Heat Exchanger Network Design using 
Pinch Analysis
Submitted to the Training In-charge, ONGC Hazira, in partial fulfilment of Summer Technical 
Training, 2014. 
K Vivek Varkey 
IIT Hyderabad
 Acknowledgements: 
We would like to thank our mentor for providing motivation and the required data in full 
detail, due to which the process was highly hassle-free. Also we would like to convey 
gratitude to the ONGC Hazira administration for arranging the training and allowing us to 
pursue this project.
 Introduction 
Any process is an energy sink, demanding high amount of heat addition or removal, that ask for individual high 
costs, especially in a large process plant. A process may employ numerous heat exchangers to bring the process 
stream at required temperatures at the various stages. Operation of these all heat exchangers with their individual 
cooling utilities and heating utilities demand a high operating cost on a daily basis. 
It is hence beneficial to strategize the heat exchanger network layout so as to couple the hot and cold process 
streams, so that minimum heat duty is to be supplied or removed using the external utilities. 
However design of an optimal heat exchanger network (HEN) so as to employ minimum operating costs keeping in 
mind the installation costs are also not out of proportion, is a complicated method. Hence several algorithms and 
famous approaches exist. 
This project employs the PINCH ANALYSIS of the energy flow to determine the optimal HEN structure for the CFU in 
the ONGC Hazira plant. The report does not deal with the derivation of the pinch technique, as several can be found 
in literature. 
The data are taken from the design report existing in the plant. The particular case deals with an inlet temperature 
of 33oC and inlet pressure of 93Kg/cm2. 
The heat duties are calculated in an ideal scenario, from the material balances, temperatures and enthalpy data 
taken from the web. 
We deal with the 5 heat exchangers namely, E701, E702, E703, E705 and E706. 
E704 is not dealt with for reasons of unavailability of data.
S 
T 
R 
I 
P 
P 
E 
R 
L 
P 
G 
Reflux 
Drum 
Preheater 
Off Gas 
Cooler 
LPG Condenser 
NGL 
Cooler 
Reboiler 
Feed 
Off Gas to 
GSU 
LPG to 
CW 
Naptha to 
storage 
 Process Flow Diagram 
Vapour 
Condensate 
Feed from Slug Catcher at 33oC and 93Kg/cm2 
PFD shown only for elements associated with this project. 
1 
5 
7 
3 
4 
2 
8 
6
 Material Balance 
Component 1 2 3 (liquid) 3(vapor) 4(liquid) 4(vapor) 5 6 7 8 
N2 0.02 0.01 0 0.01 0 0 0 0 0 0.02 
H2S 1.36 0.13 0.75 0.48 0 0 0 0 1.23 1.36 
CO2 40.46 7.15 12.11 21.2 0 0 0 0 33.31 40.46 
C1 266.56 72.14 40.82 153.6 0 0 0 0 194.42 266.56 
C2 68.51 7.5 34.02 26.99 0 0 0 0 61.01 68.51 
C3 90.32 4.76 68.18 17.38 12.69 45.56 58.25 0 27.31 32.07 
iC4 28.99 0.9 25.22 2.87 7.18 16.42 23.6 0 4.49 5.39 
nC4 45.74 1.13 41.21 3.4 13.32 25.95 39.16 0.11 5.34 6.47 
iC5 19.49 0.27 18.55 0.67 8.22 9.93 0.11 18.04 1.07 1.34 
nC5 22.4 0.26 21.54 0.6 10.14 11.04 0.01 21.17 0.96 1.22 
C6 32.03 0.18 31.54 0.31 19.39 11.95 0 31.34 0.51 0.69 
C7 38.59 0.11 38.35 0.13 28.22 10.03 0 38.25 0.23 0.34 
C8 47.64 0.07 47.51 0.06 39.45 8.02 0 47.47 0.1 0.17 
C9 25.29 0.02 25.26 0.01 22.58 2.67 0 25.25 0.02 0.04 
C10 15.5 0.01 15.49 0 14.5 0.99 0 15.49 0 0.01 
C11 17.73 0 17.73 0 17.02 0.71 0 17.73 0 0 
C12+ 7.96 0 7.96 0 7.77 0.19 0 7.96 0 0 
All flows are in Kmol/hr.
 Cp values for the required components
The values of the components in the CFU unit is 
obtained. Following which the respective mole 
fraction in each stream is also obtained. These two 
data are used simultaneously and the weighted mean 
is to obtained the overall specific heat value.
 PROCEDURE 
The method used is called Temperature Interval Method of Pinch Analysis 
1. The Cp value obtained is multiplied by the flow rate to obtain the heat capacity flow 
rate (C). 
2. Now we create the following table for the further calculation. 
The approach temperature is taken to be 10 degree Celsius. This is subtracted from 
both the inlet and outlet temperatures of the hot stream for the purpose of 
calculation, so as to obtain the pinch where the net heat exchange is supposed to 
be 0. this is put in is Tout* and Tin* in the tables.
E-702 (Stipper Reboiler) C2 
Assumptions: 
• Stripper reboiler operates at constant 156oC, and main heat consumption is to cause phase change. 
• Stripper column bottoms is saturated liquid and the outlet composition is same as the feed to the LPG column. 
• Heat duty is calculated by determining the amount of latent heat required. 
T = 156oc 
M = 344 Kmol/hr 
Q = - 959.03 KW 
E-703 (LPG Condenser) H1 
Assumptions: 
• Constant operation temperature of 43oC 
• Feed is saturated vapour from the LPG column top and outlest stream is saturated liquid. 
• Heat duty is calculated from the latent heat calculations. 
T = 43oC 
M = 121.13 Kmol/hr 
Q = + 627.55 KW
E-705 (NGL Cooler) H2 
Assumptions: 
• The inlet and outlet temperatures are respectively 175oC and 45oC. 
• The composition is known from the mass balance. 
• No phase change. 
Tf = 45oC 
Ti = 175oC 
Cp = .244 KJ/mol K 
M = 222.81 Kmol/hr 
M*Cp = 15.10 KW/K 
Q = + 1963.0 KW
E-706 (Off Gas Cooler) H3 
Assumptions: 
• Inlet and outlet temperatures are respectively 125oC and 40oC. 
• No phase change. 
• Composition is given in the material balance. 
Tf = 40oC 
Ti = 125oC 
Cp = 0.044 KJ/mol K 
M = 424.65 Kmol/hr 
M*Cp = 5.20 KW/K 
Q = + 442 KW 
Hence summing all the enthalpy requirements of the streams we get, 
C1+C2+H1+H2+H3 = + 1801.52 KW 
Hence the Q min, req is + 1801.52 KW that must the removed somewhere in the process using cooling 
utilities
 PROCEDURE 
3. Now we draw a diagram where we label the adjusted temperatures in order from 
coldest to hottest and we draw arrows to designate streams overlapping these 
temperature intervals.
 PROCEDURE 
4. Now for each interval in the previously drawn figure we find the enthalpy of each 
intervals. This is obtained by adding the heat capacity flow rates ( for hot streams it 
is considered positive and negative for the cold stream ). This is then multiplied by 
the temperature interval for which we are calculating and the values are written 
right next to the respective intervals in the figure labelled under Q. 
5. These values are then added cumulatively from the top to the bottom ( higher 
temperature to colder temperature ). The obtained data is then written in a new 
column Qres. 
6. The least value in the Qres column is called the 
pinch value and the temperature of that interval is 
called the pinch temperature. 
7. The pinch value is then added from the beginning 
of the values in Qres. It is observed that it becomes 
0 at the pinch point. 
8. The pinch value is the hot utility added and the 
final value obtained ( in this case 2005.55 ) becomes 
the cold utility required.
Hence we determine our pinch temperature as 115oC for the Cold streams and 125oC for 
the Hot streams. 
Also MER (minimum energy requirement) targets: 
• Qhot utility = -204.43 KW (to be added by steam) 
• Qcold utility = 2005.55 KW (to be removed by cooling water) 
As per the method the hot utility can only be used above pinch, and cold utility can only 
be used below pinch. 
Before we start coupling a hot stream with a cold stream we should keep one thing in 
mind. Let the specific heat flow rate of the hot stream be Ch and that of the cold stream 
be Cc . If we are trying to couple in the hot side of the pinch it has to be made sure that 
Cc > Ch of the respective streams. Similarly when we are trying to couple in the cold side of 
pinch it should be taken care that Ch > Cc . Otherwise the stream coupling will become 
infeasible. 
In the next page we draw a pinch decomposition of the streams and determine an optimal 
heat exchanger network.
PROOF 
Let us assume a counter current pair of hot and cold stream. 
Thi and Tho are respectively the inlet and outlet temperature of the hot stream. 
Tci and Tco are respectively the inlet and outlet temperature of the cold stream. 
ΔT1 and ΔT2 are respectively the difference between hot inlet and cold inlet and hot inlet 
and cold outlet. 
Q is the energy exchanged 
Ch Cc are respectively the specific heat flow rates for the hot and cold streams.
PROOF 
Q = Ch * [ Thi - Tho ] 
Q = Cc * [ Tco - Tci ] 
After rearranging we get, 
Thi - Tho = Q/Ch 
Tco - Tci = Q/Cc 
now we subtract the equations to get 
ΔT2 - ΔT1 = Q*[Cc - Ch ]/Cc Ch
PROOF 
HOT SIDE OF THE PINCH 
ΔT1 = ΔTMIN 
Hence, 
ΔT2 = ΔTmin + Q*[Cc - Ch ]/Cc Ch 
now, ΔT2 has to be greater than ΔTmin 
hence, 
Cc > Ch 
COLD SIDE OF THE PINCH 
ΔT2 = ΔT min 
Hence, 
ΔT1 = ΔTmin - Q*[Cc - Ch ]/Cc Ch 
now ΔT1 has to be greater than ΔTmin 
hence, 
Ch > Ch
In the next page we draw a pinch decomposition 
of the streams and determine an optimal heat 
exchanger network.
H1 
H2 
H3 
C1 
C2 
175 125 
156 156 
125 45 
125 
43 
43 33 
40 
43 
P 
I 
N 
C 
H 
H=204.03 
H=755 
H=272 
H= 627.55 
H=936 
H=442 
Heating Utility 
Heat Exchanger 
Cooling Utility 
Heat loads of Exchangers mentioned are in 
KW. 
Temperature is in Celsius 
Above pinch Below pinch
The above the diagram can be explained by taking an example from the actual streams. 
So in the hot side of the pinch we see a hot stream E-705 going from 175˚C to 125 ˚C with 
a heat capacity flow rate of 15.1 KW/K. this would give out a heat of 
15.1*[175-125] KW = 755 KW 
Now, when we couple this hot stream with a cold stream , E-702, this 755 KW of energy 
can be utilised by the cold stream which actually requires energy of 959.03 KW. For the 
excess 204.03 KW, that is required for the cold stream, we provide it with a hot utility of 
the same amount. 
Similarly we obtain the other hot and cold utilities required.
 Results and Discussions: 
The exchanger network drawn allows for the MER targets . The streams are coupled and 
utilities are distributed keeping in mind the temperature interval of 10, and the 
subsequent corollary that heating utilities can only be used above pinch and cooling 
utilities can only be used below pinch. 
The main purpose of this design is to allow for energy recycle, a term synonymous with 
the modern era and the global challenges we face. Though in this process it is not 
significant, some processes can self sustain themselves, that is we can achieve all 
temperature targets using very little external utilities. This leads to economic benefits and 
environmental benefits. The temperature interval method and pinch analysis as shown 
here can be utilised for all processes for energy recycle. The CFU here is only one example. 
The project has several assumptions of ideality and some figures are rounded off to allow 
for smooth calculations. Hence for application of the process, this is a rough schematic, 
more accurate measurements, considerations and calculations are required. Also 
installation costs and restrictions must be considered when determining heat exchanger 
area. 
As per the calculation shown, 
• 204.03 kW hot utilities 
• 2005.55 kW cold utilities 
is required.
 Results and Discussions: 
Current energy requirement 
Heating utilities 
QE-701 + QE-702 
= 27.2*[33-43] + [-959.03] 
= 1231.03 kW 
Cooling utilities 
QE-7.03 + QE-705 + QE-706 
= 627.55 + 15.1*[175-45] + 5.2*[125-40] 
= 3032.55 kW 
Therefore the energy that can be saved in terms of percentage :- 
• 83.4% for heating utilities 
• 33.8% for cooling utilities
 Comments: 
Upon supervision by the mentor the following drawbacks were observed :- 
1. The NGL cooler is not always in functioning. It is used generally only during abnormal 
functioning of C-702 
2. In this project it was taken into assumption that everything in reboiler vaporised at 
100%. However, this is not the case. 
3. E-704 was not used in the calculation.
 References: 
www.engineeringtoolbox.xom 
www.wikipedia.com 
www.cheresources.com 
Process and Product Design (Seeder) 
Nptel open courseware

Pinch analysis technique to optimize heat exchanger

  • 1.
    Heat Exchanger NetworkDesign using Pinch Analysis
  • 2.
    Submitted to theTraining In-charge, ONGC Hazira, in partial fulfilment of Summer Technical Training, 2014. K Vivek Varkey IIT Hyderabad
  • 3.
     Acknowledgements: Wewould like to thank our mentor for providing motivation and the required data in full detail, due to which the process was highly hassle-free. Also we would like to convey gratitude to the ONGC Hazira administration for arranging the training and allowing us to pursue this project.
  • 4.
     Introduction Anyprocess is an energy sink, demanding high amount of heat addition or removal, that ask for individual high costs, especially in a large process plant. A process may employ numerous heat exchangers to bring the process stream at required temperatures at the various stages. Operation of these all heat exchangers with their individual cooling utilities and heating utilities demand a high operating cost on a daily basis. It is hence beneficial to strategize the heat exchanger network layout so as to couple the hot and cold process streams, so that minimum heat duty is to be supplied or removed using the external utilities. However design of an optimal heat exchanger network (HEN) so as to employ minimum operating costs keeping in mind the installation costs are also not out of proportion, is a complicated method. Hence several algorithms and famous approaches exist. This project employs the PINCH ANALYSIS of the energy flow to determine the optimal HEN structure for the CFU in the ONGC Hazira plant. The report does not deal with the derivation of the pinch technique, as several can be found in literature. The data are taken from the design report existing in the plant. The particular case deals with an inlet temperature of 33oC and inlet pressure of 93Kg/cm2. The heat duties are calculated in an ideal scenario, from the material balances, temperatures and enthalpy data taken from the web. We deal with the 5 heat exchangers namely, E701, E702, E703, E705 and E706. E704 is not dealt with for reasons of unavailability of data.
  • 5.
    S T R I P P E R L P G Reflux Drum Preheater Off Gas Cooler LPG Condenser NGL Cooler Reboiler Feed Off Gas to GSU LPG to CW Naptha to storage  Process Flow Diagram Vapour Condensate Feed from Slug Catcher at 33oC and 93Kg/cm2 PFD shown only for elements associated with this project. 1 5 7 3 4 2 8 6
  • 6.
     Material Balance Component 1 2 3 (liquid) 3(vapor) 4(liquid) 4(vapor) 5 6 7 8 N2 0.02 0.01 0 0.01 0 0 0 0 0 0.02 H2S 1.36 0.13 0.75 0.48 0 0 0 0 1.23 1.36 CO2 40.46 7.15 12.11 21.2 0 0 0 0 33.31 40.46 C1 266.56 72.14 40.82 153.6 0 0 0 0 194.42 266.56 C2 68.51 7.5 34.02 26.99 0 0 0 0 61.01 68.51 C3 90.32 4.76 68.18 17.38 12.69 45.56 58.25 0 27.31 32.07 iC4 28.99 0.9 25.22 2.87 7.18 16.42 23.6 0 4.49 5.39 nC4 45.74 1.13 41.21 3.4 13.32 25.95 39.16 0.11 5.34 6.47 iC5 19.49 0.27 18.55 0.67 8.22 9.93 0.11 18.04 1.07 1.34 nC5 22.4 0.26 21.54 0.6 10.14 11.04 0.01 21.17 0.96 1.22 C6 32.03 0.18 31.54 0.31 19.39 11.95 0 31.34 0.51 0.69 C7 38.59 0.11 38.35 0.13 28.22 10.03 0 38.25 0.23 0.34 C8 47.64 0.07 47.51 0.06 39.45 8.02 0 47.47 0.1 0.17 C9 25.29 0.02 25.26 0.01 22.58 2.67 0 25.25 0.02 0.04 C10 15.5 0.01 15.49 0 14.5 0.99 0 15.49 0 0.01 C11 17.73 0 17.73 0 17.02 0.71 0 17.73 0 0 C12+ 7.96 0 7.96 0 7.77 0.19 0 7.96 0 0 All flows are in Kmol/hr.
  • 7.
     Cp valuesfor the required components
  • 8.
    The values ofthe components in the CFU unit is obtained. Following which the respective mole fraction in each stream is also obtained. These two data are used simultaneously and the weighted mean is to obtained the overall specific heat value.
  • 9.
     PROCEDURE Themethod used is called Temperature Interval Method of Pinch Analysis 1. The Cp value obtained is multiplied by the flow rate to obtain the heat capacity flow rate (C). 2. Now we create the following table for the further calculation. The approach temperature is taken to be 10 degree Celsius. This is subtracted from both the inlet and outlet temperatures of the hot stream for the purpose of calculation, so as to obtain the pinch where the net heat exchange is supposed to be 0. this is put in is Tout* and Tin* in the tables.
  • 10.
    E-702 (Stipper Reboiler)C2 Assumptions: • Stripper reboiler operates at constant 156oC, and main heat consumption is to cause phase change. • Stripper column bottoms is saturated liquid and the outlet composition is same as the feed to the LPG column. • Heat duty is calculated by determining the amount of latent heat required. T = 156oc M = 344 Kmol/hr Q = - 959.03 KW E-703 (LPG Condenser) H1 Assumptions: • Constant operation temperature of 43oC • Feed is saturated vapour from the LPG column top and outlest stream is saturated liquid. • Heat duty is calculated from the latent heat calculations. T = 43oC M = 121.13 Kmol/hr Q = + 627.55 KW
  • 11.
    E-705 (NGL Cooler)H2 Assumptions: • The inlet and outlet temperatures are respectively 175oC and 45oC. • The composition is known from the mass balance. • No phase change. Tf = 45oC Ti = 175oC Cp = .244 KJ/mol K M = 222.81 Kmol/hr M*Cp = 15.10 KW/K Q = + 1963.0 KW
  • 12.
    E-706 (Off GasCooler) H3 Assumptions: • Inlet and outlet temperatures are respectively 125oC and 40oC. • No phase change. • Composition is given in the material balance. Tf = 40oC Ti = 125oC Cp = 0.044 KJ/mol K M = 424.65 Kmol/hr M*Cp = 5.20 KW/K Q = + 442 KW Hence summing all the enthalpy requirements of the streams we get, C1+C2+H1+H2+H3 = + 1801.52 KW Hence the Q min, req is + 1801.52 KW that must the removed somewhere in the process using cooling utilities
  • 13.
     PROCEDURE 3.Now we draw a diagram where we label the adjusted temperatures in order from coldest to hottest and we draw arrows to designate streams overlapping these temperature intervals.
  • 14.
     PROCEDURE 4.Now for each interval in the previously drawn figure we find the enthalpy of each intervals. This is obtained by adding the heat capacity flow rates ( for hot streams it is considered positive and negative for the cold stream ). This is then multiplied by the temperature interval for which we are calculating and the values are written right next to the respective intervals in the figure labelled under Q. 5. These values are then added cumulatively from the top to the bottom ( higher temperature to colder temperature ). The obtained data is then written in a new column Qres. 6. The least value in the Qres column is called the pinch value and the temperature of that interval is called the pinch temperature. 7. The pinch value is then added from the beginning of the values in Qres. It is observed that it becomes 0 at the pinch point. 8. The pinch value is the hot utility added and the final value obtained ( in this case 2005.55 ) becomes the cold utility required.
  • 15.
    Hence we determineour pinch temperature as 115oC for the Cold streams and 125oC for the Hot streams. Also MER (minimum energy requirement) targets: • Qhot utility = -204.43 KW (to be added by steam) • Qcold utility = 2005.55 KW (to be removed by cooling water) As per the method the hot utility can only be used above pinch, and cold utility can only be used below pinch. Before we start coupling a hot stream with a cold stream we should keep one thing in mind. Let the specific heat flow rate of the hot stream be Ch and that of the cold stream be Cc . If we are trying to couple in the hot side of the pinch it has to be made sure that Cc > Ch of the respective streams. Similarly when we are trying to couple in the cold side of pinch it should be taken care that Ch > Cc . Otherwise the stream coupling will become infeasible. In the next page we draw a pinch decomposition of the streams and determine an optimal heat exchanger network.
  • 16.
    PROOF Let usassume a counter current pair of hot and cold stream. Thi and Tho are respectively the inlet and outlet temperature of the hot stream. Tci and Tco are respectively the inlet and outlet temperature of the cold stream. ΔT1 and ΔT2 are respectively the difference between hot inlet and cold inlet and hot inlet and cold outlet. Q is the energy exchanged Ch Cc are respectively the specific heat flow rates for the hot and cold streams.
  • 17.
    PROOF Q =Ch * [ Thi - Tho ] Q = Cc * [ Tco - Tci ] After rearranging we get, Thi - Tho = Q/Ch Tco - Tci = Q/Cc now we subtract the equations to get ΔT2 - ΔT1 = Q*[Cc - Ch ]/Cc Ch
  • 18.
    PROOF HOT SIDEOF THE PINCH ΔT1 = ΔTMIN Hence, ΔT2 = ΔTmin + Q*[Cc - Ch ]/Cc Ch now, ΔT2 has to be greater than ΔTmin hence, Cc > Ch COLD SIDE OF THE PINCH ΔT2 = ΔT min Hence, ΔT1 = ΔTmin - Q*[Cc - Ch ]/Cc Ch now ΔT1 has to be greater than ΔTmin hence, Ch > Ch
  • 19.
    In the nextpage we draw a pinch decomposition of the streams and determine an optimal heat exchanger network.
  • 20.
    H1 H2 H3 C1 C2 175 125 156 156 125 45 125 43 43 33 40 43 P I N C H H=204.03 H=755 H=272 H= 627.55 H=936 H=442 Heating Utility Heat Exchanger Cooling Utility Heat loads of Exchangers mentioned are in KW. Temperature is in Celsius Above pinch Below pinch
  • 21.
    The above thediagram can be explained by taking an example from the actual streams. So in the hot side of the pinch we see a hot stream E-705 going from 175˚C to 125 ˚C with a heat capacity flow rate of 15.1 KW/K. this would give out a heat of 15.1*[175-125] KW = 755 KW Now, when we couple this hot stream with a cold stream , E-702, this 755 KW of energy can be utilised by the cold stream which actually requires energy of 959.03 KW. For the excess 204.03 KW, that is required for the cold stream, we provide it with a hot utility of the same amount. Similarly we obtain the other hot and cold utilities required.
  • 22.
     Results andDiscussions: The exchanger network drawn allows for the MER targets . The streams are coupled and utilities are distributed keeping in mind the temperature interval of 10, and the subsequent corollary that heating utilities can only be used above pinch and cooling utilities can only be used below pinch. The main purpose of this design is to allow for energy recycle, a term synonymous with the modern era and the global challenges we face. Though in this process it is not significant, some processes can self sustain themselves, that is we can achieve all temperature targets using very little external utilities. This leads to economic benefits and environmental benefits. The temperature interval method and pinch analysis as shown here can be utilised for all processes for energy recycle. The CFU here is only one example. The project has several assumptions of ideality and some figures are rounded off to allow for smooth calculations. Hence for application of the process, this is a rough schematic, more accurate measurements, considerations and calculations are required. Also installation costs and restrictions must be considered when determining heat exchanger area. As per the calculation shown, • 204.03 kW hot utilities • 2005.55 kW cold utilities is required.
  • 23.
     Results andDiscussions: Current energy requirement Heating utilities QE-701 + QE-702 = 27.2*[33-43] + [-959.03] = 1231.03 kW Cooling utilities QE-7.03 + QE-705 + QE-706 = 627.55 + 15.1*[175-45] + 5.2*[125-40] = 3032.55 kW Therefore the energy that can be saved in terms of percentage :- • 83.4% for heating utilities • 33.8% for cooling utilities
  • 24.
     Comments: Uponsupervision by the mentor the following drawbacks were observed :- 1. The NGL cooler is not always in functioning. It is used generally only during abnormal functioning of C-702 2. In this project it was taken into assumption that everything in reboiler vaporised at 100%. However, this is not the case. 3. E-704 was not used in the calculation.
  • 25.
     References: www.engineeringtoolbox.xom www.wikipedia.com www.cheresources.com Process and Product Design (Seeder) Nptel open courseware