MEASUREMENT BY WEIRS 169 



was 12 feet with a head of 1*60 feet, screens being used above the weir 

 to equalise the velocity. Where a length approximating to 30 H is 

 impossible, one or more perforated diaphragms should be placed across 

 the stream so as to steady the motion as far as possible. 



6. Accurate determination of the head H. To measure H, water 

 should if possible be run off by an auxiliary channel until exactly level 

 with the notch sill. This level may be determined with great accuracy 

 by observing the reflection at the surface in the immediate neighbourhood 

 of the sill, since the absence of any curvature of the surface at this point, 

 indicated as it is by non-distortion of reflected objects, shows that the 

 correct level has been obtained. This level may then be read off, either 

 for rough work on a graduated staff fixed vertically in the bed of the 

 stream some 6 or 7 feet above the weir, or on a hook gauge (Fig. 153), the 

 point of which is adjusted until exactly in the surface. 



A preferable method consists in driving a stake into the bed of the 

 stream above the weir until its upper end is below the level of the sill. 

 This carries a short vertical wire, which may be filed down, until, as 

 shown by straight edge and level, its point is exactly level with the sill. 

 The water level may then be adjusted with great accuracy until this 

 point is exactly in the surface when this level may be read off on the 

 hook gauge also adjusted until its point is in the surface. 



When the weir is discharging steadily, the head can be determined, 

 either by direct reading of the graduated staff or by taking the reading 

 on the hook gauge when its point is again adjusted so as to be in the 

 surface Care should be taken when using the graduated staff that 

 allowance is made for the increase in height over the up-stream face of the 

 staff, and the decrease over the down-stream face, due to the piling up of the 

 water which occurs at a solid obstacle. For accurate work, measurements 

 of the head taken in a flowing stream are inadmissible. Quite apart from 

 the disturbance produced by any immersed object in such a stream, the 

 nature of its flow \vhich is seldom, if ever, perfectly steady, and the action 

 of the wind, produce oscillations of its surface which seriously affect the 

 possibilities of even approximately accurate results. To reduce the effect 

 of such oscillations and to avoid the disturbance caused by the presence 

 of the gauge, observations should be taken in a pit from 18 inches to 2 feet 

 square, in communication with the main stream through a pipe one or two 

 inches in diameter, opening out flush with the bed of the approach 

 channel, and perpendicular to the direction of flow. This was the method 

 adopted by Bazin, the pit being situated about 16'5 feet above the weir. 



By this means, using a hook gauge fitted with adjusting screw and 



