From G3PLX:
Dave asks how close do you have to be to see anything that isn't exactly on
frequency. The answer is 'it depends'.
The integrating vectorscope idea will produce a straight line if the signal
is exactly on frequency. If it's not exact, the line curves round into a
circle which will eventually get back to the centre of the scope in N
seconds, where 1/N Hz is the frequency error. That won't be a problem if the
signal is strong and clean - you just see a circle instead of a straight
line and you can still say 'I have received a signal'.
But the further off you are, the smaller is the resulting circle, and if the
noise level is bigger than the circle you won't see it. The closer you can
get to the right frequency the better is the sensitivity. For example if
you are 0.01Hz off, the dot will take 50 secs to trace a semicircle and get
as far out as it goes, then take another 50 secs to get back to the centre.
In this case there isn't any point in waiting more than 50 secs - if you
can't see anything in the noise after 50 secs, then you will never see it.
If you are coherent with the transmitter and the path is phase-stable (and
LF paths are), there is no limit to sensitivity, instead the limit is how
long you are prepared to wait before you see that the dot has moved out of
the noise. Scott and I can hold surface-wave LORAN lines steady in phase
within a few degrees for days on end, which means we could integrate for
weeks. Murray has no local LORAN lines but has transmitted his own coherent
carriers to ZL2AFP.
I don't see this technique being used for receiving transmitters that are
'fairly stable'. We can do that already with 'fairly stable' receivers. If
we bite the bullet and all get a GPS, then we can really start to see the
possibilities.
When we get to understand the implications of this new idea, we can start to
think about how to use it to better transmit information, but for the moment
it's just a toy so we can get the feel of it. The integrating scope is
certainly not the only way to handle coherent weak signals.
73
Peter
|
