Does anyone have a good calculation worksheet laying around in a cabinet somewhere collecting dust that they would be willing to share?
This would be for two measured distances (to known points) and single angle.
imaudigger, post: 371962, member: 7286 wrote: Does anyone have a good calculation worksheet laying around in a cabinet somewhere collecting dust that they would be willing to share?
This would be for two measured distances (to known points) and single angle.
2D or 3D?
I have an excel worksheet. I usually use the data collector so I haven't used it in a long time, but I remember it working fine. I'm not sure how to attach a spreadsheet. You could email me s m c l o y at b l u e l i n e n m dot c o m and I will reply with an attachment.
I have a SMI v5 data collection card complete with an HP48GX and memory card.............;-)
This could be for a 2D solution since the elevations are pretty straight forward.
There must be several different methods for solving a 2 point resection, because I have three different solutions (which I was not expecting).
Trimble DC
Carlson Survey
Hand calc.
.25' between Trimble and Carlson, with my hand calc. position being somewhat in the middle.
All around this isn't a very good situation - three control points set 16 years ago. One is disturbed and one has a new tree trying to swallow it.
I suppose all of the resection positions are more than likely located within the error ellipse given the fact that there is error in the "known" points.
We did check into the third disturbed point located beyond the footprint of the topo and it fit within .2'
The final product is 2' contours, so I think I can live with a couple tenths horizontal error. Situations like this are always a learning opportunity for me.
Known Points:
#803 N=4269.108, E=9217.006 HD= 278.533
#804 N=4546.050, E=9504.691 HD=216.918
Angle RT = 106-38-16
My hand calc. position is N=4329.593, E=9489.017
You have one more measurement than is needed to solve the problem, but there is significant discrepancy between your given values, so the answer will depend on what is assumed to be most accurate.
There is a couple tenths or a quarter foot of error that has to be assigned to the relative positions of 803 and 804, or else the distances and/or angle, or some combination of them.
The different solutions must have picked some givens and let others float or given them different weights.
imaudigger, post: 371962, member: 7286 wrote: Does anyone have a good calculation worksheet laying around in a cabinet somewhere collecting dust that they would be willing to share?
This would be for two measured distances (to known points) and single angle.
Bill93, post: 372017, member: 87 wrote: You have one more measurement than is needed to solve the problem, but there is significant discrepancy between your given values, so the answer will depend on what is assumed to be most accurate.
There is a couple tenths or a quarter foot of error that has to be assigned to the relative positions of 803 and 804, or else the distances and/or angle, or some combination of them.
The different solutions must have picked some givens and let others float or given them different weights.
Ya, like I said - not an optimal situation. I do have something else up my sleeve once i get some data reduced.
Field Dog - Thanks for that attachment. That was the only thing that came up quickly on a Google search. That's the form I used for my calculations.
Probably not something anybody but students do anymore these days.
Co-worker just pointed out to me where the HP48 TDS resection routine is found.
I'm wondering if one is giving weight to angles and one is giving weight to distances? I just can't figure out where in the calculations you would do this.
The small circles are the calculated positions I used to come up with my average position.
The Carlson routine seems to be doing some sort of distance-distance solution, then somehow applying the measured angle.
I realize this is a crap shoot. All of the error could and probably is in one of the "known" points, not both.
As you know better than I your accuracy needs, I provide the following comparison.
Inverse distance between 803 and 804: 399.324
Distance by Law of Cosines: 399.039
Difference between known and measured: 0.285
Length relative precision: 1 / 1403 
GeeOddMike, post: 372071, member: 677 wrote: As you know better than I your accuracy needs, I provide the following comparison.
Inverse distance between 803 and 804: 399.324
Distance by Law of Cosines: 399.039
Difference between known and measured: 0.285
Length relative precision: 1 / 1403
I would agree with your math.
I tied into some other monuments that were located when the control was originally set.
So I do have additional data to work with.
Carlson's report gives the indication that the data was pretty good...seems it is leaving out some useful information.
Northing residual = .004
Easting residual = .052
You have a discrepancy between the published coordinates and your field measurements. Note that the discrepancy in the resection points is similar to the difference between the inverse and the indirect field measurement. Something has to give. Sounds like what the site really needs is updated control. Assuming that at the end of the day someone is going to build something on the site that might need to be within a position better than 0.25+/-.
1st, this is NOT a 2 point resection. Look at any basic book on surveying to see what a 2 point resection is.
2nd, you have left out information for the reader of your post to help you with your problem.
If you have measured the distances from your unknown point(N4329.593,E9489.017; your calculations) to both points #803 and #804 then list them.
With what you give the reader has no way of knowing which side of the line ( sitting on #803 looking to #804 ) you are on. Your unknown point could be on the left side of the line; as you sit on #803 looking to #804 (a mirror image).
More information is needed.
JOHN NOLTON
Clearly there is a bust, .3' is way too much in that little distance. More control needs to be tied. Control has moved or the field data was mis-measured
lmbrls, post: 372083, member: 6823 wrote: You have a discrepancy between the published coordinates and your field measurements. Note that the discrepancy in the resection points is similar to the difference between the inverse and the indirect field measurement. Something has to give. Sounds like what the site really needs is updated control. Assuming that at the end of the day someone is going to build something on the site that might need to be within a position better than 0.25+/-.
I would agree with everything you said.
Once I am done analyzing all the data that I have available, I may decide that I need to return to the site and gather additional measurements in order to firm up the control.
JOHN NOLTON, post: 372084, member: 225 wrote:
If you have measured the distances from your unknown point(N4329.593,E9489.017; your calculations) to both points #803 and #804 then list them.
With what you give the reader has no way of knowing which side of the line... (a mirror image).
I find measured horizontal distances given as HD in his post #5. And by giving his calculated coordinates he has told us which side of the line he knows it is on.
Since there is redundant data, a least squares solution tells you a lot more than just picking things to hold and using a formula for the unknown.
JOHN NOLTON, post: 372084, member: 225 wrote: 1st, this is NOT a 2 point resection. Look at any basic book on surveying to see what a 2 point resection is.
2nd, you have left out information for the reader of your post to help you with your problem.
If you have measured the distances from your unknown point(N4329.593,E9489.017; your calculations) to both points #803 and #804 then list them.
With what you give the reader has no way of knowing which side of the line ( sitting on #803 looking to #804 ) you are on. Your unknown point could be on the left side of the line; as you sit on #803 looking to #804 (a mirror image).More information is needed.
JOHN NOLTON
John, It's a resection using two known points. Measurements are two distances and an included angle. Call it what you will.
I thought I posted all of the information. I suppose it is necessary to know that #803 was the first known point and that the angle RT was measured to the second known point #804.
Known Points:
#803 N=4269.108, E=9217.006 [hl]HD= 278.533[/hl]
#804 N=4546.050, E=9504.691 [hl]HD=216.918[/hl]
[hl]Angle RT = 106-38-16[/hl]
My hand calc. position is N=4329.593, E=9489.017
I appreciate all of the input, I really was only looking for some calculation sheets to use. I will analyze all the data I have available and make a determination as to what steps to take.
There certainly IS a bust somewhere and it is in the position of the "known" control that was set 16 years ago. I have 3 primary control points and an additional 3 monuments that were located from that control. Problem is, in 16 years the ground has moved and SOMEONE has managed to bump nearly every single monument with heavy equipment. I have a feeling a return trip will be warranted in order to tie it all together and set some new control before reducing this topo.
It would have been easier to simply go out and gather more data, but I'm glad I got the opportunity to run through the calculations myself as a check on the software. That field form really helped me pull my calculations together into an organized fashion.
Use of resection here appears to be a garbage in/ garbage out situation. Resection really only works well on high quality control. Without some input on the real quality of the coordinates of 803 and 804 one only has the 0.29 foot difference seen between them as a guess as to the quality. Imaudigger's fit to a third point within 0.2 suggests an uncertainty of 0.2 to 0.3. Imaudigger's calcd value is just the mean of the 2 positions you get if you hold 803 to 804 as the B.O.B. and shift your measured triangle to fit at 803, then at 804 (no adjustment of measurements, or accounting for uncertainties in the control coords is involved). The Carlson solution does appear to be near a dist-dist solution. How Carlson came up with those small residuals from the given data eludes me. Assuming 803 or 804 is not disturbed and the disturbed point is within the 0.29' of it's original coordinate, I'd guess the Carlson solution could be out .3 to 0.4.
For "getting yourself onto" old control, there used to be several coordinate comparison/coordinate least squares best fit routines out there (Triad; TGO; VLS Software ( http://www.google.com/patents/US7649540 ). VLS was the best in part because it paired up your points automatically. Does anyone have a recommendation for such software today?
BajaOR, post: 372102, member: 9139 wrote: Use of resection here appears to be a garbage in/ garbage out situation. Resection really only works well on high quality control. Without some input on the real quality of the coordinates of 803 and 804 one only has the 0.29 foot difference seen between them as a guess as to the quality. Imaudigger's fit to a third point within 0.2 suggests an uncertainty of 0.2 to 0.3. Imaudigger's calcd value is just the mean of the 2 positions you get if you hold 803 to 804 as the B.O.B. and shift your measured triangle to fit at 803, then at 804 (no adjustment of measurements, or accounting for uncertainties in the control coords is involved). The Carlson solution does appear to be near a dist-dist solution. How Carlson came up with those small residuals from the given data eludes me. Assuming 803 or 804 is not disturbed and the disturbed point is within the 0.29' of it's original coordinate, I'd guess the Carlson solution could be out .3 to 0.4.
For "getting yourself onto" old control, there used to be several coordinate comparison/coordinate least squares best fit routines out there (Triad; TGO; VLS Software ( http://www.google.com/patents/US7649540 ). VLS was the best in part because it paired up your points automatically. Does anyone have a recommendation for such software today?
I'd say that is a fair analysis.
Funny you mentioned Garbage in - Garbage out. This is a landfill cap survey.
[sarcasm]Don't blame me - I was just the rod man.[/sarcasm]
I thought I had it all nailed down 16 years ago!
imaudigger
Lets see 16 years ago some points were set and a local coordinate system was put on them.
Now you come back 16 years later and set on a unknown point and measure two distances and also measure the interior angle.
With this data you can check your old data (at least the distance between #803 and #804. I see no need to put coordinates on the unknown
point yet. With 2 distances and the included angle you just use the Law of Tangents, then the Law of Sines to calculate the distance between #803 and #804 (no coord. needed).
Your inverse between #803 and #804 (with your given coord.). is 399.324 ; no units were specified and none needed.
Now with your new field work ( and we done know how many times you measured the angle or with what type of instrument, nor do we know
how you measured the Horizontal distances, was the EDMI calibrated or did you use a tape?).
By using Trigonometry the calculated distance between #803 and #804 is 399.039 OR a difference of 0.285
Now this is not good but you would have to be the one that decides that. If working in a dump it would be OK(?).
I think you went about this the long way. Why did you not set on #803 and make a direct measurement to #804. You would have seen that something was wrong with the control from 16 years ago. That's if it was good 16 years ago.
JOHN NOLTON