46 / 53
14
Story
For his project assignment before starting as a master’s
student, John Fredrick Berntsen planned to crush aluminium
profiles. In came an email from SIMLab Professor Arild Holm
Clausen: “Toyota in Brussels is looking for a master’s student
to work as an intern.” John Fredrick applied and got the post.
“It was a childhood dream come true: to be able to work in the
automotive industry, to get hands on experience at a workplace
abroad, to see what it was like. It was a great opportunity to
learn,” he says. Learn he did. After surmounting some serious
obstacles, he was relieved to get some valuable results in the
end.
A WISH TO VERIFY
Toyota’s ambition was to verify the qualities of the material
model in SIMLab’s Tool Box.
To start with, they wanted to perform a test on a simple
polymer box of the kind you find in any hardware store. This
was easier said than done:
“We discovered that the boxes didn’t have the properties
described by the producer. The geometry was wrong; the
thickness and quality varied. This affected the results of our
tests significantly and reduced their value. It also stole time,”
John Fredrick confesses.
REALIST
Luckily, Ernesto Mottola at Toyota’s Technical Centre reacted
with a realist’s attitude. “This is research. There are always
unexpected results,” he said, quickly followed by suggestions
for solving the issues.
His attitude came in useful in two ways, since John Fredrick
and his Toyota mentor Yann Claude Ngueveu ran into serious
obstacles on the real thing as well.
Toyota wanted to use the lower absorber of an Avensis as a test
specimen for industrial verification of SIMLab’s polymer model.
After meticulous preparations in Brussels, John Fredrick
returned to NTNU in Trondheim for intensive days of component
testing. That’s when the real trouble started.
DISASTROUS RESULTS
“My task was to reproduce the pedestrian protection simulation
in the lab. We wanted to put the absorber in the drop tower.
That turned out to be a challenge in itself. The absorber has
complex geometry and for the test to give valuable results we
needed to control the movements. There was a risk that the
impactor would bend out of shape or break.
The first real tests were a total disaster. We started at a
very low speed, with impacts at four metres per second. The
video footage showed bending far beyond the expectations,
endangering harm to the impactor. At the same time, the cause
of the vibrations measured was not visible in the video footage.
The results were nowhere near the simulations. Something was
very wrong,” John Fredrick confesses.
A BETTER MODEL
The situation obviously put him under a lot of stress. Returning
to Brussels without results was not an option and the results
he had were useless. In the end, he and the scientific staff
at SIMLab had to improvise with pieces of wood to keep the
absorber in place. Finally, he succeeded: with surprises
alleviated, SIMLab’s polymer model showed significant
improvements from currently implemented models right from
the first simulation. It proved to have qualities that describe
the physics of polymers better than alternative models
Interestingly, in John Fredrick’s words: “It was almost scary to
see how even small details could greatly influence the results.”
TOYOTA FOLLOWS UP
Toyota’s satisfaction is illustrated by the fact that they have
decided to follow up on the work with a clear ambition to reach
industrial implementation.
“The merit of SIMLab’s material model and John Fredrick’s
activity is that realistic simulation can really help in making a
good and robust design,” says Ernesto Mottola.
CASA is also continuing. David Morin, head of the structural
joints programme, is doing further modelling.
MUTUAL BENEFIT
Professor Arild Holm Clausen, head of CASA’s Polymeric
Materials programme, stresses the mutual benefit of such
exchanges:
“This is an invaluable method for technology transfer. The first
master’s student that visited Toyota did so as a result of his
personal initiative. It was an immediate success, so the people
at Toyota were eager to continue. I hope and think this will
serve as a trigger for further exchange of scientific staff,” Holm
Clausen says.
John Fredrick Berntsen is in no doubt whatsoever about the
usefulness of his stay: “It has given me much better basis for
my present PhD work on structural joints than I would have
had otherwise.”
Over CASA’s eight year programme period, 200 master’s
students can potentially spend half a year with a partner.
John, Yann, Avensis and
POLYMERS
Potentially, a hundred years of student labour is available to the partners in SFI CASA.
Toyota has caught the essence. They’ve got help twice already.