A input 10Hz..125MHz min level 25mVrms (-19dBm)
B input min 50MHz, -20dBm
the problem is on A input only? Maybe 150MHz for the A input, being beyond
the official specifications, requires more signal on one counter than the
other. In the past they all were fine?
I have read your paper on the correlation between the real and predicted
aging of crystal oscillators. I'm trying to find a suitable model for the
Morion MV201 OCXO or the Oscilloquartz 8663 to implement a Kalman filter.

On Fri, Aug 24, 2012 at 3:34 PM, Bernd Neubig BNeubig@t-online.de wrote:

Hi Azelio,

Sorry, the counters are 53131A, not 53151A. The A and B inputs are specified
from DC to 225 MHz with a sensitivity of y20 mV (rms) up to 100 MHz, <30 mV
(rms) 100 ~ 200 MHz, and <50 mV (rms) 200 ~ 225 MHz.
The problem can be at either input alone. The problem is only in the range
between 100 MHz and <200 MHz, observed in automatic trigger mode. In manual
trigger mode the counter does not trigger at all in that freq range.

I cannot tell you an aging model for a particular OCXO, and I doubt if
anyone can ;) Aging predictions can only be made on individual OCXO, unit by
unit, and the prediction is only valid over a rather limited time. Therefore
you must continuously monitor and update your prediction.

Best regards

Bernd
DK1AG

-----Ursprüngliche Nachricht-----
Von: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] Im
Auftrag von Azelio Boriani
Gesendet: Freitag, 24. August 2012 17:28
An: Discussion of precise time and frequency measurement
Betreff: Re: [time-nuts] HP/Agilet counter 53151A trigger problems

A input 10Hz..125MHz min level 25mVrms (-19dBm) B input min 50MHz, -20dBm
the problem is on A input only? Maybe 150MHz for the A input, being beyond
the official specifications, requires more signal on one counter than the
other. In the past they all were fine?
I have read your paper on the correlation between the real and predicted
aging of crystal oscillators. I'm trying to find a suitable model for the
Morion MV201 OCXO or the Oscilloquartz 8663 to implement a Kalman filter.

On Fri, Aug 24, 2012 at 3:34 PM, Bernd Neubig BNeubig@t-online.de wrote:

lab.

time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

30mV are -13dBm so maybe it is why the -20dBm is not a suitable level for
that input. I have the 53132A and the 53181A so I can test (with the R&S
SML01). Is this problem new? Never tested before?

About the OCXO aging: OK, I'm interested in the development of a model that
can be general enough so that it can be "steered", adjusting its
parameters, during the OCXO life. The data (to adjust the parameters) will
be gathered by the continuous monitoring with the GPS.

On Mon, Aug 27, 2012 at 7:01 PM, Bernd Neubig bneubig@t-online.de wrote:

Isn't crystal aging marked by random jumps? perhaps lattice
rearrangement? Really hard if not impossible to model...
Don

Azelio Boriani

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
R. Bacon
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

So it can't be done... the possibility to discipline an OCXO taking the
action from a suitable model should help in speeding up the adjustments and
avoiding the humps as seen in the Allan deviation plot.

On Mon, Aug 27, 2012 at 9:31 PM, Don Latham djl@montana.com wrote:

There are ways of generating models, such as ARIMA, using past behavior
and the models used to tweak filters and so on. The shifts would have to
be accounted for as they occur, seems to me. Now the interesting
possibility is that a shift does not alter the underlying model for a
given crystal????
Has to be tested.
Don

Azelio Boriani

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
R. Bacon
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

Bernd,
Do the 53131A have rear inputs?  At work, we're on contract to
repair 53132As and found the rear input seriously degrades the
performance.  The basic sensitivity (100-200 MHz) is 30 mVrms.  The Rear
input is in parallel with the front panel and the sensitivity is 75
mVrms (full frequency range).
In auto, the counter will attempt to set the trigger level between
the positive and negative peaks.  In manual mode, you need to set the
trigger level and it might be misaligned.

Mike

On 8/27/2012 12:01 PM, Bernd Neubig wrote:

My starting point is the paper by Neubig on the real vs. predicted aging:
experimenting with the model and try to cut it around the Morion or the
Oscilloquartz OCXOs.

On Tue, Aug 28, 2012 at 12:15 AM, Don Latham djl@montana.com wrote:

Several decades ago, the concept of the "smart clock" arose
at what was then HP.  The idea was as discussed here to
"characterize" past aging, "predict" future aging, and
then "correct" the aging.  The goal wasn't to turn a quartz
oscillator into an atomic clock replacement, but simply
to get the oscillator through a 1 hour or so "holdover" time
during GPS outages.  It sort of worked for that very limited
purpose, but in general, past performance of HP crystals wasn't
a very good predictor of future results.  Crystals would age
in one direction for a while and possibly slow down as time
when on, but then then might start aging in the other direction.
There were also frequency jumps that were substantial and totally
random.  The reason why the HP crystals were unpredictable was
that all the deterministic processes such as mass preferentially
depositing on the crystal, so as to make the frequency age
lower, had been eliminated by years of manufacturing improvements.
The remaining processes were of the nature of quartz stress
relaxation that were very random.

Rick Karlquist

Don Latham wrote:

Could you refresh me with the reference or email the paper? I'd like to
give it a read.
Don

Azelio Boriani

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
R. Bacon
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

On 8/27/12 4:15 PM, Rick Karlquist wrote:

We see similar things in USOs  (and other components as well) for
spaceflight..

Things that people worried about 60 or more years ago just don't occur
anymore.  People used to obsessively try to match diodes in HV strings,
for instance, because the process variability was high enough to make a
difference.  These days, you get a reel of diodes and they're all pretty
much the same, and even reel to reel from month to month they don't
change much.    That's what all that 6-sigma stuff is all about, after all.

All the low hanging, and even middle hanging, fruit has been picked..

(one big exception.. ICs which are not designed for radiation tolerance
seem to have large variability in radiation tolerance..it's just not
something that's controlled for in the process)

On Mon, Aug 27, 2012 at 7:33 PM, Jim Lux jimlux@earthlink.net wrote:

We know what a OCXO is and a TCXO is.  I was at a presentation at work a
whike back and they called what you describe a "MPCXO"  or MicroProcessor
Compensated XO.    They said the characteristics were between the OCXO and
TCXO

Doesn't the thunderbolt do this.  I think it watches the aging rate of the
OCXO and adjusts during hold over.

--

Chris Albertson
Redondo Beach, California

Thank you very much, Bernd. The papers are on the stack. Now if I can
just get to them, they look very interesting.
73, Don

Bernd Neubig

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
De Erroribus Medicorum, R. Bacon, 13th century.
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

Chris,

That is a big no !  What the Thunderbolt is doing is adjusting the OCXO to keep
it aligned with a known reference from the GPS system.  The fact that a
microprocessor is involved is from a totally different perspective.

A microprocessor controlled XO is a non oven crystal oscillator system that has
additional computational control providing a bit more than just mere passive
temperature compensation.  The additional computational capability deals with
having coefficients of that particular oscillator's behavior pre coded to
compensate for the nonlinear behavior over a given temperature range.

A microprocessor controlled XO system allows for using cheap crystals with
minimum processing time and costs.  Because of limited storage space there is no
way for the system to have enough data to even try to compete with the quality of
a decent OCXO.  Beyond its initial calibration setup, it has no way of keeping it
tied to a known reference, like the Thunderbolt is doing.

Bill....WB6BNQ

Chris Albertson wrote:

www.axtal.com/data/publ/aging_e.pdf

On Tue, Aug 28, 2012 at 2:32 AM, Don Latham djl@montana.com wrote:

On 8/27/12 10:45 PM, Chris Albertson wrote:

Not really.  The MCXO has a one time calibration for frequency vs
temperature that's programmed into it (and some use very clever ways
to measure the temperature).  The disciplining algorithms in a GPSDO are
a bit smarter; some explicitly develop a model for the f vs T and apply
it, in others it's essentially embedded in a higher order filter which
takes the measured T, along with other parameters, into the filter.

I don't know if the GPSDOs try to do a time series fit/model (at least
for low order terms) to deal with things like diurnal variation.  They
could.

I should note that the MCXO approach, popularized as a "better TCXO" in
a small low power package, is also used in some software defined radios,
except that there's no separate microcontroller.  The frequency vs
temperature characteristic is just embedded in the other algorithms in
the radio's host processor.

On Mon, Aug 27, 2012 at 11:45 PM, WB6BNQ wb6bnq@cox.net wrote:

Yes but but the t-bolt "remembers" the corrections is was applying and
continues to apply them whenthe GPS signal goes away.  I'm pretty sure one
of the things it remembers is the aging rate.

I know the "real" uP controlled XOs use programmed data for aging rate and
temp compensation the t-bolt gts it's own data by comparing with GPS.

Chris Albertson
Redondo Beach, California

On Tue, Aug 28, 2012 at 6:17 AM, Jim Lux jimlux@earthlink.net wrote:

Yes, correct.  But I was talking about during hold-over.  t-bolt in
holdover is no longer a GPSDO.    It falls back on using a temperature and
aging model.

Chris Albertson
Redondo Beach, California

thanks,Azelio. Neubig kindly sent the location for the references.
Another good company, like the Wenzel site.
Don

Azelio Boriani

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
De Erroribus Medicorum, R. Bacon, 13th century.
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

Hi

I don't know that there is any data showing the TBolt adjusting for aging
during holdover. In other words: showing the holdover DAC voltage changing
at a completely constant temperature.

Bob

-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Chris Albertson
Sent: Tuesday, August 28, 2012 12:19 PM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] oscillators

On Tue, Aug 28, 2012 at 6:17 AM, Jim Lux jimlux@earthlink.net wrote:

Yes, correct.  But I was talking about during hold-over.  t-bolt in
holdover is no longer a GPSDO.    It falls back on using a temperature and
aging model.

Chris Albertson
Redondo Beach, California

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

The then Hughes Aircraft had a nice hybrid (microelectronic ) version of this
idea produced in Newport Beach. 
--- On Mon, 8/27/12, Rick Karlquist richard@karlquist.com wrote:

From: Rick Karlquist richard@karlquist.com
Subject: Re: [time-nuts] HP/Agilet counter 53131A (not 53151A) trigger problems
To: "Discussion of precise time and frequency measurement" time-nuts@febo.com
Date: Monday, August 27, 2012, 4:15 PM

Several decades ago, the concept of the "smart clock" arose
at what was then HP.  The idea was as discussed here to
"characterize" past aging, "predict" future aging, and
then "correct" the aging.  The goal wasn't to turn a quartz
oscillator into an atomic clock replacement, but simply
to get the oscillator through a 1 hour or so "holdover" time
during GPS outages.  It sort of worked for that very limited
purpose, but in general, past performance of HP crystals wasn't
a very good predictor of future results.  Crystals would age
in one direction for a while and possibly slow down as time
when on, but then then might start aging in the other direction.
There were also frequency jumps that were substantial and totally
random.  The reason why the HP crystals were unpredictable was
that all the deterministic processes such as mass preferentially
depositing on the crystal, so as to make the frequency age
lower, had been eliminated by years of manufacturing improvements.
The remaining processes were of the nature of quartz stress
relaxation that were very random.

Rick Karlquist

Don Latham wrote:

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

On 8/27/2012 11:45 PM, WB6BNQ wrote:

It doesn't use coefficients.  It has a look up table of frequency vs
temperature.

No it doesn't use a cheap crystal.  It uses a special SC cut crystal.
This crystal could very easily cost more than an OCXO crystal.

An MCXO is a very good, but expensive TCXO.  Only temperature, not aging
is corrected.

It has nothing to do with "smart clocks".

Rick Karlquist N6RK

Does anyone know about what technology is used in Swiss watches to get much better performance from
their xtals than you might expect?
I assumed that they had look up lists to insert extra counts to compensate for ambient variations,
but I have never heard any details.
cheers,
Neville Michie

On 30/08/2012, at 1:45 PM, Richard (Rick) Karlquist wrote:

nippedLe 30/08/2012 06:12, Neville Michie a écrit :

check out
http://forums.watchuseek.com/f9/thermocompensation-methods-movements-2087.html
<snipped>

--
Les chiens aboient, et la caravane passe.

Well, basically the temperature of a wrist watch is very
constant at around 90F.  That and a table lookup that gives
an adjustment for number of cycles per tick vs temperature.

All quartz watches since about 1990 are microprocessor based,
and have a table lookup.  They can only be regulated using
a special computer interface provided by the manufacturer.

-Chuck Harris

Neville Michie wrote:

On 8/29/12 8:45 PM, Richard (Rick) Karlquist wrote:

Isn't the crystal cut (and the circuit designed) in such a way that it
supports both the fundamental and the third overtone simultaneously, and
comparing the frequencies of the two (or, more accurately, fF- fthird/3)
is how they measure the temperature, which is then used to look up the
correction factor in a PROM.

Hi

The fundamental / third approach is one of several possible ways to go. You
can also run an SC on the B and C modes to get thermometer data. Early
implementations used a pair of independent blanks, one cut to be a good
thermometer. Some have even gone as far as to mount a thermistor on the
crystal.

No matter what you do, it adds cost.

Bob

-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Jim Lux
Sent: Thursday, August 30, 2012 8:55 AM
To: time-nuts@febo.com
Subject: Re: [time-nuts] oscillators

On 8/29/12 8:45 PM, Richard (Rick) Karlquist wrote:

Isn't the crystal cut (and the circuit designed) in such a way that it
supports both the fundamental and the third overtone simultaneously, and
comparing the frequencies of the two (or, more accurately, fF- fthird/3)
is how they measure the temperature, which is then used to look up the
correction factor in a PROM.

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

On 08/30/2012 06:33 PM, Bob Camp wrote:

The reason for doing it is to lower powerconsumption in critical
applications.

Cheers,
Magnus

Interesting this third overtone/fundamental3 idea but what makes the third
overtone not equal to the fundamental3? Is the crystal operated in series
or parallel resonance?

On Thu, Aug 30, 2012 at 10:30 PM, Magnus Danielson <
magnus@rubidium.dyndns.org> wrote:

Azelio Boriani wrote:

That is where the special SC cut crystal comes into play.
It's not so much that the 3rd OT is on a different frequency, but
rather that it has a different and known tempco relative to the
fundamental, so it can be used as a thermometer.  It would be
interesting to take a crystal out of a random 10811 and compare the
frequencies of the fundamental and 3rd OT vs temperature to see
if that crystal would accidentally make a good thermometer.

Rick Karlquist N6RK

Hi

It turns out that the fundamental is "not quite" 1/3 of the third overtone. You can impact the degree of offset by changing the contour of the blank. The temperature coefficient is also different on the fundamental. The net result is that you can get a pretty good thermometer reading by mixing one with the other divided by 3.

Bob

On Aug 30, 2012, at 4:37 PM, Azelio Boriani azelio.boriani@screen.it wrote:

Hi

You can do the same thing with an AT as with an SC. They both exhibit the fundamental / third offset and tempo delta.

Bob

On Aug 30, 2012, at 5:10 PM, "Rick Karlquist" richard@karlquist.com wrote:

On 08/30/2012 11:10 PM, Rick Karlquist wrote:

All the different modes have different temperature dependencies. Being
able to maintain dual oscillation on both over the range seems like a
reasonable challenge.

Cheers,
Magnus

On 8/30/12 9:33 AM, Bob Camp wrote:

I've used the "sensor on oscillator can" technique very recently on a
software defined radio.  Over the next year, I'll get some data on how
well it works in use.  The idea is to use a low power TCXO and a sensor,
rather than an OCXO, to meet a fairly tight frequency requirement (a few
hundred Hz out of 2 GHz)  The TCXO can get you do a few ppm or better.
The sensor should let me get about an order of magnitude better, and I
need to measure the temperature anyway, and it's easy to integrate with
the software waveform code.

Elecraft seems to be having some trouble with the KX3 stability for WSJT
on 6m. They also have a temp sensor on the bard, apparently and are also
trying this out. I don't know any more about it than that.
Don

Jim Lux

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
De Erroribus Medicorum, R. Bacon, 13th century.
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

Hi

If the temperature is varying slowly and there are no gradients you may get your order of magnitude over some range. You might be surprised at your TCXO. A lot of them are pretty darn good in the vicinity of room temp. You may already be an order of magnitude past your ppm or two for fairly normal temperature changes.

Bob

On Aug 30, 2012, at 8:01 PM, Jim Lux jimlux@earthlink.net wrote:

On 08/31/2012 03:12 AM, Bob Camp wrote:

I haven't seen that any of the TCXOs compensates for the temperature
hysteresis. It would be cool if they did.

The ability to handle temperature gradients can be troublesome for both
TCXOs and OCXOs.

TCXOs have become quite good these days.

Cheers,
Magnus

Hi

Some TCXO's have less hysteresis than others…...

Bob

On Aug 30, 2012, at 9:29 PM, Magnus Danielson magnus@rubidium.dyndns.org wrote:

Hi

I think they would do better to just put in a connector to drive it with a TBolt….

Bob

On Aug 30, 2012, at 8:47 PM, Don Latham djl@montana.com wrote:

On 8/30/12 6:12 PM, Bob Camp wrote:

Temp does vary slowly (the radio weighs on the order of 6kg)..

Need to hold spec (in theory) from -20 to +40C.  The oscillator runs
about 10 degrees hotter inside.

About 0.2 ppm from 5C to 40C.  +/- 1ppm worst case over the whole
temperature range.

What I'm also interested in is whether I can compensate a non TCXO 66MHz
CPU clock oscillator (even cheaper, potentially better phase noise with
a higher Q crystal, etc.)

On 8/30/12 6:29 PM, Magnus Danielson wrote:

That is the dominant error source for changes over the 5-40C.. about 0.2
ppm difference when going up from -55 to +85 and going down over the
same range.

Hi

A standard clock crystal is pretty much junk. as far as temperature performance is concerned. Even a cheap TCXO is likely to be pretty good over 25 C +/- 10C.

Bob

On Aug 30, 2012, at 11:35 PM, Jim Lux jimlux@earthlink.net wrote:

On 8/30/12 9:04 PM, Bob Camp wrote:

yes.. but, say, one had a decent SC cut with good phase noise
properties, but large (but repeatable) temperature characteristics.

Can I get good accuracy AND good phase noise, for less hassle/power/size
than an OCXO?

My radio has a TCXO and a clock oscillator, so the clock oscillator is
my test article for the temperature compensation scheme..

Bob: Great minds run in the same track :-)
Don

Bob Camp

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
De Erroribus Medicorum, R. Bacon, 13th century.
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

Hi

An SC is going to have it's temperature curve centered up around 95C or so. If it's been cut as an OCXO crystal the turns will be up there as well. By the time it gets to room temp, the delta F / delta T is moving mighty fast. Think in terms of multiple ppm/C. A typical cell phone TCXO crystal is in the sub 0.1 ppm/C range in the vicinity of room temp.
In addition, SC's are pretty hard to pull. For a normal TCXO (no DDS) something > 40 ppm of range would be needed. By the time you get the wide tune stuff in the circuit, the phase noise isn't going to be anything special.

Bob

On Aug 31, 2012, at 12:10 AM, Jim Lux jimlux@earthlink.net wrote:

On 8/31/12 7:06 AM, Bob Camp wrote:

We're not pulling the crystal.. we just let it sit.
What we do is change an NCO in the software radio implementation.
Typically, you have a coarse tune with a PLL in steps of 500kHz-1MHz,
and then you do the fine tune in software with a digital
mixer/downconverter and an NCO (which also is part of the carrier
tracking loop).

Hi

Ok, that gets you back to the basics of really major delta F / delta T slopes.

Bob

On Aug 31, 2012, at 8:55 PM, Jim Lux jimlux@earthlink.net wrote:

On 8/31/12 7:16 PM, Bob Camp wrote:

yeah.. but as long as you know what the curve is.. the NCO has a huge
range (after all, we already have to tune over >500 kHz...a few hundred
Hz isn't a big deal.

A more complex problem is doing insitu calibration from, e.g., GPS
signals or from some externally received frequency reference. (since the
radio in question can also work as a GPS receiver, eventually).

Hi

Observing a curve and being able to compensate it are often two different things. Hysteresis is one very obvious example. Another is simple sensor lag. A some what less obvious one is that the temperature performance is also influenced by the rate of change in temperature.

Here's another thing to consider:

If your crystal is running 3 ppm / C, and you are after 3.0 x 10^-11 stability at one second - You will need to either have a rate of change at ~ 1x10^-5 C/sec (0.6 mC / min) or you will need to compensate for some pretty small changes. That of course makes a bunch of assumptions ….

Bob

On Aug 31, 2012, at 11:26 PM, Jim Lux jimlux@earthlink.net wrote:

On 9/1/12 8:32 AM, Bob Camp wrote:

In this application,  the requirement for frequency accuracy has to do
with initial acquisition.. that is, you want the signal (or receiver
tuning) to be within some few hundred Hz of where it's expected to be
(because the receiver is narrow band).

The ground station typically has a Doppler predict based on orbit
knowledge, that predict has some uncertainty.  Added to the radio
frequency uncertainty.  (SNUG - Space Network User Guide, has more info)

Once you've acquired, the receiver and ground station will track (i.e.
the ground station puts in the estimated Doppler, so all you're really
tracking is the variation in the local oscillator).  (for a LEO
satellite at 2.3 GHz, the 7km/s orbital velocity already puts tens of
kHz variation on it)

(and this completely neglects that a modern radio could use something
like an FFT for acquisition)

Temperature changes are pretty slow.. I'm seeing 5-10 degree cyclical
variation over 90-100 minutes.  Actually, the bigger change is during
the warm up transient, going from off and cold to on and warm over 10
minutes or so.

In other applications, where you're not going in and out of the sun
every revolution (i.e. deep space, rather than LEO) and you were
interested in Allan deviation type measurements for gravity science
(where we're looking for 1E-13 over 100 sec sort of performance), what
we'd probably do is warm up early.. Turn it on, compensate based on the
measured temperature, and then hold the compensation fixed during the
measurement, letting the ground worry about the apparent frequency
change due to Doppler.  We'd have a high quality narrow band signal,
just at an unknown (but reasonably stable) frequency.  What the science
team is usually interested in is small relative changes in phase &
amplitude(occultations) or in small changes in frequency (Doppler, for
gravity science).

(we regularly measure velocity to cm/sec precision for outer planet
orbiters like Cassini, Juno, etc.)

Hi

I suspect that a good IT cut would probably do better than an SC in either application. In the deep space situation, a copper slug in a dewar sounds like a reasonable addition to the design.

Bob

On Sep 1, 2012, at 11:53 AM, Jim Lux jimlux@earthlink.net wrote:

On 09/01/2012 05:53 PM, Jim Lux wrote:

If you can make reasonable predictions of the heating and cooling
profile, you could use that and hopefully gain a decade or so, and any
slew in detectors and would mostly affect the remaining error.

You could also make use of uplink carrier and GPS to improve the model
state of your crystal. That way the predicted and feed forward values
could be kept fairly well adapted. Naturally, you would want a fall-back
scenario to back out for wider offset search if you failed to maintain
the model. You don't want to loose your bird due to temporary system
failure. Another alternative would be to allow for the downlink
frequency to be a hint for the uplink where the frequency is such that
you can "pull in" the bird if need to.

Cheers,
Magnus

On 9/1/12 11:00 AM, Bob Camp wrote:

If you're going to do dewars, then you're talking USOs for which the
technology is quite well developed with ovens, etc.

But that's a kilo, several watts, and a liter.

What you want is something that is better than, say, 0.1 ppm, that is
comparable in mass/power/volume/cost to an existing 1-5 ppm TCXO (about
1x2x0.5 cm)

Hi

True on the volume and weight. Not as much power as an OCXO since it's passive. At 1x 2 x 0.5 you could fix the power and weight by using an EMXO. You still would have more power than the TCXO, but no were near as much as the couple watts a USO uses.

Bob

Bob

On Sep 1, 2012, at 10:41 PM, Jim Lux jimlux@earthlink.net wrote: