[time-nuts] Best way for generating 8994.03 MHz from 2899.00042272.....MHz?

[Lux, Jim (337C)]

These days, you might want to consider using the GaAs VCOs from Hittite,
rather than the DRO. DROs are SUCH a pain to build and tune, being a
mechanical resonator in a cavity.  Everything you do seems to adversely
affect the DRO.  The MMIC VCO is just a die (or a die in a package) and it's
pretty much immune to external effects, since the resonator is built into
the oscillator. 

We built some prototypes at JPL using the VCO and a GaAs divider, and the
performance was better than DROs.

If you were building a very narrow band PLL, where tunability of the DRO
isn't needed over a wide range, the DRO might be a good solution, but still,
you have microphonics, etc. (we used to demonstrate the latter by hooking up
the output to a spectrum analyzer that has a FM demodulator, and talking to
the DRO)


On 8/14/09 5:53 PM, "Rick Karlquist" <richard at karlquist.com> wrote:

> David Bengtson wrote:
>> so there is one frequency that is X*(732/757), and he want to get X*3
>> from this? Seems like an integer-N PLL could do this pretty
>> straightforwardly, although I'd have to spend some time to figure out
>> the exact multiples. National Semiconductor has an app note on
>> frequency planning for synthsizers that would cover this.
>> 
>> Dave
> 
> Yes.  The ideal architecture would be to use a dielectric resonator
> oscillator at 8994 MHz as the output source.  Then divide its output by
> two.  Take that ~4.5 GHz signal and divide it by 757 using the divide by N
> section of an Analog Devices ADF4106.  Divide the 2998 MHz by 486
> using the divide by N section of a second ADF4106.  486 is 2/3 of
> 732.
> 
> Feed the "muxout" of the second ADF4108 into the "ref in" of the first
> ADF4108 and use the phase detector in the first ADF4108 to tune the DRO.
> (This will become clear after you read the ADF4108 data sheet :-)  The
> phase detector frequency will be about 5 MHz.
> 
> Rick Karlquist N6RK
> 
> 
> 
> _______________________________________________
> 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.
>