Journal of Accident Investigation

MATTHEW R. FOX, CARL R. SCHULTHEISZ, JAMES R. REEDER, AND BRIAN J. JENSEN
According to Airbus, the curing temperature for the CFRP
laminate was specified to be 2 0 degrees Fahrenheit. According
to Airbus material qualification data, the onset glass transition
temperature (Tg-onset) was 144 degrees Celsius in the dry
condition and 122 degrees Celsius after exposure to a climate
of 0 percent relative humidity (corresponding to a moisture
content of 0.7 weight percent). According to the engineering
drawings, the fiber volume fraction for the CFRP laminate was
60% ± 4%. The maximum volume fraction porosity permitted
in the cross-section was 2. percent.
The chemical composition of samples from each area was
assessed using IR spectroscopy; results were typical for this
composite material with no significant variances in the spectra
for each specimen.
The extent of cure and the Tg of the sample from the upper
end of the right skin panel were analyzed using MDSC, DMA,
and DSC. Portions of this sample were tested in the as-received
condition and after drying. The moisture content for the
as-received condition was approximately 0. 8 percent.
•
•
Table 1. Fifteen samples were selected from both damaged and undamaged areas
on the accident airplane vertical stabilizer.
Area Number of samples Damaged Undamaged
Right skin panel, near the aft spar 4 X
Left skin panel, near the forward spar 3 X
Right aft lug 1 X
Right forward lug 1 X
Left forward lug 1 X
Forward spar 1 X
Center spar 1 X
Aft spar 1 X
Rib 1 1 X
Rib 3 1 X
The MDSC results showed an average residual heat
value of 4. joules per gram, which corresponded to
an extent of cure that was greater than 97 percent.
The DMA results showed that in the as-received
condition, the Tg-onset measured 134 degrees Celsius,
which was between the qualification values of 144
degrees Celsius for the dry condition and 122 degrees
Celsius for the 0 percent relative humidity (0.7
percent moisture content) condition. The portion of
sample from the upper end of the right skin panel that
was tested in the dry condition had a Tg-onset of 149
degrees Celsius.
•
The extent of cure and the Tg of each sample,
including the sample from the upper end of the right
skin panel, were assessed using DSC. Results among
all samples showed no significant variance, indicating
that the extent of cure for each sample was sufficient.
MICROSTRUCTURAL EXAMINATION AND
QUANTITATIVE ANALYSIS
Sections of each sample were cut, mounted, and polished
for microstructural examination and quantitative analysis.
Cross-sections from the vertical stabilizer were prepared
and analyzed at NASA Langley and at Airbus. A typical
cross-sectional view is shown in figure 3 for a sample from the
lower end of the right skin panel.
Results of the microstructural examination and analysis
indicated that the composite structure of the vertical stabilizer
was constructed to the desired fiber volume fraction with
acceptable void content. No evidence of microcracking was
observed. The observed layups were compared to the engineering
drawings obtained from the manufacturer and, among the 1
samples, only one sample from the right forward lug showed any
discrepancies. Within the 124 layers of this sample, two layers
had orientations that were different from the drawing. Also,
two layers appeared to be missing from one position through
the thickness, but two additional layers were present at another
position. The total number of layers for each orientation in the
right forward lug was correct, and the discrepancies represented
a small fraction of the total number of layers.
12 NTSB JOURNAL OF ACCIDENT INVESTIGATION, SPRING 2006; VOLUME 2, ISSUE 1