I agree with what you say Maurizio, but there are many other
design problems that would be curable for a few dollars at the
factory.
> The bandwith of pulsed signals exists but actually most of the problem lay
> in the phase noise of the PLLs used in the transmiters that spreads some
> energy in the spectrum around the nominal frequency.
There are two QSK related problems:
1.) Many rigs use the same VCO and move it to different
frequencies when working "split". That means if you run QSK while
working split the VCO "swoops" between the receive and transmit
frequencies. Manufacturers try to delay the transmit time slightly to
allow the VCO to settle on the TX frequency. Unfortunately it often
is not quite enough and a portion of the "swoop" gets transmitted
as a "click" that extends in one direction from the transmit
frequency.
(I bought a new IC-775DSP and it solved the "swoop" problem by
running multiple VCO's at the same time, and RF switching
between the VCO's on the transmit and receive frequencies.
Unfortunately it had so much leakage in the RF switch that it
received and transmitted on multiple frequencies at the same time.
ICOM told me the only cure was to install a modified board which
turned off the unused VCO, but that would result in loss of QSK.)
2.) Another problem is poor relay and T/R switching timing, which
causes waveshape distortion.
I'm trying to work duplex for receiving on 160, and having a real
hard time finding a transceiver that does not generate transmitting
trash.
> receivers have PLLs in the heterodyne oscillators and create a local
> problem that adds to the transmitted one. Actually, unless a signal
> strenght doesn't exceed a certain level, nothing is practically noticed,
> but when the received signal is huge it's well possible to receive the
> transmitted noise as a larger occupied band or to "feel" the own receiver
> limits in the matter of phase noise and not only about its IMD
distortion.
True.
Also most ARRL and manufacturer tests fail to give a true picture
of close spaced IMD problems because test signal spacing is
placed outside roofing filter passband.
I was amazed at how poor my FT1000D was when tested with 2
kHz spacing, but fortunately it was curable for only a dollar.
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/topband.html
Submissions: topband@contesting.com
Administrative requests: topband-REQUEST@contesting.com
Problems: owner-topband@contesting.com
|