FTS4060/S24 Cesium Frequency Std
Datum 4065B Time & Frequency Standard
Fast Test Method with
4065B Frequency Offset
Symmetricom Chip Scale Atomic Clock
|Datum 4065B Cesium Time
Frequency Standard top off showing three 8 Volt
On the right the toggle switch (advance/retard and the
switches control the movable 1 PPS output in 100 ns
The BNC just below the toggle is the programmable
frequency TTL square
wave output (.1/1/5/10 MHz).
The yellow cap is on the 1 PPS master output.
The black cap is on the 1 PPS sync input BNC.
Two 7AH Gel Cell 12 Volt Batteries connected while
working on the three
8V 5 AH batteries.
Just to the right of the line cord jack is a toggle
switch which when
turned down shuts down all of the unit except for the
A much better solution to storage than circuit mods.
The field of type-N connectors has 3 ea 10 MHz out, 3
ea 5 MHz out and
3 ea 1 MHz out all 1 Vrms, 50 Ohms.
The BNC above the DB-9 is the alarm status and the
DB-9 is alarm info.
The DB-25 is a serial connector for data I/O.
This box does not have the optional telecom outputs.
The far top right BNC is the master pulse out.
Below it is the adjustable offset 1 PPS out.
|Getting ready to
them. You can see the family resemblance.
Note both units
are locked and operational.
The 4060 is s/n 1227 and should be fairly close to on
Fast High Precision Set-up of SR
This idea is from the PRS10
Frequency standard manual appendix B.
- 10 MHz Reference to BNC-T on rear panel 10 MHz Input
front panel "A" start input
- 10 MHz DUT to front panel "B" stop input
- Front panel 1 kHz TTL REF OUT to front panel EXT
- CONFIG - press SET to select cAL and press
SELECT to choose
"cLoc SourcE", use arrow keys to set rEAr.
- in the Gate field: select POS, TERM = 50 Ohms and
- for sine wave 10 MHz inputs set the A and B input
fields to: AC,
50 Ohms, Level full CCW, + Slope.
MODE to FREQ, SOURCE to B, GATE/ARM to 1 second and SAMPLE
SIZE to 1
then hold START down for a few seconds, DISPLAY to MEAN.
This display should be within 0.1 Hz of 10 MHz
Fine Frequency Measurement
This will show 1E12 in one second.
MODE to TIME
SOURCE of Start to A
GATE/ARM to +TIME and EXT
SAMPLE SIZE to 1000 (1 & 103
Now each second there will be a display of 1,000 averaged
readings. This brings the 620 precision down to 1
Datum 4065B Cesium Time & Frequency
The FTS4060/S24 is really a frequency standard and I've always
an excellent time standard.
This unit came from an eBay ad showing a Major alarm and not
locked. But when powered up it locked in about 10
after connecting the removed and taped battery wire and
additional 10 minutes the charge fault could be cleared.
Each of the three batteries is has four "X" 5 AH cells where
Cyclon cylindrical sealed lead acid cell. The 2.5
used in the O-1814
Frequency standard. The top of each battery is marked
"bad" and a
date of 10-2-01. The voltage across the pack is now 31.2
there probably is no current flowing if any one cell is bad,
testing will be needed to see what the real problem is.
Theory of Operation
This is a modern Cesium standard that uses control loops so
frequency is correct, i.e. the C-field adjustment is automatic
manual like the 4060.
Unlike the FTS 4060 where you manually set the C-Field this
monitors and adjusts based on the following items as reported
Status 5 menu:
-10 S.B. <= 160 mv
<= 40 mv
-15 S.B. <= 160 mv
Rabi-Zeeman Error -4
Ramsey Confidence +3
The AC mains power a 31 VDC supply that's diode ORed with the
battery and the external DC supply. The internal supply
is a 24
Volt lead acid battery, 12 cells of Cyclon "X" size
pure lead acid. There are a couple of major problems
- The temperature inside the case is always warm to almost
hot. That degrades the life of the lead acid battery
by an order
- If there's any venting, like happened with the Gibbs
crystal oscillator, the fumes and heat make a good
combination to etch
the traces right off printed circuit boards.
So, rather than replacing the internal lead acid battery, it
connected as an external battery. Since a diode OR gate
to combine the three power supplies a new wire needs to be run
of the unused pins on the external battery connector to bring
internal battery terminals to allow float charging.
The Red-Black Siamese cable has male 1/4" Faston connectors
those on the batteries and so plug into the internal battery
connectors. The white label says:
Remote Internal Battery D:
-24, E: +24
The Plug in the lower right of the photo connects D to black
and E to
Red. The other end of the cable has female 1/4" Faston
to plug into the battery. This solves the two major
an internal battery, i.e:
- If battery vents acid fumes they will
destroy the expensive circuitry
- The battery will last 3 to 10 times
running at room temperature
Pins D and E on J100 External DC Input were spares and so a
external battery can be added just be wiring it to pins A
and C (return). Note the external battery must
own charging circuit and should be at some voltage level
chosen by what
priority it should be used.
4065B Frequency Offset
On the FTS4065 the C-Field adjustment is made by the internal
microprocessor and there is a seperate Frequency Offset
After plotting for the time interval between the 4065B and GPS
May to 7 May (4065plot9.pdf
and finding a straight line with a slope of 472E-15 the
Offset was changed from +000000 to +000472 (a 50-50 gamble
sign should be +).
new plot was started 7 May after the change and for the first
days seemed to have worked. By starting a new plot I
constants were subtracted from the time interval so it starts
and the starting second count starts at 0.0. This way is
is a straight line it can be forced to go through (0,0).
By May 10 at 9:50 am it looked like the frequency offset had
worked. The data (plot10a.pdf
a box about +/- 10 ns high after 2 3/4 days (3E-13) but more
needed for good data.
Then the points started a climb. By 14 May (4 days
data between May 10 and 14 looks again like a nice straight
(plot10.pdf)with a slope of about 472E-15. BUT the
offset is still set at +000472. (4065BvsGPSp10b.pdf
The 4E-13 number floating on the plot is the slope after one
Just put it there so I could remember what it was. Excel
recomputes the slope as each new data point is added.
The 4065 box was rotated 90 degrees and it did make a good
Before this plot was started the frequency offset was stable
(parts in E-15) and the 4065 Frequency Offset was set to +472.
For the first couple of days it looked like that change was
the frequency offset was near zero.
But then the frequency offset returned to +472 (this with the
Offset dialed to 472) and that continued for over four days
turned the 4065 box 90 degrees clockwise (just prior to the
day 7 grid
line). That made a big change and now (May 17 2008) the
more like 167 (parts in E-15).
It may be that the Earth's magnetic field is having an
maybe just the mechanical shock has the influence.
Any thoughts what's going on? Contact
Model Numbers & Options
Setting C Field
Manual Control Voltage & Loop Gain
Government Liquidation Warning
A common misconception (and one
had until working with a Cesium standard) is that the timing
perfect. This is not the case. A Cesium standard
around the nominal frequency, but may not drift like a crystal
oscillator. A couple of terms will help when
working with this concept.
Offset - is a measure
close to the desired frequency an oscillator is running.
example an oscillator that's supposed to be at exactly 10 MHz
is off by
0.0001 Hz has an offset of 1E-11. The offset is only
valid at the
instant when it was measured. It's measure a of how well
set the frequency not so much about oscillator quality.
this is something that's under the user's control a lot of
effort go into minimizing the offset. It's common
setting the frequency of a lab grade crystal oscillator to set
at the edge of the system spec, but on the side where aging
the frequency so it at first gets better, then it's perfect,
moves to the other side of the spec. In order to do this
aging rate (i.e. stability) needs to be known.
Note that a Cesium standard may not be set to have a zero
rather most end up with an offset on the order of parts in E13
E14. The offset is known and can be backed out of
other time standards. But if the Cesium standard will be
a clock or say a transmitter, the setting the offset to the
possible value is important. The key thing is that there
aging, i.e. a time interval plot vs. GPS will be a straight
whereas a crystal or Rubidium oscillator will have a parabolic
Stability - Stability
money spec. a measure of how
well the frequency stays the same. A perfect oscillator
change frequency with time, power input, temperature, etc.,
can't get that one. The measure of how the frequency
running time is called aging. The specification on the
(Agilent) 5071A Cesium standard ($50k) is less than 1E-14 per
day. . That's to say that if it was set with a
today, by tomorrow noon it might by off frequency by
The plot at the bottom of web page
(650 days of data) shows what might be a random walk of around
minus 100 ns for an HP 5071A.
My s/n 1227 is running at abut -1.4E-14 per day. It
would be a
tad out of spec for the HP 5071A. Cesium sources are NOT
to have aging ike this.
Cesium standards are a step
Rubidium standards and are the basis of the definition of a
second. But that does not mean they are "perfect".
These are the S24 option that has a 1 MHz front panel
output, NSN 6625-01-245-3092. The official definition of
second of time is exactly 9,192,631,770 oscillations of a
between the F3 and F4 states.
was probably the first commercial Cesium standard. It
analog, no microcontrollers then. HP took over the
Varian line of
Cesium standards. Then when HP and Agilent split,
the Time and Frequency instruments and HP then became a
The FTS4060 I would call a
generation Cesium standard because it
has a micro processor that replaces a lot of analog
circuitry and is
much easier to use. There is a manual C field
needs to be set where the coarse thumb wheel is 1E-12 per
tick and the
fine wheel is maybe 1E-14 per tick.
These were purchased from
Government Liquidation with a condition code of "A1" which
that they are new. Here are some dates:
date on Lid
Possible meaning: The first
voltage measurement was done at the factory as part of the
inspection and so is close to the ship date. These dates
serial numbers are in the same order. The front panel
date may be
when the units were tested prior to being put up for
These dates seem to be over 2 years prior to the auction date
be due to how fast the government surpluses them. Note
about 14 years from the final test date to the surplus
maybe there is some number of storage years after which these
surplused, say 15 years.
s/n 1013 was shipped outside it's carton and failed to lock
received. Opening the bottom of s/n 1013 shows the
tube, Brick power
supply, and fancy 10 MHz crystal oscillator plus other
components. Behind the Cs tube is paper work indicating
replaced in 1996. There must be some reason that s/n
1013 is not
working, need a manual to find it. In the left photo at
of this page notice that the green "LOCK" light is on for the
standards (1033 & 1227) and off on the bottom (1013) one.
Model Numbers & Options
The normal FTS4060 comes in with
a 10 MHz optimized output (/201) or a 5 MHz optimized output
The /S24 option was a special unit made for the U.S. military
only a 1 Mhz output on the front and rear panels and does not
other frequencies as outputs and does not have the 1 PPS
does not have the internal or external DC supply
options. It's a
stripped down model.
All of my units (s/n 1013, 1033, 1227 have the SMA-f connector
A5 Distribution Amplifier Assembly, but not all /S24 units
connector. It's the 10 MHz output at about 4.4 V Pk-Pk.
My s/n 1013 has a rear panel with a number of plugs like it
same rear panel used for a full featured 4060, but s/n 1227
has a solid
rear panel with no plugs, so in order to install the 10 MHz
moved the alarm connector to the inside of the box and
replaced it with
the 10 MHz output.
Note that some FTS 4060 use a 5 MHz OCXO and others use a 10
061 - 1 MHz and 100 kHz RF outputs
116 - Time-of-Day Display and 1 PPS Advance/Delay
117 - 1 PPS Advance/Delay
010 - Internal Battery and Charger
015 - External Standby Battery Supply
013 - Chassis Rack Slides
Just plug in the line cord, set
switch to ON and the LOOP switch to CLOSED. After
10 minutes to 30 minutes the green LOCK light will turn on and
ALIGN pushbutton-lamp will turn off.
You can manually press the Red Operation Alarm Light/Switch to
Pressing the AC Power Reset switch will turn off the red Alarm
Pressing the Align Light/Switch may turn it off or initiate a
The Voltages shown on the meter have been scaled to fit the
to 5 volt range and are not the actual voltages in the
Near the brick power supply on the top side there's a Molex
connector with Red, Black and Orange wires. The
connector is the
same 4 terminal connector as used for hard drives in PC
computers. To get a mating connector buy a "Y" PC power
cable. You can tease out the male pins using a jeweler's
screwdriver if you don't have the extraction tool ( a hollow
fits over the male pin). The reassemble with the black
to the black ground wire in the FTS4060, Red to the +30 wire
going to the orange +5 volt wire. This makes for an easy
connect both an external DC backup supply, like the Austron
also to supply 5 volts for a 1 PPS divider.
Setting C Field
23 May 2006
Until about a week
ago I was using
the GPS 1 PPS as the start pulse and the 1 Mhz
output from the FTS4060
as the stop pulse into a Stanford Research SR620
Time Interval Counter
and doing 500 second averages. There are
some problems with this:
- The 1 MHz signal only allows a TI range of
1 micro second
before rollover occurs.
- When you get near the rollover the average
data from both sides of rollover and is very
- MOST IMPORTANT the noise is much higher
than it needs to be!
By changing the setup so that the FTS4060 10 MHz
output feeds the SR620
rear panel Reference Input (and setting the
counter to use the external
reference frequency) and then using the front
panel 1 kHz Reference
output as the stop signal two things
happen. The rollover time is
now 1 milli second (10,000 times longer) and the
noise is reduced
(probably SQRT(10000) = 100) by a huge amount.
With a 1 MHz stop if the TI is between 0 and 200
or between 800 and
1000 there is a chance of rollover points being
in the average and
between 0 and 100 or between 900 and 1000 it's
almost certain that
there will be rollover points in the
average. Because of this I'm
currently slewing s/n 1003 which was at 980 ns
and may have an optimum
C-field setting of 908 by setting the C-field at
000 where the slew
rate may be in the +20 to +40 ps/sec area so it
sill take many hours to
get the TI to about 500 ns.
Note that the 1
kHz out and the
cable between Ref Out and B in, have an
associated time delay so the TI
numbers will not match those with a direct
17 May 2006 s/n
By plotting the offset vs. C-field switch setting
it's clear that the
slope is -1E-13 per tick. This is also a great
way to see which
data points are not valid. For example the old
data point for
C-field setting 492 was +9.8E-13 which is maybe 10
times higher than
where it may end up. As of 22 May 2006 it's
-6.24E-14 with R2 of
0.84. When R2 gets up to one or two nines,
then we'll see.
Note that a Cs frequency
source is just
like any other high stability source and needs to have it's
set. The big advantage of Cs is that once set it will
like Rb or Quartz. Note this is because of the
time, and maybe is not reality.
15 Feb 2005 - The time interval plot was drifting up so an
was made to the C field. It turns out that the
adjustment was too
large. 24 hours after the adjustment for abut 6 hours
interval stayed constant within about 10 nano seconds.
indicates that even 24 hours after a C field change the
not stabilized. It really does take two to three
the unit to stablize after a C-field change.
Slew using C Field
9 March 2006 - When using a time interval counter where the
signal is the 1 PPS output from a GPS timing receiver and
signal is from the 1 MHz output of the FRS4060 the counter
has a range
of 0 to 1,000 nanoseconds. If the time interval rolls
get a saw tooth type plot. In order to slew the time
away from 0 or 1,000 you can set the C field as far as
the correct setting. For example on s/n1013 the
will be near 855, so setting to 000 causes the frequency to
about 1,400 ns per day which is an offset of 1.6E-11
but you can
see to get a 500 ns change will take about 8.5 hours.
Not as fast
as you would like. There may be other ways to slew,
but so far
Note that when the C field is misadjusted as much as
the offset is 10 to 100 times better than the best ovenized
crystal oscillators daily aging rate(1E-9 to 1E-10).
Thumb Wheel Switch Sensivity
11 Dec 2005 - s/n 1013 after complete dissassembly and
to be working, not like in Jan 2005.
With C field thumb wheel switches set for 500 the plot of
(start = GPS
1 PPS, stop= FTS4060 1 MHz zero crossing) has a positive
slope of about
5.7864 ps/sec and the plot with C field at 900 has a slope
ps/sec. So the thumbwheel switches may have a scale
Scale Factor = (7.5628 - (-1.7764)) / (900 - 500) = 0.023348
about 2E-14 per click.
See framed coment above the scale factor for s/n 1227 is
When making a Time Interval
there are different signals that can be used. The
common ones are
a 1 PPS, 1 MHz or 10 MHz. The big advantage of the 1
is that it takes a long time for a 1 second
rollover. If the TI
counter is triggered with a 1 PPS pulse (like from the
PSR10 Rb source)
and the 1 MHz output from the /S24 Cs source is used as
the stop signal
to the SR620 then the data will have a range of 0 to 1
(i.e. the period of the 1 MHz signal). If the Cs
source can be
off by as much as 1000 counts where each count is 1E-13
then it might
be off by as much as 1E-10. When the 1 MHz output is
used as the
stop signal then rollover might occur every 1000
means that the TI needs to be measured a number of times
1000 second period. You can not just measure at
by 24 hours when the source stability might be as bad as
If a 1 PPS stop signal was available from the Cs source
be no rollover problem since a source with 1E-10 stability
drift 8.6 micro seconds in 24 hours.
I have a WWV clock and when the 1 PPS from the PRS10 is
used to trigger
the A counter input the counter trigger LED is flashing at
seen by eye) that same time as the WWV clock changes
This makes it easy to tell which reading is axactly on the
manual recoring into an Excell spreadsheet.
If the 1 PPS output from a
M12+T Timing receiver had the sawtooth corrected so that
it was not
modulating the 1 PPS position then the time needed to
set a Cs standard
would be reduced by a factor of 10 or more. For
instead of an uncertatiny of +/- 50 ns on each pulse the
was +/- 1 ns then instead of needing 50,000 seconds
(13.8 hours) to see
1E-12 you would only need 1,000 seconds (16.6
WRONG #1- Since
the saw tooth
error is symetrical it gets removed when averaging is
done. On a
500 second average using the GPS 1 PPS as the start and
the SR620 1 kHz
Reference Out as the stop the standard deviation on the
group of 500 is
right at 9 nano seconds, but the mean value is
independent of the
WRONG #2 - 50 ns
is the saw
tooth size for the older 8 channel Motorola timing GPS
the M12+T only has a 9 ns saw tooth.
Direction of Change
If the time interval has a
slope then the period of the FTS4060 is increasing and
so to reduce the
period the frequency should be increased and so the
switches should be moved to a higher number. This is for
the case where
start is GPS and stop is the FTS4060.
One way to adjust the C field 3 digit thumb wheel pot is to
GPS. Although there is some jitter on the GPS 1 PPS
amounts to maybe plus and minus 50 ns (Motorola 8 chan), the
over a 24 hour
period is on the order of 1E-12. The Motorola M12+T
has about 9
ns and so is much better. The GPS receiver should be
directly. Also there a lot of jitter on the 1 MHz
much better to divide it down to 1 kHz or 1 PPS.
It seems that the scale factor for the /S24 using is 1E-13
NOT the 2E-14 in the normal FTS4060 manual.
9 Jan 2005 - Plot
10 Jan 2005 - Plot
seconds of Time
vs seconds of running time
for s/n 1013 - 5.9E-12?
12 Jan 2005 - Plot
Between 124,920 and 169,680 (12.4 hours) the 1 PPS input to
was removed and when reconnected caused a negative swing
until 248,400 seconds. But it appears that a C-Field
913 is pretty close to correct. The drift is in the
e-13 ot E-14
18 Jan 2005 - Plot of s/n1013 vs. s/n
, 1227 vs GPS & 1013 vs GPS, now using s/n
1227 as 10 MHz
ref for SR 620 counter.
s/n 1227 has it's C Field set at 600 as received from Govt
Liq. and it
appears to be moving at -3.6E-12. s/n 1013 seems to be
moving at 5E-10
more like an OCXO than a Cesium, but the Lock LED is on and
current peaks as it should. What wrong?
1 Feb 2005 - s/n 1227 - I tried to use the time interval
and the 1 Mhz output to set the C Field by getting the
offset and then
dialing in the correction (it looked like the three thumb
1E-12, 1E-13 and 1E-14), but the resulting slope after a few
observation seemed to overshoot. A better way would be
to use a
binary search where at each attempt you would half the
think I have the setting to within a single count on the
but it'll take some days to see.
10 Feb 2005 - s.n 1227 -
Still have C
field at 544
. The 10 day plot shows GPS
wandering within a
150 ns range so the poorest stability might be 1.7E-13, but
would be more like parts in E-14.
10 Feb 2005 - Enabled Ionospheric correction in GPS receiver
delta time jumped up to the 500 ns range, so this may
account for the
100 or or ns variation the last 10 days. More time
11 Feb 2005 - changed GPS to track 4 highest satellites and
elevation mask to 30 degrees.
It's very important that the GPS receiver is properly
setup to get the
best timing results.
28 Feb 2005 - The C field
has been at
for about 9 days and on average there does not
appear to be any
drift, but it's difficult to tell.
2 March 2005 - To improve the stability of the GPS 1 PPS I
the elevation mask again, this time from 30 degrees to 50
degrees. It has made a big improvement. The
deviation after 1,000 seconds worth of 1 PPS averaging is
now in the 30
ns area where before it was in the 200 ns area. During
3 days of
observation there never was a time when there were no
50 degrees. Since I'm running the GPS receiver in the
(known antenna position) only one satellite is needed for a
8 March 2005 - C filed at
After the problem with the 4060 going crazy after a beam
centering. Needed to cycle power to get good
13 March 2005 - Yesterday the counter got unplugged, but
FTS4060 nor the Austron 2100T were unplugged. Both of
instruments have warning LEDs that would indicate a loss of
power, but the FTS-4060 output frequency became more
this event. This morning I unplugged the FTS4060 for
and restarted it. After that the standard deviation on
interval improved from over 300 ns to more like 30 ns.
there are some power supply caps that need replacing or more
to be added?
Also the amount of averaging on the GPS 1 PPS needs to be
increased. At 1,000 averages the best stability that
can be seen
in one day is about
(3 * 35 ns * 2) / (SQRT(1,000) * 86400) = 7.6E-14, but
to 5,000 seconds the system improves to 3.4E-14. So
15 April 2005 - Switched to an SynPaQ/III with Motorola
receiver. This unit has 3 to 4 times less variation
than the old
8 channel UT+ GPS receiver. But there appears to be a
change in the plot
the past 5
weeks that I don't understand. The C filed has been at
20 March 2005.
28 April 2005 - the plot for s/n 1227 vs both GPS and
appears to be parabolic
some type of aging which is NOT supposed to occur whith a
source. Aging is about -3E-14 per day.
29 April 2005 - the aging
seems to be slowing down. It's now -2.2E-14/day.
1 Feb 2006 - s/n 1013 seems to be working after having all
taken apart (working on technical manual) and then put back
again on 9 Dec 2005. Changing the C-field causes a
takes about a week to settle down (now C=850) and for the
last few days
the 1 PPS has stayed within about 1 ns of the Motorola M12+T
(maybe 1ns/3 days = 4E-14).
6 Feb 2006 - s/n
is showing drift like s/n1227. The equation
for s/n 1013
y = 2.7943x2 - 302.64x + 8969.4
quality of fit is
R2 = 0.9088. The x-axis is in days and the y-axis is
The first deritive of the equation has a first term of 2 *
2.7943 * x
ns/day or +5.3E-14 drift rate.
I don't know if this is a measurement problem or a problem
9 March 2006 - The apparent parabolic aging was a
related to setting the time interval counter trigger level
(50 Ohm source and load TTL should be at 1.25 Volts, NOT 2.4
Now s/n 1013 is looking very good. Another problem may
that the Ultra Stable Oscillator coarse frequency was not
properly. It now has been centered and now looks like
offset which I'm trying to adjust to be much better.
28 April 2006 -
GPS has some noise. For example the Motorola
M12T+ has a standard
deviation of about 9 ns when 500 Time INteval
readings are averaged
(reference is some good oscillator). So you
might expect that the
noise will peak +3 sigma and -3 sigma from the mean
means that the offset you can see is about
54ns/(measurement time in
to 4 min
min to 1
- 4 days
When plotting Time Intervals in Excel you can fit a
trendline and also
get an R squared quality of fit number. R^2 should be
of nines for a good fit. If it's not then there's
17 May 2006 - there are times that last for about a couple
whee the SR620 is displaying a standard deviation for a 500
average as high as a few hundred nano seconds. I still
what causes this. Some possible things that might
cause it are:
- multipath may cause a problem if there was only one or
visable, but I would think with 3 or more visable a poor
would not cause a problem?
- no satellites at all would allow the GPS receivers 1
PPS to be
coming from it's raw crystal
- some problem with the SR620 - I have disconnected the
the external ref since it's not needed and between the
power supply, GPS receiver there's just that much more
wrong. But this has not seemed t make any
Instead of connecting the cesium 1 MHz signal to the
B input, connect
the 10 MHz signal to the counter's rear panel 10 Mhz
SEL, SET & SCALE^V to enable rear clock
input. Then connect
the front panel 1 kHz Reference TTL output to
the B (stop)
input. Now the rollover will be every 1 mS, or
a thousand times
The problem was
that the data was
getting near the 1,000 uS rollover point caused
by using a 1 Mhz signal
for the counter B (stop) input.
If you're using Julian Day numbers (maybe 6
digits) and have less that
20% of the JDN worth of data, the Excel trendline
will be in
error. Much better to subtract a very large
constant from the JDN
so that the x-axis starts from zero. This
way the trendline is
The number of data points should be the same on
either side of a
straight trend line. In my case ALL the data
points were on one
side of the line.
The LORAN-C system will
will be improved (2005) and offers a high quality time
The Austron 2100F
will work for this
The HP 5060A manual says the
frequency should be 42.82 kHz and about 1 volt RMS.
And that an
error in the Zeeman signal of 1% translates into a Cs
of 3.6E-12, so it needs to be set to within about 1
amplitude of the Zeeman signal and the C-Field
can be adjusted, with the modulation off and the loop open
to set the
C-Field, or the C-Field can be measured by adjusting the
amplitude of the Zeeman input to get maximum beam current.
Mr. Pieter Zeeman won the Nobel
along with Mr. Lorentz for explaining a
splitting in the spectral lines of light caused by magnetic
fields. This explanation was based on the new things
"electrons", but did not take into account quantum effects
like up and
down spins. His experiments and the theory by Lorentz
shed a lot
of light on what an "electron" was.
So far I don't have an audio generator that has the required
settability AND enough drive power to do this test.
Corby D Dawson and Tom Van Baak have described how the audio
for the Zeeman effect depends on the physics package and I'm
it here. The definition of the second is based on a
standard running in a zero magnetic field at sea level with
9192.631770 MHz. Real Cesium tubes run with a very
field and so their frequency is slightly off that for a
second, but the manufacturer knows how far off and allows
for it so
that the final 10 MHz or 1 PPS is exactly correct.
milli G 1
field coils in both HP and FTS Cesium tubes have the same
per milli amp constant and so the C field is determined by
how the main
frame is setup.
synthesizer that generates the frequency that's fed to the
is also in the main frame and has a frequency that matches
of the C field.
Note that as long as the C field and synthysizer are matched
other the system should work properly.
was a scheduled power outage
whicl PG&E replaced a power pole. Since I still have
the Austron 1290 Back Up power supply operational, I juse
couple of 12 Volt 7 AH gel cell batteries in series with a
schottky diode. Using the male plug from a PC hard drive
supply "Y" cable with the pins reisntalled so that black goes
(ground) and red goes to +30 and Yellow goes to orange (+5)
the diode cathode to the 4060 + 30 volt line the batteries
held up the
4060 for the 3 hours the mains power was down. Now I
the batteries and a charging them manually with a bench
The 12V 7 AH lead acid batteries are 3.75" from the bottom to
of the metal terminals. The distance from the bottom of
battery shelf to the bottom of the lid is about 3.75", so it
would be a
bad idea to try and close the lid with the batteries
there's an even more compelling reason to NOT put lead acid
in the same box as electronics. And that's because acid
from the lead acid battery will literally eat the traces off
printed circuit boards. So it's best if the batteries
are out of
It took about 2.7 AH to charge one of the 12 V 7 AH batteries
power outage was about 2.7 hours, so the FTS4060 is pulling
about 1 Amp.
But the 7 AH rating is for a 20 hour discharge (350 ma) so the
will not last 7 hours at 1 amp. I think the terminal
the end of the power outage was about 23.49 volts or 11.75
battery which is discharged. Maybe 2.7 AH is the
capacity at 1
When running from the batteries the Green Lock LED is on and
the red AC
Power Alarm LED is on as well as the red Battery LED.
position 6 still shows 0 because there is no charging current.
After AC power is restored the Power On Green and the red
are both on (press the reset button to clear the red alarm
The green lock LED is still on. No battery LEDs are
(remember the /S24 has no battery option.)
Dead New Batteries
At first one of the new batteries not only would not put out
voltage, but actually had reverse voltage across it and the
still running. This means that the switching supply
will keep the
4060 going on less than 12 volts (although it may or may not
cold 4060 on that low a voltage). The "bad" battery
like the good batteries when connected to the charging power
up a simple 12 Volt battery checker
that's just a number 1156 car tail light bulb soldered to a
that was cut in half. This pulls a couple of amps to
brightly and with only 1 amp will take some seconds to light
dimly. This works much better than the Radio Shack
tester that shows a dead battery as good.
Note that the very common 12 V 7 AH batteries come with both
(0.250") and 3/16" (0.187) quick connect type
terminals. On the
batteries I got some are 1/4" and some are 3/16". So
you need to
check each battery, even though at a quick glance they look
Under the top cover the brick power supply is on the
Beside it is a tray that could hold rechargeable
There's a 4 position Molex connector with 3 sockets installed
that are Red, Black and Orange marked "26" that's probably the
pack connector. There is a PCB behind the left metered
another PCB behind the
setup controls located behind the door on the right.
Chicago Miniature CMD series LED's.
Yellow CMD 53124A
The upper left box is marked Model 5030M/201/S25, s/n 199,
Autolock for resonators for frequency standards Feb 12, 1985
A system is disclosed for examining the response
and molecular resonators to identify and select the maximum
peak and the voltage used to cause said peak to be produced.
is fabricated of modular elements electrically connected to
board to facilitate its construction and transportation with
resonator. A microprocessor is utilized to perform the
analysis and to
generate information to select the maximum resonant peak,
system includes means to compare the value of successively
resonator outputs and to select the output with the maximum
331 is Oscillators and /3 is Molecular resonance
The idea of this invention is to use a microcontroller (RCA
1802 CMOS) to sweep the control voltage to the 10 MHz OCXO
range and watch the CBT output peaks and valleys. By
a peak with aproximate equal valued adajacent valleys on
both sides the
maximum peak can be selected and that peak used to lock the
system tying to CBT to the 10 MHz reference. The three
to Analog converter chips that are part of the A/D system
output voltage is potted in a clear compound probably to
currents. J2 is the OCXO control voltage output.
Atomic beam tube
June 29, 1976 250/251; 331/3; 331/94.1 by FTS (3967115.pdf
Other FTS Patents
Digital frequency generation in atomic frequency
using digital phase shifting February 24, 1998 331/3
Methods and apparatus for digital frequency
atomic frequency standards February 3, 1998 331/3; 331/94.1
Heater controller for atomic frequency standards
Resonator package for atomic frequency standard
May 6, 1997
System for producing spectrally pure optical
August 29, 1989 359/345
Adjustable crystal oscillator with acceleration
compensation May 13, 1986 331/156
Crystal oscillator assembly April 29, 1986 331/69
4899117 High accuracy frequency standard and clock system,
Vig; John R,
Feb 6, 1990, 331/3 ; 331/176; 331/44; 331/47; 368/202; 368/56
"Moreover, in rubidium
standards, the available C-field
adjustment range limits the useful life of the unit. For
one of the most popular rubidium frequency standards
currently on the
market, the manufacturer provides a C-field adjustment range
to +1.5.times.10.sup.-9. The aging rate of the standard is
2.times.10.sup.-10 per year. Consequently, at the specified
the limited C-field adjustment range
limits the useful life of this rubidium frequency standard
1.5.times.10.sup.-9 /2.times.10.sup.-10 =7.5 years."
5146184 Atomic clock system with improved
331/3 ; 331/79
Inside the 4060/S24
J3 is the signal coming from the
PCB of the CBT.
J4 is the
450 Hz output to the A7 X18 multiplier.
There is a 40 conductor ribbon cable connection, 2 coax cables
cable with 2
wires (Red and Black) going to A7 and TP2.
2:50 power on TP2 = 4.97 VDC and the front panel meter on 4
voltage) indicates about 5 volts. (2:50 pm)
2:56 Operation Monitor light turned off.
3:00 switching LOOP to Open and back to closed starts meter
sawtooth from 0 to 5 Volts. It takes about 21 seconds to sweep
monitor voltage from 0 to 5 Volts.
But TP2 is sitting at 4.98 VDC so must be a 5 Volt test point
some logic indicator that may be pointing to a problem.
21 July 2005 - A3-TP2 a test point to monitor the 450 Hz
The Cesium Beam Tube is on the right, marked: Cesium Beam
FTS-7103, p/n 08923-501, NSN 5960-01-214-7475.
To the left of the CBT at the rear is the 10 MHz oscillaotr,
Model 1000B. In front of the 10 MHz osc. is The A5
Amp metal box with an
SMA-f connector just behind the front panel marked J3, RF1
which may be
a 10 MHz signal that could be connected to the front or rear
divider is the A3 Alarm PCB
2 each DB-25 connectors and no RF coax connections.
D.1652 s/n 865009 (probably 1986 +...) It is not
fully populated, missing a few ICs and a number of discrete
parts that probably are part of the battery charging or
circuit. The 30 VDC brick power supply is up aginst the
I have named the DB-25m connector nearer the powr supply A3J1
DB-25m near the center divider A3J2 since there's no markings
A3J1 pins 23, 22, 24, 25, 2 and 6 are connected to the Monitor
wheel switch positions 1 thorough 6 respectively and the
goes through the front panel meter to ground.
The Battery Charge, AC Power Alrarm, Battery On andAC Power On
connected to A3J1 pins 4, 5, 6 and 9 respectively.
Five of the wires on A3J2 are connected to the 5030 Assembly
A3J1 pins 1, 13 and 18 and connected to A3J2 pins 1,2,5,6,8,13
The Physics package might be
the combination of the Cesium beam tube, the Times 51
the Interface PCB since the latter two items are bolted to the
the Cesium Beam Tube.
The Physics Package is in turn a part of the 5030
addition to the Physics Package the 5030 assembly has Most of
except the PS1 30 Volt power supply and the A3 Alarm 5 x 6"
the front and
rear panels. The 5030 Assembly is 16 x 7.75 x 5 inches.
four 5/32" hex cap bolts, being
careful to not let the 5030 assembly crash and move it so that
easily get to the SMA connectors and the #2 Philips screws on
Check to see that the 3 Coax cables are marked 4 (Zeeman audio
(Rear 1 MHz out) and 7(Front 1 MHz out) that mate to
J4, J5 and J7, then disconnect these SMA cables.
Remove the two "D" connectors using a #2 Philips screwdriver
the 5030 Assembly free of the chassis.
Note It is an easy job to
5030 Assembly and that may allow using the complete 5030
/S24 unit to bring a defunct FTS4060 back on line.
This can be
done in a few minutes. But I don't know where the
modules are located on a full featured 4060. If they
are on the
right side ( the 5030 is on the left side) then it would be
easy. If they are in the way of removing the 5030
it would take longer.
10 MHz OCXO.
A1A5 Dist AMP
At the upper right is the A5
Distribution Amplifier. This amy be an A5/S24 with the
Mhz output missing.
A1A5 & A1A7 Sub Assembly
cable ends that will be disconnected, then
by removing 2 (+) screws and loosening 2 (-) captive
disconnecting some connectors (no soldering needed) the
& A7 assembly can easily be removed.
A1A5 Distribution Amp
units have an A5 amplifier that has a open SMA-f connector
forward and that connector has a 10 MHz signal that's
about 4.4 Volts
Pk-Pk. But other /S24 units have the connector and
parts removed and so don't have the 10 MHz easy to
The cable from the A5 10 Mhz output to the rear panel is
long, SMA(m) on the A5 end and a bulkhead BNC(f) for the
A1A2 Mother Board
seen once the A5+A7 sub assembly is
removed. All the components in the 5030 assembly
interface to the
mother board. This greatly minimizes the wiring
There may be a dozen components on the mother board.
Max dimensions are about 12" x 5" although it's "L"
the A1A9 input filter at attached to the
The right hand narrow part is just to get the 40 conductor
lined up with the A1A3 uP board.
A1A6 Ultra-stable Oscillator
). This is a brick about 3x3x7inches with
connections on one of the 3x3" ends. Part number is
05818-119. There's a DB-9 connector
with the following pinout:
Voltage in (-10 to +10)
(coarse tune hot)
The oven insulation is my means of a dewar. The
rate might be <1E-10 per day when new, but can get
below 1E-12 after
running for some time. The phase noise is lower than
-134 dB at
10 Hz, -144 dB at 100 Hz and -157 dB at 1 kHz.
The 10 Mhz output is from a right angle SMB connector
down. (All the small coax is terminated with 50 Ohm
connectors in the FTS4060).
On top of the 1000B (p/n 05818-119) there's a coarse
To remove the USO three 1/4" nuts need to be removed that
are below the
A3 uP board and the connectors disconnected.
on a hand picked 100B:
A1A7 x18 Mult & Mixer
the Times 18
Frequency Multiplier (10 MHz in, 180 MHz out) and
As seen in the photo the connectors are: 10 MHz in,
with Black, Gnd, and Red wires going to J4 on
the A3 Microprocessor board. Cable with Black, Red,
(ground) & blue wires soldered to feedthroughs going to
J4 on Cesium Beam Tube motherboard.. 12.6 MHz input &
A1A8 Cs Power Supply
A8 Power Supply for the Cesium Beam Tube that
includes the two HIGH VOLTAGE outputs. The bottom of
this PCB is
the top left front when the top cover is removed. You
want to have your hands anywhere near this board when power
A1A9 DC Input Filter
and should be able to slide out, it's trapped by
the female thread fitting used to attach the 5030 sub
assembly to the
chassis. It has a dual electrolytic cap, a single
cap, a diode and an inductor.
A2 RF Assebmly
3 x 7" A2/S24 RF Assembly (56219-05280-011
Assy 05281) that takes in 10 MHz and outputs 1 MHz. On
featured 4060 this board would also output 100 kHz and 10
may that the 74LS90 divide by 10 circuit could be
10 MHz outputs instead of the two 1 MHz outputs that are on
A1A3 Micro Processor Board
input (J2) that takes in the error signal from the
Cs interface board. It also has a coax EFT output
(J3) to drive
the Ultra-stable 10 MHz oscillator (A6). The 450 Hz
generated on this board and feeds A7. 40 pin
connector J1 has a
number of analog signal inputs and outputs as will as
With J1 pointing up in the photo at left the two TO-5 cans
in the upper
left are the +15 and -15 volt supplies for the analog Op
sample/hold and DAC circuits comprsing the left analog
part fo the
The uP is an 1802.
DAC1006 D/A converters are used both for A/D conversion
comparator and for D/A output to the meter and
A1A4 12.6 MHz Synthesizer
same size board, the A1A4 12.6 MHz
Synthesizer (56219 Assy) that shares the same 14 conductor
and has a single coax cable that goes to the times 18
Multiplier. There are a half dozen Synchronous
ICs on this board.
A1A1 Cesium Beam Tube Assembly (Physics Package)
the photo gives it a curved appearance
because of perspective.
A1 Cesium beam tube is held at each
end by an angle bracket that has 3 large philips screws
holding it to
the 5030 frame. One of these is under the A8 power
supply and the
other is under the A3 uP board.
The x51 Multiplier and the Cesium Beam Tube Interface PCB
to the tube.
Connections to the rest of the system are by means of:
Red & white High voltage wires, twisted pair of orange
wires to Cs
coax with 180 MHz from A7 to X51 mult.
26 conductor ribbon cable to mother board W5
Coax with error signal from interface board to A3 uP
wires coming out of the CBT are
soldered. The E26 Test Point is missing. The
from theDetector Heater are labeled along with the "E"
number of the
X51 Microwave Multiplier
The x51 Microwave multiplier gets it's RF input from the
Multiplier - Mixer and feeds it's output to the waveguide
the CBT. The Red, Green and Black wires come from
PCB Rails but no Card Edge Connectors
The PCBs are held by rails, but there are no card edge
the PCBs. All connections are made by coldered wires,
connectors (typcially standard 50 Ohm SMB), or rectangular
connectors. There is an unused pair of rails above the
microprocessor PCB, but if another board is used there it
would ned to
have notches to clear the to coax connectors coming from the
Manual Control Voltage & Loop Gain
In Appendix A of the Operation
it describes how to manually set the Control Voltage and Loop
Gain. The symptom indicating that this needs to be done
when the Monitor is set to 4, Control Voltage, the needle
ramps up and
jumps down and this is repeated over and over. This was
symptom my unit had so I manually adjusted these two settings
follows after over 30 minutes of warm up:
- Open door and turn off "MOD" and set "LOOP" to OPEN,
- At the same time press "Align" (behind the door) and
Alarm" (next to the Monitor LED). This stops the
- Set the "Manual Scan" switch (behind the door) to
Voltage" and set the Monitor switch to 4 (control Voltage)
and use the
"Manual Scan Increase/Decrease" switch to center the
- Set the "Manual Scan" switch (behind the door) to to
and set the Monitor switch to 3 (Beam) and and use the
Increase/Decrease" switch to center the meter.
- Switch the "MOD" and "LOOP" switches back to On and
In my case this caused the Lock LED to turn on and stay
26 Feb 2006 - Note after
the C field set on s/n 1013, there was a power failure
lasting about 2
seconds. But 1013 had been running for many months and
working well. After the power came back on the lock
LED did not
light. After giving it about 4 hours still with no
the above procedure was used to set the control voltage and
and afterwards the lock LED turned on as soon as the closed
mod on switches were thrown.
28 Apr 2006 - You really can only center the Control Voltage
front panel controls (as described above in the OPEN
there's a continuous search or the yellow monitor LED is on
green LOCK LED, then it's time to adjust the coarse
frequency of the
crystal oscillator. This can be done with the source
locked. Remove the bottom cover and with the monitor
position 4 (CONTROL) note the reading. In my case it
3. Then adjust the coarse pot on the USO to bring the
little to the other side of 2.5, in my case to 1.6.
control voltage gets closer to 2.5 the yellow monitor LED
will go out
leaving only the two green LED for LOCK and AC power.
Hint: leave the door slightly open by turning the screw all
out, closing the door and then turning the screw 1/4
way if the ALIGN lamp turns on you can see it. You
the door open, but for me it's in the way of other stuff.
Manual Loop Gain
PCB there are 2 10-turn pots. The one
close the the SMB connector is R9 and should not be
one next to R9 is R5 and is the manual beam current
should be set so that the voltage between E23 (ground) and E26
point) is 1.8 +/- 0.2 VDC. But I can't find E26 on my
Where is it?
Also note in the photo there is a small screwdriver adjustment
of the 10 MHz OCXO that has been relocated from the front so
do NOT need to remove the 5030 Assembly to get access to this
Serial Number 1013 has had the Green Lock LED on for about 3
hours. If the line cord is removed and plugged back in
seconds the unit Lock LED turnes on in about 10 minutes.
The manual implies that you
the loop gain or beam current adjustments on the fly.
But on the
two occasions that I have tried to do that the FTS4060/S24
confused. The fix has been to pull the line plug for
seconds and restart. Without the restart the control
makes a good accessory. It
provides a UTC clock, 1 PPS output and a check on the
stability of the
FTS4060. Also if the reference input fails the 2100T
lock and need to be manually re started and so it's a good
the output of the reference source.
If you have one of the /S24
you tell me if you have the 10 MHz output and your serial
If you have brought a unit back to life tell me what you