[time-nuts] Fast frequency counting question

Bruce Griffiths bruce.griffiths at xtra.co.nz
Sun May 4 23:00:37 EDT 2008

Luis Cupido wrote:
> Hi Bruce,
> I don't know what you mean by low resolution
> but I can easily think of more than 12 bit at
> higher than 150Ms/s.
An  ENOB of 12 is probably inadequate for useful single shot 
measurements of a 40MHz sinewave .
>  > will tend to limit the effective resolution in determining the zero
>  > crossing locations.
> by no means zero crossing is to be used. we do have
> much more information than that... phase info is all over
> in all sampled points not just on zero crossings.
> This is why I've said some transform to produce a spectrogram.
The application is probably more sensitive to zero crossing jitter than 
anything else.
If one directly samples the oscillator output directly the waveform may 
not be even approximately sinusoidal.
With a logic level output oscillator a much higher sampling rate than 
100Msps may be required to adequately sample the transitions adequately 
unless a bandpass filter is used.
In this case (nanosecond rise and fall times) a lower resolution ADC 
with a multi-gigahertz sampling rate may be useful.
Really need more information on oscillator waveform etc.
> Maybe this is the reason why you mention the need of so many
> integrations and the "unpractical" comment, as you are thinking
> of using only a tiny fraction of the information available.
> ...
> interesting topic :-)
> Luis Cupido.
> p.s.
> however even tens of thousands of measurements
> of 1ms windows is a few minutes !!!
No, I intended that all samples be used, even sampling a sinewave at 
100MHz or so wont achieve picosecond noise and resolution in determining 
the zero crossing times unless all samples are used.

However, you wont be able to restart the oscillator every millisecond 
without affecting the transient response.
Depending on the data collection system, it may take much longer than a 
millisec to process or transfer the data from each transient.
An expensive system that can make such measurements quickly (100/sec or 
so) probably isnt justified until the problem can be shown to be real 
and not just a customer measurement system artifact.

Obtaining the averaged oscillator startup transient response is perhaps 
not useful in that the GPS receiver may not be particularly sensitive to 
A sampling jitter of much less than 1ps or so is probably not warranted 
given the likely oscillator phase noise characteristics.

Sampling the low pass /bandpass filtered output of a mixer will convolve 
the oscillator transient response with that of the filter, since direct 
sampling is likely to achieve all the resolution required if a 
sufficiently low jitter sampling clock and ADC are used, using a mixer 
probably adds unnecessary complexity and empty resolution.


More information about the time-nuts mailing list