Software Suite for
Durability, Damage Tolerance, Reliability & Life Prediction
Enhances FEA Solvers MSC Nastran*, ABAQUS, ANSYS,
RADIOSS & LSDYNA
* Best Performance and Verified Solutions with MSC Nastran
This Week's Feature Composite Example
Material Qualification and Certification Determine Allowables by Means of Virtual Simulation Combined With Limited Testing

Abasis and Bbasis strength values are critical to reduce risk in structural design of composite aircraft structures.
A previous newsletter presented a novel approach to determine A and
Bbasis allowables for composite materials (click
here for previous newsletter). This newsletter provides
additional details of the A and Bbasis allowable generation and
describes a link between virtual testing and design carpet
plots.
The calculation of allowables for polymer matrix composite for aerospace applications is governed by FAA and Military Handbook 17E standards and rules. The process is costly and time consuming as large numbers of coupon tests are inevitable (example: the generation of allowables for the IM7 fiber and 5250 resin for the F22 program cost close to 100 million dollars). To accelerate the prediction of allowables and reduce the number of coupon tests, GENOA combines multiscale composite modeling with progressive failure analysis (PFA), probabilistic analysis and minimum test data to determine A and Bbasis values. Figure 1 shows the multiscale PFA. The PFA is used to produce virtual scatter data using probabilistic analysis. Figure 2 shows the process flowchart. The scatter in material strength is determined by iterating on coefficient of variations (COV) of random variables from single or multiple sources of uncertainties (i.e. fundamental material properties and fabrication variables). The iterative process replicates scatter in the strength value obtained from the test of one coupon of each material batch. If the scatter is unknown, then maximum of 10% coefficient of variation can be chosen as per the FAA regulations for composite materials. However, when test data variation is known, then a COV can be estimated, as shown in Figures 3 and 4. In Figure 3, a COV of 0.06 provides a good estimate of the test data and in Figure 4, a COV of 0.01 is used. The methodology is applicable to notched (Figure 5) and unnotched coupons and structures and has the potential of reducing the coupon count for testing by over 60%.
It is important to have design envelopes before and after the A and BBasis values are generated. The design envelop is a graphical representation of the variation in material properties (stiffness, Poisson's ratio, and strength) with variation in ply angle [e.g., 0/+45/45/90] distribution in the laminate. The graphical representation is referred to as a 'Carpet Plot' in industrial practice. A typical Carpet Plot is shown in Figure 5. The carpet plot is a powerful tool which can be used as a design reference. In GENOA, a carpet plot is generated automatically after virtually simulating ASTM standard or other chosen testing methods (Figure 1). First, the layup in the laminate is automatically varied and ASTM or user defined tests are virtually simulated for each laminate layup. This is an automated process, requiring minimal user interface. Next, the virtual test database is filtered for the carpet plot information desired and plotted graphically. The carpet plots generated need not be limited to A and Bbasis allowables. Other carpet plot options include laminate stiffnesses, first ply failure and final failure. If test data is available, then test results can be plotted on top of the virtually generated carpet plots for verification.
More focus on virtual testing and carpet plot utilization to reduce physical test matrix reduction will be shown in upcoming future newsletters.

Figure
2  Flowchart for generating Abasis and BBasis Allowables in GENOA using testing standards for ASTM and MILHDBK
17E and FAA 
Figure 3 
Cumulative Density Functions (CDFs) with
assumed coefficients of Variation of 0.06,
0.075, 0.10 for compressive composites
unnotched composite coupon [1] 
Figure 4 
Scatter from test and simulation for the
strength of the openhole coupon along with
GENOA predicted A and BBasis values [2] 
Figure 5  Carpet plot for various percent of 45 deg plies is useful to reduce coupon testing 
References:
1. G. Abumeri, M. Garg, and M. R.
Talagani, A Computational Approach for Predicting A and BBasis Allowables for Polymer Composites, SAMPE Fall Conference, TN, 2008.
Click here to email us for the
technical publication.
2. M. R. Talagani, Z. Gurdal, and F. Abdi, S. Verhoef ?Obtaining Abasis and Bbasis Allowable Values for OpenHole Specimens Using Virtual testing? AIAAC2007127, 4. Ankara International Aerospace Conference, 1012 September, 2007 ?
METU, Ankara.
Click here to email us for the
technical publication.
If you do not want to receive this newsletter, please click here. If you still have trouble unsubscribing from our website, please email genoa@ascgenoa.com with a "UNSUBSCRIBE" subject header.