Laminate Elastic Properties

[yusf]

Hi all;

I am trying to extract the basic elastic constants of a laminate consist of many layers making use of composite plate theory.In other words after calculation of ABBD matrix, i find EX EY GXY Vxy etc.

I performed many trials to find the effect of orientation, single layer 's properties on global laminate elastic properties.

I noticed that as long as the thickness and elastic properties are identical for each layer oriented in 0/90 the global EX equals to single ex of individual layers for those also identical in all mechanical and geometrical prop's.

Let me give example:

LowerCase :Single Layer
UpperCase :Laminate

ex:1 thck:1 mm

1 0
1 90
gives global: EX:1 Thck:2 mm ( extracted from A)

1 0
1 90
1 0
1 90

gives global: EX:1 Thck:4 mm ( extracted from A)


Could you please comment about my conclusion? (in fact i am having confusion coz if i apply a tension test i knw that the thicker laminate will be too stiff and thiner one will be weak compared to thicker.But how can theoretically two different layup give me same Elastic moduli?)

???
"
From this results i concluded that for simple tension test regardless of the thickness of the laminate (thinner or thicker) the global elastic moduli of laminate equals to single layer and this is true if orientations are 0/90"

???

Also i noticed that In MIL -HDBK 17 A and from the experienced users from here; the suggested orientation for tension test is 0/90. Can this result be due to above theoretical results ( if i am not wrong)?

I attached a file in which i want to ask which stress strain diagram represents the given laminate condition if a tension test would have been done.
If i am not clear please write me i can explain my quesiton in detail.


thanks in advance

 

[ESPcomposites]

Well, your terms seem to be a bit "all over the place". You seem to mix moduli, stiffness, and strength (weak), and that may be the problem?

Perhaps this will help though. The effective modulus is a material property and both the thin and thick laminates can have the same moduli.

Stiffness is not a material properly, but rather a combination of modulus and geometry. F=kx, where k is the stiffness, and for a simple laminate is AE/L = (width*thickness*modulus)/Length. So the thick laminate will have a greater stiffness, even though the modulus is the same.

Your stress-strain curves will show the slope of the modulus. If you plotted a load-deflection curve, the slope would be stiffness.

Brian

http://www.espcomposites.com

 

[SWComposites]

"if i apply a tension test i knw that the thicker laminate will be too stiff and thiner one will be weak compared to thicker." > how do you "know" this??

Both of your stress-stain plots are not correct. Both laminates should have the same modulus (slope) and should fail at approximately the same stress or strain value.

SW

 

[yusf]

Hi thank you very much for your quick answer..

I know that i confused too much and reading your comments to understand well.

ESP
you noted:
The effective modulus is a material property and both the thin and thick laminates can have the same moduli.

and how about the failure load which one fails first and at what strain value? Can it be approximately be calculated analytically or by FEM?


SW
i meant that the thicker laminate should be stronger (stiffer) so that it will resist much more load than thinner one ( i may confused again) since i dont have any test data i am trying to make conclusions by analytical calculations but i could not match the real test and analytical ( together with fem).

Is there any sample problem ( analytical or fem ) or tutorial solved to better understand the real test and fem calculations.

Think that there is no test data and I want to perform a real test simulation as much as i can by using ABAQUS..How can i perform it? Any guide or example problem? I am extracting results bu

Please pardon me if i am asking so simple questions ..I want to refresh myself and avoid doing any mistakes while performing FEA.

Need your guidance..

Thanks in advance

 

[yusf]

And Also i would like to add to understand;

Is it true that the Elastic Moduli extracted from laminate theory will be same with the test results somehow conducted on same laminate?

thanks

 

[ESPcomposites]

Few comments:

- What SW said is correct. Both should fail at about the same strain level.

- Moduli from lamination theory should match the test data well.

- For failure prediction, you can look in the aeronautics/aerospace forums as well. We have had some discussions there.

Brian

http://www.espcomposites.com

 

[yusf]

Thank you Brain;


Could you please comment ...
I have looked some more to theoretical basics and realized that i can also find the failure strain if i know the failure load ( at which the laminate fails during real test) by using the following laminate theory formulation (right?)

[Nxx] = [A11] [A12] [0] [exx]
[0] = [A21] [A22] [0] * [eyy]
[0] = [0] [0] [A33] [exy]

For example for real tensile test Nx will be the line force value and if i calculate the laminate extension matrix A ( for the laminate being tested with well defined orientations and material properties) by the following manipulation i can extract the failure strain exx;

inv(A)* [Nij]=[eij]

I want to be sure that if real test for composites will also watch above calculations.


Then i want to make some other detailed calculations and tabulate the failure load failure strain for specific laminate with various material and orientations so that i will end up with strong theoretical basics and i will then be able to analyze my real design wiht FEM and understand if it fails or not.

Could you please make comment on my procedure above.If you have some suggestions for FEA based composite analysis and the failure predictions.

for example i have a 3 mm laminate test data but i want to be sure in with calculations that my other 1.2 mm laminate fails under some other load conditions wihtout carrying a real test.In other words i want to take 3 mm real test data as reference.?s it possible?

I am also checking other forums that you mentioned thanks.

Thanks in advance.

 

[yusf]

Hello All;

I think i am not clear with my above question and i would like to kindly ask in other words..

My question is especially in the field of testing of composite materials and making use of the datas obtained from those tests.

If i test a 2.5 mm thick specimen and get result how can i use the faiure data obtained from test for my let's say 3 to 5 mm thick composite design.

Sould i perform other specimen tests that are in 3 ,4 and 5 mm thick? in order to model all thickness region as single lamina in ABAQUS.?

Should i perform analytical lamination theory to find each individual layer propertis by guessing until precise EXX EYY at which the whole laminate appearent modulus equals to the test value?

AND THEN modelling other thick regions by calibrated EXX EYY?

Any idea? thanks in advance...

 

[SWComposites]

First, stiffness and strength are different properties; their use in analysis is different.

If i test a 2.5 mm thick specimen and get result how can i use the failure data obtained from test for my let's say 3 to 5 mm thick composite design.
> calculate the failure stress or strain and use that value for the other thicknesses

Should i perform other specimen tests that are in 3 ,4 and 5 mm thick? in order to model all thickness region as single lamina in ABAQUS.?
> probably not necessary

Should i perform analytical lamination theory to find each individual layer properties by guessing until precise EXX EYY at which the whole laminate apparent modulus equals to the test value?
> you could do that but it may not be accurate as you need 4 independent stiffness properties for ply (E1, E2, nu12, G12); or a better approach would be to test [0], [90] and [+45/-45]s laminates to get E1 and nu12, E2 and G12, respectively

AND THEN modelling other thick regions by calibrated EXX EYY?
>yes, use the E1, E2, nu12, G12 values from above as the ply stiffness properties in analysis

SW

 

[yusf]

SW

Thank you very much for your kind help.

I would like to ask to be sure if you dont mind;

SW:> calculate the failure stress or strain and use that value for the other thicknesses

Me:>Now i understood.As Brain and you stated aboove regardless of the thickness, the laminates will fail at same strain value.So if i find all layer strain values by using FEM for different thick regions i can compare those strains with test and understand whether it fails or not.

And lastly as far as i followed from here i think it is agreed that the tsai-wu failure criteria will not give accurate results that match with real test for the analysis of laminates following above way namely modeling it as single layer (3 mm thick laminate as single layer ) do you think the failure index extracted from such analysis can be used or it has meaning in terms of failure?

In other words ,FEM also use plate theory and extract failure index (FI) for each indiviual layers what if my single layer in FEM has 10 layers in reality? does FI have meaning in that condition?

I look forward to hearing from you.
Thank you again