[OR] Template cross sections identifying linework not intersecting the cross section. Why?

Working on the circulatory alignment of a roundabout, I have multiple templates dropped:

  • where there is the circulatory roadway plus full curb and gutter adjacent to a splitter island (and usually, some bit of the splitter island)
  • where there is the circulatory roadway plus more pavement to tie to the approach and departures
  • where there is transition between the two above
  • some others I don't want to get into.

The problem in question is that at some splitter islands (but not all), where it should be finding lateral and vertical keypoints along attached linework, instead of finding this linework it for some reason connects to one of the other splitter islands where it is not in the same cross section or even close.

Here I have a good island and a bad island. In the first picture, the top left is supposed to have the surface of a splitter island. I could understand if it was a matter of just not being able to find the linework that should define the island segment, but I would then expect it to just terminate there or fail or something. I should not expect it to locate part of an island more than 5 feet away.

 x

  • What is the target for the end conditions of the splitter island template? Are you targeting a Linear XYZ element from the DGN file?

    Can you attach the DGN for this roundabout model so we can see how that corridor is set up and the template is solving to the point in a separate element down at the different location?


    For more information about the Road and Site design tools, visit the Road and Site design WIKI at: http://communities.bentley.com/products/road___site_design/w/road_and_site_design__wiki

  • I'm targeting a feature (xyz), not a specific element. I think I know where you're going with this, because I believe I've seen that if the target is a specified linear element, that target is connected, no matter the disposition. But I'm having trouble understanding why it should extend to any element that doesn't cross its cross section whether the target is specified or not.

    Other than a secure upload as I've done on another thread, I don't know how to attach files on the thread, so I'll do that in a minute. It'll be called 3d-circul-final.dgn.
  • I tried looking at the corridor but the corridor itself appears to be in a referenced model so I cannot look at the Corridor Objects to determine how it is getting constructed. Is the template in the corridor a single template for the circular area, and using a point control to tie into a curb linear object? Then it is shooting out toward the single point in that other Splitter island? Should these two areas that are shooting down to the splitter island at the bottom left continue to solve to some other object after tying into a curb?

    Can you zip all the reference files and attach them so we can see how it is constructed? You can either attach them through the secure file upload or if you select the option below your reply in the forum "Advanced Reply Editor" you can attach files or media to the post. You could attach the zip file here.


    For more information about the Road and Site design tools, visit the Road and Site design WIKI at: http://communities.bentley.com/products/road___site_design/w/road_and_site_design__wiki

  • No, the corridor isn't using a single template for the circular area in this file (though I've made that happen in another file in a more recent attempt to get this thing to work for me). I'm not sure why you wouldn't be able to see the corridor. Whether it is in a referenced model or not, it is in that same file. I can see the template drop boundaries in either model. What other files might be necessary? Edge of pavement? Alignment? The 3D break lines? Apparently the maximum upload size is 20 MB +/-. My zip is a tad bigger than that. I don't know what you need.

    Yes and no: I was using a point control to tie to a behind-the-curb point, but I was using a feature description target (xyz), not a named linear target.

    Yes, some backs of curb are shooting toward a single point in another splitter island. (This still happens in my latest attempt, too.)

    I'm not sure I understand your question "Should these two areas that are shooting down to the splitter island at the bottom left continue to solve to some other object after tying into a curb?"

    I think the answer is Yes: (Essentially) everywhere a back of curb exists, it should be able to find another point behind it representing the break in the concrete of the splitter island. Where it does find that break line, it should be drawing to where it intersects the cross section being studied. Where it does not find that line, it should stop at the back of curb.

    What is happening, however, is that in each template drop where I have the splitter islands, either all of the cross sections are drawn correctly or they are all incorrectly seeking on specific point another island's back of curb or splitter island break line. (The same or something similar still happens in my latest attempt, except that it must not be a template-drop problem alone, since as I said, in my latest attempt it's all one template drop.)

    I cannot comprehend that any point should be expected to be found except as a target intersects with the cross section.

  • I just realized I could do a secure upload. The file is called 3d-circul-final.zip.