200
9 Characterisation of Flow Properties at the Surface
are present in propulsive nozzles, in the base region of missiles or space launchers,
to quote a few examples. Since it involves transfer phenomena, like skin friction, the
heat flux is a delicate quantity to predict accurately and therefore a good candidate
for the validation of the calculation codes. The heat transfer distributions can also
reveal changes in the nature of the flow, for example the transition of a boundary
layer from laminar to turbulent state.
In addition of resolving the physical quantities on the surface of the models, the
measurement of their deformation increases the level of complexity during wind
tunnel testing. Indeed, significant load causes non-negligible deformation of the
structure. For example, in the wind tunnel the wing can be twisted by 1°, a nonnegligible amount and compromises the comparison with a CFD calculation which
can show significant variations. This increases the necessity of characterising the
deformations so as to perform the simulation on the real shape of the model under
the aerodynamic loads. This previously underestimated effect is now systematically
evaluated.
9.2 Measurement of Pressure at the Wall
9.2.1 Pressure Scanning Systems
The measurement of the pressure on the surface of a model is conventionally carried
out through small orifices or tappings (diameter of 0.3–0.5 mm) connected to a
sensor or transducers via a tubing. This technique is well known and there is a
wide variety of transducers based on the magnitude of pressure to be measured, the
response time (case of unsteady measurements), the size and cost. Due to the large
number of tappings on a model in order to have detailed information on the pressure
distribution on its surface, several tappings are most often connected to a common
transducer via a scanning device. The pressure transducer is installed in the scanner
that is mechanically driven, the pressure ports of the scanner being successively
in communication with the transducers. Figure 9.1 shows a ScanivalveTM scanner
with 4 heads of 48 channels each. The scanner most often includes a calibration valve
bringing the sensor into communication with known pressure prior to tests. The main
disadvantages of these scanners are their size, often prohibiting their installation in
the model and their relatively slow scanning speed (few readings per seconds).
The electronic scanner enabling a considerable increase in the scanning rate (up to
20,000 readings per second) consists of a set of transducers, each in communication
with its own pressure port. The transducer is generally a Wheatstone bridge diffused
in a monocrystalline silicon crystal by means of a semiconductor type treatment
(see below). The voltage outputs of the transducers are sent to multiplexers and can
be digitised selectively. Electronic scanners are quite miniaturised in the way to be
imbedded in small scale models and positioned as close as possible to the tappings
to reduce the response time. The number of sensors per scanner is generally between
20 and 50. Figure 9.2 shows a 16-channel electronic pressure scanner with reference
and calibration.
9 Characterisation of Flow Properties at the Surface
are present in propulsive nozzles, in the base region of missiles or space launchers,
to quote a few examples. Since it involves transfer phenomena, like skin friction, the
heat flux is a delicate quantity to predict accurately and therefore a good candidate
for the validation of the calculation codes. The heat transfer distributions can also
reveal changes in the nature of the flow, for example the transition of a boundary
layer from laminar to turbulent state.
In addition of resolving the physical quantities on the surface of the models, the
measurement of their deformation increases the level of complexity during wind
tunnel testing. Indeed, significant load causes non-negligible deformation of the
structure. For example, in the wind tunnel the wing can be twisted by 1°, a nonnegligible amount and compromises the comparison with a CFD calculation which
can show significant variations. This increases the necessity of characterising the
deformations so as to perform the simulation on the real shape of the model under
the aerodynamic loads. This previously underestimated effect is now systematically
evaluated.
9.2 Measurement of Pressure at the Wall
9.2.1 Pressure Scanning Systems
The measurement of the pressure on the surface of a model is conventionally carried
out through small orifices or tappings (diameter of 0.3–0.5 mm) connected to a
sensor or transducers via a tubing. This technique is well known and there is a
wide variety of transducers based on the magnitude of pressure to be measured, the
response time (case of unsteady measurements), the size and cost. Due to the large
number of tappings on a model in order to have detailed information on the pressure
distribution on its surface, several tappings are most often connected to a common
transducer via a scanning device. The pressure transducer is installed in the scanner
that is mechanically driven, the pressure ports of the scanner being successively
in communication with the transducers. Figure 9.1 shows a ScanivalveTM scanner
with 4 heads of 48 channels each. The scanner most often includes a calibration valve
bringing the sensor into communication with known pressure prior to tests. The main
disadvantages of these scanners are their size, often prohibiting their installation in
the model and their relatively slow scanning speed (few readings per seconds).
The electronic scanner enabling a considerable increase in the scanning rate (up to
20,000 readings per second) consists of a set of transducers, each in communication
with its own pressure port. The transducer is generally a Wheatstone bridge diffused
in a monocrystalline silicon crystal by means of a semiconductor type treatment
(see below). The voltage outputs of the transducers are sent to multiplexers and can
be digitised selectively. Electronic scanners are quite miniaturised in the way to be
imbedded in small scale models and positioned as close as possible to the tappings
to reduce the response time. The number of sensors per scanner is generally between
20 and 50. Figure 9.2 shows a 16-channel electronic pressure scanner with reference
and calibration.