Page:Aerial Flight - Volume 1 - Aerodynamics - Frederick Lanchester - 1906.djvu/412

experiment, the effective value of $\xi$ is approximately .09, with a probable error of less than 10 per cent, plus or minus.

Roughened surfaces. June 23rd, 1907 (later).

Wind, as before.

Friction plane, covered Oakey's No. 2${\tfrac {1}{2}}$ glass paper.

	Length.		Breadth.		Leverage.
Normal plane	2.5″	$\times$	1″	$\times$	3.25	= 8.1
Friction plane	12.25″	$\times$	4.6″	$\times$	8.12	= 458

$\xi ={\frac {8.1}{458}}=.0177$

On the face of it the method of the aerodynamic balance is so direct and straightforward as to leave no possibility of doubt as to the validity of the results. Under constant wind conditions it seems that a change of 10 per cent, can be readily detected, and it appears, therefore, fair to assume that this is the outside limit of experimental error.

If it had been found possible to conduct experiments in still air with the instrument in motion, it would have been difficult to resist the above conclusions; it is, however, by no means certain that, under the conditions of wind reaction experiment, the matter is quite as simple as it appears.

There seems to be some possibility that the normal plane and friction plane are not equally affected by the wind turbulence; it is even uncertain whether the influence of turbulence is in the same direction in the two cases. We know the effect of turbulence in the case of the normal plane is probably to increase the pressure reaction (§ 131), but it is by no means established that the effect is the same in the case of skin-friction. If the direction of motion of turbulence were confined to the line of motion of the main translation, it would certainly seem that the influence of turbulence would be to increase the frictional drag. If, as is actually the case, the motion of turbulence have a component at right angles to the friction plane, it is conceivable that it will give rise to discontinuity in the system of flow, which, on the

390