460 



NATURE 



[June 



[920 



mine experimentally the complete reaction of the 

 air on the moving shell when the directions of its 

 axis and the motion of its centre of gravity no 

 longer coincide. In such a case the angle between 

 these two directions is called the yaw. Until 

 recently the reaction on a yawed shell had never 

 been studied experimentally. The necessary data, 

 however, can be obtained by observation and 

 analysis of the initial motion of the shell in the 

 first few hundred feet after leaving the muzzle 

 of the gun, for in this interval the axis of a shell 

 oscillates periodically over an appreciable range 

 of yaw.i The motion is a little complicated, and 

 its interpretation is not yet completely worked out 

 in terms of the reaction of the air. Moreover, a 

 really satisfactory experimental method has not 

 yet been devised. But a start has been made on 

 the problem, and approximate values of the more 

 important components have been determined. ^ 



The somewhat crude experimental method so 

 far used consists in firing a shell through a series 

 of cardboard screens. The shape of the hole in 

 the card determines the size and direction of the 

 yaw at the instant of passing through the card. 

 From such observations the motion of the axis 

 can be plotted out against the time (if the velocity 

 of the shell is known), and the period of its oscil- 

 lations determined. The disturbing effect of the 

 cards themselves can be determined by suitable 

 control experiments and roughly estimated. Two 

 specirnen observed curves ^ traced out by a point 

 on the axis of the shell relative to the centre of 

 gravity are shown in Figs, i and 2. These 

 two paths are strictly comparative, as the only 

 difference between their circumstances is an altera- 

 tion of the axial spin. The slowly spinning shell 

 (Fig. i) has oscillations of comparatively long 

 period and large amplitude. These curves are 

 closely analogous to the curves which represent 

 the oscillations of a spinning top near its vertical 

 position. They differ only in showing slight damp- 

 ing and variation of period. 



Let us consider further this analogy between 

 a shell and an ideal spinning top. The centre of 

 gravity of the shell and the point of support of 

 the top are analogous ; so are the moments of 

 inertia about these points and the axial spins. 

 To the direction of motion in the case of the shell 

 corresponds the vertical in the case of the top ; 

 to the disturbing couple due to the reaction of the 

 air on a yawed shell corresponds the gravity 

 couple on a displaced top. The analogy so far 

 is practically exact ; it is modified by the following 

 Jacts : — 



(i) That the centre of gravity is not a fixed 

 point like the point of a top, for its velocity varies 

 both in magnitude and in direction under the 

 reaction of the air ; it describes a helical curve, 

 thus modifying the couple. 



^ 1 Such experiments are described in a forthcoming paper in the Royal 

 Society Transactions by R. H. Fowler, E. G. Gallop, C. N. H. Lock, and 

 H^W. Riphm g^ . 



- The fijrces on a model shell at rest in a steady current of air of low 

 velocity can also be measured directly in a wind channel ; the results are 

 probably applicable to a shell moving at velocities up to 700 f.s. 



3 The observation* were made for the Ordnance Committee at H. M. S. 

 Excellent, Portsmouth. 



NO. 2641, VOL. 105] 



(2) That an appreciable frictional couple exists 

 which, in conjunction with the helical motion of 

 the centre of gravity, serves to damp out the axial 

 oscillations completely. 



(3) That, in addition to (i) above, the magni- 



/^ Yertical 



Radial iScale of 

 Degrees of Yeav. 



FiG. I. — Observed path of the nose of a shell, muzzle velocity 

 1565 f.s. Rifling I turn in 40 diameters ol the bore. Total 

 time taken to describe curve shown 0.38 sec. 



tude and direction of the velocity of the shell are 

 steadily altered by gravity. 



Experiments so far carried out have determined 

 approximately the values of the couple analogous 

 to the gravity couple for velocities from 900 f.s. 



f^ Verticci£ 



nadiaJt Scale of 

 Degrees ofYccw. 



Fig. 2. — Observed path of the nose of a shell, 

 muzzle velocity 1563 f.s. Rifling i turn 

 in 30 diameters of the bore. Total time 

 taken to describe curve shown 0.20 sec. 

 [The scale of Fig. 2 is three times that of 

 Fig. 1.] ^ 



to 2200 f.s. for two different shapes of shell, when 

 the yaw is not too large. By determining these 

 couples, for. various differerit positions- of the 

 centre of gravity, rough values of the resulting 

 sideways thrust on a yawed shell were deduced. 



