[time-nuts] OCXO Vref (in)stability plot

[David]

On Mon, 22 Apr 2013 16:13:17 +0000, "Poul-Henning Kamp"
<phk at phk.freebsd.dk> wrote:

>In message <trinity-7a7ee757-89f7-4b92-8884-9435668fef15-1366646349138 at 3capp-gm
>x-bs21>, "Jochen Frieling" writes:
>
>>your finding is quite interesting. I would like to replicate your
>>measurements on some other voltage sources and Vref sources.
>>Could you please elaborate on the setup, settings and acquisition
>>process?
>
>There really isn't anything to it:  Power the OCXO from a good
>clean supply (HP6626A in my case), measure the voltage on the
>VREF pin with the best voltmeter you have.
>
>In my case a HP3458A with "NDIG=8", "NPLC=20"

I have used the same technique to optimize 0.1 to 10 Hz noise in low
noise amplifiers, references, and power supplies.  Conveniently the
standard deviation is equal to the RMS so the calculations are easy
and I had good agreement with the datasheet values of various parts.