[time-nuts] OT: xtal osc PN

Mike Feher mfeher at eozinc.com
Sun Sep 19 19:37:11 UTC 2010

Frank -

Great idea, so obvious I did not think of it. If you mix the 20 and 22 you
will only get 3 dB degradation or still very close to the -131 dBc/Hz
relative to the 10811A. As I mentioned before the architecture is relevant.
I have found that mixing does not cause any noticeable degradation, and I
used to go all the way up to 45 GHz on military programs where it was very
critical. At the frequencies you are talking about I doubt if the amplifiers
will have any appreciable degradation either. Of course you have to keep
levels in perspective, as you will not do better than kT. I also do not
believe that dividers will have much impact. After all, a DDS is a
divider/counter and accumulator, and PN is usually considered to be very
close to  20logN better at the output than the reference, however, DDS does
have spurious at most frequencies, but that is a discussion for another
time. I still think your original thought is your best approach. Fast, easy
and less than $100, even if you do use a used 10811A. 73 - Mike 

Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731


-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of francesco messineo
Sent: Sunday, September 19, 2010 3:00 PM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] OT: xtal osc PN

Hi Mike,

On 9/19/10, Mike Feher <mfeher at eozinc.com> wrote:
> Well, if one just looks at the spec of the 10811A for relative
> it is -140 dBc/Hz at 100 Hz offset at 10 MHz. Realistically, probably a
> little better. From that it would be real easy to generate the frequencies
> Frank is looking for, obviously 20 would be easy but would be only -134
> dBc/Hz at 100 Hz away. 22 would be easy by diving the 10 and mixing it
> the 20, assuming the divide by 5 has a very small contribution, the PN of
> the resultant at 2 MHz is also theoretically 20logN better, so, mixing
> also give close to -134 dBc at 100 Hz away. 42 can easily be generated by
> doubling the 20, to get -128 dBc at 100 Hz and then mixing with the same 2
> MHz to get the 42 MHz, still resulting in almost -128 dBc/Hz at 42 MHz.
> Obviously filters will have to be used to get rid of the unwanted lower
> mixing products. Depending on the architecture used, as stated below,
> further multiplications will again decrease these numbers by 20logN.
> Filtering at the IF with a 250 to 500 Hz filter is not going to do
> to the 100 Hz numbers. I assume the mode of communications here is CW,
> the narrow filter at IF. 73 - Mike

this exact approach (well maybe mixing 20 and 22 to obtain 42 MHz
instead of doubling 20 MHz then mixing) was my first choice but I'm
afraid I wouldn't be able to do all these steps without making any
mistake like chosing the right mixer, right filters and
It was really what I wanted to do, but looks like I could spend too
much time doing it and ending with a poorer result than carefully
chosen xtal oscillators.

Thanks for suggesting it though :-)

Frank IZ8DWF

time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to
and follow the instructions there.

More information about the time-nuts mailing list