[time-nuts] OCXO sensitive to gravity

Lux, Jim (337C) james.p.lux at jpl.nasa.gov
Thu Aug 13 23:22:45 UTC 2009


> -----Original Message-----
> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
> Behalf Of Lux, Jim (337C)
> Sent: Thursday, August 13, 2009 4:01 PM
> To: richard at karlquist.com; Discussion of precise time and frequency
> measurement
> Subject: Re: [time-nuts] OCXO sensitive to gravity
> 
> > -----Original Message-----
> > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]
> On
> > Behalf Of Rick Karlquist
> > Sent: Thursday, August 13, 2009 3:35 PM
> > To: Discussion of precise time and frequency measurement
> > Subject: Re: [time-nuts] OCXO sensitive to gravity
> >
> > Tom Duckworth wrote:
> > > The orientation change is due more to the earth's magnetic flux
> > effect on
> > > the oscillator, and less so from gravity.
> > >
> > > Tom
> > > Tom Duckworth
> > > tomduck at comcast.net
> >
> > Sorry, this is simply incorrect.  Magnetic flux from the
> > earth has no effect on quartz oscillators.  There is no
> > mechanism there.  Acceleration definitely affects quartz.
> >
> >
> 
> Hmm. We're investigating just this at work, in connection with things
> destined to orbit Jupiter, where we expect to see a periodically
> varying flux as the spacecraft spins.  The quartz is insensitive, but
> the mounting isn't necessarily insensitive to magnetic fields,
> especially if any magnetic materials (i.e. Kovar seems to be of
> interest) are used.
> 
> If you have any circuits sensitive to DC offsets, then a varying
> magnetic field can cause problems (e.g. the VCO input in a PLL).
> 
> I suppose it depends on "how good" the oscillator is expected to be...
> 

Found some typical specs for a USO (these are from a 2001 paper by Candelier, et al)
 5E-13/Gauss  (they got on the order of 1E-12/Gauss in their tests for the non-magnetic structure, but 10E-12 to 60E-12/gauss for the previous design)

Sami Asmar, here at JPL, published a paper back in 1997 summarizing the state of the art performance, and gave performances from 5E-13/G (for an oscillator from APL) to 5E-11/G (for a Rb source on Huygens)

http://tmo.jpl.nasa.gov/progress_report/42-129/129F.pdf
A newer paper by Sami and his colleagues, at
http://trs-new.jpl.nasa.gov/dspace/bitstream/2014/8057/1/03-3544.pdf
has essentially the same data.


http://adsabs.harvard.edu/abs/1997ESASP.395..299R is a ESA paper on state of the art in 1999.



Now... none of these effects would be of the magnitude reported by the original poster. Measuring a 1E-11 error in a 10MHz source isn't probably something you could "see" in a instantaneous sort of sense.

At JPL, *we* care about them because we use these oscillators in radio science measurements where you want an Allan deviation of <4E-16 in 1000 seconds to measure things like Jupiter's gravity to infer the internal structure or to compare radio transit time at 8.4 GHz and 32GHz to look for gravity waves, and in that context, a 1E-13 jiggle is pretty big.  Even for run of the mill spacecraft nav purposes, you care, but not quite to that level.



More information about the time-nuts mailing list