Show messages: All
1-5
6-19
From: Michael Gibson
Hi Peer,
re:
> Is there such a thing as a "sheet metal body" in the STEP standard
Not that I know of. Maybe when SolidWorks has an object in sheet metal mode it writes out some particular attributes that their importer is looking for.
Is there any example STEP file that works?
- Michael
From: Michael Gibson
Hi Peer,
re this part:
> I'm surprised by the prohibition on solid bodies being stitched together since I didn't think that was a thing in the first place.
Yeah that's kind of a strange thing to say. Usually that term "stitched" means surfaces being attached to each other along a shared edge.
I think they just mean that they don't want multiple solids to be packaged together in a single STEP file.
- Michael
From: BurrMan
Hi Michael,
Yes, programs like 360 and solidwrks have a "Sheet metal Mode or workbench". The STEP file will contain other info like bend data and flattened entities.
From: pressure (PEER)
Hi Michael,
good.stp works, but MoI.stp fails
MoI.stp is the result of opening good.stp with MoI and then saving as .stp
The MoI header is ok: the problem is in the Data section. I created a hybrid file with the MoI header and the good data and that works, but using the good header and the MoI data fails.
- Peer
Attachments:
good.stp
MoI.stp
From: Michael Gibson
Hi Peer, what do you have set in moi.ini under:
[NURBS Export]
WriteAnalyticCurves=n
WriteAnalyticSurfaces=n
Do you get any different behavior if those are set to =y ?
Does "good.stp" come from sheet metal mode in SolidWorks?
- Michael
From: christian (CHRI)
hi PEER
" Have you successfully uploaded a 3D .STEP file to SendCutSend using that plugin ?"
No I didn't do it
What is important for me is not to unfold but to provide the manufacturer a correct 3D part
I let the supplier unfold
( the "losses" when folding and the unflattening parts depend from the folding tools witch will be used and the material quality ).
Excuse me for my bad english, i'm french
christian
From: pressure (PEER)
Hi Michael,
Now we're getting somewhere. The washer works if it's exported with analytic surfaces:
WriteAnalyticCurves=n
WriteAnalyticSurfaces=n
- fail
WriteAnalyticCurves=y
WriteAnalyticSurfaces=y
- success
WriteAnalyticCurves=y
WriteAnalyticSurfaces=n
- fail
WriteAnalyticCurves=n
WriteAnalyticSurfaces=y
- success
But, going back to the bent part we started with, .stp files of your "119 R063 sendcutsend3.3dm" and "119 R063 sendcutsend2.3dm" exported from MoI with WriteAnalyticCurves=y and WriteAnalyticSurfaces=y fail with the same error when uploaded to SendCutSend.
If I try uploading a .stp exported from "119 R063 sendcutsend.3dm" that I originally posted with the non-analytic planes generated by Sweep, but now exported with WriteAnalyticCurves=y and WriteAnalyticSurfaces=y, I get a new error "This STEP/STP file has an inconsistent thickness and cannot be manufactured."
Re:
> Does "good.stp" come from sheet metal mode in SolidWorks?
I downloaded it from McMaster-Carr. No idea of the software origin beyond what's in the header.
- Peer
From: pressure (PEER)
Hi Christian,
Re:
> the "losses" when folding and the unflattening parts depend from the
> folding tools witch will be used and the material quality
Yes that's exactly why I want to send the manufacturer a 3D model. I want the manufacturer to calculate the bending losses for me.
- Peer
From: pressure (PEER)
OK I got to the bottom of the problem. Surfaces generated by Sweep are non-analytic. Not just planar faces, but also the cylindrical faces at the bend.
Attached is a model made with Extrude and Revolve that works. Should have thought of this earlier since it's not the first time that a non-analytic result from Sweep has tripped me up.
- Peer
Attachments:
full analytic bend.3dm
full analytic bend.stp
From: Michael Gibson
That's great Peer! It would be cool to see some of the metal parts when you get them made.
- Michael
From: Mik (MIKULAS)
Hi Peer,
Does full analytic bend.3dm and full analytic bend.stp exported with this NURBS Export settings?
WriteAnalyticCurves=y
WriteAnalyticSurfaces=y
Hi Michael,
I would like to deeply understand of NURBS Export settings, may I ask you for your explanation in which case we should use "y" or "n" parameter in MoI ini file?
Thank you very much.
Mik
From: Michael Gibson
Hi Mik,
re:
> I would like to deeply understand of NURBS Export settings, may I ask you for your explanation
> in which case we should use "y" or "n" parameter in MoI ini file?
The case where you could benefit from switching it is if a receiving application works better with one or the other.
It depends on the behavior of the application you're sending it to.
When WriteAnalyticSurfaces=n then surfaces will be written to STEP format as a general B_SPLINE_SURFACE entity. When WriteAnalyticSurfaces=y then surfaces that are planes, cylinders, cones, or spheres will be written as a specialized entity type like PLANE, CYLINDER_SURFACE, etc...
Most of the time if the receiving app cares about analytic geometry they will examine general surfaces and automatically detect if they are actually planes, cylinders, cones, or spheres so it doesn't make any difference. But in certain cases (as Peer encountered here) that might not be how the importer works.
- Michael
From: pressure (PEER)
Hi Michael,
Re:
> It would be cool to see some of the metal parts when you get them made.
Sure here's a pair of sheet metal brackets I made by hand using MoI dimensioned drawings.
- Peer
From: Mik (MIKULAS)
Hi Michael,
OK, I understand.
Thank you.
Mik
Show messages: All
1-5
6-19