[time-nuts] Fast frequency counting question

Magnus Danielson magnus at rubidium.dyndns.org
Sun May 4 20:28:56 EDT 2008


From: "Tom Van Baak" <tvb at LeapSecond.com>
Subject: Re: [time-nuts] Fast frequency counting question
Date: Sun, 4 May 2008 17:18:03 -0700
Message-ID: <003401c8ae45$80679e70$0800a8c0 at pc52>

> > Hi all,
> > 
> > We have a problem at work where one of our engineers needs to measure
> > the start-up chirp in an oscillator destined for a GPS application.
> > 
> > We are looking for ppb level frequency resolution with sample periods
> > around 10 - 20us (yes, that's us, not ms!). The chirp is all over well
> > inside 1ms.
> 
> Would the end-user allow you to gate the oscillator output
> for the first 1 ms thereby avoiding the chirp you're trying to
> measure?

I considerd this... but I think it would be better if the end user built their
design to handle the initial 1 ms chirp. So re-designing the oscillator would
be the wrong way around it and it would not really solve other similar
problems they could be having from the same source, the power turnon.

They could probably get a much more reliable result if they just let things
settle a bit before they start depending on them.

I have already indicated that I consider it to be a red herring chase...
But the measurement needs to be done never the less...

Also, it needs to be done multiple times to feel the variance.

John is certainly on the right track here...

Cheers,
Magnus



More information about the time-nuts mailing list