CDB3062

Chemical Engineering Laboratory III

January 2020

LONG REPORT

Title : EXPERIMENT 1: HEAT EXCHANGER

Group : 4G

Group members : ABDUL CARIM COMANDA MOMADE        25563      

FARRAH ATIKAH ABD RAHMAN       25312      

SITI NUR FATIHAH BINTI ADNAN       25398      

NASRIN FATINI BINTI AHMAD FAUZI       17005968      

MUHAMMAD ANWAR BIN HAIZAL       25202      

Coordinator : Dr. Serene Lock Sow Mun

Lab Demonstrator : Mr. Afif Asyraf

Date of Submission : 12 March 2020

1
TABLE OF CONTENTS

1.0 ABSTRACT .......................................................................................................................................... 3

2.0 LITERATURE REVIEW ......................................................................................................................... 4

3.0 PROBLEM STATEMENT AND OBJECTIVE ........................................................................................... 7

4.0 METHODOLOGY ................................................................................................................................ 8

5.0 RESULT AND DISCUSSION ............................................................................................................... 10

6.0 CONCLUSION ................................................................................................................................... 16

7.0 REFERENCES .................................................................................................................................... 17

2
1.0 ABSTRACT

This paper is discussing about the performance of temperature control in heat exchanger such
as P, PI and PID controller. The shell and tube heat exchanger are the most common type of
heat exchanger in petrochemical industries due to it suitability for low- and high-pressure
applications. One fluid flow through the shell surrounding the tubes to transfer heat between
two fluids. Heat is transferred from the tube fluid to shell fluid to remove heat, or from the shell
fluid to the tube fluid to heat the material inside. Hence the main objective of this experiment
is to identify the characteristic of Proportional (P), Proportional and Integral (PI) and
Proportional, Integral and Derivative (PID) Controller on a temperature process control loop.
Besides that, also to determine the correct tuning data for the particular controller because each
of the process have different tuning value. The controller received signal from Resistance
Temperature Detector (RTD when disturbance is introduced to the new setpoint 40℃ and the
response of P, PI, PID controller respectively recorded in control chart. The characteristics of
P, PI and PID are analyzed and discussed in the report. Based on this experiment, controllers
performance are evaluated using its responses to the process such as overshoot, Rise time,
Settling time and the Steady state response. From discussion, it can be justified that the 30s
value of Integral used for this parameter set has the shortest time response compared to other
parameters which eventually make PI controller with second parameter is the most suitable
heat exchanger controller.

3
2.0 LITERATURE REVIEW
According Çengel, Ghajar, & Kanoglu (2011), Heat exchanger are devices that
facilitate the exchange of heat between two fluids that are at the different temperature while
keeping them from mixing from each other. Also, heat exchanger is used to transfer thermal
energy (enthalpy) between two or more fluids, between a solid surface and a fluid, or between
solid particulates and a fluid at different temperature and in thermal contact (Nimankar &
Dahake, 2016). Heat exchanger are extensively used in food processing industry, dairy
industry, biochemical processing, pharmaceuticals, chemical plants, and petroleum plant to
name a few (Agarwal, Sikand, & Shanthi V., 2014)

Heat exchangers are classified according to; Contact Types, Surface Compactness,
Number of Fluids, Flow Arrangement, and Construction Features. Based on Shah & Sekulić
(2003) in many heat exchangers the fluids are separated by a heat transfer surface and ideally
the do not mix are referred as direct transfer type or recuperators. In contrast, exchanger in
which there is intermittent heat exchange between the hot and cold fluid via thermal energy
storage and release through the surface or matric are referred as indirect transfer type or
regenerators. This type of heat exchanger usually has fluid leakage from one fluid stream to
the other due to pressure difference and matrix rotation or valve switching. During designing
of heat exchanger, construction features are taken into consideration. Four major construction
type are tubular, plate type, extended surface and regenerative exchangers.

As agreed by Dubey, Verma, Verma, & Srivastava (2014) and Çengel, Ghajar, &
Kanoglu (2011) the most common type of heat exchanger in oil refineries and industrial
processes and suit for high pressure application is shell and tube heat exchanger. This heat
exchanger consists of a vessel with different sizes number of tubes inside and heat transfer
between these tubes together and with the shell side through tube walls. (Abd, Kareem, & Naji,
2018). As its name implies, this type of heat exchanger consists of a shell (a large pressure
vessel) with a bundle of tubes inside it. One fluid runs through the tubes, and another fluid
flows over the tubes (through the shell) to transfer heat between the two fluids (Dubey, Verma,
Verma, & Srivastava, 2014). Round tube in various shape are used in heat exchanger and the
most common are tube bundles with straight and U-tubes (Shah & Sekulić, 2003) while for the
shell part, shell is made from a circular pipe if the shell diameter is less than about 0.6m (2 ft)
and is made from a metal plate rolled and welded longitudinally for shell diameters greater
than 0.6m (2 ft). The E shell is the most common, due to its low cost and simplicity, and has
the highest log-mean temperature-difference correction factor F (Nimankar & Dahake, 2016).

4
Baffles are commonly placed in the shell to force the shell-side fluid to flow across the shell to
enhance heat transfer and to maintain uniform spacing between the tubes.

Figure 2.1 The schematic of a shell and tube heat exchanger (Singh & Kumar, 2014)
Based on Figure 2.2, The input cold water is supplied from the overheat tank to the
shell side of the heat exchanger. Steam is supplied to the tube side of the heat exchanger. RTD
is used to measure the output temperature of the heat exchanger and is connected to the
transmitter. The transmitter helps to reduce the noise in measurement. The data from the
transmitter is updated in the PC based controller using a data acquisition (DAQ) device. The
PC based controller processes the error signal and computes the appropriate control signal. The
controller unit sends the corresponding control signal to current to pressure converter via
another DAQ device. The current to pressure converter converts the current output of PC based
controller to appropriate pressure signal so that the steam valve can be actuated in a proper
manner (Padhee, 2014)

Figure 2.2 The schematic

diagram of temperature control heat exchanger (Padhee, 2014)

5
 Heat transfer in heat exchanger is typically efficient but controlling the temperature of
the fluid being heated at specific and stable setpoint can be challenging. The major disturbance
that can affect the process fluid outlet temperature is changes in flow rate of fluid, fluid inlet
temperature and change in steam flow rate (Yehia, 2016). The control objective is to maintain
the process fluid outlet temperatre. In most of the papers the process is controlled by traditional
feedback control or a feed forward control is in which, it is observed that each controller
respond differently at various conditions Hence it is necessary to find an optimal solution for
effectively controlling the process at various process conditions and also during sudden
disturbances. A controllers performance are evaluated using its responses to the process such
as overshoot, Rise time, Settling time and the Steady state response. (Saranya,
Thirumarumurugan, & Sivakumar, 2017)

6
3.0 PROBLEM STATEMENT AND OBJECTIVE

Process industries generate large amount of heat that needs to be transferred. Thus, heat
exchanger is one of the important unit operation in plant as it provide a medium for heat
ultilization between the two fluid which are hot stream and cold stream. For the precise heat
transfer,the temperature and the flowrate of hot fluid and cold fluid passing through the heat
exchanger should be monitored and controlled efficeintly. However,there are difficulties to
design a controller with suitable value of Proportional(P), Integral(I) and Derivative (D) to
reach certain setpoint as it involve load disturbances in form changing in temperature which
are flucntuating througout the process. Despite of that, three paramteres of PID controller
mostly tuned by emprical tuning like ziegler-Nicholas but it not always suitable for every kind
of process dynamics as certain process have different long dead time and oscillatory behaviour.
As to solve the problem , closed loop with the identification the process variable, manipulated
variable and controlled variable needed to be done. Hence,corresponding controller graph that
are generated from sets of values for P,PI and PID values were studied to determine which
controller is suitable to be implimented for heat exchanger control system.

7
4.0 METHODOLOGY

Startup Procedure
1. HV231, HV234, HV235, HV236,HV241,HV243,HV244,HV245 and HV247 ensured
to be opened while HV232,HV233,HV240,HV242,HV248,HV250,HV251,HV252,
HV253,HV237,HV238 and HV239 are closed.
2. Cold Water Tank T-201 and Hot Water Tank T-203 are filled with water respectively.
3. The instrument air regulator RG1 was set to 2.8 bar while main panel was switched on.
4. After Hot Water Tank T-203 reached 65ºC, pump P211 and P213 was switched on and
flowrate meter, FI201 was set to the 10 LPM using HV240.
5. Throughout the experiment , the temperature Hot Water Tank T-203 and level are
monitored and maintained between 75ºC to 88ºC and 75% of the tank respectively.

Proportional, Integral and Derivative Heat Exchanger Control

1. The values of P=100, I=60 and D=60 were entered to the control system.
2. The set point (SV) was set to 30ºC, the control was put in manual mode and the valve
opening (MV) was adjusted accordingly for the temperature of cold water (PV) reach
the setpoint.
3. As the PV has been reached, new set point was set to 40ºC and the controller was set
to automatic mode.
4. The time to reach the setpoint and the controller graph were being recorded.
5. Repeat steps 1 to 4 for the experiment for the values P=100, I= 60 and D=300 .

Proportional and Integral Heat Exchanger Control

1. The values of P=100, I=120 and D=0 were entered to the control system.
2. The set point (SV) was set to 30ºC, the control was put in manual mode and the valve
opening (MV) was adjusted accordingly for the temperature of cold water (PV) reach
the setpoint.
3. As the PV has been reached, new set point was set to 40ºC and the controller was set
to automatic mode.
4. The time to reach the setpoint and the controller graph were being recorded.
5. Repeat steps 1 to 4 for the experiment for the values P=100, I= 30 and D=0

8
Proportional only Heat Exchanger Control
1. The values of P=100, I=9999 and D=0 were entered to the control system.
2. The set point (SV) was set to 30ºC, the control was put in manual mode and the valve
opening (MV) was adjusted accordingly for the temperature of cold water (PV) reach
the setpoint.
3. As the PV has been reached, new set point was set to 40ºC and the controller was set
to automatic mode.
4. The time to reach the setpoint and the controller graph were being recorded.

9
5.0 RESULT AND DISCUSSION

5.1 PROPORTIONAL ONLY HEAT EXCHANGER CONTROL

For the first part of the experiment which was conducted to determine the response of
the P control using only PB (Proportional Bandwidth) for tuning purposes. The PB values set
to 100%, Ti to 9999s and Td to 0s. Since proportional action produces offset which basically
means that it will calculate the errors produced by subtracting the set point value and the
respond value. When using only P-Controller, the system will not reach the set point, or it will
take very long time to reach the desired set point with right tuning data.

Figure 5.1: P control response with set point of 40˚C and PB value of 100

According to Figure 5.1, obtained from the experiment, after setting the cold-water
outlet at 30˚C and stabilizing it, a step change was introduced and a set point of 40˚C was set.
After approximately 13 minutes the process showed a very slow and insignificant response to
the change in the system. This slow response just shows how slow a P-controller is when used
alone in a system. It is possible that the system reaches the set point, but that would take a very
long time and would possibly require a different tuning parameter. Moreover, small oscillations
can be observed from the figure above which suggests that the stability of the controller is not
the best. The poor result obtained are due to controller errors where theoretically it will
calculate offset value.

10
5.2 PROPORTIONAL AND INTERGRAL HEAT EXCHANGER CONTROL

The second part of this experiment is to test the effectiveness of PI controller using different
set of parameter values. The first set integral value is P=100, I=120s and D=0s while for the
second parameter was set to P=100,I=30s and D=0s. The different of parameters will resulted
in different output of the graph and different time response for each parameters. At first, the
set point was introduced at 30°C and after achive stability new setpoint is introduced which is
40°C whereby this apply for both parameter set.

Figure 5.2.1: Graph of tuning PI controller using 120 value of Integral

Based on the graph 5.2.1,for the first parameter set for PI controller which are P=100,
I=120 and D=0, it can be observed from right to the left side of the graph, as the setpoint was
set to 30 °C, the flowrate of hot water would approach the setpoint and after some time, the
new setpoint was set as 40°C to observe the process response of the particular parameter set
for PI controller. The output line produced in the graph has significantly approched the value
of setpoint which proved that the offset has been eliminated by introducing Integral controller.
Theoretically, Integral controller is used to eliminate or reduce offset also known as steady
state error which produced by uncorrected error that has accumulated by Proportional
controller by continuously integration the area under the error curve. As it can be seen on the
graph, the output has low steepness in achieveing the new setpoint that resulted in longer
settling time which is 20.47 minutes. This can justified that the 120 value of Integral used for
this parameter set is not suitable as heat exchanger controller.

11
 Figure 5.2.2: Graph of tuning PI controller using 30 value of Integral
Based on the graph 5.2.2, for the second parameter set for PI controller which are
P=100, I=30 and D=0,from the right side of the graph, the setpoint of 30 °C was introduced
,and the flowrate of hot water reduced to approach the setpoint and towards the left side, the
new setpoint was set as 40°C to observe the process response of the particular parameter set
for PI controller. The offset has been eliminated as the output line produced in the graph has
significantly approched the value of setpoint. Theoretically, Integral controller is used to
eliminate or reduce offset also known as steady state error which produced by uncorrected error
that has accumulated by Proportional controller by continuously integrating the area under the
error curve. However, the contrast of using the PI controller without introducing Derivative
controller, the overshoot can be observed from the output graph, this is because PI controller
itself has no ability to predict future error as fulfilled by Derivative controlller. Nevertheless,
since the overshoot produce is very small, it can be neglected as the defect since the graph
showed the output line for Integral value of 30s is more steeper compared to previous PI
controller parameter in achieveing the new setpoint that resulted in shorter time response which
is 13.27 minutes. Hence, the shorter time response produced, the more optimum and favourable
system of the controller used.

12
5.3 PROPORTIONAL, INTEGRAL AND DERIVATIVES HEAT EXCHANGER
CONTROL
PID controllers was used in this experiment by having two different set of parameters
of P, I and D. The system was set to 30℃ at first and new setpoint was introduced which is
40℃. The time taken for the system to reach its new setpoint from 30℃ to 40℃ was recorded
as it would determine how fast the response to correct the system to any changes happen in the
real situation. The faster the time taken to reach the new setpoint, the better the control system
of the process. Since the PID control is using both Integral and Derivatives control, the function
of PID control should be as below:

Where the functions include both the integral time constant and the derivatives time constant.
This function later can be changed to the transfer function to the respect of its error, steady
state process variables or even to its disturbance by using Laplace Transforms formulas. As the
integral time constant introduced is supposed to eliminate the offset by the P control, the
derivatives time constant is supposed to help fasten the process as the integral time will cost
the process to have some delay in it since it will integrate the process in the function of time.
The PID controller is divided by two common types which is the ideal PID controllers and the
real/actual PID controllers. The formula for the transfer function of the ideal controller is given
by:

The ideal PID-controller is not suitable for direct field interaction; therefore, it is called the
non-interactive PID-controller, where to cater the other type of controller, the real PID
controller should be used. Meanwhile the real PID controller formula is given by:

The real PID-controller is especially designed for direct field interaction and is
therefore called the interactive PID-controller. Due to internal filtering in the derivative block,
the effects of electrical noise on the PV input are greatly reduced.

13
 As mentioned above, the experiment used two set of different parameters for P,I and D
values. The graph below is the results of the experiments using PID control using the first set
of parameters of P,I and D:

Figure 5.3.1: PID control graph using first set parameters

The first parameters of PID controllers used is P=100, I=60 and D=60. The time taken
for the system to reach 40℃ was recorded as 16.21 minutes. As can be seen from the graph,
the steepness of the graph is low. It shows that the parameters used for this experiment will
produced slow time taken for the process to correct the system which is not favourable
compared to the fast response from the system. Although the system did not oscillate, the
system shows some overshoot when it reaches the setpoint but later being corrected by the
controller to have the temperature of the outlet product tank to be the same as the setpoint
which is the 40℃. The overshoot is because this heat exchanger involved the temperature
control which considered as slow response control where the systems requires some time to
correct the temperature for it to be steady at the setpoint as it reached the setpoint. The graph
below is the results of PID control using the second set of parameters for different P,I and D
values:

14
 Figure 5.321: PID control graph using second set parameters

The second set of parameters used in this PID controller is P=100, I=60 and D=300.
There are obvious change in the D value compared to the first parameter used where the D
value changed from 60 to 300. Theoretically, the derivative term assesses “how fast” error in
the process is changing. As the rate of error either increases or decreases so too does the size
of the derivative response (Inc., 2018). However, the result of this experiment differs from the
theoretical part as the time taken for the process to reach 40℃ is slower compared to the first
parameter of D value which is 60. This might because of the human time response error when
the students recording the time taken of the process to reach its setpoint. Therefore, the time
recorded contains such error. Since PI control produce faster time take for the system to
response, PID control is not suitable to be used for this heat exchanger. The optimum controller
should produce the fastest time taken for the system to meet its setpoint.

5.4 SELECTING BEST CONTROLLER FOR HEAT EXCHANGER.

Therefore, from discussion of other controller paramater, it can be justified that the 30 value
of Integral used for this parameter set has the shortest time response compared to other
parameters which eventually make PI controller with second parameter is the most suitable
heat exchanger controller. Although theoretically, heat exchager is temperature driven control
process which is a slow process, it is suitable to use PID controller. However, since there is no
fixed controller for every process and there also might happened because of human time
response error when the students recording the time taken of the process to reach its setpoint,
hence experimentally result in PI controller with second parameter to produce fastest response
time.

15
6.0 CONCLUSION

After the experiment had been carried out , it can be said that the objectives of the
experiment have been achieved as all the characteristics of Proportional only control
(P),Proportional Band and Integral (I) and Proportional Band ,Integral and Derivative Action
(PID) had been determined and discussed thoroughly.

For the first part of the experiment which is used Proportional only Control(P), it was
found that the controller exhibit offset and significantly slow response time even after 13
minutes toward setpoint as it is because the Proportional control just detect the error produce
by system only but poor in induce correction mechanism. While for the second part of the
experiment which is Proportional Band and Integral (PI) for both sets of values, it is found that
both graph , offset produced by Proportional (P) was eliminated by the value of Integral
inserted as it undergoes continuous integration of the area under the curve lead to the better
converging behavior towards setpoint. Despite of the second sets of value of PI has fastest
settling time, the graph shows overshoot behavior as the value of Derivative does not include
in this part of the experiment. For the third part of the experiment which is Proportional Band,
Integral and Derivative, it was observed that the graph controller significantly converging
without oscillatory behavior towards setpoint as it eliminates offset and there is overshoot
produced by the system but corrected by the Derivative Action (D). However, both PID
controller display longer settling time than PI controller as it is the result from the disturbance
produced which is the temperature of hot water fluctuating through out the experiment.

In conclusion, PI controller is the best choice for this experiment as it eliminates offset,
fastest settling time and less overshoot. However, for the future preferences and wide
application in industry, PID should be favorable as eliminate offset, fast settling time and less
overshoot in theory but the value of PID needed to be further tuning appropriately to produce
robust performance in control system for heat exchanger

16
7.0 REFERENCES

1. Nimankar, S. J., & Dahake, S. K. (2016). Review of Heat Exchangers. Global Journal
of Engineering Science and Researchers, 3(12), 81–92.
2. Çengel Yunus A., Ghajar, A. J., & Kanoglu, M. (2011). Heat and mass transfer:
fundamentals and applications. New York: McGraw Hill Higher Education.
3. Shah, R. K., & Sekulić, D. P. (2003). Fundamental of Heat Exchanger Design.
Hoboken, New Jersey: John Wiley & Sons, Inc.
4. Agarwal, P. V., Sikand, A. V., & Shanthi V. (2014). Application of Heat Exchangers
In Bioprocess Industry: A Review. International Journal of Pharmacy and
Pharmaceutical Sciences, 6(1), 24–28.
5. Dubey, V. V. P., Verma, R. R., Verma, P. S., & Srivastava, A. (2014). Performance
Analysis of Shell & Tube Type Heat Exchanger under the Effect of Varied Operating
Conditions. IOSR Journal of Mechanical and Civil Engineering, 11(3), 08–17. doi:
10.9790/1684-11360817
6. Abd, A. A., Kareem, M. Q., & Naji, S. Z. (2018). Performance analysis of shell and
tube heat exchanger: Parametric study. Case Studies in Thermal Engineering, 12, 563–
568.
7. Singh, G., & Kumar, H. (2014). Computational Fluid Dynamics Analysis of Shell and
Tube Heat Exchanger. Journal of Civil Engineering and Environmental
Technology, 1(3), 66–70.
8. Yehia, S. (2016, January 5). Applying heat exchanger control strategies. Retrieved from
https://www.controleng.com/articles/applying-heat-exchanger-control-strategies/
9. Saranya, S. N., Thirumarumurugan, M., & Sivakumar, V. M. (2017). An Optimal
Analysis of Controller Strategies for Different Heat Exchangers – A Review. Middle-
East Journal of Scientific Research, 25(4), 761–775.
10. Padhee, S. (2014). Controller Design for Temperature Control of Heat Exchanger
System: Simulation Studies. WSEAS TRANSACTIONS on SYSTEMS and
CONTROL, 9, 485–491.

17