798 APPENDIX III 



stock, and condition of piezometer orifices shall be such that no piezometer 

 shall vary in its readings by more than 10 per cent of the velocity head 

 from the average of all the piezometers. The piezometer orifices shall be 

 flush with the inside surface of the penstock wall, the passages shall be normal 

 to the wall, and the wall shall be smooth and parallel with the flow in the 

 vicinity of the orifices. The orifices shall be | inch in diameter. 



The velocity at each point in the penstock shall be computed by the 

 formula V = \/2gh, in which h represents the difference in feet between the 

 total dynamic pressure recorded by the Pitot tube at that point and the 

 average static pressure recorded by the piezometers. The velocities so 

 determined shall be plotted as ordinates against values of the areas of the 

 sections of the penstock corresponding to the points of measurement as 

 abscissas, a smooth curve being drawn through the points obtained. The 

 mean velocity in the penstock will then be taken as the mean ordinate of 

 the above curve multiplied by 0.976. This coefficient is based on the average 

 of various comparative tests, and is required to correct for oblique or sinuous 

 flow under the usual conditions in straight penstocks. 



When the length of straight run of penstock is insufficient or when the 

 flow is disturbed by a severe bend or obstruction upstream from the tube 

 or when the average velocity is less than 5 feet per second, the above coef- 

 ficient will not apply correctly, the correct value being considerably lower 

 in such cases, which do not, therefore, come within the scope of this code. 

 The coefficient corresponds to a tube, the point of which is f inch in diam- 

 eter with a i inch hole, the face being normal to the axis, and at least 3 inches 

 from the nearest surface of the supporting pipe. 



(d) By the Screen or Diaphragm Method. When the screen method is 

 used a sufficient length of straight flume of uniform cross-section shall be 

 constructed with a close-fitting screen filling the cross-section. Provision 

 shall be made for accurately observing the velocity of the screen, preferably 

 by electric contacts and chronograph. The length of run of the screen shall 

 be sufficiently in excess of the portion used for measurement to provide 

 ample space for starting and stopping the screen, so as to insure uniform 

 conditions over the measured portion of the run. In determining the dis- 

 charge the velocity of the screen shall be multiplied by an area intermediate 

 between the net immersed area of the moving screen and the average area 

 of stream cross-section of the portion of the channel traversed. The varia- 

 tion of the level in the flume shall be observed during the course of the run 

 and the average elevation shall be used in determining the area. 



(e) By Titration or Chemical Method. When the chemical method is 

 used in measuring discharge, care shall be taken to insure that at the point 

 of introducing the dosing solution no portion of the solution shall be carried 

 off by back currents and shall therefore fail to pass to the sampling station, 

 and that the sampling station shall be so placed that no pollution shall be 

 caused by reverse currents, causing fresh water to pass the station from down- 

 stream. When necessary, owing to a short length of mixing passage or lack 

 of sufficient disturbance to cause thorough mixing, the dosing pipes shall 

 be so placed that an equal degree of concentration over the entire section 

 of the sampling station shall be obtained. Samples shall be taken from 



