Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

NX implicit modeling

Status
Not open for further replies.

BubbaK

Civil/Environmental
Feb 13, 2014
220
0
0
JP
NX 1953 has a new implicit modeling capability. Has anyone used it? I’m especially interested in the “general equation” option. How do you enter the equation? The documentation says nothing, except that some built-in examples are available. What do these examples look like?
 
Replies continue below

Recommended for you

Hi Bubbak!

If you get a chance, check out the interview with the PM, which includes some slides and demo footage of it being used.


The initial target is designers for Additive Manufacturing.

Some of the built-ins include Triply Periodic Minimal Surfaces (TPMS) for Gyroids, Swartz, Neovius, Schoen, and Diamond.

There are also implicit primitives shapes (Block, Sphere, etc...) and Edits (Unite, Subtract, Intersect, and Shell/Offset).

For General Equation, there are ~a dozen pre-defined equations or you can enter your own (e.g. using X, Y, and Z cartesian coordinates for the geometry, e.g. for the canned "Sphere" the equation is: x^2 + y^2 + z^2 -1.
 
Thanks @CADPat. The format of the equation is interesting. Looks like I’m supposed to write “x^2” for the square of x. Not many languages work that way. The only two I know are Visual Basic and Knowledge Fusion. So, I wonder how I should write cosine, sinh, square root, absolute value, etc. Am I only allowed a single line of code. If so, can it include a function call (that would be a huge plus). Again, nothing in the docs, AFAICS.
 
Isn't : "x^2 + y^2 + z^2 -1" regular NX expressions ?
I think this is based on how math is entered in Unix, I might be wrong there. If John Baker listens , he can hopefully clarify .

Regards,
Tomas




 
BubbaK said:
Looks like I’m supposed to write “x^2” for the square of x. Not many languages work that way.

I prefer it to the C# method of exponentiation...
I don't know about Unix, but I agree with Toost, it appears to follow the convention used in the NX expression system.

www.nxjournaling.com
 
I prefer x^2, too. I notice that one of the built-in examples uses "hypsin" instead of the more usual "sinh". This suggests that the syntax is either KF or a regular old NX expression (the two are almost the same, anyway).

Here's a few fun ones to play with

x^4 + y^4 + z^4 - (x^2 + y^2 + z^2)

2*y*(y*y - 3*x*x)*(1 - z*z) + (x*x + y*y)^2 - (1 - z*z)*(9*z*z - 1)

max(-x, -y, -z, x+y+z-1)

(x*x + y*y - 1)*z - x*x + y*y


Use box size = 2 for the first three, and box size = 3 for the last one.

But it looks like we only get one line of code, which is a pretty severe restriction. Bummer.


 
Status
Not open for further replies.
Back
Top