Mr. J. D. Fry on a New Micromanometer. 497' 



together. In the process of drying the tension considerably 

 increases, and when the membrane is thoroughly dry it 

 presents a uniform taut surface. The petrol-resin varnish 

 is then applied. 



The special attachment for the measurement of the pressure 

 is set up in the following way : — In front of one face of 

 the membrane a light mirror m is hung by a quartz bifilar 

 suspension, the points of support of which are not at the 

 same level. Thus one fibre is attached to a short style 

 projecting from the centre of the membrane, and the other- 

 fibre to a point above it which is rigidly fixed to the iron 

 ring. When a small difference of pressure is set up between 

 the two faces of the membrane, the style follows the motion 

 of the centre, and the mirror is rotated through a small angle. 

 The angle of rotation for a given pressure difference depends 

 on the horizontal distance S between the two points of 

 support of the mirror ; the manometer is therefore mounted 

 on a metal base fitted with levelling screws, and by means 

 of these S and consequently the sensitiveness of the gauge 

 may be varied over a wide range. The upper limit of 

 sensitiveness is reached when S is practically zero and 

 instability sets in. 



Disks of plate glass C are cemented to recesses which have 

 been turned one in each casting. In this way an air-tight 

 chamber is obtained in front of each face of the membrane. 

 Glass tubes d, cemented to holes through each ring, enable 

 excess pressure to be applied to either chamber. 



In testing for leak through the membrane a soap-bubble- 

 affords a very convenient means of detection. The manometer,, 

 being rendered sufficiently unsensitive, is connected on one- 

 side to the interior of a hemispherical soap-bubble blown on 

 a flat plate, the other side being left open to the atmosphere.. 

 This produces a deflexion on the instrument which, in the? 

 absence of leak, remains steady, but which increases if the 

 bubble grows smaller owing to passage of air through the- 

 membrane. 



In dealing with small pressure differences, it is found 

 necessary to make all the connexions with tubes of reason- 

 ably large diameter, since with tubes of small bore effects 

 due to viscosity become troublesome, and the manometer- 

 takes a considerable time to attain its final reading. 



Calibration of the Instrument. 



The instrument is obviously one for which the constant 

 cannot be found a priori. It has therefore to be calibrated, 

 Phil. Mag. S. 6. Vol. 25. No. 148. April 1913. 2 M 



