Rhino *.3dm as import to VSP

167 views
Skip to first unread message

kenkad

unread,
Dec 6, 2012, 4:49:15 PM12/6/12
to ope...@googlegroups.com
Hello,
I thought that VSP import of fuselage shapes from 3D CAD were best as *.3dm rather than *.stl because the smoothness of the shape was better retained in this way (nurbs).  I seem to remember this from discussions at the Nashville AIAA conference.  So finally we are setting up some K-12 sessions using VSP and the students have their fuselage shapes free form created (Solid Edge 3D CAD) and got Rhino to convert the parasolid to *.3dm.  But, to my surprise when I downloaded VSP for the students, the import does not seem to accept *.3dm.  Anyone care to comment on what our next step should be?  Are we forced to use an *.stl file and if so how do we know what the resolution should be.  The students do plenty of 3D printing, but this seems to be a different issue.
Kenkad

Rob McDonald

unread,
Dec 6, 2012, 7:34:41 PM12/6/12
to ope...@googlegroups.com
Kenkad,

I'm sorry if there was some misunderstanding along the way. You are
correct, VSP does not import *.3dm or any other smooth surface file
formats. All of VSP's 3dm support is export-only.

VSP can only import 3D models in either STL (triangles) or HRM
(wireframe) format. The STL import is probably what most people would
find useful.

Of course, any imported file is brought in as a static entity -- it is
not parametric. Furthermore, the imported triangle mesh can not be
used for all of the standard VSP operations (mesh generation, etc).

I have had success importing STL files and then overlaying a native
VSP component and adjusting the parameters until I get a satisfactory
mesh. This is a slow manual process, but can work well if you need it
to.

We have research plans to help automate this process. This may be
what you heard about at ASM that made you think we had a more
comprehensive import capability. So far, all of that work is in the
future.

I hope this helps clear up what VSP can do for you,

Rob

kenkad

unread,
Dec 6, 2012, 9:00:17 PM12/6/12
to ope...@googlegroups.com
Rob,
I am not nor would I pretend to be a 3D CAD expert.  But, here in Huntsville, we certainly do have such people (Siemens PLC, etc.).  If the imported testelated file 
is coarse (large triangles that you would certainly not send to a 3D printer), then what does VSP do with the triangle end points to create a reasonably smooth surface?  I am a bit concerned with the statement that 'an STL file is imported and then native VSP components are overlaid" and that "this is a slow manual process".  K-12 students are going to be asking me (and the instructor) why we are going backwards in technology.  Free form in 3D CAD is the thing.  That is one of the most important attributes of the new SpaceClaim product and other 3D CADs are going in that direction as well.  That is one of the reasons I was also asking about a VSP person here in the Huntsville area.
Kenkad


Rob McDonald

unread,
Dec 6, 2012, 9:27:42 PM12/6/12
to ope...@googlegroups.com
Ken,

I wouldn't say that VSP is a step backward in technology.

However, VSP is not CAD. VSP will never be CAD. VSP does not want to
be CAD. If VSP wanted to be CAD, it would never have been created.

CAD is an incredible tool for certain jobs. I wouldn't dream of using
anything else for those jobs.

However, in my opinion, CAD is a terrible tool for some other jobs.
VSP is a different approach with a different philosophy and a
different design. VSP purposefully gives up on some of the strengths
of CAD in order to enable a fundamentally different approach.

In the design of every tool, tradeoff decisions are made. VSP's
reason for existence is the fact that it is not CAD.

As for your problem...

Lets assume someone is pulling in a STL file in order to use it as a
guide for making an equivalent VSP model. The resolution for that STL
file isn't very critical. The STL file just acts as a guide for the
VSP model. If the STL file is good enough for rapid prototyping, it
is more than good enough for this purpose.

However, if you are aiming towards a workflow that will involve a lot
of back-and-forth between CAD and VSP, then you are probably doing it
wrong. Usually, I see a workflow that starts with VSP and if a design
moves far enough along, it may make the transition to CAD.

The need for a CAD to VSP workflow is much more a specialty thing --
someone has a legacy model in CAD that they want to do a quick
parametric study in VSP...

Rob

Mark Moore

unread,
Dec 6, 2012, 9:51:33 PM12/6/12
to ope...@googlegroups.com
You need to back up and exPlain why you starting with a cad fuselage and importing it into vsp - what are you trying to accomplish?

It is far easier and faster to build the fuselage in vsp with parameters, with any cross sectional shape able to be modeled.  So you are going backwards from what most use this tool, where we quickly generate geometries in vsp (with even a 10 year old without any cad experience being able to do it) and then exporting to cad when more detail is required.

Mark


On Thursday, December 6, 2012, kenkad wrote:
Rob,
I am not nor would I pretend to be a 3D CAD expert.  But, here in Huntsville, we certainly do have such people (Siemens PLC, etc.).  If the imported testelated file 
is coarse (large triangles that you would certainly not send to a 3D printer), then what does VSP do with the triangle end points to create a reasonably smooth surface?  I am a bit concerned with the statement that 'an STL file is imported and then native VSP components are Poverlaid" and that "this is a slow manual process".  K-12 students are going to be asking me (and the instructor) why we are going backwards in technology.  Free form in 3D CAD is the thing.  That is one of the most important attributes of the new SpaceClaim product and other 3D CADs are going in that direction as well.  That is one of the reasons I was also asking about a VSP person here in the Huntsville area.
Kenkad


Reply all
Reply to author
Forward
0 new messages