help!!! how does a composite material be calculated?

[gregario]

Hi.

I'm trying to get into a dark world for me... .

the calculation of elements in composite material. . .

premise that before the crisis in the company we were following a course for the use of cososworks (now simulation...), but that with the advent of the crisis in 2009 is a completely abandoned speech... unfortunately!!!! ! ! ! ! ! ! !

and that I would like to grow personally and then with these new knowledge have an extra chance for the change of work, I would like to ask you for help.. .

I would like to calculate (I know the loads and their layout) a carbon fiber bicycle frame, but I don't know which side to start with the definition of the material and the "construction" of skins with diversified orientation on my frame.

How do I proceed?

-better a "empty" solid body of the hypothesized thickness or work with surface bodies?

- How do you simulate skin?

-What mechanical characteristics are to be attributed to the fiber?

I apologize for my ignorance, but in this frum there are very competent and prepared people who allow me to disturb. . .

thanks in advance for every possible answer

 

[meccanicamg]

hi gregario,
Have you tried to look on the sw guide?
something interesting you find it in English googolando ...
I believe that composite material is identified as a multilayer and not as a base material with anegates of fibres with anisotropic characteristics.

before proceeding with modeling you must have clear ideas on the material and its characteristics, the fact of modeling comes after in my opinion.

It's a shame that with the crisis you don't take the course anymore. the self-taught service can give its good fruits but have someone who shows you the tricks... always convenient.

I never needed to use the surfaces and for me it is an unknown world but I think they are two equivalent approaches even if then to develop the project they take different paths.

I don't know anymore:

 

[cesius79]

hi....I am approaching the fem method so I can't tell you much but I can definitely tell you that cosmos sucks....I explain, who chews a little finite elements starts laughing if you tell them that you have made a verification with cosmos... is extremely limited and because of the poor possibility of meshare and to set various parameters the results obtained are not very reliable. .
I repeat, I speak more for indirect experience than for personal knowledge!
and I add that if you do not first have a deep "mechanical" knowledge of the material you are facing.. Let it be...there are those who before trying in such things read booklets and booklets. I don't think anyone can give you exhaustive answers on two lines in a forum...

 

[meccanicamg]

I agree with cesius, finite element simulators integrated in modelers have the advantage of switching from model 3d to its simulation but they carry with them different limits. Just think about the modules that has integrated inventor and take the complete separate ansys package... There's no comparison. we say that integrated ones are good for standardized activities, known and common materials. It is also true that those who know a lot more than we would say that almost everything can be done with cosmos...

 

[Er Presidente]

Hi.

I'm trying to get into a dark world for me... .

the calculation of elements in composite material. . .

premise that before the crisis in the company we were following a course for the use of cososworks (now simulation...), but that with the advent of the crisis in 2009 is a completely abandoned speech... unfortunately!!!! ! ! ! ! ! ! !

and that I would like to grow personally and then with these new knowledge have an extra chance for the change of work, I would like to ask you for help.. .

I would like to calculate (I know the loads and their layout) a carbon fiber bicycle frame, but I don't know which side to start with the definition of the material and the "construction" of skins with diversified orientation on my frame.

How do I proceed?

-better a "empty" solid body of the hypothesized thickness or work with surface bodies?

- How do you simulate skin?

-What mechanical characteristics are to be attributed to the fiber?

I apologize for my ignorance, but in this frum there are very competent and prepared people who allow me to disturb. . .

thanks in advance for every possible answer

minefield!

Surely it is better to work with solids and then it would be better to shape the stratifications in the most significant knots, with ad hoc model.

It's not easy for anything, but simulation must be simplified, and simplifying it's easy to catch chants.

to avoid harming an approach that I have tried is to simulate the structure to verify the performance of the loads (quality analysis) and then apply the loads in detailed models of the critical zones.

It's a job that needs so much experience, but it can become a job if you can do it.

Hi.

 

[Nicolas]

Hi.

I'm trying to get into a dark world for me... .

the calculation of elements in composite material. . .

premise that before the crisis in the company we were following a course for the use of cososworks (now simulation...), but that with the advent of the crisis in 2009 is a completely abandoned speech... unfortunately!!!! ! ! ! ! ! ! !

and that I would like to grow personally and then with these new knowledge have an extra chance for the change of work, I would like to ask you for help.. .

I would like to calculate (I know the loads and their layout) a carbon fiber bicycle frame, but I don't know which side to start with the definition of the material and the "construction" of skins with diversified orientation on my frame.

How do I proceed?

-better a "empty" solid body of the hypothesized thickness or work with surface bodies?

- How do you simulate skin?

-What mechanical characteristics are to be attributed to the fiber?

I apologize for my ignorance, but in this frum there are very competent and prepared people who allow me to disturb. . .

thanks in advance for every possible answer

hi gregario,
in the sector there are specific sw for composites. one for all could be laminated tools (but there are others).
the greatest problem of composites is that they are not easily modelable (cf. er president) with standard sw fem. even if you can risk having to spend a lot of time taking your hand.
however we say that:
1. fiber can simulate it using the data sheet of suppliers in which they give you, beyond the fiber, also the mechanical characteristics of the resin. We do not forget that the resin is charged with the transfer of the loads between layers of fiber and that it participates in the determination of the total final characteristics. resin, besides, does not participate in normal efforts on the fiber plane and this makes it an anisotropic material (especially orthotropic).
2. characteristics, depending on the application sector, are affected by reduced coefficients such as temperature, mixture and weight. These coefficients come to decur design value even up to 50% (see aeronautical sector and mil-hdbk17: the bible of composite design).
3. lay-up layers. in the first approximation you can represent the solid to design using an isotropic material. At this point, analyzing the stresses, you can make a first estimate of lamination. if you have a specific sw, then you can proceed in subsequent approximations. If you do not have such a sw, then there is to be tribulated. and not little.

if you surf the internet you will also find a freeware sw that allows you to calculate the Young module of a lamination. you can integrate this to the fem.

bye:

 

[stef_design]

I agree with cesius, finite element simulators integrated in modelers have the advantage of switching from model 3d to its simulation but they carry with them different limits.

In recent times, for high-level cads, things are combined.
for example we can find the couple:
nx + nastran
Cat + abaqus

and for inventor:
inv + algor

cmq in principle is always better to divide the two things (cad/cae)
Hi.

 

[Onda]

composites are typically shaped (I would say in 95% of cases) with shell medals.
normally shell models are made on the mold surface and not on the average surface, so you can simulate different lamination planes without changing the pattern.
I recommend checking the direction of the normal elements to ensure you have the material on the correct side.
It seems to me that the complete version of cosmos can model composites, although I use cosmos only for metal materials and patran/nastran to model composite structures. Moreover I would like to say that it is absolutely not a bad soluture, that you do really remarkable things with cosmos and it is not to be underestimated, it is not a game even if so it can seem in its plastic interface of solidworks.
I've known people who do text-line analysis with cosmos, and I didn't seem to have much to envy than nastran.

In any case, for your model, you must be able to insert the properties of each single layer and its orientation.
otherwise, as nicolas rightly says, you can use a program to calculate laminate properties and consider a single constant thickness.
a program with very low cost (29$) but suitable for purpose is http://www.thelaminator.net/#download.
you must then make a property for each area where you have a variation of laminate and then calculate the correct elastic modules.
I always recommend working with shell models rather than solids. are much faster to solve and much more precise!
to change a thickness just type a number and do not remodel everything!.
Moreover, especially if you have fine thicknesses, you should meshare so thick that you would not just have a supercomputer to solve, while you do not have this limitation with a shell model, where you just have to be careful that your elements do not disconnect too much from geometry, especially if you use elements of the first order, without midside knots.
wave