Page 10 - Shimadzu Journal vol.2 Issue3

New Technology

Adoption of a 10 MN Fatigue Testing Machine to Expand the Evaluation
Ability of Composite Materials for Aerospace Structure

Sunao Sugimoto
Advanced Composite Research Center, Institute of Aeronautical Technology, Japan Aerospace Exploration Agency

In recent years, there has been a rapid increase in applications for carbon we succeeded in very accurately predicting, within a 1 % margin of error,
fiber reinforced plastics and other composite materials for commercial the load at which delamination will start at the end of the stringers.
passenger aircraft. Even as much as half of the weight of some aircraft
As shown in Fig. 2, the need for the testing of large-scale samples that
being used today consists of composite materials. When building aircraft,
simulate aircraft structures has been felt both within and outside JAXA. In
not only do strict safety requirements need to be met, but the need for
particular, in order to perform a compression test, a large-scale
lightness also should be satisfied. For these reasons, there is little leeway
anti-buckling jig needs to be placed around the test sample, and in many
when it comes to safety as compared with other structures. The
cases, the space between the columns of the testing machine needs to be
Composite Materials Handbook-17 (CMH-17) provides information and
quite wide. Additionally, in the area of research and development for
guidance necessary to design and fabricate aircraft structure from
rocket structure, there is a need for load tests that can handle a portion of
composite materials.
a full-scale structure. Moreover, due to the increasing size of test samples
As a method for assuring the strength of composite material structures, and the technologically advanced nature of materials, the need for a
CMH-17 offers the concept of a Building Block Approach (BBA), as shown testing machine that is capable of handling much greater loads and
in Fig. 1. In this approach, strength data for "coupons (strip-shaped provides a sufficiently wide amount of space between its columns is
samples)" is acquired first. Subsequently, the mechanical properties of increasing.
"elements" such as C type and T type specimens are measured and then
At JAXA, however, the existing testing machine could only perform static
structural members "details" and "subcomponents" are measured.
tests up to 3 MN and fatigue tests up to 2.5 MN. The space between the
Finally, strength measurements for the target "components" are 600 kN
columns of the testing machine was also narrow. Consequently, in (Initiation)
performed to produce the accumulated strength data. This method allows
February, 2014, we installed a Shimadzu Servopulser 10 MN fatigue
the strength of the target structure to be assured statistically.
testing machine (EHF-UV8MN-830). An overall view of this testing
Thousands of tests need to be performed like this because composite machine appears in the photograph in Fig. 3; Table 1 lists its specifications.
materials are inherently anisotropic and non-homogeneous. Compared
Immediately after adopting this 10 MN fatigue testing machine, it was 685 kN
with metallic materials, they exhibit multiple types of coexisting strengths.
used to test a 2.5 m dia. cylindrical shell with a composite lattice structure,
From a statistical point of view, there is an "A" value, which means that at
which is a research model of a rocket structure in the Space Transportation
a 95 % confidence level, 99 % or more of the test samples are strong
Program by Space Transportation Mission Directorate, JAXA. Additionally,
enough to withstand destruction. There is also a "B" value, which means
at the Advanced Composite Research Center, we are also proceeding with
that at a 95 % confidence level, 90 % or more of the test samples are
preparations for the compression strength testing of three stringer
strong enough to withstand destruction. In the field of composite
stiffened panels, as part of our research into low-cost composite materials.
materials, the "B" value is most commonly used. In order to obtain the
value for "B," n = 29 test samples are required for each item being tested.
In many cases, this is rounded up to 30. Moreover, in the case of
composite materials, there is a close connection between the COMPONENTS
manufacturing process and strength or inconsistencies in the strength. For
this reason, five independent batches are commonly made. From each
batch, six test samples are selected to form a total of 30 samples to be
tested. With this, the uncertainties of the manufacturing process can be NON-GENERIC SPECIMENS SUB-COMPONENTS
also evaluated. STRUCTURAL FEATURES
At the Advanced Composite Research Center, Institute of Aeronautical
Technology of JAXA, research aimed at expanding the applications for DETAILS
composite materials in aircraft, rockets, etc. is underway. So far, in
research aimed at lowering the cost of composite materials for aircraft, we
have implemented Vacuum-assisted Resin Transfer Molding (VaRTM)
research for main wing box structures. As part of this, we have, as shown ELEMENTS
in Fig. 1, performed testing starting at the coupon level, proceeded
onward to test full-scale simulations of the wing structure, and then
performed finite element analysis corresponding to this. This follows the GENERIC SPECIMENS COUPONS DATA BASE
BBA. As one example, we have used a 2.5 MN hydraulic fatigue testing
machine to perform a fatigue test on a large stiffened panel that simulates
a portion of the lower skin panel of a main wing, and then have
performed finite element analysis corresponding to this. Fig. 2 shows the
results. In this fatigue test, as a result of the prior finite element analysis, Fig. 1 Conceptual diagram of the building block approach (BBA)

88