Fusion Educators (Read-Only)
Are you an educator who uses Fusion (formerly Fusion 360) in their courses in secondary and post-secondary? This is the official Autodesk forum for educators like yourself to share the success you are having with Fusion in the classroom.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Bar Bending Modeling with generative design

8 REPLIES 8
Reply
Message 1 of 9
mech18392087
710 Views, 8 Replies

Bar Bending Modeling with generative design

I am trying to model a bar bend, of a square 8x8 geometry which length is 120 mm, almost like the picture below.
So what I did is, I designed the bar, and the three poles. The starting shape I chose is the bar, the two poles under the bar had to be constrained, using the fixed constraint, and for the upper pole, I applied the force I wanted and a fixed constraint allowing it to only move up and down along the x axis, just like shown at the picture. All those areas had to be marked by preserve geometry in order to make them able to receive constraints and forces, therefore the final shape became attached to the poles which is far away from the modelling of a bar bend. Any ideas how to pull this through? Thanks in advance.

8 REPLIES 8
Message 2 of 9

@mech18392087 -

 

I am unsure I am following your questions so I will ask some clarifying questions. 
1. You mention you have started with 8x8 geometry. Are you just looking to understand the stresses and deflections in that 8x8 bar under a testing scenario as you have pictured above, or are you trying to design a lighter-weight shape that will carry that same load? 

2. You mention pulling through. Are you trying to form the bar around the loading point or just see how much it deforms? 

I think a key thing for us to figure out is if generative design is the right solution or if we should be doing a linear or non-linear stress analysis in the Simulation workspace. I think answers to the above questions will help us narrow down on the best toolset. 

 

Thanks, 

Mike Smell

Sr. Product Manager, Fusion 360

 

Message 3 of 9

Thanks for the reply. Let me clear up any misunderstandings. My goal is to create a lighter object using generative design, which will be tested at a bar bending machine, therefore I need to apply the same loads and constraints and conditions like the bar bending machine normally would. As you understand, in real life, the bar bends and slides across the two lower poles, as the upper pole gets lowered. So my goal here is to create a lighter object using generative design, but i need the object to be created for the bar bending test, therefore I need to approach real bar bending conditions.
Message 4 of 9
mech18392087
in reply to: mech18392087

the lighter produced design will later be printed and tested at the bar bending machine

Message 5 of 9

@mech18392087 

 

Thanks for the clarity. Given your description, I think you were on the right track with just a few small tweaks. 

The biggest thing I see in the image is that your have the preserves modeled outside of the 8x8 beam itself. Preserves should represent geometry that is part of the final design, not the fixture in this case. I would also encourage you to think about how you apply your boundary conditions to get a proper response at those contact points. Depending on how you constrain them, you may get higher or lower stress levels, so some experimentation to get the right response will be required. Here is an example of how I would approach a problem like this. I just put a random load on this, but this should give you an idea of what I mean about the preserves being part of the final shape vs. the fixture. I hope this helps. 

 

https://a360.co/3FROL7h

 

Thanks, 

Mike 

Message 6 of 9
mech18392087
in reply to: mech18392087

Thank you for your time.
I have already tried this approach. However the problem is 1. that it is not modelled closely to real life because it doesn't have the poles that the bar bending machine has and the loads and constraints wont be applied to the same little area as they normally would  2. The final shape will have those 3 blocks which are preserves 3 the two lower preserves wont be able to slide down and to the side just like in real life.
My goal is to manually contribute to the final geometry as little as possible and let the program produce the results by itself, while trying to approach real life conditions, i.e. the bar bending machine.

Message 7 of 9

@mech18392087 - 

 

If your test is going to deform the bar so much that the positions of the boundary conditions will change significantly then generative design might not be the right approach, as it is assuming linear materials and small deflections like a traditional linear static stress analysis. I have one other idea in mind using load cases that might help you get a bit closer, but it will still be making some assumptions and probably not be a super accurate final solution if a lot of non-linearity is involved. 

Message 8 of 9

@mech18392087 

 

Here is another sample to take a look at.  https://a360.co/3FROL7h Please look at Generative Model 2 and Study 2. I am using multiple load cases, with the boundary conditions in a different position in each load case to simulate the support on the bar getting closer to the ends as the load is applied. I also made a flat preserve on the entire bottom, which might be overkill and it might only need to be continuous for a distance at the constraint locations. I will leave that decision to you. I think this is a way to get close to what you are after. Remember, Generative Design is looking at linear static stress results to validate against your desired safety factor so if your physical testing is pushing the bar into the nonlinear region of the stress strain curve, that will not be accounted for in the shape you are generating. 

 

Good luck!

Mike Smell

Message 9 of 9
mech18392087
in reply to: mech18392087

Thank you for your time, however I cant accept this as solution because the manual contribution to the final geometry due to the preserves, is big. Also I don't think that a normal bar bend is entering the realm of non-linear static stress as the displacement is not that big, as If you were bending a fishing pole for example. Either way you provided great insight into the problem and that was very helpful.

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

You've got Fusion.


Autodesk Design & Make Report