[time-nuts] smallest rubidium
wb6bnq at cox.net
Mon May 23 21:17:57 UTC 2011
You should have said so in the first place. Unless you are sure that the reference
oscillator is the base for all the generated frequencies, it would not make sense to
install a Rb source in the radio. You are correct concerning most of the
"so-called" higher stability options in that they are barely worth the price asked
for them and only meet specs in a tightly controlled environment.
As for the size, you would be better served using a high quality miniature oven
controlled crystal oscillator (OCXO). A number of companies produce such items,
with some having extremely good specs. If you do go that route, select one that has
an electronic frequency control (EFC) port so you can have the option of external
control if you want it. You would have to design and build your own phase locking
circuit unless some company's oscillator offers that.
However, do not expect that having a PERFECT reference oscillator precisely on 10 or
20 MHz will put you perfectly on frequency as you will be disappointed. In a fully
DDS generated (all signals from a or multiple DDS's) radio scheme they only come
"CLOSE." You will never ever get "right" on because of the binary nature and
resolution of the DDS. Especially with the older radios as the DDS resolution was
lower than can be had these days.
So it is a question of "What is good enough" for the intended product. In the
Amateur radio world One hertz is considered the best basic step size. Many radios
only displayed and used 10 hertz step size which can be easily had with a 28 bit DDS
(a 0.1xx hertz error). Many of the most modern radios are only using a 32 bit DDS
which translates into a 0.01xx hertz error.
By the way, the error factor of the DDS is not consistent, again due to its binary
nature. If you utilize the Analog Devices web site ADIsimDDS app you will see that
the error offset is all over the map. The best that can be done is to make sure the
error factor is outside the measurement frame by decade or more.
Keep in mind that the clock frequency for the DDS determines the number of sampling
points for the constructed signal. So as you climb higher in the frequency in
relation to the clock of the DDS, the number of sampling points go down. Here is
the URL for the Analog Devices simulator :
"Dr. David Kirkby" wrote:
> I've bought an amateur radio transceiver - Kenwood TS-940S. This has a 20 MHz
> crystal osciallator (not 10 as I stated before), but it was optionally available
> with a 20 MHz TCXO, called SO-1 which sat on a small (how small?) circuit board.
> But these TCXO's seem to be like rocking horse dung, so I wondered about putting
> my own TCXO on a board. Then the idea of perhaps using a rubidium hit me.
> Some enterprising sole is selling on eBay a circuit board which replaces the
> SO-1, and has a 10 MHz reference input. But of course that means you need to
> rely on having the external reference. Having it built in would be nice, but I
> don't think there's enough room.
> Perhaps an OCXO might be practical - better than a TCXO, but not as big as a
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts