The Venturi Flume 
9 
The fact that ohstructions mig-ht occur in the flume is of con- 
siderable importance, and actual tests have been made tO' ascertain 
the effect of placing- such obstructions in the flume in both the con- 
verging* and diverging sections. The preliminary investigations 
made in 1915 on the V-notch flume brought out the fact that placing 
a 2-by 4-inch timber on edge, nailed to the floor of the flume at a 
point 3 feet above the throat section, did not increase the discharge 
by more than one per cent for any depth. Similar tests on a V-notch 
flume installed at the field laboratory at Bellvue, Colorado', showed 
the same result. In addition, it was found that placing the 2-by 4- 
inch timber on the downstream end of the flume had no apprecia- 
ble effect. 
Recent tests on a 3-foot rectangular flume show that 2-by 4- 
inch timbers nailed vertically to the sides of the converging sections, 
above the upper gage connections with' 2-inch projection, had no 
effect; also the 2-by 4-inch timber on edge on the floor of the flume 
at a distance one foot above this gage connection had no effect, but 
when the timber was placed one foot below, or downstream from the 
connection, the effect was to increase the upper head so as to indi- 
cate approximately 4 per cent increase in the discharge when flow- 
ing IOV 2 second feet. Small strips placed transversely on the floor, 
projecting one-half inch, both in the converging and diverging sec- 
tions, had no effect on the discharge. The deposit of sand and silt 
Upon the floor of the flume has been found to be of small conse- 
quence, due to the fact that, as the converging section decreases in 
area, the velocity increases and the tendency for settlement or 
deposit is eliminated. Large floating masses would no doubt lodge 
in the throat section if the rectangular flume were installed, but this 
trouble could be effectively met by the installation of the trapezoidal 
section which would pass the lodging mass by the increased or 
accumulated head above the structure. 
Requests have been made for information and data relative to 
Venturi flume^ having capacities for more than one thousand sec- 
ond feet, but facilities and equipment are not available at present 
for undertaking the calibration of flumes of more than 30 second 
feet capacity. The law of flow, as has been determined from our 
experimental data, is applicable to sizes up to and including the five 
foot throat, but only approximations can be made for larger sized 
flumes by the application of the known law of flow. It is thought 
that the present range of calibration of flumes will meet the ordi- 
nary demand, and special cases requiring exceptional capacity will 
necessitate individual rating or calibration. 
