This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Lecture 6 Economic Pipe Selection Method
I. Introduction
•
•
•
• The economic pipe selection method (Chapter 8 of the textbook) is used to
balance fixed (initial) costs for pipe with annual energy costs for pumping
With larger pipe sizes the average flow velocity for a given discharge
decreases, causing a corresponding decrease in friction loss
This reduces the head on the pump, and energy can be saved
However, larger pipes cost more to purchase Cost to t a l minimum
to t a l
Energy costs =
annualized fixed costs fixed energy
Pipe Size (diameter) •
•
•
•
• To balance these costs and find the minimum cost we will annualize the fixed
costs, compare with annual energy (pumping) costs
We can also graph the results so that pipe diameters can be selected
according to their maximum flow rate
We will take into account interest rates and inflation rates to make the
comparison
This is basically an “engineering economics” problem, specially adapted to
the selection of pipe sizes
This method involves the following principal steps:
1. Determine the equivalent annual cost for purchasing each available pipe
size
2. Determine the annual energy cost of pumping Sprinkle & Trickle Irrigation Lectures Page 59 Merkley & Allen 3. Balance the annual costs for adjacent pipe sizes
4. Construct a graph of system flow rate versus section flow rate on a loglog scale for adjacent pipe sizes
•
•
•
• We will use the method to calculate “cutoff” points between adjacent pipe
sizes so that we know which size is more economical for a particular flow
rate
We will use HP and kW units for power, where about ¾ of a kW equals a HP
Recall that a Watt (W) is defined as a joule/second, or a Nm per second
Multiply W by elapsed time to obtain Newtonmeters (“work”, or “energy”) II. Economic Pipe Selection Method Calculations
1. Select a period of time over which comparisons will be made between fixed
and annual costs. This will be called the useful life of the system, n, in years.
•
• 2. The “useful life” is a subjective value, subject to opinion and financial
amortization conditions
This value could alternatively be specified in months, or other time period,
but the following calculations would have to be consistent with the choice For several different pipe sizes, calculate the uniform annual cost of pipe per
unit length of pipe.
•
•
•
•
• •
• A unit length of 100 (m or ft) is convenient because J is in m/100 m or
ft/100 ft, and you want a fair comparison (the actual pipe lengths from the
supplier are irrelevant for these calculations)
You must use consistent units ($/100 ft or $/100 m) throughout the
calculations, otherwise the ∆J values will be incorrect (see Step 11 below)
So, you need to know the cost per unit length for different pipe sizes
PVC pipe is sometimes priced by weight of the plastic material (weight
per unit length depends on diameter and wall thickness)
You also need to know the annual interest rate upon which to base the
calculations; this value will take into account the time value of money,
whereby you can make a fair comparison of the cost of a loan versus the
cost of financing it “up front” yourself
In any case, we want an equivalent uniform annual cost of the pipe over
the life of the pipeline
Convert fixed costs to equivalent uniform annual costs, UAC, by using the
“capital recovery factor”, CRF UAC = P ( CRF ) Merkley & Allen Page 60 (73) Sprinkle & Trickle Irrigation Lectures i (1 + i ) n CRF = (74) (1 + i) − 1
n where P is the cost per unit length of pipe; i is the annual interest rate
(fraction); and n is the number of years (useful life)
• Of course, i could also be the monthly interest rate with n in months, etc. •
• Make a table of UAC values for different pipe sizes, per unit length of pipe
The CRF value is the same for all pipe sizes, but P will change depending
on the pipe size
Now you have the equivalent annual cost for each of the different pipe
sizes • 3. Determine the number of operating (pumping) hours per year, Ot: Ot =
•
• 4. (irrigated area)(gross annual depth)
= hrs / year
(system capacity) (75) Note that the maximum possible value of Ot is 8,760 hrs/year (for 365
days)
Note also that the “gross depth” is annual, so if there is more than one
growing season per calendar year, you need to include the sum of the
gross depths for each season (or fraction thereof) Determine the pumping plant efficiency:
• The total plant efficiency is the product of pump efficiency, Epump, and
motor efficiency, Emotor Ep = EpumpEmotor
• (76) This is equal to the ratio of “water horsepower”, WHP, to “brake
horsepower”, BHP (Epump ≡ WHP/BHP) Sprinkle & Trickle Irrigation Lectures Page 61 Merkley & Allen •
•
•
•
• Think of BHP as the power going into the pump through a spinning shaft,
and WHP is what you get out of the pump – since the pump is not 100%
efficient in energy conversion, WHP < BHP
WHP and BHP are archaic and confusing terms, but are still in wide use
Emotor will usually be 92% or higher (about 98% with newer motors and
larger capacity motors)
Epump depends on the pump design and on the operating point (Q vs.
TDH)
WHP is defined as: WHP = QH
102 (77) where Q is in lps; H is in m of head; and WHP is in kilowatts (kW)
•
• If you use m in the above equation, UAC must be in $/100 m
If you use ft in the above equation, UAC must be in $/100 ft •
• Note that for fluid flow, “power” can be expressed as ρgQH = γQH
Observe that 1,000/g = 1,000/9.81 ≈ 102, for the above units (other
conversion values cancel each other and only the 102 remains)
The denominator changes from 102 to 3,960 for Q in gpm, H in ft, and
WHP in HP • 5. Determine the present annual energy cost: E= OtCf
Ep (78) where Cf is the cost of “fuel”
•
• •
• For electricity, the value of Cf is usually in dollars per kWh, and the value
used in the above equation may need to be an “average” based on
potentially complex billing schedules from the power company
For example, in addition to the energy you actually consume in an electric
motor, you may have to pay a monthly fee for the installed capacity to
delivery a certain number of kW, plus an annual fee, plus different timeofday rates, and others
Fuels such as diesel can also be factored into these equations, but the
power output per liter of fuel must be estimated, and this depends partly
on the engine and on the maintenance of the engine
The units of E are dollars per WHP per year, or dollars per kW per year;
so it is a marginal cost that depends on the number of kW actually
required Merkley & Allen Page 62 Sprinkle & Trickle Irrigation Lectures 6. Determine the marginal equipment cost:
•
•
•
• Note that Cf can include the “marginal” cost for the pump and power unit
(usually an electric motor)
In other words, if a larger pump & motor costs more than a smaller pump,
then Cf should reflect that, so the full cost of friction loss is considered
If you have higher friction loss, you may have to pay more for energy to
pump, but you may also have to buy a larger pump and/or power unit
(motor or engine)
It sort of analogous to the Utah Power & Light monthly power charge,
based solely on the capacity to deliver a certain amount of power
Cf ($/kWh) = energy cost + marginal cost for a larger pump & motor
where “marginal” is the incremental unit cost of making a change in the
size of a component •
•
• This is not really an “energy” cost per se, but it is something that can be
taken into account when balancing the fixed costs of the pipe (it falls
under the operating costs category, increasing for decreasing pipe costs)
That is, maybe you can pay a little more for a larger pipe size and avoid
the need to buy a bigger pump, power unit and other equipment
To calculate the marginal annual cost of a pump & motor: MAC = ( CRF $big − $small ( ) Ot kWbig − kWsmall ) (79) where MAC has the same units as Cf; and $big $small is the difference in
pump+motor+equipment costs for two different capacities
•
•
•
• The difference in fixed purchase price is annualized over the life of the
system by multiplying by the CRF, as previously calculated
The difference in pump size is expressed as ∆BHP, where ∆BHP is the
difference in brake horsepower, expressed as kW
To determine the appropriate pump size, base the smaller pump size on a
low friction system (or low pressure system)
For BHP in kW: BHP = Sprinkle & Trickle Irrigation Lectures QsHpump
102 Ep Page 63 (80) Merkley & Allen •
•
•
•
•
•
•
• 7. Round the BHP up to the next larger available pump+motor+equipment
size to determine the size of the larger pump
Then, the larger pump size is computed as the next larger available pump
size as compared to the smaller pump
Then, compute the MAC as shown above
The total pump cost should include the total present cost for the pump,
motor, electrical switching equipment (if appropriate) and installation
Cf is then computed by adding the cost per kWh for energy
Note that this procedure to determine MAC is approximate because the
marginal costs for a larger pump+motor+equipment will depend on the
magnitude of the required power change
Using $big $small to determine MAC only takes into account two (possibly
adjacent) capacities; going beyond these will likely change the marginal
rate
However, at least we have a simple procedure to attempt to account for
this potentially real cost Determine the equivalent annualized cost factor:
• This factor takes inflation into account: ⎡ (1 + e )n − (1 + i )n ⎤ ⎡
⎤
i
⎥⎢
⎥
EAE = ⎢
e −i
⎢
⎥ ⎢ (1 + i )n − 1⎥
⎣
⎦⎣
⎦ (81) where e is the annual inflation rate (fraction), i is the annual interest rate
(fraction), and n is in years
•
• 8. Notice that for e = 0, EAE = unity (this makes sense)
Notice also that the above equation has a mathematical singularity for e =
i (but i is usually greater than e) Determine the equivalent annual energy cost: E ' = (EAE)(E)
•
•
• (82) This is an adjustment on E for the expected inflation rate
No one really knows how the inflation rate might change in the future
How do you know when to change to a larger pipe size (based on a
certain sectional flow rate)?
Beginning with a smaller pipe size (e.g. selected based
on maximum velocity limits), you would change to a
larger pipe size along a section of pipeline if the Merkley & Allen Page 64 Sprinkle & Trickle Irrigation Lectures difference in cost for the next larger pipe size is less than
the difference in energy (pumping) savings
•
9. Recall that the velocity limit is usually taken to be about 5 fps, or 1.5 m/s Determine the difference in WHP between adjacent pipe sizes by equating the
annual plus annualized fixed costs for two adjacent pipe sizes: E' (HPs1 ) + UACs1 = E ' (HPs2 ) + UACs2
or, ∆WHPs1−s2 =
•
•
• (83) (UACs2 − UACs1)
E' (84) The subscript s1 is for the smaller of the two pipe sizes
The units of the numerator might be $/100 m per year; the units of the
denominator might be $/kW per year
This is the WHP (energy) savings needed to offset the annualized fixed
cost difference for purchasing two adjacent pipe sizes; it is the economic
balance point 10. Determine the difference in friction loss gradient between adjacent pipe sizes: ⎛ ∆WHPs1− s2 ⎞
∆Js1− s2 = 102 ⎜
⎟
Qs
⎝
⎠
•
•
•
•
•
• (85) This is the head loss difference needed to balance fixed and annual costs
for the two adjacent pipe sizes
The coefficient 102 is for Qs in lps, and ∆WHP in kW
You can also put Qs in gpm, and ∆WHP in HP, then substitute 3,960 for
102, and you will get exactly the same value for ∆J
As before, ∆J is a head loss gradient, in head per 100 units of length (m
or ft, or any other unit)
Thus, ∆J is a dimensionless “percentage”: head, H, can be in m, and
when you define a unit length (e.g. 100 m), the H per unit meter becomes
dimensionless
This is why you can calculate ∆J using any consistent units and you will
get the same result 11. Calculate the flow rate corresponding to this head loss difference:
• Using the HazenWilliams equation: Sprinkle & Trickle Irrigation Lectures Page 65 Merkley & Allen ∆J = Js1 − Js2 6⎛ 1.852 q⎞
= 16.42(10) ⎜ ⎟
⎝C⎠ 4
−
(Ds14.87 − Ds−2.87 ) (86) where q is in lps, and D is the inside diameter of the pipe in cm
• Or, using the DarcyWeisbach equation: ∆J =
• 800 f q2
gπ 2 −
5
(Ds15 − Ds−2 ) Solve for the flow rate, q (with q in lps; D in cm): ⎡
∆J
q = C⎢
⎢ 16.42(10)6 D−4.87 − D−4.87
s1
s2
⎣ ( •
• (87) ) ⎤
⎥
⎥
⎦ 0.54 (88) This is the flow rate for which either size (Ds1 or Ds2) will be the most
economical (it is the balancing point between the two adjacent pipe sizes)
For a larger flow rate you would choose size Ds2, and vice versa 12. Repeat steps 8 through 11 for all other adjacent pipe sizes.
13. You can optionally create a graph with a loglog scale with the system flow
rate, Qs, on the ordinate and the section flow rate, q, on the abscissa:
•
•
•
•
• • Plot a point at Qs and q for each of the adjacent pipe sizes
Draw a straight diagonal line from lower left to upper right corner
Draw a straight line at a slope of 1.852 (or 2.0 for DarcyWeisbach)
through each of the points
The slope will be different if the log scale on the axes are not the same
distance (e.g. if you do the plot on a spreadsheet, the ordinate and
abscissa may be different lengths, even if the same number of log cycles).
In constructing the graph, you can get additional points by changing the
system flow rate, but in doing so you should also increase the area, A, so
that Ot is approximately the same as before. It doesn’t make sense to
change the system flow rate arbitrarily.
Your graph should look similar to the one shown below Merkley & Allen Page 66 Sprinkle & Trickle Irrigation Lectures 14. Applying the graph.
•
•
•
• Find the needed flow rate in a given section of the pipe, q, make an
intersection with the maximum system capacity (Qs, on the ordinate), then
see which pipe size it is
You can use the graph for different system capacities, assuming you are
considering different total irrigated areas, or different crop and or climate
values
Otherwise, you can just skip step 13 and just do the calculations on a
spreadsheet for the particular Qs value that you are interested in
The graph is perhaps didactic, but doesn’t need to be constructed to
apply this economic pipe selection method III. Notes on the Use of this Method
1. If any of the economic factors (interest rate, inflation rate, useful life of the
system) change, the lines on the graph will shift up or down, but the slope
remains the same (equal to the inverse of the velocity exponent for the head
loss equation: 1.852 for HazenWilliams and 2.0 for DarcyWeisbach).
2. Computer programs have been developed to use this and other economic
pipe selection methods, without the need for constructing a graphical solution
on loglog paper. You could write such a program yourself.
3. The economic pipe selection method presented above is not necessarily
valid for:
• looping pipe networks
• very steep downhill slopes
• non“worst case” pipeline branches
Sprinkle & Trickle Irrigation Lectures Page 67 Merkley & Allen 4. For loops, the flow might go in one direction some of the time, and in the
opposite direction at other times. For steep downhill slopes it is not
necessary to balance annual operation costs with initial costs because there
is essentially no cost associated with the development of pressure – there is
no need for pumping. Non“worst case” pipeline branches may not have the
same pumping requirements (see below). 5. Note that the equivalent annual pipe cost considers the annual interest rate,
but not inflation. This is because financing the purchase of the pipe would be
done at the time of purchase, and we are assuming a fixed interest rate. The
uncertainty in this type of financing is assumed by the lending agency.
6. This method is not normally used for designing pipe sizes in laterals. For
one thing, it might recommend too many different sizes (inconvenient for
operation of periodicmove systems). Another reason is that we usually use
different criteria to design laterals (the “20%” rule on pressure variation).
7. Other factors could be included in the analysis. For example, there may be
certain taxes or tax credits that enter into the decision making process. In
general, the analysis procedure in determining pipe sizes can get as
complicated as you want it to – but higher complexity is better justified for
larger, more expensive irrigation systems. Merkley & Allen Page 68 Sprinkle & Trickle Irrigation Lectures IV. Other Pipe Sizing Methods
• Other methods used to size pipes include the following:
1. Unit head loss method: the designer specifies a limit on the
allowable head loss per unit length of pipe
2. Maximum velocity method: the designer specifies a maximum
average velocity of flow in the pipe (about 5 to 7 ft/s, or 1.5 to 2.0 m/s)
3. Percent head loss method: the designer sets the maximum pressure
variation in a section of the pipe, similar to the 20%Pa rule for lateral
pipe sizing •
•
•
•
• It is often a good idea to apply more than one pipe selection method and
compare the results
For example, don’t accept a recommendation from the economic selection
method if it will give you a flow velocity of more than about 10 ft/s (3 m/s),
otherwise you may have water hammer problems during operation
However, it is usually advisable to at least apply the economic selection
method unless the energy costs are very low
In many cases, the same pipe sizes will be selected, even when applying
different methods
For a given average velocity, V, in a circular pipe, and discharge, Q, the
required inside pipe diameter is: D=
• 4Q
πV (89) The following tables show maximum flow rates for specified average velocity
limits and different pipe inside diameters Sprinkle & Trickle Irrigation Lectures Page 69 Merkley & Allen Gallons per Minute
D (inch)
A (ft2)
0.5 0.00136
0.75 0.00307
1 0.00545
1.25 0.00852
1.5 0.01227
2 0.02182
3 0.04909
4 0.08727
5 0.13635
6 0.19635
8 0.34907
10 0.54542
12 0.78540
15 1.22718
18 1.76715
20 2.18166
25 3.40885
30 4.90874
40 8.72665
50 13.63538 Litres per Second Velocity Limit
5 fp s
7 fps
3.1
4.3
6.9
9.6
12.2
17.1
19.1
26.8
27.5
38.6
49.0
68.5
110
154
196
274
306
428
441
617
783
1,097
1,224
1,714
1,763
2,468
2,754
3,856
3,966
5,552
4,896
6,855
7,650
10,711
11,017
15,423
19,585
27,419
30,602
42,843 Cubic Feet per Second
D (ft)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20 A (ft2)
0.785
3.142
7.069
12.566
19.635
28.274
38.485
50.265
63.617
78.540
95.033
113.097
132.732
153.938
176.715
201.062
226.980
254.469
283.529
314.159 Merkley & Allen Velocity Limit
5 fp s
7 fps
3.93
5.50
15.71
21.99
35.34
49.48
62.83
87.96
98.17
137.44
141.37
197.92
192.42
269.39
251.33
351.86
318.09
445.32
392.70
549.78
475.17
665.23
565.49
791.68
663.66
929.13
769.69 1,077.57
883.57 1,237.00
1,005.31 1,407.43
1,134.90 1,588.86
1,272.35 1,781.28
1,417.64 1,984.70
1,570.80 2,199.11 Page 70 D (mm)
10
20
25
30
40
50
75
100
120
150
200
250
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2700
2800
2900
3000
3100
3200
3300
3400 A (m2)
0.00008
0.00031
0.00049
0.00071
0.00126
0.00196
0.00442
0.00785
0.01131
0.01767
0.03142
0.04909
0.07069
0.12566
0.19635
0.28274
0.38485
0.50265
0.63617
0.78540
0.95033
1.13097
1.32732
1.53938
1.76715
2.01062
2.26980
2.54469
2.83529
3.14159
3.46361
3.80133
4.15476
4.52389
4.90874
5.30929
5.72555
6.15752
6.60520
7.06858
7.54768
8.04248
8.55299
9.07920 Velocity Limit
1.5 m/s
2 m/s
0.1
0.2
0.5
0.6
0.7
1.0
1.1
1.4
1.9
2.5
2.9
3.9
6.6
8.8
11.8
15.7
17.0
22.6
26.5
35.3
47.1
62.8
73.6
98.2
106
141
188
251
295
393
424
565
577
770
754
1,005
954
1,272
1,178
1,571
1,425
1,901
1,696
2,262
1,991
2,655
2,309
3,079
2,651
3,534
3,016
4,021
3,405
4,540
3,817
5,089
4,253
5,671
4,712
6,283
5,195
6,927
5,702
7,603
6,232
8,310
6,786
9,048
7,363
9,817
7,964
10,619
8,588
11,451
9,236
12,315
9,908
13,210
10,603
14,137
11,322
15,095
12,064
16,085
12,829
17,106
13,619
18,158 Sprinkle & Trickle Irrigation Lectures ...
View Full
Document
 Fall '03
 Sprinkle

Click to edit the document details