44 BULLETIN 852, TJ. S. DEPARTMENT OE AGEICULTTJUE. 



capacity is greater than a value of n of 0.012 would indicate. Air 

 troubles in gauge No. 1 caused inconsistent results for this series, and 

 for this reason full weight should not be assigned to the tests. The 

 values of C s range from 0.380 to 0.434, and the corresponding values 

 of n from 0.0116 down to 0.0101. During the season of 1917 the 

 writer again visited this pipe while it was carrying far more water 

 than for any of the observations made in 1915, but the air troubles 

 were even greater than before, so experimentation was considered 

 out of the question. This pipe shows the need of "air chimneys" 

 near the intake if maximum efficiency is to be reached. 



No. 37, Experiment S-59. — 120-inch monolithic concrete pipe, 

 Whitney siphon, Los Angeles Aqueduct, California. 1 — The Whitney 

 siphon, on the Saugus division, is 955 feet long, 10 feet inside diameter, 

 with a shell 9 inches thick, and is subject to a head of about 75 feet. It 

 conveys water for municipal and irrigation purposes across a narrow 

 canyon, between two flow-line tunnels. The sides of the canyon have 

 a slope of about 3 feet horizontal to 1 foot vertical. 



The pipe was constructed in place, smooth wood forms being used 

 for both inside and outside surfaces. 



The interior of the pipe was treated with a finish coat of rich cement 

 mortar. The grit of the sand in this coat was very noticeable to 

 the touch at the time of experiments in 1916. 



A reach of the siphon pipe below the level of the floor of the flow- 

 line at both inlet and outlet was chosen for test. Holes were drilled 

 through the top of the shell and one-eighth-inch iron-pipe nipples, 

 each 18 inches long, were cemented into the shell. (The hole lor a 

 one-eighth-inch pipe is nearly one-fourth inch in diameter.) Two 

 piezometer tubes of type A, exactly alike in construction, were 

 thrust into the pipe through the nipples. The tapering end of these 

 tubes was made flexible by a small rubber-hose joint with the main 

 tube, so that the current might hold each piezometer in the direction 

 of flow by means of the flattened taper, like a vane. 



The flow-line channel is tapered into the round section of the 

 siphon pipe by means of transition sections at each end of the siphon. 

 Between the transition sections and the flow-line sections manholes 

 are placed at both inlet and outlet. 



The tap for gauge No. 1 piezometer was 59.2 feet downstream 

 from the manhole at the inlet, while the tap at gauge No. 2 was 61.1 

 feet upstream from the manhole at the outlet. The holes in each 

 piezometer tube were 3.6 feet downstream from the taps in the pipe. 

 This made the reach between piezometers 857.4 feet long. Water 

 columns were used at both gauges. 



The nominal size of the pipe was accepted in the computations. 

 The velocities were determined by timing the passage of solutions of 

 fluorescein, injected at gauge No. 1 and observed at the manhole 

 beyond gauge No. 2, correction being made for the fact that the 

 channel is not full for part of the distance between gauge No. 2 and 

 the manhole. 



Experience with this pipe showed that satisfactory tests can not 

 be made on a reach of pipe that is relatively short, compared with the 

 size of the pipe, especially for low velocities. The loss of head for 



1 Construction of the Los Angeles Aqueduct, Final Report, Los Angeles, Calif., 1916, pp. 210, 214. 

 Engin. Con. July 3, 1912, p. 20. 



