Run this journal file and you will get several ACIS API message = inconsistent edge-face relationships errors. This is not an important error to me, but I want to inform the team:
reset
create frustum height 0.1 radius 0.081 top 0.02
Volume 1 copy move z 0.01
webcut Volume 2 tool Body 1
delete Body 1
move Volume 2 z 0.0510 include_merged
move Volume 3 z -0.0510 include_merged
vol all size 0.01
mesh vol all
delete Volume 3
imprint mesh onto body all
stitch body all
compress
The issue you’re getting appears to be down in the ACIS geometry kernel from Dassault. That may be hard to fix, but IDK.
Also, imprint mesh onto body only works for sheet bodies, not solid volumes, so it would probably be best to use build uspline from mesh but I see that also fails to build a U-spline.
I’ve also tried using partitioning the geometry (see attached journal file) and that has some success, however it encounters the infinite loop bug discussed in Difficulty with sweep mesh. There is a fix for that bug, but we’re waiting for a new Windows build of Flex – once I’ve got the new build I’ll test it out, unless @caleb wants to beat me to it.
hollowCone.jou (1.3 KB)
Thanks for the help. I have had the most luck by cutting the center out of the cone and cutting it into fourths.
Hi Kurtis,
I’m moving this over to a Jira bug so we can get it into a sprint.
The command imprint mesh onto body all was not added with solids in mind. I don’t believe it has been tested on solids at all. If you are trying to build a 3d spline on a volumetric mesh you should skip the imprint mesh command, and use the build uspline from mesh command instead of the build uspline command.
Also, currently we only support swept solids with a single source surface. Unfortunately this geometry is not well suited for a single-surface sweep. If the circular cap was cut out of the end of this cone, the rest of the geometry could be swept and a spline built.
Caleb, in that case should I be able to fit a volumetric spline if I cut the cone with a cylinder and then for an extrude on the resulting hollow cone? Like I did in this journal file.
reset
create frustum height 0.1 radius 0.081 top 0.02
Volume 1 copy move z 0.01
webcut Volume 2 tool Body 1
delete Body 1
move Volume 2 z 0.0510 include_merged
move Volume 3 z -0.0510 include_merged
webcut volume 2 sweep surface 8 perpendicular inward through_all
delete Volume 4
webcut volume 2 with general plane zplane move dz 0.1
delete volume 2
webcut volume all with general plane xplane
webcut volume all with general plane yplane
webcut volume 8 10 6 3 with cylinder radius 0.015 axis z
imprint all
merge all
mesh Surface 93 Surface 105 Surface 113 Surface 85
volume 10 8 3 6 scheme Sweep source surface 85 113 105 93 target surface 85 113 105 93 83 115 103 95 sweep transform least squares
volume 10 8 3 6 autosmooth target on fixed imprints off smart smooth off
mesh volume 10 8 3
vol all size 0.01
mesh vol all
build uspline from mesh
That should do it, except that you need to make sure that each volume is meshed with the sweep scheme. Some of those in the journal file above are submapped.
This variation on your journal file treats the hollow cone as one swept volume, and it builds the uspline fine, after a bug fix that we recently implemented. I’m not sure if you have that fix or not.
reset
create frustum height 0.1 radius 0.081 top 0.02
Volume 1 copy move z 0.01
webcut Volume 2 tool Body 1
delete Body 1
move Volume 2 z 0.0510 include_merged
move Volume 3 z -0.0510 include_merged
webcut volume 2 sweep surface 8 perpendicular inward through_all
delete Volume 4
webcut volume 2 with general plane zplane move dz 0.1
delete volume 2
delete vol 3
vol all size 0.01
mesh Surface 19
volume 5 scheme Sweep source surface 19 target surface 20 sweep transform least squares
volume 5 autosmooth target on fixed imprints off smart smooth off
vol all size 0.01
mesh vol all
build uspline from mesh
It does not build a Uspline for me so I do not have the latest version. I would certainly love to have the latest update if it is available for friends.
I have a couple of example problems I’d like you to take a look at. I may be able to provide a small amount of funding to your statement of work to do the problem. Can you give me a quote?
Here are journal files that I have made that will generate the geometries in trellis.
blank.jou (616 Bytes) cone_punch_die.jou (1.1 KB) sphere_punch_die.jou (364 Bytes)
This is an at least twenty character test post.
Kevin, I counted the characters in that statement.
As did I, because that looked like a lot more than 20 characters… I counted 34 37 characters… did I pass the test?
I fixed it so that all messages to this discourse form now appear in zulip, so that we can respond faster.
Testing external forum again.
