The N.A.C.A. High-Speed Wind Tunnel and Tests of Six Propeller Sections Page: 1 of 22

                                
REPORT No. 463
THE N.A.C.A. HIGH-SPEED WIND TUNNEL AND TESTS OF SIX PROPELLER SECTIONS
By JoHN STACK

SUMMARY
This report gives a description of the high-speed wind
tunnel of the National Advisory Committee for Aero-
nautics. The operation of the tunnel is also described
and the method of presenting the data is given. An
account of an investigation of the aerodynamic properties
of six propeller sections is included.
The tunnel is operated on the induction-jet principle.
Compressed air discharged through an annular nozzle
surrounding the tunnel downstream from the test section
induces a flow of air from the atmosphere through the
test section of the tunnel where the models are placed.
The forces on the model are measured by a 3-component
pholwto-recording balance.
The test results included herein comprise measure-
ments of the lift, drag, and pitching moments of six air-
foils.   The sections chosen for tests have thickness
ratios of 0.06, 0.08, and 0.10; three are based on the
Olark Y profile and three on the R.A.F. 6 profile.  The
tests were made over a wide speed range and for several
angles of attack, varying from that of zero lift to 10,
in order to investigate the effects of compressibility on
the airfoil characteristics.
The data obtained indicate that the Clark Y airfoils
are superior to the R.A.F. 6 airfoils for propeller appli-
cations except for high-pitch propellers operating at
low values of V/nD.    The effects of compressibility on
the airfoil characteristics are large and important. As
the speed of the air flowing past an airfoil is increased
the lift, drag, and moment coefficients undergo a small
numerical increase which continues until a compressi-
bility burble occurs. As the speed is increased further,
the breakdown of the flow corresponding to the com-
pressibility burble is evidenced by a marked decrease in
the lift coefficient and a rapid increase in the drag
coefficient. The speed at which the compressibility
burble occurs is dependent on the angle of attack and
the airfoil thickness; increasing either causes the com-
pressibility burble to occur at lower speeds. A com-
parison of these data with the theoretical work of Glauert
and Ackeret as regards the nature and amount of the
effects of compressibility on the lift-curve slope sub-
stantiates the theory for speeds below'that at which the
compressibility burble occurs.

A computation of propeller characteristics based on
these results is compared with the experimental results
on a full-scale propeller. The reasons for differences
are discussed and recommendations for future work are
given.
INTRODUCTION
The advantages of model testing as an aid to the
solution of full-scale problems are often neutralized by
the inaccurate reproduction of the full-scale flow in the
model test. The conditions which must be fulfilled in
the model test so that the results may be directly
applicable to the full-scale problem are twofold. First,
the model must be geometrically similar to the full-scale
object-a condition usually obtained-and second, the
model flow pattern must be similar to the full-scale flow
pattern-a condition generally not fulfilled. The prin-
cipal factors that determine flow similarity are the Rey-
nolds Number pVl/p and the compressibility factor
V/V, where V is the velocity of sound in the gas. The
first of these two factors, the Reynolds Number,
expresses the ratio of the mass forces to the viscous
forces in the gas. It is essential that this ratio have
the same value for the model flow as for the full-scale
flow. The second factor, the ratio of the velocity of
the body to the velocity of sound in the gas V/V,
indicates to what extent the flow is affected by the
compressibility of the gas. For most applications the
effects of variations in the value of this ratio are
neglected because the velocities of the air streams in
most wind tunnels are of the same order of magnitude
as the velocities of most aircraft and the effect of the
differences in the value of this factor between the
model flow and the full-scale flow is therefore small.
In addition, the speeds common to most aircraft are
low with respect to the velocity of sound in air and the
corresponding pressure differences are likewise small.
A knowledge of the compressibility phenomenon is
essential, however, because the tip speeds of propellers
now in use are commonly in the neighborhood of the
velocity of sound. Further, the speeds that have been
attained by racing airplanes are as high as half the
velocity of sound. Even at ordinary airplane speeds
the effects of compressibility should not be disregarded
if accurate measurements are desired.
399