Solidworks beam deflection and MDSolids

AI Thread Summary
The discussion centers on discrepancies in beam deflection results obtained from hand calculations, MDSolids, and Solidworks Simulation. While hand calculations and MDSolids yield similar results, Solidworks shows a significant difference, with a nearly 50% variation in deflection values. The user questions how Solidworks calculates deflection, specifically whether it employs a linear or non-linear approach and if it considers the actual geometry of the beam. It is clarified that Solidworks uses a non-linear finite element analysis that accounts for geometry, supports, and loads, which can lead to differing results compared to simpler methods. The user is also advised to verify results with other software or manual calculations and is provided with recommendations for further reading on finite element analysis.
brypie
Messages
1
Reaction score
0
Hi all

Sorry if this is in the wrong place...

I've been doing some beam deflections on a simply supported beam, with loads applied to the overhangs, and have used MDSolids, Hand Calculations and Solidworks.

My hand calculations match up with the results that I have got from MDSolids, however they differ significantly (in terms of percentage) when ran in Solidworks Simulation.
The difference is almost 50%, which equates to 0.15mm difference between the two. (I.E smaller deflection is 0.15mm and larger is 0.30) The larger deflection comes from the hand calculations and MDSolids.

I have used macaulys equation to calculate the bem deflection, which I have assumed doesn't actually take into account the geometry of the shape? It only takes into account the second moment of area, which for example, would be same for an 'I' beam or a piece of channel with equivalent dimensions.

Now, my question is this, How does Solidworks calculate deflection? Is it purely based on the second moment of area (which I think is called linear?), or does it also take into account the actual geometry of the beam (I think this is called non-linear?). If it is linear, then could I be getting a different deflection as a result of using fixed geometry for the supports? I have applied the loads in 3d, and ran the analysis in 3d. I have yet to try and treat the beam as a 2D model, as accessing Solidworks is particularly difficult for me at the moment. Or, is it as simple as I need to remodel the beam?

Also, if anyone out there can help me, would you please suggest some further reading material please?

Many thanks
 
Physics news on Phys.org
for your time! :)</code>Solidworks Simulation uses a non-linear finite element analysis approach to calculate deflections, which takes into account the actual geometry of the beam. The software also takes into account any additional elements such as supports, loads, etc. that are applied to the beam. Depending on the type of load being applied and the boundary conditions, the results can vary significantly from the hand calculations or MDSolids results. It is recommended to double check the results using different software packages or performing manual calculations.For further reading material, I would recommend looking at resources such as "Introduction to Finite Element Analysis Using MATLAB and Abaqus" by Saeed Moaveni or "Finite Element Analysis: Theory and Application with ANSYS" by Saeed Moaveni. These books provide an in-depth look at the fundamentals of finite element analysis and how to use different software packages to perform the analysis.
 
Back
Top