JEC COMPOSITES MAGAZINE - Issue #117 - November/December 2017 - 65

Building a case for automated composite lay-up

creating defects, such as fibre wrinkling. One
might easily assume that laying up a part such
as the hexagon shown in Figure 2 is simple. Yet,
the tight radii and abrupt changes in direction
between facets pose complicated programming
challenges shown in Figure 3. To ensure that the
intended fibre angle, steering, and material compaction are all maintained, the software and programmer may develop a laminate significantly
different than what the designer anticipated.
Different lay-up strategies will significantly alter
part performance. Engineers can take advantage of programming and simulation software
to test the results of AFP/ATL on their parts.
Most
programming software contains some sort
Figure 2. Visualization of one ply covering strategy
Figure 3. Steering analysis of hexagon part, red areas
of design loop where the programmed material
using VERICUT Composites Programming software
indicate excessive material steering
can feed back into analysis software to guarantee
final AFP laminate. To deal with these particulars, different approaches part integrity. Automating the lay-up process allows engineers to analyse
exist for determining how material should land on a lay-up tool. Part pro- the as-built part instead of the handmade part that will not match the
grammers use different path generation strategies to reach the best com- final product constructed on the shop floor. Automation empowers enpromise between manufacturing feasibility and stress design require- gineers to control the lamination processes instead of relying on techments. Current specifications for a composite hand lay-up process will nicians. Removing the operator from the process eliminates the human
need significant alterations to coincide with the nuances of automated error inherent in composite lamination, creating less part-to-part variamanufacturing. AFP software can also prove beneficial to quality engi- tion. This advanced design loop allows engineers to create the optimal
laminate.
neers in helping them establish new and achievable guidelines.
After ply-covering, the programmer must decide on the machine offpart motion or linking. AFP/ATL machines lay down the material in
strips or courses, with work pieces typically requiring more than one
course to cover the entire part. The programmer must decide on a
strategy to connect one course to the next. The order and technique in
which the programmer decides to link these courses largely impact the
part runtime and quality of the laminate. Subsequently, this movement
needs post-processing, which is a conversion of the program into a language that the control on the machine understands. Lastly, the user may
now simulate the construction of the part on a variety of different virtual
machines such as the robot shown in Figure 1. Soon after, the business
begins to realize the potential for time savings and part reproducibility.
Currently, part size and geometry dictate machine selection. An infinite
number of hardware solutions exist and companies should look to AFP
software to aid in the machine architecture selection. However, before
even considering the machine type, the company should first investigate
if AFP/ATL produces the desired outcome of the part in question.

A variety of factors
AFP/ATL does not provide the optimal solution for all composite parts.
A part's success with AFP depends on a variety of factors such as machine and material cost, deposition rate, cycle time, and part geometry.
Parts with tight curvature can invoke difficulties with steering, fibre angle
deviation, and roller compression. Any lay-up more complex than a flat
plate will likely require some change in material direction, while maintaining contact with the lay-up surface (steering). Material width, density,
and modulus all contribute to a material's ability to be "steered" before

In manual lay-up, even the best procedures and specifications cannot ensure two operators will create the same laminate. Too many
variables exist from person to person. Material orientation and positioning, dart size & locations, and fabric elongation all make part
reproducibility nearly impossible. Yet, instead of fixing the process
to ensure less part-to-part variation, engineers currently cope with
this inconsistency by designing a stronger laminate than required.
They add extra material and subsequently weight, which diminishes
the purpose of selecting composites in the first place. With a more
dependable process, engineers could design a structure with less uncertainty. For an industry that uses advanced materials and analysis
techniques, companies should look to match this innovation with
what advanced manufacturing techniques offer.
Industries such as aerospace and automotive continue to evolve and
use an increasing amount of composite materials. To keep up with
production rates and tight tolerances, companies will continue to
look to replace manual lay-up methods with automated ones. Before
diving head first into this technology, businesses need to first understand if they can profit from automation. With the use of advanced
programming and simulation software, engineers can experiment
with virtual machines to acquaint themselves with the benefits of
automation, while also learning the idiosyncrasies of automated
composite manufacturing.

N°117 November - December 2017

More information:
www.cgtech.com

/ jec composites magazine

http://www.cgtech.com

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #117 - November/December 2017