Create TIN Zpts and "3D Breakline effect"

[jrwbell 0]

Hi Everyone.

I am new user of Create ZTIN Zpts and I have a query about "3D breakline effect".

The manual states:

"© If there are any points snapped to the line’s vertices, the elevations of these points are used to set the elevations at all the vertices generated along the line. In this way, a 3D breakline effect can be produced within the TIN. If there are no points snapped to the line, the line’s elevation is used giving the effect of a horizontal line."

My experiment to try and understand these was to try to generate a simple earthern bund. I used is a triplet of point objects with elevation at 5 m spacing (the centre "run of points" being the top of the bund), a polyline along the top of the bund and a polygon around the outside in a 2d_tin file (read in using Create TIN Zpts WRITE TIN == mi\2d_ztin_M03_Test_014.MIF).

I've tried the point objects in a separate file and embedded in the 2d_tin. I've also tried the polyline snapped to the first and the last node of the top of the bund, and snapped to every point.

The M03_5m_014_sh_obj_check.mif file indicates that the triangulation is working fine (and semed to work fine with points in separate file and embedded and with and without snapping the polyline to the points), but I was hoping to achieve a 2d_ZLN RIDGE effect.

My question is should I expect Create TIN Zpts to work in the same way as 2d_ZLN RIDGE works.

regards

Justin

[Bill Syme 0]

Hi Justin

Good question! And the answer is no.

When a TIN is created the Zpts are interpolated according to where they lie on the TIN's surface. Therefore if Zpts do not lie exactly on top of the ridge, the crest of the ridge will not be correctly represented. While for a line object, the crest of the line (ridge) is transferred to the nearest Zpts (the number of Zpts effected depending on whether the line is a thin, thick or wide line), therefore, the crest of the ridge is correctly represented.

In your case, if it's important for the breaklines in your TIN to have their crests correctly represented, then copy the relevant lines into another layer and use "Read MI Z Line" or "Read MI Z Shape" (making sure you've removed the keyword "TIN") after the "Create TIN Zpts" command. If your elevation points are in a separate layer, there is no need to copy these (just refer to the layer in both commands).

I hope that makes sense!

Cheers

Bill

[dannyduong 0]

Hi everyone,

With regards to viewing the TIN triangles in the sh_obj_check.mif layer, has anyone come across the issue of the triangles not showing up?

I've been running the same .tcf/.tgc but with the addition/removal of 2d_ztins at times. Sometimes the triangles show up and at other times they don't. Looking through the .tlf shows that the 2d_ztin has been read in and has a certain number of TIN triangles. The actual 2d_ztin shape does affect the .2dm, so it appears to be read in properly, but I'd like to check the triangulation.

I'm running TUFLOW 2010-10-AF DP (64-bit).

Any tips would be good.

Cheers,

Danny

[Sam Marginson 0]

Hi Danny. I'm encountering the same problem with regard to triangles not showing up in the sh_obj check file

I'm running 2011-09-AF-w64 with the command "Create TIN Zpts WRITE TIN == <relative file path & name>"

[peteraylett 0]

Hi folks,

As far as I can see, the first time you ask TUFLOW to to the triangulation, it will generate the triangles in the check files, but thereafter when it uses the .xf4/xf8 file(s) it won't add the feature into the sh_obj check file. As you've noted, you can still check it has had the desired impact by looking at the .2dm, or the zsh_zpts check etc. If you'd like to see the triangulation every time (not just the first time) you can either ask TUFLOW not to use xf files (XF Files == OFF), but that may slow down your initialisation, or use "Create TIN Zpts WRITE TIN == <relative file path & name>" as Sam suggests, to give you a triangulation you can look at in SMS (or wherever). If you just can't remember when you first read that file (and thus where the check files with it in might be), you could just re-save the input file, so the save date is more recent than the xf file, and run again and it will turn up for that run...

For my preference, if it could be made such that it's in the check files every time (please Bill!) that'd be good.

Hope that helps,

PHA.

[Sam Marginson 0]

if it also wrote a check file containing triangles & Z-points modified by TINs read in using the Read TIN Zpts command, that too would be fantastic!

[Jacky 0]

On 7/4/2012 at 0:45 AM, peteraylett said:

Hi folks,

 

As far as I can see, the first time you ask TUFLOW to to the triangulation, it will generate the triangles in the check files, but thereafter when it uses the .xf4/xf8 file(s) it won't add the feature into the sh_obj check file. As you've noted, you can still check it has had the desired impact by looking at the .2dm, or the zsh_zpts check etc. If you'd like to see the triangulation every time (not just the first time) you can either ask TUFLOW not to use xf files (XF Files == OFF), but that may slow down your initialisation, or use "Create TIN Zpts WRITE TIN == <relative file path & name>" as Sam suggests, to give you a triangulation you can look at in SMS (or wherever). If you just can't remember when you first read that file (and thus where the check files with it in might be), you could just re-save the input file, so the save date is more recent than the xf file, and run again and it will turn up for that run...

 

For my preference, if it could be made such that it's in the check files every time (please Bill!) that'd be good.

 

Hope that helps,

PHA.

omg, thank you so much. Clarifies everything. I've been pulling my hair out for the longest time ever trying to find out why triangles arent showing! These are the tips and tricks that are never documented in manuals and can only known be from experience and theyre the ones thats always so frustrating to figure out!

[peteraylett 0]

lol, you're very welcome Jacky! It's good to know to the old posts are still useful!