James said:
Paul wrote on Thu, 23 Sep 2010 17:36:45 -0400:
It is FSU234 and I think is a fairly recent model.
The model number could be F5U234 ("five" instead of "esse")
http://cache-www.belkin.com/support/dl/p74386ea_f5u234ea_uk_hr.pdf
If you're currently using the wall adapter with the hub, try
disconnecting the wall adapter and running the hub in "bus powered"
mode.
In bus powered mode, the 500mA from the computer, powers the hub plus
up to four 100mA loads.
That won't work with 2.5" hard drives connected to the hub, as those
typically try to draw 500mA from the bus/cable as well. They're "high
power" loads. So this debugging technique can't work, with 2.5" drives
as test devices (very few 2.5" drives come with their own wall adapters).
But a 3.5" drive, has its own power source inside, and only a couple
milliamps are drawn from the hub in that case. A 3.5" drive is no
load at all.
Disconnecting the hub adapter, for a test, is viable if you have only
3.5" hard drives connected. Give that a try.
One problem self-powered hubs can have, is "shoving power up the cable".
The hub really should have some protection for that, but it is hard
to arrange (the usage of diodes, won't fit well within the voltage
drop budget of the USB spec). By disconnecting the power supply from
your hub, I'd attempting to see if the problem is related to reverse
power flow.
I have one reference schematic for a hub here, and they use a relay
to disconnect the 5V lead on the upstream USB cable (to the computer).
The relay is activated, by the presence of the wall adapter 5V feed
on the hub. So when that hub design is wall powered, the relay inside
the hub "clicks" and opens the 5V wire path from the computer,
so the power can't flow backwards. If the wall adapter is de-energized,
then the relay allows 5V from the upsteam cable to flow again.
But relays cost money, and while that is a technically excellent
solution (no voltage drop etc, meets USB spec), it costs too much
to put in actual hubs.
I cannot relate the two issues. Reverse power flow, shouldn't prevent
the shutdown of the computer. But it's one thing to check.
The second possibility, would be related to power management software. But
I can't imagine a USB device using standard Microsoft drivers, refusing
to shut down. So that doesn't make much sense either. I'd have to bet
on a reverse power flow issue of some sort, instead. The program
UVCView, might be used to review config info coming from the hub, but
I don't know what would constitute "bad" info, if I had to look.
The program "dumppo", can be used to look for claims of power
management driver incompatibility in regular computer devices, but
I'm not aware of USB stuff appearing there. And I also haven't seen
any examples posted in recent years, of anything actually getting
listed by dumppo. Any manufacturer worth their salt, isn't going to
be admitting they can't get their hardware to work on an ACPI computer.
That would have stopped ten years ago.
Paul