[time-nuts] Bodnar "Precision Frequency Reference (GPS Clock)" AND LeoNTP Networked Time NTP Server Questions

Magnus Danielson magnus at rubidium.dyndns.org
Fri May 18 08:23:21 EDT 2018


Hi,

On 05/18/2018 02:16 PM, Dana Whitlow wrote:
> Caution, folks, about USB cable radiation.  While the intended signals
> flowing through the
> cable presumably contribute a bit to the overall picture, common-mode
> currents on the cable
> are the most likely cause of severe radiation problems.  These currents
> arise not merely from
> intended USB signals, but from *all* the digital activity within the
> device, and will be present
> whether or not USB communications are going on.
> 
> The keys to controlling this problem are:
> 
> 1) Shielding on the USB cable.
> 2) Proper design inside the USB device, rarely done beyond the point of
> (barely) meeting
>      government regulations, which are far too forgiving IMHO.
> 3) Use of a ferrite choke on the cable to reduce those residual currents
> that get by anyway.
> 
> Note that only item 3 might be under the control of the user to any useful
> extent.
> 
> So please don't just assume that using a USB cable for charging only solves
> anything.
> 
> Note that the above comments are also generally applicable to *all *cables,
> including power
> cables, that are plugged into *any* digital device.

I could not agree more.

Shielded cables helps a lot. We used to have issues at the radio club
with Ethernet cables, but as we shifted to shielded cables with reduced
the noise level considerably.

Ferrite helps to suppress common mode currents, but the milage may vary
depending ont he ferrite material.

There is USB isolators, which breaks "ground loops" (sigh), and that do
helps on the low frequency side of things, but not necessarilly at HF,
where the isolation is just a capacitive gap which couples fairly well.

One thing that does help is not to have unnecessary rise/fall times of
signals.

Cheers,
Magnus



More information about the time-nuts mailing list