Page 131 - Buried Pipe Design

106 Chapter Three

For the soil to be used as embedment, a series of simple laboratory
tests can be run to produce data similar to those shown in Fig. 3.20.
However, experience has shown that data given in Fig. 3.20 are rep-
resentative of most soils and can be used for design. Thus it is evident
that soil density is the most important parameter in limiting pipe
deflection.

Empirical method
Each of the methods discussed so far for determining load and deflec-
tion has a theoretical basis, and except for the prism load theory, all
require experimental investigation to determine the unknown con-
stants. In the past several years, techniques have evolved whereby a
model or prototype pipe is tested until failure occurs and the total per-
formance of the pipe is studied. Suppose a pipe is to be designed with
a certain earth cover in an embankment. Without a pipe in place, no
arching occurs, and the soil pressure at any height is easily calculated
(the prism theory load at that depth). When a pipe having good flexi-
bility is in place, the static pressure will not be greater than the prism
load pressure applied. Trying to calculate the actual pressure has frus-
trated researchers for years. If a pipe is installed in a prism load con-
dition (e.g., soil cell), the resulting deformation can be monitored
without the need to calculate the actual static pressure.
This procedure has been used with great success at various research
laboratories such as at Utah State University under the direction of
Reynold K. Watkins and at the U.S. Bureau of Reclamation under the
direction of Amster K. Howard. Data obtained in this manner can be
used directly in the design of soil-pipe systems and in the prediction of
overall performance. The possibility of buckling, overdeflection, and
wall crushing is evaluated simultaneously by actual tests. No attempt
to explain the soil-pipe interaction phenomenon is necessary in the use
of this method, and the end results leave nothing to be estimated on
the basis of judgment.
For example, if tests show that for a given soil compaction at 25 ft
(7.6 m) of cover a flexible pipe deflects 3 percent, and in every other
way performs well, the actual load on the pipe and the soil modulus
are academic. Thus, a pipe installation can be designed with a known
factor of safety provided that enough empirical test data are available.
In collection of these data, pipe was installed in a manner similar to
that used in actual practice, and the height of cover increased until
performance levels were exceeded. The procedure was repeated many
times, and a reliable empirical curve of pipe performance versus
height of fill was plotted. The use of these empirical curves or data
eliminates the need to determine the actual soil pressure since the