Grading infill TIN lines pass over feature lines

[Citadel2012]

Can anybody tell me why the triangles of this surface (composed of infills only) pass over certain feature lines and triangulate nicely to others? I set the tesselation spacing at 1' and the angle at 3 deg in the automatic grading surface. The TIN lines pass over my face of curb and triangulate directly to the back of curb, which is causing contour mis-shaping.

My workaround for this includes adding lots of elevation points to each line; however, the feature lines have no elevation points other than the P.I.'s and I would prefer to keep it that way.  I am really wondering why the triangles are behaving this way.

(win 7 64 bit, C3D 2012)

[colton.determan]

The feature lines are on a different site. Make sure they are all on the same site and the triangles won't pass through the break line. The other possilbe problem is that your surface is set to exclude elevations less than X and your feature line at zero or something really different from what it should be.

[Citadel2012]

---

@colton.determan wrote:
The feature lines are on a different site. Make sure they are all on the same site and the triangles won't pass through the break line. The other possilbe problem is that your surface is set to exclude elevations less than X and your feature line at zero or something really different from what it should be.

---

The feature lines are all in the same site, which is the same site as the grading. My surface is not set to exclude any elevations. The feature lines are all at the correct elevations. As you can see in the original post's image, some of the triangles ARE hitting the face of curb line.

Like I said, I can 100% work around this by adding extra elevation points, but it is more important to me to understand why the triangles are behaving this way.

[jmayo-EE]

Are you sure that the FLine's close to each other? No gaps/leaky buckets?

Did you try lowering the tessellation values?

[Citadel2012]

---

@jmayo wrote:

Are you sure that the FLine's close to each other? No gaps/leaky buckets?

Did you try lowering the tessellation values?

---

I'm sure the feature lines close, hence why my infill is working correctly. The tessalation spacing and angle to start was 10' and 3 deg respectively; I lowered the spacing to 1' to better model tight radius curves, and changing the angle to 1 deg seemed to have no effect on the triangulation across this feature line.

I checked the linework to make sure there were no phantom vertices or zero length segments. 

It may help to say that I designed the surface by tweaking the face of curb grades and used the stepped offset tool to create a uniform 0.5' curb. The triangles DID pick up the elevations along the face of curb feature line until I created the offset; once the infills updated to the new feature line, the triangles jumped the face feature. 

[jmayo-EE]

Is the FLine defined in the srf as a breakline? If so try removing it or adjusting the build order, if not maybe adding it will get you going.

[Citadel2012]

It is not a breakline, but the grading infill treats all features in the grading site as breaklines. Changing the tessalation spacing from 10' to 1' does the same job as lowering the mid-ordinate distance if the features had been added as breaklines. I am trying not to add/change anything in the surface definition, because as of now the entire definition is the grading group and nothing more (real clean and stable). 

Is there some sort of preference for triangulation in grading infill objects? 

[Neilw_05]

Are you sure there is an infill between the 2 curb lines? You may have done something to cause it to dissolve. Maybe have a look in a 3D orientation.

[Citadel2012]

---

@Neilw wrote:
Are you sure there is an infill between the 2 curb lines? You may have done something to cause it to dissolve. Maybe have a look in a 3D orientation.

---

Yes, I am sure. I believe that If there was no infill in between face and back of curb, the triangles could not possibly cross. 

As you can see in the below image, this issue appears only in arcs in my feature lines. You can also see that there are no holes in the surface, or gaps without infills, or zero elevations.

[Neilw_05]

It's hard to diagnose without a file. Can you clip out a piece and post it?

[Joe-Bouza]

I think it is a leaky bucket inside a bucket.

Do grading have a presidence order, similar to surface definitions?

[david.zavislan]

Do you get breakline error messages in the event viewer when the surface is built/rebuilt? 

If you do, go into the surface's properties command.  On the Definition tab, set the Allow crossing breaklines build option to Yes. 

[Joe-Bouza]

I believe the OP has explained that the surface definition consist of a Grading object only. The breaklines are FL withing the gradings site and triangulation is not being honored

[Citadel2012]

We are waiting to hear back from someone from Autodesk about this issue.

In theory, one closed feature line can serve as the bounds of one singular infill object, which then can be modified by adding feature lines in the same site to model a project. Each one of those feature lines that is within the boundaries of the grading infill and is also in the same site would serve as a breakline, with the grading group surface controlling the triangulation between features.

With this being said, no feature lines need to close, nor should close in this situation, because they are all bounded within one grading infill. A feature line or set of features that closed would create a new boundary for which another infill would be needed to include it in the grading surface.

In my particular case, I have bounded several areas and am using several infills to create a continuous surface. I have found through some playing around that the curb will triangulate correctly if the infill in between in the front and back of curb is placed first and the infills in front and behind the curb are placed afterwards. If I create the infills in front and behind the curb first, then try to create the infill in between the face/back of curb, the triangles will jump the face of curb.

[Neilw_05]

I just encountered this exact problem when adding an infill between two curved wall featurelines. If I add the infill the triangulation ignores one of the lines. I can't find anything amiss in the linework. If I delete the infills on the outsides of the wall, the infill inside works properly. I've tried reapplying the infills in different sequences but it makes no difference. My workaround is to omit the infill between the wall lines and add the featurelines as breaklines to the surface.

This appears to be a defect.

[Neilw_05]

I decided to experiment and see if I could find a trigger to make the behavior stop. My first suspicion was a featureline that closed off the end of the wall. It was connected to a featureline that represents the end of a driveway at the top of the wall. When I broke the FL at the junction and made it a separate piece, the wall infill worked properly.

I don't have an explanation for the behavior but I suspect you might have a similar case.

[Neilw_05]

I've continued to run tests to see if I could find any consistency to the behavior, but nothing makes any sense. Even with the end cap as separate pieces the wall still exhibits the problem.

At this point all I can do is blame it on the quirkiness of gradings and move on.

Once again a lot of time wasted on troubleshooting.

[randyspear6624]

I have a similar issue but much quirkier...

I've got 6 lots in a row comprised of feature lines infills and grading features. The surface will not generate on one lot but that lot is not specific; usually the last one I pick to infill will not triangulate.

There doesn't seem to be anything wrong with the FL's or gradings as it will triangulate on all lots depending on the order I pick them. Like I said, the last one usually doesn't triangulate however if I delete the grading infill from a lot in the middle, then the last picked lot will triangulate. Unfortunately, redoing the infill to the middle lot does not get it's triangulation back for it.

Any suggestions?

C3D 2015 SP3

[bwolskisdca]

Would love to see an answer to this problem.  This is annoying bug.

[mzjensen]

This is an old thread, but I'm on C3D 2020 and still see the same issue (see attached GIF). Has anyone found a fix in the last few years?