Cooling Water System Design
Cooling Water System Design
Jin-Kuk Kim
Supervisor : Professor Robin Smith
First, a brief description of the problem of cooling water system design is given,
together with the main issues motivating this work.
In this presentation, the cooling tower and the cooling water network is first
examined separately to discuss the nature of cooling water system design. Then the
design of overall cooling water systems is investigated.
Design options for heat load distribution are explained for debottlenecking.
Recirculation water
Cooling
Tower HEN
Makeup Cooling water
Blowdown
Cooling towers are usually used where water is used as a cooling medium. When
the use of fresh cold water is limited, the continuous re-circulating and re-cooling of
cooling water is common.
The cooling water system mainly consists of the cooling tower and heat exchanger
network (HEN). The cooling water used in the HEN returns to the cooling tower
where the hot return cooling water is cooled.
HE 3
new HE
&
CT performance ?
HE 4 Process
changes
CW Combined
CT water/energy
Network Performance
Design pinch approach
Interaction
System changes for the cooling water system can be expected. For example, a new
heat exchanger is introduced into the HEN, or the heat duty of coolers is changed,
or process constraints limit the current operating conditions. These process changes
influence the conditions of the return cooling water and consequently affect the
cooling tower performance. But we don’t know exactly how the cooling water
system is affected with the new conditions and how the cooling water network
design affect the cooling system.
The cooling system has interactions between the cooling water network and the
cooling tower performance. As the cooling water system has energy implications, a
combined water and energy pinch analysis should be used to investigate the
interactions of the overall system.
Objectives
Approach Range
• CT effectiveness
QACTUAL Twbt Water
effectiveness(I) = Temp
QMAX
The cooling tower removes heat from hot water by direct contact with air. The
cooling is accomplished by a combination of sensible heat transfer and the
evaporation of water.
The cooling tower performance is influenced by air and water conditions and
packing design. These factors determine how much driving force can be obtained
with given conditions. This humidification process is operated most effectively by
keeping the driving force as large as possible.
The cooling tower effectiveness is defined as the ratio of actual heat removal to the
maximum attainable heat removal. This value measures the thermal performance of
cooling tower.
Cooling Tower Model
Counter-current cooling tower with mechanical air-draft
• Assumptions
q Adiabatic operation in cooling tower
q Air and water flowrates are constant
q No drift and leakage loss
q No effect of the air fan placement
q Interfacial areas are equal for heat and mass transfer
q Temeprature variation has no effect on the transfer coefficients
q Thermodynamic properties are constant across the cross
section of the tower
In this work, the cooling tower is assumed to be the counter-current contacting type
between air and water and air flows by mechanical fan.
The model needs to predict the temperature of exit water from the tower for given
design and operating conditions.
To illustrate the working principle of cooling towers and predict cooling tower
efficiency, a one-dimensional steady-state model with simple but sufficient
accuracy is developed.
G
Water heat balance (^) TG+dTG L+dL
H+dH TL+dTL
LCPL dTL = {GCPL dW − hL a ⋅ dz}(Ti − TL ) W+dW P
The mechanism of cooling is a combination of heat and mass transfer. Latent heat is
carried across the interface between water and air by diffusion of water vapour. The
heat is also transferred by temperature difference between water and air.
The figure represents a section of the tower with differential height dZ and shows
the water and air flow, which are separated by the interface. The phenomena of
mass and heat transfer are modelled as transfer coefficients multiplied by driving
forces based on the interface temperature. To find the interface temperature, heat
balances are set up for the overall control volume (1), water (2) and air side (3).
From these three equations, the equation for the interface temperature is obtained.
But, without the differential value of humidity and air temperature, the interface
temperature cannot be determined.
dW kG a
Air Humidity = (Wi − W ) G2
TG2 water
dz G H2 L2
W2 TL2
dTG h a
Air Temperature = G (Ti − TG )
dz GCS
G
= L (TL − Ti )
Water Temperature dTL ha TG+dTG
H+dH L+dL
dz LCPL W+dW TL+dTL
G L dz
TG TL
Absolute Humidity H
W
MW pS ì B ü z
Wi = pS = exp í A − ý
M Air (P − pS ) î C +T
The water to be cooled enters the top of the tower and the cooling air is either
induced or forced through the tower from the bottom to the top.
Air humidity can be represented by the first equation, which represents mass
transfer of water vapour from the interface to the air. Air and water temperatures
can be represented in the same manner as the first equation.
These differential equations need the value of interface temperature. But the
differential increase of humidity and air temperature is also needed to calculate the
interface temperature. This means an iterative method is necessary to obtain the
value of interface temperature.
Assume TL1
Z = Z0
Adjust
TL1
Assume Ti Z =Z+dZ
No
No
Calculate
dTG/dZ,dW/dZ
Adjust Z = Zmax / TL2 -TL2,CAL / < M
Ti Yes
CalculateTi
Yes
Calculate
/ Ti -Ti,CAL / < M TG,W,TL Stop
No Yes
This slide shows the flowchart for the model, which finds the exit water and air
conditions when the inlet air and water conditions are given.
First the exit water temperature (TL1) is assumed and then numerically integrated
from bottom to top of the tower (Zo - Zmax). The role of inside loop is to find the
interface temperature at every differential increment.
The calculated inlet temperature is compared with the real inlet temperature (TL2).
The value of exit water temperature (TL1) updated if the condition is not satisfied.
The proposed model is accurate enough within 1% error on the basis of the limited
data available.
Cooling Water System Model
E
• CT Model F2
Evaporation T2
T1 = f (F2,T2,TWBT)
F1 = f (F2,T2,TWBT) CT CW
Network
E = f (F2,T2,TWBT) QHEN
F1
T1 Makeup
• Makeup / blowdown
Cold
F0 = F1 - B + M Blowdown F0
T0
F0T0 = (F1 - B)T1 + MTM B M TM
As the blowdown and makeup have an effect on the heat and mass balances of the
cooling water system, the overall model of the cooling water system should include
the blowdown and makeup.
Makeup water is often added to the cooling tower basin before cold blowdown is
taken. But the effect of makeup location is not significant.
We now introduce a key concept in the design and operation of cooling towers
known as the cycles of concentration. The cycles of concentration (CC) is defined
as the ratio of the concentration of a soluble component in the blowdown stream to
that in the makeup stream. The blowdown and makeup are calculated from the
evaporation loss and cycles of concentration.
CT Modelling Result 1
34
32
Water Outlet Tem
30
28
lower CW outlet temperature
26
24 52
50
W a 2.0 48
ter 46
In l
ur e
44
e t F 1.5 42 ra t
l ow 40 pe
ra t 1.0 m
e 38 Te
36 l et
0.5 34
te r In
Wa
This graph shows how the cooling water outlet temperature is affected when the
water inlet conditions are changed. When the cooling water inlet conditions have a
high temperature and low flowrate, the cooling tower removes more heat from
water and obtains a lower cooling water outlet temperature.
CT Modelling Result 2
0.8
Here is another graph to show the cooling tower performance. This graph shows
how the effectiveness of the cooling changes when the water inlet conditions are
changed. When the inlet cooling water has conditions with high temperature and
low flowrate, the effectiveness of the cooling tower is high and cooling tower
removes more heat from the water.
3. Design Aspects of Cooling Water
Networks
HE 3
CW HE 4
Supply line
Maximise CW return flowrate
Minimise CW return temperature
Q
No systematic methods have been suggested to deal with the design aspects of
cooling water networks.
The current practice for cooling water network design is based on parallel
configurations. In a parallel configuration, the cooling water demand is determined
by minimising the flowrate to the individual heat exchangers. But the cooling water
return flowrate is maximised and the cooling water return temperature is minimised
because there is no-reuse of cooling water in a parallel design.
When the cooling water system is bottlenecked, a parallel design is not flexible in
dealing with various process restrictions.
What is the best CW network design ?
• Not all of cooling duties require CW at the CW supply temperature
HE 1 HE 1
HE 2 HE 2
Parallel Series
• Series Arrangements
Increase Heat removal
CW return temperature
of cooling tower
Decrease can be expected
CW recirculation flowrate to increase
A series arrangement, in which cooling water is reused in the network, makes the
return temperature of the cooling water higher and the flowrate lower. From the
results of the cooling tower model, the heat removal in the cooling tower can be
expected to increase with these conditions.
If we change the design configuration from parallel to series arrangements, the
cooling tower can manage more heat load on the coolers.
CW Network Design
• Stream data of heat exchangers using CW
2 50 40 100 1000
3 85 40 40 1800
4 85 65 10 200
The data for hot process streams are represented as CP values. A CP value, which is
the value of heat capacity multiplied by flowrate of cooling water, is used for this
study. The heat capacity of cooling water is assumed constant through all the
temperature range.
Until Now.....
• Parallel Configuration : No CW re-use
HE 1
CW from CT HE 2 CW to CT
HE 4
This cooling water network is designed with a parallel configuration. The parallel
configuration follows once-through policy, which means there is no re-use of
cooling water.
But
C(ppm) C(ppm)
C 800 800
Process
Cout, Composite curve
max
400 400
In water pinch analysis the mass load of contaminant is transferred to the water
stream, which becomes more contaminated and exits the operation as wastewater.
The “limiting water profile” is defined in terms of the maximum inlet and outlet
concentrations for the water stream. These limits may be imposed by a number of
different process restrictions: mass transfer driving force, corrosion, fouling, water
treatment and so on. This profile defines a boundary between feasible and infeasible
operation.
Once we have the limiting water profile for every water-using operation, we can
proceed to construct a limiting overall curve by combining all the individual
profiles into a single composite curve. This curve represents how the total system
would behave if it were a single water-using operation.
400 2
3
100 PP 45.7 kg/s 100ppm
44.3 kg/s 100ppm 4
0
M (g/s) 0 kg/s 44.3 kg/s 45.7 kg/s
P
90 kg/s
0 ppm
Water Mains
10 - 24 Cooling Water System Design
Some similarity from water pinch analysis can be observed from this representation
of heat exchanger. The limiting cooling water profile is defined in terms of the
maximum inlet and outlet temperatures for the cooling water stream. These
allowable temperatures are limited by the “minimum temperature approach” which
avoids unacceptably high heat exchanger area. These limits may be imposed by
other considerations: corrosion, fouling, cooling water treatment, safety, etc.
This profile defines a boundary between feasible and infeasible operation. Any
cooling water supply line at or below this profile results in a feasible operation.
Also, the limiting cooling water profile allows us to analyse the heat exchanger on a
common basis because this profile represents simultaneously the water and energy
characteristics.
Limiting Cooling Water Data
2 30 40 100 1000
3 30 75 40 1800
4 55 75 10 200
-=∆Tmin = 10 oC
- CW inlet temperature = 20 oC
10 - 26 Cooling Water System Design
The limiting cooling water data can be extracted from the hot process stream data.
The minimum temperature approach is assumed to be 10 °C. The supply
temperature of cooling water is taken to be 20 °C.
Construction of Limiting CW Composite Curve
T(_C) T(_C)
75 75 Cooling Water
Composite Curve
55 57.8 oC
55
Cooling
Water
40 40 Supply Line
for Max.
Re-use
20 20
From the limiting cooling water profile for every heat exchanger, the cooling water
composite curve can be constructed by combining all individual profiles into a
single curve. In this construction, the inlet and outlet temperatures of each operation
define temperature intervals. By combining operations within temperature intervals,
the limiting cooling water curve is obtained. This curve represents how the total
system would behave if it were a single cooling-water-using operation.
By maximising the outlet temperature of the cooling water supply line, the cooling
water demand is minimised and the cooling water re-use is maximised. Each point
where the supply line touches the composite curve creates a pinch in the design. It is
important to note that the interpretation of the pinch does not represent a zero
driving force of heat transfer, as all constraints were built into the limiting cooling
water profile.
Cooling Water Network Design
• Use Kuo and Smith’s method
P PP PPP
90kW/ C o o
45.7 kW/ C 0 kW/ oC
45.7 kW/ oC 75 oC
T(_C) 75 oC
20 oC 40 oC
1
75 PPP
2
55
40 PP 3
45.7 kW/ oC 40 oC
20 44.3 kW/ oC 40 oC
4
P
90kW/ oC 0 kW/ oC 44.3 kW/ oC 45.7 kW/ oC
20 oC Q(kW) 20 oC 40 oC 75 oC
To achieve the minimum cooling water flowrate target in a design, Kuo and Smith’s
water main method is applied to cooling water network design.
As in water pinch analysis, we decompose the problem into distinct regions and
identify the cooling water requirements for each region independently by cutting off
pockets and calculating the individual cooling water flowrate requirements.
In this way the re-use opportunities are identified and the network design achieves
the target.
Design of CW Network with Maximum Re-use
CP = 20
HE1 HE3
CW from CT CP = 20 CW to CT
CP = 90 kW/oC CP = 90 kW/oC
Tin = 20 oC CP = 5.7
HE2 HE4 Tout = 57.8 oC
CP = 50
CP = 44.3
% - 15.4 % + 11.2%
As the cooling water is re-used between heat exchangers, the outlet temperature is
increased and flowrate is decreased when maximum re-use design is compared with
the parallel configuration.
Constraints on CW Return Temperature
• CW treatment problem limits the CW return temperature.
T CW Composite Curve T
Maximum Re-use
No Pinch How to
Temperature limitation
design ?
No Re-use
New CW
supply line
Q Q
The design for maximum re-use of cooling water is not enough to complete the
development of the design methodology for cooling water networks.
The purpose of the water supply line targeting is different between water system
design and cooling water network design. Water system focuses on the
minimisation of contaminated water to the environment, which forces the design of
the water network into the minimum consumption of water.
For cooling water networks, there are interactions between the subsystems. The
minimum consumption of cooling water cannot guarantee the optimal design of
cooling water network and therefore the optimal cooling water supply target may be
different from that for maximum reuse.
Moreover, the cooling water system cannot operate beyond a specific return cooling
water temperature, because the hot return cooling water temperature might cause
fouling problem, corrosion or problems with the cooling tower packing. It is
common practice to introduce temperature constraints for the return cooling water.
The cooling water supply line makes no pinch point with the limiting cooling water
composite curve when there is a temperature constraint or maximum re-use supply
target is different from the desired one.
CW Network Design without Pinch
T T Modified
Modify CW composite
Feasible Region
CW
Composite
New CW Pinch
Migration New Pinch
Supply Line
Q Q
Without Pinch With Pinch
“Kuo and Smith’s water main method” is limited and needs to be extended to
design problems without a pinch.
The limiting cooling water profile limits the temperature and flowrate of cooling
water. Any cooling water supply line below the limiting profile will result in a
feasible design.
The cooling water composite curve, which represents the total behavior of the
cooling water network, can be modified in the feasible region, which is enclosed by
the new desired cooling water supply line and the cooling water composite curve.
If the cooling water composite curve can be modified to make a pinch point with
the new desired cooling water supply line, the cooling water network problem can
be changed into a design problem with a pinch.
The “pinch migration” method is used here to convert the cooling water network
into a pinch problem with the new desired supply line.
CW Composite Curve Modification
T
T
1. Heat Load
CW Composite Shift Energy
Penalty
1 Q
2 T
Q 2.Temperature
Shift
Q
Two ways of “pinch migration” could be considered for finding a new pinch.
The first is “heat load shift” in which the cooling water composite curve moves
along the heat load axis.
The second is “temperature shift” in which the cooling water composite curve
moves along the temperature axis.
The “temperature shift” method is preferred because the “heat load shift” method
results in an energy penalty.
Limiting CW Profile Modification
T[_C ]
Feasible
region
for modification
Temperature
Shift Q[kW]
T[_C ] T[_C ]
Original
Increase Original
Modified CP ∆Tshift Modified
∆Tshift
T*
T*
Increase CP
Temperature
Q[kW] Limitation Q[kW]
The “temperature shift” method is chosen to modify the cooling water composite
curve. The next problem is how to find the new pinch and how to modify the
composite curve.
The first stage is to shift the temperature of the limiting cooling water profile for
that amount of temperature shift (which is calculated in the next slide).
It may happen that the modified profile will cross the supply line and we need
another step.
The second stage is to increase the flowrate of the limiting water profile when the
shifted-profile is restricted by temperature limitations. By increasing the cooling
water flowrate, the limiting cooling water profile is modified to satisfy the
temperature limitations.
The new pinch, with the new cooling water supply line, is calculated from a simple
mass balance equation.
The new calculated pinch of 38.5 °C is migrated from original pinch of 40°C. It is
necessary for the individual profiles to apply a “temperature shift” for modification
of the composite curve. Cooling water streams 1, 2 and 3 take part in forming the
original pinch, which means streams 1,2 and 3 are the candidates for “temperature
shift”. The limiting cooling water data for streams 1, 2 and 3 are modified by a
1.5°C temperature shift.
For streams 2 and 3 it is simple to get the new limiting cooling water profile by
shifting the temperature. But for stream 1, we need to find the increase the flowrate
because the 20°C of cooling water supply temperature restricts the temperature shift
of the limiting data.
• Temperature-shifted • New composite curve
limiting CW data
T(_C)
Heat Tcw, in Tcw, out CP Q Temperature-shifted
Exchanger [oC] [oC] [kW/oC] [kW] CW composite
(: modified)
1800 3400
Q(kW)
After modification of each limiting profile, the new limiting cooling water data are
given. For stream 1, the CP is increased from 20 kW/°C to 21.6 kW/°C.
The modified cooling water composite has a pinch with the desired cooling water
supply line. Now the “cooling water main method” can be applied to the cooling
water network design.
CW Network Design without Pinch
• Target temperature = 55 oC
CP =16.1
CP = 21.6
HE1 HE4
CP =5.5
CW from CT CP = 21.6 CW to CT
CP =35.6
The “temperature shift” and “pinch migration” methods enable design with any
target temperature.
Comparison of CW Network Design
• Maximum Re-use
CP = 20
HE1 HE3
CW from CT CP = 20 CW to CT
CP = 90 kW/oC CP = 90 kW/oC
Tin = 20 oC CP = 5.7
HE2 HE4 Tout = 57.8 oC
CP = 50
CP = 21.6
HE1 HE4
CP =5.5
CW from CT CP = 21.6 CW to CT
o
CP = 97.1 kW/ C CP = 97.1 kW/oC
Tin = 20 oC Tout = 55 oC
HE2 HE3
CP = 54.0
CP =35.6
The cooling water network has different design configurations with different target
temperatures. The new design methodology for cooling water networks can achieve
the desired temperature target.
4. Cooling Water System Design
HE 1
HE 2
Cooling Tower is
HE 3
Bottlenecked
CW Network
new HE
• Parallel Design
HE 1
HE 2 CW Flowrate/
Heat Load
Additional CT
HE 3
new HE
Increased
Required ?
10 - 39 Cooling Water System Design
When cooling water networks need to increase the heat load of individual coolers or
to introduce new heat exchangers into an existing system, the cooling water system
can become bottlenecked. As the increase of cooling duty influences the cooling
tower performance and the cooling system has interactions, the cooling water
network should be modified if no penalty is desired to the new system.
When the traditional parallel arrangement is applied with the new operating
conditions, the water flowrate and the heat load of cooling tower consequently
increase. If there are no other design options than parallel arrangements, an
additional cooling tower is needed to satisfy the new bottlenecked conditions
because the existing cooling tower has limitations to the performance and the heat
removal capacity.
• Base Case
Evaporation
15.1 t/h
Air
732.24 t/h 732.24 t/h
TWBT = 23.9 oC CT
TDBT = 29.4 oC
HE 1
28.8 oC
HE 2 43.3 oC
7.6 t/h
HE 3
Blowdown Makeup
22.7 t/h CW Network
10 oC Cycles of
Concentration = 3
The debottlenecking procedure for the cooling water system is illustrated with an
example. The base case has three heat exchangers.
Introduce New Heat Exchanger to CW System
• Limiting Cooling Water Data
Heat Tcw, in Tcw, out CP Q
Exchanger [oC] [oC] [kW/oC] [kW]
1 28.8 37 200 1640.2
The limiting cooling water data is given for the base case. In this example, a new
heat exchanger is introduced into the base case, which makes the cooling water
system bottlenecked.
Parallel Design of CW Network
HE 1
HE 1
HE 2
HE 2
HE 3
HE 3 New
Heat Exchanger HE 4
New outlet conditions for the cooling water are shown when the parallel
arrangement is applied to the design of cooling water network with the new heat
exchanger. The flowrate, temperature and the heat load of the cooling tower are
increased and therefore the cooling tower performance and heat removal can be
influenced.
Results of Parallel Design Method
HE 1
CW from CT HE 2 CW to CT
When the parallel design method is preferred, some checks should be carried out
before targeting of cooling water systems.
First, the cooling water inlet temperature (Tin = 30.4 °C) is hotter than the desired
inlet temperature (28.8 °C) under the parallel arrangement. Second, the cooling heat
load of the HEN (15.6 MW) is bigger than heat removal of the cooling system (14.6
MW).
These results mean that additional cooling plus the current system is needed to cool
the cooling water below the maximum permissible inlet temperature (28.8 °C) and
to remove another 1 MW of cooling duty.
From the previous results, two guidelines for the design of cooling water systems
can be suggested.
The first is that the heat removal of the cooling tower can be increased by changing
the inlet cooling water conditions from high flowrate and low temperature to low
flowrate and high temperature.
The second is the new cooing water design methodology which enables change of
the cooling water network design from a parallel arrangement to a series
arrangement.
By changing from a parallel to a cooling water re-use design, the heat removal of
cooling tower can be increased without any energy penalty to the current system.
Find the Feasible CW Supply Line
T(_C) T(_C)
CW Supply Line
Cooling Water for Maximum Reuse
Composite Curve 51.3 oC 51.3 Same Heat Load
(Q = 15.6 MW)
Feasible
CW Supply
44.1 oC Line
From the limiting cooling water data, the cooling water composite curve can be
constructed. The cooling water network can be changed within a feasible region
which is bounded by the maximum re-use supply line and the parallel design supply
line. The parallel design supply line is a low bound as there is no reuse of cooling
water.
The feasible cooling tower supply line represents the attainable outlet conditions of
cooling water by changing the design configuration.
Effect of CW Inlet Conditions to CT
T(_C)
Same Heat Load of Heat removal of
B cooling water network Case cooling water system
51.3 (Q = 15.6 MW)
Parallel(A) 14.61 MW
Max.
Reuse(B) 15.69 MW
44.1 A
Target 15.60 MW
692.6 1020.9
CP(kW/_C)
As the cooling water inlet conditions to the cooling tower affects the cooling tower
performance, it is necessary to know how the inlet conditions affect the cooling
water system.
The cooling water supply line has the same heat load (15.6 MW) from the
viewpoint of the cooling water network. But the heat removal of the cooling system
is changed with the different inlet conditions of the cooling tower. The heat removal
of the cooling water system is increased as the design configuration changes from
parallel to maximum re-use.
It is obvious that the target supply conditions of the cooling water, which is not
known yet, should lie somewhere on the feasible cooling water supply line.
The lower inlet temperature to the cooling water network, the more heat will be
removed in the cooling tower. So it is not necessary to achieve a temperature lower
than 28.8 °C in this example.
Targeting of CT Supply Conditions
Feasible CW supply line
T(_C)
M
B : Maximum reuse
«
CW Composite curve
«
B: Target condition 50.3 oC
«
BN : Parallel design
A « [no reuse]
Initial Isothermal line of 28.8 Target CW Supply line
condition cooling system CP = 725 kW/oC
outlet temperature
CT water inlet flowrate 15.6
Q(MW)
Target (B*) : CT water inlet temperature = 50.3 oC
CT water inlet CP = 725 kW/oC
× CW inlet temperature to HEN = 28.8 oC (point B*)
The next stage is to find the target supply conditions of the cooling tower.
The feasible cooling water supply line can move from BN to BM. The target
conditions which satisfy the desired temperature to the cooling water network (28.8
°C) can be found by changing the cooling water supply conditions from BN to BM.
The target conditions for the cooling water supply line can be found by using a
cooling system model.
The heat removal of the cooling system is the same as the heat load of cooling water
network at the target conditions (B*) where the inlet temperature to cooling water
network is satisfied. Target conditions are given by the intersection between the
feasible cooling water supply line and the isothermal line of cooling system outlet
temperature.
Apply the Temperature Shift and Pinch Migration
• Find New Pinch • Limiting CW Profile modification
T(_C)
Temperature
T Shift
52.7_C T
∆Tshift 51.8_C
50.3
48_C
T* T* 47.1 _C
The target conditions for debottlenecking have been found and the next stage is to
design the cooling water network with target conditions.
As the new cooling water supply line has no pinch with the limiting composite
curve, the temperature shift and pinch migration methods are applied to this case as
explained in previous section.
The new pinch point is calculated and then the limiting cooling water profile
modification is applied.
CW Network Design
T(_C)
Modified CW
CW Composite Curve Composite
Curve
50.3 oC
CP =104.3
15.6
Q(MW)
CP = 236.1
HE1 HE4
CW from CT CW to CT
CP =200
CP = 68.2
o CP =310 Tout =
CP = 725 kW/ C
Tin = 28.8 oC HE2 HE3 50.3oC
CP =488.9
CP =110.7
The proposed design of cooling water network is shown which satisfies the target
conditions.
Design of New Cooling Water System
HE1 HE4
28.8 oC
9.5 t/h HE2 HE3
Blowdown Makeup
28.3 t/h CW Network
10 oC
The final design can be evolved for network simplicity but this will require a
penalty in the system performance.
But
HE 1 HE 1
HE 2 Cooling Water HE 2
Re-use
Pressure drop increases
• Pump characteristics Flowrate decreases
Efficiency
Flowrate
10 - 50 Cooling Water System Design
5. Heat Load Distribution of Cooling
Water Systems
HE1 HE4
28.8 oC
9.5 t/h HE2 HE3
Blowdown Makeup
28.3 t/h CW Network Cycles of
10 oC Concentration = 3
Re-circulating cooling systems discharge less water to the environment than once-
through cooling systems. However, the re-circulating cooling system has inherent
disadvantages in terms of cooling water treatment problems because re-circulating
systems are susceptible to fouling and corrosion.
Three phenomena are usually considered in cooling systems: scaling, corrosion and
biological fouling. Fouling reduces the heat transfer efficiency and increases
maintenance and operating costs. But fouling is related to temperature.
The debottlenecking procedures of cooling water systems maintain high
temperature and low flowrate of return cooling water to increase the heat removal
capacity of the cooling tower. However, the increase in temperature is not favoured
from the viewpoint of water treatment in cooling systems. So cooling water systems
require temperature limitations when cooling water treatment becomes a problem.
Let us revisit the previous example.
The final design of debottlenecked cooling water systems was 50.3 °C for the return
temperature. This temperature increase may cause water treatment problems.
Effect of Return Temperature Constraint
B : Target condition
«
Temperature « BC : Best
44.1 oC Constraint
«
« BN : Parallel design
A
28.8 CW Supply line Initial
for Parallel Design condition 29.3_C
28.8_C
Q(MW) CT water
15.6 inlet
---- Isothermal line of flowrate
cooling system
outlet temperature
The best CW supply conditons (BC )
1. CW inlet temp. to HEN (29.3 oC) > Desired Temp(B* =28.8 oC)
2. Heat Load of HEN (15.6 MW) > Heat load of Cooling System(15.2 MW)
The previous target temperature (50.3 °C) is now higher than the acceptable
temperature limitation (47 °C), which means that the heat removal of the cooling
system cannot be obtained by changing network design. The flowrate of the cooling
water supply line cannot decrease beyond the 47°C temperature limitation. So the
maximum heat removal of the cooling water system under given conditions occurs
when the target temperature has reached the temperature limitation.
The best cooling water supply conditions are at point BC but these conditions still
cannot satisfy the requirements of the cooling water network. From the cooling
system model, the cooling water inlet temperature (29.3 °C) is higher than desired
inlet temperature (28.8 °C). Furthermore, the heat removal for best conditions (15.2
MW) cannot satisfy the heat load of the cooling water network (15.6 MW).
Other design options should be incorporated along with best cooling water supply
conditions.
New Design Option
• Change of CT operating conditions
Increase the air flowrate of cooling tower
In this example, the cooling water system with return temperature limitation needs
additional cooling. The overall heat load of the cooling water network is too high
for the cooling tower alone. So some amount of heat load should be dissipated by
other heat transfer equipment or by introducing other options.
One design option is changing of cooling tower operating conditions. This can
increase the driving force of the cooling if we increase the air flowrate of the
cooling tower.
The heat load of the cooling tower can be reduced by changing the cooling water
supply conditions, as some amount of heat load is distributed by the other design
options.
If we decrease the temperature or flowrate of the cooling tower inlet conditions, the
heat duty of the existing cooling tower is decreased.
Increase Air Flowrate
Evaporation
18.9 t/h
QR = 15.6 MW
CT
736.9 t/h
28.8 oC 47 oC
CW
Network
9.4 t/h 10 oC
Cycles of
Blowdown Makeup CWN Concentration = 3
Q =15.6 MW
28.3 t/h
The more the driving force of the cooling is increased, the more heat is removed
and the lower the outlet cooling water temperature obtained. As the water flowrate
is fixed for this case, an increase of air flowate is an alternative way to maintain the
driving force for cooling.
The heat duty of the cooling tower is increased by increasing the air flowrate.
Hot Blowdown Extraction
Hot
Blowdown
Evaporation T2 FH
F2
T1
CT F1 T1 = T2
CW F2 = F1 - FH
Network
Cold Makeup
Blowdown
The temperature of the return cooling water does not change after and before hot
blowdown but the flowrate is decreased by the amount of hot blowdown.
That is the clue to find the amount of hot blowdown.
• Find a CT inlet flowrate
CPCT,in = ? Cooling
System TCW,IN = 28.8 oC
TCT,in = 47 oC Model
22.4 1224.5 714.5 14374.2 15598.7 28.8
What is needed next is to find what amount of hot blowdown should be extracted
from cooling system to solve the heat load problem. By using cooling system
model, the amount of supply flowrate of cooling water and hot blowdown flowrate
can be found.
The table shows the iterative procedure to find the target flowrate of hot blowdown.
As the amount of hot blowdown is increased, the heat duty of the cooling tower is
decreased. At the target conditions, the cooling water system achieves the
requirements of temperature and heat load.
• Design with hot blowdown extraction
Hot HB
The final design with hot blowdown extraction is shown. The return flowrate of
cooling water changes from 736.9 t/h to 714.5 t/h by hot blowdown extraction. In
this case, the required hot blowdown exceeds the original cold blowdown flowrate,
which results in an increase of make-up water and decrease of cycles of
concentration.
Introduction of Air Heat Exchanger
T2
QAHE
Evaporation
F2 ∆T = T1-T2
T1
CT AHE F1
F2 = F1
CW
T1 > T2
Cold Network
Blowdown
Cycles of Concentration = 3
Makeup
For the hot blowdown extraction case, the heat load of the cooling tower is
distributed by reducing the flowrate of cooling water supply. There is one other
variable to be considered for heat load distribution. If the temperature of the hot
return cooling water can be lower, the heat duty of the cooling tower would also be
reduced.
To decrease the return cooling water temperature, air heat exchangers may be
installed between the cooling tower and the cooling water network. The flowrate of
return hot cooling water does not change after and before the air heat exchanger but
the temperature is decreased by the amount of heat removed by air heat exchanger.
That is the clue to find the amount of heat removal by air heat exchanger.
• Find a CT inlet temperature
3.9 3342.5 43.1 12257.3 15598.8 28.8
Now consider how to find the amount of heat to be removed by the air heat
exchanger. By using the cooling system model, the amount of heat load in the air
heat exchanger and in the cooling tower are targeted.
The table shows the iterative procedure to find the target heat load of the air heat
exchanger. As the amount of heat removed by the air heat exchanger increases, the
heat duty of the cooling tower decreases. At the target conditions, the cooling water
system achieves the requirements of temperature and heat load.
• Design with Air Heat Exchanger
QAHE =3.3 MW
Evaporation
15.0 t/h
43.1 oC
QR =12.3 MW CT AHE
736.9 t/h
47 oC
28.8 oC CW
Cold Network
Blowdown
7.5 t/h
Makeup QCWN =15.6 MW
22.5 t/h
The final design with an air heat exchanger is shown. The return temperature of the
cooling water changes from 47 °C to 43.1 °C by the air heat exchanger. As the air
heat exchanger influences only temperature of the hot return cooling water in this
case, there is no change in the cycles of concentration.
Comparision of Heat Load Distribution
The hot blowdown is a more effective method than the air heat exchanger from the
viewpoint of energy. The capacity of cooling tower is used more effectively in the
hot blowdown case. But hot blowdown has a penalty of make-up water increase and
thermal pollution to the environment.
6. Conclusions
Combined
water and energy
pinch method
HE4
New CW Network
In this presentation, a robust cooling tower model has been developed and a new
cooling water network design methodology has been suggested. The design of
cooling water systems has been investigated.