796 APPENDIX III 



understood to be efficiencies computed by the use of the formula and coef- 

 ficients here given. 



where Q = quantity in cubic feet per second; 

 L = length of weir in feet; 

 h = observed head above crest in feet. 



P is the height of the crest above the bottom of the channel of approach ' 

 in feet. 



To facilitate computations, all corrections for velocity of approach have 

 been included within the coefficients as given; these are therefore to be 

 used in the formula stated above, the observed head being used without 

 modification. 



Note: The above coefficients are the averages of values computed by 

 the following three formulas: 



(1) Bazin, 



(2) Rehbock, 



~ -- 



3 



(3) Fteley-Stearns, 



Q =3.3lL(A+1.5/i,) 3 / 2 +0.007L, 

 in which Jit, = head due to velocity of approach. 



The weir shall be sharp crested, with smooth, vertical crest wall, complete 

 crest contraction, and free overfall. Complete aeration of the nappe shall 

 be secured and observation of the crest conditions and form of nappe shall 

 be made during the test to avoid defective conditions such as adhering 

 nappe, disturbed or turbulent flow, or surging. The sidewalls of the channel 

 shall be smooth and parallel and shall extend downstream beyond the over- 

 fall above the level of the crest. 



Weirs of a length exceeding approximately twenty times the head (ex- 

 cepting in cases where the velocity of approach is extremely low); or weirs 

 of moderate crest length having high velocities of approach; or those in 

 which the velocity of approach is irregularly distributed, or in which the 

 leading channel is subject to action of the wind, should either be subdivided 

 into a number of sections or the head should be observed not only at both 

 sides but also at intermediate points across the channel of approach. The 

 elevation of the crest should be measured at short intervals of its length 

 in determining the zero readings of the hook gauges. 



(6) By Current Meter. When the discharge is measured by current 

 meter, observations shall be taken by two different types of meter, one type 

 having preferably such characteristics that it will slightly over-register 

 under conditions of turbulent or oblique flow, and the other type having 



