Error Ellipse Semi-major axis for traverses on state plane coordinates

So in doing some research, I think the problem may be that I'm not applying a scale factor to the job after setting up the GPS points.?ÿ Any thoughts?

If that's the case, then the proper setup would be:

1. establish two control points on state plane with the GPS equipment (points #1 and 2).?ÿ
2. Measure the grid distance between those points based on the point coordinates.
3. Measure the ground distance between the points?ÿwith the total station.
4. Scale Factor = Grid Distance / Ground Distance
5. Apply Scale Factor to the job
6. Begin the traverse

?ÿ

Any thoughts? Would that account for the large standard deviations I'm finding in my "passing" adjustment?

There is no Az line in your state plan input file, but there is one in the assumed coordinates file.?ÿ Try adding an azimuth for your intial backsight.

I think there's no AZ line in my state plane file because it ran the traverse setting up and backsighting known points.

The least squares adjustment needs more than a single NE coordinate for control.?ÿ The statistics are getting blasted because your other two coordinate pairs are completely free to adjust.?ÿ I don't think the Civil 3D adjustment program allows you to enter standard deviations for the coordinates like Star*Net does, so you have to put in an azimuth, either fixed or with a standard deviation, in order for the final error ellipse statistics to shake down correctly.

Ah ok, thanks. I'll give that a shot and will post the results. Thank you for your help.

Thank you Peter! That worked.?ÿ Edited FBK file and results attached.

Here's what I added to the fieldbook file for a calculated azimuth of 174-25'-31.5" from point 2 to point 1:

!Setup Date: 2018/02/27 19:54:58 Method: Known backsight point
!NEZ 2 3014439.0953 679349.5202 206.6129 "T."
STN 2 4.340
AZ 2 1 174.25315
!NEZ 1 3014339.9314 679359.2022 205.5210 "T."
BS 1 174.25247
F1 VA 1 174.25247 99.636 90.09235 "T."
!Delta horiz dist: 0.0003 Delta height: 0.0001

?ÿ

For some reason Civil 3D calculated the azimuth as "174-25-24.7".?ÿ I'm?ÿ not sure why what I entered in to the fieldbook file ends up about 7 seconds off but I edited the input file before running the adjustment and that seems to correct it.

I do not use Civil 3D for adjustments, I am a StarNet person.?ÿ The two adjustments had different ChiSquare Values.?ÿ I don't know how Civil 3D determines this value.?ÿ A ChiSquare confidence interval is determined by the probability value and the degrees of freedom.?ÿ If the surveys are identical (discounting the additional azimuth) you should have very similar statistics.?ÿ I don't know if this difference is indicating an issue but just check your pre-adjustment statistics setting, sometimes the defaults are not what you would use.?ÿ Just a thought.

Thanks. Yeah I don't understand the chi-square value in Civil 3D either. In StarNET it had to fall between the upper and lower bounds in order to pass and it was always a number around 1. I don't know how Civil 3D calculates it. I'll double check my inputs.?ÿ I actually had a discussion on Autodesk's message board about what the correct settings for my equipment: https://forums.autodesk.com/t5/autocad-civil-3d-forum/set-leica-ts15-16-in-equipment-database/m-p/7800307#M363839

?ÿ

I use a Leica TS-15/16 total station with Leica GPR-111 prisms.?ÿ Best I can tell, the civil 3D settings should be as follows:

UNITS

?ÿ

Distance: US FOOT

Angle: Degrees DMS (DDD.MMSSS)

?ÿ

ANGLE

?ÿ

Horizontal Collimation:?ÿ0.00001

Angle Type:?ÿRight?ÿ

Vertical Collimation: 0.00001

Vertical Angle Type:?ÿZenith

?ÿ

EDM

?ÿ

Refractive Index:?ÿ658?ÿ

Wave Constant:?ÿ105.45

Offset:?ÿ0

Measuring Device:?ÿScope

?ÿ

PRISM?ÿ(Leica GPR-111 prism)

?ÿ

Titlting:?ÿYES?ÿ

Prism Offset:?ÿ-34.40

Prism Constant:?ÿ-34.40

?ÿ

STANDARD?ÿDEVIATIONS

?ÿ

EDM Proportional Error:?ÿ1.50?ÿ

EDM Constant Error:?ÿ0.0010

Centering Error:?ÿ0.0015

Target Alignment Error:?ÿ0.0010

Pointing Error:?ÿ0.005?ÿ

Horizontal Circle: 0.00010

Directional Error:?ÿ0.00486

Vertical Error:?ÿ0.00486

Prism Height Error:?ÿ0.001524

Theodolite Height Error:?ÿ0.001524

For the Least Squares settings, I have these standard settings (would love some input):

Create Input File = YES

Input file name = ƒ??network nameƒ?

Network adjustment type = 3-Dimensional

Maximum number of iterations = 10

Coordinate convergence = 0.010000000

Confidence level = 95%

Perform Blunder Detection = Yes

Update Survey Database = No

So it appears that the ChiSquare value is the calculated value from the survey based on this equation,?ÿX² = v²p².?ÿ Civil 3D uses?ÿonly a one tailed (upper) test in a least square adjustment.?ÿ Your survey,?ÿChi Square Value : 46.14271.?ÿ From a ChiSquare table for 54 degrees of freedom at 5% probability (0.05) the critical upper value is 72.15.?ÿ So your ChiSquare value is below the critical value you pass the ChiSquare test.?ÿ ?ÿLike you said StarNet shows a interval, an upper and lower limit as it uses a two tailed test (upper and lower).?ÿ?ÿShould the Chi-Square test fail on the upper bound, it is assumed this failure is due to non-random errors, blunders, uncorrected systemic error, excessive residuals, or improper weighting.?ÿ Should the Chi-Square test fail on its lower bound it may be due to over optimistic error estimation or weighting.?ÿ So not utilizing the lower tail does not cause any harm, when I first started using StarNet it was an upper bound test.?ÿ?ÿ

See this link for the ChiSquare value explanation

https://knowledge.autodesk.com/support/autocad-civil-3d/learn-explore/caas/CloudHelp/cloudhelp/2016/ENU/Civil3D-UserGuide/files/GUID-19E35B77-72A1-4C3B-ABCC-4B3F26AB2DA7-htm.html

That's good to know.?ÿ Thank you for looking in to that.