I was concerned when I read that the design of the composite cylinder appears to have been subcontracted to the manufacturer, who then appears to have used an add on FEA program to a well known modelling program for the analysis.
The entire system of cylinder, rings, end caps, porthole etc...
I can't see anything in the original post to suggest this was modelled with symmetry. In fact it might be difficult to hang the piece on it if each jig had two legs. I think the problem is you are applying the load 200" from the jig, whereas the piece will be relatively rigid. It would be more...
Alter the local axes at each joint so that one axis is parallel to the sling, and then add a fixity in that direction. Dont know Ansys workbench so it may be called something else like skew system, or local system etc.
If they are rigid links the model and results should be linear and proportional. The only explanation I can think of for the difference is if some of the loads have not been scaled up the same e.g. self weight loads or the composite action has been modelled as a staged construction with slab...
Am surprised that the composite bridge girder model you describe has any significant nonlinearity. I would stay with your initial instinct and check carefully there are no other problems with your model. I don't know what program you are using but are the "gap" elements intended to be rigid...
Interesting, and strange. I wonder if your time step is too big, its interesting to try it smaller, plus it converges a lot quicker. You will see at some stages there are very big deflections.
I think the cable element is the best element here anyway, and if you put compression limits in frame...
Thanks for the second model. I couldn't download the first model, maybe you can post it again. I notice some of the cables are slightly offset from the yz plane.
In the cable element model the insulator appears to be fully fixed at one end, translation and rotation.
In the frame element model...
Its not an ideal method, because as you have realised the localised thickness will be difficult to model well. Some codes have shell to solid transition elements, which may be a better way to go.
You need minimum boundary conditions in x y z directions otherwise you will get free body movement which will be a problem for the solver. If you balance the loads, the reactions at the 3 fixities can be minimal, or zero if you worked out the pressure very accurately.
Would applying the water pressure as a load (triangular pressure distribution) be appropriate, then have minimum boundary conditions to prevent rigid body movement?
Somebody here seems to have a similar problem. I don't use Abaqus but in my software one has to define the element group as isobeam and then axisymmetic shell as the sub type.
http://comments.gmane.org/gmane.comp.mathematics.abaqus.user/16965
If you are talking about load factors obtained from an Eigenvalue buckling analysis (as against a non linear analysis), you need to remember that these are only the ratio of applied loads to elastic critical loads, and do not consider the yield stress if this gives a capacity less than the...
In my code buckling load factor = 1/(1-eigenvalue), and eigenvalues are extracted in ascending order. Therefore if a load factor is negative it will only have been extracted after failure to obtain the requested number of positive load factors. In my experience in many cases this will indicate...
