Tricks To Obtain Radius Values When Analysis Tools Fail 7

[Marlborough]

Curious if anyone else has any dirty tricks to obtain approx. radius values when NX analysis tools fail.

I place 3 points using "points on curve" and then create an arc segment using the "Start, End, Point on Arc" option and then measure the radius of the arc.

 

[SDETERS]

Could you try picking resize blend with the synchronous modeling this is if you are trying to find a radius of a surface?

 

[cowski]

When an analysis tool fails or returns strange results, the first thing I do is run "examine geometry". Usually it is a problem with the model that is causing the analysis tool to fail.

www.nxjournaling.com

 

[Marlborough]

Thanks for the reply. In most cases, it doesn't detect the face as a blend.

 

[jerry1423]

So is the edge a tolerant curve ?

 

[CAD2015]

SDETERS ,
Nice trick, thanks!

MZ7DYJ

 

[jerry1423]

and to add what SDETERS did, I sometimes use resize face in synchronous modeling to quickly see the diameter size

 

[JohnRBaker]

This is also what the Synchronous Modeling function 'Optimize Face' can do for you, but it allows you to slect as many faces as you wish and offers an option to first 'clean' the faces before 'optimizing' them. Note however that this will remove all the features from the body so proceed with caution (you will be warned).

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

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

 

[CAD2015]

"Resize face", (suggested by Jerry1423) doesn't always work.
Optimize Face is still a mistery for me; Unfortunately, the Help Menu doesn't provide any related info.

MZ7DYJ

 

[JohnRBaker]

'Optimize Face' is a sort of 'Heal Geometry' like function that you can use on-the-fly. Not sure that helps all that much, but...

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

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

 

[joycejo]

I use Derived Curve - Isoparametric Curve and select the face. If it is a cylindrical or conical face you will get an arc you can get information on.


John Joyce
N.C. Programming Supervisor
Barnes Aerospace, Windsor CT
NX7.5, NX9.0, NX10.0(Testing)
Vericut7.3.3

 

[BubbaK]

Which analysis function are you using, and what sort of "failure" are you seeing? Are you measuring radius of a curve, or an edge, or radius of a surface? There are many radius analysis functions, and I can't imagine why they all would fail. If you can get a radius by constructing a little circular arc, then the code could do that, too. It ain't rocket science.

 

[jerry1423]

You are right BuubaK, but there are quicker was to do it; I guess just trying to find the path of least resistance.

 

[JohnRBaker]

If you can get a radius by constructing a little circular arc, then the code could do that, too. It ain't rocket science.

Do you really think it's a good idea for an 'analysis' function to return information that is the NOT correct? After all, if an edge/face was NOT circular, giving you the impression that it was by reporting an explicit RADIUS value could lead to making assumptions which might result in problems down stream.

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

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

 

[braddles90]

I find extracting the edge then using the simplify function works pretty well. Tends to give you a bunch of different arcs that you can then measure the radius of pretty quickly

 

[CAD2015]

Mraddles90,

What do you mean by "Simply function" ?

MZ7DYJ

 

[JohnRBaker]

Under Curves, there is a function called 'Simplify Curves' which will try to replaces freeforms curves, like a Spline, with a series of best fit Arcs/Lines. It uses the Modeling Tolerance to determine how close the original curve does the new set of curves match. Note that this function does NOT produce an associative feature but rather a simple set of 'dumb' curves.

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

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

 

[CAD2015]

Thanks, Mr. Backer.
I knew about the "Simplify Curves", but I thought that Mraddles90 were thinking about something else.
I have developed a habit of experiencing all the option for achieving a specific goal.......

MZ7DYJ

 

[BubbaK]

John R. Baker wrote:

> Do you really think it's a good idea for an 'analysis' function to return information
> that is the NOT correct? After all, if an edge/face was NOT circular, giving you the
> impression that it was by reporting an explicit RADIUS value could lead to making
> assumptions which might result in problems down stream.

Radius of curvature makes sense for *any* curve. The value computed from the little arc is an approximation of the true radius of curvature, but, if done properly, the approximation will be a very good one. NX (and other CAD systems) use approximations all the time, and I don't see anything wrong with it.

The user is asking "what's the radius of curvature", and that question can be answered by using the arc-approximation technique. The user is not asking "is this curve circular" -- that's a different question.

 

[JohnRBaker]

You're correct, however I took the suggestion as being that NX should, without doing anything special, return WHAT the radius WOULD have been had the edge actually been an arc. That to me would be misleading. Now if you're willing to accept the 'instantaneous' radius at some arbitrary point on a freeform curve, NX has a function for that named 'Local Radius'. You'll find it along with the rest of the 'Measurement' tools on the Analysis Ribbon tab.

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

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