[IGSMAIL-3741]: IGS Timescales

[Ken Senior]

******************************************************************************
IGS Electronic Mail      15 Feb 13:17:40 PST 2002      Message Number 3741
******************************************************************************

Author:  Ken Senior

Dear Colleagues,

With reference to the new IGS clock products described in IGS Mails #3470
and #3615, we have made several more changes to the system which generates
the Rapid (IGRT) and Final (IGST) IGS timescales.  The complete set of
plots and .clk product files have been regenerated for each timescale from
the beginning of receiver clock availability (late 2000).  We expect 
this to
be the last major enhancement to the system.  Any future changes should not
require replacement of timescale data already posted.

The changes implemented are:

* The upper limit of weights each clock can carry in the ensemble is now 
set
 nominally to the maximum of {0.1, 2.5/# masers, 2.5/# clocks}.  This 
change
 was necessary particularly for the Rapid timescale IGRT which has suffered
 from a number of data dropouts (sometimes resulting in only 1 or 2 masers
 available for certain days).

* By request, Allan deviation plots are now included for all clocks 
(including
 crystals).

* Chi-Square statistics using the appropriate equivalent degrees of freedom
 for overlapping Allan deviation computations are now performed for the
 calculation of confidence limits in the Allan deviation plots.

Results are available at
  http://clockdev.usno.navy.mil/igst    and
  ftp://clockdev.usno.navy.mil/igst

Regards,
Ken Senior

Time Service Department--Clock Development
U.S. Naval Observatory