46 
BULLETIN 1373, U. S. DEPARTMENT OF AGRICULTURE 
the dust-collecting and floor-sweep s} T stenis operate satisfactorily; 
before the valves were installed, the distant floor-sweep hoods did not 
have enough suction to pick up grain. The gates are placed in the 
branches as close as possible and parallel to the main pipe. The 
amount of suction and consequently the amount of air to be handled 
by the branch line is regulated by the opening of the slide, which fits 
tight into a groove and remains in any position it is set. By sealing 
or locking the slide in the desired position an inspector can control 
the suction and air velocity in a branch supplying a hood. 
Another method of controlling the velocity in the branches is by 
means of a butterfly valve (fig. 37) inserted in the branch, preferably 
as near as possible to the main pipe. This valve, designated the 
United States Department of Agriculture velocitv control valve, 
may be locked in the desired position. 
EFFECT ON GRAEV WEIGHTS 
Observations have been made in various elevators throughout the 
country to determine what effect the use of suction as applied by 
modern dust-collecting equipment has on grain weights. 
Fig. 37.— United States Department of Agriculture velocity control valve 
Tests were conducted by the Bureau of Chemistry in an export 
elevator equipped with a dust-collecting system that applied suction 
to the scale hoppers, garners, belt loaders, and boots. One of the 
scale hoppers was filled with No. 2 hard wheat and weighed. The 
grain was then dumped and spouted into a storage bin, from which 
it was later transferred to a shipping belt to be elevated for reweigh- 
ing. Four runs were made, two with suction and two without 
suction. The garner bin was swept clean before each weighing, and 
the first weight of each run was taken under the same conditions as 
the last weight of the preceding run. Four suction lines not con- 
nected to hoods acted directly on the grain: A 3-inch pipe connection 
at the belt loader, a 3-inch pipe at the elevator boot, a 6-inch line to 
the garner bin, and a 6-inch connection to the scale hopper. The 
results of these tests are shown in Table 7. 
The shrinkage losses varied whether suction was used or not. On 
the whole, the shrinkage loss was slightly greater when the fans 
