STAR*NET Users with Versions of SURV CE or SURC PC which are older than release 2.62 may discover that the Positional Tolerances section in their listing is reporting that many points are failing the test even though every other aspect of the project appears to be well within tolerances.
This can be caused by an issue that has been discovered recently. Carlson SURVCE records the weighting of a vector measurement in the G2 and G3 Lines in the RW5 file. Depending on the receiver type and SURVCE release, the weighting may be recorded as either Covariance or as a Standard Error Correlation.
STAR*NET always inserts the inline option:
.GPS WEIGHT COVARIANCE
...at the top of the gps file when importing the vectors. If your equipment is recording the G2 and G3 lines as Standard Error Correlations it will be important that you replace the inline option with:
.GPS WEIGHT STDERRCORR
...in order for the vectors to be weighted appropriately.
Please check with Carlson support to confirm if your vectors are being reported as Standard Errors or Covariances. Carlson has reported that as of version 2.62 all vector weighting will be reported as covariances, but results may vary if you have an older version.
While the only way to be sure this is an issue is to ensure you are running SURV CE 2.62 or newer, the example shown below may help:
A typical vector file with the error correlation reported incorrectly G2 and G3 lines with noticably different values. These lines are typically shown in scientific notation so in the example below the values in the G3 line are in the order of 1 hundredth or one thousandth the value in the G2 line:
A typical vector file with the error correlation reported correctly has G2 and G3 lines with similar magnitudes:
James Johnston
Comments