THE MEASURINC; OF AIR WITH SPECIAL 

 REFERENCE TO COMPRESSORS. 



Bv Carl janssen. 



The measuring of large quantities of air, although of great 

 importance to mining and industrial firms using compressed air 

 for various purposes, has not been brought to a satisfactory- 

 state of accuracy until lately. The reasons for this are twofold: 

 firstly the great difficulties met with as soon as com])arati\ely 

 larger quantities of air had to be measured, and secondly, the 

 feature that the reciprocating compressor could work as a dis- 

 placement air meter of, at first sight, considerable accuracy, so 

 that the necessity for creating new methods of measuring air 

 was not felt to such an extent as to induce imi)r()vcment. 



With recii)rocating comi)ressors il seemed sufticient to 

 count the re\olutions during a certain time and to find the 

 volumetric efficiency in order to determine the amount i:>f air 

 drawn into the machine — the volumetric efficiency l)eing defined 

 as the ratio of the length of suction line as sliown l)y the indi- 

 cator diagram and the total length of diagram. 



This method howexer does not give the actual weight of 

 air .which is in the low pressure cylinder at the 1)eginning of 

 the com])ression stroke. With high power reciprocating com- 

 pressors the jacket cooling has \ery little influence on the air- 

 drawn into the machine. Its main purpose is to preyent the oil 

 vin tlie cylinder liner from carbonizing. The temperature of 

 the cylinder walls during the suction stroke is higher than the 

 t.em])erature of the entering air, so that the air will 1)e heated 

 during the suction stroke and the tem])erature of the air inside 

 the cvlinder at the ])eginning of the compression strolsc will, 

 therefore, be considerably diff'erent from the tem])eraturL- meas- 

 ured during the suction stroke at the compressor inlet fianye 

 or in the suction chamber. The density of the air inside the 

 cvlinder being, on account of its higher temperature, lower than 

 the density of the air outside the. compressor, the weight of air 

 calculated from the tem])erature in the suction chamber, is 

 therefore, too high, i.e.. the weight of air discharged by the 

 compressor must be smaller than the weight found from tlie 

 inlet temperature and the volumes determined from indicator 

 cards. 



Assuming an atmospheric pressure of 12 lb. and an increase 

 i)f tem])erature from 75° F. to 8^° F., when the air is flowing 

 through the suction chaml^er and the port to the cylinder, we 

 find the weight of 1 cu. ft. of air to decrease from 0.060616. to 

 00595 lb. and the error resulting from this to be equal to i.8^f . 

 The wei'.iht of air actuallv drawn in will be further reduced 



