1896.] 



on New Researches on Liquid Air. 



141 



working it is better to use a needlo stopcock, worked from the outside 

 by a screw passing through the middle of the coil of pipe. 



In testing the individual coils as to the amount of air passed per 

 minute under different pressures, the arrangement of apparatus shown 

 in the Plate 7 was used. 



A is a bottle of compressed air, to which the copper pipe B is 

 attached. This coiled pipe first passes through the vessel C con- 

 taining water, in order to equalise the temperature, and then through 

 the cork D into the glass vacuum vessel E, when it is led by a large 

 number of convolutions to the bottom, terminating in a minute pin- 

 hole valve F. The released air passes from F right up through the 

 coils and out of the vent by the copper tube G, which in its turn 

 passes through a vessel H similar in its object to C, and is then 

 conducted to a measuring meter Z J. 



The following table gives the results of a series of experiments 

 made on one coil as to the rate of discharge of air at different 

 pressures : — 



Pressure in Atmo- 

 spheres. 



65 

 105 

 155 

 198 

 210 

 250 

 287 

 290 



Cubic Feet per Minute 



Measured under Atmosphere 



at 15°. 



22 

 0-42 

 0-63 

 0-79 

 0-84 

 1-00 

 115 

 1-18 



The results show that the rate of air discharge through a fine 

 aperture is directly proportionate to the pressure, or the velocity 

 with which the gas on the high-pressure side enters the orifice, is 

 independent of the density. Actual measurements of the size of the 

 needle-hole resulted in proving that the real velocity of the air 

 entering the aperture on the high-pressure side was about 500 feet 

 per second. In all these experiments the temperature of the coil was 

 not allowed to get so low as to produce any visible trace of conden- 

 sation in the air jet. Just before liquefaction the rate of dischart^e 

 of air through the same aperture may be doubled, the pressure re- 

 maining steady, owing to change in the viscosity of the gas and other 

 actions taking place at low temperatures. The above measurements 

 can only be regarded as representing the general working of such 

 regenerating coils. 



A double coil of pipe has advantages in the conduct of some 

 experiments. The efficiency is small, not exceeding the liquefaction 

 of 2 to 5 per cent, of the air passing, but it is a quick method of 



