Twist a solid body in NX 8.5

[Bilal91]

Hello, can I twist a solid body in NX 8.5 ?

Actually, first I had a sketch where I swept it along a helix. After that I used the Draft option but unfortunately the body is no more twisted as it was(it's an airfoil).

Can I know twist it again ??

 

[JohnRBaker]

It would help if we had either the actual model (that would be best) or at least a picture showing what your model looks like and some idea of what you're looking for as a result.

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.

 

[johnwizz]

Sounds to me like you need to have either a starting section which is at the starting angle you want, then an ending section that is at the finished angle. Or you may use 2 different guide curves, one for the leading edge and one for the trailing edge.

 

[johnwizz]

Here is a pic of what I am refering to with 2 guides. The trailing edge is a line along the Y axis and the leading edge is rotated 5 deg in the Y-Z plane.

 

[Bilal91]

Here is the file.
The airfoil now has a certain twist.
Why when I want to increase the Offset value (which is minimum now is 0.06 mm), the twist will not show as original ?

For example, I want the Offset to be +1.5 mm with same twist.

 

[johnwizz]

What makes you think the twist has changed when you offset the surface? It does not appear to change to me.

 

[Bilal91]

Yes it is changing, try to increase and decrease the Offset..

 

[JohnRBaker]

The problem here is math. The offset of a surface is not always what you think it is.

Yes, in certain cases everything works for you as you would expect that it would. For example if you were to offset a Spherical/Cylindrical/Toroidal/Planar surface/face then the results would another Spherical/Cylindrical/Toroidal/Planar surface/face. Unfortunately that's NOT the case when you offset a so-called freeform surface/face. For example, the offset of a NURB surface is NOT another NURB surface. It's something else, but it's no longer a NURB.

This is also true in the 2D world. For example, the offset of a Circle is another Circle, but the offset of an Ellipse is NOT an Ellipse. Now you might get away with small offsets and still get a result that is close enough, but eventually you'll reach a point where things fall apart. Now I don't mean that an operation will fail, just that it gets to the point where the approximations are no longer "good enough". I suspect that that's what's happening in this case.

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.

 

[Toost]

In this case it's not the Math that John speaks about, it's the intitial sketches which rotates when the expression "Airfoil_Thickness_Offset" changes.
This is indeed strange but it does not cover the fact that NONE of the two sketches are even close to be fully constrained, they are both fully free to move X and Y and rotate.
- Go back and make the sketches fully constrained.
In the attached image i have added a datum CSYS such that i have something to constrain against... Then the offset is 2 mm.

Regards,
Tomas

 

[Bilal91]

I used instead of Offset feature, the Thicken feature. It worked good.
Insert ---> Offset/Scale ---> Thicken.

But at the same time, could you please Mr. Toost attach here the NX file which you modified.
Thank you.

 

[Toost]

Here's the partfile.
Again I looked at it and the rotation is probably due to that the expression : "Airfoil_Thickness_Offset"
is used in both sketches as :
Chord_Length=37.7-2*Airfoil_Thickness_Offset
This makes the sketch move.

Then still both sketches are free to move XY + rotation. ( you can try open one of the sketches and drag something , it will move.)


Regards,
Tomas

 

[Bilal91]

Thank you