modelling of RC bea...
 
Notifications
Clear all

[Solved] modelling of RC beam under 3 point bending test

11 Posts
3 Users
2 Likes
1,120 Views
Mahmoud
Posts: 20
Topic starter
(@mahmoud)
Paid Intern
Joined: 3 years ago

Hi everybody, I am a beginner user of Ls-Dyna. I am trying to simulate reinforced concrete beam under 3 point bending test according to attached file. I want to capture midspan load - displacement curve of beam. After analyzing force outputs for center point of loading plate are zero. I defined rigid material for loading plate.

Please get a guidance.

 reinforced concrete beam
10 Replies




Posts: 61
(@icebreakersours)
Lab Tech
Joined: 3 years ago

There isn't sufficient information in the post to give a good answer so I must ask clarifying questions instead. Does the analysis execute to completion? Can you visualize the deformation of the beam in D3PLOTs? What database cards have you requested? Do you know how to plot a force vs. displacement curve?

Reply
2 Replies
Mahmoud
(@mahmoud)
Joined: 3 years ago

Paid Intern
Posts: 20

@icebreakersours

I solved the problem this way:

I defined the nodes of concrete surface that are in contact with the loading plate as a set and activated the Nodal_Force_Group keyword for this set in order to recorded the force.

Regardless of accuracy of the simulation results about maximum load carrying capacity of beam, general shape of obtained graph is similar to experimental results as seen in the attached file.ِ Despite the hourglass setting this type of deformation exists in model. 

 

 

 

 RC beam
Reply
Mahmoud
(@mahmoud)
Joined: 3 years ago

Paid Intern
Posts: 20

the deformation of beam also is attached. and the requested database includes:

ASCII option: NODFOR, NODOUT, RCFORC

BINARY_D3PLOT and

NODAL_FORCE_GROUP

thank you in advance for your attention. Please provide any suggestion for further guidance.

 deformation
Reply




Posts: 61
(@icebreakersours)
Lab Tech
Joined: 3 years ago

Yes, you have got a solution. However, it is quite far from a credible one because of the obvious hourglassing. Reduce the rate at which load is applied, check the hourglassing control cards, element formulation in the section card, etc. and ensure hourglassing is gone before you can begin your journey towards a credible numerical solution. It is a bit like "happiness"; you never truly get there 😉

Reply
1 Reply
Mahmoud
(@mahmoud)
Joined: 3 years ago

Paid Intern
Posts: 20

@icebreakersours 

Happiness is a concept in human mind. 😎 

Thank you Sir.

Reply




Mahmoud
Posts: 20
Topic starter
(@mahmoud)
Paid Intern
Joined: 3 years ago

I am trying to simulate seismic behavior of reinforced concrete cantilever column. A constant axial load is applied at the free end of column. Lateral loading is a cyclic displacement that applies near the free end. I want to derive load-displacement hysteresis loops. I use Mat _ Winfrith (mat084) for concrete modelling. To consider the effects of concrete crushing in compression I defined MAt_ADD_EROSION with minimum principle strain at failure (MNEPS) as an erosion criterion.Finally the deformed shape of column after analyzing is completely illogical.

Can anyone please advise on this?

Reply
3 Replies
Mahmoud
(@mahmoud)
Joined: 3 years ago

Paid Intern
Posts: 20

the modeling file(k file) also is attached.

Reply
Negative Volume
Admin
(@negativevolume)
Joined: 5 years ago

CEO
Posts: 639

@mahmoud it looks like you created a new topic for this new question. Therefore, I am removing your response here. Also, please copy your follow-up comment with the .zip attachment to your new topic. 

@icebreakersours if you don't mind, could you copy your response to the new topic located below?

https://feassistant.com/forums/ls-dyna/modelling-of-reinforced-concrete-column/#post-1315

Reply
Mahmoud
(@mahmoud)
Joined: 3 years ago

Paid Intern
Posts: 20

@negativevolume 

Thank you. I am very eager to hear your comments on the subject.

Reply




Posts: 61
(@icebreakersours)
Lab Tech
Joined: 3 years ago

I have not had the chance to look at the job but, clearly, the results make no sense. Seeing the elements fly away makes me think that the input acceleration is too high which suggests the rate of loading is unrealistic. Use a smooth spline (or whichever function you prefer) to prescribe the loading so that the acceleration is physical. Now, it may turn out that the time scale is too slow and, if that is the case, then increase the mass scaling as long as the ratio of physical and non-physical energies continues to make sense.

Reply







Share: