[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [linrad] CS5396...a board or two that use it
> how real is really the need in EME for such huge dinamic range?
> Except if you have a really big local QRM problem, I guess the
> dynamic of EME signals should easily be covered by 8bit
> sampling, provided you set the levels carefully?
NO NO!!! EME stations form activity centers and it is
common to have a few more EME-ers within 50 km.
Then 8bit will not be enough. 16 bit is marginal. My old
system with 16 bits was saturated the last hour in
contests because then SM5DCX would point his main lobe
into my back lobe from a distance of about 30km.
In free space I would need -160dBc/Hz to handle
SM5DCX assuming F/B=15dB.
With the actual flat land we have between us I think
-130 to -140dBc/Hz would be adequate. That is what
we can get in spectral purity from good standard
With the Delta44 and the RX2500 the noise floor at 2.5MHz
is in the order of -147dBc/Hz.
Going directly from 144 to 2.5MHz leads to several problems
and I have decided to do it by use of several conversions.
Each frequency conversion will add a little noise so at the
144MHz input the noise floor would be at say -144dBc/Hz.
The associated noise figure is about 15dB and it will be
necessary to add a mast mounted preamp and a second
RF amplifier in the shack. If you want to have a system
noise figure around 0.25dB the noise floor will degrade
to about -124dBc/Hz.
I can live with that but it will cause some problems
now and then. With a 15dB improvement using CS5396
the limiting factor would be the purity of the transmitted
spectrum of my fellow EME-ers (until they get a Linrad
In terrestrial weak signal communication it is easier.
Placing the offending signals outside the visible
passband improves by 15 or 20 dB and another 10dB can
be obtained by accepting 1dB loss of system NF.
On the other hand there are many more strong stations
in terrestrial activity and although the ERP power is
typically 10 to 20 dB lower I may have their main lobes
pointing right into my own main lobe.
Leif / SM5BSZ