Single Total Station Instrument Setup Boundary Survey

Great question. I wish I could answer it for you. I once did a boundary survey, not for a plat, by running a traditional traverse. My new boss criticized me for wasting time and said I could have located all of the corners from a single TS setup.

All critical measurements of a boundary survey should have some element of redundancy. The closure of a traverse provides a check of all angles and distances (but not side shots). For a radial survey, redundancy can be provided by shooting all points a second time from a second radial point at a significant distance from the first radial point.

Another option (not as good as what @pls8xx wrote) if there is not another good spot where you can shoot everything would be to reset the TS at a different HI and BS one of the new points and reshoot everything including the original BS. Then compare coordinates between the 2 setups. Also I would be doubling (2 faces) all shots.

@pls8xx "The closure of a traverse provides a check of all angles and distances (but not side shots)." This is one reason for the original question. You can have a near perfect traverse and still flub a sideshot to a corner. Often times, a corner can only be seen from one of the traverse points. Are you suggesting to BS the rear traverse point, set the pin, then BS the forward traverse point, and re-shoot the pin as a check?

"For a radial survey, redundancy can be provided by shooting all points a second time from a second radial point at a significant distance from the first radial point." So, are you saying you think it would be okay to have only two setup points, occupy one and backsight the other, shoot all corners, then switch the occupied and backsight points, and shoot the corners again?  

Posted by: @stillplumb

↑

You can have a near perfect traverse and still flub a sideshot to a corner.

This is why you should always double an angle to a side shot.

Posted by: @stillplumb

↑

You can have a near perfect traverse and still flub a sideshot to a corner. Often times, a corner can only be seen from one of the traverse points.

You either take a chance on a bad sideshot or you do the work necessary to have redundancy. The only reason a corner can only be seen from one of the traverse points is because the person doing the traverse didn't do enough work to make it seen by 2 points. If it can be seen by 1 traverse point, you can traverse into it and back out of it and make it a part of the traverse. Yes, this takes more work. It also produces better results than the results you get without redundancy.

I guess I don't understand the question.  If you can see all corners from one setup, why not just BS the first (always D/R), FS the second, BS the second, FS the third, BS the third...etc. until you finally BS the last and FS the first.  That's a closed traverse with the distances between all the setups being 0.00', if you want to calc it that way.  LS adjustment and you're done.

@jim-frame Agree, Did not see the problem. I have done that at times on small lot surveys. setup , located all the corners with doubles, do my checks and comps, set my POL's and move to the next job.

We have similar issues in Australia. When you sit your board exam, they want to see a misclose on a traverse. They don't like it when you show up and state that there is no misclose because you traversed using the resection method and performed a least squares adjustment at each station. The way around this is to traverse using the resection method back to your original station (or thereabouts) then take a shot to one of your original control points. Then compare the deltas, find the linear difference and divide by the overall traverse length. But they still don't like it. I think they own shares in the Bowditch company.

It seems odd to me that the PU of a GNSS survey is acceptable, but radiating points from a total station is not. After all, the angular and distance SDs from a total station are the equivalent of the PU from a GNSS survey anyway when radiating points from a total station.

One survey in my entire career had the ability to shoot every corner from one setup and in fact it wasn't even my survey, just one I was helping on.  Anything beyond one point has cumulative error that is being introduced.  Whatever your state rules might be, the actual location of all corners in relation to one another is more accurately represented by the measurements from one setup (check local listings).

Posted by: @stillplumb

↑

@pls8xx "The closure of a traverse provides a check of all angles and distances (but not side shots)." This is one reason for the original question. You can have a near perfect traverse and still flub a sideshot to a corner. Often times, a corner can only be seen from one of the traverse points. Are you suggesting to BS the rear traverse point, set the pin, then BS the forward traverse point, and re-shoot the pin as a check?

"For a radial survey, redundancy can be provided by shooting all points a second time from a second radial point at a significant distance from the first radial point." So, are you saying you think it would be okay to have only two setup points, occupy one and backsight the other, shoot all corners, then switch the occupied and backsight points, and shoot the corners again?  

 

No to the first question, yes to the second, provided there is a significant distance between the two occupied points.

In a case where the corner monument can't be occupied, I might compare the coordinates after doing something like this:

With the change in survey equipment over the last 50 years, the 'when' and 'how' to apply redundancy has also changed. Modern equipment has reduced the chance of human error but may have increased the likelihood of machine error.

Redundancy is not a one-shoe-fits-all item. It should be tailored to the situation. This is not a problem for surveyors, it's part of being a professional. 

Posted by: @landbutcher464mhz

↑

Another option (not as good as what @pls8xx wrote) if there is not another good spot where you can shoot everything would be to reset the TS at a different HI and BS one of the new points and reshoot everything including the original BS. Then compare coordinates between the 2 setups. Also I would be doubling (2 faces) all shots.

this can actually be done to produce a closure. Simply back sight the other point after breaking setup. Return all the angles the first set vs second set can produce an angular misclosure as the sum should equal 360. Any centering error Changes or in disproves one the distance misclosure.  Not saying this is allowed or acceptable but in some basic elementary text books this is the first step in teaching the principles of a raw closure for precisian. One of the reasons of Intent of a traverse though goes beyond the closure. It was based on principles that the relative relationship of each corner to the next around a property were relative within reason. Hence not always historically were closures required   I believe in TN the statement on plats is you are certifying to the raw unadjusted closure not the adjusted one. This proves out in retracing old surveys how one side fits well but yet the other side also fits well but they do not fit together well on an old boundary yet the monuments are what they are. Sometimes well very often math is just an exercise. Not to take away from math along with good practices and weighing the intent with what is discovered sometimes is more common sense than all the math and technology along with the understanding of how it was originally established and with what techniques and equipment for that day. We still retrace many parcels that have no math whatsoever. Along jones line to bills lines and such. 

Locating everything from a single setup is acceptable in NC as long as you can calculate the relative positional accuracy and it meets the specs. Generally, a direct and reverse shot with three shots averaged from each face will get you to the ALTA standard of .07' + 50ppm, assuming your equipment is calibrated at least once a year. If you have a couple of irons set within five feet from each other you'd want to repeat shots to those, maybe three D&R shots, as the "relative" part of the positional accuracy requirement could bite you otherwise. 

Posted by: @murphy

↑

Generally, a direct and reverse shot with three shots averaged from each face will get you to the ALTA standard of .07' + 50ppm, assuming your equipment is calibrated at least once a year. If you have a couple of irons set within five feet from each other you'd want to repeat shots to those, maybe three D&R shots, as the "relative" part of the positional accuracy requirement could bite you otherwise. 

Maybe I'm missing something, but this seems like overkill.  As a test, I set up a dummy network -- a square 1,000' on a side with an instrument point inside the box -- and ran a Star*Net pre-analysis using my usual field equipment and methods:  A GeoMax Zoom 80R (2" horizonal angles, 0.005' centering for both gun and glass), D/R to each point.  Set up at 1, BS 2, turn to 3, then BS 3 and turn to 4, etc, finally closing from 6 to 2.  (I threw in point 6 about 7' from point 2 to check the relative positioning between nearby points.)

The pre-analysis indicates a 2-sigma semi-major axis of less than 0.06' for all points, including the relative between 2 and 6, thus in compliance with ALTA specs.

           Adjusted Bearings (DMS) and Horizontal Distances (FeetUS)
           =========================================================
                 (Relative Confidence of Bearing is in Seconds)

From       To               Bearing    Distance       95% RelConfidence
                                                    Brg    Dist       PPM
1          2          S33-41-24.24W    901.3878    8.95   0.0316   35.0047
1          3          N63-26-05.82W    559.0170   10.52   0.0431   77.0683
1          4          N63-26-05.82E    559.0170   12.46   0.0431   77.0683
1          5          S33-41-24.24E    901.3878   11.78   0.0446   49.5042
1          6          S33-36-04.45W    894.4551   10.56   0.0446   49.8529
2          3          N00-00-00.00E   1000.0000    0.00   0.0377   37.6932
2          6          N45-00-00.00E      7.0711  804.41   0.0538 7608.0266

                               Error Propagation
                               =================

                Station Coordinate Standard Deviations (FeetUS)

Station                     N             E             Elev
1                         0.000000      0.000000      0.000000
2                         0.016439      0.012291      0.007950
3                         0.017159      0.012291      0.008910
4                         0.014634      0.016908      0.008910
5                         0.019135      0.020210      0.011242
6                         0.018368      0.018557      0.011190

                   Station Coordinate Error Ellipses (FeetUS)
                            Confidence Region = 95%

Station                 Semi-Major    Semi-Minor   Azimuth of       Elev
                            Axis          Axis     Major Axis
1                         0.000000      0.000000       0-00       0.000000
2                         0.042740      0.026409     154-37       0.015581
3                         0.048981      0.016434     146-54       0.017462
4                         0.043083      0.033761      63-26       0.017462
5                         0.051478      0.044622      56-19       0.022035
6                         0.045784      0.044591     123-36       0.021932

                        Relative Error Ellipses (FeetUS)
                            Confidence Region = 95%

Stations                Semi-Major    Semi-Minor   Azimuth of     Vertical
From       To               Axis          Axis     Major Axis
2          6              0.054640      0.025866      33-33       0.026903

@jim-frame Ya beat me to it !

There is much to be said for maintaining a high level of precision. But I always thought that the value of redundancy was eliminating blunders. We should recognize that operations that produce high accuracy are still done by humans.

Two man crew, instrument man, rod man (limited experience) 

Friday afternoon, rod man thinking about weekend plans with his girlfriend

Set up at 1, BS 2, turn to 3, then BS 3 and turn to 4, etc,

Instrument man can't see the ground where rod man is on point 3

Rod man is actually on a nearby goat stake with old flagging rather than the corner the Instrument man found.

Math checks out, looks good.

Redundancy - men trade jobs

Former Instrument man goes to all the right corners

Former Rod man repeats the work but doesn't get great accuracy, but it's obvious that something is wrong when the coordinates don't match. 

@pls8xx Well, your specific scenario would hold true for anytime of surveying. Idk about creating a hypothetical situation jus tto prove the survey could fail. 

Posted by: @gary_g

↑

@pls8xx Well, your specific scenario would hold true for anytime of surveying. Idk about creating a hypothetical situation jus tto prove the survey could fail. 

 

I didn't mean to be critical of any method of surveying. What I did want to point out is that precision and redundancy are two different items of importance, and the one doesn't work as a replacement for the other. 

One valid form of blunder trapping that should not be discounted is comparing the results of your ties with record dimensions. Only if such values are unsatisfactory do we need to pull out the stops on other means of achieving redundancy.

I also note that these traverse closure specifications expect you to make your ties in a manner that would normally yield the specified closure, not to run some manufactured BS contrivance. If the regulation is ambiguous, it leaves open just exactly how it is to be followed. And if your state law still specifies accuracy in terms of traverse closures it is about 25 years past due time to amend that law.