Search results for query: *

Be careful when using C3D8I also - it is much more sensitive to aspect ratio (likes to be a perfect cube). Check your artificial strain energy.

No. See below for example. Again, this is clearly defined in the Abaqus Documentation, which you clearly did not read or did not understand. Also, you script is very clearly just a copied dump of the .rpy file, so you could also open CAE and make the change in there and then copy the relevant...

When you create the step abaqus automatically creates a field output and a history output object. By default, these will write output for a bunch of variables at every single increment at every single node/element (whole model). You need to modify this to only select whatever variable is of...

A more serious problem is that Abaqus will fail to mesh that part with the current settings, as indicated by the orange color. In order to produce a mesh of sufficient quality you will need to do some careful and clever partitioning (which is difficult but doable with python), or use tets.

Yes you're fine. Many posts here go unanswered, for any number of reasons I suppose. Probably because internet.

The documentation quite clearly tells you how to add execution options via the .execute command, I doubt this is new in recent version(s).... Abaqus 2016 Analysis User's Guide - 20.2.4 - aStudy.execute(): So, this would look like: aStudy.execute(token, execOptions='cpus=X')

I have tried similar in the past and was also unsuccessful, although it was more of fiddling around for me and not a requirement so I dropped it. You may try contacting Abaqus support who are usually quite responsive, surely they will give you a definitive answer.

Thanks Ryan. I just came back here to post same, you are certainly on the ball!

Hi, So I've successfully used a series of commands to generate a native solid hexahedral mesh (nodes and elements) by first creating nodes and then using Mesh->Between(F11) to define connectivity and generate the elements. However then when trying to automate the process with as an API, I get...

Could you try using python methods such as getByBoundingCylinder(), or findAt() if you know the coordinates?

Can't comment much on your failure criteria, you will have to figure that out. Reason for negative principal stress is because your beam is bending, and you have axial (along the length of the beam) compression on the top surface. Your mesh is not suitable for this, both element type and...

Maybe I'm missing something, but doesn't this model exhibit quarter symmetry? You could model one quarter and just use appropriate boundary conditions?

I agree Mustaine that is a better approach. I was just giving an option in case OP didn't want to re-run the analysis :)

I have done it manually with python by using the surface.elements connectivity, then screening out the nodes that aren't in surface.nodes. This won't guarantee that the final numbering is in any particular order that would correspond to a 2D element though.

The image above has quads only. You have quads and tets so that's one reason.

The same way you created the initial box. Extrude solid. Not sure why you are struggling so much with this. There are a wide range of ways to arrive at the geometry you're describing. Here's two: 1. Create larger cube, then partition top/bottom/sides with planes to create the internal volume...

To add to this - you are using a section integrated "before analysis", meaning linear. So from my understanding it doesn't make sense to use a nonlinear property with that, hence the limitation. A possible workaround would be to use much smaller elements with non-tapered sections integrated...

Glad it worked and you confirmed the direction at least. In terms of scripting to define multiple sections, you are the one who decides the number of elements. You should know the total length, and so the section definitions will be a function of the number of divisions that YOU decide...

Look at the deflection/stress distributions? You might need to use more than one element.

Fix one end and apply a load/moment or enforced displacement/rotation at the other end and look at the result. Should be obvious which direction the taper is in.