LS-DYNA TUTORIAL 14: Delamination Test and Cohesive Elements

[@dakshitha92]

Oh my god, I just learned the filter option in prepost. What a lifesaver!!!!! Thank you

[@AmeenTopa]

No problem man. Life has been very busy lately. More videos will come after I am done with some paper publications and reviews..

[@kaanferikel]

amazing stuff thank you so much for this video, I will use the info I learned from this video to apply it on to some balistic stuff. Thanks mate

[@AmeenTopa]

Thank you! I hope you found my work useful!

[@athina46]

Dear Ameet, thank you for the video. I have a question concerning the MAT138 outputs. I am interested in studing the forces on the cohesive layer. If i model using the Contact algorithm i can study the forces from RCFORC. How can i do the same for the cohesive model? Thank you!

[@AmeenTopa]

You are welcome. If you are using cohesive model, then you can try using ascii nodforc. You can check the manual for more about this keyword.

[@kamerun_o_mulema4349]

Hi, using the second approach with the tiebreak contact what is the tensile failure stress? If we want to use the tiebreak contact with two different linear elastic materials, which failure stress (NFSL) should we insert as the tensile failure stress? Thank you.

[@AmeenTopa]

You can read more about this in my blog entry here: lsdynauniverse.blogspot.com/2021/03/delamination-and-tiebreak-contact.html

[@mateuszjuzun3915]

Thanks for sharing tutorial. Which group of consistent units is used in your model?

[@AmeenTopa]

Kg, mm, ms. My favorite units system 😀

[@sadamegeri]

Thank you very much. It's another great video

[@AmeenTopa]

So nice of you. Please share my videos so they can reach and benefit more people!

[@loukasath.9787]

Thank you for the helpfull video Ameen! I am trying to model a single lap adhesive joint with the same methology you explained in the video. However, the analysis doesn't start because of the error "The bottom of cohesive element ... is not attached to an adjacent shell, tshell or solid element - nor - is part of a tied or tiebreak contact definition" for all the adhesive elements, even though I have merged the nodes between the substrates and the adhesive. Do you have any idea what I'm doing wrong? Is it neccessary to use a tie contact?

[@AmeenTopa]

From the error message, I believe there are some cohesive elements with "free ends". All cohesive elements should be connecting between two non-cohesive elements.

[@loukasath.9787]

@@AmeenTopa Thank you, that was the problem! I was using two layers of cohesive elements. Also, have you made any video in which you reduce the computational time of the simulation? As I said, I am modelling a single lap joint with a wave-shaped overlap area, the model consist of 70,800 elements and the total time is around 5 hours, but i think that it's quite a long time (I'm a new user)

[@AmeenTopa]

@@loukasath.9787 I don't have any videos on that specific topic. The computational time depends on the timestep size and termination time. Smaller timesteps are necessary when the simulation is dynamic and the explicit solver is used. For long duration events (more than 1 second in real time), the implicit solver is used which allows for bigger timesteps. Check my tutorial on axial tube crush.

[@周泽宇-q6q]

哦，天哪，这个视频对我来说太及时了，首先感谢你的视频！最近，我一直在研究粘性冰材料和结构之间的碰撞。我想问您，在六面体八节点冰单元的内聚力模型中使用了什么 ELFORM？再次感谢！

[@AmeenTopa]

@@周泽宇-q6q sorry, I don't understand.

[@ZooHanoi]

Hello Ameen, thanks for the video. Which solver did you use to run the simulation?

[@AmeenTopa]

LS-DYNA SMP Double Precision R11

[@seyhanokuyan277]

Can I use the cohesive elements as mortar between bricks for any masonry shear wall? and also Can I use MAT_193 for cohesive elements?

[@AmeenTopa]

You can use MAT DRUCKER PRAGER but the cohesive elements cannot be in ELFORM = 19 or 20, as I did in this video. You can use MAT193 with ELFORM = 2.

[@edeplessnesss]

Thanks Ameen. many papers perform cohesive zone modelling by writing a subroutine. Is there any way of this in Ls-dyna?

[@AmeenTopa]

Hi, do you mean user defined material? If yes, then I don't know of anyone who did this before but I believe it is possible.

[@mohammadraza9381]

Thanks for this tutorial Ameen, I am currently trying to model an adhesive joint but because of a mismatched mesh between the adhesive and the adherend I'm having issues running the simulation. Do you have any tutorial on using tied contacts to connect two mismatching meshes?

[@AmeenTopa]

You are welcome. Check out my tutorial on Concrete Wrapped with Composite Laminates Also, my tutorial on Three Point Bending

[@mohammadraza9381]

@@AmeenTopa Hey Ameen, I checked out your other tutorials and created my model. Unfortunately, when I run my model it stops after a certain time do you have any idea why that might be happening?

[@AmeenTopa]

What does the error says?

[@mohammadraza9381]

@@AmeenTopa There is no error really. It goes up to 29010000 on the left hand side column and just doesn't progress further. My email is [e-mail] if you want to contact me so I can send you a screenshot. Thanks

[@AmeenTopa]

@@mohammadraza9381 sorry I just noticed your comment. Have you figure out the solution?

[@noamaanshaikh5702]

hi, can i use abaqus inp file in LS-Dyna only to use as a solver? i have CZM in my abaqus model.

[@AmeenTopa]

@@noamaanshaikh5702 as per my knowledge, you need to convert the model to k file format first.

[@raphaelfaustino2607]

Nice material, Ameen! Is there a way to create mesh for coesive in order to be finer than mesh for the plates? I know that for same mesh size is possible as you did...thanks

[@AmeenTopa]

You can make a finer mesh of the cohesive and connected to the plates using Tied Contact. However, using this approach, the nodes of the cohesive part are not merged with the plates. Thus, you cannot follow the same steps as I used in the tutorial. You need to create the elements for each part separated (cannot do in step). With that being said, for Delamination test, it is always advised to have the same mesh size for both cohesive elements and plates.

[@raphaelfaustino2607]

@@AmeenTopa I'm trying to model a drop test with delamination, using TSHELL or SOLID for plates and solid Coesive elements for interface. For tiebreak contact it worked fine but for the another aproach for delamnination(coesive elements) with material model, like MAT_138, I have some doubts about contact Between Indenter/plates and interfaces. Is that possible to send you my model? I tried: automatic_surface_to_surface (Indenter and plates) Eroding_surface_to_surface( plates and cohesives / plates and plates) Must I create some tie contact between plates and interface?? Thanks a lot for the help

[@AmeenTopa]

You mean the same model as in this video? Sorry I don't have it anymore as this was very old tutorial.

[@AmeenTopa]

If the plates and interface have merged nodes at their boundary (like on this tutorial), this no need for tiebreak or tied contact

[@raphaelfaustino2607]

@@AmeenTopa thanks a lot for your help!

[@bengisucetinoglu5647]

HI, thanks for video. I became an error and it is*** Error 20264 (STR+264) Material 10 using *MAT_COHESIVE_MIXED_MODE has inconsistent material properties resultin in a negative displacement for the peak traction for mixed mode loading. u/L= 0.0000E+00 is smaller than 1. I cant find an answer. Could you please help me?

[@AmeenTopa]

You can check the user manual vol 2. You will see how this u/L is calculated.

[@bengisucetinoglu5647]

@@AmeenTopa Thanks a lot for your answer😀 you helped me a lot with your video.🙂🙂🙂

[@mono7891]

Great video. Thank you

[@AmeenTopa]

You are welcome.

[@sazidkhan4796]

Nice work . Can you make a video on compact tension (CT) specimen of fracture toughness testing and how we calculate j integral in that specimen

[@AmeenTopa]

Dear Sazid, fracture mechanics is very deep topic and I have no strong background about it. For your reference, please check the following: www.dynamore.de/de/download/papers/2015-ls-dyna-europ/documents/sessions-e-5-8/non-linear-fracture-mechanics-in-ls-dyna-and-ls-prepost/view

[@sazidkhan4796]

Thanks . Your videos are very useful. Keep making new videos 👍

[@AmeenTopa]

@@sazidkhan4796 you are welcome bro. Sure will do 😀😀

[@senabeyter421]

Hello Ameen, I did tie break contact shell dcb model but when my model is not peeling out from cohesive part. Separation is only from initial length. How could i solve this problem? Could you give me advice please?

[@AmeenTopa]

Hi there, here things that few things that you need to check: Initial penetration between the shell parts The values of the normal and shear strength/strains are accurate. Offset between the shells. Also, if both beams are shell, you can try constrained_tied_break where you need to put only the value of strain at failure

[@senabeyter421]

@@AmeenTopa Thank you very much Ameen

[@senabeyter421]

@@AmeenTopa Ameen I want to ask a final question. This card worked in tiebreak but if i use for cohesive shell dcb, is it available similar card? Because when i look initial penetrations, automatic fixing changes my model shape and manual button can not be activated in model editing. Thank you for your patient.

[@AmeenTopa]

If you are modelling the two beams as shell elements, you try: 1 - Contact tiebreak offset, 2 - Constrained tiebreak, or 3 - contact tied offset with cohesive elements between the two parts.

[@aminetoumi4009]

Hello , Thank for the tutoriel . please do you have an idea to how identify mat 138 parameters . Thank's

[@AmeenTopa]

hi, you may search for it in the literature, you will find them!

[@srihari74]

Great work.thanks for sharing

[@AmeenTopa]

Thanks for watching!

[@nurshufinah3170]

Hello, I am doing my FYP project this semester about the investigation of delamination on the composite structure. There are some problems that I need to overcome and I do not have strong knowledge about this ls dyna software. Can you help me? I would like to ask you some questions.

[@AmeenTopa]

Yes sure. What are you questions?

[@nurshufinah3170]

@@AmeenTopa I follow your step to model my DCB but the material parameters i put based on paper that I refer to and some problems occurs when the ls dyna cannot simulate because it said the value being insert is not valid or too big especially in poisson ratio part. Another problem is the result for my delamination i want graph force vs opening displacement of the composite but i do not know how to do it because in this tutorial only shows graph with force vs time only. I have search how to get the graph from ls dyna that I want in the internet but still cannot manage to find how to do it. I hope you can help me and can I contact you through email because maybe there are still some question I want to ask and I can show you my sample what problems I have faced.

[@AmeenTopa]

Put your email here and I will follow up in my free time.

[@nurshufinah3170]

@@AmeenTopa my email is [e-mail] and may I know this form delamination tutuorial do you refer any paper all the parameters? Can I get the link of the paper?

[@AmeenTopa]

The paper I referred to can be found from this link: www.researchgate.net/profile/Adnan_Gheryani

[@malithchanaka3004]

Is Hourglass not needed for this simulation?

[@AmeenTopa]

Not all simulations require an hourglass control. First, run the analysis and see if you observe any hourglassing deformation or significant hourglass energy. Otherwise, the simulations should be fine.

[@malithchanaka3004]

@@AmeenTopaThank you sir.

[@MoHonda90]

How did you get the values for NFLS and SFLS?

[@AmeenTopa]

Check the description of the video. I put some references there. The values chosen in this example might be arbitrary.

[@叶梦清痕]

Thanks Ameen，learn a lot from your video , thanks for your shearing

[@AmeenTopa]

My pleasure! Please share my videos so more can benefit! :D

[@叶梦清痕]

@ Of course, my pleasure!

[@sandeepkumar-um3cl]

Ameen brother , I have one bullet part in separate ls dyna file. And one solid plate in another ls dyna file . I have imported both the file . But I want change the orientation and position of bullet .. How to do that ?? Please reply ..... tomorrow is my presentation...... Everything depends upon you.....

[@AmeenTopa]

Dear Kumar, You can use Element Tools, Translate, Rotate. Check my tutorial on "Concrete + Wrecking Ball" to see how I used these tools.

[@sandeepkumar-um3cl]

@@AmeenTopa thanks a lot brother ... you are doing a great and priceless job.... Keep it doing......

[@sandeepkumar-um3cl]

@@AmeenTopa brother suppose I want to impact this composite ( which you have created in this delamination video) with a bullet then how to do that?. I have used bullet as a slave . I want to select eintre composite ( upper layer + clay+ lower layer) as my master segment ID . How to do that ? Or I have to select upper plate as my master ID. Please reply. Iam getting termination error. Please help Can you make a simulation of bullet striking the body armor.? Note:- the body armor can be modelled as consistent of three layer 1) 1st layer is of ceramic 2) 2nd layer of clay 3) third layer is composite. Similar to your delamination model. Just difference is upper material is ceramic while lower material is made from composite. Can you do this simulation? Can I send you my ls pre post file? please give me your mail I'd..

[@AmeenTopa]

Dear Kumar, First of all, I don't give my email here. You can see from the comment of others. You put your email here, and I will contact you in my free time. Secondly, yes it is possible. For the contact with many parts as target, you can use part set as the master. Surface to Surface contact or Eroding contact if you have element failure. Lastly, yes, the three layers simulation is possible. Nothing difficult about it. Just see my other videos about impact.

[@sandeepkumar-um3cl]

@@AmeenTopa thankyou brother , I followed for your valuable advice . I have applied set part for multiple target .. My email ID Is [e-mail]. please contact me when you're free . So that I can send you my thesis ... Because there is termination error in my ls dyna file. Please help me to remove it..