Any Starnet users import Carlson CE RW5 files . Doing some testing and I found a few things that might be a bug , I am new to the software . When I import Carlson GPS vectors it strips the descriptions from the points, also appears the vector standard errors are not being read in correct and is causing large minor and major error ellipses , that are blown out of proportion .
Yes, we found the same thing and spent quite a bit of time with Microsurvey and Carlson. They confirmed this, but have not fixed it. If you change the header of the GPS file to show standard error, it makes the results appear correct.
I heard back from the Carlson support team, they have confirmed as we suspected, that the G2 line is expressing the GPS error as "Standard Error" the way that your TVG vectors are expressed. So changing the .gps weight inline IS a valid solution in your case.
Here was some more information about what they discovered, for future reference:
Hello,
In talking with the Development Team they found out in SurvCE in the RW5 raw file the only Receivers that were recording proper G2 variance values were Leica and Ashtech brands.
They have resolved this issue in our current SurvCE 2.80.xx BETA version so that all Receiver models store Variance into the G2 record of the RW5 raw file.
Since this customer is using a Topcon GPS receiver with Version 2.56 it appears all the G2 records are "Standard Deviation" rather then "Variance.
If we parse the "--SurvCE Version " in the RW5 file and its 2.80.21 or less we could then parse the "--Equipment: " field to see if it matches the Ashtech and Leica list below.
If its Version 2.80.21 or less and its not on the Ashtech or Leica list then we would need to convert the G2 records from Standard Deviation into Variance values for the output of the Ascii *.GPS file.
If its Version 2.80.22 or higher or if its an older Version but its showing to be a model from the list below then the G2 records should not be modified since they will already contain Variance values.
That would mean for example Version 2.07 or 2.09 or 2.50.22 or 2.56 or 2.61.4 would all be checked for Equipment type but version 2.80.35 or 3.00 or 3.02 would not be checked.
The G2 is the only vector record that was affected by this and we shouldn’t modify any other vector records. The G0,G1 and G3 records are valid and match the Specification that was provided for all versions and all equipment types.
Here is the Equipment values that are already properly using Variance in the G2 record in older SurvCE as they would appear in the “Equipment:” field in the RW5 raw file:
ProMark 500
ProFlex 500
Z-Max
Z-Xtreme
ZSurveyor
Z12 or Sensor
Leica Viva – GS10/GS15
Leica GPS1200 – Smart Antenna
Leica GPS1200 – GX1230
Leica SR530
Also be nice if they publicized this instead of me spending all weekend trying to figure out why I cannot get a decent solution . makes no sense at all, who testing the software before it get in front of the user .
What about the description stripping did they say they were going to fix it , hell thats a biggie . I guess I can insert into excel and fix , but why should I have to.
Also, be aware that when turning sets, the angle displayed in the final average in the Raw Data is the actual angle read (average).
So, if you turn three sets, and zero at the start, turn to say 90-00-00, and keep on turning sets, if your second set starts with the gun reading say 359-59-55, and your third set starts with the backsight reading at 359-59-50, you could very well have an average angle of 89-59-55, and if you look in the Raw Data you see your backsight at 359-59-55.
I dont know if there is any way to fix this, other than zeroing every time you sight the backsight (which Carlson doesn't like). It keeps me from using the field book...
If I am doing something in my settings to create this, please let me know, but I think at this point it is an actual "issue".
-David
