[time-nuts] smallest rubidium

[Jim Lux]

On 5/23/11 6:07 AM, Dr. David Kirkby wrote:
> On 05/23/11 09:31 AM, Rob Kimberley wrote:
>> Try
>> http://www.symmetricom.com/media/files/downloads/product-datasheets/DS_SA.22
>>
>> c.pdf
>>
>> Rob Kimberley
>
> Thank you. I don't have know what board area I have available, but I
> think think its going to be considerably less than that size. I think
> the only option will be to use an TCXO, which to be honest is good
> enough, but if there were small rubidiums, I would consider using one.
> But it seems they are not as small as I would have liked.
>
>

Depending on your actual needs/other stuff available in your device, a 
non-TCXO and a temperature sensor might do as well. The non-TC means 
that the crystal can be higher Q, so better close in phase noise, the 
temperature sensor means you can calibrate the variation with 
temperature (assuming your downstream application can work with 
"knowledge" of the frequency.. if you're trying to generate 10 MHz, then 
that doesn't help as much)