sagging cable

barrti

Hi,

I am trying to get the sagging cable tutorial to work because I am trying to model a flying fox cable. I think its ok but have a few questions. The tutorial saggingcable.liml in the mecway folder is using square section steel but the tutorial suggests line elements? Which is best? Any pointers here? Can one apply forces in the middle etc?

Victor

Hello barrti

The line elements are what has the square section, so it uses both. You could also change it to circular to be closer to a real cable. The only geometric property of the section that it's using is the cross-sectional area.

Yes, you can apply other forces, for example to simulate the load on a flying fox. Just start it after the tensioning, such as at the same time as gravity. See attachment for example.

barrti

Thanks Victor
Is it possible (referring to your file) to provide an x displacement constraint at the end of the cable (i.e at x=10). This will give two fixed ends as well as a 1000 N cable pretension. If this was done, the final cable tension will be more than the initial tension.


BTW, I am very impressed so far with Mecway. It gives great value. I'm looking forward to more features whether via Calculix or whatever.

Do you plan to allow non-linear geometric analysis of plate structures sometime? I am interested in aircraft type structures which have stress stiffening occurring at times.

Cheers.

Victor

Here's my attempt at having both ends fixed but no extra tension. It uses the same trick of having a tension force that initially stiffens it to prevent unconstrained rigid body motion, but at the final time step, this artificial tensioning force is removed, leaving only gravity. I haven't validated it against any other result so I'm not sure if I overlooked some important physics or not.

This would represent a cable which, when it's not under tension, has the same length as the distance between supports. I'm not sure how you would add another tension on top of that.

I don't plan to include shells or plates in non-linear but that will be supported by the CalculiX interface.

barrti

Thanks Victor. I'll give it a shot.