Looking for 468-DC GOES receiver info
Hi Poul:
Yes, at day 42 I set the elevation mask to 30 degrees. I have TAC32 and
it makes an elevation plot so 30 degrees is well above any of the
horizon objects. Today I'm thinking about setting it to 60 degrees
since only one satellite is needed for timing and this may get rid of
multipath.
Have Fun,
Brooke Clarke, N6GCE
--
w/Java http://www.PRC68.com
w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com
Poul-Henning Kamp wrote:
In message 4223C82C.4030108@pacific.net, Brooke Clarke writes:
Another clue that it's GPS based is to notice that the time of day
when the big changes occur are about the same. The major lines are
each at midnight. Maybe it's related to which GPS satellites are
being used?
Have you tried setting a mask-angle in your GPS ?
Try 10 or 15 degree depending on how cluttered your horizon is.
Hi Brian:
I've been thinking about bringing my laptop to the FTS4060 and running
TAC32 and watching to see if there's a correlation between the time
interval changes and a change in which satellites are being tracked.
But even with the jumps if there is no drift then I know the C field is
properly set, it's just going to take longer. Also the LORAN-C is
looking better and better as a precision time transfer method. It's
been 5 days since Middletown came back on the air and the 2100T Timing
receiver was restarted and the offset is only 20 nano seconds.
20E-9/(5*86400) = 4.6E-14.
Have Fun,
Brooke
Brian Kirby wrote:
Its possible, its multipath. Depending on your receiver, you can shut
off reception from individual GPS PRNs. Check to see which ones are
in use at the time, and then turn one off. 12 Hours later, turn it
back on and turn off another, etc.
Also, if you can turn off GPS PRNs, you can pick at time common with
NIST and use the GPS archive at
http://www.boulder.nist.gov/timefreq/service/gpstrace.htm to compare
time (use view track to see one satellite by itself).
Poul-Henning Kamp wrote:
In message 4223C82C.4030108@pacific.net, Brooke Clarke writes:
Another clue that it's GPS based is to notice that the time of day
when the big changes occur are about the same. The major lines are
each at midnight. Maybe it's related to which GPS satellites are
being used?
Have you tried setting a mask-angle in your GPS ?
Try 10 or 15 degree depending on how cluttered your horizon is.
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Hi:
Since 20 March 2005 the C field has not been changed. Although there's been some glitches that caused offsets it sure looks to me like a parabolic curve. See:
http://www.pacificsites.com/~brooke/pdf/sn1227_CF568.pdf
Is this indicative of a frequency drift?
Scratching head,
Brooke Clarke, N6GCE
--
w/Java http://www.PRC68.com
w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com
Hi again:
I've stitched the plot and fitted a parabolic curve. The equation is:
Y = -1.1367X^2 + 213.66X -927.7 where Y is in ns and X is in days, so
converting the two first constants by dividing by 86400 gives:
drift = -1.31E-14 per day
offset = 2.47E-12
What I was checking was to see if this is a cesium or crystal
oscillator, and I think it's way too good to be a crystal.
Still scratching,
Brooke
Brooke Clarke wrote:
Hi:
Since 20 March 2005 the C field has not been changed. Although
there's been some glitches that caused offsets it sure looks to me
like a parabolic curve. See:
http://www.pacificsites.com/~brooke/pdf/sn1227_CF568.pdf
Is this indicative of a frequency drift?
Scratching head,
Brooke Clarke, N6GCE
Hi:
I've been reading up on real world Cesium standards and they do drift.
For example see the first paragraph on page 4 of:
Introduction to time and frequency metrology by Judah Levine
REVIEW OF SCIENTIFIC INSTRUMENTS VOLUME 70, NUMBER 6 JUNE 1999
http://www.boulder.nist.gov/timefreq/general/pdf/1288.pdf
Cesium is just like a crystal oscillator in that it has an offset and a
drift. The above article says:
"A good commercial cesium frequency standard, for example, might
exhibit fractional frequency fluctuations of 2E-14 for averaging
times of about one day. The frequency of the same device might
differ from the SI definition by 1E-13 or more (sometimes much
more), and this frequency offset may change slowly with time as the
device ages. (This frequency offset is what remains after
corrections for the perturbations mentioned above have been applied.
If no corrections are applied, the fractional frequency offset is
usually dominated by Zeeman effects, which can be as large as 1E-10.)"
The FTS4060 time interval is following the following equation (it takes
about a month to get this equation):
y = -1.2594x2 + 236.37x - 10318
where Y is in ns and X is the Day Of the Year. The first term is the
fractional frequency stability, i.e. drift rate and is
1.14E-14 per day which is pretty good. The HP-Agilent 5071A is
specified at <1E-14 per day.
Note that the fractional frequency stability of a good lab grade crystal
standard is about 1E-10 per day, so Cesium is 10,000 times better, but
still has drift.
This explains a lot about why setting the C field near 1E-14 is
difficult, the frequency is changing all the time.
Now Having Fun,
Brooke Clarke, N6GCE
--
w/Java http://www.PRC68.com
w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com
The FTS4060 time interval is following the following equation (it takes
about a month to get this equation):
y = -1.2594x2 + 236.37x - 10318
where Y is in ns and X is the Day Of the Year. The first term is the
fractional frequency stability, i.e. drift rate and is
1.14E-14 per day which is pretty good.
Do be careful here. Excel will blindly report
equations with 5 significant digits no matter
what the data looks like.
Here's something to try: break up your data
into three 10-day segments and see how well
the x2 term of the equations agree.
Or convert phase to frequency and then plot
30 days of frequency. If you have real drift it
should be clear from this plot.
DougH, JohnA, and I have Stable32 which makes
this a snap if you want to send any of us the raw
phase data.
The HP-Agilent 5071A is specified at <1E-14 per day.
Note that the fractional frequency stability of a good lab grade crystal
standard is about 1E-10 per day, so Cesium is 10,000 times better, but
still has drift.
All frequency standards have frequency instabilities.
Hydrogen masers, Quartz, and Rubidium have drift,
but Cesium standards are generally considered to
have zero drift. That's one reason UTC is based on
that technology.
And note that fractional frequency [in]stability is not
the same thing as frequency drift. I can go into this
in more detail if you wish.
This explains a lot about why setting the C field near 1E-14 is
difficult, the frequency is changing all the time.
If you make frequency plots in addition to phase plots
you will see dramatically why setting the C-field of
your 4060 to 1e-14 is hopeless. Frequency plots will
graphically show frequency instability (the width of
the line) and frequency drift (the slope of the line).
/tvb
Now Having Fun,
Brooke Clarke, N6GCE
Hi Tom:
Think of it this way. Offset is a measure of how well someone has set
an oscillator. Stability is a measure of how well the oscillator
maintains it's frequency. Stability is the money spec. A lab grade
crystal oscillator may have a stability of parts in E-10 per day. That
means that if the offset is set to zero a day later the frequency may by
off by parts in E-10.
The world in not a perfect place and any Cesium oscillator also has
aging, it's not perfect, that's why there's a spec on the 5071A saying
less than 1E-14 per day. NIST has a fountain Cesium where they say less
than 1E-15 per day. That number is the stability NOT the offset. To
see the stability you need to accumulate enough data to see the
parabolic shape. It took me over a month after I got the wrinkles out
of my test system.
Can you hookup a Cesium source and make a time interval measurement for
a period of a month? Do you have software that will tell you the
stability (not offset) of the oscillator being tested?
Having Fun,
Brooke
Tom Van Baak wrote:
The FTS4060 time interval is following the following equation (it takes
about a month to get this equation):
y = -1.2594x2 + 236.37x - 10318
where Y is in ns and X is the Day Of the Year. The first term is the
fractional frequency stability, i.e. drift rate and is
1.14E-14 per day which is pretty good.
Do be careful here. Excel will blindly report
equations with 5 significant digits no matter
what the data looks like.
Here's something to try: break up your data
into three 10-day segments and see how well
the x2 term of the equations agree.
Or convert phase to frequency and then plot
30 days of frequency. If you have real drift it
should be clear from this plot.
DougH, JohnA, and I have Stable32 which makes
this a snap if you want to send any of us the raw
phase data.
The HP-Agilent 5071A is specified at <1E-14 per day.
Note that the fractional frequency stability of a good lab grade crystal
standard is about 1E-10 per day, so Cesium is 10,000 times better, but
still has drift.
All frequency standards have frequency instabilities.
Hydrogen masers, Quartz, and Rubidium have drift,
but Cesium standards are generally considered to
have zero drift. That's one reason UTC is based on
that technology.
And note that fractional frequency [in]stability is not
the same thing as frequency drift. I can go into this
in more detail if you wish.
This explains a lot about why setting the C field near 1E-14 is
difficult, the frequency is changing all the time.
If you make frequency plots in addition to phase plots
you will see dramatically why setting the C-field of
your 4060 to 1e-14 is hopeless. Frequency plots will
graphically show frequency instability (the width of
the line) and frequency drift (the slope of the line).
/tvb
Now Having Fun,
Brooke Clarke, N6GCE
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
At 08:32 PM 4/18/2005, Brooke Clarke wrote...
Can you hookup a Cesium source and make a time interval measurement for a period of a month?
Against what, and which one is drifting? If UTC is based on Cs, and Cs has intrinsic drift, then so does UTC, so who's measuring who?
Perhaps you need to measure against PSR 1937+21.
I wanted to follow up on this thread even though
Brooke and I have talked offline.
To the question -- could I measured a 5071A for a
month and what is its stability? -- attached are
200-day 5071A frequency plots; one is hourly
averages (2freq2.gif) and the other daily averages
(2freq4.gif). The plots are flat (no frequency drift).
Also attached (1sigma1.gif) is a stability plot going
out to tau 4 million seconds (45 days). You can see
these high-perf 5071A are in the 15's and still have
not quite hit their noise floor.
As for accurate definitions of stability, aging, and drift
have a look at pages like:
http://www.ieee-uffc.org/freqcontrol/vigaging91/introduc.htm
http://www.ieee-uffc.org/freqcontrol/quartz/vig/vigaging.htm
http://www.accubeat.com/glossary.asp
It is true that no frequency standard is perfect and all
have finite stability. Because of instability, frequency
will in fact wander around. But drift is something else;
it's the systematic (think gradual, continuous, unbounded
monotonic, sometimes linear) change in frequency over
time -- and this is typically not observed in a Cesium
standard.
/tvb
My understanding from talking to the experts is
that no aging, or environmental effects are observable
on the 5071A, and measurements have been made into
the 10^-15 range. Also, no systematic frequency offset
has been observed down to at least 1x10^-14. IE,
frequency offset between different 5071A's is randomly
distributed with a mean within 10-14 of laboratory
standards, such as Cs fountains. Knowing the way we
overdesigned the 5071A, I am not surprised that it
is this good.
Rick Karlquist
(member of 5071A design team)