[MBDyn-users] MBDyn and Ropes-Pulley

MBDyn users list mbdyn-users at mbdyn.org
Fri Sep 2 21:30:02 CEST 2022


Thanks a lot Louis and Reinhard for your hints about how to address the 
problem.

I had a look at the discussion on "Constitutive law for a chain". Looks 
remote from my use case. The three node beam elements for the rope / 
rigid bodies for the pulley / triangular contacts looks to me like 
overkill.

Algebraic constraint for the distances between the center points of 
several pulleys, looks the more promising...

All options seem to me at "MBDyn expert level". I will try fist an 
approximate  workaround with distance constraints to start with.

Thanks a lot to have taken time to explain and propose those solutions.

Best regards,

JM

Le 02/09/2022 à 21:16, MBDyn users list a écrit :
> Dear JM,
>
> You could use three node beam elements for the rope and rigid bodies 
> for the pulley. The most difficult part is how to connect the rope to 
> the pulley. One option would be to use module-triangular_contact which 
> can be found here:
>
> https://public.gitlab.polimi.it/DAER/mbdyn/-/tree/develop/modules/module-triangular_contact 
>
>
> It allows you to model unilateral contact and friction between several 
> nodes (e.g. all the nodes of your rope) and a rigid surface (e.g. the 
> surface of your pulley which is touching the rope). In your case you 
> should define a contact between each node of the rope, using the 
> radius of the rope cross section as spherical contact radius, and the 
> pulley it is attached to. The surface of the pulley must be meshed by 
> flat triangles (e.g. by using Gmsh). When you create the mesh for your 
> rope, you have to ensure that a sufficient number of nodes from the 
> rope is touching the pulley all the time. Otherwise you may loose the 
> contact between the rope and the pulley. So, if the length of your 
> rope is large compared to the radius of your pulley, this approach 
> would become quite expensive.
>
> Another much simpler and less expensive approach would be to develop a 
> new loadable element which imposes an algebraic constraint for the 
> distances between the center points of several pulleys. So, you would 
> create one structural node at the center point of each pulley and at 
> the end points of the rope. Then the algebraic constraint would 
> enforce that the total length of the rope is unchanged or you could 
> consider a certain expansion of the rope proportional to the tensile 
> load. The exact definition of the algebraic constraint will depend on 
> the topology of your hoist. I'm attaching a very simple example of 
> such an algebraic constraint. Of course the corresponding Lagrange 
> multipliers must be added to the equilibrium equations as well.
>
> Best regards,
>
> Reinhard Resch
>
> On 01.09.22 19:25, MBDyn users list wrote:
>> Dear all,
>>
>> I looked at MBDyn documentation and examples and did not identified 
>> functions to model and simulate ropes - pulleys assemblies (for 
>> exemple for mainsail hoist). Is there something that I could have 
>> overlooked?
>>
>> The use case is not straightforward to model as a "distance" as there 
>> are multiple pulleys attached to different locations, different 
>> purchase in those, with some angles between the different parts...
>>
>> Maybe this does not fall in the category of problems to be addressed 
>> by a tool like MBDyn.
>>
>> Best regards,
>>
>> JM
>>
>> _______________________________________________
>> MBDyn-users mailing list
>> MBDyn-users at mbdyn.org
>> https://mail.mbdyn.org/cgi-bin/mailman/listinfo/mbdyn-users
>
> _______________________________________________
> MBDyn-users mailing list
> MBDyn-users at mbdyn.org
> https://mail.mbdyn.org/cgi-bin/mailman/listinfo/mbdyn-users


More information about the MBDyn-users mailing list