fixing/ patching an imported solid

someone showed this to me. then i used the command myself. now i
can't find it.

i have an imported IGES surface or solid model (it was a solid in the
software that created it) that shows the "!" error.

and it was related to right-clicking, then using the menu that pops up.

but i still can't find it.

HELP !

thanks.

 

thank you, jk !

on this model, it shows it as not needing fixing. then i get the
message that says, under "what's wrong", "imported entity has 6 check
entity errors".

the IGES file i did this before, a bunch of the faces weren't
displaying. that was the time it showed it needing fixing, and "it"
fixed "it" (sWorks patched up my model so it looked allright with no
error messages.)

yeah, it's a good feature. guess i'm getting pretty biased in terms of
liking solidworks.

anyway, it might be an acceptable defect, on this new model. not sure
till i fix it, if it can be fixed.

on the imperfect model that is displaying, i'm using axis, point/plane,
to add an axis normal to the surface. again to find the center of the
spherical surface.

 

My suggestion would be to get a better export from the originating software. Say,
this was exported from Pro/e, there's no need for you to be messing around with
surfaces in SW if it was a solid in Pro/e. All you'd do is export it as an IGES
solid from Pro/e and that would be the end of your troubles in SW. All the solid
modelling kernels ~ ASUS, Granite, Parasolid ~ all understand B-rep, even if they
don't all agree on surfaces. (Lots of it has to do with how each handles accuracy,
not to be confused with precision, but related to it.) So, if, say you were doing
an IGES export from Pro/e, you'd check (instead of surfaces) the Solids and Shells
check boxes from the export interface.

 

Is this because some *systems* that *use* a specific kernel
or because certain specific kernels (ACIS comes to mind) don't
all support higher degrees of NURBS?

 

All the solid modelling kernels ...

Not sure what David means by that. Not sure if all systems support type
186 objects (in fact know of one, Rhino, that doesn't) or if it's a "kernel
function".

Actually, ACIS "supports" (by previously stated definition) higher degree
surfaces. I don't have any documentation but, once upon a time, did a
little testing by reading and writing a data set of degree 3 thru 11
surfaces. MDT 6 (believe it's ACIS 6) supports degree 7 max (converts deg
9 & 11 to 7). Inventor (ACIS 7) exports all unchanged.

-------------------

This biggest problem with IGES is synching export and import options, data
types, etc. Large percentage of the time STEP will be less prone to
trouble.

====================

 

The kernel must support the storage of the surface data but that
does not mean that the applications software (which must read & write
it) does. A kernel might support NURBS of degree 10, as an example,
but the applications software only degree 3. Such a system would
be limited to degree three I think.
Types 180-186 are manifold solid B-Rep entities which are not
directly needed for surfaces to be translated & reknit AFAIK.

That's going to change the actual surfaces as it cannot be an exact
mapping.

IIRC ParaSolid supports higher degrees. UG & other systems do.

What does SW support?

Probably less flexible too then <g>.

 

This biggest problem with IGES is

Could be. I don't know, but would venture a guess that STEP is just more
coherent because it was started after more data types were established
(some degree of standardization among the various major software
developers?). I've never been quite sure why there are so many different
ways of representing, say, a trimmed cylindrical surface in IGES (not so
much a statement about IGES as it is about my low level of understanding).

[What I do find incomprehensible; that most present day developers don't
bundle (at least fully AP203 enabled?) STEP translators with their basic
software packages. Money grubbers.]