JEC COMPOSITES MAGAZINE - Issue #120 - April 2018 - 61

lmproving composite manufacturing reliability and yield with affordable models and simulations

Workforce training
For companies that want to develop this modelling and simulation capability in-house, the learning curve can be substantially reduced by creating teaching modules that will make the engineers familiar with the use
of the software program and help them explore various paths to design
their process. This option makes sense for a company that has committed to developing a series of parts using the liquid composite moulding
process. Processing engineers and technicians with higher composite
manufacturing skills can directly handle a number of issues at the manufacturing floor without the support of their design office. Significant
time can thus be saved.
Consulting
Companies that do not have the necessary engineering expertise or
computational resources can, as a first step, outsource the process design. They can do this gradually and move forward to the next step as
they build confidence. For example, the first step may be to get an idea of
the fill time and the fabrics to use in the process. It could also be to find
what process makes sense: RTM, VARTM or RTM Light, for example.
The next step could be to perform flow simulations and learn from them
as to the resin flow behaviour and use this information to design the process. Both sophisticated software and composite manufacturing science
imply steep learning curves. Hybrid consulting-training can help bridge
the gap between in-depth understanding and technological reality.
Conclusions and future outlook
To stay competitive in the marketplace, advanced composite manufacturing industries will need to reduce labour and improve their efficiency
and yield. Most SMEs have ignored or have not been able to afford science-based approaches to manufacturing and continue to use trial and
error, which is known to be time consuming with cost overruns and has
been a big deterrent for replacing metal parts with composite ones.
Commercial process simulation tools based on the fundamentals of
composite processing have been available for a couple of decades and are
finding acceptance by large companies as risk mitigation tools. However,
they need to invest in licenses for such simulation software, in the characterization of the material properties needed as input for such simulations, and in engineer resources to understand, learn and execute the
simulations. Composite design platforms also need to be integrated with
process simulation so that the designer can iterate the design not only
for properties but also for ease of manufacturing. Although very crucial,
this issue has eluded software developers due to the different approaches used for CAD managing and meshing needs for process simulations.
Once such a tool is developed, it will initially be useful for companies
that already have in-house CAD software and expert designers.
If simpler tools become available, providing better guidance to improve
the manufacturing process, SMEs will be on board as well. Once they
understand the relevance of such tools and the potential cost savings,
one can wean them off trial-and-error approaches. Hence the need to
develop such tools for processes, for the characterization of material
properties and/or for the creation of a database that can be made universally available. However, as part complexity increases, simple tools

Fig. 5: Conceptual schematic from CAD to automated manufacturing using sensors
and simulations [11]

can only provide rudimentary know-how but process simulations can
provide better insight and more in-depth understanding. As SMEs do
not have the resources to execute such simulations, these requirements
can be bypassed by developing turnkey automated simulations that will
provide the process design for a robust process by integrating the various
intermediate steps without requiring any additional input.
The future outlook for advanced composite manufacturing could be
quite promising if the simulations are integrated with the manufacturing
equipment and sensors are embedded in the tool and allow for on-line
process control, as conceptually illustrated in Figure 5.
For example, in liquid composite moulding processes, sensors could
be embedded in the mould to detect resin arrival, and this information
could work in tandem with the simulation so that auxiliary actuators direct the resin flow to resin-starved regions, making the process robust by
taking online action to account for variability in the process or material
parameters. Some of these features have been demonstrated at the laboratory scale [11-14] and implemented in injection moulding equipment.
This technology is ripe for being transferred to liquid composite moulding equipment.
However, no equipment maker is ready to invest as they do not see a
growing market demand for this feature. On the other hand, the market demand is limited as the cost of composites is high due to lack of
automation and production is marred with uncertainly and part-to-part
variability due to the trial-and-error approach.
This chicken-and-egg problem can be resolved if large aerospace and automotive companies start to replace many of their structures with composites requiring the development of such equipment with integrated
sensors, simulations and automation, encouraging equipment makers to
fabricate such equipment.
SMEs can then gain from the use of "smart" equipment to make consistent parts that could compete in price and quality with traditional metal
components.

More information:
www.ec-nantes.fr

N°120 April 2018

/ jec composites magazine

http://www.ec-nantes.fr

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #120 - April 2018