In practice, a composite structure is far more complex than a similar metallic design. There are, quite simply, more variables involved!To modify a composite product design could involve some of the following (in addition to the ‘usual’ parameters of changing product shape etc)•Modify existing plys - changing angles to suit load paths•Add more plys - changing angles, thickness•Make the plys symmetric or unsymmetrical as appropriate•Change the matrix properties•Change the material/materials used (Carbon type, Glass, etc)•Change the weave/chop etcThe above list is not exhaustive and only relates to modifications to the Composite. There are far more variables to change than if an Isotropic material were used (metal) where non of the above are even considerations.The simple reality is that it is not really possible to design an optimised composite structure without the use of Predictive Engineering tools (typically, but not limited to, FEA). Of course its is perfectly possible to design a composite structure without FEA but, without an extensive test program, it will almost certainly be ‘Over Engineered’. It is highly probable that avoiding ‘Over Engineering’ was one of the main drivers in choosing a composite material in the first instance!The typical design work flow for a composite product would/could be as follows.1.Create ‘best guess’ design (Possibly using Laminate Carpet Plots to select E, G etc)2.Initial FEA3.Look carefully at the deflections, Principle Stresses (and, importantly, directions), Lamina stresses, ply failure etc etc4.Re-Work the design (adding more plys, changing angles due to stress directions etc, etc)5.Re-Analyse6.Repeat from 3 until stress/deflections etc are acceptable7.Manufacture8.Test9.Feedback data to FEA (Modification of FE model due to Draping in manufacture etc)10.Repeat from 3 if neededRe-design could be done with the aid of comprehensive NTD and possible strain gauge work but the Designer would be ‘in the dark’ with respect to inter-lamina stresses and how close the composite was to failure.The above is simplified but is a methodical approach. Without the use of FEA it is almost impossible to achieve stage 4 above and be confident that any problems had been solved. Design and Predictive Engineering
In practice, a composite structure is far more complex than a similar metallic design. There are, quite simply, more variables involved!
To modify a composite product design could involve some of the following (in addition to the ‘usual’ parameters of changing product shape etc)
•
Modify existing plys - changing angles to suit load paths
•
Add more plys - changing angles, thickness
•
Make the plys symmetric or unsymmetrical as appropriate
•
Change the matrix properties
•
Change the material/materials used (Carbon type, Glass, etc)
•
Change the weave/chop etc
The above list is not exhaustive and only relates to modifications to the Composite. There are far more variables to change than if an Isotropic
material were used (metal) where non of the above are even considerations.
The simple reality is that it is not really possible to design an optimised composite structure without the use of Predictive Engineering tools (typically,
but not limited to, FEA).
Of course its is perfectly possible to design a composite structure without FEA but, without an extensive test program, it will almost certainly be ‘Over
Engineered’. It is highly probable that avoiding ‘Over Engineering’ was one of the main drivers in choosing a composite material in the first instance!
The typical design work flow for a composite product would/could be as follows.
1.
Create ‘best guess’ design (Possibly using Laminate Carpet Plots to select E, G etc)
2.
Initial FEA
3.
Look carefully at the deflections, Principle Stresses (and, importantly, directions), Lamina stresses, ply failure etc etc
4.
Re-Work the design (adding more plys, changing angles due to stress directions etc, etc)
5.
Re-Analyse
6.
Repeat from 3 until stress/deflections etc are acceptable
7.
Manufacture
8.
Test
9.
Feedback data to FEA (Modification of FE model due to Draping in manufacture etc)
10.
Repeat from 3 if needed
Re-design could be done with the aid of comprehensive NTD and possible strain gauge work but the Designer would be ‘in the dark’ with respect to
inter-lamina stresses and how close the composite was to failure.
The above is simplified but is a methodical approach. Without the use of FEA it is almost impossible to achieve stage 4 above and be confident that
any problems had been solved.
Design and Predictive Engineering
