PDA

View Full Version : USB Hub problem.

tefleming
19 Jan 2004, 9:48pm
Is there any reason that some of my USB devices won't work off a hub?

My Ati Remote Wonder and Intel Wireless desktop controller won't work on the hub (worked fine on port)

But my HP printer works fine off either.

Windows autodetects the hub fine. Is there an alternate driver I could try?
TheSmJ
19 Jan 2004, 10:02pm
Some devices just dislike being on a hub. I forget the reason why.
Enverex
19 Jan 2004, 10:05pm
Is it a self-powered or host powered hub?
tefleming
19 Jan 2004, 10:06pm
It's a self-powered, Belkin F5U101
Geeky1
19 Jan 2004, 10:12pm
That's probably why. USB ports are spec'd for 500ma @ 5v. Putting too many devices on a self-powered hub results in more power draw than the USB port can handle.
tefleming
19 Jan 2004, 10:13pm
It's self-powered (not bus powered)
It plugs into the wall
Geeky1
19 Jan 2004, 10:17pm
Oops. :doh:

Don't mind me- I can't read today.
tefleming
19 Jan 2004, 10:29pm
Alright, in case this wasn't wierd enough, I unplugged the power cable and:

It all (Remote, Receiver, Printer) works fine...

I'm so confused...
shwaip
19 Jan 2004, 10:31pm
there may be a switch on the hub to switch between self and bus powered...make sure it was set to the right thing.
Straight_Man
19 Jan 2004, 10:45pm
Most HP printers do not draw USB power, and it looks like your HP is actually backfeeding power into hub USB power line-- If works, leave it unplugged, the dang printer is powering the remote and the Receiver through the hub.... :D

One way to try and isolate, would be to turn off and unplug printer from USB, do receiver and remote still work??? If not, the HP is feeding power into hub instead of drawing power, from its power supply, or hub is drawing juice from printer USB power trace.... While not exactly to USB standards, this has been known to happen and not hurt unless too much feeds back in.... If they still work, then they are not overloading the standard power draw limits either, and no harm done....

John.
seatech1
23 Oct 2004, 4:57am
Read this article
http://www.short-media.com/review.php?r=242

This is probably your problem. I had the same problem with a linksys usb hub and a lacie external usb hard drive. When I used other devices that drew their power via the usb, it caused me to have corrupted data written to the lacie drive.
Straight_Man
23 Oct 2004, 5:15am
Read this article
http://www.short-media.com/review.php?r=242

This is probably your problem. I had the same problem with a linksys usb hub and a lacie external usb hard drive. When I used other devices that drew their power via the usb, it caused me to have corrupted data written to the lacie drive.


It can be that also, you are limited to one port's worth of power draw for two devices AS WELL AS having one internal port (not root hub) try to handle two devices. Second, if the wiring gets at all crossed, and from motherboard to motherboard the pinouts can vary and NOT be to USB standards as nomrally the onboard ports are not necessarily to standard while the front plugin ports NEED to be to USB standard as to position and what comes into them, you can get all sorts of crosstalk that results in what you had. Power feedback also causes interference and\or crosstalk sometimes, the hubs typically do not have resistors to prevent backfeed of power into hub. At guess, the LaCie was gettign signals for two devices or cables got crossed. Not to put down SeaTech by any means-- that is absolutely not the intent here, but a power feedback results in a signal into the hub that the hub has to dump or feed out as a 5V signal high...

And if it is not connected to external power with a ground, it dumps, if it can, to any fground it can find. No wonder your USB HD had problems. A CD Burner, DVD Burner, video adapter, etc would also get crosstalk in the sense of a power load on a data line tossing things out of whack and with DC that could be a continuous high that the device was never wired to filter. This is the equiv of a hot line shorted to a data pinout or an interference from the field effect in a chip or component caused by a power load it was not intended to use.

It could be what seatech said, or in fact a backfeed tossing the hub chips or circuitry semiwild that caused the LaCie data storage mess.
seatech1
23 Oct 2004, 10:31pm
No offense taken. You sound like you know what you're talking about. So you're saying that usb hubs don't have a resistor (or diode) to prevent the power from feeding back? Didn't know that. I wonder if it's possible to put an in-line diode on a cable between the powered device and the hub to cure the prob? And the next question is, does somebody make such a thing?