I am conducting an investigation into a composite structure which best lends itself to 3D analysis. I have created a global ply layup, then extruded this layup using the automatic tools.
Having created the zones I've noticed they are a bit... well odd. The image below shows typical zoning behaviour.
For what I consider to be a relatively simple laminate using straight forward lamina's (43 in total) NX has presented me with 136 different "zones" which range from 1 to 375 elements and don't seem to have any obvious structure / reasoning behind them.
I am fairly new to NX and this approach to modelling laminates so have a couple of questions.
1) What do the zones actually do / what are they for?
2) Can I manually control them so they are tidier to help me further on in the analysis?
Thanks for any help,
Solved! Go to Solution.
Each zone points to a physical property, or real constant, or section, depending on your solver.
It is normal for fibers to deviate when draped on doubly curved surfaces, creating more zones.
if your geometry is simple, if you do not have many ply drops or if you are not using the uni or woven solvers, then you should not have many zones.
A typical omission consists in not defining a consistent material orientation, so that the fiber angles change from element to element because the material orientation changes. Select your 2 D meshes, right click check material orientation. You can select a vector to define the MOV under mesh associated data. Then update your layups, recompute zones and you should se fewer.
you can also edit the laminate physical property and increase the angle tolerance, which will also create fewer zones at the expense of accuracy.
Thanks for your reply, I checked and yes, two of my faces had material orientations out by 90 deg to the rest of the laminate, although still consistant within the faces themselves. Other faces were all consistent with one another.
Having updated the mesh so that all material orientations are now consistent, and increased the tolerence from 10 to 20 degrees I have noticed the zoning has got stranger (see image below). Although I do have less zones (down to 58), those zones still vary significantly in number of elements and the location of elements within the zones.
The geometry is more or less three flat plates with the two outer at an angle to the central plate. I have 9 drop offs in total, but for of them are to simulate core tapering. The zoning doesn't seem to coincide around those drop offs. In the image above, all except the core layers are present.
One question on the tolerence, what impact does this have on accuracy? Is there a typical expectation / relationship between change in tolerance angle and solution outputs?
I should add I am using the Uni solvers, this is because the laminates we use are all UD's so it seemed sensible to do so.
Although being fairly new to the software I acknowledge there may be nuances that I am not aware of!
I just thought I would mention that the angle tolerence was indeed the issue. Obviously what I consider to be a large angle and what NX CAE considers to be a large angle was out of kilter.
Thanks for your help.