From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Analysis of fan power in relation to speed, exploring blade design solutions like variable pitch and twisting.
| Identifier | ExFiles\Box 75\3\ scan0140 | |
| Date | 19th September 1924 guessed | |
| -2- Contd. (2) Rate of increase of power with speed. The speed of the car and the speed of the fan increase at the same rate. It therefore follows that the power absorbed by the fan will increase as the cube of the speed of the fan, and the efficiency of the fan will remain constant. To make the power increase at any {other} rate involves either:- (i) Making the speed of the fan increase at a slower rate than the speed of the car, in which case the power could be made to decrease near to zero at high speeds. (ii) Using a variable pitch propeller. The first solution is not practical, and I assume that you do not wish to fit a variable pitch propeller on a motor car chassis. The only partial solution left is to so design the blades that they twist under load in such a way as to decrease pitch and so decrease the value of the torque coefficient Kq in the expression Torque = pKq n²D⁵. The twist is approximately proportional to N². Hence if there is any twist at all it will get greater as the speed goes up, this keeping down the power absorbed. It is quite impossible by this means to prevent the power increasing to some considerable extent since the thrust must increase if the blade twist is to increase. This decreasing pitch effect is dangerous in a wooden screw since the blades tend to split. It can be obtained in your present type of construction if you sweep back the vlades opposite to the direction of rotation and make them of thin gauge steel. I doubt whether aluminium thick enough for general strength would be sufficiently flexible. The twisting effect and also the efficiency would be greater if you rounded off the tip on the leading edge. With the blade angles I have recommended the efficiency would increase slightly as the blade twisted, or alternatively, as the belt slipped (within limits). If you have any facilities for doing so it would be interesting to try three amounts of sweep-back and to measure the power absorbed at various speeds in still air over as wide a range as possible. I should add that if the back pressure in the bonnet mounts up at high speed this makes matters worse, since Kq in the expression Torque - pKq n²D⁵ will increase and the power will increase at a greater rate than the cube of the speed. | ||