NY Long Island basically is a LPD system and very close to zero elevation. The grid to ground differences in 3.2 miles was less than one foot. The bridge girders were pre-assembled in Coeymans, near Albany NY, in 400' spans, then barged down river ~120 miles to the bridge site.
It would be nice to have SPC to be 100ppm. Mine range between 150ppm to 800ppm. 0.8' in 1000' is less than cloth tape accuracy. I can't think of a construction project designed on grid. The weird thing is that since GPS surveyors have had a MORE difficult time using site control.
@leegreen Do you really lay out grid coordinates? Or do you apply the appropriate CSF then layout. If the 2nd then you are not working on grid. What benefit is there in designing on SPC? I know of many possible pitfalls and I do not think the benefits even come close to outweighing the pitfalls. SPC have their use but applying them to construction sites is unwise in my view. At a low elevation the possible impact of mistakes is greatly reduced however at 6000' they become not only noticeable but can be costly. The potential for costly mistakes is greatly reduced by designing and working on a local system. A system with no scale factor.
You are not much of a scientist if you can't work with meters and feet. However if as a scientist you make a mistake then the spacecraft you sent to Mars becomes lost because you didn't work with the correct units. A huge costly mistake that was unnecessary and occurred only because someone felt compelled to introduce unneeded complexity. (Mars Orbiter $125 million)?ÿ
@leegreen Do you really lay out grid coordinates? Or do you apply the appropriate CSF then layout. If the 2nd then you are not working on grid. What benefit is there in designing on SPC?
Yes, we do work in SPC on NYSDOT projects. Set the controller to the correct SPC and the scale factors are applied for each EDM distance measured. All coordinates remain in Grid. Combined Scale Factors are only applied on the distances, never scale the coordinates. So yes, we are on Grid Coordinates. I didn't realize this method was so foreign. In the controller, you can choose to set a constant scale factor or have it calc the CSF at the setup point.
There are no 6000ft elevations in NY State. Mount Marcy is the?ÿhighest point?ÿin?ÿNew York, with an?ÿelevation?ÿof ?ñ5,344 feet.?ÿ
@leegreen That's interesting. I have entered the CSF into my gun to create ground distances, but only on projects that were small enough that the distances really did not matter. Typically, as shown on the insert which our DOT produced, we would scale from the origin 0,0 by the CSF for all of the coordinates so that they would now reflect ground distances. Unlike the DOT we would then truncate the resulting ground coordinates so there would be no confusion as to whether or not they were SPC. By entering the CSF to only alter the edm distances you are essentially scaling your coordinates around your occupy point instead of the origin. The resulting coordinates now cannot be scaled back down to state plane unless you first subtract your occupied coordinates. We always scale the coordinates around 0,0 if we want to easily convert between ground and SPC. Some governmental agencies require our clients to submit SPC so it is important for us to always be able to convert both ways.
I was taught to Never scale coordinates. CSF is only applied to distances. Odd that a DOT would not understand this. The CSF changes at every where on the grid. If you scale about 0,0 then how would anyone calculate the CSF? You end up with a bastardized coordinate system no one else can follow without the proper Metadata.?ÿ
I ran into a landfill project where someone scaled about zero, but didn't truncate and no Metadata. The coordinates were 200ft from true SPC, resulting in the contractor to drill gas wells in the wrong location.?ÿ
?ÿ
@leegreen Lee I agree coords that look like SPC but aren't can cause problems. Yes the CSF changes everywhere on the grid, thus 1 scale factor only is used for the project. Typically the center of the project both vertically and horizontally is used to determine the sole scale factor to be used on the project. Scale factors are not changed just because you occupy a different point. By scaling the coords you maintain ground distances between points. When you scale only distances you corrupt the relative distances between points. 2 points due East of your setup that should have 100' (ground between them) can be set out and will be ok. The same with 2 points due west of your set up. However if you set 1 point due east and 1 point due west of your set up they will be further apart than the intended 100'. You will have effectively applied the scale factor twice. You will have increased the Easting of the East point by the scale factor but you will have reduced the Easting of the West point by the scale factor. What should happen is the scale factor should be applied to both Eastings in a positive manner but by scaling distances you are applying the scale factor positvely to one coordinate and negatively to the other. The same is true for Northings.
Using the DOT scale factor above 1 SPC point 2500' east of the occupied point and one 2500' west should have 5001.7278 feet between them on the ground. Scaled distances though would put 5001.8277' feet between them. Not many projects have such a high scale factor nor do many projects have a problem handling a 0.1' of error in 5000'. It is not surprising that both your approach and my approach do not cause actual problems that have to be dealt with.
I find it ludicrous that they have the project elevation to 0.0001 meters and also the CSF to .00001 ppm.?ÿ
Unfortunately the PA Turnpike Commission does something similar to alter state plane coordinates to make them "project coordinates". The metadata then sometimes gets lost, and since they look like SPC people assume they are SPC.?ÿ
Just so much confusion. When you divide a state plane distance by its associated average combined factor, you aren't scaling anything about anything. What you are doing is determining the ground distance between those two points. That distance should match a tape measure or a total station measure very closely.
In many instances, a project scale factor is not associated with any known point. Instead, an average scale factor is computed for the project and an average elevation factor is also computed. The product of these two becomes the "project scale factor" but there is likely no point on the ground that shares these two parameters. Thus, the project is not centered around any known point.
I'm far removed from the nuts and bolts and skills of surveying and the software that drives surveys. However, I believe that I read somewhere on this forum (I think from Kent) that StarNet accommodates the procedure that @leegreen uses. NGS publications define that as the proper approach.
Unquestionably, adjusting coordinates by a single combined factor is mathematically equivalent to adjusting plane distances by the same single combined factor. However, the former practice destroys the connections from plane to ellipsoid to ground while the latter one preserves them. In both cases the use of the term "project scale factor" instead of "project combined factor" is misleading and adds to the confusion.
If you want plane coordinates different from state plane coordinates, then to me an LDP is the superior choice. If you want to base the survey on state plane and you want the most accurate individual distances that you can calculate, then Lee Green's approach is the best one.
Not too sure about Star*Net. I had a project recently where I had both GPS vectors and total station observations, which is very typical for me. I surveyed an existing network that had what looked like SPC, and did a free adjustment. There were four control points, two at each end, and I shot directly between two, one on each side (420 meters). The results of the adjustment agreed very well in orientation, but the distance between the two control points was off by 0.017 m, which was exactly the difference between grid and ground (sub mm difference). It appears that the previous survey was done at ground.?ÿ
So, I simply make a local system in starnet with scale factor=1.000000000. However, by doing that it will not accept GPS because it has no way to process/transform/apply the covariance terms. I had never done a local grid system in Star*Net, but I can't figure out a way to do it otherwise. And if I use a defined SPC zone then I have a scale issue. Any ideas on how to deal with this??ÿ
Here is the network, red is total station, blue is GPS:
Here is what the four control points look like (the other pedestal on that side is in the far background at the same elevation:
Two of them were up on?ÿ a ledge. The steps and roof are all for the pedestal control point...
Another interesting thing is that they drilled a hole through the concrete in order to sight one of the control points.
?ÿ
Here's one old source that alludes to the feature (see figure 3 and go to the bullet points on the last page):?ÿ https://www.johnson-frank.com/pdf/StarNet%20ProSuveyor%2010-96.pdf
I remember a discussion with Kent on grid to ground. He was adamant that no adjustments to state plane coordinates was ever necessary and, I think, he had an example where his starnet vectors were all adjusted to ground by their unique scale and elevation factors.
I'll dig back through the archives and see what I can find.
Sorry!
Here's a link to a Purdue example. The relevant material is on document page 20 with references to prior pages. I used the search term "elevation factor" within the document to cut to the chase.
https://engineering.purdue.edu/~bethel/sn6.pdf
Note the early versions of starnet in both cases.
?ÿ
Go down to the first post by Kent. He tells us what but not how.
I use Topcon Magnet Field / Magnet Office to navigate in SPC using GNSS and EDM all projects. As noted the software will calculate the CSF using the Occupied and BS points, then apply the scale factor only to the EDM distances. All coordinates remain in GRID. When inversing between grid coordinates the software will provide distances in grid, ground, and geodetic.
I think many old software packages did not work well with SPC between EDM and GPS, or perhaps many surveyors just have not learned to use SPC & CSF correctly.
?ÿ
I sent a question to Star*net tech support, and it turns out to be a simple solution. Starnet has a .multiply inline option, so I would apply the scale factor for the adjustment, and then scale back at the end. Or I could scale the given coordinates to grid, do the adjustment, then reverse the process.?ÿ
The Combined Factor is different at every point on the grid. By scaling your coordinates using one CF you are not producing ground distances except at the point the given CF is for. You are essentially choosing one approximate CF for your entire project. If the project is relatively small and flat then it may be sufficiently accurate although not mathematically rigorous. If your project is large or has large differences in elevation then it will matter.
I set my controller to the correct map projection and zone. Raw measurements are stored but the controller calculates SPC coordinates, not bastardized "modifrickinfried" (Loyal's term) coordinates. The data can later be processed through the office software and be projected however you want to do it.
