[Therion] Therion Survex Loop Closure Anomoly

[Bruce]

I have a mystery, whereby an overland survey between two fixed points gives
a correct pdf plan output with;

 

loop-closure survex

 

(I know it is 'correct' because the aerial photograph image matches the
features surveyed) but if I specify;

 

loop-closure therion

 

then one of the fixed points is about 150m misplaced in plan and the
surveyed features are clearly in the wrong place.

 

In both scenarios therion.log reports identical average loop errors and
specific loop error statistics.  The error over a 500m loop is about 2.3%,
mainly due to a 12m error in elevation (can not trust those gps readings).
The plan error if one of the points has its 'fixity' removed is about 4m,
and the plots produced are correct regardless of the loop closure method.

 

It is a rather large dataset with many underground loops, and four fixed
surface points.  Two of which have overland surveys between them.  Only one
of the overland surveys has this problem.

Using Therion 5.3.5.  Can replicate this result on three computers.

 

Any theories as to what might be going on and how to locate the issue?

 

Bruce

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20101115/8d32e7f0/attachment.htm>