APPENDIX III 797 



characteristics such that it will under-register under similar conditions. The 

 true velocity obtained by reducing the meter readings on the basis of their 

 still -water ratings may then be taken as a weighted mean between the two 

 series of observations. 



As a basis for arriving at the proper weighting of diverging meter results, 

 the instruments in question should, in addition to their regular still-water 

 ratings, be given simultaneous oscillation or angularity tests at several 

 velocities near those which will probably be experienced during tests. By 

 means of the resulting data, curves showing the over- and under-registering 

 characteristics of each meter may be plotted for varying degrees of obliquity 

 or velocities of oscillation. The total deviation of the two meters may then 

 be noted for any obliquity or lateral velocity. When the relative deviation 

 of the two meters is observed in the field, the curves will then indicate the 

 proportions in which the total deviation should be divided to give the proper 

 correction for each meter. 



The point method of observation shall be used and sufficient points shall 

 be obtained to enable both vertical and horizontal velocity curves to be 

 plotted for all portions of the section of measurement. The average velocity 

 shall be determined from these curves by planimeter. 



The section of measurement shall be rectangular and smooth flow con- 

 ditions shall be obtained. It is recommended that in order to avoid abnor- 

 mally long durations of run a number of meters of each type be used simul- 

 taneously. The elevation of water shall be continuously observed during 

 the current meter measurement by stilling boxes, piezometers, or other 

 reliable means. If the supporting rods for the meters are in the same plane 

 as the meters, the area of these rods shall be subtracted from the wetted 

 area of the flume in calculating the quantity. The meter should preferably 

 be supported by rods placed a sufficient distance behind them to avoid any 

 obstructive effect. When a heavy mast or supporting frame is used, it 

 should be designed to offer a minimum disturbance, and should be located 

 several feet downstream from the meters. 



(c) By Pitot Tube. When the Pitot tube method is used, the Pitot 

 tube shall be located in a straight run of penstock or conduit, at a distance 

 equal to at least ten pipe diameters from any upstream bend and at least 

 five diameters from a downstream bend. When the observation is made 

 in a circular pipe or penstock, at least two Pitot tubes shall be arranged 

 to traverse two relatively perpendicular diameters, but in the case of very 

 large penstocks or those having unsymmetrical flow, Pitot tubes shall be 

 arranged to traverse completely or partially the intermediate diameters, 

 giving traverses at forty-five degree intervals. 



In determining the velocity in the penstock by the Pitot tubes the static 

 pressure over the cross-section shall be measured by from four to eight 

 carefully constructed piezometers equally spaced around the wall of the pen- 

 stock at a section 1 foot in advance of the Pitot tube section to avoid the 

 effect of the Pitot tube supporting structure, the penstock being of uniform 

 cross-section between the piezometers and the points of the Pitot tubes. 

 All piezometers shall be connected to separate gauges. The conditions of 

 measurement, including velocity distribution, length of straight run of pen- 



