for the purposes of our psn uses (plotting quake location etc) what is
the error budget for timing?  I've never seen this discussed.  



There must be a fundamental limit caused by the variation in travel
times through the earth.  Ie, if the same quake happened twice some
thousands of miles away, what would the typical variation in travel
time through the earth be?



Another limit will be the algorthms used to calculate the distance from
the S and P waves.  Also, the accuracy on our traces with which we
place the S and P pointers.  I suspect that this is the largest source
of error and may be the equivalent of many many seconds.



Ian Smith



Larry Cochrane wrote:


  
The time stamping of the data is preformed on the A/D board not on the
computer running WinSDR. If you want very accurate timing, down to a few
milliseconds, you will need to use GPS. The GPS receiver connects up to the
A/D board so it can uses the highly accuracy 1 pulse per second signal
generated by the GPS receiver to time stamp the data. See
http://www.seismicnet.com/gps/index.html for a GPS timing system that will
work with my A/D board.

You can use the computer running WinSDR to supply timing information to the
A/D board. In the WinSDR system settings dialog box select "Local" timing
reference type. When this timing mode is selected the A/D board interrogates
the computer for the current time of day.

I found that the NT port of NTP works very well under Win2K. If you have a
DSL line, and can find a good time server that is near by, you can achieve a
timing accuracy of around 10 to 20 milliseconds. While searching around I
found this Windows port of NTP:
http://home.att.net/~Tom.Horsley/ntptime.html

Larry Cochrane
Redwood City, PSN


----- Original Message ----- 
From: "tdick" <dickthomas01@.............>
To: <psn-l@..............>
Cc: "Kevin Bach" <bachcottage@...........>
Sent: Wednesday, April 14, 2004 6:52 PM
Subject: computer time sync


  
  

    
I am using Dimension 4 to sync the time on the computer that runs only
WinSDR. My time seems to vary. In an event yesterday, St. Louis network
    
  
  
and
  
  

    
my computer were 20 seconds off for the same event. Any suggestions?


__________________________________________________________

Public Seismic Network Mailing List (PSN-L)

To leave this list email PSN-L-REQUEST@.............. with
the body of the message (first line only): unsubscribe
See http://www.seismicnet.com/maillist.html for more information.

    
  
  

__________________________________________________________

Public Seismic Network Mailing List (PSN-L)

To leave this list email PSN-L-REQUEST@.............. with 
the body of the message (first line only): unsubscribe
See http://www.seismicnet.com/maillist.html for more information.