EWB@WSU: Campbell Farm – Yakama Indian Reservation, WA

Pump Selection Outline

EWB@WSU: Campbell Farm

Yakama, WA Indian Reservation

Pump Selection Outline

Surveying: Joshua Horky and Alex McDonald

Demand Study, Head Calculations & Surveying Calculations: Joshua Horky

CAD Drawings, System Compilation, & Final Pump Selection:Alex McDonald

Edited By: Alex McDonald

June 15, 2005

Table of Contents

  1. Introduction……………………………………………………………………3-4
  2. Fig 1: Proposed Potable Waterline
  3. Fig 2: Potable Well System Schematic
  4. Demand Study………………………………………………………………5-10
  5. Campbell Farm Water System Statistics
  6. Fig 3: Diurnal Demand Curve
  7. Demand Estimate for Current Population
  8. Demand Estimate for Population Expansion
  9. Results
  10. Fig 4: Population Demand Curve 60 gpcpd
  11. Fig 5: Consumption rate Demand Curve pop. of 64
  12. Fig 6: Consumption rate Demand Curve pop. of 83
  13. Demand Summary
  14. Determining Head…………………………………………………………...11-14
  15. Theoretical Basis
  16. Fig 7: Total Head Diagram
  17. Calculations Summary
  18. Discussion
  19. Total Head Calculation
  20. About SQE Pumps…………………………………………………………..15-22
  21. Theoretical Basis
  22. Reading Typical Pump Curves
  23. Fig 8: Typical Pump Curve Demonstration
  24. Reading Grundfos Pump Curves
  25. Fig 9: Typical Pump Curve Demonstration
  26. Max Production
  27. 10 SQE05B-200
  28. Fig 10: 10 SQE05B-200 at max production
  29. 15 SQE07B-180
  30. Fig 11: 15 SQE07B-180 at max production
  31. PressureRange
  32. 10 SQE05B-200
  33. Fig 12: 10 SQE05B-200 at 10 GPM
  34. 15 SQE07B-180
  35. Fig 13: 15 SQE07B-180 at 10 GPM
  36. Variable Pressure
  37. Using the CU 301
  38. Fig 14: CU 301 Control Box
  39. Pump Selection………………………………………………………………22-25
  40. Pressure Tank
  41. Grundfos SQE Series
  42. Fig 15: System Design
  43. Pump Specifications
  44. Fig 16: SQE Cutaway
  45. Projected Cost
  46. Special Thanks………………………………………………………………26-27
  47. Works Cited………………………………………………………………….....28

I) Introduction

Directors of Campbell Farm, David Hacker and Sheri, have enlisted the help of EWB@WSU to design and implement a new potable well system. The system is designed per the specifications of the Washington Department of Health (DOH) permit process for Group A-TNC systems.

The Campbell Farm is a Presbyterian non-profit organization that serves as a mission station, hosting people from across the country that come to participate in service projects on the Yakama Indian Reservation.

Danish pump manufacturer Grundfos comes highly recommended by faculty associated with Engineers Without Borders at WashingtonStateUniversity for their quality of product and solid warranty, and thus are the pumps reviewed.

The purpose of this paper is to compile a list of appropriate well pumps for the Campbell Farm potable well system, and demonstrate the reason they were chosen. Pump selection has been based on the head required to transport water from its location at the bottom of the well to the top of a 3/4 in PVC waterline raised 17 ft above ground. It is assumed that all horizontal mainline water extensions from the Flexcon bladder tank will be using 2-inch PVC piping to minimize headloss due to friction. This discovery is to be communicated in this paper. It is expected that the Campbell Farm and Youth Education Services should be able to select a pump based on the information in the following pages. Figures 1 and 2 have been included to help illustrate the total system design for the Campbell Farm well system.

Fig 1: Draft of the Campbell Farm’s proposed potable waterline, pipeline shown in blue

Fig 2: Draft of the Campbell Farm’s potable well system schematic by EWB@WSU.*

*Note: Insulation/heating of the 4.5 gallon pressure tank, chlorination, and the complete design of the pump house for the well system is yet to be determined, and therefore subject to change. Chlorination may not be required, and will be determined upon the completion of water testing by Cascade Analytical. Pump house schematics are not required for the DOH permit application.

II) Demand Study

Campbell Farm Water System Statistics:

Variables provided by the Campbell Farm

Current Usage

  1. Farm House
  2. 2 Full Baths (sink, toilet, shower)
  3. 1 Kitchen
  4. 5 people Year Round
  5. 2 additional People in summer
  6. Trailer
  7. 1 Full Bath
  8. 1 Kitchen
  9. 2 People Year Round
  10. Peach Haven
  11. 1 Full Bath
  12. 1 Kitchen
  13. 8-10 People maximum temporary guests
  14. Spitzenburg
  15. 1 Full Bath
  16. 1 Kitchen
  17. 12-14 People maximum temporary guests
  18. Hop kiln
  19. 5 Baths (one with 2 sinks, 3 stalls, and 2 showers)
  20. 1 Commercial Kitchen
  21. 1 Kitchen in apartment
  22. 29 People Temporary Guests
  23. 2 People in Apt Year Round

Expansion

  1. 2 Additional Staff Houses
  2. 1 Full Bath each
  3. 1 Full Kitchen each
  4. 2 People each year round
  5. 9 Temporary Guest Houses (20X20 in size)
  6. 5 Houses with Full Bathrooms for 8 people each
  7. 4 For families (2-4 people) with Full Bathrooms, Kitchenettes (sink, microwave, small refrigerator)
  8. Chapel
  9. Public Restrooms with a couple stalls and sinks each for men and women
  10. ConferenceBuilding
  11. Commercial Kitchen
  12. Public Restrooms with 4 sinks and stalls, and shower each for men and women

Fig 3: Diurnal Demand Curve used for Campbell Farm’s projected Total GPM Demand. Curve produced by the American Water Works Association (AWWA).

The diurnal curve used to project the Total GPM Demand for extended period simulations was assumed to be equivalent to the American Water Works Association (AWWA) curve shown in Figure 3.

Diurnal curve is used as follows:

  1. The horizontal axis represents a 24-hour time period.
  2. The vertical axis represents a system multiplier.
  3. A determined average system demand is set to the 1.0 multiplier on the graph. System demand averages are based on gallons per capita per day as rated by an individual’s type of residency – or how long a person lives on the Campbell Farm, and what waterworks activities they are expected to perform.
  4. Then the average demand is multiplied by the highest point on the curve. It is important to multiply it by the highest peak, because that is the rate at which the demand on the Farm’s water system is at its highest. This value is 175% of the average of estimated gallons per day consumption of the Farm.
  5. The peak value is then converted into gallons per minute through simple unit conversion.
  6. All components of the system must be able to perform at that peak level – i.e. a pump must be able to meet the highest demanded flow in order for it to be able to operate at all times.
  7. Demonstrations of this process are found on page 7 (detailing current demand) and page 8 (detailing expansion demand). Graphs of the current demand are included on pages 9 and 10.

Demand Estimate for Current Campbell Farm Population

Given:
Current
full time / 9 / people
seasonal / 55 / people
Total / 64
Also / 1 commercial Kitchen
2 sets of public restrooms, including 1 shower
Find: / Average flow per day
Peak flow per hour
Assumptions:
gpcpd = / 100 / full time staff/permanent residents
gpcpd = / 60 / seasonal campers/non-permanent residents
Solution:
average flow per day = / 4200 / gallons
peak flow per hour = / 175 / % of average daily flow
= / 7350 / gallons/day
Q1 / = / 5.1 / gallons/minute

Demand Estimate for Campbell Farm’s Population Expansion

Given:
Current
full time / 9 / people
seasonal / 55 / people / 64
Expansion
Full time / 4 / people
seasonal / 56 / people / 60
Total / 124 people
Also / 1 commercial Kitchen
2 sets of public restrooms, including 1 shower
Find: / Average flow per day
Peak flow per hour
Assumptions:
gpcpd = / 100 / full time staff/permanent residents
gpcpd = / 60 / seasonal campers/non-permanent residents
Solution:
average flow per day = / 7960 / gallons
peak flow per hour = / 175 / % of average daily flow
= / 13930 / gallons/day
Q2 / = / 9.7 / gallons/minute

Results: (gpcpd = gallons per capita per day). Figures 4, 5, and 6 (placed below) are graphical representations of simulations run using current demand estimates from page 7.

Fig 4: Population Demand Curve assuming 60 gpcpd

Figure 4 shows how water demand increases as a function of population when the rate at which a person utilizes water is assumed to 60 gpcpd. The horizontal, solid, red line represents the maximum amount of water that can be legally removed from the ground each day because of current permitting limitations. This corresponds to a maximum population of 83 people, which meets the 5000 gallons per day DOH requirement.

Fig 5: Consumption rate Demand Curve assuming a population of 64 (current pop.)

Figure 5 shows how demand increases as a function of rate of personal utilization of water when the population is assumed to be 64 people (the current maximum seasonal population). Again, the red line represents the legal limit and here corresponds to 78 gpcpd.

Fig 6: Consumption rate Demand Curve assuming a population of 83 (max legal at 60gpcpd)

Figure 6 shows how demand increases as a function of rate of personal utilization of water when the population is assumed to be 83 people (the maximum anticipated seasonal population after expansion). Yet again, the red line represents the legal limit and here corresponds to 60 gpcpd

Demand Summary:

According to the DOH, the standard rate of consumption is 100 gpcpd (gallons per capita per day). This number has been reduced to 60 gpcpd for non-residents because of the anticipated lower consumption rate of the campers (laundry, etc is done at home by guests).

The following is in regards to the demand estimates on pages 7 and 8:

The daily consumption rate has been kept below 5000 gallons/day for the current camp size because a daily rate beyond 5000 gallons/day will require the farm to obtain a water right from the Washington State Department of Ecology, which is not likely to happen before the project must be completed this summer (2005). The proposed demands allow the camp to carry on at its present size and demand without obtaining additional water rights.

The system currently demands 4200 gallons/day and a peak supply rate of 5.1 GPM. The expanded system will demand 7960 gallons/day and a peak supply rate of 9.7 GPM. These are the basic demands which must be considered in the feasibility study.

Q1 (current system demand) = 5.1 GPM

Q2 (expansion system demand) = 9.7 GPM

Therefore Q (total system demand) is less than 10 GPM or 0.02228 ft3/s

III) Determining Head

Theoretical Basis

Water does not move itself from a low elevation to a relatively high elevation--this requires energy. This energy can be added to hydraulic systems through a pump. A pump is expected to be lowered into the well and will receive energy in the form of electricity to be transmitted to the water mechanically. This can be communicated mathematically in the following equation.

Where:

E1is the energy of the water in at the bottom of the well

E2 is the energy of the water at the top of the 0.75 in pipe raised 17 ft in the air

wshaftin is the work done on the water by the pump

The calculations included in this paper facilitate determination of the work required from the pump to move the water to the second energy level E2.

The pump demand is primarily based upon two (2) things:

  1. Q (flow measured in units of volume per unit time i.e. g/m, m3/s, or ft3/s)
  2. H (energy measured in units of distance i.e. m or ft)

Work can be expressed in feet of water because (p = ρgh)

The required head is a function of several things:

  • Required elevation differential
  • Required pressure differential
  • Required velocity differential
  • Headloss in system
  • Minor losses (due to valves, bends, diameter changes, etc.)
  • Major losses (due to friction)
  • Major and minor losses are calculated with a spreadsheet prepared previously attached here.

The Q – Campbell Farm’s total system demand – has been determined by a previous demand study by EWB@WSU (please see Demand Study pages 5-10) to be less than 10 gallons per minute (GPM) or 0.02228 ft3/s. From the estimated Q value and the assumptions made from Fig 3: Total Head Diagram, pages 11-13 will demonstrate how to determine the value of H – total head required by the pump to operate.

Fig 7: Total Head Diagram used for calculating head created by EWB@WSU

Calculations Summary

The required elevation differential is a function of the depth of the well, the minimum predicted static level of water in the well, and the height of the water in the storage tank.

Minimum static level of water = 17 ft below waterline

Pump depth = 90 ft below waterline

Height of water in Hop Kiln = 20 ft above waterline

Total elevation differential = 90 ft + 20 ft = 110 ft

Required pressure differential depends on the pressure at the pump in the bottom of the well and upon the pressure required at the top of the 0.75 in PVC line. The pressure at the bottom of the well is a function of the height of the water above the pump in the well. The pump is to be placed 90 feet below the waterline and the static water level may drop as low as 17 ft below the waterline. The water is to be pumped to 20 ft above the waterline. The pressure at the pump is 110 ft - 37 ft = 73 ft. The required pressure at the top of the 0.75 in PVC line is 60 PSI = (~2.307) (60) = ~ 138 ft.

Total pressure differential = 138 ft - 73 ft = 65 ft

The required velocity differential is zero (0) since the water is to be at rest in the bladder tank.

The headloss due to friction is a function of the length, diameter, and material of the pipe as well as the flow through it. The total length of the potable pipe system is approximately 446 ft. The friction fluctuates based on the pipeline’s various inner diameters (which are different than their ratings) and the 10 GPM of water flowing though them. The headloss due to friction has been determined to be ~6 ft by the spreadsheet attached above.

Total headloss due to friction = 6 ft

The headloss due to fittings in the pump house is comprised of the gate valves, pressure release valve, water meter, pressure/bladder tank, and pressure transducer. Negligible headloss from the pressure tank, because it acts as a burp tank to absorb back pressure. The pressure transducer creates negligible loss because it detects pressure via a spring loaded system that T’s into the pipeline, creating a small dimple on the line – which is insignificant for this system. The gate valves 0.16 ft of head between them. Thus, the total head from this system is too low to count.

The headloss due to bends, diameter changes, and other constraints is a function of velocity, diameter, and k values. The system is conservatively assumed to have seven 90˚ threaded bends of varying diameters which have been determined to cost ~1 ft of head for the bends and ~1 ft of head for the diameter change (by the spreadsheet attached above).

Total headloss due to bends, diameter and other = (1 + 1) ft = 2 ft

Discussion

The conclusions concerning head losses are approximate values. The calculations have assumed a 20 ft vertical lift above waterline at 60PSI for the 0.75 in PVC Hop Kiln. Additional losses will undoubtedly occur due to bends and friction from a greater length of the 3/4 in pipe maintained in the internal plumbing of the Hop Kiln than the 20 ft stated. However, this loss is not expected to bring the Hop Kiln’s water pressure below the Washington Department of Health’s (DOH) requirement of 30 PSI. It is assumed that all horizontal mainline water extensions from the Flexcon bladder tank will be using 2-inch PVC piping to minimize headloss due to friction.

Total Head Calculation

The total input required from the pump which must move water from the bottom of the well to the top of the 0.75 in PVC line can now be calculated and expressed.

H (total required head) = (110 + 65 + 6 + 2) ft = 183 ft

IV) About SQE Pumps

Theoretical Basis

Water does not move itself from a low elevation to a relatively high elevation--this requires energy. This energy can be added to hydraulic systems through a pump. A pump is expected to be lowered into the well and will receive energy in the form of electricity to be transmitted to the water mechanically. Selection of pumps for potable water systems is primarily based upon two (2) things:

  1. Q (flow measured in units of volume per unit time i.e. g/m, m3/s, or ft3/s)
  2. H (energy measured in units of distance i.e. m or ft)

According to an earlier demand study produced EWB@WSU and expansion variables provided by the Campbell Farm, the Farm will not require a flow rate over 10 GPM for its entire potable system demand before and after expansion. Selection of the Grundfos SQE pumps was based on the estimated system demand and total required head of 183 ft.

  1. Q (total system demand) is less than 10 GPM.
  2. H (total required head) = 183 ft

Reading Typical Pump Curves

Fig 8: TypicalPump Curve Demonstration, graph from us.grundfos.com

At left is an example of a typical pump curve. The H (total head required - ft) is demonstrated on the vertical axis, and the Q (total system demand – GPM) on the horizontal axis. The shaded region illustrates at what power levels the pump can operate. To read the graph, simply find system Q and H values, and find where they intersect on the graph. If your values intersect outside of the region, then they exceed the possible function of the pump. It is typically best to select a pump whose operational values are 2/3rds of the way down the curve according to Grundfos technical engineers.

Reading Grundfos Pump Curves

Fig 9: GrundfosPump curve demonstration, graph from us.grundfos.com

At right is an example of a pump curve to be found on us.grundfos.com. The H (total head required - ft) is demonstrated on the vertical axis, and the Q (total system demand – GPM) on the horizontal axis. Below is another graph that illustrates at what power levels the pump can operate. To read the graph, simply find system Q and H values, and find where they intersect on the graph. If your values intersect outside of the region, then they exceed the possible function of the pump. It is typically best to select a pump whose operational values are 2/3rds of the way down the curve according to Grundfos technical engineers.