View Single Post
Jul22-10, 06:27 PM
Sci Advisor
P: 1,498
Unfortunately I cannot give you any full text as I no longer have access to my old files at my recently-previous position. To be honest, the (code) sections above constitute essentially all of an input file that I was doing.

Let me try and clarify a little based on what your saying.

Yes, the bearing nodes are the nodes which correspond to the actual bearing-to-shaft interface on the shaft. I typically have this defined as a line component, and then I NSLL.

OK, to your second part, you must have missed the CERIG part. We are going to couple that non-fixed, mass element-added node which we will define as the master node in the CERIG command to all of the "bearing nodes".

Now, what is very important is that while we have elements only in 2D, we have nodes in 3D (see general axisymmetric elements and nodal planes). These additional nodal planes are created when we issue the NAXIS command. After we select our line component which corresponds to the bearing, we need to select the rest of the nodes around the azimuth that correspond to the same radial and axial location.

As far as setting contraints, it's really not that bad, because all we're going to fix is the "foundation". In the example I have above, I'm assuming that the there is only one "spring" between the shaft and ground. I could have easily coupled a few springs in-line to simulate the bearings, bearing housing, outer casing, stand, etc. In the end though, we're only going to fix one single node.