Advances in Science and Technology

Research Journal
Volume 9, No. 26, June 2015, pages 89–95 Research Article
DOI: 10.12913/22998624/2370

EXPERIMENTAL INVESTIGATION OF THE IMPACT OF FLIGHT SPEED

ON DRAG FORCE IN THE AUTOGYRO MODEL

Zbigniew Czyż1, Paweł Magryta1, Marcin Szlachetka2 

1
Department of Thermodynamics, Fluid Mechanics and Aviation Propulsion Systems, Faculty of Mechanical
Engineering, Lublin University of Technology, Nadbystrzycka 36, 20-618 Lublin, Poland, e-mail: z.czyz@pollub.
pl; [email protected]
2
Department of Mechanics and Mechanical Engineering, Pope John Paul II State School of Higher Education
in Biala Podlaska, Sidorska 95/97, 21-500 Biała Podlaska, Poland, e-mail: [email protected]

Received: 2015.04.07 ABSTRACT

Accepted: 2015.05.08 The paper presents the experimental investigation of the impact of velocity on drag
Published: 2015.06.01 force in the autogyro model. One of the methods which simulate motion of the flying
object consists of using a wind tunnel. In this case, test object is stationary and the
motion of air is forced by e.g. a special fan. The costs related with renting and the wind
tunnel service are still very high. In this paper, the motion of the autogyro with respect
to the air, was produced by fixing this model with scale to measure the drag force on
the passenger car roof. The position of the object relative to the vehicle was checked
on the basis of numerical analysis of the airflow around this vehicle. Based on the in-
vestigations, the field of velocity and pressure, and air flow formed around the contour
of the vehicle which have been chosen, were determined. In addition, the drag force
characteristic was determined as a function of velocity and it was compared with the
values from the numerical analysis. This research is a form of verifying opportunities
for this type of research on vehicles. The conclusions derived from the analysis of the
results will be used in the future to carry out further research.

Keywords: autogyro, aerodynamics, CFD, wind tunnel, airflow.

INTRODUCTION object is limited by the size of the wind tunnel test

section. Often, the object must be scaled in order
Aerodynamic experimental investigation al- to ensure the correct proportions in relation to the
low to obtain aerodynamic characteristics, to measuring chamber [1].
verify and investigate the stability, and to verify The dimensions of the tunnel are the barrier
hazardous flight conditions of aircraft or other re- in the case studies, which should be reflected
search objects. One way to simulate the motion in real scale. If the model test results are to be
of flying objects is the use of the wind tunnel. In referenced to the real objects certain dependen-
this case, the test object is stationary relative to cies, which are called the similarity laws must be
the ground and the air movement is forced by a complied. They take into account the geometri-
special fan. This allows us to perform a lot more cal, kinematic and dynamic criteria. Geometric
research than in real flight conditions. Of course, similarity is a consequence of the need to intro-
in addition to the wind tunnel, advanced measur- duce, the so-called, linear scale. Linear scale is
ing equipment that measure both the loads acting defined as the ratio of the linear dimension of the
on the test object and visualization of flow around model corresponding to the linear dimension of
it are often required. Research and measurement the real object. Geometric similarity is relatively
in wind tunnels associated with high costs of rent easy to obtain. But in the case of kinematic simi-
and service. In addition, the size of the research larity, kinematics of individual components in-

89
Advances in Science and Technology Research Journal Vol. 9 (26) 2015

cluding the mapping of all the twists and fluctua- RESEARCH OBJECT AND METHODOLOGY
tions in the position of objects relative to the air-
flow should also be similar. Dynamic similarity The object of the research was autogyro mod-
is much harder to keep. This type of similarity is el, called Gyro - One, which is shown in Figure 1.
observed when the forces acting on the real ob- This is an unmanned RC aircraft, which is pro-
ject and the model are consistent with the terms pelled using 1000 W electric engine with a maxi-
of similarities. For this purpose, it is necessary mum take-off weight of about 4 kg. Autogyro is
to keep the number of criterion, among which over 1.050 mm in length and 510 mm in height.
we can find the Mach number, the coefficient of Main rotor diameter is 1800 mm. More details
forward velocity, Froude number, Strouhal num- about the object are added in a different paper [3].
ber or Reynolds and Locke number [2].
The more we reduce the object size, the
number of criterial and transform results on a
real object are more difficult to keep. Depart-
ment of Thermodynamics, Fluid Mechanics and
Aviation Propulsion Systems at Lublin Univer-
sity of Technology is equipped with a GUNT
HM 170 wind tunnel which has the cross-sec-
tion of the test chamber 292×292 mm respec-
tively. In order to examine EM-11 Orka plane
with a span of 13.5 m in this tunnel, it should
be reduced at least 67 times. If we do so, the
Reynolds number, which is the ratio of the aero-
dynamic forces of inertia to air viscosity should Fig. 1. Research object Gyro - One
increase the flow rate by the same times. Even
in the case of the minimum velocity for the air- Autogyro is mounted on the balance scale
craft, which is 90 km/h, the velocity in wind equipped with a strain gauge for measuring an
tunnel should be equal to 6030 km/h. Only in additional resistance force (Fig. 2). Autogyro is
this case, this number does not meet the crite- fixed to the mast (1) mounted on a linear bearing
rion (with the same working medium). It is also (6). Under the pressure of the air flow it produces
not clear if the fact that the phenomena occur- the tension on the strain gauge (2) by a rod (7).
ring at the actual flow around the aircraft and The signal from the strain gauge is processed by
the model will be the same. In addition, such a measuring cards and read using a program devel-
high flow velocity are not possible to achieve oped in LabVIEW software.
using this wind tunnel and we will be not able The whole structure was mounted on a ve-
to meet different criterion numbers. Recogniz- hicle with a luggage rack. For the experiment,
ing the need to study larger objects, it was de- Opel Vectra sedan B was used. Autogyro was po-
cided to verify the alternative method, in which
the movement of the autogyro (research object)
to air was produced by fixing the autogyro with
scales to measure the drag force on the car roof.
So it is a moving car that forces the air flow
around the test object. This method, depending
on the used scale makes possible to test indi-
vidual components provides a measure of the
aerodynamic forces and moments.
The results of resistance force obtained from
the experiment were compared to the results ob-
tained by numerical calculations. Numerically
obtained velocity field of pressure and air flow
Fig. 2. View of balance scale, 1. mast, 2. strain gauge
formed around the contours of the vehicles al- to measure the power of drag force, 3. rails, 4. plat-
lowed accurately assess the measurement in con- form for scales to measure lift, 5. stringers, 6. linear
ditions that were shown. bearings

90
 Advances in Science and Technology Research Journal Vol. 9 (26) 2015

sitioned at a height of about 670 mm measured

from the surface of the roof. Despite the complet-
ed tests including the main rotor rotation, this pa-
per presents only the results of the research object
without the main rotor blades.
Tests were carried out in a qualified Technician
Improvement Centre of Driving – Moto Park in
Ulez. This place is adopted from the old military
airport. There is a main starting lane with a length of
1500 m and a width of 30 m. It allowed for carrying
out the tests in which the influence of the ground is
smaller than on random street. The decision to un-
dertake research on a rented former military airport
Fig. 3. View autogyro placed on a vehicle
has also been taken for security reasons.

NUMERICAL METHOD AND BOUNDARY

CONDITIONS

Experimental studies have been extended by

a simulate flow around the vehicle and autogy-
ro. As the main object of interest is the autogyro Fig. 4. The two-dimensional draft of a vehicle Opel
model, it was decided that for this object the cal- Vectra B – side view [4]
culations will be carried out in 3D and in the case
of a vehicle a 2D simulation will be made, to find
out the field of flow around the vehicle in the lon- Autogyro 3D model was done in CATIA V5
gitudinal plane passing through the center of the software. Figure 5 shows the views of a solid
car, which may help in the analysis of results. For model of the autogyro. The model is without head
this purpose, the contour of the vehicle (Fig. 4) rotor, blades and an engine. In addition, aircraft
was traced in Catia v5 and then imported to De- interior components, such as accumulators, re-
sign Modeler (Ansys Fluent). After completing ceiver, cabling were not included in the modeling.
the grid, the computing model was loaded into the It is justified by the type of performed analysis,
solver (Ansys Fluent). Table 1 shows the general which is an external flow around the geometry of
settings for the numerical analysis of flow around the tested object. Further simplification involved
the vehicle geometry. small, non-significant changes like fasteners –

Table 1. General settings of numerical analysis for the Table 2. General settings of numerical analysis for the
airflow around the vehicle airflow around the autogyro geometry
Basic Type Pressure-based Basic Type Pressure-based
parameter Time Steady parameters Time Steady
Viscous Viscous
Models k-omega (2-eqn) SST Models k-omega (2-eqn) SST
model model
Fluid Air Fluid Air
Materials Density Incompressible-ideal-gas Materials Density Incompressible-ideal-gas
Viscosity Constant 1.7894e-05 Viscosity Constant 1.7894e-05
Velocity 72 [km/h] Velocity 0–90 [km/h]
Turbulent Turbulent
Inlet 10 Inlet 1
intensity [%] intensity [%]
Hydraulic Turbulent length
14 0.28
Boundary diameter [m] Boundary scale [m]
conditions gauge conditions gauge
Pressure-outlet Pressure-outlet
pressure 0 Pa pressure 0 Pa
Turbulent Turbulent
Outlet 10 Outlet 1
intensity [%] intensity [%]
Hydraulic Turbulent length
14 0.28
diameter [m] scale [m]

91
Advances in Science and Technology Research Journal Vol. 9 (26) 2015

Fig. 5. Geometric model of the autogyro prepared for the CFD analysis

bolts, nuts, and holes. Table 2 shows the general shows the pressure field around the vehicle. We
settings for the numerical analysis of flow around can observe that at the transition between the
the autogyro geometry. surface of the glass and the roof area, the pres-
sure is about –0.6713 kPa. Static pressure drop
causes an increase in the flow rate as it is shown
RESULTS AND DISCUSSIONS in Figure 9. Undisturbed air stream velocity is
20 m/s (the boundary condition of numerical
Figure 6 shows the drag force as a function of analysis), and increases nearly doubled in the
velocity of the moving vehicle. Results recorded largest vacuum.
from several testes were approximated with a sec- In order to determine the velocity field around
ond-degree polynomial line. the autogyro, additional figure 10 was shown, in
The drag force from experimental research as which autogyro model was applied into the veloc-
a function of the velocity of vehicle were com- ity field map.
pared with the results of numerical calculations. It can be seen that the velocity field, where the
It turns out that the drag force of the numerical autogyro is mounted is not uniform. In the lower
calculations account for about 50% of the experi- part of the object of the velocity is about 29 m/s
mental results. while at the top it decreases to about 25 m/s. For
In order to explain such a large differences it simplicity, we can assume that the average speed
was decided to investigate the velocity and the acting on the model is 27 m/s. If we consider this
pressure field around the used vehicle. Figure 8 value of the velocity in the calculation of the drag

Fig. 6. The drag force as a function of velocity obtained from experimental studies

92
 Advances in Science and Technology Research Journal Vol. 9 (26) 2015

Fig. 7. Comparison of the drag forces as a function of the vehicle velocity

for the calculation model and experiment

Fig. 8. Pressure field around the vehicle

Fig. 9. Velocity field around the vehicle

93
Advances in Science and Technology Research Journal Vol. 9 (26) 2015

Fig. 10. Velocity field around the vehicle with the autogyro

Fig. 11. Air streams around the vehicle

force (1) it turns out, that the component ν2 in nu- CONCLUSIONS

merical computation is equal to 400 m2/s2 (for 20
m/s) which corresponds to 54.9% in relation to Confrontation of experimental methods for
the actual test (729 m2/s2 for the 27 m/s). measuring aerodynamic forces with the numerical
Px = 0.5 ∙ Cx ∙ ρ ∙ A ∙ v2 (1) method allows for the verification of the test results.
As it is shown in the article, the alternative way of
Except for the velocity and pressure field research is less expensive than wind tunnel tests,
we should pay attention also to the shape of the however, it requires consideration of a number of
air streams which surround the vehicle. It turns external factors affecting the results. Computational
out that even the addition of significant changes model almost always differs from real experiments.
in velocity and pressure in the cross-sectional It has idealized conditions that it makes easy to re-
shape, the air flow is also affected (there is a de- peat the steady boundary or initial conditions. In
viation of flow) and in a relatively large distance case of real research, we often receive a scatter of
from the roof surface. Deviation of the veloc- results, from which we draw a trend line. Not fo-
ity vector indeed affect the aerodynamic force cusing on the drawbacks of numerical analysis, we
component. can conclude that the disparity of experimental and

94
 Advances in Science and Technology Research Journal Vol. 9 (26) 2015

CFD simulation results is due to a big impact of air REFERENCES

velocity around the vehicle. The geometric model
used in the numerical analyzes was slightly different 1. Krysztofiak G.: Tunelowe badania wpływu uster-
from the object used in the experiment. Some of the zenia motylkowego na charakterystyki aerodynam-
elements that produce additional drag force (mast, iczne modelu kadłuba wiatrakowca. Modelowanie
components mounted on slides of linear bearings, inżynierskie 42, 2011, 227–232.
rotor head) were not properly covered. 2. Krysztofiak G.: Kryteria podobieństwa aerody-
Research allows to estimate the value of namicznego w tunelowych badaniach modeli
aerodynamic force components. A more detailed wirników wiropłatowych statków powietrznych.
analysis with the additional use of CFD allows for Modelowanie inżynierskie 42, 2011, 217–226.
the correct interpretation of the results. Further 3. Czyż Z., Wendeker M., Raczyński R.: Koncep-
research should take into account all the above- cja hybrydowego statku powietrznego z napędem
described issues with particular emphasis on the wielowirnikowym. Logistyka 6, 2014, 2936–2945.
selection of a vehicle and platform on which the 4. http://opelastra24.eu.interiowo.pl/modele/opel_
test object will be attached. vectra_b_wymiary.html

95