FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1914
1914 - 0201.PDF
FEBRUARY 21, 1914- (/ycHT] speeds. And, therefore, if the square law assumptions hold with sufficient accuracy, the experimental mean pitch is constant; that is, it is a constant multiple of the diameter, in the same way as the mean chord pitch. Another way of putting the same thing is that the closeness of the experimental mean pitch to a constant value at different speeds is only a further indication of the degree of approximation to which the square law assumpiions.are true. In the course of his paper before this Society last year, Mr. Low gave some figures taken from the Bulletin of the Koutchino Laboratory, which only showed a variation of ± 2-5 per cent, in the value of the experimental mean pitch over a considerable speed range ; this variation is probably covered to some extent by experi mental errors. In the course of the propeller experiments carried out at the National Physical Laboratory, it has been found invariably that the experimental mean pitch for a propeller is constant within the limits of experimental accuracy for a speed range of 15 to 50 feet per second. Very little experimental data is available on this particular point, but that which has been quoted indicates that the departure from the square law is very s-light under these conditions, and that the experimental mean pitch may be taken as constant for practical purposes. Another objection advanced is that it is impossible to determine the value of the experimental mean pitch during the course of the design of a propeller. But recent experiments all tend to show that the value obtained by actual test differs very slightly from that obtained by calculation from the no lift lines of the various sections, which are used instead of the pressure face chords to determine a mean value of the pitch for design purposes. Methods of Propeller Design.—It is well to give a little considera tion to the various methods that are in use for designing propellers ; these may be divided roughly into two groups. The first of these, which has been used for the design of almost all existing water pro pellers, is to design from previous experience. When a new propeller is required to work under conditions differing slightly from any previously encountered, a propeller that has actually proved itself to be successful under somewhat similar working conditions is chosen as a model. The new propeller is then designed by making slight alterations from this selected model. This method has proved very successful for marine propeller work, the only limitation in the method being that the variation made between any two successive propellers must, of necessity, be small. It must not be forgotten that there is, at the present time, a big difference between the sciences of marine and aerial propulsion, which difference is merely a question of the relative ages of the two sciences. No doubt one of the chief reasons for the success of this method, when applied to the design of water propellers, is that the number of different propellers working successfully under various conditions is very large ; and it is generally an easy matter to select a propeller that shall serve as a model for slight alterations or variations. But aerial propulsion is not yet in this happy state ; the science is so new that there has not been time for sufficient data to accumulate to allow this method of design to be applied with certainty of success. The Aerofoil Analogy.—The result has been that a new and more powerful method of design has been evolved by aeronautical 201 Al- tO O / / i f / f • *—^** - ! f\_ Fio.5. 1 taf. engineers ; this method may be termed the blade element theory. This was first developed by Lanchester and Drzewiecki, but has since been considerably modified by subsequent designers. The method is so well known and was so ably expounded by Mr. Low, in a paper read before this Society last year, that it is only necessary to mention the chief points at present. The root assumption underlying the theory is that any small element of a propeller blade (cut off by two cylinders, concentric with the shaft) is subject to the same forces as an element of an aerofoil of equal length (cut off by two planes parallel to the wind direction), the resultant velocity of the element and the inclination of the element to the direction of the resultant velocity being the same in the two cases. In the case of the propeller blade element the velocity considered is the resultant of the translational velocity and the peripheral velocity at the radius of the element. It is immediately apparent that this is a very far-reaching assumption, and it is now necessary to investigate to what extent it is justified by the results obtained. This practical test, as to whether the method gives satisfactory results when carefully applied, must always finally decide the application of theories of this kind. The method, as developed by Drzewiecki and others, goes a little farther than the above assumption. It is shown that the efficiency of a propeller will be a maximum, when every element is inclined at such an angle to the direction of its resultant motion that, if used as an aerofoil, it would give a maximum value for the ratio of lift to drag. Hence it has been usual to assume a sort of mean section for the propeller blade ; and from experiments on aerofoils in a wind channel, to determine the inclination of this mean section, such that it will give a maximum value of the ratio of lift to drag. The propeller is then so designed that, under its working con ditions, every element of the blade is inclined at this best angle of attack to its resultant velocity, so that under these conditions the over-all efficiency of the propeller shall be a maximum. The forces on each element can then be determined and the total force on the propeller blade be obtained by integration. In Mr. Lanchester's method the integration is graphical in order that the width of the blade may be varied to fit in with other considerations. Drzewiecki, however, worked out his equations on the assumption of a blade of constant width; and Mr. Low last year showed us the equations with the width of the blade expressed as a function of the radius. The object in both these cases was evidently to obtain expressions for the forces on the elements of the blade that could be integrated without resource to graphical methods. The speaker would like to put in a plea for a further extension of the method, as one strong objection to the method in its present form is that the section of a large portion of the blade must differ very considerably from the mean section taken ; this is necessary from considerations of strength and stiffness. But results of tests on so many aerofoils, of very widely varying sections, have now been published, that it should not be difficult to pick out a series to repre sent almost any propeller blade at various points along the radius. The resultant force on the blade could then be obtained by graphical integration with a much closer approximation to its actual value than is possible by the method of taking a mean section for the blade. If this source of error be eliminated in this manner, the comparison between the forces obtained on a propeller and those obtained by integration from the results of tests on the appropriate sections in a
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
