I have had success entering old field notes by Jim Conkright (from our files) into StarNet.
Most of the data is doubled-angles right and he closed the horizon. This latest project it looks like he used a 10" theodolite (or maybe a total station, it is dated in 1984). He balanced the angles for example: 94-41-10, 189-22-10 = 94-41-05 & 265-19-00, 170-38-10 = 265-19-05, M=94-41-00 (to make the angles add up to 360-00-00). I use his meaned angle. He also measured the distance in face 1 and face 2 and meaned it to a horizontal distance which is what I used. He didn't measure HIs and THs so the best I can get is a 2D adjustment.
I just have one issue I'm not sure what is the best way to handle it. Common practice was to set up on a centerline monument, backsight another then produce the line in the opposite direction. He doesn't note how he did this, I assume he plunged the scope, maybe did a double center. StarNet doesn't directly allow for this so I put the line in like this:
M 3-101-102 180-00-00 608.08 ! 0.01 'SetPK
I hold the 180 degree angle fixed but maybe I should let it float a little?
I have the standard errors of the angles set at 20", the statistical information looks like:
Adjustment Statistical Summary
==============================
Iterations = 2
Number of Stations = 26
Number of Observations = 53
Number of Unknowns = 50
Number of Redundant Obs = 3
Observation Count Sum Squares Error
of StdRes Factor
Angles 26 2.941 1.414
Distances 26 0.464 0.561
Az/Bearings 1 0.000 0.000
Total 53 3.405 1.065
The Chi-Square Test at 5.00% Level Passed
Lower/Upper Bounds (0.268/1.765)
I'd subject the 180 to the same standard error as the other angles. Double-centered or not, those flops aren't perfect.
Dave Karoly, post: 411778, member: 94 wrote: I hold the 180 degree angle fixed but maybe I should let it float a little?
One way to handle the standard errors of line projection would be to create a special instrument in the instrument library that is invoked using the inline command .INSTRUMENT ____________.
If he didn't record the angles, it seems unlikely that he made the projection by measuring angles to points straddling line and interpolating the distances from each to the point on line from the angles (that presumably were slightly more and slightly less than 180å¡00'00").
So, the instrument that you'd set up could use a standard error of nominally zero (0.1", say) for the angle of 180-00-00 representing the projection of the centerline and rely upon the target centering errors to capture the uncertainty in the line produced.
I always considered double flopping a scope to prolong a line as much more precise than turning an angle.
Paul in PA
Kent McMillan, post: 411792, member: 3 wrote: One way to handle the standard errors of line projection would be to create a special instrument in the instrument library that is invoked using the inline command .INSTRUMENT ____________.
If he didn't record the angles, it seems unlikely that he made the projection by measuring angles to points straddling line and interpolating the distances from each to the point on line from the angles (that presumably were slightly more and slightly less than 180å¡00'00").
So, the instrument that you'd set up could use a standard error of nominally zero (0.1", say) for the angle of 180-00-00 representing the projection of the centerline and rely upon the target centering errors to capture the uncertainty in the line produced.
StarNet allows standard errors to be inserted into the M line between the distance and description. So I could put say 1 0.01 for 1" angular standard error and 0.01 distance standard error.
Dave Karoly, post: 411812, member: 94 wrote: StarNet allows standard errors to be inserted into the M line between the distance and description. So I could put say 1 0.01 for 1" angular standard error and 0.01 distance standard error.
Yes, but it would depend upon techniques used whether the same target cenering errors used in weighting the other angles would be appropriate. At first impression, it seems reasonable.
Paul in PA, post: 411808, member: 236 wrote: I always considered double flopping a scope to prolong a line as much more precise than turning an angle.
It very much depends upon the distances over which a line is being prolonged. Over short ranges where the forward target is a plumb bob point or something similar at pavement level and with no refraction, that would be likely true. Over longer ranges where the forward target is above the ground and refraction comes into play, most likely not.
Line projection by theodolite angles to points straddling line becomes a more precise method as distances lengthen. It was used, for example, in running the boundary between Alaska and Canada in lieu of chasing "Left" and "Right" from ridge to ridge along the boundary. Once the angles to the straddle points are measured and the point on line between them is set, it's light work to verify the correctness of the point by theodolite angles to it.
