1 | Colorado Replenishment Pipeline
Colorado Replenishment Pipeline
Design Proposal
ME 555 – Thermal Systems Design
College of Engineering
San Diego State University
Prepared for
Dr. AsfawBeyene
Prepared by
Jonathan Butbul, Eric Miller, Nam Ngo
November 1, 2010
Contents
1.Scope
2.Objective
3.Major Assumptions
4.Background Data
5.Design Concept
5.1.Pipeline Route
5.2.Piping System
5.3.Pumping System
5.4.Water Treatment System
6.Cost Estimates
7.Other Concerns
8.References
Abstract
A system is proposed that will pump water from the Pacific Ocean into the Colorado River as it crosses the US border into Mexico. This system uses over 216 pumps and 172 miles of piping to increase the flow rate at the pipeline’s exit by 15,189 cubic feet per second. Finally, the cost of the system is estimated to evaluate the economic feasibility.
1. Scope
The Colorado River spans five states, running approximately 1,450 miles from Colorado to northwest Mexico. It is an essential source of water for the majority of Southwest America, as well as a large provider of clean and renewable energy. In recent decades, the river discharge has been modified various dams and diversion for irrigation. Construction of these dams has dramatically altered the surrounding geography; several lakes and reservoirs have been created along the river’s path. Due to an international agreement with Mexico, the United Statesis required to maintain the Colorado River at a minimum flow rate of 2,139 cubic feet per second upon crossing into Mexico[3].
1.Objective
The objective of this project is to design a system that will pump water from the Pacific Ocean to the Colorado River, in order to restore the flow rate near the U.S-Mexico border to it average flow rate upstream near the Hoover dam. The water must be desalinized and treated, so that its cleanliness is roughly equal to that of the Colorado River.
2.Major Assumptions
In order to simplify the analysis and design of this system, several major assumptions will be made.
Assumption 1.Pipeline Route
The pipeline route will follow the shortest path along a major highway from the Pacific Ocean near San Diego to the Colorado River near Yuma, Arizona. The elevation profile will consist of an initial horizontal section, a climb to the maximum elevation, another horizontal section, a descent to the final elevation, and a final horizontal section.
Assumption 2.Installation and Maintenance
The proposed design will not describe installation and maintenance procedures. The only exception will be the decision to route the pipeline along a major highway for ease of access in case of necessary maintenance.
Assumption 3.Pipes and Joints
The proposed design will not address issues related to thermal expansion of the pipes or joints. Minor head losses (e.g., those due to pipe joints, connections) will be neglected. Major head loss calculations will only consider pipeline segments S2 and S3, and will assume that their combined length is the entire distance of the pipeline route, multiplied by the number of pipelines in parallel.
Assumption 4.Pump Operating Conditions
Because limited data is available on the type of pumps required for this project, pump operating conditions will be assumed for each pumping station. Details about assumed operating conditions can be found in Section 5.2.
Assumption 5. Water Treatment Facilities
The proposed water treatment design will assume similar operating conditions and facilities to that of an existing desalinization plant in current operation. Startup and variable costs will be scaled up proportionally to the flow rates of the existing and proposed plants. The water treatment process will only describe the removal of salt, and the cost and operating conditions of the plant will be assumed to clean the water to the same quality as the model existing plant.
3.Background Data
3.1.Geological Data
The dam directly preceding the Hoover dam on the Colorado River is the Glen Canyon Dam. The volume flow rate is measured near the exit of this dam at a station calledLee’s Ferry. The average flow rate at this station from 1945 to 2009was 17,850 cubic feet per second [2].
As the Colorado River crosses the US-Mexico border, the volume flow rate is estimated to be 2,661 cubic feet per second [3].
3.2.Geographic Data
The shortest highway route from the Pacific Ocean to the Colorado River at the US-Mexico Border is 172 miles along Interstate Highway 8. The maximum elevation along this route is 4,199 feet above mean-sea-level, which occurs 72 miles along the route. A map of the pipeline route is shown below in Figure 1.
Figure 1. The selected route (bright pink) follows I-8 from San Diego to Yuma. The elevation profile for the selected route is plotted below. The path of the Colorado River is highlighted in light blue. The US-Mexico International Border is shown in yellow.
Design Concept
3.3.Overall Concept
A top-level system schematic is shown in Figure2. The proposed design begins with offshore pumping station P1, which pumps seawater from the Pacific Ocean to the desalinization plant. At the desalinization plant, the majority of the salt is removed, and the fresh water is collected into reservoir R1. Next, the secondary pumping station P2 provides most of the dynamic head necessary to cross the Sierra Nevada mountain range on the way toward the Colorado River. A third pumping station P3 provides the rest of the dynamic head necessary for the last segment, which empties into the Colorado River approximately 5 miles north of the US-Mexico International Border.
For more details about the piping stations, the pipeline segments, and the desalinization plant, please see Sections 5.3., 5.4., and 5.5., respectively.
Figure 2. System Schematic
3.4.Pumping System
The pumping required for this system is split into three large stations.
The first station, P1 pumps saltwater from the Pacific Ocean inland to the desalinization plant. The inlet to the desalinization plant will be no more than 30 feet above sea-level (). The desalinization plant processes water at an 80 percent throughput efficiency (see Section 5.5), meaning pumping station P1 must pump more water than any other station in order to meet the required flow rate for the rest of the system (Equation 1).
A set of wet pit column pumps (Goulds Pumps, Seneca Falls, NY) will be connected in parallel to meet the required flow rate for station P1. Because limited performance data is available for this pump,operating conditions will be assumed at 90 percent of the maximum published capacity and 5 percent of the maximum published head [Ref: Goulds Pump Catalog]. Therefore, each wet pit column pump will impart 1,003 cfs of seawater with 30 feet of dynamic head, for a total power output of 267 kW (Equations 2-4).The energetic efficiency of each pump at these operating conditions will be assumed at 80 percent, thus the input power required will be 334kW (Equation 5). A total of 19 pumps will be required to meet the required flow rate for this station (Equation 6), so the total input power required for station P1 pumps is 6.35 MW (Equation 7).
The second station, P2 will pump fresh water from reservoir R1 up and over most of the Sierra Nevada mountain range. This station will impart into the water most of the dynamic head required to reach the Colorado River. The flow rate required of this pumping station is the same as the total flow rate required of the entire system ().
Pumping station P2 will use high-head vertical industrial turbine pumps (Goulds Pumps) operating in parallel. Because limited information is available about these pumps, operating conditions will be assumed at 75 percent of the maximum published capacity and 75 percent of the maximum published head. Therefore, each pump will impart 109cfs of water with 2625feet of head, for an output power of 2.46 MW per pump (Equations8-10). The energetic efficiency at these operating conditions is assumed to be 85 percent, thus each pump requires 2.895 MW power at its input (Equation 11). In order to meet flow rate requirements for this station, 140 pumps must be connected in parallel (Equation 12), for a total input power of 405.3MW at pumping station P2 (Equation 13).
The third and final station, P3 will be located 2550 feet up the Sierra Nevada mountain range, and will pump the water the rest of the way to the Colorado River. The entire system flow rate will pass through Station P3 (). The max elevation of the pipeline is 4,200 feet, and the initial estimate of major and minor pipe losses is 1000 ft. So, Station P3 must impart 2575 feet of head into the water for it to reach the Colorado River (Equation 14).
Pumping station P3 will use the same wet pit column pumps from Station P1 (Goulds Pumps). Because limited information is available about these pumps, operating conditions will be assumed at 75 percent of the maximum published capacity and 75 percent of the maximum published head. Therefore, each pump will impart 835.5cfs of water with 675 feet of head, for an output power of 4.9 MW per pump (Equations 15-17). The energetic efficiency at these operating conditions is assumed to be 85 percent, thus each pump will require5.727 MW of power at its input (Equation 18). In order to meet flow rate requirements for this station, 19pumps must be connected in parallel (Equation 19), and in order to meet head requirements for this station, 4 pumps must be connected in series (Equation 20), for a total input power of 435.2 MW at pumping station P3 (Equation 21).
3.5.Piping System
The proposed piping design for segments S2 and S3 will use commercial steel pipes that are 184 inches in diameter (Pipe Industries Corp., Commerce City, CO). Because these pipes are so large, the relative roughness is very small (less than 0.0001), thus the Haaland Equation for friction factor is appropriate. The friction factor is therefore equal to 0.00894.
Pipe InformationTotal Number of Pipes / 19
Flow through each Pipe / 799.42 CFS
Pipe Cross-Sectional Area / 184.655 ft2
Velocity through Single Pipe / 4.329245
Reynolds Number / 6008732.3= 6.00*106
ε/D / .ooo15in/184in= 8.152*10-7
Using the Darcy-Weisbach formula, the major losses are calculated to be 67 feet in each pipeline in segment S2 and 241 feet for each pipeline in segment S3.
3.6.Water Treatment System
The reverse osmosis process removes large dissolved molecules from fluid solutions by applying a high pressure to the fluid when it’s on one side of a selective membrane. The particles are then left behind in the volume in front of the membrane, while the purified fluid passes through.
Figure 3: Basic scheme of desalination by RO [6]
The process can also be made more efficient by recovering some of the flow energy from the brine waste product via a turbine, which can power the pumps that run the filtration process.
Figure 4: Recover Energy using a turbine [6]
The largest desalinization plant in the US, located in Tampa, Florida has a maximum capacity of approximately 45.78 cubic feet per second [7]. Assuming an 80 throughput percentage, this proposed design will require 18,986 cubic feet per second (Equation 1). This is 415 times the flow capacity of the Tampa plant.
The Tampa plant cost 150 million to start up, and gets .659 cents per cubic meter of drinkable water[7]. Scaling up these figures due to the increase in flow, the estimated cost per year to desalinate 18,986 cubic feet per second of water will be $105 million annually
- Estimated cost of startup : 1 billion dollars
- .659 cents per cubic meter
- 18986 cfs to 538 m^3/s
- 538m^3/s*.659cent/m^3 = $3.36/sec
- Annual Cost = 31536000 (s/year)*$3.33/sec = $105,960,960/year
4.Cost Estimates
Because the prices of many of the selected components were not available, the prices below are all very rough estimates.
The wet pit column pumps from Goulds Pumps were estimated to cost $150,000. The total cost of 171 of these pumps will be $25,650,000
$150000/pump* 171pumps = $2565000
The selected piping is estimated to cost $1000/ft. The total cost of four 172-mile or 908,160 feet for 19 pipelines will be $17,255,040,000.
172mi*5280ft/mi*$1000/ft* 19 = $17,255,040,000
The total power required by the three pumping stations is 846 MW. Assuming continuous operation, the total energy required every year by these stations will be 26.78 GWh. Assuming a cost of $0.10/KWh, the energy required to run the pumping stations will cost $2,678 billion every year.
846MW*3600s/hr*8765.81277hr/year = 26.78GWh
26.78GWh*(10^6 KWh/gwh)*($0.10/KWh) = 2678 *10^9/year
Hoover Dam generates 4.2 billion KWh annually [5], according to our calculation the power needed to run these pumps is 26.78 GWh, or .2678 billion KWh, 15 times less than the Hoover dam.
4.2/.2678 = 15.68
5.Other Concerns
Steady State:
Once the system reaches steady state, the head that must be overcome is no longer the maximum elevation; rather it is the major losses plus the height difference of the final destination and the origin. This head is a considerably less. To take advantage of this phenomenon, turbines can be positioned along the down-slope of the mountain range to generate power, or multiple pumps can be turned off to save power. The flow would have to be diverted around these shut-down pumps.
Pump Failure:
There are 18 lines, and hundreds of pumps that make up this intricate system, it is only a minor problem if one or two pumpsfail. The real problem come in when a pump connected in series breaks down, a water dam would be created because water cannot flow through a broken pump.To remedy this problem an emergency water diversion scheme must be established.
Noise Pollution:
One pump is already a nuisance, but having over a hundred pumps located in close proximity is a real big problem.
Environmental Changes:
This project is the equivalent of constructing a river. Because such a massive amount of water is being pumped to one location, it will create drastic changes in the surrounding environment. The Hoover dam was built not only to supply energy, but also to prevent floods. In order to prevent an overflow of water, the flow would have to be increased slowly at first, and controlled to supply less or more water when needed.
Real Estate Costs:
The desalination plant itself would take up prime real estate in an already crowded city, so allocating space for it would be a very expensive undertaking.
6.Conclusion
Due to the vast distance, high elevation, and enormous flow that must be pumped, the cost for the energy alone to run this system would be astronomical. It is not a realistic situation to pump the water all the way to the Colorado. If the government were to seek a system that would increase the flow rate of the river as it enters Mexico, they should consider this system as a very last resort.
7.References
[1] Hodge, B.K., & Taylor, R.P. (1999). Analysis and design of energy systems. Upper Saddle
River, New Jersey: Prentice Hall.
[2] Water data report: lee’s ferry test station. (n.d.). Retrieved from
on
November 1, 2010.
[3Pacific institute recommendations report. (n.d.). Retrieved from
on November 1,
2010.
[5] Hoover dam frequently asked questions. (n.d.). Retrieved from
on November 1, 2010.
[6]Lachish, U. (n.d.). Optimizing the efficiency of reverse osmosis seawater desalination.
Retrieved from on November 1, 2010.
[7] Tampa bay seawater desalinization plant, florida. Retrieved from
technology.net/projects/tampa/ on November 1, 2010.
Department of Mechanical Engineering | San Diego State University
Copyright 2010