Laminate Stiffness, Test vs Predicted

[timtimtim44]

Hello all.
When comparing the stiffness results from a tensile test of a lay-up (0/90/+/-45 in differenct combinations) with the stiffness predicted using classic laminate theory (CLT), I have noticed that it's not uncommon for CLT to overpredict the stiffness. The measured stiffness is usually around 10% less than the predicted value, this behaviour has appeared when testing both unidirectional and woven material. There is some difference in thickness but the difference is never large enough to justify such a difference in stiffness.
The values that we use as a basis for our models have been experimentally determined.

Is 10% accuracy as good as it gets with CLT or is this a particular behaviour for multi-oriented laminates that CLT does not capture?
Any input/thoughts or experience from eng-tips members would be much appreciated.

 

[SWComposites]

No, you should be getting much closer. Over what stran range is your lamina and laminate moduli? Are you using lamina tensile moduli? What is the fiber? Resin? At what temperature?

 

[SWComposites]

Also, are you using measured thicknesses to calc moduli, or a consistent nominal thickness? How are you experimentally measuring strain? What test method standards are you using?

 

[timtimtim44]

We are testing according to ASTM D3039M, mostly high modulus fibers cured at 180°C with 55% Vf for fabric and 58% for woven. But similar results is also observed when testing T700 UD laminate with higher resin contents (roughly 48%). We normally measure strain with extensometers, but we have also performed some samples with back-to-back strain gages to obtain similar stiffness values. All tests are performed at RT.

Regarding thickness I can provide the following example data. (Tests performed with the same batch of material).

0° woven test coupons, E1t = 111 GPa average (min =109 GPa), E2t = 106 GPa average (min = 102 GPa) with CPT between 0,29-,305 mm with the avearge being 0,3 mm.
[0/0/45/0/0]s laminate results in a laminate modulus of 93 GPa (experimentally determined), measured mean CPT 0,307 mm. CLT predicts around 100-103 GPa.

 

[ESPcomposites]

Here is some additional data. For the NCAMP IM7/8552 unidirectional material system (public domain), there was an average difference of 3% between the directly measured laminate tension/compression elastic moduli (3 different laminates at 3 environments) and the elastic properties determined from the ply properties and CLT (15 total data points).

Brian

http://www.espcomposites.com

 

[SWComposites]

Your CLT does not seem correct. I put E1=111 GPa, E2=106 GPa, G12=4 GPa, nu12 = 0.06 into my laminate code for [0/0/45/0/0]s and get Ex = 98.9 GPa. Correct this from 0.3 mm CPT to 0.307 mm CPT gives 96.6 GPa. Only off by ~ 4%.

But, what strain range did you use for lamina E1 and E2 moduli? What strain range did you use for laminate Ex moduli measurement? Are they the same?