i 4 AEROPLANE PERFORMANCE CALCULATIONS 



Method of Correcting. Considering, for instance, the cor- 

 rection for rounded wing tips, the procedure is as follows : tests 

 carried out at the National Physical Laboratory on the effect of 

 rounding off various lengths of wing tip are available. A glance 

 at these tests shows that, broadly speaking, the rounding has a 

 favourable effect both on the values of / L and L/D. From the 

 published figures it is possible by cross plotting to obtain a series 

 of curves, one for each of a set of values of X, giving, on a base 

 of the span of the rounded tip divided by the chord, the value of 

 L/D divided by that for the standard case of zero rounding (i.e. 

 square tips). This has been done and the resulting curves (and 

 a dotted curve for the k^ max correction) are given on pages 92 

 to 95. The method of applying the correction is to take the 

 value of k^ max for the standard case and multiply it by the correction 

 given by the dotted curve, and to take the value of L/D for the 

 standard case at a given value of X and multiply it by the correction 

 given by the appropriate curve, the curves in each case being 

 read at a point corresponding to the number of chords rounded 

 off at each wing tip, of course. The results obtained are the 

 values corrected for wing tip shape. 



On pages 88 to 97 curves are given for finding all the 

 necessary corrections. The various corrections are, of course, all 

 to be multiplied together. 



The weak point in the system is that corrections for, say, 

 wing tip shape, obtained on a certain wing section, are not, 

 strictly speaking, applicable to a different wing section. It is, 

 however, impossible to get a more accurate prediction until all 

 the standard corrections have been repeated on all standard 

 wings which would be a gigantic task. 



There is another but less serious weakness in the method, 

 namely, the assumption that the errors can be dealt with separately 

 and their combined effect found by multiplication. This, how- 

 ever, is an accepted scientific principle of approximation, and 

 there is no need to worry about it. The best defence, perhaps, 

 of the complete procedure is that it gives close predictions of 

 the performance of aeroplanes which after all is all that is 

 required. 



When a test on a model of the wing to be used is available, 

 the model corresponding to the wing in aspect ratio, gap/chord 

 stagger and wing tip shape, and differing from it only in size 

 and in the fact that the channel test speed is below the flying 

 speed of the machine, it is, of course, only necessary to apply 

 the correction for dimensions. 



