Saving as iges file flips surface normals

[MIL-TFP-41]

Ok, this should not be tough, but it is giving me fits. When I save the attached revolved solid as an iges it flips the normals on half the surfaces. Plays hell with a cmm program that is looking for the top of a surface.

We try as a .stp file also, we get the same result in the cmm program (it apparently converts the solid into surfaces)

What really sucks is you can fix all the surfaces in the .iges file, resave as an iges, and it will still open up bad. I suspect this is because of the untrimmed surfaces that are hidden.

Just wondering if anyone else has seen this behavior ever...and if there is a fix.

test.mcam

[JParis]

This doesn't answer your question, sorry about that, it does though prompt a question....

Why use iges?

It's technically not outdated but it is an old process that is fraught with issues and limitations, I'm just curious.

[#Rekd™]

Igs or iges = I Guess 

[Aaron Eberhard]

Yeah, it's been a while since I've had to deal with iguess, which is great for me.  Sorry you're suffering, though!

It's been even longer since I looked at the spec, but if I recall, the iguess surface normal is simply determined by which "direction" the surface is described in (i.e., is it going left to right or right to left kinda thing).   You can see the same thing if you create this as a revolved surface in Mastercam:

That said, I'd recommend revolving this as a surface, then you can use Surface  > Normals > Orient to quickly align them.    Since they're native surfaces, the normal will survive the export to iges (see attached file).   The problem you have otherwise is you're always going through the parasolids kernel for surface manipulation, whereas if you deal directly with native surfaces the whole time you avoid conversion.

You can see the difference in file size if you Save As this solid as an iguess, it'll be ~500kb.  If you save a revolved surface as the same, it'll be 20kb.

surfaces.igs

[MIL-TFP-41]

23 hours ago, JParis said:

This doesn't answer your question, sorry about that, it does though prompt a question....

Why use iges?

It's technically not outdated but it is an old process that is fraught with issues and limitations, I'm just curious.

Like I said, we try to use .stp, but the cmm software converts those into surfaces and sometimes flips the normals on those models. At least with an iguess file one can open it in mastercam & see if the normals are flipped.  Next step is to try their parasolid translator & see if that gives us better results.