I've seen several other ideas posted with similar concepts. I decided to summarize what I would like to see, and some potential issues that users should consider.
I started by looking at the old MLINE object, and the MLSTYLE command which is used to define what these objects look like. Basically you define a series of offsets with individual colours and linetypes, and save that as a multiline style. When you draw a Multiline it automatically creates mitered corners. This feature is very old, none of the newer tools for editing polylines or featurelines work for this.
In concept, I'm trying to create something that looks like an MLINE, but includes elevation differences at the corners, using the same basic process as the Adjacent Elevations by Reference command. The user would specify the layer, colour, style, etc for each featureline and the offset from the previous one to build the FL set. These FL sets could be saved for reuse later. The user would select a master FL and apply the FL set to create a grouping like the ones in the sample drawing.
The closest thing we have in Civil3d to the MLSTYLE command is probably the band set dialog. The user would create a FL set in pretty much the same way by specifying an offset, vertical difference, style, layer, colour etc for each FL in the set. A FL set would always use the same site as the master that it's applied to. There would be no independent editing of the lines in the set, the only way to change them would be editing the master, or making changes to the FL set definition and reapplying it. Note – the ability to change a single offset or delta Z at a single location would be as a “nice to have” feature but not mandatory. Users could “extract” features from a set to get independent featurelines for custom editing.
As I was building a sample drawing to show what I have in mind I realized that there will be some issues at PI points which must be addressed. I think that it's important for the users to weigh in on how these issues should be resolved. The example that I’m using is a basic curb and gutter with attached sidewalk. The master FL is the edge of pavement.
Offset Delta Z
Master 0.0 0.0
Gutterline 0.40 -0.008
Top of Curb 0.05 0.15
Back of Curb 0.15 0.0
Back of Sidewalk 1.50 0.03
CHAMFERS:
At Point A1 I’ve created a basic chamfer for an outside corner. The two red lines are perpendicular to the master, and the delta Z values are applied at each PI on the white FL’s This means that the width of the curb and sidewalk is “wrong” at the corner but it does follow the basic rules. At point A2 I tried something different, so that the red projection lines are parallel to the master, and both the offset distances and projected slopes are maintained. I can see circumstances where either of these solutions would be appropriate, so we may have to include something in the FL set to specify which method to use. Inside corners are probably easier, point B is the only solution that I can see for this.
FILLETS
Civil3d already uses fillets in grading projection, and I don’t see any reason to change how the software would calculate in these cases. The only real oddity I found was the scenario at D2, where the radius of the curve doesn’t allow for fillets on the inner featurelines. At the top of curb we have two points which are both projecting to a single intersection; I used the average of the projected elevations for this, and then continued with a chamfer scenario similar to B.
MITERS
Civil3d already uses miters for grading inside corners, and the calculated elevations appear to work the same way that I did at point F. There is no existing method for calculating elevations on outside miter corners but if we use the same logic then the grade of all the projected featurelines is flat around the corner – see point E1 for an example. Another solution might be something like E2, where all of the projection slopes are the same, even though the distances at the corner are different than they are at the ends.
Steve
Please use the Accept as Solution or Kudo buttons when appropriate