[time-nuts] Thought experiment on a low cost timing board

Poul-Henning Kamp phk at phk.freebsd.dk
Sat Feb 25 15:04:40 EST 2006

In message <20060225194724.68A6DBDE0 at ip-64-139-1-69.sjc.megapath.net>, Hal Murray writes:

>> They have two 12 bit D/As which are summed through a 1:1000 network to
>> give around 20 bits of effective resolution at the cost of some
>> discontinuities.
>> Because the ratio is only a quarter of the full ratio of the
>> individual D/A's, they can always center the interresting range in the
>> lower D/A's range. 
>Neat.  Thanks.
>How well does that work for controlling something like an OCXO from a GPS?  
>I'd expect that the shifts due to temperature or supply voltage on the big 
>D/A would be fast enough to cause problems.

It obviously work for SRS in the PRS10 and since the EFC voltage changes on
a decent OCXO are very small, I'd expect it to work equally well for us.

But yes, you'd need to pay attention to filtering of the powersupply.
If you use ratiometric ("resistance output") DACs and the OCXO's
Vref, I don't think the tempco is going to be much of a problem.

But then again, I usually run into all sorts of problems once I try my
theories out in practice.

And that's at least half of the fun :-)

Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.

More information about the time-nuts mailing list