Pi 4 and USB C

I send you specs with links to sources and you again spit the same sh*t. It is not polite at all.

Except it doesn't

Indeed

--
?Progress is precisely that which rules and regulations did not foresee,? 

  ? Ludwig von Mises

I was talking about pre-4 Pis, they have USB 2 and the specification on that (except for some 'power only' ones, which the Pi's USB isn't) is a maximum of 500mA. So cables which are perfectly OK for the USB

2.0 specification don't necessarily have to be able to carry the current required by the Pi.

I agree that with the advent of USB C/3 on the Pi 4 the Pi is well within the specifications for those cables/interfaces so cables conforming to the specifications should be capable of carrying the required current.

--
Chris Green

All Pi versions that have USB 2.0 interfaces take more current than is allowed by the USB 2.0 don't they? They're not exceptional, lots of USB 2.0 interfaced devices do the same.

--
Chris Green

Chris, you are again mixing up things. The one thing is the power supply (we are talking about) the other thing is the USB port on the device - here you are right it would supply about 0.5A

What we were *actually* talking about was the *cable*. What I was originally trying to say was that USB 2.0 *cables* are not rated to carry the current that the Pi takes.

I said that even a Pi 2 I am running takes more current than many/most USB 2.0 cables carry without a significant voltage drop. I found that only about 1 in 4 cables kept my Pi 2 happy. The point was that these cables fulfil the USB 2.0 specification but are not heavy enough to carry the current a Pi requires.

--
Chris Green

No he's not, he is pointing out that cables made to spec for USB-2.0 may not be able to carry more than 0.5A because they don't have to and therefore will likely cause problems if used to power a Pi no matter how good the supply. In this he is quite correct - although it took me a while to grok the point he's making.

--
Steve O'Hara-Smith                          |   Directable Mirror Arrays 
C:\>WIN                                     | A better way to focus the sun 
The computer obeys and wins.                |    licences available see 
You lose and Bill collects.                 |    http://www.sohara.org/

Yes, exactly! Sorry if my original comment wasn't too clear.

... and, as I said, when I tried some of the many USB micro-B cables I have only a minority were able to power my Pi 2 satisfactorily even though the power supply was more than capable.

--
Chris Green

OK, now I understand. It might depend on the cables, because as mentioned recent USB2 cables (mainly provided with phone chargers) are capable of caring the load (2A*5V=10W)

regards

On Sun, 21 Jun 2020 13:43:44 +0200, Deloptes declaimed the following:

And do those cables also provide data transfer? I've had a few cell phones that came with power-only USB-2 cables -- no data pins. {And strangely, one of those would NOT charge using a common USB-2 charger [vs phone dedicated charger] but WOULD charge when plugged into a laptop at work... and I don't think the laptop had those special "charge mode" USB ports)

--
	Wulfraed                 Dennis Lee Bieber         AF6VN 
	wlfraed@ix.netcom.com    http://wlfraed.microdiversity.freeddns.org/

All cables I received in the past 10y with the mobile phones I bought provide charge and data. I bought 5 phones to be precise. I bought the RPi4 last year and did some research (homework) what power supply and cable do I need to power it up. This is the only one using USB-C ATM although I had one other phone using USB-C for few weeks.

Charging on the USB2 of a computer port is, as explained above, done at

0.5A - it takes ages.

Yes, if you look into it the USB 'standard' is quite complicated in that cables can be 'power only' and in that case there may be resistors across pins to indicate this to the devices at the ends.

This is only for 'old fashioned' mini- or micro-B cables, USB C is a whole new ball game, much more complex.

--
Chris Green

Indeed. A good example of how *not* to develop a protocol. Define a poor standard to start, without a clear upgrade path, then keep piling on complexity :(

--
W J G

Not everything is about systemd. Its just that poettering wishes it was.

--
?Progress is precisely that which rules and regulations did not foresee,? 

  ? Ludwig von Mises

We blew past the 500 mA limit long ago, well before the introduction of the Raspberry Pi. The iPhone was out 5 years earlier and needed 1 A for full-rate charging. By the time the Raspberry Pi was introduced, there was a rather large field of devices needing 1, 1.5, 2, or more amps to run and/or charge. The tricks used to signal current draw with only passive components have even been added to the USB specs, so they really aren't "more than USB," as you put it.

_/_ / v \ Scott Alfter (remove the obvious to send mail) (IIGS(

Top-posting! \_^_/ >What's the most annoying thing on Usenet?

No, but a USB 2.0 cable that also has data lines really only has to carry 500mA to be 'to spec.'. As you say lots of tricks (I'd call them bodges personally) have been added to the USB spec. to handle all the various power requirement additions but to expect any/all cables to be able to handle them all is a bit hopeful.

All I was originally pointing out was that it's (very) often the cable is the issue when running a Pi, you can have a very capable power supply but if the cable isn't good enough it's not going to work reliably.

--
Chris Green

might be true, but in my opinion it is exaggerated (I think it is the right word in English). A cable is a copper wire to help electrons pass through. What else can go wrong with the cable? It will just melt down :)

But if I am not wrong for a cable to handle 5V at 2A with length of 2m and drop of 0.5% the wire should be approx 6sqmm (diam. approx. 2.5mm) for 3A it should be 8.5sqmm (diam. approx. 3.2mm).

Someone mentioned additional passive electronics in the cable - now this is mean. At the end it might be best option to soldier your own cables :/ However the HP one is powereing the RPi4 just fine. I personally assume the electronics would be in the power supply and in the device.. AFAIK RPi would not run if there is not enough power instead of melting the cable :)

No idea why you had such bad experience with powering from USB power supply.

Sorry, but that simply isn't true. The USB specs say quite a lot about USB cables - it's all freely available to the general public, so I recommend that you download at least one of the specs and read the section on cables.

You may well realise, after you do so, that lots of "USB cables" out there really don't conform with the USB specifications. A lot of them are clearly too thin.

David

This isn't a purely theoretical problem. One of my devices (a Hudl2) will only charge properly if I use a short 15cm cable, and then only with some chargers. I *guess* it's really (perhaps non-conformantly) fussy about getting a full 5V even when pulling its full charging current.

This is a pain if I want to use it while it's charging, so I'm planning to get a longer cable advertised as using thick power conductors, and see if that helps.

John