Truss Thermal Expansion

FatmacFatmac
I frequently use thermal expansion of truss elements in my yacht rig models to vary tension of rigging elements. Generaly this presents no problems but some times when I'm looking for large changes in length of an element (over 100% or more) I run into convergences issues. Is there a limit on how big a change in length one can make? I have attached a simple model were I was investigating these thermal expansion affects on a single truss element and was interested to see that when I trying to double its length by thermal expansion the the analysis did not complete -can anyone help explain why? When I applied a small axial tensile force the operation completer

Comments

  • disladisla
    edited July 5
    100% seems a lot for an Elastic material. If you don't mind triggering nonlinear analysis you could try with any hyperelastic material for the truss.


  • FatmacFatmac
    Thank you Disla for the tip -i tried the hyperelastic details you suggested and indeed it does converge much better (see attached video). I'm unfamiliar with these types of materials and although the solution now converges I could not find a set of coefficients that produced a stiff enough material for my needs -perhaps it is not possible for hyperelastic materials to have high stiffness? It got me thinking about another work around -I've now gone back to using pre-tension with length adjustment for these large changes in length - not quite as elegant to look at (see 2nd video) but appears to do what I want and I can control the stiffness of the rigging element.
  • VictorVictor
    @disla, I seem to remember you found some problems with very high strains in linear materials before?

    Although that's usually physically unrealistic, in this case, the mechanical strain is near zero so I'd say it should be a reasonable approach.

    Models with no load like this sometimes don't converge. But that seems to be separate from the problem of using this in an actual loaded structure.

    Be careful with *PRE-TENSION SECTION because it doesn't update its orientation for large rotations. It looks like you can see that happening in your mpeg where two elements move out of collinearity but remain parallel.
  • disladisla
    edited July 6
    You can also try to add the dummy load as pressure. Like "Squeeze" the truss along its longitudinal surface. Small enough not to alter the overall capacity. (10Pa)

    @Victor. Right. Large Strains require other strain framework to get realistic results. Hyperelastic unfortunately has the default compressibility bug that has been reported when model is set up in SI units.
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register