Flexural properties of laminated composites are layer dependent. For example, the flexural strength of a 4 layer/ 3mm thick/ 0.5b volume fraction is different from a 24 layer/ 3mm thick/0.5 volume fraction composite , just because the individual layers are of different thickness. Is there a predictive model ? ( Assume t hat the tests are done at the same speed and same l/d ratio.
Dear Padmanabhan Krishnan, as far as I remember the common Approach builds on linear elasticity for Composite structures and considers the contributions layer by layer. On the oter Hand FEM Software provides more detailed answers to specific Problems and applications. Material properties can be taken from uni-directional materials a.s.o. I you want to have more / more Details, I think the present Situation for developing something more than this is rather good.
Though there are interesting papers by Carl Zwben ( layer effect in flexure) and Akkerman ( woven fabric elastic constants), even for linear materials flexure is seldom linear. My worry is that there are no linear and nonlinear models ( Separately) that can predict layer effect for any combination, though there are glorious books on even the influence of weave geometry. on tensile and compressive behaviour.. How do we predict using models, to evaluate flexural strength , modulus and strain at failure for laminated structures at any static speed of testing and L/d ratio ? Present FEM softwares do not have a single model to bank on ! Some times a linear estimate is made for a material that would give a non linear sigma-epsilon curve in flexure.. In sandwich composites , how much ever I try, I get up to 27 % error for having chosen a different weave, GSM and number of layers even for the same vol frac, beam thickness and L/d ratio.. How do we go about it ?
If you are talking about linear elastic behavior, then you can obtain the flexural properties from the D matrix. If you want the so-called flexural strength, then you need apply a certain failure to the most stressed layers. A general-purpose constitutive modeling such as SwiftComp can help you. The code can be freely launched in the cloud at https://cdmhub.org/resources/scstandard.
https://cdmhub.org/resources/scstandard
Good answer Wenbin Yu ... Agree with you... Also Rüdiger Rentsch .....
Padmanabhan Krishnan ... Good question.. Please share me the best answer might you trust...
Regards…
- Badges
- Science topic
- Similar topics
- Vectorization