dwn said:
Paul, I really appreciate your suggestions and the explanation too. Assume I get
this one, is there any chance, should it overheat and fried my motherboard? I
searched a few Newsgroups on "USB" way back to 2005, a few laptop users had
their motherboard fried! Thank you again.
I don't have any way to analyze the designs remotely. I can't tell you,
whether the customer review information of that nature is correct or not.
But you can take that information as a warning at least, that such
a possibility exists.
If you have a cheap or old computer to test with, use it first. And
see if there is any bizarre behavior (like after you select "Shutdown"
in Windows, and the hub is connected and powered by the AC adapter,
you can't start your computer with the front button). If you're satisfied
with the tests with the "disposable" test computer, then you can try the
expensive or hard to repair laptop.
*******
You could probably rig up a test, to see if power is available
on the main (Type B) port of the hub, by using a light bulb between VCC and
GND pins of the USB connector. A small bulb, like a 6V 100mA or so,
might be suitable, to detect if 5V power is available. Using a multimeter
may not be sufficient, because the multimeter doesn't apply a load and
you can't easily tell whether there is sufficient leakage that way to
cause a problem. You really need a "load test", to distinguish microampere
leakage, from milliamps or higher leakage.
For example, this bulb doesn't draw much current, and might be
suitable for a quick check. But to use this, you'd need to know
which pins are VCC and GND on the new hub.
http://www.radioshack.com/product/index.jsp?productId=2102814
In the Type B picture here, it would be pin 1 and pin 4, that
could have 5V power on them. You'd connect the adapter to the
hub, then access pins 1 and 4 of the Type B (square) connector,
with your light bulb. I looked at the Type B connector on
my USB ZIP drive, and it has those numbers stamped right in
the plastic. That is what I used to verify the numbering scheme
at least. (I don't have a hub that I can use, for this test.)
http://www.usbman.com/Guides/BelkinMotherboard_cable_assembly.htm
____
/2 1\ Pin 1 = VCC = 5V As viewed by looking at the
|3 4| Pin 4 = GND = 0V connector on the hub.
+----+
There might be 5V between pin 1 and pin 4, even on a good device.
But it should not be able to source milliamps of current. If a
small light bulb will light up, then you know it can source power
(and you don't particularly want that).
The example Radio Shack bulb is rated for 25mA (a tiny load). When
the filament is cold, the current draw could be double that. Since the 5V
supply is not equal to the 6V the bulb needs, it won't glow
quite as brightly as it would if connected to a 6V source.
(You're more likely to be able to buy a 6V bulb, than a 5V
bulb, which is why I picked that one.)
But it would be good enough as a quick check, as to whether
the hub is properly isolated or not. The hub should not push
power backwards on pins 1 and 4. The bulb should *not* light up,
when connected to pin 1 and pin 4. The reference schematic I've
looked at, for such designs, uses a small relay for isolation.
If there really was a relay inside the hub, you'd hear a "click"
from inside the hub, when the AC adapter is connected to the
barrel power jack (and has AC power available to it).
Look for the text string "Au9254A21 USB Hub Buspower and Selfpower Changed"
here, for an example of an isolation method using a relay. When the AC
adapter is plugged in, the VCC_UP from the square Type B connector
is disconnected from the hub. PDF page 16, in the upper right
hand corner, shows the upstream (Type B, square) connector and
its connections. The relay is used to disconnect VCC_UP for isolation
when "self powered".
http://logout.sh/computers/projects/usbradio/AU9254 R2-020108.pdf
Paul
