[time-nuts] Cesium C Field Set?

John Ackermann N8UR jra at febo.com
Fri Feb 11 12:31:16 EST 2005

Brooke Clarke wrote:

> John: do you get more accuracy (or just more resolution) by using the 
> external chart recorder output?

I've found the frequency error reading from the Austron to be unreliable 
at low offsets -- I suspect because it's being averaged over a 
relatively short time (see below).

> Anyone:  What time period is used in the Austron 2100x LORAN-C receiver 
> for the frequency offset display?  I know the actual offset time 
> interval is zeroed at start up or can be zeroed manually, but it's not 
> clear to me how the offset value is calculated.

The Austron output (both chart recorder and frequency error meter) have 
selectable averaging from 200GRI to 3200GRI.  I don't have the manual 
here at work to get the exact command, but you can set the averaging. 
I'm now using the shortest average as the multiple effect of long GRI 
averaging plus reading averaging was causing some odd plots.

One problem with Loran is that you do get some one-time jumps presumably 
because of steering at the Loran station.  So it's sort of like WWVB in 
that you have to eyeball the chart to be certain the results make sense. 
  But that said, it seems that after several days of readings the linear 
offset between GPS and Loran seem to converge at the e13 level.  I'm 
still tweaking to get into the 14s, so I don't have an answer on that 
part yet.  (There seems to be some backlash or something in the C field 
pot on my 5061A; I made a minor adjustment that should have increased 
the frequency, but the negative offset got greater!)

> Anyone:  When making a time interval mesurement what effect does an 
> error in the counter time base have on the result?

I asked this question a few weeks ago and the answers I got are that as 
long as the absolute time interval between start and stop is small -- in 
the few us range -- the stability of the counter timebase won't matter 
as its impact is imperceptible over such short intervals.  But if you 
get up into the milliseconds, it could have an impact.


More information about the time-nuts mailing list