I recently ran into this problem with Carlson 2014 (2012 too). The 3d Breakline generated by the field to finish routine has a foot bust in it. Has anyone else ever run into this?
Thanks!
I digitized off of your screen shot and completely replicated your "problem". I put it in quotes because it is actually correct. Carlson is interpolating the verticality of the curve between the sloped approach and departure.
It has been my experience that you usually need a shot on the POC to minimize interpolations on things that matter.
Why would it not straight grade between the points that were shot in the field?
I recreated the basic layout using your coding and elevations and mine gives the correct elevations that you were looking for. I'm using Carlson 2013 on AutoCAD 2013. Are you using the Intellicad or embeded version?
Probably because if it did some people would say "why doesn't it interpolate the vertical curve?" 😉
Seriously, though, perhaps there is a toggle?
Stephen, his ahead tangent is pretty short, I think that's what's causing the hump.
I think Plumb Bill is right. Look for an interpolate/extrapolate option.
I agree that this looks like a vertical curve because of how it's drawn. If this were a road profile I'd agree with your line of thinking. However, this is a horizontal curve on a sidewalk. For the software to do anything besides a straight grade between points along a horizontal curve would be insane. Plus I can't find any controls under field to finish that have anything to do with vertical curves.
I agree Stephen, there is no way that this is a "feature". This is in 2014 running in C3D. I tested it in 2012 and got the same result. Carlson tech support informed me that they replicated it with my data in 2015 too.
I can only replicate it in a horizontal curve. If I strip out the horizontal curve control codes it draws the 3D polyline correctly.
It may be mostly a horizontal curve in reality, but sidewalks need to be drawn as breaklines and as such the software is attempting to take into account the possible 3D nature of the curve.
I'm curious as to why I'm not seeing the issue in 2013. Would you mind sending me your data? stephen at wardlandsurveying dot com
On the way.
> .... This is in 2014 running in C3D....
This may well be one of C3d's many "undocumented features".
We just replicated it on the 2012 stand alone version too, so it's not C3D.
With your exact coordinates I was able to replicate the problem in Carlson 2013 on AutoCAD 2013. After much tinkering and testing, it seems that certain combinations of horizontal curve geometry and changes in grade cause Carlson to go boom.
The top example is one I created where Carlson appears to create a straight grade from the PC to the point-on-curve and a different straight grade from the point-on-curve to the PT. This is how it should handle grade changes between points on horizontal curves.
The middle example is Edward's original example.
The bottom example uses Edward's horizontal coordinates with elevations tweaked by me. Notice the little tail sticking out to the right of the PT. My extreme elevation changes caused Carlson to overshoot the PT and loop back. See the zoomed in version of the PT below.
Based on that last example alone, I'd say there's an issue with how they're calculating grades on 3d polylines in field to finish.
All of the above examples are from Carlson 2013 running on AutoCAD 2013 created with a single run of field to finish so that I could be sure all settings were identical.
It could be the issue has always been there but is really only apparent with the heavily lopsided POC in Edward's example. Perhaps the extrapolation closely resembles the slope when the POC is more balanced. I've tested the interpolation along an arc in 2013 recently and it looked good.
It definitely only shows up with certain combinations of horizontal and vertical geometry. For example, if I raise Edward's PT by .40, it runs both sides of the curve as straight grades even though the shots are creating the curve are still unequally spaced.
I'd be curious to see how they have it programmed.
Interesting. That is what we found too. It is a combination of the horizontal curve, the placement of the POC, and the vertical deltas that make it happen. FYI, the reason that the POC was so lopsided was that there was a grade break there at the driveway approach. This needs to be fixed asap. Between this and the GPS vertical bust I found in SurvCE earlier, I am losing confidence in their software.