[time-nuts] Checking accuracy of Rubidium standards

Bruce Griffiths bruce.griffiths at xtra.co.nz
Sat Nov 8 22:18:51 UTC 2008


Neville Michie wrote:
> Hi,
> I have a plan which involves the dividing down of the 10MHz of a GPSDO
> and a rubidium (LPRO) to about 1MHz or 100kHZ and applying them to a
> XOR or D latch to get a PWM signal that can be averaged for a strip  
> chart recorder or
> 12 bit analogue data logger. The DC output gives a range of 5 volts  
> for one
> microsecond or 10 microseconds phase difference and folds back if  
> this difference is exceeded.
> The data from the datalogger is in a format that a spreadsheet can use.
> With time and phase measurements I wonder how hard it is to get Allen  
> variance.
> I realise the PWM method requires a low pass filter and this will  
> prevent short period
> variances from being calculated.
> cheers, Neville Michie
>
>   
Neville

A low pass filter or equivalent means of limiting/defining the
measurement bandwidth is required when measuring ADEV.

The limitation with all such digital phase detectors is the nonlinearity
as the ends of the range are approached.
It is possible to achieve a resolution of around 4E-11/Tau with such
phase detectors if a higher resolution ratiometric ADC is used.


Bruce





More information about the Time-nuts_lists.febo.com mailing list