billion running on 7296 computer cores. To put
this in context, if we had run this on a typical dual
core desktop computer, itwould have taken a little
over 12½ years to solve.
In the next issue we will look at how we took this
philosophy and used it to help the FIA formulate
the 2021 regulations.defunct team and so had the basic spine and
running gear of our future windtunnel model
at a very low cost.
We decided thatthe majority of our basic
research would be done using computational fluid
dynamics (CFD) because, without the regulatory
restrictions that teams have imposed on their
work, we would be able to use much greater
computing power, and hence model sophistication
to investigate the problem. We engaged some
outside companies to provide the computing
power and some of the basic methodology and
then set about investigating the metricswe would
apply determining the goodness or otherwise of
the wake. This has led to somepretty extreme
computing power being appliedto our two-car
unsteady flow simulations. During one quiet
weekend when we were able tograb a lot of idle
machines we ran 14 cases simultaneouslywhich,
including following cars in seven cases was a total
of 21 cars. This led to a CFD cell count of over 2.7peaky aerodynamics. The final solutioneliminates someof the more
complex suspension elements, aswell as putting some restrictions on
geometry while still allowing this to be an area of development.
Finally, we knew that aerodynamics had to be an important
performance differentiator. Not only has thisbeen at part of
the DNAof F 1 for over 50 years, but
experience ofthe amount of effort and
cost expendedby teams on so-called
homologated formulae to perfect ‘fixed’
aerodynamics proved that over-regulation
here waspointless. In addition, aerodynamic
development is very visible and attractive to
fans, as well as being the most cost-effective
way of adding performance to a car.
However, it was also feltthat one aspect
of the current cars that was detracting from
the spectacle was their inability to race in
close proximity. We put a solution to this
right at the top of our ‘to-do’ list. My first
task was to populate our small aerodynamic
group and put inplace the tools needed to
do the required research.
I was very fortunate to be able to recruit
Jason Somerville to lead theteam and get
some extremely capable guys to assist him.
We also bought Manor’s 50% windtunnel
model from the administrators of theAERODYNAMICDEVELOPMENT IS VERY VISIBLE AND
ATTRACTIVE TO FANS, AS WELLAS BEING THE MOST
COST-EFFECTIVE WAY OF ADDING PERFORMANCE
INSIDER
UNDER THE HOODTop of the ‘to-do’ list wasto try
and solve the inability of cars
to follow each other closely26 F1 RACING JANUARY 2020
Manor’s 50% windtunnel model was an early
purchaseby Symonds and his aero group