46 REPORT 1871. 



hitherto prevented the employment of the air-manometer, and at the same time 

 to be more accurate and unalterable in its working than the spi-ing gauges now 

 commonly used for steam-boilers. lu the first place, it is a great defect in the 

 common air-gauge that the divisions on the manometric tube diminish rapidly at 

 high pressm-es, and consequently the reading becomes less and less accurate the 

 higher the pressure. The new steam-gauge, on the contrary, possesses the same 

 degree of accuracy at all pressures, and even enables us to make the accuracy of 

 reading greater at higher pressures.^ 



Another serious defect of the air-manometer is the liability to rupture of the 

 narrow column of mercury when the steam is suddenly shut off or turned on. 

 This is entirely avoided in the present instrument by the use of two closed vessels 

 communicating with each other only by very narrow capillary tubes. Finally, the 

 small column of mercury enclosed iu the glass tube of common air-manometers is 

 subject to capillary depression, and to the disturbing eflects of heat upon the air- 

 bulb and upon the mercuiy. 



In the instrument now to be described it is sought to avoid these defects by not 

 using capillary tubes for the manometer, and by disposing the air and mercury in 

 such a way as to make the effect of heat insensible. 



The air-tube of the manometer consists of a series of tubes of equal length, but 

 different diameters, joined together by means of a blowpipe, and ending at the 

 top in a glass bulb. The lower end is connected by an air-tight screw, joined 

 with the first of two iron vessels containing each mercury or some other liquid, 

 and communicating only by a very narrow capillary tube or channel. 



The manometric tube is sealed at the bottom, but there are two fiue capillary 

 openings through the side at points below the sui'face of the mercury or other 

 liquid contained in the two iron vessels. Hence the communication of pressure 

 from the steam or other compressed gas, whose pressure is to be measured, and 

 which presses directly upon the surface of the liquid in the second iron vessel, can 

 only take place through a system of two capillary channels ; and the resistance 

 which these channels oppose to the motion of the mercury, by which they are 

 filled, makes it impossible for sudden changes to occur in the height of the mano- 

 metric column, and thus entirely prevents the di\'ision of the column or the entry 

 of steam or gas into the manometer. 



The capacities of the tubes and of the globe, which compose the manometric 

 tubes, are so adjusted that they decrease in the same ratio in which the pressure 

 increases, which is e^adently what is required by Mariotte's law in order that an 

 increase of pressure of one atmosphere may cause the first tube to be filled by the 

 enclosing liquid, and that a further increase of pressure of the same amount may 

 cause the second tube to be filled, and so on, each equal increment of pressure 

 causing the same rise of the liquid in the manometric tube. This adjustment of 

 the capacities is efi'ected as follows : — Let the capacity of a manometer, to be di- 

 vided so as to show pressures up to, say, four atmospheres, be called unity, and let 

 ^i) i^-i) '^si find Wj be the capacities of the first, second, and third tube and of the 

 terminal globe respectively, then we have — 



^i + ^2+*'3+'^4 = l for one atmosphere. 

 ^2+ ^"3 + "4 =2 foi' two atmospheres. 

 v.^-\-Vi = ^ for three „ 

 ^i=i for four „ 



This gives for the capacities of the tubes and their radii : — 



1 



1 .. 1 



1 .. 1 





"3 — tV- 



V, = t'7t= 



2.3 ' VCttA 



1 .. 1 



3.4' ' VI2^ 

 ]_ 



t',= 



1 



