Help needed to setup project

[Yohnny Lim]

Hi,

I am new to this software, but I have experience using MicroSurvey Star*Net.

I tried converting a small survey data from total station into the SALSA .lsa data format and run the adjustment, but I am getting this error below:


I've attached the .proj file here, appreciate if anybody is able to point out the errors in the .proj file.

Thanks in advance.

[Yohnny Lim]

Attached the adjustment log as a reference, 

17:43:07.030 lsapreprocessor starting... COMPLETED

lsasolver, Ver. 0.6.0 1/11/16, SALSA Ver. 1.15.0 Aug 15 2022 10:25:59, Run 2025/01/08 at 17:43:07

Output directed to log file C:/Users/YohnnyLim/Desktop/LSA/survey.out

lsasolver is terminating with code 18: Unable to open geoid file C:\Program Files\SALSA\bin/../data/geoid/egm2008_2.5m.und

lsasolver timing: 0.002 seconds. (0.000 sec)

17:43:07.308 lsapost starting... COMPLETED

ERROR - INVALID LATITUDE LIMITS - LATITUDES MUST BE IN RANGE [-90.000000, 90.000000]

lsapost unable to find .bin file output from solver.

ERRORS ENCOUNTERED:

ERROR - INVALID LATITUDE LIMITS - LATITUDES MUST BE IN RANGE [-90.000000, 90.000000]

--- Wed Jan 8 17:43:07 2025 -------------------------------------------------------------

there's another error message saying unable to open the geoid file.

[Clark Hughes]

Hi Yohnny,

Thank you for reaching out. Taking a quick look at your project file, I see you're using the POSC record to specify positions for the project. POSC stands for "Position, Cartesian" and the corresponding 3-element vector is specified as [X, Y, Z] relative to the reference frame origin, which is the center of the earth. So, for example

     5141.1420 1098.5030 54.2740 m

is going to be a few km from the earth's center and probably not what you intended.

For reference: https://en.wikipedia.org/wiki/Earth-centered,_Earth-fixed_coordinate_system

Without digging deeper I'm not sure whether the latitude error leads to the geoid error or vice-versa, but either way, both are likely stemming from the project's points all being near the center of the earth and probably spanning the poles.... For this project I'll suggest using POSG (Position, Geodetic) and specify your known positions in lat/lon/ht. At that point, the map should give you a sanity check that the points are in the right place, and you can carry on from there.

Best,

Clark

[Yohnny Lim]

Thanks for the quick response Clark, however our reference points are all in local projections (meters). Is there a way I can use POSG with meters coordinates?

Thanks.

Yohnny

[Clark Hughes]

Hi  Yohnny,

The short answer is no. POSG coordinates are expressed a [Lat Lon Ht] (polar coordinates) where lat/lon are angles of course....

Maybe you can back up and share what you're trying to do. Specifically, where did these fixed coordinates come from? Are they derived from GPS or some other measurement? If so it would seem you'd want to include those measurements (and associated uncertainty) in SALSA as opposed to asserting the positions as fixed. At any rate, if you can share the provenance of those points' coordinates it may help us give you some better guidance.

Clark

[Yohnny Lim]

These coordinates were arbitrary coordinates used locally for a monitoring setup, its not based on any coordinate system, I see we can use NONE on the geoid, but when I try to adjust the network the SALSA software crashes.

[Clark Hughes]

Ok, I understand that the points are not based on any [conventional] coordinate system but unless you literally pulled them out of thin air, they must be based on somebody's measurements? Again, any insight may allow us to give you some better guidance.

Regarding the geoid, I agree that 'none' may be appropriate if you're not doing geodetic work, but the reason SALSA is failing is because the points in the network are nonphysical, i.e., wrapping around the center of the earth.

Regards,

Clark

[Yohnny Lim]

Hi Clark, I just verified with the surveyor, their coordinates are based on GDM2000 (Malaysia). Can we use GDM2000 coordinate system to process the data?

[Clark Hughes]

Hi Yohnny,

That info should help. Let me play with that project when I get home this evening; I think there's a not-too-complicated way to work with those coordinates. In the mean time, you might try looking for a GDM2000 to (and from) WGS 84 converter. If you have the capability to convert to/from WGS 84 that would be ideal.

Clark