[time-nuts] Testing frequency using NTP Bruce GPS ps

Mike Monett XDE-L2G3 at myamail.com
Mon Oct 13 03:56:31 UTC 2008


  "Tom Van Baak" <tvb at LeapSecond.com> wrote:

  >Mike,

  > Comments below.

  >> I have to learn more about how you do your measurements. A 53132A
  >> is way  out  of  my price range at the moment. But  I  do  have a
  >> 53310A which should give comparable results.

  > Yes, if  you can get a clean series of 1PPS  sub-ns  time interval
  > measurements over GPIB from your 53310A then you're all set.

  No luck. It is very good for looking at PLL loop dynamics,  which is
  the reason I bought it.

  But it doesn't seem to be useful for Allan variance, so I'll have to
  make my own system. I'll use the digital mixer I posted earlier.

  > What time/frequency reference do you have available  against which
  > to compare your GPS receiver?

  I ordered a rubidium, two OCXO's, and two Oncore UT Plus GPS cards.

  I plan on making a standard 1PPS phase/frequency detector to lock to
  the 1PPS.

  The second version will use the same model GPS card, but it will use
  a new method of locking to the 1PPS.

  I'll compare  the  results with the rubidium  and  OCXO's  locked to
  either GPS.

  >> I am  interested in learning more about these errors. Is  there a
  >> list somewhere that describes the sources of error, and  what can
  >> be done to reduce it? I keep coming across hints of doing this in
  >> software, but  it  would seem that has to be done  in  the signal
  >> processing part  of  the receiver. But we  don't  have  access to
  >> that. Is  there  anything else we can do to help  remove  some of
  >> these errors?

  > This is  a  large topic. Google for words:  gps  error  budget and
  > you'll begin  your  long   journey.   Also  look  for  topics like
  > ionosphere multipath  troposphere  ephemeris.  A  typical  OEM GPS
  > receiver board  will  get you within 100 ns  of  UTC  without much
  > effort. Here's a couple of pages to get you started:

  >http://edu-observatory.org/gps/gps_accuracy.html
  >http://www.trimble.com/gps/howgps-error2.shtml
  >http://www.cmtinc.com/gpsbook/chap6.html
  >http://www.ias.ac.in/currsci/oct102007/927.pdf
  >http://www.usace.army.mil/publications/eng-manuals/em1110-1-1003/c-4.pdf
 
>http://seismo.berkeley.edu/~battag/GAMITwrkshp/lecturenotes/unit1/unit1.html

  Thanks very  much  for  the links.  That  information  will  be very
  helpful.

  [...]

  > Your GPS  receiver  outputs a correction  value  in  fixed integer
  > units of ns; no less. Other words loosely used on this context are
  > resolution or quantum or precision.

  Now I am confused. I thought it used the internal crystal  clock for
  the timing,  and  selected which pulse to output for  the  1PPS, and
  this is what causes the sawtooth. If so, the timing is determined by
  the crystal.

  [...]

  > When you write your software to correct sawtooth errors you'll see
  > where GPS or TIC sign errors and off-by-one sample errors can bite
  > you. Not all receivers do it the same way, either, so make sure to
  > study the manual(s).

  I don't use software to correct the sawtooth. It is a whole  new way
  of locking to the 1PPS.

  A lot of the errors will simply vanish. But I'll still be vulnerable
  to ionosphere  delay, constellation changes, and pigeons  sitting on
  the antenna.

  >/tvb

  Thanks for your help!

  Best Regards,

  Mike Monett



More information about the time-nuts mailing list