JEC COMPOSITES MAGAZINE - Issue #96 - April 2015 - 31

Fig. 2: Holistic quality control approach for CFRP rib manufacturing within the LOCOMACHS project

In-situ ultrasound-based quality
control and fullfield distortion
measurement

The assessment of the DLR rib production shows the cost of this production.
Manufacturing engineering and the effort
spent on spring-in compensation are

Holistic cost
assessment
model
To conduct a holistic
cost assessment, it
is most important
to look at an actual

Fig. 3: Cost assessment for the ribs made by DLR

No96 April 2015 / jec

Feature Aeronautics

Ultrasound provides important advantages
compared to other process monitoring
methods as the sound waves can propagate
through the mould wall and therefore do
not require direct contact with the part.
So they do not affect the surface quality of
the part or vacuum integrity of the mould.
Conventional ultrasound transducers for
this application are bulky, expensive and not
sufficiently reliable due to unstable coupling
between the sensor and the mould. This led
to the development of adapted ultrasound
sensors better suited for CFRP process
monitoring under industrial conditions. The
sensors consist of a piezoelectric element
mounted directly onto the mould. They
are small and low cost, thus allowing a high
number of sensors and simple integration,
often subsequently into existing tools [3, 4].
Based on the interaction of the waves with
the composite, the most important process
parameters - like resin arrival, impregnation,
DoC, gelation, vitrification and laminate
thickness - can be obtained by advanced
data processing. The flow front can be
detected in pulse-echo mode, the same
sensor acting as transmitter and receiver,
where a notable drop in the amplitude of the
reflected sound signal occurs on resin arrival.
By smart placement of the sensors, resin arrival detection can be used to spot the risks of
race tracking and trapped air, and to ensure
complete impregnation. The cure correlates
with sound velocity, which is calculated

by the time of flight of the sound waves
propagating through the laminate thickness. Other than the often used dielectric
sensors, not only the DoC at the surface, but
a mean over the part thickness is acquired.
Thus, the autoclave process is controlled
based on in-situ cure data. The laminate
thickness is obtained from the time of flight
and calibration trials to blank out the cure
influence. The laminate thickness, and hence
the fibre-volume fraction, is controlled by
adjusting the differential pressure between
the autoclave and the resin pot to the target
value. The dimensional fidelity of the
manufactured ribs was evaluated using an
optical GOM Atos fullfield measurement
system. The measurement procedure utilizes
a novel reference-element concept which
allows consistent referencing throughout
the overall process chain and the ability
to unambiguously compare CAD, FE and
measurement results, since best-fit operations are avoided. The evaluation of the
first rib set demonstrated a homogeneous
thickness distribution
throughout the part,
and a full compensation of spring-in
distortions.

production process with the consistency
to manufacture composite components.
Virtually designed production processes
usually lack this consistency. Even though
the production process considered was run
in a laboratory, it satisfies this requirement.
Holistic cost assessment means that all the
relevant parameters of the laboratory-based
manufacturing of the LOCOMACHS ribs
were recorded and measured accurately,
except for the building services. These
could not be assessed due to the specific
cost accounting methods used within DLR.
To fulfil this vast requirement of 100%
documentation, two main paths were used.
On the one hand, the non-recurring costs
(NRC) of all production process steps had
to be evaluated, while on the other hand
a detailed recording of the manufacturing
itself had to be guaranteed to detect all types
of recurring costs (RC). The NRC include
both specific investments, such as the
mould, and non-specific ones like the autoclave. The mould, and similar components,
were included in a fictional production
scenario of 125 rib sets, respectively for 125
aircrafts depreciated over 5 years. The investment in the autoclave and other equipment
was broken down into investment/day with
depreciation over 5 years and allocated to
the days of usage correspondingly.
RC and NRC costs were assigned to each
process step individually to visualize both
cost-driving process steps and cost-driving
cost types, as illustrated in figure 3.

composites magazine 31

JEC COMPOSITES MAGAZINE - Issue #96 - April 2015

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #96 - April 2015