Quasi-static Analysis with Abaqus/Explicit
Вставка
- Опубліковано 12 сер 2024
- Quasi-static analysis is a useful tool for studying systems which undergo very large deformations. While the implicit solver in Abaqus/Standard would seem like the obvious choice due to the globally static response of the system, Abaqus/Explicit is much more efficient for this type of problem.
Consider a rigid punch contacting a deformable plate. The punch is moved at 2.5mm/min, covering a total distance of 12mm. Since explicit dynamics works with very small time increments, the quasi-static problem has to be condensed into a manageable time frame, otherwise the simulation could easily require several million increments to complete.
In this tutorial I will demonstrate how we can model the above experiment using the appropriate techniques which produces good quality results within a reasonable time frame.
Extremely Good video for Quasi-static analysis.. Thank you.
Thank you so much , have a good new year !
Thank a lot. It was very useful and fruitful.
Great stuff! I liked the part where you discuss energies and volumetric locking. Thanks!
Thanks for your great work!
very informative. thanks a lot
Great video.Thanks a lot.
very nice explanation, thank you
thank you very much for this video
Thank you so much
please i send you the zip file i hope to get your help. on the other hand are you have any other video really your approch is very amazing .
This video is very informative. Thank you so much. I have some questions in my problem. I have a RC slab of 120mm thick , simply supported at four edges and at the center of slab one isolated square column of size 150x150x420mm (150mm projection from both top and bottom face of the slab). I want to analyse the structure in quasi static analysis by introducing velocity through the column and required the result in the form of failure load and corresponding crack pattern.how is it possible in ABAQUS Explicit?
very wonderful video. Thank you very much please can you upload another video for multi layer material behavior under the same loading condition kind regard
Hi, thanks for the nice tutorial. I wonder if the Standard implicit solver would have the same solution as Dynamic explicit in this case. Have you done one with Standard implicit?
Dear Sir
Excellent Video. Would it be possible for you to provide us with either input deck or CAE file. I would use this as reference to model another problem which is similar.
Regards
Thanks for the excellent explanation, sir. I have the question, did you define speed on RP in the punch? Based on the description in this video, you say "The punch is moved at 2.5mm/min, covering a total distance of 12mm". Appreciate your answer
Thank you for your video.
I have I question to ask. If my simulation include the fracture or failure, and if the procedure of quasi static simulation is the same as your video? I have create a individual set of impacted part, and to evaluate the ratio of kinetic energy and internal energy, the loading displacement with the same smooth step amplitude, but the initial kinetic energy is not under 5% of internal energy. The fracture part flied away with velocity, although I want to simulate the quasi static experiment, the fracture with kinemic energy. Could you provide a video that illustrate this simulation that with fracture(solid element or beam element)?
sir
ur vedio is helpfull for me. and have given excellent informatiom.i have question on how to calculate mass scaling for the multilayer composites
if we would use total time instead of step time, the displacement rate changed? what would change?
Hi, thanks for video. Can u help me in modelling?i faced many problems
Hello. Thanks for the helpful video! I was wondering why you decided to do an Explicit analysis rather than a Standard one.
Thanks again!
No reason in particular. Both implicit and explicit dynamics will do the job, but I thought it would be interesting to show the mass scaling functionality.
Hello ! About Mass scaling, how do you define your factor ?
The factor should be the maximum value at which the kinetic energy does not exceed a few % of the total energy. There's no single value and usually has to be determined through a trial-and-error process. It's a balance between mass scaling and the the velocity of the punch (and thus the step time).
I have 3 questions if someone can answer I would appreciate it:
1- What is the point of using enceastre BC on solid elements since they don't have rotational dofs, isnt pinned is more accurate?
2- How does the solver recognize the relation between Reference Point and Discrete Rigid Part, I did not see any relation definition such as interaction given to them ?
3- Why punch has the slave surface and blank has the master one, shouldn't it be other way around since the blank deforms it should have slave surface and rigid punch should have master surface ?
Thanks.
Great video.. Thanks a lot.
PS: how much time did the analysis take to complete?
Hi Adity. It's so many years since I ran this analysis and I don't have the original files. The job was ran on a fairly average computer so I expect it completed within an hour.
Спасибо за уроки, а неявная динамика есть, у меня глубоко нелинейная задача, но медленно протекающий процесс - квази статика
Sir How did you calculate the mass of the indentor??
Could you please tell me if you have constrained the Reference Point of rigid body, if so what type of constraint did you use. Or just adding the Boundary Conditions Displacement would do.
I am doing something similar but with the CF.
- Thank you
I think this question is answered at 08:44. The punch is being moved in displacement control, therefore a displacement boundary condition is used.
how did you know you can reduce the K.E by factor of 63,000? at 22:40
Hello!
Is there any way to make that both elements deform?
HI Marco,
At 1:56, in the Create Part dialogue, you have the option to create the part as "Deformable". This allows for elements calculations on the punch.
L
Could someone tell me why the mass density is 7.87e-9? I think he used the units of Kg/mm^3, is it should be 7.87e-6?
Hi Huan,
Density is expressed in Tonnes/mm3 in the SI(mm) system and Kg/m3 in the SI(m) system. Since I used SI(mm) for this model, the value of 7.87e-9 is correct.
In abaqus , Kg is used with m however Tonne is used with mm. Hence, density units are (kg/m3) and Tonne/mm3 respectively.
Hi sir please I do this tutorial on Abaqus version 13 and complete all stage of building the model but when I submit the data for analysis it is not work and there is erro message come out Error in job formingpunch: Abaqus/Explicit Analysis exited with an error - Please see the status file for possible error messages if the file exists.
Job formingpunch aborted due to errors.please can I send the file to you to chek what error is don with my file because I cannot find the error soursekind regard
+ahmed alqayssi Hi, you could send a zip of your working directory to helpSpaceMATLAB@gmail.com. I don't support Abaqus users as a hobby so I can't guarantee an instant reply. However I'll aim to take a look :)
HSM
+helpSpaceMATLAB thank you very much sir for your help .
Thank you very much for your beautiful video. I have some questions. When we define the time period of step, do we need to estimate the first natural frequency of the model ?
2. How to get the contact force of plate, because some time designer want to get relationship between Force - Displacement ?
3. How to reduce the vibration like this figure ? I mean, make figure Force - Displacement more smoothly,...
goo.gl/photos/XLUKtNqivUvL4cut5
+Thắng Đỗ Quang Estimating the first natural frequency gives you a rough idea of the expected deformation to a certain and predictable wave shape, but a lot of it comes down to intuition. For a simple beam in bending (say the beam is impacted by another beam), it makes sense to use the 1st Eigenfrequency as a guide because the expected deformation mode matches the natural mode of the structure. However, for more complex geometry it's not necessarily as simple. The stable time increment is a more numerically robust guideline for determining the maximum increment (and hence acceptable step time), because it's related directly to the element characteristic length and stiffness, so it's still easy to estimate even for complex geometry.
The contact force can be extracted from an Abaqus/Explicit analysis using the field output surface variable CFORCE. This includes the normal (CNORMF) and shear (CSHEARF) components. For a given node you can then generate a history from the .odb field and plot it against the corresponding U component.
I can't give a specific answer to your third question because I don't know how the data was obtained/generated. Noise like this can originate from many sources (e.g. local instabilities, mesh quality, contact issues, artificial energy, poor time increment, insufficient material damping etc.)
+Thắng Đỗ Quang You could also consult this paper: www.bodietech.com/pdfs/DSP_for_Explicit_FEA_AUC1999.pdf which discusses methods for reducing noise in Explicit results.
+helpSpaceMATLAB thank you very much for your kind explanations. I'm more understand for these methods.