Excellent RTK Survey on Perfect Site

> As may be seen the RTK survey was extremely accurate, in many cases better than 1:5000.
I'd bet that had more to do with the care and attention the corners were set with than RTK.

I assume you are trying to compare your work with the RTK work. In fact, what I see is the difference between you and another Texas surveyor and it has nothing to do with RTK. :-S

> > As may be seen the RTK survey was extremely accurate, in many cases better than 1:5000.
> I'd bet that had more to do with the care and attention the corners were set with than RTK.

The fence posts definitely were quite casually located. I don't think that any were tied from measured offsets, but instead whichever position the rover pole could be held that was nearest to the post (while naturally half a post diameter from its center) was considered to be the position of the post.

The rod and cap markers generally appear to have been set with enough craft that I'd suspect there were other problems in the mix. I haven't taken the time to autopsy the effort by determining whether the fence posts fit the record geometry among themselves much better (within nominally half a fence post diameter) than they do in relation to the rod and cap monuments.

Given the lapse of time between the initial "design" survey and the stakeout, and the possibility that the RTK base was shifted around during the survey, generating coordinates from the localization-du-jour, it seems like business as usual in RTK World.

> I assume you are trying to compare your work with the RTK work.

Not trying. That is actually a comparison of the results of an RTK survey as shown by the record of the survey to what is really on the ground. The quality of the monumentation was significantly above average even if the quality of the RTK-derived survey measurements was distinctly substandard.

Did you adjust the corners to fit the record, set new monuments in the correct place or did you call the offending surveyor to come adjust his corners?

> Did you adjust the corners to fit the record, set new monuments in the correct place or did you call the offending surveyor to come adjust his corners?

No, but I did have some fun restoring some monuments that had gotten graded out along a road and scratching my head over one striking FUBAR that had placed a marker on a road easement line more than 4 ft. west of the position indicated on the plat. When RTK is at work, it isn't surprising that there will be the wild misses.

> The survey appears to have been done in several stages

The first opportunity for a less than diligent surveyor to step in the proverbial steaming pile.

Tis a poor craftsman that blames his tools.

Loyal

> Tis a poor craftsman that blames his tools.

But if all you've got is a hammer, everything looks like a nail. I'm just doing the autopsy and it appears to me that somebody has been selling magic hammers to folks unfamiliar with nails. :>

"Blaming His Tools"

Kent,

"...comparisons of record to actual..."
Is there any reason to assume that YOUR measurements are superior to another surveyor's?

"...a modern subdivision that was laid out with RTK..."
So on that Record of Survey was it noted thusly: "Eat it Kent, we done it all with RTK!"?

"...distances being surface distances computed from exact grid distances using a Combined Scale Factor that differs from actual by less than 10ppm..."
No surveyor should use the words "exact" or "actual", because it implies a certain level of arrogance. How "exact" were the distances if you felt you needed to "adjust" them with an arbitrarily chosen "average" scale factor?

"...bearings referring to grid North of the Texas Coordinate System..."
The couple of degrees difference between the other surveyor's bearings and your "actual-true-exact-perfect" bearings looks suspiciously like mapping angle, and the distance difference might be merely a difference in choice of "average" Combined Scale Factor. I'd want to look at ALL the data before I make a denunciation.
Information lacking here for BOTH surveys:
Which Texas Coordinate System?
Reference System? NAD 27 or NAD 83?
Datum/Frame Tag? 86? HARN? 2007? 2011?

Opinion Alert:
Low-Distortion Projections don't suffer from Convergence and Scale problems as greatly as State Plane does.

Dave

"Blaming His Tools"

> "...comparisons of record to actual..."
> Is there any reason to assume that YOUR measurements are superior to another surveyor's?

Yes. That's what using the error analysis functions of least squares survey adjustment software such as, say, Star*Net, makes possible.

> "...a modern subdivision that was laid out with RTK..."
> So on that Record of Survey was it noted thusly: "Eat it Kent, we done it all with RTK!"?

No, the plat indicated it was surveyed via RTK and my previous conversations with the responsible surveyor confirmed that.

> No surveyor should use the words "exact" or "actual", because it implies a certain level of arrogance.

Not really. A grid distance with an uncertainty of 0.02 ft. is exact in comparision to a distance with discrepancies of more than 0.20 ft. In the context of a surface distance computed from a grid distance using an average Combined Scale Factor, the grid distance is essentially exact and the surface distance is inexact in the sense that there is known (small) error in it as a result of the variation in height from the project average. The effects from variation in height are not something that a custom projection eliminates, BTW.

> "...bearings referring to grid North of the Texas Coordinate System..."
> The couple of degrees difference between the other surveyor's bearings and your "actual-true-exact-perfect" bearings looks suspiciously like mapping angle, and the distance difference might be merely a difference in choice of "average" Combined Scale Factor. I'd want to look at ALL the data before I make a denunciation.

Yes, obviously the surveyor who laid out the subdivision wasn't working in the SPCS. The North direction to which the bearings on the plat refer was almost certainly geodetic North at some unspecified point where an RTK base was set up to kick things off.

> Information lacking here for BOTH surveys:
> Which Texas Coordinate System?
> Reference System? NAD 27 or NAD 83?
> Datum/Frame Tag? 86? HARN? 2007? 2011?

None of that makes a significant difference as far as comparing surface distances to distances noted on the plat.

> Low-Distortion Projections don't suffer from Convergence and Scale problems as greatly as State Plane does.

Well, a custom projection would still have had to deal with the variation in scale from change in elevation and would be a pig with lipstick. The advantages to using standard projections in that area are huge since one routinely deals with the relationships of things at separations of ten or more miles for which someone else has reported coordinates.

> I would have to agree with other posters that something seems a bit whacky, even for RTK. To be a that far off means no fix, or mixed/improper base, etc - not reading the data is a failure of the operator and not the tool.

Well, to begin with, the RTK was apparently set up to generate coordinates in a projection grid oriented geodetic North at the RTK base and approximately at an ellipsoid height of 0.0 as derived from an autonomous position. That last bit was apparently something the surveyor was unaware of since there is no indication of that on the plat.

The principal lesson is that if you're going to use RTK, use a standard projections that you can describe. If you don't want to use a standard projection, you assume the burden of documenting whatever projection you cooked up in a way that some future surveyor will be able to convert it to a more useful projection.

If the responsible surveyor had simply instructed the RTK controller to use the proper zone of the Texas Coordinate System of 1983, even accepting an autonomous position for that of the base, the end result in terms of grid bearings and grid distances would be virtually indistinguishable from those of a survey meticulously connected to NAD83.

> And I'd have to agree with another post, words like "exact" and "actual" do kinda sound like those that would be used by overzealous salespeople of magic tools.

You're probably thinking in terms of RTK World where everything is speculative. In the world of post-processed rapid static GPS solutions and conventional measurements, one can speak with confidence of what is actually on the ground and its relative exactness compared to some faulty value.

Relative Exactness

Kent,

"...relative exactness..."???

A new Kent-ism!

Dave

Relative Exactness

> "...relative exactness..."

Not really. The whole idea of exactness depends upon context, i.e. in relation to what. For example, for construction, anything below the threshold of construction tolerance is essentially exact. In comparison to RTK-derived length measurements with errors of a foot or more, a distance with an uncertainty of 0.02 ft. is essentially exact.

Relative Exactness

> Even the professionals that work in the realm of metrology, down to 20,000th of an inch do not use terms like exact and actual (unless they are in sales). It would be like an engineer saying there is absolutely no chance of structural failure - would get laughed right out of their presentation. "Relative exactness" sounds a bit like a "feeling" rather than science

I love that you RTKers are so averse to the idea of exactness, although I can understand certainly understand why that would be. :> In the real world of professional land surveying, as uncertainties dip below 5mm, exactness is descriptive in that no better measurement produces a significantly different result.

Relative Exactness

Kent McMillan ExactoNetworks are built with pride to the most exacting standards. We guarantee your complete satisfaction with your ExactoNetwork or we will gladly refund your purchase price. Simply return any unused portion of the product to P.O. Box 5MM, Austin, TX.

Relative Exactness

> Kent McMillan ExactoNetworks are built with pride to the most exacting standards.

The beauty part is that it doesn't take that much effort to manage the relative exactness of a survey as long as you have realistic estimates of the uncertainties in the various component measurements. If an RTK controller has the means to output the GPS vector with its uncertainties and a surveyor has a means of testing whether those uncertainties are realistic or are either optimistic or pessimistic, then the relative exactness of even an RTK survey can be evaluated.

Obviously, systemic RTK failures such as misusing a projection, would be harder to detect unless one cracked out a total station and compared actual scale to projection scale on a sufficiently long line to give a meaningful result.

Relative Exactness

> Give people some credit.

I eagerly await the day when I find that an RTK survey has been made in a way that isn't markedly substandard in some significant way. The reason why I mention the low quality of RTK work is because that is pretty much all that presents itself.

As for the time element, it took a five-minute occupation of the rod and cap monuments to get a position with an uncertainty of 0.02 ft. In all the amounted to 5 x 25 minutes = about 2 hrs. The folks who want to argue that an extra 2 hours to locate 25 monuments is a deal breaker that will cause them to go out of business are welcome to do it. They do have to realize, however, how subpro it sounds.