508 EXPERIMENT STATION RECORD. 



in Madras to irrigate 10 acres of land planted to almost any crop when the water has 

 to be lifted 25 ft." 



The author quotes the results of Perry's experiments on windmills, carried out in 

 this country, and says: 



"These results can be very approximately expressed in the following way: The 

 maximum work which can be done by a windmill in a 10-niile breeze is equal to one 

 foot-pound per second per square foot of wind surface. A 16-ft, wind wheel has an 

 area of 201 sq. ft., and should therefore supply in a 10-mile wind 201 foot-pounds 

 of work per second, equivalent to 0.373 horsepower. From the equation y = 188 x, 

 it is found that the actual work represented by the water lifted is 0. 198 horsepower, 

 or 53 per cent of the maximum work that can be obtained from the wind. Allow- 

 ing for the friction of the gearing and the loss of energy in the pump, the efficiency 

 is satisfactory and the load on the pump is evidently suitable. 



" A careful consideration of these results leads to several very important conclu- 

 sions, which are evidently confirmed by the experiments on this mill. Although 

 the maximum work which can be done by a windmill is proportional to the cube of 

 the wind velocity, the actual amount of work which it does when pumping water is 

 proportional to the wind velocity, because the work done depends upon the number 

 of strokes of the pump, and the number of strokes of the pump depends upon the 

 velocity of rotation of the windmill, and this, we have seen, is approximately pro- 

 portional to the wind velocity. This result is fully borne out by these experiments, 

 since the work done is found to be exactly proportional to the wind velocity. 



' ' The practical meaning of this is that a windmill works with the greatest efficiency 

 when the velocity of the wind is just sufficient to keep it in steady motion, and that 

 at any other higher velocity of the wind but a portion of the useful work which the 

 windmill could do is utilized. If a convenient arrangement could be devised whereby 

 the load put upon the windmill varied with the square of the wind velocity the work 

 done by these machines would be much greater. An attempt to realize this is made 

 by the manufacturers who provide means whereby three different lengths of stroke 

 of pump can be obtained; but in practice it is found inconvenient to alter the length 

 of the stroke. The most satisfactory way to vary the load on a windmill is to pro- 

 vide it with two pumps worked through a rocking lever fixed at ground level. One 

 pump can be permanently attached to the rocking lever and the other whenever the 

 wind velocity is sufficient to justify doing so." 



By examining all the records of hourly wind movements the author determined 

 the total number of hours during the year when the wind was strong enough to 

 have driven a second pump in addition to the first, and also the number of hours 

 when a third pump could also have been driven. He found that, neglecting any 

 work done by the wind when the average velocity was below 8 miles per hour, the 

 total quantity of water raised by the windmill during the year would have been 

 increased by 52.4 per cent if another pump could have been attached when the wind 

 was favorable, and the addition of a third pump when the wind was sufficiently 

 strong would have increased the work done by a further 24.3 per cent. 



The author also examined the results that would have been obtained by the use of 

 a 10-in. pump as compared with the 8-in. The 10-in. pump would have required a 

 wind velocity of over 9 miles per hour before it could operate, but in the stronger 

 winds it would have had so much greater capacity than the 8-in. that the total 

 amount of water raised would have been practically the same for each month 

 throughout the year as with the 8-in. pump. "All windmills arranged for lifting 

 water should be arranged so as to be able to drive two pumps during periods when 

 the wind velocity is sufficiently high." 





