Richard Prosser <PICLIST@mitvma.mit.edu> wrote:

> I am in the situation where it looks as if I will have to run a
> USB "Normal"   Rate = 12MBit/sec connection through a 3kV isolation
> block. Has anyone done this of has any ideas as to how easy it
> might be. The data is encoded NRZI with bit stuffing to prevent
> long runs of 1 or zero so a transformer may be adequate - but the
> control systems breaks these rules to identify control codes - I
> think. Also, one line of the differential pair is pulled high to
> indicate the data rate - so this would need to be detected &
> reinserted separately.
>
> It can obviously be done with sufficient complexity, but I am
> wondering if anyone has had success with a simple way?

Richard:

My first recommendation is to put the isolation between your
instrument/sensor/whatever and the USB microcontroller, rather than
between the microcontroller and the PC.

If you can't do that, you have a difficult problem to solve.
Isolating a USB device from a USB host isn't easy, since it's hard to
determine the direction (host-to-device or device-to-host) of the USB
messages.

Opto-isolated USB hubs do exist, however.  B&B Electronics, for
example, sell an opto-isolated 4-port USB hub; more information is
at:

    http://www.bb-elec.com/product.asp?sku=UISOHUB4&dept_id=145

-Andy

=== Andrew Warren -- aiw@cypress.com
=== Principal Design Engineer
=== Cypress Semiconductor Corporation
===
=== Opinions expressed above do not
=== necessarily represent those of
=== Cypress Semiconductor Corporation

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.