JEC COMPOSITES MAGAZINE - Issue #72 - April/May 2012 - 61

factor to the pristine specimen strength data is no longer valid. Thus using a CAI allowable of 406 MPa for the laminate with 50% 0° plies as established using a blanket knockdown would result in using a compression strength value that is actually 26% too high. This would indicate an under-designed structure. Although a detailed discussion is beyond the scope of this article, note in Figure 2 the lower amount of scatter in the data of the damaged specimens. This is typical of notched laminates since the failure is being forced to occur at the damage zone rather than occurring randomly. When the A or B basis values are calculated from notched specimens rather than pristine specimens, a higher value of the damaged strength allowable can be expected. In addition, testing of notched laminates or sub-elements is much easier to perform since grip breaks (tension) or end brooming (compression) is typically not an issue. Conversely, in order not to over-design a composite structure, if one is not bound by the conventional π/4 family of laminates, it may be possible to obtain a lay-up that optimizes the strength value of interest with considerations for damage. Figure 3 shows data by Park [2] in which an open-hole laminate has optimum compression strength with a [0/90/+25/-25] lay-up rather than a more traditional [0/90/+45/-45] lay-up. The increase in notched tensile strength is even more impressive. By designing to

notched strengths, the strength of a laminate that must satisfy a damage tolerance requirement has the potential to be optimized. Figure 3 suggests that a higher material strength value can be used in the 0° direction if the ±45° plies are replaced by ± 25° plies (as long as the laminate has positive strength margins in all of the other directions that it will be loaded in the structure). Weight can be saved since the laminate can be made thinner and still meet the strength value requirement with the given design load. In conclusion, if a composite structure can never be damaged, then performing a multitude of tests on pristine laminates and applying the appropriate A or B basis statistics and required factor of safety is appropriate. However, if a composite structure might be damaged (which is typically assumed for nearly all strength critical structures), much of the time and expense of testing pristine lamina and laminates can possibly be eliminated. The use of a reduced materials characterization test matrix is application dependent and would not be appropriate for large programmes in which the composite structures are of the same fibre/resin system and will experience a multitude of loading scenarios. However, in many cases testing with damaged laminates could be the starting point of a material characterization programme to develop strength allowables. n
More information: alan.t.nettles@nasa.gov References [1]. Anderson, T.C., “Certification Cost Reduction Using Compression-After-Impact Testing”, Composite Structures: Theory and Practice, ASTM STP 1383, P. Grant and C.Q. Rousseau, Eds., American Society for Testing and materials, West Conshohocken, PA, 2000, pp. 29-46. [2]. Park, J.W., “Mic-Mac/Finite Element Analysis” in: Strength and Life of Composites, Stephen W. Tsai, Ed., JEC Composites, 2008.

Fig. 3. Theoretical open-hole strength values as predicted by Park [2]

No72 April - May 2012 /

jec composites magazine 61

JEC COMPOSITES MAGAZINE - Issue #72 - April/May 2012

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #72 - April/May 2012