Dear Dr. Michael, Very interesting video but kindly correct me if I am mistaken as I think there is a typing error in the equation for calculating the dimension of RVE; S = D.sqrt (Pi/2.Vf) as it should be S = D/2.sqrt (Pi/Vf) or S = R.sqrt (Pi/Vf); while R is the radius of the fibre. Thanks in advance

@MichaelOkereke

2 жыл бұрын

Hello @John Nabil, good observation. You are right if there was only one fibre within the RVE... that is a classic square RVE with fibres only at the corners. In this case, it is like a body-centred square unit cell with a central fibre and the four corner fibres. This leads to the equation that I used. If you look at it from first principles and derive the equation, for a unit cell (with square arrangement) comprising of N fibres, with edge length, S, fibre diameter, Df and volume fraction, Vf, the expression of S becomes S = Df*sqrt( (N*pi) / (4*Vf) ). For N=1, S is exactly as yours but for N =2, then you find that mine is correct.

@joonabil

2 жыл бұрын

Dear Dr. Michael, Thanks a lot for your clear detail explanation as I was thinking that you derived this equation for a unit cell for one centred fibre as it is mentioned on the title of the video “Square Unit Cell”. Now I understood that it is for a unit cell with a central fibre and the four corner fibres which means in total two fibres as you modelled it in the video.

Thank you for this great video! I have been waiting for your next video continuously.

@MichaelOkereke

2 жыл бұрын

Hello @Enidentictok, so nice to hear that. Cheers!

Thank you, Dr. Okereke, for the informative videos. Although I already learned how to model it, it is good to redo it. In my studies, I used hexagonal, square, and random fiber distribution both in 2D and 3D cases.

@MichaelOkereke

2 жыл бұрын

Wonderful @Gideon Lyngdoh! It is always good to re-learn things. Did you think about random cases modelling? I believe this is the next step and see if the randomness has an effect on the effective properties?

super!!!! thanks

@MichaelOkereke

2 жыл бұрын

Cheers Tony

Thank you sir, your videos have helped me a lot in learning abaqus... always looking forward for your next video...respect from India

@MichaelOkereke

2 жыл бұрын

Thanks

Thanks You Dr Okereke . Can you please do a video or explain how we can construct a RVE with interphase between the 2 main phases .

@MichaelOkereke

2 жыл бұрын

Hello @Dura, thanks for the comment. Yes, this is something I will do in future. Please keep watching out for the video, when it gets published.

Thank you Dr. Okereke for this amazing video. Would you please make another video on woven fabric structure ?

@MichaelOkereke

2 жыл бұрын

Hello @Ajay Jayswal, thanks for this. I am definitely going to consider making such a video as a lot of people are asking for it. Thanks.

very cool

@MichaelOkereke

2 жыл бұрын

Cheers!

Thank you Dr. Okereke for this amazing video, should we apply PBC in bi-axial loading

@MichaelOkereke

11 ай бұрын

Yes we do, I think you can. I will think about this and see if there is a purpose in making a video where I show RVE of a composite (2D), and apply biaxial loading as well as PBC. It should be interesting. Let me know by the number of likes to this comment, who wants to see such a video.

Thank you very much Dr. Michael I am very grateful to you for your answer I am very interested in your valuable lectures My question is in the picture attached to this message about how to determine the value (percentage %) for each material in the preparation of the composite material, whether layer, slice or cube. That is, anywhere in the choices of the Abacus program when we specify these values, that is, when we limit the dimensions and density, and with what Young's modulus and the Poisson's Modulust, because I searched a lot and did not find anyone who explained it. Where I noticed that in the video I posted, the value of polypropylene was 40% and fiber was 60%, how did this happen? This is in the case of two materials, but if it is three materials, it means two solid materials with a polymeric material, and I hope that you will help me with it, and I am very grateful to you

@MichaelOkereke

Жыл бұрын

I understand the question now. With regards to the volume fraction of the fibre and matrix constituents, you have to determine this yourself when you design the composite. It is a volume fraction that means in creating the geometry, you have to decide how much of the volumetric space (virtual domain) is going to be occupied by the fibre and how much by the matrix. The volume fraction of 40% fibre means that only 40% of the volume will be occupied by the fibre. How do you then prepresent this? So, if you have a space of 100 x 100 x 100 overall for the composite, you ahve to find how many fibres will occupy the fibre to give you a 40% volume fraction. You have to make some calculations and if you look at my videos on RVE modelling of UD composites, you will see how to do the calculations. If the calculation shows that 5 fibres are required (for example), then you make sure in the 100x100x100 space, you create cylinders that represent the fibres based on their radius that will cover 40% of the volume. The remaining space will then be 60% which is the matrix. For a three-constituent system, the same things holds. Clearly, where you need to get the percentages i.e. 40% and 60% will come from a manufacturer's data sheet for the material you are using. In the case of the videos you refer to, I got them from the publication referred in the video which stated what percentage is the volume fraction for the fibre (E-glass) and the matrix (polypropylene). For the system you are testing, please find those percentages yourself. If you do not have it, then you can decide how you want to give percentages to the parts provided all of them add up to 100%. That is my suggestion.

Thank you very much for sharing the insightful video. However, I'm curious about the accuracy of the elastic parameters estimated using the constrained normal displacement approach as a substitute for periodic boundary conditions. I noticed you also shared a video about implementing periodic boundaries, and now I'm confused about the differences between these two types of constraints. Please help me out; this issue has been troubling me for quite some time.

@MichaelOkereke

7 ай бұрын

Hello, thanks for the comment. There is no need to be confused. Whether Periodic Boundary Condition (PBCs) or Dirichlet Boundary conditions (DBCs) (what you describe as constrained normal displacement approach), both will give elastic properties. The literature says you get better and quicker convergence to elastic properties using PBCs rather than DBCs. The DBCs approach is the most common way people impose BC on domains to extract effective properties. The use of PBCs is particularly suited for heterogeneous media where stiffness mismatch between constituents exists so we cannot guarantee homogeneous deformation of the edges/surfaces. For such instance, it is better to use a PBC which will accommodate for the stiffness mismatch between the constituents. If you do decide to use a DBC, what you will get is an over-constrianed solution with an upper bound value of the elastic properties (i.e. slightly higher than it should be) than where you use a PBC. So, in principle, there is no reason for concern. You can check out my publication about this principle in this paper: Okereke, M. I., and A. I. Akpoyomare. "A virtual framework for prediction of full-field elastic response of unidirectional composites." Computational materials science 70 (2013): 82-99, doi.org/10.1016/j.commatsci.2012.12.036

@user-nr1qm8ie3h

7 ай бұрын

Thank you, I understand now. You are amazing!@@MichaelOkereke

Dear Dr.Michael, if I need a Hexagonal packing unit,but one fiber and four corner fibers like this,what formulas do i need? How to determine the fiber diameter of a given L W H?Thank you so much!

@MichaelOkereke

2 ай бұрын

Please check my other videos on unit cell and see if you can find resources to help you.

hexagonal and square

How type of boundary conditions.thank you so much

@MichaelOkereke

7 ай бұрын

It is Dirichlet (displacement) boundary condition.

i want hexagonal and inclined at various angle(15,30,45,60..etc) fiber distribution.

@MichaelOkereke

Жыл бұрын

If you follow my video, just making sure you change angle of the fibres, then this will do it. Check out my video on Bouligand structures (manti schrimp) or multi-layer composite, it will help you proceed.

I have another question. Do you have a plan to deal with RVE modelling of woven composites similarly to this video? I hope to see the next

@MichaelOkereke

2 жыл бұрын

Hello @Enidentictok, yes this is in the pipeline and please keep watching out for that.

Hi! i have a 3D ortotropic material (wood) and i need to modelling in multiple scales. Can i do it with abaqus or ansys?

@MichaelOkereke

Жыл бұрын

Of course, with careful thought, you can definitely model this in ABAQUS.

sir at 9.38 minutes how did you deselect a matrix، Which button do you press

@MichaelOkereke

Жыл бұрын

First select all components in the model at once. Then, by holding down the Shift and or the Ctrl button on the keyboard and click on the matrix to deselect it leaving only the fibre. Hope this helps.

@sofianebarbara3322

Жыл бұрын

Thank you sir

May I know where is the video in card can guide me create all load cases

@MichaelOkereke

Жыл бұрын

I have a few videos on creating load cases. Check this kzread.info/dash/bejne/h6GI2a-KdKbQd5c.html&si=EnSIkaIECMiOmarE

Can I use these equations especially the equation S to MODELLINg one fibre

@MichaelOkereke

Жыл бұрын

Hello @ilhem Sahnoun, thanks for the comment. It is possible to model one fibre but you will use a different form of the equation than the one I provided here. I answered similar question to a comment by John Nabil. Check the pin comments under this vidoe and you will see my comments on John Nabil's query. It will give you a format for the equation for calculating one fibre. Let me know if that helps.

@ilhemsahnoun

Жыл бұрын

@@MichaelOkereke THANK YOU DR

Hi Dr. Okereke, Is there any reference papers about the size of the square unit cell model?

@MichaelOkereke

2 жыл бұрын

Not really @Chenyi, it all depends on the size of the fibre diameter. You can refer to this old paper of mine for fibre dimensions and RVE size: doi.org/10.1016/j.commatsci.2012.12.036

@chenyili5733

2 жыл бұрын

@@MichaelOkereke Thanks so much~!!

@chenyili5733

2 жыл бұрын

@@MichaelOkereke What about the size of the square unit cell model? Say, how to get the equations for calculating the dimension of the square unit cell?

@MichaelOkereke

Жыл бұрын

This will require some sort of calculations. I suggest you get a textbook on composite micromechanics and find the equation. It is aimilar equation as th eone used in this video, the only thing missing is the central fibres.

how i can calculate volume fraction. please

@MichaelOkereke

Жыл бұрын

Hello @ilhem sahnoun, I think you can use the equation I gave in the video at the time frame of 2:45. If you re-define it so that Volume fraction, Vf becomes the subject of the formula, then you can determine volume fraction from that equation. Good luck.

Hello Dear Dr. Michale for the fiber arrangement calculation S = Df*sqrt( (N*pi) / (4*Vf) ), is there any solid refference like journal paper or book?. I am eagerly waiting for your response.

@MichaelOkereke

Жыл бұрын

Hello @Mazed npu, there is this book which captures this very well. It is *Mechanics of Composite Materials, Second Edition* by *Autar K. Kaw* and published by CRC Taylor & Francis. You want to check page 224-225, and Chapter 3: Micromechanics of a Lamina . Hope that helps.

@mazednpu3834

Жыл бұрын

@@MichaelOkereke Thanks a lot sir

Hi, Dr. Okereke. Thanks for sharing these informative videos. Is there any way to calculate strength calculation for Z-direction (fiber direction) tensile strength? Can I contact you via your mail? Best regards.

@MichaelOkereke

2 жыл бұрын

Hello @Cagatay, thanks for your comment. Yes, the strength can be determined from the stress-strain plot. The models do not have a damage model or cohesize zone incorporated so strength predictions will be excessive for the current model. With some modifications to the model setup, you can easily determine the strength values. Please send me an email via the contact email on my About page of this channel. Thanks.