I believe that I have successfully “ingested” Forma data into Civil3D. My two coordinate systems are
UTM-12N
AZ83-CF
After “import” the geometry is still off approx 300m in X and 500m in Y.
I assume this is some kind of “grid to ground” conversion factor.
I assume that this is the “best we can do for right now” and that a manual “grid to ground” translation is required at this time.
Any thoughts?
Can you provide a bit more on the question? Perhaps a sample file to illustrate with?
I haven’t found any discrepancy beyond the accuracy of the survey data (i.e. if the GIS data rounds to the nearest meter, or the building heights are incorrect), so you might be missing a transform for the geometry.
I believe there is a “grid to ground” scale factor required to properly align geometry from FORMA to NAD83 grid systems. For example, if you import OSM data and FORMA data into C3D, they don’t line up. I think it is because there is a scale factor that might need to be contemplated. This is due to the curvature of the earth, and is a common “problem” that must be overcome with grid to ground survey data.
To test, you only need to import OSM data and FORMA data into the same C3D.dwg.
As a no Civil 3D user… how does one import OSM data?
The last time I checked Dynamo for Civil imports stuff around the “site origin”, not about a particular a coordinate system. As such you have to transform it from the site origin to the coordinate system you’re working in.
I believe there is a “grid to ground” scale factor required to properly align geometry from FORMA to NAD83 grid systems. For example, if you import OSM data and FORMA data into C3D, they don’t line up. I think it is because there is a scale factor that might need to be contemplated. This is due to the curvature of the earth, and is a common “problem” that must be overcome with grid to ground survey data.
To test, you only need to import OSM data and FORMA data into the same C3D.dwg.
The Forma geometry was translated properly using the ref point given from the forma site components. The resulting geometry was x=300M and Y=500m off. This is a similar outcome one will observe when converting geometry from grid to ground.
When I’ve worked with GIS data outside of C3D and in the context of forma this hasn’t happened. In fact it’s just lined up after transforming to the same coordaintes (either GIS to the forma project or forma project to the GIS.
That said, I’m not a Civil 3D user so statements like “import OSM data” might be leading to a discrepancy. Maybe try bringing one building of GIS data into forma (there are importer available), bring the same GIS data into Civil 3D, write a Dynamo graph to read the geometry in each and do the comparison. I’m guessing that will be similar to the Forma site transform (or whatever the name is now), but I can’t confirm as I don’t have GIS data to import, nor do I know how to do so in Civil 3D. If you invite me to a project I can try and look tomorrow though.
The Civil Nodes contains a node to import Open Streetmap data, which must have been downloaded first as OSM file. To import it in Civil 3D, a Coordinate System must have been set, else there is no conversion from Lat/Lon to projected coordinates possible. As far as I know, OSM is 2D so there is no grid to ground scale factor involved.
If there are differences between applications, maybe different Coordinate Systems were used, the applications might have different definitions, or other conversion mistakes or roundings were made.
The geometry has the same length and vertex count coming from each domain…..so I know that we “very close”. Here is a snap shot of what it looks like in model space.
How did you transform the geometry from Forma coordinates to project ones?
I used the X and Y translation from the FORMA SITE “REFPOINT” in dynamo.
I noticed some weird stuff with coordinate systems in Forma & Revit.
Basically trying to push models from Revit into Forma- despite coordinate system being set correctly in Revit and it working with other GIS systems, couldn’t get it to work in Forma.
I gave up on Forma for this & other reasons- it just does not seem like a production-ready tool just yet (I see it as having potential, but needs a heap more development)
Immediately, I do think FORMA is a great way to get aerial imagery into REVIT.
Additionally, if you put a “terrain pad” at a lower elevation like so…..it will give you the opportunity to populate the “void” with proposed civil terrain content in the context of REVIT.
@ddrennen2QDK2 : I do not think there is a linear scale between NAD83 and UTM12N. I think you need to use a coordinate transformation. Most of Forma currently uses proj for this and I’ve previously written about how to use it in Dynamo.
@zachri.jensen : has there been any conversations around exposing a coordinate projection node in the Civil 3D dynamo package? Would be interesting to have access to the functions built into Civil 3D. That would make these issues easier to deal with and more robust for the future.
Not that I can share publicly right now. This is the best I can do:
