WHAT IS THE BASIC REASON TO DO SWAP EDGE IN SURFACE

[wardhanraj]

Basically when we have 4v or more points triangles can be formed in 2 or more ways.Does swapping edges will give different results?

[autoMick]

Yes, it makes a difference, particularly if you want to display the surface as contours. It depends a lot on the features of the surface, the pattern of your survey points, and what you are hoping to get out of it. Your edges are like little breaklines running between your survey points which are running either a ridge or a valley between those points. I always try and look at my survey points with an aerial photo underlay and check that my triangles are appropriate. You can control triangulation with swapping edge and adding breaklines (including proximity breaklines). You can get a much more accurate and more presentable surface with attention to triangulation. Use a surface style that displays triangles and contours, then experiment.

Cheers

Mick

[Joe-Bouza]

Yes, it proves that all contours are lies

[autoMick]

Yeah... and I wouldn't rely on surface slope arrows either. 

[cwr-pae]

I have found the surface slope arrows are always 100% correct ... for showing the direction of the low side of the triangles in the tin. 

[Jeew-m]

Dear Friend,

Get a paper which is rectangular. So we have 4 corners. Say each corner is a point. Let us assume top left corner to be point 1 then top right point 2, bottom right is point 3 and bottom left is point 4. 

Now turn the paper through the line which connects point 2 and point 4. You have 2 triangles.

Then make it flat and turn again through the line which connects point 1 and point 3. still you have 2 triangles

Feel the difference of triangles and overall shape you get each time.

That is what you call swapping edges.

Civil 3d does not know the actual site. But you know it. By swapping edges you can make real site conditions on your surface.

Thanks

[Anonymous]

ok, hello again.

Surface swapping edges allows you to cleanup problem areas, based on a site visit or just plain old common sense.

attached are three images.  two show the original surface - one with contours and the other with the triangles turned on.

the other image show the revised surface - with revised contrours.  Big differnence.

this shows what a big difference between the two - and they use the very same survey data. . . . . .

Knowledge is a dangerious thing;  use swapping edges carefully and preferrably after a site visit with the original topo.

best of luck

nonbeard13

[ESchomberg]

I have personally found that you can flip triangles and keep flipping them to try and make your contours look like you think they should look, but essentially it is partially guess work. I have also found that you can keep flipping until you have a big mess.

I personally have found that if you add 3D feature lines as break lines in the appropriate places such as grade breaks, top of bank, toe of slope etc. You do not have to flip any triangles. They come out looking like they are supposed to. The idea of swapping edges is that the computer does not know which elevations to pull from but in a sense, neither do we without careful study of every surrounding elevation. However the 3d feature lines guide the program to pull the contours to the correct place.

There may be rare cases where flipping triangles are necessary. But like I said this is my opinion based on limited experience drawing topos.

[AllenJessup]

One instance where I often Swap Edges is topo of an existing road. The Surveyor shoots TCs, BCs and CLs. Sometimes if they don't line up quite right you get a contour that looks like a boomerang on one side of the CL. Swapping an edge in that case can give you the more expected rounded V configurations. You're correct that care must be taken. If you try chasing a swap across a whole site. You're probably going to end up with trouble. I usually try to keep it to a couple of swaps. I stop at 3 or 4 if it hasn't made things better and then undo the ones I have made.

Experience does help. I started shooting Topo with a Transit and Stadia Rod then interpolating with a scale and calculator. I got to the point where I would interpolate the 10 footers and freehand the 2 footers. My old boss taught me from the beginning that "All contours are lies".

[jeff_rivers]

As another user posted in the images, there are times when C3D cannot figure out how to triangulate multiple points in close proximity to each other. 

I've noticed this most with corridor models where I don't get good triangulation around the curb returns, especially as the curb gets farther away from the centerline (and is thus not perpendicular to the centerline).  I'll always look carefully at my FG surface around the radius returns and flip faces as necessary.  I don't have any images to show, but if I did they would look very much like the ones @Anonymous posted. 

I would think the triangulation algorithm would make use of the corridor featureline codes, and know not to triangulate from a flowline point to a top back of curb point, but it doesn't, unless the corridor surface is built explicitly from the corridor featurelines.  

[jordan]

depending on the type of work you do it may be more useful at times compared to others. Say for instance you get a topo from a surveyor and he surveys the lip, flow line and back of curb. many times if he is lazy or doesn't catch it, the TIN will go from the lip of gutter to back of curb, then you end up with a broken flow line. Another scenario is if I am fortunate enough to get a surface from and engineer, many times we get jagged contours when between a building pad and road or ditch. This especially happens when I have a flat pad. I go through and smooth these before we make 3D models for our GPS equipment. When I added new feature lines, sometimes it breaks changes the TIN line representing a corridor, sidewalk, channel, etc... When this happens the crest of a ditch may get broken and the topo of the building pad will TIN to  the ditch flowline, and you end up with a broken crest.