MightyMoe, post: 368456, member: 700 wrote: If you're using a Base/Rover setup then you don't do anything, set up on your elevation point and go. The file should be correctly projected, the geoid model applied and you hold the elevation of the point you are on.
Never, never, never calibrate, it's a bad idea, especially vertically. Now if you are in a network??
Don't use them so I don't know, but I would advise to use Base/Rover for tight site work anyway.
I agree, and at least with NY RTN you would need to compute a factor for a given area to get to NAVD88; the RTN is not directly tied to it.
I would typically recommend NOT using a tilted plane localization. Depending on the software, it is possible to use multiple points for vertical that are used for vertical translation only. This process would keep the tilt values at 0, and average the translation of each of the multiple points that are fixed.
Again, depending on software, a localization can be used with a datum that is based on orthometric heights (using a geoid model) or a datum that is based on ellipsoid heights. A localization is a comparison of two 3-dimensional Cartesian coordinate systems and the resulting comparison is a list of transformation parameters that mathematically convert one system of coordinates to the other system of coordinates with the lowest possible residuals based on least squares.
As Loyal so eloquently depicted, the contours of the Geoid can be significant. These changing contours represent changing slopes. Imagine providing a client a topographic survey of the Grand Canyon based on 4 shots. This would be somewhat analogous to making a best fit plane to cover a larger area with varying mass. On the other hand, using a single point, or even multiple points only for translation, would be more analogous to translating to an existing DTM surface (the geoid) and trusting the DTM surface for cuts and fills across a major earth work project. The tilted plane that results from a multipoint localization is just a single plane. The DTM is composed of multiple planes. It could be that the DTM is inaccurate at times. This is why it is good to tie in those 4 points for QC. The DTM however has more resolution in most cases that a single plane that you would create using inclination values in the localization.
I would recommend creating a tilted plane only in areas where the Geoid is deficient. But be aware, there is probably some local gravitational anomaly that is causing the geoid to be inaccurate in this sort of area, and the tilted plane will be subject to the same disturbance.
I doubt that I can add much to what has already been said, but here is my take on the subject. A single point vertical adjustment is going to adjust every point vertically by the amount of difference between the control elevation and the observed elevation, period. The relative vertical precisions of the points will be entirely dependent on the quality of the geoid model that you're using. I have found this method to be particularly useful when using the RTN on surveys that were required to be tied to a particular benchmark.
As has been described, a multi-point vertical calibration creates an inclined plane through the site. I always use a geoid model in this scenario as well, to best compensate for geoidal undulation. There are several potential pitfalls to using multiple points. You specifically mentioned four points; this is fine as long as the four are roughly at the corners of the site. If they have a linear relationship you can easily get a tilting plane and the error can get real ugly real quick as you move away from that line. For obvious reasons two and three point vertical calibrations should NEVER be performed (message me if you need an explanation of this) Also four points offers no redundancy - if one of the points is bad you're back to the three point scenario. So in order to have a best fit, proper redundancy, and truly meaningful residuals, I prefer to have a minimum of five points, and the more the better (to a point) depending on the size of the project. The points need to be distributed about the perimeter, and I like to have one or two near the center as well.
Calibrations work well as long as you play by the rules. If you start working "outside the box" you're extrapolating and the errors will generally increase exponentially as you move further away. Trimble has guidelines for the maximum recommended tilt value for the calculated vertical plane; I'd have to look it up, my Trimble Trainer days were a few years back. The only time that I ever use a site calibration these days is when I absolutely have to be on a plant grid system, but surveying down here doesn't have some of the challenges that are found in other parts of the country.
This is great topic. Wish we could get others to understand this. For some reason nearly every Trimble crew I talk to on a NYSDOT job says they must use a calibration. Yet every NYSDOT I've been on works fine with a NAD83 projection and GEOID. I had one client in Buffalo area tell me that his hired PLS and/or crew re-calibrated the control 7 times inside a 12 month period.
leegreen, post: 368559, member: 2332 wrote: ...re-calibrated the control 7 times inside a 12 month period.
I don't even understand why anybody have to calibrate (localize) more than once, if ever, on a project.
That sounds like big problem.
Kevin,
I agree. It did create problems. But for the life of me, there was no way I could convince the PLS that this procedure was not working, and not at all needed. I sure it had a lot to do with me NOT being a PLS, only a technician. One reason why I plan on taking the exam(s).
leegreen, post: 368559, member: 2332 wrote: This is great topic. Wish we could get others to understand this. For some reason nearly every Trimble crew I talk to on a NYSDOT job says they must use a calibration. Yet every NYSDOT I've been on works fine with a NAD83 projection and GEOID. I had one client in Buffalo area tell me that his hired PLS and/or crew re-calibrated the control 7 times inside a 12 month period.
If I'm not mistaken, Trimble's SCS900 software requires a calibration. I don't think it will let you even open a site without one....
I would think that using a mark with a known NAVD 88 elevation and a valid ellipsoid height would be the best-case scenario for a 1 point vertical localization.
Using a handful of user established control points without known elevation/ellipsoid data is still relying on any given geoid model.
My interpretation would be that holding a handful of self-established control marks for vertical localization is creating a false sense of security. All one would be doing is analyzing the quality of GPS observations in comparison to the chosen geoid.
If you calibrate/localize onto existing control, I don't think you are still using the geoid file.
It's one or the other.
Steve Emberson, post: 368567, member: 181 wrote: If I'm not mistaken, Trimble's SCS900 software requires a calibration. I don't think it will let you even open a site without one....
Steve,
This is along the lines of what I am constantly being told by these Trimble users.
Is this really true?
This does appear to be the root of the problem.
Why would Trimble NOT allow the use of a Projection with a GEOID?
The best practice for vertical control of existing elevations has been shown to be using a Geoid model applied to the elevations. This is simple enough to do. I have generally been using Geoid09 for the last 5-6 years, I also have some projects that have Geoid03, both work well, but Geoid09 seems to do better.
Up until three years ago the FEMA maps in the area were NGVD29 with an extensive set of bench marks to control the maps.
Setting on a 29 bench applying the Geoid model to the 29 ELEVATION was shown to give by far the best results with GPS. There is no calibration or localization to this process.
As we move forward I'm including Geoid12B in newer projects, and it seems to capture undulations in the Geoid even better than Geoid09. It must be stated that it can't work to use "raw" numbers from CORS or OPUS, this will not produce elevations that match NAVD88 as precisely as holding the actual record elevations. Of course this should be obvious. If I send a point into OPUS it will give me one number, if I do my own calculation from the closest CORS it gives me another number, and the actual bench is a third number. The last one I did, the OPUS # was .14' low, the CORS # was .07' low (not too bad).
If what you need is record elevations tied to local control then a calibration is not as accurate as actually occupying the points and using them with a Geoid model, again this is obvious.
It isn't necessary to use NAD83(2011) numbers from CORS in concert with GEOID12B, you can of course, but you won't "hit" older bench marks, maybe in some areas, not here, not yet.
Andy J, post: 368571, member: 44 wrote: If you calibrate/localize onto existing control, I don't think you are still using the geoid file.
It's one or the other.
I guess it depends on your software. In Trimble Access I can use the geoid model with a vertical site calibration. This is desirable because you're calibrating the elevations of the control to the ellipsoid heights of the observations and if you don't use the geoid in an area with significant undulation you'll get error.
leegreen, post: 368573, member: 2332 wrote: Steve,
This is along the lines of what I am constantly being told by these Trimble users.
Is this really true?
This does appear to be the root of the problem.
Why would Trimble NOT allow the use of a Projection with a GEOID?
It is absolutely true in their software up to v2.9x. I haven't dove in to the v3.x yet, it possibly could. I think the reason behind it is keeping the software dumbed down for contractors requiring them to have survey control professionally set before they could just take off, kind of a big brother deal. I assume their thinking is contractors wouldn't have the people in place to use geoids, select proper zones, etc. It also could stem from the iron curtain between the survey and construction sides of Trimble products.....
Andy J, post: 368571, member: 44 wrote: If you calibrate/localize onto existing control, I don't think you are still using the geoid file.
It's one or the other.
As a point of clarification... If your original control is a product of the geoid to begin with, you are still relying on the geoid; don't fool yourself.
Of course the source and accuracy of the control varies. What is any given surveyor using for "existing control"? I would assert that depends on what was done to establish said "existing control". Since this discussion is centered around the vertical component of localization my point is that 4 assumed points with orthometric heights derived from a geoid are not equal to more rigorously established control (say a series of NGS bench marks).
I think there are 2 common practices: 1)some combination of OPUS positioning including NGS published control points (bench marks) to establish site control
2)control marks set by a surveyor and assumed to be perfect and held as "truth" for a given project
My experience shows that users localizing to multiple vertical points tend to be using assumed coordinate systems. The experience of others may be different.
In my experience, those avoiding localization or using only one vertical point and verifying the vertical position of other control tend to have a better grasp on the control networks and the geodesy behind them.
I hope that clarifies what I was attempting to state.
Steve Emberson, post: 368592, member: 181 wrote: It is absolutely true in their software up to v2.9x. I haven't dove in to the v3.x yet, it possibly could. I think the reason behind it is keeping the software dumbed down for contractors requiring them to have survey control professionally set before they could just take off, kind of a big brother deal. I assume their thinking is contractors wouldn't have the people in place to use geoids, select proper zones, etc. It also could stem from the iron curtain between the survey and construction sides of Trimble products.....
Steve,
Thanks for the info. I do notice that Trimble likes the iron curtain approach (as you call it) to construction vs. survey. The first questions you are asked by a Trimble tech. or dealer is along those lines. Topcon also has a dumb down version of GNSS software for construction, it is called Pocket3d. But no iron curtain. It is very powerful, yet simple to use. No limitations and 100% fully compatible with the survey software.
Kevin Samuel, post: 368594, member: 96 wrote: As a point of clarification... If your original control is a product of the geoid to begin with, you are still relying on the geoid; don't fool yourself.
Of course the source and accuracy of the control varies. What is any given surveyor using for "existing control"? I would assert that depends on what was done to establish said "existing control". Since this discussion is centered around the vertical component of localization my point is that 4 assumed points with orthometric heights derived from a geoid are not equal to more rigorously established control (say a series of NGS bench marks).
I think there are 2 common practices: 1)some combination of OPUS positioning including NGS published control points (bench marks) to establish site control
2)control marks set by a surveyor and assumed to be perfect and held as "truth" for a given projectMy experience shows that users localizing to multiple vertical points tend to be using assumed coordinate systems. The experience of others may be different.
In my experience, those avoiding localization or using only one vertical point and verifying the vertical position of other control tend to have a better grasp on the control networks and the geodesy behind them.
I hope that clarifies what I was attempting to state.
You are making a very good point, if the site was surveyed using GPS and there is metadata explaining how, then the construction surveyor should also use those same methods.
I would caution to say that if the surveyor calibrated somehow to create the underlying data, there may be no way to recreate that calibration.
Calibration if done, should only be done one time, then never again. Some have been taught to use calibration as a check and I've heard of some surveyors calibrating every day.
algallop, post: 368370, member: 8394 wrote: My boss, a licensed surveyor for many years, taught me that in order to get accurate vertical readings while performing an RTK survey, you have to calibrate on four or more points with good vertical solutions. However in practice, I have done RTK surveys calibrating on only one point vertically and checked it with a level and/or total station, and found that the WGS '84 projection (both in SurveyPro software and Trimble Access) works pretty well establishing an accurate vertical plane.
I would like to perform topographic surveys with our R10 equipment using a 1-point vertical calibration, but first I have to convince my boss that it is accurate... Can you guys give me some convincing feedback that it's safe to do this?
I'm sure it's not as simple as I am making it sound, and I'm not an expert in GPS technology or anything. I just know that in my experience, a RTK 1-point vertical calibration is just as accurate as a level or a total station (at least when you are only working within a relatively small area, maybe not over miles and miles).
Thanks in advance,
Arthur
It's okay. The plane will work. When you use multiple points, you have to stabilize the plane. So, two points, allows the plane to tilt. Four points, surrounding the project, will do the same thing as 1 point. The problem with calibrations is that it can mess with stuff. If you get outside the relevant range of the calibration, things can go badly, very quickly.
To that end, I recommend doing neither. If you don't trust the elevations from the R10 or OPUS, go shoot a vertical point and simply translate the elevations at the end of the survey, in the office. I've had amazing results doing this, adjusting the points to OPUS verticals over the years. Generally less than a tenth when compared to other ones.
The four point thingy is a throwback to what the dealers were told to teach us. Understanding what it's doing will allow one to think outside the box.
Jon Collins, post: 368427, member: 11135 wrote: I'm with your boss. While others have good points as far as very small sites, checkins, good geoids etc....why do something half ass. Calibrate the right way. Why put your reputation on the line? I've followed way too many of these on construction projects, one arrogant idiot made me drag him in front of half dozen city engineers, contractors VP, the engineer he worked for, my 2 bosses....all to show I can't follow his 1 point survey. He had done an autonomous job, even worse. Funny thing was I think my whole fee was 4k. Small job. Big problems.
[USER=11135]@Jon Collins[/USER]
Why calibrate unless you are trying to fit an old database? What's wrong with NAVD88?
Loyal, post: 368455, member: 228 wrote: Here's an image of what Geoid12a looks like in Utah. Basically a contour map (relative to the NAD83 Ellipsoid) expressed in 0.1 meter contours. Not too many "planes" in the larger sense, but in the postage stamp sense, one could probably get away with an incline plane locally.
Loyal,
Is this map available for the state of Texas that you know of? It sure would be handy for instructional purposes.
Just googling "geoid contour map texas" I found a few but nowhere near the detail of the Utah map.
James
FWIW
We do alot of construction layout using RTK networks here in Florida mainly 1 ran by lengmann and occasionally FPRN run by FDOT.
MY boss PLS for 30+ years will never do a localization, there was one site about 5 square miles only time in the past 3 years we've done one.
Now the only time we use if for elevations is rough topo on soft surfaces i.e. grass/dirt etc. where a tenth of error can come from how far you push the pole in the ground.
Now me guy who has a BS in computer science goes out without boss man I usually localize to 2 points from NGS.
ON my own time I've been working on testing the 2 networks and seeing which ones and which corrections give me better vertical and which better horizontal. Now I had a scare that has reduced my confidence in my setup.
Shot a NGS point with GPS and elevation on it came within about a tenth on all networks and correction types all good there. Now the scary part my HD (Horizontal distance) was off by a TENTH for all networks and corrections on network was off one way the other network was off the other. May just be that one point I test against has been disturbed, but its disturbing me. Laying out curb inlet CB's starting to scare me a little bit.
my 2Cents....