I have a Revit file with a rotation to true north of 20°. I have exportet two files to IFC, one with "Internal" and one with "Current Shared coordinates". When I linked the IFCs back into Revit I was surprised to see that the "Internal"-file was rotated and the "Shared ..." was not. In Solibri it's the opposite.
I found the rotation in a IFC entity called IfcGeometricRepresentationContext, but only in the "Internal" ifc file. Is this expected behavior?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for posting. Here's some more information about how project north is defined on an element in the two options. See if it makes sense or you disagree.
When you nominate internal or project base point as the basis of the element transforms, the true north direction and origin placement are defined on the representation context. If the shared coordinates are nominated, then this data is included explicitly on the element transformations.
I've attached some files to demonstrate this. Let me know if this at least helps in understanding where the information is defined, even if it's not helping yet with how other software process this (including Revit). You can test the import with my 3rd part importer, which can import relative to shared coordinates or internal (but I have to check if I had also an explicit option to consider or ignore the context setout).
For the "Internal" export I was expecting no offset of the geometry (It is as expected: IfcLocalPlacement = Point(0,0,0))
For the "Survey Point" export I was expecting both an offset (point) and a rotation (direction). The offset is as expected, but I also expected to see a direction. I guess it's because I assosiate the survey point with shared coordinates.
I guess it makes sense that when the geometry is rotated towards true north in IfcLocalPlacement, the angle to true north is 0 in IfcGeometricalRepresentationContext and vice versa.
To sum up: To me the export makes sense, except for the lack of rotation when exported with the "Survey Point" option. I didn't expect the geometry to be rotated in the import.
Thanks,
Einar
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thank you for the example files. I think I'm starting to get this now. Does my answer to Angel make any sense? I didn't realize that moving the PBP unclipped did anything to the IFC export. Did you use the official exporter? I sent you a request for a trial version for your Revit addin..
Thanks,
Einar
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've been helping Angel and his team with some improvements to the IFC exporter to Revit, and I was the one who primarily developed the changes we're seeing. Exporting with the shared coordinates as the basis should generate the same model as before these changes, when project north/true north rotation was including on the element.
If you can clarify a bit further what you think is not correct with the resulting files, we can consider if the implementation needs adjustment.
Cheers,
Jon
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
My personal preference would be that the "survey point" option should behave like the "shared coordinates". By that I mean the rotation should be included in the element. But that is only a feeling/personal expectation.
Thank you very much for the contributions!
Regards,
Einar
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have a Revit file with a rotation to true north of 20°. I have exportet two files to IFC, one with "Internal" and one with "Current Shared coordinates". When I linked the IFCs back into Revit I was surprised to see that the "Internal"-file was rotated and the "Shared ..." was not. In Solibri it's the opposite.
I found the rotation in a IFC entity called IfcGeometricRepresentationContext, but only in the "Internal" ifc file. Is this expected behavior?
Hi Einar,
Let me discuss this internally. Forgetting what happens when you link back into Revit, what did you expect to see exported into the IFC file?
Regards,
Angel
Hi Einar,
Thanks for posting. Here's some more information about how project north is defined on an element in the two options. See if it makes sense or you disagree.
When you nominate internal or project base point as the basis of the element transforms, the true north direction and origin placement are defined on the representation context. If the shared coordinates are nominated, then this data is included explicitly on the element transformations.
I've attached some files to demonstrate this. Let me know if this at least helps in understanding where the information is defined, even if it's not helping yet with how other software process this (including Revit). You can test the import with my 3rd part importer, which can import relative to shared coordinates or internal (but I have to check if I had also an explicit option to consider or ignore the context setout).
Cheers,
Jon
Hello Angel,
I guess it makes sense that when the geometry is rotated towards true north in IfcLocalPlacement, the angle to true north is 0 in IfcGeometricalRepresentationContext and vice versa.
To sum up: To me the export makes sense, except for the lack of rotation when exported with the "Survey Point" option. I didn't expect the geometry to be rotated in the import.
Thanks,
Einar
Hi Jon!
Thank you for the example files. I think I'm starting to get this now. Does my answer to Angel make any sense? I didn't realize that moving the PBP unclipped did anything to the IFC export. Did you use the official exporter? I sent you a request for a trial version for your Revit addin..
Thanks,
Einar
Hi Einar,
I've been helping Angel and his team with some improvements to the IFC exporter to Revit, and I was the one who primarily developed the changes we're seeing. Exporting with the shared coordinates as the basis should generate the same model as before these changes, when project north/true north rotation was including on the element.
If you can clarify a bit further what you think is not correct with the resulting files, we can consider if the implementation needs adjustment.
Cheers,
Jon
Hi Jon!
The export seems to be correct to me.
My personal preference would be that the "survey point" option should behave like the "shared coordinates". By that I mean the rotation should be included in the element. But that is only a feeling/personal expectation.
Thank you very much for the contributions!
Regards,
Einar