Learning Forum

[yungpv]

Hello people,

I'm new to Ansys or CAE in general and I decided to model a pulley or bollard system with ropes and friction. First I wanted to create a bollard model with a rope and check the results manually with the Euler-Eytelwein formula.

I faced a few problems. First I tried to model the rope as "cable" or "link" in Ansys by extracting it as a beam in SpaceClaim and changing it while PreProcessing to "cable" or link". This failed because Ansys didn't detect connections/contacts between the bollard and the rope.

Then I modelled the rope in SpaceClaim as "Solid". The connection was detected and I can choose the friction etc. However, the results with the solid are not satisfactory at all. Is there any way to model friction/contact with links or cables, mabye changing the solid afterwards? Or did I do things wrong with my solid modelling?

Explanations for the boundary conditions:

Displacement 2 : Top right face of the solid is held in every direction

Displacement: Shell surface of the lower left part is held in x,y directions in order to keep it straight

Fixed Support: The inner surface of the bollard/pulley is held

Force: Force acts downwards on the left side surface area of the "rope"

My problem here is actually that the fixed support in the bollard is bearing most of the load and not the reaction force in the rope, which is due to the "solid" behavior.

I'm sorry for the long question and hope that you can help me.

[peteroznewman]

nThank you for a detailed post. I think you will find the discussion linked to below relevant. It shows a coil spring torque clutch, which has a similar friction characteristic with the rope on the bollard.n/forum/discussion/10032/contact-step-control/p1nYou can see that beam elements were used, not solid elements. Beam elements are appropriate for a wire coil spring because the material is isotopic.nRope is not isotopic, it bends with almost no stiffness, but is very stiff axially. so you don't want to use beam elements. Link and Cable elements are appropriate for rope. These are not simple to use, but you have a good simple problem to learn how to apply them. LINK180 elements connect nodes that have only translational DOF so there is zero bending stiffness between two adjacent elements. That means they are only stable when under tension. If you try to push them, the solution fails, since you can't push a rope.nhttps://en.wikipedia.org/wiki/Capstan_equationnI read the Wikipage above and am thinking about the best way to see evidence of the Capstan equation in a simulation.n

[yungpv]

I really appreciate your answer. nI read your linked thread. At first I tried to model the rope as you said as beam in SpaceClaim and then changed it to truss or cable in Mechanical. But my problem there was that the contact was not detected by Space Claim, so it didn't work. Does Ansys not detect/model contacts between solids and beams? Or can I change the solid in Ansys afterwards to LINK180 but how would that work? My initial idea was to change the beam to link with only tension or cable to see the results there. nI appreciate your help and hope that I'm not doing too bad. n

[peteroznewman]

nnYou should read some of the posts about LINK180.nhttps://www.google.com/search?q=site%3Aforum.ansys.com+link180&oq=site%3Aforum.ansys.com+link180&aqs=chrome..69i57j69i58.1428j0j15&sourceid=chrome&ie=UTF-8n

[yungpv]

Array nHey, I'm back again. I managed to model the beam as LINK180 and define the contact correctly. Thank you, I really appreciate your help. nNo Seperation works totally fine here but the frictionless/frictional models do not converge. It does converge if I set the beam type to beam again but not with link/truss and no compression turnt on (which is then LINK180). nThe correct project is project B (bollard with rope )in the Workbench file. nKind regards nArrayn