Sub region causing contours to shift

[Parkinson]

Hi,
I attach topo with a sub-region that I did today.
I noticed that the addition of the sub-region caused some contours to shift. You can see the difference if you delete the sub-region.
My feeling is that sub-regions shouldn't do this.
I would love some feedback on whether this ia a bug or I have done something wrong. Either way I don't think it is intentional (or desirable!) for the contours to shift of their own accord.
Regards,
Martin

[Scott Hopkins]

I have also noticed this. It happens some of the time but not all of the time. Usually the shift is small enough that I can ignore it but I can imagine circumstances where this could cause a major headache.

[David Conant]

Revit draws contour lines after it triangulates the toposurface. It finds a constant height path across triangles and draws the contour. contour segments When a sub-region is created, Revit re-triangulates so that triangles don't cross the boundary. Since this causes some shift in the triangles, it can also cause some shift in the position of contour lines. The effect will be most noticeable for gentle slopes where a small vertical shift can result in a larger horizontal one. In the surface itself, the change in elevation of any point on the surface due to re-triangulation is never more than an inch or two (2-5cm).

[Parkinson]

Thanks David,
However some of the levels in the example I attached previously seem to alter by more than half a metre.
Could it me a units thing? The units are set to millimetres.
Regards,
Martin

[LRaiz]

Martin,
Units selection does not affect shifting of contour lines when making subregions. There are two factors that influence the distance of such shifts: (A) slope and (B) size of topo surface triangles before the split. Smaller slopes result in greater shifts. On the other hand the smaller the size of triangles the less contours will shift. This is all understandable because the precision of original surface model is limited by the number of points that define it.

The fact that contours shift a noticeable distance is not necessarily an indication of a problem. For example if original topo surface is defined by just a dozen of points but covers an area of a square kilometer and varies in elevation by only several centimeters then contours may shift a great deal. The drawings showing contours on subdivided top surfaces should be OK for as long as the inaccuracy of result is less than imprecision of the input and imprecision of the grading bulldozer blade. Nonetheless if you do want to make contours shift smaller distances then you need to increase the number of points that define topo surface especially in the neighborhood of split boundary.

HTH

[Parkinson]

Thanks LR,
That makes sense. I think I only had about one point defining a large area.
Regards,
Martin