July 13, ign] 



NATURE 



61 



considerable time to complete, but results of practical 

 importance have been obtained in the determination of the 

 negative or positive righting moments acting on models 

 -of airships of different forms. If an elongated model of 

 the customary fish-shaped form be supported in a current 

 so that it can turn about an axis through its centre of 

 gravity, it tends to set itself at right angles to the current ; 

 when it is oblique to the current a moment acts on it 

 tending to increase the obliquity. The amount of this 

 moment has been investigated for different angles of 

 obliquity in the case of several models ; combined with the 

 measurements of "lift" and "drift," this enables the 

 magnitude and line of action of the resultant force on the 

 model at any obliquity to be determined. 



The next step was to find the amount of fin area 

 necessary, and the best position for the fins, to give a 

 positive in place of a negative righting moment. Experi- 

 ments for this purpose have been carried out, and have 

 led to interesting results. It was found, even with a 

 considerably elongated tail, that if the fin were placed 

 towards the rear of the tail and close to the body, the 

 portion of the fin nearest the body was comparatively in- 

 active, owing to the slow movement of the stream in this 

 region. This slow motion of the stream near the tail was 

 confirmed by photographs taken to investigate the nature 

 of the flow past fish-shaped models. 



These experiments have enabled the amount of fin area 

 necessary to give a positive righting moment to be deter- 

 mined. Other questions which have to be investigated in 

 connection with the general problem of stability are the 

 effect of the instability of the wake, which does not leave 

 the tail symmetrically when an airship body is moving 

 parallel to its axis, and the " damping " action of the 

 air as regards any motion which involves rotation, as 

 when oscillations are set up, or in turning. Apparatus has 

 been constructed with the aid of which it is hoped that 

 the damping coefficient may be determined, and the effect 

 of wake instability examined. 



Air Resistance of Wires and Ropes. — The experiments 

 which have been carried out on wires and ropes have also 

 furnished results which will probably be of general interest. 

 The tests were made on a large number of wires and ropes, 

 including smooth wires of diameters ranging from 0-04 to 

 0-25 inch, with wire ropes of five or six strands and hemp 

 ropes of three strands of diameters ranging from o-i to 

 0-6 inch. 



The air resistance of the stranded ropes, per unit of the 

 sectional area exposed to the wind, is found to be of 

 approximately the same amount as that for small square 

 plates. No great difference was found between wire ropes 

 and hemp ropes at the same velocity. In the case of 

 smooth wires, the resistance per unit of sectional area is 

 appreciably less, the difference being of the order of 20 per 

 cent. 



Experiments were also made on the air resistance of 

 vibrating wires ; no appreciable effect on the air resist- 

 ance was found at the vibration velocities reached, whether 

 the wires were made to vibrate in a plane parallel to the 

 direction of motion or perpendicular to it. It may be 

 inferred, therefore, that in practice the air resistance of 

 wires can be calculated on the basis of the values given 

 for stationary wires in the table printed in the account of 

 these experiments given in the technical report. 



It is of interest to note that the values found at the 

 National Physical Laboratory are in close agreement with 

 the results obtained, also during the past year, for the 

 resistance of stationary wires and ropes at the well-known 

 aerodynamical laboratory at Gottingen under the direc- 

 tion of Prof. Prandtl. The work on airship models, and 

 the results for the resistance of inclined plates, are also 

 in general accordance with the observations of a similar 

 character which have been made at Gottingen. 



Wind Resistance of a Radiator of Honcvcomb Type. — 

 Experiments have also been made on the wind resistance 

 of the honeycomb form of radiator. For the purpose of 

 these tests, a scale model was made and its resistance 

 compared in the wind channel with that of a solid block 

 of the same external dimensions. The conclusion was that 

 the wind resistance of such a radiator, in which the net 

 area is about 25 per cent, of the total area, is approxi- 



NO. 2176, VOL. 87] 



mately one half that of a flat board of the same 

 dimensions. 



It was considered of some interest to determine also the 

 velocity of the air flow through the tubes of the honey- 

 comb, and its variation with the length of the tube. In 

 these experiments the tubes of the actual radiator were 

 employed, and with a tube length of about 4 inches the 

 wind velocity in the tube was found to be about three- 

 fourths of the mean wind velocity outside. Reducing the 

 length of the tube by one half produced an increase of 

 only about 15 per cent, in the air velocity through the 

 tube. 



The general conclusion was that the honeycomb form of 

 radiator is fairly efficient, and it does not appear that any 

 considerable increase in efficiency can be obtained by 

 diminishing the length of the tubes or by increasing the 

 ratio of diameter to length beyond the value, viz. 1 : 12, 

 which obtained in the type tested. 



Other Tests in the Air Channel. — Among the other 

 investigations which have been made in the air channel 

 may be mentioned a series of tests on models of dirigible 

 sheds of different design to determine the resultant force 

 on each due to the wind : tests to determine the wind 

 resistance of a model gondola ; and a number of experi- 

 ments on a model of the girder designed by Fabre and 

 used in the new type of Paulhan aeroplane. The experi- 

 ments on this girder were directed to the determination of 

 its head resistance at various angles to the relative wind 

 and also of the lift obtained from it when inclined about 

 an axis parallel to its length. It was found that the 

 efficiency of the girder, regarded as a small biplane, was 

 about 50 per cent. 



Small Water Channel for Visual and Photographic 

 Work. — A small water channel has been constructed with 

 a view to the investigation of the nature of the flow round 

 an obstacle in a fluid medium. In this a steady stream of 

 water is kept in motion, into which small models of plates, 

 aerofoils, airship bodies, &c, can lie introduced, and the 

 nature of the flow can be studied with the aid of colouring 

 matter added locally to the water. 



With this apparatus interesting photographs have been 

 obtained of the flow past plates and balloon models. 

 These have shown that even for an elongated fish-shaped 

 airship model the relative velocity of flow near the tail is 

 considerably less than in the main stream, thus explain- 

 ing the relative inefficiency, as regards the production of a 

 righting moment, of the portion of a stabilising plane close 

 to the body in this region. 



Some valuable information has also been obtained with 

 this apparatus as to the eddy formation in the rear of 

 plane and curved plates, and the experiments on these will 

 be continued. 



Wind Pressure on Square Plates. — In connection with 

 questions arising out of the model tests and the determina- 

 tion of the correction factor, if any, to be applied in pass- 

 ing from the results obtained in small model experiments 

 to the corresponding full-scale values, an examination has 

 been made by Messrs. Bairstow and Booth, of the National 

 Physical Laboratory, into the experimental results obtained 

 by different observers for the air pressure on square plates. 

 Both Eiffel and Stanton in their experiments on square 

 plates have found that the wind resistance per square 

 foot of a small plate is less than that of a large plate, 

 the difference, according to Stanton, as between plates 

 2 inches square tested in the wind channel and plates 

 10 feet square exposed in the open being about 20 per 

 cent. 



In the report of the committee for the year iqoq-10 

 (p. 38) Lord Rayleigh pointed out the general form which, 

 according to dimensional theory, the law of variation of 

 resistance with dimensions must assume, and showed that 

 such a variation as found by Stanton for square plates 

 involved also a departure from the law according to which 

 the resistance of a plate in a current of air is taken to be 

 proportional to the square of the velocity. Messrs. Bair- 

 stow and Booth have shown that a formula can be found, 

 falling under the general type indicated by Lord Rayleigh, 

 which accurately represents the results both of Eiffel and 

 Stanton over the whole range to which their experiments 

 extended when both the dimensions of the plate used and 



