[time-nuts] Terminators on empty outputs/connectors
magnus at rubidium.dyndns.org
Sat May 26 05:55:03 EDT 2018
On 05/24/2018 10:41 AM, Clay Autery wrote:
> Putting in a hardware order...
> Question: Is it recommended to put terminators on all unused GPS Distro
> amp ports? Output ports on GPSDO and NTP servers (10MHz and PPS).
All GPS splittes I have looked at so far uses the Wilkinson
power-splitter. It provides some isolations between the output ports, or
rather, the ideal scenario has perfect isolation, the real world not so
much naturally. If you consider a two-port splitter where you leave one
end open, and ignoring the fact that the splitter math actually wants a
50 ohm load to perform correctly, there will be some leakage over to the
next port and full leakage back to the source. Usually we can assume
that the source direction wave gets dampen in the amplifier sitting
there to compensate gain or for a passive one dampens in the cable back
and forth as well as not getting a perfect reflection in the antenna
source impedance. The direct leakage to the next port is likely to
dominate. Regardless, this will behave as multipath for the GPS signal
and just contribute to the multipath environment that the antenna
already have. It's only when you have a really crap splitter that you
need to be careful. For carrier phase measurements, you need to be more
careful about signals, but mostly for stability of signals.
So, for most cases, I don't bother on my splitters. I would only care if
I would really try to get stable and accurate measures, but not because
it would have much of an impact, but just because I don't want a simple
to fix thing do me any harm.
More information about the time-nuts