[time-nuts] USB over optical fiber

[jimlux]

On 12/6/19 9:10 AM, Dana Whitlow wrote:
> Even without any funny business with boost converters (or whatever). those
> copper wires
> make a nice antenna for coupling noise from both ends (PC and USB
> accessory) into
> the environment inside the chamber.  Unless, of course, the cable is
> well-shielded, which
> you did not mention.
> 
> I suppose it's too late to ask, but did you perhance try adding some
> clamp-on ferrite chokes
> to the cable?
> 

We had a similar problem doing some testing on hardware in a EMI/EMC 
chamber where we wanted to measure fields well below the MIL-STD-461 
RE102 limits at HF frequencies.

Get some 31 mix toroids in the 2.4" diameter, and you can get a lot of 
turns of the cable through it and the hole is big enough to clear most 
connectors.  There are also clamp-on forms.  31 mix has a nice wide 
range, although I don't know if it goes down to 100kHz.. there might be 
a better mix.
You definitely don't want the usual VHF suppression mixes like 43 or 61. 
  But maybe mix 78, which is optimized for <2MHz

In this kind of application you want to look at mu'' (the lossy part of 
permeability) as well as mu' (the non lossy part)

Or, as you plan to do, run stuff off batteries.