Bjorn’s Corner: Aircraft drag reduction, Part 5

November 17, 2017, ©. Leeham Co: In the last Corner, we described how the Wright Brothers (bicycle manufacturers in Dayton (OH)) decided to research their own aerodynamic data with the help of their own designed-and-built wind tunnel.

The wind tunnel was not more advanced than what had been done before. But their measurement system was. It built on their bicycle test setup, Figure 1.

Figure 1. The Wright Brothers’ arrangement for testing wing shapes on their bicycle autumn 1901. Source: NASA.

The Wright Brothers wind tunnel experiments

To design a flying machine, you need to calculate how large wing you need to get the necessary lift. You also need to calculate how much drag the machine would produce, to size the forward motion system.

At the turn of the 20th century, it was understood the calculations looked like:

Lift or Drag = “the force of air”  times  “the air velocity” squared,  times  “the wing area”  times  “the wing efficiency factor.”

If we write this as a formula we have L = c * V2 * S * Cl

Where:

L = Lift force, or for Drag force we use D

c = constant, representing the force of air (the air pressure constant)

V = speed of air

S = wing or drag area

Cl  = wing efficiency in generating lift per unit area. For drag, it would be Cd, the amount of generated drag per unit area

The problem at the time was the constant c. Smeaton estimated this in Whirling arm experiments, but it was 72% too high. Many at the time had started questioning this constant, which had ruled for 140 years.

One of the reasons the Wrights wanted to do their own measurements was to set the correct value for c. Secondly, they wanted to understand Cl and Cd for a number of different wing forms, so they could settle on a suitable one for their gliders and later for manned flight machines.

A smart measurement method

The measurement of the force of a wing or other object in a wind tunnel was either done with spring scales (Whenham) or weight balances (Lilienthal, Phillips). The problem was, to fix the Cl or Cd to a correct value, you needed to know c if you had the force. The Wrights found a way around this.

If they built a balance, which as counterweight had a known aerodynamic object (for example a flat plate blocking the air), they could classify their test objects characteristics to the known characteristics of this object. With it, c would be working on both sides and wouldn’t have to be known.

They first tested their idea using a bicycle wheel, Figure 1. They placed a wing element on the wheel which would be parallel to the wind when they cycled. Then a flat plate that would be perpendicular to the wind at a 90° angle. When the wheel rotated so the objects balanced each other, they could deduce the forces based on the moment arms for the balanced state (equal and opposite torques existed when the two forces and their moment arms around the wheel centre were equal).

As it wasn’t easy to measure the wheel’s equilibrium angle while cycling, they moved the test to a similar balance in their wind tunnel, Figure 2.

Figure 2. Lift force balance placed in the wind tunnel. Angle scale is read trough tunnel top window. Source: NASA.

The balance’s function is easier to understand from Figure 3.

Figure 3. The principle of Wright’s wind tunnel balance. Source: NASA.

The model (here a wing) forces the light brown stick to the right in the tunnel (seen from above through the tunnel’s glass plate). This swings the drag plates to the right as well. Their collective moment arm gets longer around the blue pivot points. At some angle (read on the dial), the lifting torque and the drag torque from the plates are equal. Then read the dial, and calculate the lift versus drag ratio.

Hundreds of tests

During September to December 1901, the brothers tested over 200 airfoil types, Figure 4. They could quickly produce these from steel plate in their bicycle shop.

Figure 4. A sample of the airfoils tested. Source: NASA.

When they had done an overall classification, the best candidates were more thoroughly tested with new and more precise balances. In the end, the data for the best 48 wing models were catalogued.

With the tests, the Wrights now had the most detailed aerodynamic data in the world for objects needed for building aircraft.

Their clever balances measured Lift/Drag ratios directly and through further tests with other balances, Lift and Drag coefficients.

The Lift/Drag data was the most useful information. With this single value, one can deduce that if one needs to lift 200lb (the weight of a person), an L/D ratio 10 wing would produce 20lb of drag.

More knowledge needed

By now the Wrights had knowledge of what wing profiles and shapes to use. But they also needed knowledge of how to control the aircraft and how to propel it forward. This is the discussion for the next Corner.

4 Comments on “Bjorn’s Corner: Aircraft drag reduction, Part 5”

1. Thank you very much, Björn. Actually, it helps a lot to understand how the science of aerodynamics was started, and how difficult it get to understand all the forces that work into a flight.
It also helps me better unde4rstand why the airships were so much developed before, since Archimedes effect was so well known before.

2. Thank you, great information.

3. Sometimes it takes someone practical to bust out of a block.

Those guys certainly were down to earth even through they took us to the skies!

4. Wait, there is something I missed. I was given to understand that the Wright Brothers’ (disconcerting) discovery was that the center of pressure of a curved airfoil moved forward as the angle of attack increased and back as it decreased, making their glider tail heavy in a climb (something few of their predecessors had managed before) and nose heavy in a dive. The reason for so many test wings in their wind tunnel was to find a profile that made that problem manageable. Have I been misled all these years?