USB power vs AC/DC adapter

Hard disk drives take (typically) over 500 mA to start up. So there's more current draw than any standard USB port offers. Two plugs, means you get the limit of available current from each port, and that can run the drive.

Laptop computers with only one port are ... the other problem. The DC adapter is voltage and current limited, cannot (1) reach high voltage to reverse-bias the USB power transistor (there's a transistor to switch off or throttle the USB power, used for sleep mode etc.), or (2) supply any significant power to a powered-down USB interface connector (it won't power up your laptop).

I suspect the adapter would be set for 5.1V, just to remove the USB power pins from the

Hmm, so VUSB is actually protected by the USB protocol from reverse biasing, even if it's shorted together with the DC jack's 5VDC?

Also, the startup power of most 2.5" HDDs are closer to 4-5W, or

800-1A rather than 500mA. Seek/read/write current draw is closer to 500mA. This is from measurements as well as various drive datasheets/ tech specs. So, isn't it gonna be a problem even starting up a drive even when you use a USB Y cable? Assuming the controller and any other chips in the enclosure also draw some amount of power. Makes me wonder how these enclosures even work!

Makes me wonder

When I first decided to "learn" USB, I was convinced that USB had a

500mA limit. I must have read that in a dozen places. (And purchased several books on the subject!)

But there's some trick to this... And as with everything else USB, its complicated. The book I need is at the house. I'll look this trick up for you and post it tomorrow. Just because the drive needs more than 500mA does NOT mean it won't run 800mA from USB ...contrary to everything I once read about power management on USB hubs, controllers and devices.

As for the 5V jack, I'll bet this is handled via USB also(?). In other words, if power supplied by jack, it's not power-managed ("supplied") by USB.

Like they say: "The price of simplicity is complexity."

-mpm

Perhaps that is the case, but imagine if you're plugging the drive with the 2 USB connectors into a self powered hub. The hub would only be provideed 500mA from the PC USB port, and would only provide 100mA downstream to connected devices(if the USB specifications are to be believed). What happens then? I've done some googling also and read that some earlier Macs do provide more than 500mA on the USB port, but that newer Macs actually do limit the current to 500mA. Add into that confusion some non-compliant USB ports(front panel or otherwise) that provide even less than 500mA, and it sounds like a real mess for external USB enclosures manufacturers!

I've looked up some USB-ATA bridge chip reference designs from Cypress/ NEC and they have a jumper to indicate if it's bus or self powered, i.e. only VBUS or the DC jack is connected to the LDO and never both at the same time. Now obviously this wouldn't work too well with enclosures since we can't assume users are smart enough to switch it too and from, and it's too much of a hassle anyway. I've never seen an enclosure that has any kind of such switch. I wouldn't be surprised if manufacturers just shorted the 2 pins, hope the 2 USB connectors are enough, hope nobody uses the DC jack and call it a day!

I'd be interested to read what your book has to say though...

Sorry in the first paragraph I meant bus powered rather than self powered hub.

Sorry. Could not locate what I was looking for. I am missing one of my USB design books..., perhaps it was in there?

Worth mentioning:

1 - I might be crazy.??

2 - Everywhere I read last night: USB is limited to 500mA max per port, and only powered hubs and controllers can deliver that. (Unpowered hubs can not.) Duh?!... Nothing new here.

3 - I did review a circuit that sat on the USB port and gobbled up as much power as it could, delivering the load whatever it needed, and storing the leftover "elsewhere" (presumably a battery or capacitor bank). When needed, the load could draw on this reserve. Note: This is not the trick I recall reading about... But it's a pretty cool circuit.

4 - As others have posted, several USB manufacturers do not seem to enforce the 500mA restrictions. I have not confirmed this, nor the comment about laptops generally enforcing same. (Which seems logical as they run from portable batteries and attention would likely be paid to this.)

5 - The notion of a peripheral acting as 2 devices, enumerating for high power on both ports and then combining the power to effectively circumvent the USB per-port (i.e., per "device") limitation is a valid approach. Power comes from both USB connectors, data from just one. Unfortunately, not much help for PC / Laptops with a single USB port.

6 - IBM has a "PoweredUSB" approach which is basically a USB with a custom-configured power plug adjacent to (and made an integral part of) the USB jack. Of course, your laptop/PC isn't going to have this, and neither is anything sold at CompUSA, etc... And again, this is NOT what I remembered reading about circumventing the 500mA limit.

So with that, I guess that senility is setting in, or I just plain read it wrong. (Or the source material itself was wrong...)

Until I see the doc (?), I guess I'm going to go back to my original

100mA / 500mA / 500uA belief system. -mpm