Beam / Truss usage

eric_c

Hello,
First congratulation for making a such easy to use FEM software (bought it yesterday).

After experimenting with beam and truss for a truss structure, the result was good. Then I tried to refine them x2 several time. The result was to soft. Could you confirm that it is wrong to refine truss or beam ?

Thanks

Victor

Hello eric_c

If you refine a mesh made of truss elements, the new elements will also be truss elements, so they'll have too much freedom to rotate and move. It sounds like you should make all of them beam elements and apply "flexible joint" at the ends which you want to be pin jointed. In that way, refinement will maintain the same behavior.

eric_c

Thanks !
Indeed I'm working on a telescope building.
I was comparing with result of this model (end of page)
http://www.cruxis.com/scope/scope1100_trusstubes.htm

He compare result with a formula and FEM model. In MecWay I got the same result if I'm not refining the structure. With refinement there is 3 time more bending. That was the reason of my first message.
After investigation if I remove the weight of tube (I'm using gravity in MecWay) with only a mass at the end of telescope the result are the same as the Cruxis one! This mean the FEM model used on Cruxis probably did not take in account the selfweight of tube, but only the load at the end of tube...

I've also compare with other Truss Serrier formula, the result are in between Cruxis result and MecWay.
So all in all, even if I don't know where is the right result. I realise that how the model is describe has huge importance. ( I'm new to FEM)

Victor

That's a pretty clear validation of both results, not to mention looks like quite a powerful telescope! The Cruxis page does mention not taking into account weight distribution of the tubes, so perhaps they lumped the tube weight at the ends.