I found this to be true a few years ago when searching for a
digitally controlled low noise VFO for a particular project. As I
recall spurs in the DDS system were around 80db down or so out
a few hundred kilohertz. Hardly a "low noise" VFO. They be
better today. About the same time I owned a Japanese
transceiver with DDS LO system. Although the close in phase
noise was pretty low, that radio was full of spurs and phantom
signals. I don't know if that was the result of the DDS or not.
As a solution to my problem I settled on a FLL [frequency lock
loop] which, when used with a good voltage tuned oscillator,
makes a very low noise VFO with digital control capability [A/B
vfo's, memory, etc.] and almost zero drift. See patent no.
6,175,280 for a full description if you are interested.
Testing it as a remote VFO for my Corsair II, I can find no
difference in phase noise performance between the Corsair VFO
and the FLL, and the FLL system is completely free of spurs.
It does have two drawbacks relative to PLL or DDS systems.
Close in phase noise is not reduced by action of the loop, so one
must start with a good quality VCO. And frequency switching
speed is somewhat slower than PLL/DDS systems. So if you are
switching from one end of the band to the other, it may take a
half second or so to get there. That second drawback will keep
it from being used in many applications.
In the Corsair application, I run the FLL at 40 to 44 Mhz and
divided by 8 to get 5 to 5.5 Mhz for injection into the Corsair.
That division results in nearly 18 db of phase noise reduction
resulting in a very quiet digital LO.
Which brings me to the Ten Tec scheme of generating the LO
signal at high frequency and dividing down to reduce phase noise
[and step size] at the same time. [They did not do this in the
Paragon by the way] Noise reduction is 20 log (divide ratio).
Generate high......divide down is simple and turns in good numbers
when quiet dividers with low jitter are used.
I wonder if they used that scheme in the Orion main receiver
LO? According to the published specs for the Orion, the LO
phase noise on the main receiver is typically -136 dBc/HZ from
0.5 to 20 kHz. To put that into perspective, a quick look at the
pretty quiet Icom 756 PRO II phase noise plot by the ARRL,
shows around -125 dBc/HZ at 2 kHz from the carrier and rising
as you get closer. The Orion should sound better than any of the
synthesized rigs out there and maybe even better than a
Corsair! :-) Did they generate high and divide low? Maybe a
DDS/PLL system? FLL? No, I think no FLL. :-) It will be
interesting to see how they get that performance.
Can hardly wait 'till the Orion hits the street.
-73-
-Lee-
On 27 Aug 2002 at 2:09, Dr. Gerald N. Johnson, electr wrote:
> DDS can be good close in and when not stretched too far, but DDS is
> not at all clean, there are so many spurs it usually takes a PLL (with
> the attendant noisy VCO) to act as a spur filter.
|