Slope Analysis - Reverse Face Normals

[itaylor]

NX 8.0.3.4
Using Analysis > Shape > Slope to check a solid model for tooling draft angle and undercuts, the results show many faces with the Face Normals needing to be reversed. Can we correct this for the entire model in one operation insted of selecting each face individually.

 

[cowski]

If you need to reverse face normals, I'd wager it isn't a solid model you are dealing with...

www.nxjournaling.com

 

[itaylor]

Thanks for the quick response.
It's definitely a Solid... Information > Object... Solid Body. I Analyzed the Parent Sewn surfaces that where used to create the main Thickened Body and I get a consistent result with correct Surface normals as one would expect.
The main body is a Thickened Sheet... 3mm thick with united Extrudes, thickened bodies etc.
Various features in the Analysis are shown in the attached image, I can't show you the entire part. We get the same Surface Normal prolem results on small simple Shelled bodies with united extrude features. I wasn't too worried about it personally as the Magenta / Blue told me it was the correct angle, just with reversed normals, but the Trainee Designers were pulling their hair out trying to get a correct result.
Regards
Ian

 

[cowski]

Make sure that only the solid is shown when you do the slope analysis. If the original sheet bodies are also shown and selected during analysis, you can get some strange overlapping results.

Also, use "examine geometry" to make sure there are no errors with your geometry.

www.nxjournaling.com

 

[SDETERS]

Have you looked into using the Molded part Validation tools for this?

 

[itaylor]

Done and done again just to be sure!
Examined Geometry, it returned a Consistency error... I fixed that, re-examined... Analyzed Slope... same outcome!!!
We are a large GM Design supplier.
Using TcAE 9.1.2.6 & the latest GM Toolkit. I'm 15 yrs on UG.

 

[itaylor]

We don't have Mold Wizard Licenses we've never needed them before.
The inconsistency only seems to have begun with NX8 although I didn't pay too particular attention when it began.

 

[cowski]

At the risk of sounding obvious, double check the pull direction vector?

Without seeing the file, I'm out of ideas. I'd suggest sending it to GTAC for a look if needed. I understand that they have confidentiality agreements in place to handle these situations.

www.nxjournaling.com

 

[itaylor]

Ok thanks, we'll persevere. If we find the answer I'll post it.
Have checked the vector may times over the weeks, had used the Absolute WCS selecting the vector and changed to the line I use for Draft direction with no difference.
Many thanks for all of your input.

 

[KhimaniMohiki]

Will draft analysis not work? Ive just tried slope analysis on a part with varying face normals and using draft analysis seems to do a pretty good job of re-aligning them.

Khimani Mohiki
Design Engineer - Aston Martin
NX8.5

 

[itaylor]

Yes I already did a Draft Analysis. The issue there is that some areas have 0.5, others 1.0, 3.0... draft and it will only tell us if the Draft is within the 0 - X Limit. Slope Analysis shows it all at a glance. Plus with multiple Sliders that also need to be checked, the Draft Analysis is slow to update on a large, complex part. Specify Interior Position doesn't solve it because of the part complexity. We've put the issue to GTAC.
Thanks again.

 

[itaylor]

GTAC are reviewing this 'Enhancement Request' for inclusion in future versions...!

Thanks to all

 

[cowski]

I'm confused, why are they considering it an 'enhancement request'?

www.nxjournaling.com

 

[itaylor]

Good point... hence my exclamaition mark in my last comment.
I didn't want to say something I shouldn't.
It worked fine in NX/.5 and before.
T Erbe (says hi!) is our NX Administrator and he was just as puzzled as you. He predicted it would be that kind of reply.
Who am I to argue with GTAC..!
Cheers

 

[JohnRBaker]

It's an 'Enhancement' because the code is problably working exactly as it was designed to work. Now you might want to argue (I've done it myself on many occasions) that perhaps the 'design' was not entirely correct to begin with, but officially, if the code, as designed and written, is working as expected, then any change can NOT be classified as a 'bug fix' since technically there was NO error. You have to remember that as an ISO compliant organization, we have to follow strict and documented procedures, which are subject to audit, and one of the critical issues is that we have to accurately and explicitly classify the REASON for every code modification.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.