[time-nuts] TBolt signal levels

Charles Steinmetz csteinmetz at yandex.com
Thu Jun 19 15:36:29 EDT 2014


Tom wrote:

>I'm testing several Trimble Thunderbolts here with several different 
>antenna setups: indoor/outdoor, low-gain/high-gain, patch/helix, 
>plain/choke-ring.
>
>Has anyone done a comparative analysis of antenna make/model/type 
>against signal levels for this receiver?
>
>Or is there a rule of thumb for what is consider "good" vs. "bad" reception?

Be careful when you say "signal levels" in connection with a 
Thunderbolt.  The figure reported by the Thunderbolt is NOT the RF 
signal level, it is the carrier to noise ratio.  The c/n is mostly a 
function of the antenna location, and only a weak function of the 
antenna and LNA.  So mostly you switch antennas and things don't seem 
to change much unless the new antenna is in a different 
location.  Unless you are looking at the antenna output with a 
spectrum analyzer or frequency-selective voltmeter/power meter, you 
will not know the relative signal levels from the different 
antennas.  But c/n is what is important, so that's all good.

Lots of Lady Heather screen shots have been posted on the list 
showing accumulated signal strengths (commands SAS or SAD).  To be 
"good," you want to have large swaths of the plot in cool colors 
(blues and greens) (c/n over 40 dBc), and preferably mostly greens 
(c/n over 45 dBc).  Look at Warren's screen shots -- his antenna 
location is pretty good.  Contrast with Dave's screen shots -- his 
antenna location is adequate but not really very good.

Finally, note that not all similar c/n plots represent equally good 
reception.  Multipath can look like "carrier" to the Thunderbolt 
rather than "noise," so an antenna location with substantial 
multipath can look just as "good" as a location that is free from 
multipath.  Depending on the direction from which the reflections 
arrive, an antenna with good multipath rejection (e.g., a choke-ring 
design) may give superior results without looking any better on the c/n plot.

Best regards,

Charles





More information about the time-nuts mailing list