STAR*NET V6 - Circe

Chapter 8 Analysis of Adjustment Output
The answer to a non-convergence problem may be as easy as simply increasing the
number of iterations. Because of the geometry of the network, or possibly because of
some poor initial estimates of approximate coordinates, some networks just take a larger
number of iterations to converge. If the adjustment still fails to converge, even after
increasing the iterations to a much larger number, say 20, there is likely some problem in
your data which needs to be found and fixed.
8.10 Coordinate Changes from Entered Provisionals (Optional)
STAR*NET optionally lists differences between hand-entered provisional coordinates
and the final adjusted coordinates. This option is particularly useful in deformation
studies where the purpose is to determine how much specific stations moved since the
previous survey.
To do this, simply enter the specific adjusted coordinates from your previous survey as
provisional (approximate) coordinates for a new survey, and request this optional listing
section. Only changes from these hand-entered provisional coordinates will be shown in
this output listing section, not the remaining approximate coordinates computed by the
program.
Coordinate Changes from Entered Provisionals (US Feet)
Station dN dE dZ
319 0.0000 0.0000 0.0000
46 -0.5022 -0.2278 -1.6420
552 -1.0064 0.9188 0.9973
559 -1.0477 0.7678 0.9115
560 0.2513 0.4325 0.2234
Coordinate Changes from Entered Provisionals (3D Example)
When entering provisional coordinates to be used as the basis for computing the changes
as described above, it is important to place these coordinate values at the beginning of
your data so that they will be read “first” during the processing.
Why is this? When STAR*NET reads data and finds more than one set of approximate
coordinates for a particular station, only the “first” coordinate values are remembered.
The remaining are ignored. Therefore, placing your provisional coordinates first in the
data assures you that some other approximate coordinates may be already existing later
in the data will not corrupt your desired results.
You might consider creating your “provisional” coordinates in a separate file, and
placing this file as the first entry in your data file list. Then at some future time when you
want to update the provisional coordinates for a subsequent run, simply replace that
file’s contents with the newer provisional coordinates.
126