Mar. 6 ,1916 
Flow through Weir Notches 
1057 
duration of the experiments. The error in time in operating the shear 
gate and the stop watch is only a small fraction of a second. 
The calibrated tanks cover an area 55 feet square, divided by 12-inch 
vertical-sided concrete walls into one tank 27 by 55 feet, two tanks each 
2 2 >H by 27 feet, and a channel 6 by 27 feet, which is connected with each 
tank by a 14-inch circular orifice placed on the floor line and controlled 
by a gate. The tanks are 8 % feet deep. Their floors are all at the 
same elevation, and they have a combined capacity of more than 22,000 
cubic feet available for experimental purposes. The tanks have been 
carefully calibrated, corrections having been made for all irregularities, 
gate openings, rods, etc., and tables have been prepared giving the 
capacity at each 0.001 foot in elevation. A brass rod 1 inch in diameter 
and 9 feet long was placed in a vertical position near one comer in each 
calibrated tank, being held out from the wall about 6 inches by iron 
brackets set in the concrete (fig. 3). Holes drilled in these rods at 
carefully measured intervals of about 18 inches serve as datum points 
when the quantity of water in the tanks is being measured. The eleva¬ 
tion of the water in the tanks is determined to 0.001 foot by means of a 
hook gauge having fixed to its back a heavy clamp provided with a pin 
which fits snugly into the holes in the rod. A steel ladder was placed 
adjacent to the brass standard rods in each tank and anchored to the 
concrete. The platform shown in figure 3 is 20 by 24 inches and can be 
lowered close to the water surface and secured to any of the ladders by 
means of hooks. The funnel-shaped arrangement attached to the plat¬ 
form has a ^-inch hole in the bottom and can be adjusted so as to form a 
stilling basin for the hook gauge. With the water levels at the beginning 
and the end of the experiment determined by means of the standard 
rod and hook gauge, the volume run during the experiment can be 
determined readily from the calibration tables. 
Unless otherwise stated, the experiments recorded in this publication 
were made with notches the edges of which were one-sixteenth inch or 
less in thickness. The notch plates used were constructed either entirely 
of brass or of steel with brass notch edges. The crests and sides of the 
notches were dressed to true angles and straight lines, and by means of a 
micrometer caliper were calibrated to an allowable divergence of 0.002 
inch from a straight line. The triangular notches were dressed to tem¬ 
plates. The plate containing the notch under observation was placed 
in a vertical posi^on in the T frame in the bulkhead of the weir box, 
and the crests of rectangular and Cipolletti notches were leveled to 
within 0.001 foot by means of a 12-inch steel-frame level, upon which a 
bubble division indicated a variation of 0.0004 foot for a length of 1 foot. 
The inner face of the bulkhead was flush with the crest of the notch. 
The triangular notch plates were placed so .that a vertical line would 
bisect the angle formed by the sides of the notch. In all the experi- 
