[time-nuts] Hadamard variance
bruce.griffiths at xtra.co.nz
Thu Apr 9 13:51:41 UTC 2009
Just dont get too carried away.
Remember that the Hadamard deviation is only insensitive to linear
If the drift is quadratic for example then it will affect the Hadamard
Steve Rooke wrote:
> Hi Tom,
> 2009/4/9 Tom Van Baak <tvb at leapsecond.com>:
>> I use it sometimes when I need to. But note that in most cases
>> you do NOT want to ignore drift. If you measure an OCXO for
>> the purpose of using it in a clock or appliance or radio or test
>> equipment you really do want to know if it has drift or not. ADEV
>> will show this, while HDEV will not. So you have to be careful
>> about using statistics that deliberately and quietly ignore effects
>> that may be important to your application.
> Indeed, I was thinking that HDEV would be a good tool to characterise
> free running OCXOs with it's insensitivity to drift but, of course, I
> would use ADEV to measure the performance of a GPS locked system or
> one running in holdover mode.
>> But before you run off and use HDEV for everything note that
>> the other practice that is far more common -- simply remove
>> frequency drift from the raw data before computing an ADEV
>> on the residuals. If you look at plots in professional journals you
>> will often find comments to the effect that phase, frequency, or
>> drift offsets have been added or removed prior to making said
>> phase, frequency, or stability plots.
> I had no intent to use HDEV exclusively, it seems like a useful tool
> to analyse free-running oscillators to measure the affects of noise
> while screening out drift (which we have some means of handling in
> holdover circuits). As a selection tool it seemed quite useful and I
> was asking if others felt the same way.
> Agreed, it is possible to factor out drift by pre-processing the data
> and then using just ADEV to compare all aspects of any open or closed
>> Here, to see the difference that HDEV makes (or not) see:
> It seems to have an effect removing some of what must be drift with
> the OCXO plot but adds nothing to the PPS one. Do I take it that the
> OCXO was free-running and the PPS was locked to GPS, as this would
> account for the differences?
>> The command line program that I use (ADEV3) these days:
>> Tool for ADEV, MDEV, HDEV:
>> Source code (compiles in windows, bsd, or linux)
> Thanks for the pointers, I'll have a look at this instead.
More information about the time-nuts