USB chargers, anyone ?

Does everyone out there think Wikipedia is a pile of lies and useless twaddle?

Scroll down to the Physical Appearance section.

VBUS is supposed to be loaded with no more than 0.5 amperes (0.9 for USB

Perhaps I should have read a bit further. This is probably what you're looking for...

Charging ports and accessory charging adapters The USB Battery Charging Specification of 2007 defines new types of USB ports, e.g., charging ports.[44] As compared to standard downstream ports, where a portable device can only draw more than 100 mA current after digital negotiation with the host or hub, charging ports can supply currents above

Charging ports exist in two flavors: charging downstream ports (CDP), supporting data transfers as well, and dedicated charging ports (DCP), without data support. A portable device can recognize the type of USB port from the way the D+ and D- pins are connected. For example, on a dedicated charging port, the D+ and D- pins are shorted. With charging downstream ports, current passing through the thin ground wire may interfere with high-speed data signals. Therefore, current draw may not exceed 900 mA during high-speed data transfer. A dedicated charge port may have a rated current between 0.5 and 1.5 A. There is no upper limit for the rated current of a charging downstream port, as long as the connector can handle the current (standard USB 2.0 A-connectors are rated at 1.5 A).

Before the battery charging specification was defined, there was no standardized way for the portable device to inquire how much current was available. For example, Apple's iPod and iPhone chargers indicate the available current by voltages on the D- and D+ lines. When D+ = D- = 2V, the device may pull up to 500 mA. When D+ = 2.0 V and D- = 2.8 V, the device may pull up to 1000 mA of current.[45]

Dedicated charging ports can be found on USB power adapters that convert utility power or another power source - e.g., a car's electrical system - to run attached devices and battery packs. On a host (such as a laptop computer) with both standard and charging USB ports, the charging ports should be labeled as such.[44]

To support simultaneous charge and sync, even if the communication port doesn't support charging a demanding device, so called accessory charging adapters are introduced, where a charging port and a communication port can be combined into a single port.

The Battery Charging Specification 1.2 of 2010 [12] makes clear, that there are safety limits to the rated current at 5 A coming from USB 2.0. On the other hand several changes are made and limits are increasing including allowing 1.5 A on charging ports for unconfigured devices, allowing high speed communication while having a current up to 1.5 A and allowing a maximum current of 5 A.

AFAIK, in order to pull more than 100 mA from a USB port, a device is supposed to do a digital dance with the host to request the higher capacity, up to 500 mA. There may be (read: probably are) dedicated charging ports that will cheerfully source 500 mA (or more?) to a "dumb" device without requiring the dialog.

Check out this page for the resistors that are needed to turn on most USB sources.

Isn't it sad all this garbage is needed to send 5 volts down a cable?

Needing three resistors is sad?

yeah, it is.

USB is not like ethernet where adding power happened later on and there is a chance to do weird stuff like plug two PoE switches into each other with a patch cable. USB doesn't support "illegal" scenarios like this due to the connectors, so any extra signalling is just bullshit to lockout devices or make you get stuff you should not need.

I've seen the USB 2.0 specs, and they're beyond unbelievable. You'd think what appears to be a simple interface would take only ten pages or so to define, but it takes 10 or 20 times that.

USB might be "over-defined", but at least it works. I can't think of a USB product I've owned that didn't work correctly. From my point of view, this is not an accident. Somebody did a lot of thinking to make sure there weren't any "gotchas" in the system specification.

USB

be

black

and

A couple cents worth of resistors to turn on the port is a simple and sensible way to protect the port. It was designed that way, not added at a later date. Look at the data sheets for some of the hub ICs.

Thanks for that William

On the face of it, then, it would seem that a voltage on the data pins is used to inform the USB device how much current it can take, which in the case of the 'port' being on a 1 amp charger, would be the full spec whack of

Arfa

Cheers

Arfa

USB

it's

USB

to be

thin

wires,

were

black

and

they

when

to protect the port from what? A device that doesn't work anyways?

hub ICs and resistor networks to provide 5 volts and a few hundred mA is just too hard?

Can anybody confirm that the clowns that made the miniDIN connector were involved in the creation of USB?

There is a typo on that document that I don't have the time to find the correct info for. This (1st paragraph, second last sentence) doesn't make sense:

"A charging port supplies up to 500 mA at 5 V, up to the rated current at 3.6 V or more,..."

Couldn't find a ready explanation for the rated current, but 3.6A may be what the author meant, however I see later where the rated current is more like 1.5A with 5A maximum.

Someone with more time can find the correct reference and fix that article...

John :-#)#

William Sommerwerck wrote:

...

almost the first 10 years of USB were complete crap.

every chipset for it was buggy (even big name ones like intel), it was slow, and devices other than a mouse were really pushing it. OS support was junk too.

My favorite knockout combination was Windows 98 on a Hitachi laptop. There was no chance in hell anything USB would ever work on that combo.

I had some digital camera with USB 1 or 1.1. it was actually slower than using the irDA port on the camera, which was emulating 115kb serial. yes, an LED and photodiode behind some plastic window emulating a 1960s method of connecting teletypes to furniture sized computers blew away the mess known as USB.

these days, usb mostly works. It's still obnoxious when devices can't handle being moved between ports though, which is proof USB or everybody that writes device drivers are completely stupid.

Does moving a scannwe between my front and rear USB port really require a driver reinstallation? WTF?

That might be true. My first computer with USB was purchased in 2001, and I never had problems with USB.

In theory, no, but that's the way it is.

If you go to Device Manager, select Show Hidden Devices, and rummage through the listings (especially Universal Serial Bus Controllers), you'll see a lot of grayed USB devices. Many of these represent the /same/ device attached to a /different/ USB jack.

In other words, the USB system organizes by ports rather than devices. If you plug a device you've used before into a different port, USB doesn't know this, and installs the driver for use by the new port. I don't know why it was done this way.

If this bothers you, keep track of which jack you plug a device into. Make sure that devices that often connected to the computer at the same time don't use the same jack.

By the way, it's okay to "prune" the listing of entries for devices you're not using any more, or are duplicates. The worst that can happen is that the device will stop working until you remove/replug the cable. A few months back I deleted something like a hundred entries, without wrecking anything.

That's almost entirely an operating-system issue... not USB per se.

It's a question of how does an application locate/identify/address a specific USB peripheral.

USB has a tree hierarchy... controllers, hubs (possibly several layers of them) and devices. Applications generally want to open a device using a fixed, repeatable identifier. Depending on the OS and the application, this identifier may be based on the bus topology (which controller, hub chain, and port) or on the unique characteristics of the device itself (model, manufacturer, type, serial number).

It sounds as if Windows is creating a unique, long-term identity for the device based on both of these - the device itself, and the bus topology to reach it - and that changing to a different port "orphans" the identity. It might be Windows itself doing this, or the manufacturer-supplied device driver.

Other operating systems (e.g. Linux) can assign a unique identity based solely on the device's unique characteristics, assuming that it has enough to make it unique (e.g. a real serial number), and the same identity will be assigned to it no matter what port it's plugged into.

Thanks for the explanations.

I don't know how long you've been using personal computers, but it seems you've never had to set up COM ports, which can be a major pain in the toches. USB largely delivers on its promise of "plug and play". Whatever problems it might have, they are offset by the fact that USB works with little or no fussing.

The only problem I can remember is a product that said you shouldn't install the driver until you've plugged in the device, when it was actually the other way around. Or maybe vice-versa.

I used a dumb terminal before I had a complete computer, so I'm no stranger to serial and all the goofy breakout boxes, null modem adapters and chain of adapters almost foot long to get stuff to work. It was all pretty stupid, but about as good as it gets for stuff that old.

that right there is the sign something inherently doesn't work correctly. If you did just plug something in, you'd end up in a world of hurt most of the time too- again, it's totaly senseless and the sign of a stupid committee at work.

Imagine your outlet exploding if you plugged in a new appliance before registering the device with your circuit breaker panel.

Or maybe an network device that if plugged into your hub/route/switch would completely destroy your network unless you installed magic drivers or code works or whatever somewhere else first.

It's all stupid and there's no excuse for it at all.