Low Cost, High Accuracy GPS Timing

1
Its About Time !!!!!
2
Clock Calibration for VLBI2010Portions have
been adapted from Timing for VLBI presented
at IVS TOW MeetingHaystack May 9-12, 2005

• Tom Clark
• NASA/GSFC NVI
• mailto K3IO_at_verizon.net

VLBI2010 Working Group Haystack Sept 15, 2006
3
Oscillators and Clocks
There is a difference between Frequency and Time

• The Oscillator
• Pendulum
• Escapement Wheel
• Crystal Oscillator
• Oscillator Locked to Atomic Transition
• Rubidium (_at_ 6.8 GHz)
• Cesium (_at_ 9.1 GHz)
• Hydrogen Masers (_at_ 1.4 GHz)

Events that occur with a defined nsec -- minutes
FREQUENCY

• The Integrator Display Clocks
• Gears
• Electronic Counters
• Time transferred from outside (GPS)
• The Rotating Earth (i.e. UT1 sundials)

Long-Term seconds - years
TIMING
4
What Timing Performance Does VLBI Need?

• The VLBI community (Radio Astronomy and Geodesy)
  uses Hydrogen Masers at 40-50 remote sites all
  around the world. To achieve 10 signal
  coherence for 1000 seconds at 10 GHz we need the
  two oscillators at the ends of the interferometer
  to maintain relative stability of ?
  10/(360?1010Hz?103sec) ? 2.8?10-15 _at_ 1000 sec
  
• In Geodetic applications, the station clocks are
  modeled at relative levels 30 psec over a day ?
  30?10-12/86400 sec ? 3.5?10-16 _at_ 1 day
• To correlate data acquired at 16Mb/s, station
  timing at relative levels 50 nsec or better is
  needed. After a few days of inactivity, this
  requires ? 50?10-9/ 106 sec ? 5?10-14 _at_ 106 sec
  
• Since VLBI defines UT1-UTC , we need to
  control the accuracy of our knowledge of
  UTC(USNO) - UTC(VLBI) to 100 nsec or
  better.

A
B
C
5
The Allan Variance A graphical look at clock
performance
FREQUENCY
TIME
C
A
B
6
Why do we need to worry about Absolute Time
(i.e. Accuracy) in VLBI?

• To get the correlators to line up for efficient
  processing, the relative time between stations
  should be known to 100 nsec.
• In the past, geodetic and astronomical VLBI data
  processing has been done by fitting data with
  station clock polynomials over a day of
  observing, and then discarding these results as
  nuisance parameters or instrumental constants
  that are not needed for determining baseline
  lengths, source structure, etc.
• The uncalibrated and unknown offsets now range
  from 1-10 ?sec at many VLBI stations.
• If VLBI2010 is to produce accurate UT1 as a
  major data product, then absolute clocks need
  to be a fundamental design consideration.

7
Why do we need to worry about Absolute Time
(i.e. Accuracy) in VLBI?

• The MAIN reason for worrying about absolute
  time is to relate the position of the earth to
  the position of the sun, planets stars
• Generating Sidereal Time to point antennas
  (especially big arrays, including VLBI!).
• Measuring UT1(i.e. Sundial Time), Nutation
  Precession to observe changes due to
  redistribution of mass in/on the earth over long
  periods of time.
• Knowing the position of the earth with respect
  to the moon planets to support interplanetary
  navigation.
• To improve the accuracy of GPS/GALILEO/GLONASS
  navigation
• etc . . . . . .

8
Why do we need to worry about Absolute Time
(i.e. Accuracy) in VLBI?

• At the stations this means that we will need to
  pay more attention to timing elements like
• Frequency Standard and Station Timing, including
  changes within a one-day experiment.
• The lengths of all cables in the signal timing
  paths.
• The geometry of the feed/receiver to the antenna,
  including deformation with pointing
  temperature.
• Calibration of instrumental delays inside the
  receiver and backend. The development of new
  instrumentation is needed.
• The care with which system changes are reported
  to the correlators and the data analysts.

9
VLBI Analysis assumes the intersection of axes as
the fundamental reference point.
VLBIs REAL Clocks (1) Fundamental
reference point, Geometry Cables
The Real Signal Path
Remember the lengths of the red - - - cables
contribute to Clock (1)
10
VLBIs REAL Clocks (2) The Microwave IF
Signal Path Phase Calibrator
CONTROL ROOM
H-Maser
ON ANTENNA
UP
Phase Cal Ground Unit Monitors Cable Length
Changes -- UP Down
Cable Length Transponder
5 MHz
DOWN
5 MHz
Divide by 5
Phase Cal Counter
IF Signals to Control Room
1 MHz
Quasar
Pulse Generator
This is the Phase Cal data clock that is used
to analyze VLBI data. Note The 1/?sec pulse has
a 200 nsec ambiguity because of ?5 stage.
1 Pulse/?sec
Microwave Receiver
Remember the length of every red cable and the
properties of every red box contributes to Clock
(2)
11
VLBIs REAL Clocks (3) Converting IF Signals
into Bits
This is the clock that the correlator uses to
make fringes
H-Maser
IF From Microwave Receiver
5 MHz
5 MHz
Formatter Clock
IF Distributor
Video Converter
Clipper/ Sampler
Recorder
Remember the length of every red cable and the
properties of every red box contributes to Clock
(3)
12
VLBIs REAL Clocks (4) Synchronizing the bits
with GPS to establish UTCVLBI minus
UTCUSNO
H-Maser
5 MHz
1 PPS
GPS Constellation
Formatter Clock
Counter 1
Counter 2
GPS Antenna
1 PPS
Initial Sync
GPS TIMING CLOCK (like my TAC)
Remember the length of every red cable the
properties of every red box contributes to Clock
(4)
13
For the VLBI2010 Era

• IMHO, We must insure that all four of these
  different types of clocks used by VLBI are
  calibrated throughout the data acquisition,
  correlation and data processing chain at every
  station. These clocks need to be harmonized at
  the 100 nsec level at each station.
• This will allow VLBI2010 to be a reliable source
  of UT1 at the (hopefully) sub-?sec level, free
  from biases and long-term drifts with no
  network-to-network day-to-day mismatch seams.
  

14
One Possible Solution Calibrate small antennas
very accurately then use them to transfer
calibration to the more difficult
stationsThis is taken from the latest Japanese
IVS NICT-TDC NEWS No.27 _at_ http//www.nict.go.jp/w
/w114/stsi/ivstdc/news_27/pdf/tdcnews_27.pdf