[time-nuts] Least costly 10 MHz reference solution

[Magnus Danielson]

Russ,

Welcome!

On 01/23/2013 05:48 PM, Russ Ramirez wrote:
> Greetings,
>
> I have been reading what I can find on Rubidium and GPSDO approaches, but
> there are some fine points that do not make it clear which is the best
> 'bang for the buck' solution. My requirement/desire is to have a 10 MHz
> standard for my lab that I can trust to an accuracy of 7 decimal places (10
> ppb?), so anything that is good to a few ppb is certainly adequate for what
> I am looking for. I have a OCXO unit that is voltage adjustable - for
> example, adjusting this to 10.0000000 MHz per my HP 5334A requires -12.71V.
>
> So the simple (maybe) question is, should I go for a Rubidium disciplined
> unit, or go with a home-brew GPSDO solution using the Vectron OCXO I
> already have? My main cause of confusion is ignorance concerning all the
> GPS solutions out there with 1pps outputs, to use in a GPSDO, and which
> ones jitter too much to be useful (solutions under $50 exist).
>
> Thanks in advance.

A rubidium or GPSDO such as Thunderbolt can be found fairly cheaply.
If you go for a Thunderbolt, get one with antenna as a kit, mostly 
because it is a handy way to get started. For better stability you can 
get a better antenna later, if the need would occur.

The rubidium should give you the precision you need straight out of the 
box, unless it has "issues". In order to control if it has issues, 
having the ability to at least compare to GPS becomes obvious, so you 
end up wanting that GPSDO anyway. You can get both for reachable money 
anyway, if you look around long enough.

Doing a home-cooked GPSDO is fun naturally, and there is an art in 
low-budget designs giving fair amount of performance.

Cheers,
Magnus