. An iPhone 5 suffers from this problem for example.

 

Then you need to take into account the different types of fast charging. One way is to just turn up the amount of amps available to the device i.e a phone that used a 1A charger stick it in a 2A one and it will charge faster.  But if you have a Samsung then you need to be aware that it actually will LOWER the amps on fast charge...but it will increase the voltage in order to achieve the same effect. 

 

Put simply charging devices is not an easy task to do yourself with your own hardware. A much neater solution would be to simply hide a 4 or 6 port power cord behind something and run long USB cables to the devices you want to charge. You could even use extension cables into a box so that it looks like you have your own custom charging block, but really it's just a bunch of device charges hidden inside a case. (take care with cooling if you are doing this).

[T3Kgamer]

USB PD, the "box" is going to include fast charging but on dedicated ports

I like your idea of the hidden multi-port, but you did answer my main and pretty much only concern.
I'll probably regulate each port in that case, or each of a few. I was mostly wondering if I could just put the unregulated 5v 15+a into the board and call it good, i'll probably use some sort of current limiting component to do this instead.

[T3Kgamer]

Just now, James Evens said:

Take a look at USB PD ICs. They will take care of all the protocol stuff and you basically just provide power.

Definitely will. didn't even realize there was a specific IC for that I just figured they used a 5v?a regulator of some sort

[mariushm]

It's more complicated than that.

 

Originally, USB 2.0 was supposed to allow maximum 5v at 0.5A but to save money and simplify circuits on most motherboards there were four connectors grouped together and protected by a resettable fuse that tripped at over 2A of current.

So, a device plugged in the computer could often take more than 0.5A of current, up to around 2A of current, if the other devices in the other usb ports that were grouped together would not pull significant amount of current.

 

USB 3.0 raised the limits to 5v and 0.9A.

 

However, various phone companies implemented phone chargers which could give more than 0.5a or 0.9A to the phone. The battery charging controller could figure out how much current the usb charger can provide by checking the voltages on the data wires in the usb port.

So for example - and these may not be the correct numbers - if the battery charging chip measures 2.5v on the data receive wire and 2.8v on the data transmit wire, then it knows it can safely take up to 1A of current from the charger. If it's 2v on both, maybe this tells it can take up to 2A.

Samsung had their own combinations of voltages, Apple had their own combinations of voltages.

 

The usb guys came up to clean the mess with their usb battery charging specifications and now you can get small chips you can put inside a charger and in a device and these chips can talk through the usb cable and the chip in a device can "learn" what the charger is capable of - instead of measuring two voltages on the data wires, it actually gets data through the data wires and receives the maximum capabilities of a charger.

Here's the battery charging specification from usb : https://composter.com.ua/documents/BC1.2_FINAL.pdf

 

after usb 3, they came up with usb power delivery, which allows up to 100w ... basically by default, you get 5v and 0.9A or a bit more, and a chip on the other end of the cable can "negociate" with the charger or usb controller on the other end to get more current or higher voltage ... for example, you could get 5v, 12v or 20v  and current up to 5A, if my memory is correct. So for example, you could get 12v x 5A  = 60 watts through a proper usb cable, and power a lcd monitor or whatever. 

 

Note this is different than QuickCharge, which is a custom protocol and custom chips ... but it works the same... a chip in the phone can "talk" to a chip in the charger and tell it  hey i can handle 9v or 12v instead of 5v, so start sending me 9v for now and if all goes well in a few seconds I'll bump it up to 12v and go from there. 

The idea is that high currents means high losses in thin cables... so if you want high currents to charge batteries faster, you need thick cables. A work around that is to raise the voltage.. so instead of sending 5v at 3A = 15w and have 2 watts lost in the usb cable due to being too thin, you may send 12v at 1.5A = 18w and you probably lose only 1w or less in the conversion from 12v down to 4.2v or whatever the battery needs.

 

 

[T3Kgamer]

2 minutes ago, mariushm said:

Spoiler

 

It's more complicated than that.

 

Originally, USB 2.0 was supposed to allow maximum 5v at 0.5A but to save money and simplify circuits on most motherboards there were four connectors grouped together and protected by a resettable fuse that tripped at over 2A of current.

So, a device plugged in the computer could often take more than 0.5A of current, up to around 2A of current, if the other devices in the other usb ports that were grouped together would not pull significant amount of current.

 

USB 3.0 raised the limits to 5v and 0.9A.

 

However, various phone companies implemented phone chargers which could give more than 0.5a or 0.9A to the phone. The battery charging controller could figure out how much current the usb charger can provide by checking the voltages on the data wires in the usb port.

So for example - and these may not be the correct numbers - if the battery charging chip measures 2.5v on the data receive wire and 2.8v on the data transmit wire, then it knows it can safely take up to 1A of current from the charger. If it's 2v on both, maybe this tells it can take up to 2A.

Samsung had their own combinations of voltages, Apple had their own combinations of voltages.

 

The usb guys came up to clean the mess with their usb battery charging specifications and now you can get small chips you can put inside a charger and in a device and these chips can talk through the usb cable and the chip in a device can "learn" what the charger is capable of - instead of measuring two voltages on the data wires, it actually gets data through the data wires and receives the maximum capabilities of a charger.

Here's the battery charging specification from usb : https://composter.com.ua/documents/BC1.2_FINAL.pdf

 

after usb 3, they came up with usb power delivery, which allows up to 100w ... basically by default, you get 5v and 0.9A or a bit more, and a chip on the other end of the cable can "negociate" with the charger or usb controller on the other end to get more current or higher voltage ... for example, you could get 5v, 12v or 20v  and current up to 5A, if my memory is correct. So for example, you could get 12v x 5A  = 60 watts through a proper usb cable, and power a lcd monitor or whatever. 

 

Note this is different than QuickCharge, which is a custom protocol and custom chips ... but it works the same... a chip in the phone can "talk" to a chip in the charger and tell it  hey i can handle 9v or 12v instead of 5v, so start sending me 9v for now and if all goes well in a few seconds I'll bump it up to 12v and go from there. 

The idea is that high currents means high losses in thin cables... so if you want high currents to charge batteries faster, you need thick cables. A work around that is to raise the voltage.. so instead of sending 5v at 3A = 15w and have 2 watts lost in the usb cable due to being too thin, you may send 12v at 1.5A = 18w and you probably lose only 1w or less in the conversion from 12v down to 4.2v or whatever the battery needs.

 

 

 

That's extremely useful information. I thought USB PD was just 5v and quick charge was all the fancy 5v 12v 20v