Any advice on galvanic isolation for USB?

[BEP]

I have a need for complete USB isolation up to 5kV surges. USB power will be provided by a separate (already isolated) good quality power supply.

There are plenty of products for sale but I build my own. Has anyone succeeded with this task and willing to share ideas?

Thanks,

WW

[Peterae]

Not sure if this is of any use to you WW

http://www.analog.com/static/imported-files/data_sheets/ADuM4160.pdf
http://search.digikey.com/scripts/DkSearch/dksus.dll?pname?site=uk;lang=en;name=ADUM4160BRWZ-ND

[BEP]

Thanks.

I'm using that device now since early last year. I'm on my fourth chip 

It certainly eliminates ground loops but tends to fail after several high voltage hits below the published ratings.

Do you have USB isolation for your test equipment?

[Peterae]

Hi WW
I have never used USB isolation GULP i have been OK so far, but i guess most of what i have done is low voltage, low power.

I guess you could say that chip must be doing it's job, if the equipment down stream is OK afterwards.

[BEP]

I guess you could say that chip must be doing it's job, if the equipment down stream is OK afterwards.

That's the problem 

It took out an old laptop and two USB hubs. I have TVS's between lines and between lines and ground at key points. I was just after preventing ground loops as these can really screw up results.

Maybe I'll use WiFi isolation 

[BEP]

It appears WiFi will work.  ZigBee on a PIC with a USB host port is fine for lower data rates.(I'm collecting data on a circuit wrought with transients using the ADCs in the PIC).

Nothing special about the CUTs. Just a load of problems with ground loops affecting the test outcome.