I'm designing a hoist and inside this design is a wheel with a concave face (see attachment) for a steel cable.
I want to suppress the cable and put a fixture on the light blue face.
What fixture should be most suitable?
thx in advance.
Can you explain fixture, it is interpreted several ways depending what you are doing. Are you referencing a fixture used for a subsequent machining operation or else?
I mean a fixture, a constraint to connect the wheel with the surroundings, so no machining but a fixture you use in FEA
This is exactly something I will be covering in the next book! I work with chains-over-sheaves on a regular basis. The problem is similar - we have a tension-only element (rope, cable, chain) pulling on a free rotating body.
In this case it is stated that the sheave is to be a boundary condition, with a fixture on it. We don't have to be shy about fudging the geometry. In Solidworks Simulation cylinders are a very flexible thing. I put a cylindrical surface in the middle and put a fixed hinge fixture on it. This will fix the shape in space and only let it spin about the X axis.
One could also put a hinge fixture on it or pick DOF to constrain with an "Advanced(On Cylindrical Faces)" fixture.
Using the split face model, the fixture can be confined to the half where the cable is expected to lay. Here a uni-directional fixture is put on the bottom half, using the top plane as reference geometry.
This fixture allows the sheave to slide and rotate in the X-Z plane, so the stub shafts will have to be connected to or in contact with other machine members.
My favorite way to load a rotating member with chains coming off it is to literally model the chain halves and tie them to a body off in space.
The sheave is split again to get verticies on the inside of the curve. The fixture body is as wide as the distance between the verticies on the sheave, and it is several times that length away from the sheave. That body is fixed with a hinge, so that it can rotate in space but nothing else. The connectors here are 'extension-only' springs, but rigid links can be used if one is sure that the cables will remain in tension.
In this setup the cable forces are guaranteed to be equal* and the sheave is free to rotate (for small displacements). Stress peaks will appear at the point connections, so the sheave as a rated bit of hardware must be evaluated separately.
All these methods fix the sheave in space to some extent. They are only really good if the sheave is near the primary machine fixture. I think it would be more common to put a load on the sheave than to fix it, but this may be right for your machine. I also use this setup for connected pairs of hydraulic cylinders which truly are the machine fixture.
* - There is a critical flaw in rotating connectors in SW 2018 and 2019; forces will NOT be equal about a rotating body in these cases. Using general contact is a partial workaround. A fix has been promised in 2020 SPR#: 1122636.
I did something similar. Modelled a steel cable and fitted it in the concave opening and then put fixtures on the cable sections.
It worked for me
What are you trying to find out about the sheave? Lets assume you want to assess the structural suitability of the sheave to loads applied from the tension of some chain or cable with some wrap angle resisted by say a shaft in the centre bore. If it were me and the above is the situation then you have a bunch of options. You could apply the load as forces on a split line at the appropriate location with the appropriate angle. if you want the rotational degree of freedom free on the sheave then symmetry could be helpful. Or you could use inertial relief. To get the rotational degree of freedom to be free you can use a remote load/rigid restraint option where you can control all six dof at a single node - one of the most useful things in Sim IMHO. You could use the link connector to simulate the presence of cables between components as long as you ignore the stress riser so produced where they terminate and cable axial flexibility is not important. The springs Shawn uses also have the stress riser issue, the spring option being a better way to go if the cable element stiffness matters. There is a whole host of ways to get where you want to go. It would be a lot easier to get you there if we had a better idea of exactly what you want to know about the sheave, what the load case is and what sort of margin of safety you want at the end of the day. The biggest benefit of laying it like I suggest is that you will be asking yourself good engineering questions and often the right questions will lead you to the answer, maybe with a bit of help. If the FOS required is large, typically ~4.5 for hoisting gear, then the modeling fidelity become a bit less onerous. That said though, good modeling practice is always a good objective and getting good BC's and load paths is the hardest thing to get right in FEA, in my view at any rate.
Retrieving data ...