2.6 The Main Sections of a Wind Tunnel
45
In general, wind tunnels can be classified into two categories:
– Industrial wind tunnels often large in size, are intended for testing more representative configurations such as the model of a complete aircraft or even real full
scale vehicles in automotive applications.
– Research wind tunnels are of small or medium size, devoted to fundamental
aerodynamic studies with the aim of characterising particular complex phenomena
such as laminar to turbulent transition, separation, shock-wave/boundary-layer
interaction, turbulence properties, etc. or to evaluate new concepts, for example in
the field of flow control.
However, the distinction is sometimes arbitrary as fundamental tests can be also
performed in large wind tunnels to achieve realistic Reynolds numbers or work on a
larger model better suited for the measurement of fine flow structures or quantities.
Conversely, applied studies are often performed in research wind tunnels because of
their better flow quality and cost of running.
2.7 Design and Manufacturing of Wind Tunnel Models
The large wind tunnels built until the middle of the 20th century, such as S1MA at
ONERA, Modane-Avrieux (see Sect. 4.3.1) were intended to test full scale aircraft
while avoiding problems related to Reynolds number in particular. The progress in
numerical simulation to handle some of the issues related to wind tunnel testing of
larger full scale models to satisfy the demand of the market and the cost of running
these facilities increased rapidly. This led to serious reduction usage of these facilities
and several of them being made redundant. However, the automotive sector did not
suffer similar trend as there are more modern wind tunnels for testing full scale
vehicles even under more severe climatic conditions (see Sects. 3.2 and 3.3). But
for the aerospace industry, the use of scaled down models is the general rule, with
all the complexity due to dynamic similarity covered in this book. The design and
construction of a wind tunnel model is therefore a task with very high added value
on which the quality of the results depends to a large extent.
Geometric representativeness is the major requirement. It is still necessary to
choose the adequate level of representation since all the details of a real plane cannot
be represented and would not be, given the Reynolds number effects. Conversely,
additional features might be required to trigger boundary-layer transition. It is also
necessary to take into account the deformation of the model during the tests. The
degree of roughness of the aerodynamic surfaces must be also well controlled during
the manufacturing process which sometimes conditions machining technology or
might require manual finishing.
As weight is generally not a problem in the wind tunnel, models are often made
of solid steel to minimise deformation (see Fig. 2.12). Still, some deformations can
be significant and it is necessary to identify them during the tests (see Sect. 9.6).
On the other hand, to limit the weight of the models and the cost of realisation,
45
In general, wind tunnels can be classified into two categories:
– Industrial wind tunnels often large in size, are intended for testing more representative configurations such as the model of a complete aircraft or even real full
scale vehicles in automotive applications.
– Research wind tunnels are of small or medium size, devoted to fundamental
aerodynamic studies with the aim of characterising particular complex phenomena
such as laminar to turbulent transition, separation, shock-wave/boundary-layer
interaction, turbulence properties, etc. or to evaluate new concepts, for example in
the field of flow control.
However, the distinction is sometimes arbitrary as fundamental tests can be also
performed in large wind tunnels to achieve realistic Reynolds numbers or work on a
larger model better suited for the measurement of fine flow structures or quantities.
Conversely, applied studies are often performed in research wind tunnels because of
their better flow quality and cost of running.
2.7 Design and Manufacturing of Wind Tunnel Models
The large wind tunnels built until the middle of the 20th century, such as S1MA at
ONERA, Modane-Avrieux (see Sect. 4.3.1) were intended to test full scale aircraft
while avoiding problems related to Reynolds number in particular. The progress in
numerical simulation to handle some of the issues related to wind tunnel testing of
larger full scale models to satisfy the demand of the market and the cost of running
these facilities increased rapidly. This led to serious reduction usage of these facilities
and several of them being made redundant. However, the automotive sector did not
suffer similar trend as there are more modern wind tunnels for testing full scale
vehicles even under more severe climatic conditions (see Sects. 3.2 and 3.3). But
for the aerospace industry, the use of scaled down models is the general rule, with
all the complexity due to dynamic similarity covered in this book. The design and
construction of a wind tunnel model is therefore a task with very high added value
on which the quality of the results depends to a large extent.
Geometric representativeness is the major requirement. It is still necessary to
choose the adequate level of representation since all the details of a real plane cannot
be represented and would not be, given the Reynolds number effects. Conversely,
additional features might be required to trigger boundary-layer transition. It is also
necessary to take into account the deformation of the model during the tests. The
degree of roughness of the aerodynamic surfaces must be also well controlled during
the manufacturing process which sometimes conditions machining technology or
might require manual finishing.
As weight is generally not a problem in the wind tunnel, models are often made
of solid steel to minimise deformation (see Fig. 2.12). Still, some deformations can
be significant and it is necessary to identify them during the tests (see Sect. 9.6).
On the other hand, to limit the weight of the models and the cost of realisation,