THE FLOW OF WATER IN CONCRETE PIPE. 87 
for velocity of approach. The depth of water upstream from the weir 
crest was 3.40 feet and the end contractions 1.50 feet and 1.55 feet, 
respectively. For all discharges less than 17.0 second-feet, the 
bottom and end contractions were in excess of two times the head 
on the weir. For greater discharges than the above, the end con- 
tractions are not sufficient to give standard conditions. The mean 
velocity for any particular run of water was taken as the discharge 
divided by the mean of the areas of the water sections at the two 
ends of the reach. The slope was taken as the constructed slope of 
the pipe line, a fall of 1 foot per 1,000 feet of pipe. 1 
The reach tested consisted of about equal amounts of tangent 
and curve, as is typical of this line, but did not include any of the 
siphon pipes. There is so much curvature on this line (50 per cent) 
and the individual curves are as a rule so sharp that the values of the 
coefficients of retardation may be taken as for a very smooth pipe with 
an excess of curvature. In other words, the average pipe with the 
same type of construction would probably show even more favorable 
carrying qualities. 
No. 55, Experiment Ekle-2. — 42-inch reinforced concrete flow line, 
Victoria Aqueduct, Vancouver Island, British Columbia, Canada. — 
With the following modifications the same discussion applies to this 
reach as to No. 54 above: 
After flowing for 27 miles through a flow-line pipe, except for six 
inverted siphons, laid on a uniform grade of 1 foot fall per 1,000 
feet of pipe, the water wells up through 110 holes 4 inches square 
from an outlet chamber just upstream from an 8.04-foot weir similar 
in construction to the one at the intake, except that the depth from 
the perforated floor to the crest of the weir is 1.88 feet and the end 
contractions are 5.5 feet each. For all observations the end con- 
tractions were more than twice the head on the weir; but the bottom 
contraction did not conform to standard conditions for a discharge 
greater than 17 second-feet. 
This series of observations was made on a reach of pipe about 800 
feet long, just upstream from the weir described above where the 
discharge was measured. 
No. 58, Experiment JBL-1. — Tunnel No. 15, San Gabriel plant, 
Pacific Light & Power Co., California. — Just before starting the con- 
struction of the Los Angeles Aqueduct a series of experiments was 
made on both open an covered concrete channels, located in southern 
California. The measurements, made by Charles H. Lee and D. L. 
Reaburn, were reported by J. B. Lippincott 2 under whose direction 
the experiments were conducted. (See p. 11 for Mr. Lippincott 's 
conclusions.) Of this tunnel he writes: 
The section is rectangular, 4| feet wide and 4 feet deep, with a semicircular arch, 
and finished with a 1 to 3 cement-mortar plaster. 
There was no vegetable growth in the tunnel which could be felt or seen. The 
tunnel has been in use for eight years. There is a slight curve at the upper portal. 
Discharge measurement was made on tangent above the curve. Twenty feet beyond 
lower portal there was a sharp angle in alignment. 
1 In the opinion of the writer the slope should have been taken as the fall of the surface of the water in 
the pipe, corrected for changes in the velocity head due to changes in the areas of the water sections at the 
two ends of the reach. However, his own experience on this same pipe indicates that practically the con- 
structed value for the slope is obtained when determined by the surface fall with the above corrections. 
(See column 13, Table 11, opposite pipe No. 56.) 
2 Observations to determine the value os Cand n as used in the Kutter formula by J. B. Lippincott, 
Engin. News, June G, 1907, vol. 57, p. 612. 
