Learning Forum

[glyn]

Hi Everyone, We are using Discovery Live, to model a real-world Three-Point Bend test. We conducted real tests to establish young's modulus of materials, however when we input these into the simulation we get diverging results across the loading range. The problem in summary: at lower loads the model deflects more than real life, and at higher loads it deflects less than real-life. Setup: It is a 2metre piece, the unsupported region spans 1.9m. Essentially we have a frame resting on two cylinders, and a load applied centrally in the midpoint of the span. We have modelled the supports as: cylindrical at one end, and planar on the other (across a small surface). We *believe* the planar support might be the misrepresentative boundary condition which is affecting the model. In an ideal situation we would like each support to be both cylindrical and able to move in a planar fashion. But we cannot see this as an option in the list of supports.   In the attached image you will see a side view of the setup. Note: we can see the ends of the beam lift up from the supports, however they are not symmetrical, with one side lifting higher than the other... again leading us to believe that different modelled supports are a source of this.   Any help / pointers in setting up a representative 3-point bend test would be greatly appreciated.   Thank you in advance.   Glyn

[Subashni Ravichandran]

Hello glyn_griffiths

A Cylindrical Support constrains radial motion while allowing rotational and axial motion on a selected cylindrical face while a Planar Support constrains motion normal to the surface on a selected face.

In this case, I would recommend using Cylindrical support at both ends.

While applying the cylindrical support, you may need to apply it individually to both ends.  

Please let me know if you face any issue with that.

[glyn]

Hi Subashni ,

 

Thank you for your reply - we did in fact start with this approach.

However, from our understanding of the behaviour, the cylindrical supports prevents the section from having any planar movement in the x axis. This would mean the body is subject to more external forces, it's effectively been asked to stretch between the supports, where as in reality it would slide somewhat?

 

Kind Regards,

 

Glyn

[Subashni Ravichandran]

Hello glyn_griffiths

Cylindrical support as mentioned constrains the Radial and the axial motion although it may allow some amount of sliding in the tangential direction.

With the available options, this is the closest representation you can get. I would recommend you to run the simulation at high fidelity for better accuracy.