SECT. II.] 



PROPERTIES OF STEAM. 



101 



FIG. 14. 



169. When C A is the height of a uniform 

 atmosphere, C D the height of the chimney, and 

 D E the quantity the air expands by the heat it 

 receives in passing through the fire, the height E D, 

 or F G its equal, represents the height of the co- 

 lumn of air which produces the motion, and the 

 velocity will be that a heavy body would acquire 

 by falling through the height F G. If the whole 

 of C A were empty, then B H, the height of the 

 atmosphere, would be the height through which the 

 body must fall to acquire the velocity with which 

 the air would move into the tube, provided it suf- 

 fered no contraction at the entrance ; but such a 

 contraction is well known to take place in air as 

 well as in water. 



170. When this is applied to a chimney, the smoke being sometimes of a 

 density different from common air at the same pressure and temperature, the same 

 excess of temperature will produce a greater or less effect in proportion as it is of 

 less or greater density than common air. This will be found by subtracting from 

 the expansion the specific gravity of the smoke or vapour, that of air of the same 

 temperature and pressure being unity. Or it may be done by an allowance of a 

 portion of the temperature for the difference of density : either method gives the 

 same result when properly calculated. In this case I intend to adopt the former 

 method. The latter is followed in my book ' On Warming and Ventilating 

 Buildings.' 1 



171. Let h be the height in feet from the place where the flue enters to the 

 top of the chimney ; e = the bulk to which one foot of air increases by the change 



1 The principles of calculation followed, both in this and in the work referred to, are perfectly 

 identical with those employed by Mr. Gilbert, in an excellent paper on the subject in the 

 ' Quarterly Journal of Science,' vol. xiii. p. 113 : but the notation and methods of managing the 

 processes are different ; and Mr. Gilbert's mode of calculating the expansion does not afford quite 

 satisfactory results : besides, he makes no allowances for the contractions and loss of force in 

 curvilinear motion. I mention the circumstance, because some people compare and criticise, and 

 imagine those things to be different which are in reality identical, as may easily be shown by 

 putting both in the same notation, and reducing by the rules of algebra. The great object of a 

 practical analyst is to render the final equation as easy of application as possible. As to those who 

 question principles, it is rather unfortunate for them to question those established principles of 

 pneumatics which are confirmed by experiment. It is only when theory and experiment do not 

 agree, that the principles can be called in question. 



