Isn't that just the Allan Deviation data? Symmetricom has datasheets on their website for their various modules. They have a GPS discplined quartz oscillator in several flavors.
http://www.symmetricom.com/media/files/downloads/product-datasheets/ds%20XLi%20Options%202.pdf

Something else to consider is doing post processing.. Use a nice quiet 10MHz oscillator for your source/sampling clock, and record the 1PPS from the GPS receiver as well as your unknown, then figure out after the fact what the oscillator was doing.

Jim

Matt Ettus wrote:

Matt

If the application is somewhat analogous to VLBI, then the maximum
(uncorrectable ie random) allowable carrier frequency phase errors
between receivers depends on the integration time.
Maximum integration times for VLBI are typically 10,000 sec or less,
with corresponding relative ADEV (after correction for drift etc) on the
order of 1E-14 (or better) @ 10,000 sec for a carrier frequency of a few
GHz.

What is the integration time in your application?

What is the carrier frequency?

For short integration times most rubidium standards are much noisier
than a good OCXO.
GPS carrier phase techniques will allow lower noise comparisons of LO
phase differences than a GPS derived PPS based system.

Bruce

On Mon, Jan 5, 2009 at 12:13 PM, Lux, James P james.p.lux@jpl.nasa.gov wrote:

I don't think Allan Deviation is the right measure.  First, standard
Allan dev numbers won't take environmental differences into account.
Also, isn't Allan Dev measured vs. a better reference?

Unfortunately, post processing isn't possible in this app, since it is
a real-time communications application.

Thanks,
Matt

On Mon, Jan 5, 2009 at 1:54 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:

We are trying to receive data which is modulated at rates which vary,
but are in the 100 Hz to 100 kHz range, so the integration times are
very short, all below 10ms.

30 MHz and below for now, but we'd like to be able to go as high as
500 MHz or so.

Don't the Rb standards end up disciplining an OCXO so you get the
benefit of the crystal at the shortest integration times?

Thanks,
Matt

Environmental differences, as in the environment of the box? Or the propagation path?
For the box, the short term is going to be dominated by the OCXO, isn't it? So it's relatively environment immune.

Allan Dev is the box against itself,in the adjacent time slice, but wouldn't that be pretty similar to a box against an identical box, because the noise is assumed to be uncorrelated from time slice to time slice.

OH.. This is like doing a distributed phased array for EME (or, as N5BF wants to do, EVE)

On the basis of your other mail a few minutes later, you're looking at very short integration time (e.g. 10ms), so wouldn't phase noise be your more appropriate thing to look at, and for that, you're almost certainly looking at the basic properties of the quartz oscillator.

OTOH, if you're trying to a phased uplink sort of thing, then you are concerned about longer time intervals (e.g. you want the relative phases of Tx1 and Tx2 to be constant over intervals of seconds)

Matt

Matt Ettus wrote:

What's your tolerance for phase variations between a pair of LOs?
Its on the order of 50 degrees or so for VLBI.
At 500MHz this corresponds to about 140ps.
The corresponding relative ADEV is around 1E-8 @ tau = 10ms.
Almost any good crystal oscillator can achieve this.

Yes, but compare their resultant short term ADEV with that of a good OCXO.
They can be 1-2 orders of magnitude worse than the better OCXOs.
Most of them use a relatively short time constant FLL to steer the the
crystal oscillator to the passive Rubidium cell.
A few (typically older more expensive ones) can be relatively quiet.
An external low noise OCXO phase locked to the rubidium standard with a
relatively long time constant can be used achieve lower ADEV at sort
averaging times if the operating environment allows this.

Whether any of this is necessary depends on your applications tolerance
to short term phase variations between LOs.

Bruce

Matt Ettus wrote:

Not so, ADEV actually includes environmental effects during the measurement.
However, data sheet ADEV specs may have been obtained in a more closely
controlled environment that your intended use.
ADEV is measured between oscillators one of which may be better than the
other.
The corresponding ADEV of each oscillator may then be derived from the
relative ADEV data if some assumptions are made.
One can use the datasheet ADEV to estimate worst case ADEV for pairs of
such oscillators.
To gauge the relative ADEV between oscillator pairs in your environment
you will need to make some measurements as the various unspecified
thermal time constants of the OCXO will be significant.

Bruce

On Mon, Jan 5, 2009 at 2:56 PM, Lux, James P james.p.lux@jpl.nasa.gov wrote:

Yes, a very similar application.

Well, basically, I need to be able to coherently add signals from
multiple locations without first looking at those signals to determine
what the phase error is.

Yes, we would need the relative phase to be constant over time scales
up through about a second or two.

Matt

On Mon, Jan 5, 2009 at 3:02 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:

As you said, our ADEV requirements are pretty loose and could be met
by any decent crystal.

But Said just told us that he sees about 25ns difference between
units, which is about 180 times worse than 140 ps.

In VLBI I think that they can deal with an unknown phase error between
the two LOs as long as that phase error remains relatively constant
over the period of integration.  That is why they care about ADEV.
They just correlate the two signals, and subtract out that constant
phase error.  My problem is that we can't just do that in this
application.  We need to know that phase error a priori.

Matt