How I can design a composite material 4

[nadim22]

Hi,
Please how I can design a composite material and with which software?
Can I design this type of material with abaqus?If yes which method I use?

 

[RPstress]

Your question is a bit vague. If you mean design a structure made out of laminated fiber-matrix materials, then I suggest reading parts of some of the books and websites which have been recommended in this forum in the past.

Reasonable websites are

http://home.comcast.net/~brandihampson/cd/index.html

and

http://www.mse.mtu.edu/~drjohn/my4150/.

(Also

http://www.mse.mtu.edu/~drjohn/my4150/compositesdesign/index.html

appears to be the same material as the first site.)

Using the Eng-Tips Google search box at the top of the page and looking for composites introductory will find some book recommendations.

Yes, Abaqus can be used for analysing structures made out of composite materials.

 

[nadim22]

Hello RPstress,

Thanks RPstress ,in fact I 'm a bignner in this topic.

 

[SWComposites]

So what are you specifically designing? You do not design with an FE code; you use FE to analyze a design.

 

[nadim22]

I want to design a composite fiberglass which is used to manufacture boats and analyse it to see his behavior in fatigue.So, I don't know if I should design it with Catia and transport it in format step to abaqus to analyse it, or I should use only abaqus I saw that in module proprety and exactly in create section when I select shell I find composite.

 

[RPstress]

For something with the level of complexity of a boat hull you're probably better off designing in CAD rather than attempting to use Abaqus's preprocessor.

I can't remember if Abaqus has a STEP translator, but either STEP or IGES will be a suitable translation tool. As Abaqus is now owned by Dassault it may get a more direct translator for Catia.

Abaqus itself is probably overkill for this sort of analysis. A simpler FE solver would probably be adequate.

I don't know much about the behaviour of glass in fatigue. You will need some materials data from somewhere. Someone more involved in the marine industry may be able to help more.

Eric Greene's book at

http://www.marinecomposites.com/

may be helpful if you are new to boat design. Experienced work colleagues would also normally be necessary. You may be eligible for a free subscription at

http://proboat.imirus.com/

.

From my limited experience of Abaqus I think a one week training course would be appropriate if you lack experience in its use.

 

[ESPcomposites]

So say you develop an accurate FEM, with composite properties. What particular information do you hope to gain from this? How do you propose to actually solve your fatigue analysis?

A FEM will only give you stresses and/or loads, but the more important part is what you do with this information.

I get the impression that you think the FEM will actually solve the problem you are asking. In the aircraft industry, we use the FEM to simply find the loads in the structure. From there, we apply loads to a separate detail level analysis. There is a good reason for this. The detail level analysis can account for semi-empirical methods, details not included in the FEM, etc.

I suppose the other way to do this was be to obtain an allowable gross strain and survey the FEM for it. However, if you are already at the point of not having a good detail method, I don't think it will be too important if your FEM is "composite" or not (provided you are using standard type symmetric and balanced layups). I suspect a much simpler model using "effective properties" would be sufficient, but you will need to be judge of this.

I have run across some technical papers on glass fiber fatigue so you may be able to find something (i.e. google, etc.). I will have to look again, but they behaved rather differently than carbon fiber, which is the focus material of most studies.



Brian

http://www.espcomposites.com

 

[nadim22]

Hello all,
Thanks for your help ESPcomposites and RPstress.
In reality, I don't understand what do you mean by "detail level analysis".
I find in this attachment that it is possible to model oligo-cyclic fatigue with composite material.
Please have you any idea how I can find the table of "Edit composite Layup" in abaqus.
Thanks again.

 

[ESPcomposites]

Nadim,

I think you are under the impression that FEM = composite analysis. FEM is only a tool that can assist you with analysis and in general it can help to enhance your accuracy, etc. For those getting started, the pretty pictures are way to attract customers, but it may be misleading as well.

But at the core of it, FEM is not really "analysis" by itself. A "detail analysis" means that you would extract the load from the FEM and then apply that to a structural detail, such a hole, joint, etc. Will you need to consider this for the hull? Will you need to consider the effects of impact damage?

For example, let us consider a hole. You may not actually model the hole in the FEM. Instead, you would look at the loads in that region and then apply this to a detail model capable of considering the local stress field around the hole. Even if you were to model the hole in the FEM, it will only be mildly helpful if you don't know how to apply a failure criteria for it, with the material system you are working on. Similarly, there are special failure modes to consider when going around bends, etc. that the FEM will not be able to directly solve for you.

Without knowledge of structural analysis, an advanced FEM software is not likely to give the desired result of "enhanced accuracy", which is the major objective. That being said, you could perhaps find a strain allowable based on a technical paper. You could then apply this globally to the FEM and make sure your structure is capable of the applied loads. But in my opinion, there is no need for special composite software solutions if you are going to do that.

Brian

http://www.espcomposites.com

 

[RPstress]

My French isn't good enough to be entirely certain, but the 'oligo-cyclic' fatigue analysis appears to be about delamination growth. This is a relatively sophisticated area, not entirely suited to beginners. Having said that a two day course might be adequate to give you an idea of what FE analysis with Abaqus could do for you.

ESP's comments about not needing special composite solutions are appropriate. I would start out with the basics of structural and stress analysis methods, unless you already have a bit of grounding in these. Fairly basic analysis is usually all that is needed for simple marine structures. (Apologies if I'm offending any naval architects out there. When I briefly studied naval architecture 30 years ago analysis was deliberately kept fairly simple, not to say simplistic. I've no doubt things can get quite complicated these days, but the basics should still apply.)

 

[cpinz]

Nadim22

some basics, first:
oligo-cyclic fatigue means damage in a material subject to cyclic PLASTIC STRAINS, typically under reversed stress.
For metals, it means to have fracture in a very limited range of cycles (Nr < 100000). In such conditions, the material is subject to the so-called "hysteresys cycle" whoose form depends by the softening or the hardening of the material itself. (end of basics)

Now...
- do you really want to analyze a composite in plastic strain ?
- are you sure that your boat has to be verified in the plastic range ?..and for so few cycles ?

My suggestion:
- go to the basics of structural design (..and not of FEM !) and materials behaviour;
- go to the basics of composites (..think only in elastic terms, let it be plastic range);
- try to understand the loads acting on a boat.

After that, you will be able to answer your question by yourself

Cpinz

 

[stanier]

However much FEA you use you will end up building one and testing it until it breaks.

 

[Composites_FEA_Guru]

Hi nadim22,

If you have generated a FEA model inside of Abaqus and have an output file (.odb) that contains the lamina stresses in your composite, then you can use Fe-Safe/Composites to determine the cycles-to-failure of your part. This works well when you don't have a simple shapes and don't have the time to spend digging through thousands of reference papers for a equation that suits your needs.