[time-nuts] Allan Variance versus Allan Deviation

William H. Fite omniryx at gmail.com
Mon Mar 28 19:33:52 UTC 2011


Magnus is quite right.  Variance is almost always just a value on its way to
becoming another value.  Squared units make little sense in most
applications.

Bill (the statistician)

On Mon, Mar 28, 2011 at 1:10 PM, Magnus Danielson <
magnus at rubidium.dyndns.org> wrote:

> On 03/28/2011 02:36 PM, Bruce Griffiths wrote:
>
>> Martyn Smith wrote:
>>
>>> Hello,
>>>
>>> I'm trying to compare the stability specifications of two oscillators.
>>>
>>> The first spec is 2 x 10E-11 Allan Variance.
>>> The second spec is 3 x 10E-11 Allan Deviation.
>>>
>>> As Allan deviation (ADEV), is the square root of Allan variance, I
>>> thought the first spec is the better of the two, but a collegue
>>> disagrees, saying 3x10E-11 Allan Deviation is a better spec than 2 x
>>> 10E-11 Allan Variance.
>>>
>>> I'm sure the million experts at there will give me the correct answer
>>>
>>> Regards
>>>
>>> Steve Jones
>>>
>>>
>>> _______________________________________________
>>> 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.
>>>
>>> Strictly you are correct, however some early datasheets (eg Crystek, )
>> say Allan variance whilst they actually tabulated its square root or
>> Allan deviation.
>>
>
> It's also common to have plots saying "Allan Variance" and plotting the
> Allan Deviation.
>
> Since Allan Deviation is the square root of Allan Variance, 2E-11 Allan
> Variance is about 4.5E-6 in Allan Deviation so an Allan Deviation of 3E-11
> is better than an Allan Variance of 2E-11... by far.
>
> Use Allan Deviation through-out.
>
> Cheers,
> Magnus
>
>
> _______________________________________________
> 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.
>


More information about the time-nuts mailing list