4/10/2019 CE 4995 Honors

Project: Age in
Water Distribution
Systems

Harry Vaslo
FT9873
 1

Executive Summary
Harry Vaslo of Senior Design Group 3 was to research ways in which his water distribution system for a
hotel will change after 50 years, and provide possible solutions to whatever problems he finds may
arise. His research pointed to the likelihood that his water distribution system’s pipes will have a change
in their roughness coefficients and have difficulty providing chlorine residual. Harry applied his
discoveries to the EPANET model of his site and discovered that the pipe material he was required to
use, ductile iron, is effective at maintaining its original roughness factor and actually provides a better
chlorine residual after decades of use. While there are some pipe materials that are even better at
providing a consistent chlorine concentration, Harry determines that changing his model is unnecessary.
 2

Contents
Executive Summary....................................................................................................................................... 1
Introduction .................................................................................................................................................. 3
Results of Research ....................................................................................................................................... 3
Research Conclusion and Application of Research to Water System ........................................................... 6
Current Water System Design....................................................................................................................... 7
Initial Chlorine Concentrations ..................................................................................................................... 8
Chlorine Concentrations after 50-years...................................................................................................... 11
Analysis of Results & Possible Design Changes........................................................................................... 13
References .................................................................................................................................................. 14
 3

Introduction
Harry Vaslo was responsible for designing a water supply system for a hotel design in his Senior Design
course. For his project, there were several design parameters he had to follow, such as pipe material
and depth, as well as any regulations put in place by the City of Detroit and State of Michigan. However,
it is of Harry’s opinion that the rules dictating his design may not have been the most appropriate for
increasing the longevity of the system. In this report, Harry details the ways that age effects a water
supply system. He then applies his research to the EPANET model of his water system to simulate what
his system will look like after 50 years. He concludes with the results of his findings and design choices
he could make now to prevent the negative effects of age.

Results of Research
As a water distribution system ages, and the wall materials of the system’s pipes corrode, one can
expect changes in pipe roughness and chlorine decay rates in the system. As water is run through a
distribution system, it loses some of its initial chlorine concentration as a result of the chlorine reacting
with the pipe material and biofilm forming on pipe walls (Al-Jaseer, 2007). Both pipe material and pipe
diameter are factors in the chlorine decay rates of a pipe, as evidenced in Figures 1 through 4 below:

Figure 1. First-order chlorine wall decay constants

as a function of pipe age for a variety of pipe
diameters. Pipe material is cast-iron (Al-Jaseer,
2007).
 4

Figure 2. First-order chlorine wall decay constants

as a function of pipe age for a variety of pipe
diameters. Pipe material is steel (Al-Jaseer, 2007).

Figure 3. First-order chlorine wall decay constants

as a function of pipe age for a variety of pipe
diameters. Pipe material is cement-lined ductile
iron (Al-Jaseer, 2007).
 5

Figure 4. First-order chlorine wall decay constants

as a function of pipe age for a variety of pipe
diameters. Pipe material is PVC (Al-Jaseer, 2007).

The chlorine decay rates of a pipe change drastically depending on the pipe material and diameter, but
this change is not necessarily always negative. Over a 55-year period, depending on the pipe material,
wall decay constants could decrease as much as 92% or increase as much as 431% (Al-Jaseer, 2017).
Therefore, the diameters and materials of pipes chosen at the beginning of the construction of a new
water distribution system could decide whether the system will still be able to provide clean drinking
water 50 years later.

To simulate age in a pipe, the Hazen-Williams roughness coefficient is adjusted. The Hazen-Williams
roughness coefficient is reduced to imitate the degradation of pipe walls (Ormsbee and Wood, 1986).
The roughness coefficient of a pipe determines the friction losses that water experiences when running
through it; for example, its flow and velocity (Davis, 2010). Predicting a new Hazen-Williams value is
difficult, as it is a factor of pipe material, condition, and age (Liou, 1998). The change in Hazen-Williams
coefficient is also not a linear; the Hazen-Williams coefficient of a pipe will decrease quickly over the
first 20-years of age, but the decrease is slower and less pronounced as the pipe age approaches 50
(Sharp and Walski, 1988). This relationship is illustrated in the figure below:
 6

Figure 5. Plot of change in Hazen-Williams

coefficient with respect to age and pipe diameter.
Plot created off of tests using cast-iron pipe(Sharp
and Walski, 1988).

Research Conclusion and Application of Research to Water System

Harry was required to use ductile iron pipe as the pipe material for his system. As evidenced in Figure 3,
ductile iron distinguishes itself from materials such as cast-iron and steel by having a high wall decay
factor when new, but dropping significantly after years of usage. In addition, after more research was
conducted to determine how much the Hazen-Williams factor of ductile-iron pipe drops after a
prolonged period of time, Harry found that the Hazen-Williams coefficient of roughness only decreases
2.5 units per decade, according to the Detroit Water Master Plan (PVCPIPE Association, 2017).
Therefore, after 50 years the roughness coefficients of the system’s pipes will only drop by 12.5. These
two discoveries point to Harry’s hypothesis of ductile iron not being the most suitable pipe material for
increasing the longevity of his hotel’s water supply system to be wrong, in terms of concerns about the
system being able to effectively provide chlorination to the hotel’s guests. However, any conclusion
cannot be made until the conditions the hotel’s water supply system will be in after 50 years are
simulated in EPANET.
 7

Current Water System Design

A more comprehensive description and a sketch of Harry’s system can be found in the final submission
for Senior Design Group 3. The hotel water main has two connections to the water line running
underneath East Jefferson Ave. The lines leading from these connections meet and form a looped
system. There are two lines leading into the hotel: one for normal water usage, and one for fire-fighting.
All pipes are 8 inches in diameter, except for the pipes representing the water main under East Jefferson
Ave. There are three fire hydrants on site. It was required that the pipe material be ductile iron.

The water supply system was modeled in EPANET to determine its effectiveness at supplying water to
the hotel. Connections to the water main under East Jefferson Avenue were simulated by having two
junctions connected to the same reservoir through an extremely small length of pipe (0.00001 feet).
Harry was to assume that the surrounding system could provide flow and pressure required for
domestic and fire-fighting usage. The scope of the project involved seeing that pressures between 30
and 80 psi, and ideally 40 psi, were met at junctions representing the building and hydrants at both
normal use and fire-flow. The project did not require testing for chlorine concentrations.

Figure 6. EPANET model with site background

 8

Figure 7. EPANET model without site background

Initial Chlorine Concentrations

The model needs to have a chlorine concentration, which is accomplished by doing the following:

Figure 8. In “Quality
Options” change
Parameter from “None”
to “Chemical.”
 9

Figure 9. Give each

junction an Initial Quality
of 0.6

Figure 10. Go to
“Reactions Options” and
change the “Global Bulk
Coefficient” to -30. This
value comes from Figure
3; as most of the pipes in
Harry’s system are 8-inch
diameter (200 mm), a
decay rate of 30/d can be
estimated.
 10

Figure 10. Give reservoir

an Initial Quality of 0.6

With these values saved to the model, the model is then run for six hours and data from the junctions
and pipes is collected:

Figure 11. The state of each junction after 6-hours

 11

Figure 12. The state of each pipe

after 6-hours

Chlorine Concentrations after 50-years

To simulate how the system will look after 50-years, the following parameters must be changed:

Figure 13. Roughness coefficients

in each pipe must be reduced
from 140 to 127.5, a reduction of
12.5
 12

Figure 14. Using Figure 3, change the

Global Bulk Coefficient from -30 to -5

Run the model again and chart the chlorine concentrations in the pipes and junctions:

Figure 15. The state of each junction after 6-hours

 13

Figure 13. The state of each pipe

after 6-hours

Analysis of Results & Possible Design Changes

The chlorine concentrations in junctions are higher and the reaction rates in the pipes are lower in the
50-year model than in the new model. There is also no noticeable change in flow or velocity between
either model. Therefore, we can conclude that chlorine residual will actually be better 50 years from
now than right after construction. This raises questions as to what can be changed in Harry’s design to
improve its longevity. The purpose of this report was to address concerns that Harry’s water system
might not be effective at providing water to his hotel’s guests after a prolonged period of time. He found
that water systems primarily struggle with providing properly chlorinated water after years of use.
However, due to the unique properties of his pipe material, ductile iron, chlorine concentrations are
actually improved after years of use, and the change in roughness of the pipe has done virtually nothing
to change flow in the pipes. According to Figure 4, PVC pipe has an even lower chlorine decay rate than
ductile iron after prolonged use. Replacing Harry’s ductile iron pipe with PVC pipe is the only real
improvement to the model that can be deduced from his research, and even then, depending on cost
and regualtions, even this solution might be unnecessary.
 14

References
Al-Jasser, A. O. (2007). “Chlorine decay in drinking-water transmission and distribution systems: Pipe
service age effect.” ScienceDirect, Water Research, <https://doi.org/10.1016/j.watres.2006.08.032>

Davis, M. L. (2010). Water and Wastewater Engineering. McGraw Hill.

Liou, C. P. (1998). “Limitations and Proper Use of the Hazen-Williams Equation.” ASCE,
<https://ascelibrary.org/doi/pdf/10.1061/(ASCE)0733-9429(1998)124:9(951)>

Ormsbee, L. E., & Wood, D. J. (1986). “Explicit pipe network calibration.” Journal of Water Resources
Planning and Management, 112(2), ASCE, 166-182.
<https://ascelibrary.org/doi/pdf/10.1061/(ASCE)0733-9496(1986)112%3A2(166)>

“Ductile Iron Pipe's Hazen-Williams Flow Coefficient Declines Over Time.” (2017). PVC Pipe Association
Technical Brief. <https://www.uni-bell.org/Portals/0/ResourceFile/ductile-iron-pipe%E2%80%99s-hazen-
williams-flow-coefficient-declines-over-time.pdf>

Sharp, W. W., and Walski, T. M. (1988). “Predicting Internal Roughness in Water Mains.” JSTOR,
American Water Works Association, <www.jstor.org/stable/41292059>