Grid VS. Ground

[Anonymous]

I thought I would go ahead and start this thread b/c my other tread "Grid to ground coordinates" opened up a can of worms. I'm not biased ether way. I do strictly design. I do not survey, construct or map. I am curious which is best. As I type this I have a feeling it could get ugly. So please be respectful of each others opinions and please state facts and sound arguments. After a few good arguments I may declare a winner, If that's possible and I feel like sticking my neck out.

Thanks
Joe

[Sinc]

Yep. We're in kind of a "limbo" phase right now, where we know the problems of the "old school" methods, but haven't quite transitioned yet into a better solution...

LDPs are a very nice solution. But when we do construction surveying, we often have little choice in the matter, unless we were also involved at the very beginning of the project. Unfortunately, the way things usually work, the construction phase hits and goes up for bid, and we get just the construction staking contract, and that's the first we see of a project. In that case, we generally need to "go with the flow", and just use whatever system is already in use.

So that means that, as Surveyors, we basically need to understand ALL of these options, because we may need to use any one of them at any time. Eventually, though, I hope we get to the point where we have lots of "official" LDPs all over the place, like what Oregon is moving toward, and they start to gain widespread adoption. That would solve a whole bunch of problems we have now.

And if the industry starts centralizing on LDPs, then maybe we can start getting Autodesk to fix support for them in Map and Civil 3D, so that we have more support for them built-in to the software. That will definitely make things easier. 🙂

-- Sinc
http://www.ejsurveying.com
http://www.quuxsoft.com

[Anonymous]

Would you have any guidance on creating a custom coordinate system? My goal is to display Map data, images, shape files, etc... at a fixed scale of NAD83. We simply apply a scale factor to the NAD83 coordinates to create our ground coordinates.

[Sinc]

This might help:

http://www.quuxsoft.com/videos/DefineCustomGrid.aspx

Note that if you scroll down to the bottom of the page, I include some notes about creating the scale factor.

-- Sinc
http://www.ejsurveying.com
http://www.quuxsoft.com

[Anonymous]

Have you created 2 parallel custom systems Sinc? I've only used single
parallels for my custom systems, mainly because I'm not sure how to
determine what the parameters should be for a 2 parallel system. A 2
parallel system is probably overkill for a small project but it might be a
good choice for a regional system.

[Sinc]

In a sense, that's what I'm doing.

That step where I "multiply by 0.999999" has essentially the same effect. Basically, I identify a scale factor using my project elevation. Then I multiply by 0.999999 to drop the center of the projection a little bit below ground level, creating essentially the same thing as a double-parallel system. It's just easier to identify the parameters using a single-parallel Lambert system and scale factor, so that's what I typically do. And in my experience so far, it also seems to be the most common choice in general among people who are using LDPs, probably because it is so simple. That is, unless you are trying to align your LDP to a road or a river. In that case, you may want to use a transverse or oblique Mercator instead.

The two-parallel system is pretty much the same thing, except you pick your parallels directly. In that case, it just helps to keep in mind the general rules of thumb for distortion. Here's the max separation between parallels to maintain a given accuracy:

1:10,000 = 1°37' (0.9999)
1:100,000 = 0°31' (0.99999)
1:500,000 = 0°14' (0.999998)
1:1,000,000 = 0°10' (0.999999)

The number in parentheses is that additional scale factor used to drop a single-parallel system, and indicates the equivalent value to achieve similar accuracies when using that method.

-- Sinc
http://www.ejsurveying.com
http://www.quuxsoft.com

[Anonymous]

"Then I multiply by 0.999999 to drop the center of the projection a little
bit below ground level, creating essentially the same thing as a
double-parallel system." I didn't grasp that at first. It makes sense.

The main difference vs. 2 parallel is in the latter you can be more specific
in establishing the points where the scale facor is 1, if it matters.

[stacy.dunn]

Sinc,
How would you define a tranverse mercator projection in map?

[Sinc]

Basically the same way you would the single-parallel Lambert I use in that video.

First, select your datum or ellipsoid - at the moment, here in the US, we most-often use NAD83 as the datum. Then select Transverse Mercator as your projection type, and enter the projection parameters. For a Transverse Mercator, the parameters are as shown in the attached image.

-- Sinc
http://www.ejsurveying.com
http://www.quuxsoft.com

[phillip.hines]

Do you happen to have another link to your grid/ground paper? I'd like to read it but the current link is dead. 

[AllenJessup]

Unfortunatly Sinc hasn't been with us for quite a few years. Possibly @Jeff_M could help

[sboon]

I was afraid of this, so I posted a copy of it to this forum back in 2014.

https://forums.autodesk.com/t5/civil-3d-forum/working-with-grid-coordinates-richard-j-sincovec/m-p/4...

[Jeff_M]

I have preserved some of the demos and papers Sinc made HERE.

[phillip.hines]

Thank you.