[time-nuts] ok, newbie questions

[bg at lysator.liu.se]

> Yes I agree a newer thunderbolt would surely suffice for me and probably
> also the ocxo in my 8662A synth
>
> But I am still academically curious about the impact of more channels of
> satellites? What is the value of these extra sats?
> Thanks!

In a mobile scenario, you need measurements to four or more satellites to
solve for x,y,z and time. A ground vehicle travelling in areas with lot of
antenna masking - high buildings close to the road or trees - blocking
line of sight from antenna to the satellites, will want to track all
satellites in view. And since especially low elevation satellites will
come and go, you want to pick these up really quick once visable.

Thus you want as many channels (correlators) as possible.

For a stationary timing scenario, (x,y,z) might be known - or averaged
over many measurements (site survey). This gives that in a timing mode the
receiver only needs to track one satellite.

The GPS receiver evolution moved from 1-channel multiplexing, to 6
channels, and 8, 12 and even more. While there might be timing variants of
generic navigation receiver, the core correlator chipsets are shared.

So even if the timing receiver does not benefit that much from 8+
channels, once the navigation receivers got more channels, the timing
versions had to follow.

Modern timing receiver have better hardware to put the 1PPS signal where
intended. Se http://www.leapsecond.com/pages/vp/sawtooth.htm for a
comparison of semi modern receiver and an old one. This might have a
bigger impact than the number of channels.

--

    Björn