Indian Institute of Technology Patna

Department of Mechanical
Engineering

Shell and Tube Heat Exchanger

ME – 315 Heat and Mass Transfer

Instructor – Dr. Rishi Raj
SUBMITTED BY: Group 5

Hemant Surariya 1901ME26 (Theory and discussion )

Makineni Ajay Kesava Chowdary 1901ME40 (Calculations and Results)

Mudavath Vijay 1901ME45 (Objective,Equipment Required and

Precautions)

Ottikunta Santhosh 1901ME46 (Calculations, Results )

Raj Kumar 1901ME50 ( Procedure and Observations Table )

Suganth R 1901ME61 ( Error and Conclusion )

OBJECTIVE :
● To determine the overall heat transfer coefficient for shell and tube
heat exchangers using LMTD method and Epsilon-NTU method.
● To determine the heat exchanger effectiveness for shell and tube
heat exchanger.

EQUIPMENT REQUIRED :
Shell and tube experiments : Shell and tube set up,instant water
heater,rotameters and sensor board to measure the corresponding
temperatures.

SHELL AND TUBE HEAT EXCHANGER APPARATUS:

THEORY:

A shell and tube exchanger consists of a number of tubes mounted inside a

cylindrical shell. Two fluids can exchange heat, one fluid flows over the
outside of the tubes while the second fluid flows through the tubes. The
fluids can be single or two phases and can flow in a parallel or a
cross/counter flow arrangement.
The rate of heat transfer depends on the thermal conductivity of the
dividing wall and convective heat transfer coefficient between the fluids and
wall. Heat transfer rate also changes according to the boundary conditions
like insulated wall condition or adiabatic condition. Heat transfer in a heat
exchanger usually involves convection in each fluid and conduction through
the wall separating the two fluids. Heat exchangers are classified based on
various features such as transfer process, fluid flow direction, number of
fluids, surface compactness, heat transfer mechanism, construction etc.

There are two ways of analysing Heat Exchangers

● LMTD method
● NTU- epsilon approach

LMTD method:

This method is used when the inlet and outlet temperatures of both hot and
cold streams are know
𝑄 = 𝐹𝑈𝐴∆𝑇𝑚

Where Q is the Heat Transferred between the hot and cold fluids

U is the overall Heat Transfer Coefficient.

A is the total area available for heat exchange

∆𝑇𝑚 is the log mean temperature difference

Now for parallel flow,∆𝑇𝑚 is given by

∆𝑇1−∆𝑇2
∆𝑇𝑚 = ∆𝑇1 where ∆𝑇1 = 𝑇ℎ,𝑖 − 𝑇𝑐,𝑖 𝑎𝑛𝑑 ∆𝑇2 = 𝑇ℎ,𝑜 − 𝑇𝑐,𝑜
𝑙𝑛( ∆𝑇2 )

And for Counter flow , ∆𝑇𝑚is given by

∆𝑇1−∆𝑇2
∆𝑇𝑚 = ∆𝑇1 where ∆𝑇1 = 𝑇ℎ,𝑖 − 𝑇𝑐,𝑜 𝑎𝑛𝑑 ∆𝑇2 = 𝑇ℎ,𝑜 − 𝑇𝑐,𝑖
𝑙𝑛( ∆𝑇2 )
And Overall Heat Transfer Coefficient (U) for Double Pipe Heat

Exchanger is given by
𝐶ℎ(𝑇ℎ,𝑖−𝑇ℎ,𝑜)
𝑈 = 𝐴∆𝑇𝑚

With ro and ri being the outer and inner radii of the inner tube of the double
pipe heat exchanger

Ao is the outer surface area of the inner tube and L is the length of the inner
tube.

NTU-epsilon method:

This method is used when only the inlet temperatures of the two streams
are given Heat Exchanger Effectiveness is given by
𝑞
ϵ= 𝑞𝑚𝑎𝑥

With,
𝑞𝑎𝑐𝑡𝑢𝑎𝑙 = 𝐶ℎ(𝑇ℎ,𝑖 − 𝑇𝑐,𝑖),
𝑞𝑚𝑎𝑥 = 𝐶𝑚𝑖𝑛(𝑇ℎ,𝑖 − 𝑇𝑐,𝑖)
And after simplification Effectiveness can also be written as:

For parallel flow arrangement

1−𝑒𝑥𝑝(−𝑁𝑇𝑈(1+𝐶𝑟))
ϵ= 1+𝐶𝑟

For counter flow arrangement

1−𝑒𝑥𝑝(−𝑁𝑇𝑈(1−𝐶𝑟))
ϵ= 1−𝐶𝑟𝑒𝑥𝑝(−𝑁𝑇𝑈(1−𝐶𝑟))
The applications of double pipe heat exchangers are a lot, and it is hard to
encompass all, and it is all because of the practical, elegant, and simple
design. A few of them are as below:

● Power Generation
● HVAC
● Marine Applications
● Pulp and Paper
● Refrigeration
● Pharmaceuticals
● Air Processing and Compressor Cooling
● Metals and Mining

EXPERIMENTAL SET-UP :

Parameters Materials Dimensions

Shell (cold water) Mild Steel Di =210mm
Do =220mm
Tubes (hot water) Copper (k=385 W/m-k) di =12.7mm
do =13.95mm
Number of Passes(n) 2
Length of the heat 0.5 m
exchanger(L)
Number of tubes 32

PROCEDURE :-

1. This experiments will be carried out for at least three sets of mass
flow rates of cold water keeping the hot water flow rate constant.
 2. And then Set the flow of water through two pipes (inner and outer ) by
opening the valves V1 and V2. The flow rate may be set to some
desired value using the Rotameters R1 and R2.
3. Now , switch on the geyser.
4. Record the temperature of the entering and leaving hot and cold
streams using 4 different thermocouples and digital temperature
indicator placed on the control panel, only after the steady - state is
reached.
5. Repeat the experiment for other flow rates of cold water.

OBSERVATIONS TABLE :-

Set 1 :-

HOT WATER (°C) COLD WATER (°C)

FLOW RATE (LPH) = 220 FLOW RATE (LPH) = 200

Time (min ) Th,i (T4) Th,o (T2) Tc,i (T3) Tc,o (T1)

t=0 38 35 25 26

t= 10 38 32 25 30

t = 20 38 32 25 30

Set 2 :-

HOT WATER (°C) COLD WATER (°C)

FLOW RATE (LPH) = 100 FLOW RATE (LPH) = 150

Time (min ) Th,i (T4) Th,o (T2) Tc,i (T3) Tc,o (T1)

t=0 38 32 25 30

t= 10 50 34 25 33

t = 20 48 34 25 33
Set 3 :-

HOT WATER (°C) COLD WATER (°C)

FLOW RATE (LPH) = 80 FLOW RATE (LPH) = 240

Time (min ) Th,i (T4) Th,o (T2) Tc,i (T3) Tc,o (T1)

t=0 48 34 25 33

t= 10 48 32 25 30

t = 20 48 32 25 30

CALCULATION:

For Set1

For Set-1 we have,

ṁℎ = 220 𝑙𝑝ℎ = 0. 0616 𝑘𝑔/𝑠
𝑇ℎ,𝑖 = 38 ℃ 𝑇ℎ,𝑜 = 32 ℃
ṁ𝑐 = 200 𝑙𝑝ℎ = 0. 056 𝑘𝑔/𝑠
𝑇𝑐,𝑖 = 25 ℃ 𝑇𝑐,𝑜 = 30 ℃
𝑇ℎ,𝑖 + 𝑇ℎ,𝑜 𝑇𝑐,𝑖 + 𝑇𝑐,𝑜
𝑇𝑚,ℎ = 2
= 35 ℃ = 308 𝐾 𝑇𝑚,𝑐 = 2
= 27. 5 ℃ = 300. 5 𝐾

Calculating the properties for hot water at 𝑇𝑚,ℎ (308 K)

From Table A.6 from book, interpolating between 305 K and 310 K,
we get,
3
−3 𝑚
𝑣ℎ = 1. 0062 × 10 𝑘𝑔
3 𝐽
𝑐𝑝,ℎ = 4. 178 × 10 𝑘𝑔·𝐾
−6 𝑁·𝑠
𝜇ℎ = 724. 6 × 10 2
𝑚
 −3 𝑊
𝑘ℎ = 624. 8 × 10 𝑚·𝐾
𝑃𝑟ℎ = 4. 852

Now we will calculate the properties of cold water at 𝑇𝑚,𝑐(300.5 K)

From Table A.6 from book, interpolating between 300 K and 305 K,
we get,
3
−3 𝑚
𝑣𝑐 = 1. 0032 × 10 𝑘𝑔
3 𝐽
𝑐𝑝,𝑐 = 4. 1789 × 10 𝑘𝑔·𝐾
−6 𝑁·𝑠
𝜇𝑐 = 846. 4 × 10 2
𝑚
−3 𝑊
𝑘𝑐 = 613. 7 × 10 𝑚·𝐾
𝑃𝑟𝑐 = 5. 767

Theoretical approach:

Convective heat transfer coefficient for inner tubes

−3
𝑑𝑖 = 12. 7 × 10 𝑚
ṁ 0.0616
ṁ = 32ℎ = 32
= 0. 001925 𝑘𝑔/𝑠
4ṁ 4×0.001925
𝑅𝑒𝐷 = π𝑑 𝜇 = −3 −6 = 266. 342
𝑖 3.14×12.7×10 ×724.6×10
Hence we can say that the flow inside the tubes is laminar. 214.5

Assuming fully developed, constant Heat flux, we can say that

𝑁𝑢𝐷 = 4. 36
𝑘 624.8×10
−3
𝑊
ℎ𝑖 = 4. 36 × 𝑑ℎ = 4. 36 × −3 = 214. 5 2
𝑖 12.7×10 𝑚 .𝐾

Convective heat transfer coefficient for shells

The shell can be assumed to behave as annulus
So
 −3 −3
𝐷𝑖 = 210 × 10 𝑚 𝑑𝑜 = 13. 95 × 10 𝑚
−3
𝐷ℎ = 𝐷𝑖 − 𝑑𝑜 = 196. 05 × 10 𝑚
−3
ṁ𝑐×10
ṁ = 3600×𝑣𝑐
𝑘𝑔/𝑠
−3
200×10
⇒ −3 = 0. 0554 𝑘𝑔/𝑠
3600×1.0032×10

Reynold’s number,
4ṁ 4×0.0554
𝑅𝑒𝐷 = π(𝐷 −𝑑 )𝜇 = −3 −6 ⇒ 425. 302
𝑖 𝑜 3.14×196.05×10 ×846.4×10

So the flow inside the shell is laminar

𝑑𝑜 13.95
Now 𝐷𝑖
= 210
= 0. 06643

From Table 8.3 in the book

Interpolating between diameter ratio 0.05 and 0.10 for 𝑁𝑢𝑖, we get
𝑁𝑢𝑖 = 15. 87126
𝑘𝑐 613.7×10
−3
2
Therefore, ℎ𝑜 = 𝐷ℎ
× 𝑁𝑢𝑖 = −3 × 15. 87126 = 49. 62 𝑊/𝑚 . 𝐾
196.05×10

overall heat transfer coefficient

1
𝑈=
( 1
𝐴𝑜 ℎ 𝐴
𝑖 𝑖
+
𝑙𝑛(𝑟𝑜/𝑟𝑖)
2𝜋𝐿𝑘
+ℎ𝐴
1
𝑜 𝑜
)
'
Where, 𝐴𝑜 = 2π𝑁𝐿 𝑑𝑜
'
𝐴𝑖 = 2π𝑁𝐿 𝑑𝑖
'
𝐿 = 2𝑁𝐿
'
Where, 𝐿 =Length per pass of the shell and tube heat exchanger.
N = Number of tubes.
𝑑𝑜 =Outer diameter of inner tube.
 𝑑𝑖 =Inner diameter of inner tube.
So, we can write the expression for U as

1
U=
( 1
ℎ𝑜
+
𝑑𝑜𝑙𝑛(𝑑𝑜/𝑑𝑖)
𝑘
+ℎ
1
𝑖
( ))
𝑑𝑜
𝑑𝑖

Putting all the above obtained values, we get the value of U

𝑊
𝑈𝑡ℎ = 37. 617 2
𝑚 ·𝐾

Using Log-Mean Temperature Difference (LMTD) method:

−3 3
200×10 ×4.1789×10 ×(30−25)
Q = ṁcCp,c( Tc,o - Tc,i ) = −3
3600×1.0032×10

Q = 1157.1028 w
Another relation for Q is
Q = 𝐹𝑈𝐴(△𝑇)𝐿𝑀𝑇𝐷
Where,
F = Correction factor
A = Total Area of Heat exchanger
U = Overall heat transfer coefficient
(△𝑇)𝐿𝑀𝑇𝐷 = Log mean temperature difference based on counter-flow
arrangement
F can be calculated from the table using P and R values.
𝑇ℎ,𝑜 − 𝑇ℎ,𝑖 32−38
P= 𝑇𝑐,𝑖−𝑇ℎ,𝑖
= 25−38
= 0. 9615
𝑇𝑐,𝑖−𝑇𝑐,𝑜 25−30
R= 𝑇ℎ,0−𝑇ℎ,𝑖
= 32−38
= 0. 833

Based on these values obtained above, we get F = 0.92

((38−30)−(32−25)) 1
(△𝑇)𝐿𝑀𝑇𝐷 = 38−30 = 8 = 7. 49
𝑙𝑛( 32−25
) 𝑙𝑛( 7 )
and since there are two passes so,
A = 2πd oNL’
𝑄 1157.1028
U = 𝐹𝐴(△𝑇) = −3 = 119.754
𝐿𝑀𝑇𝐷 0.92×2×3.14×13.95×10 ×32×0.5×7.49

𝑊
𝑈𝐿𝑀𝑇𝐷 = 119.754 2
𝑚 ·𝐾

Using ε-NTU method

−3 3
220×10 ×4.17×10 𝑊
Ch= ṁhcph = −3 = 253.7989 𝑘
3600×1.0062×10
−3 3
200×10 ×4.17×10 𝑊
Cc=ṁccpc= −3 = 231.4205 𝑘
3600×1.0062×10

𝐶ℎ < 𝐶𝑐, so 𝐶𝑚𝑖𝑛 = 𝐶𝑐and 𝐶𝑚𝑎𝑥 = 𝐶ℎ

Cr= Cmin/Cmax= 0.912

Now
𝑄𝑎𝑐𝑡𝑢𝑎𝑙 𝑄𝑎𝑐𝑡𝑢𝑎𝑙
ε= 𝑄𝑚𝑎𝑥
= 𝐶𝑚𝑖𝑛·(𝑇ℎ𝑖 − 𝑇𝑐𝑖)
1157.1028
⇒ 231.42 · (38−25)
= 0.3846

Using Table 11.4 we have the relation for NTU of one shell pass with even
tube passes
𝐸−1
NTU1=-(1+C2r)-0.5 ln( 𝐸 + 1 )
where
2
ε1
− (1+𝐶𝑟)
E= 2 0.5
(1+𝐶𝑟 )

we have, ε = 0.3546 , Cr = 0.912

we get,
NTU = 0.646608
 𝑈𝐴
𝐶𝑚𝑖𝑛
=0.646608
and since there are two passes so,
A = 2πd oNL’

0.646608×231.42056
U= −3
2×32×0.5×13.95×10

𝑊
𝑈𝑛𝑡𝑢= 106.701 2
𝑚 ·𝐾

RESULTS:
𝑊
All values of U are in 2
𝑚 ·𝐾

set1 set2 set3

U theoretical 37.617 39.81 39.68
U LMTD 119.754 100.24 94.45
U NTU-e 106.701 86.086 76.95
%Error in U LMTD 218.35 151.79 138.03
%Error in U NTU-e 183.65 116.24 93.926
Effectiveness 0.385 0.524 0.615

DISCUSSION:
● The results show that the effectiveness of the exchanger device is
associate with the flow rate of the cold fluid
 ● The flow rate should be less so that heat exchange between the
fluids is more efficient but in practical use we may not be able to
reduce the flow rate beyond a threshold in that case increasing the
surface area of the tube inner tubes by increasing the length or
diameter would improve the heat transfer.

● We can observe that heat lost by the hot fluid is not equal to the heat
gained by cold fluid.
i.e; 𝐶ℎ(𝑇ℎ,𝑖 − 𝑇ℎ,𝑜) ≠ 𝐶𝑐(𝑇𝑐,𝑜 − 𝑇𝑐,𝑖)
So in our calculation for NTU- method we can say that appropriate
answers can’t be obtained from this method.

● The difference between the theoretical value and the experimental

value is Expected due to many errors and approximations
Considered while calculating.

SOURCES OF ERROR:

● In the whole experiment the main thing we have considered is

specific heat capacity and convection heat transfer coefficient of the
water is constant, but in reality, these values depend on the
temperature which gives us major error.

● Our calculation is done on the belief that Shell and Tube Heat
Exchanger is insulated but there will be heat transfer between the
system and the surroundings which causes the error .

● Fouling which occurs in Shell and Tube Heat Exchanger is the

formation of unwanted material deposits on heat transfer surfaces
with the passage of time. By forming a kind of insulation on
heat-transfer surfaces It reduces heat transfer and also the value of
thermal resistance is affected which we are not considering in the
calculation.
 ● There might be some fluctuations in readings and instrumental errors
in Thermometers and Rotameters.

CONCLUSION:

In this experiment, our team members have learned different equipment

that are required for the experiment,procedure , theory & the necessary
assumptions to determine the overall heat transfer coefficient using LMTD
method and Epsilon-NTU method & To determine the heat exchanger
effectiveness for shell and tube heat exchanger.

From the results table, It is observed that U theoretical is almost constant

for different mass flow rates,it's anticipated as our nusselt number is
constant for fully developed laminar flow, so heat transfer coefficient will
depend on thermal conductivity of water. since for different mass flow rates
the average Temperature doesn’t change much, the thermal conductivity of
water also doesn’t have huge impact. so heat transfer coefficient will
remain almost constant.

PRECAUTIONS :
● Always keep the apparatus free from dust.
● Never switch ON main power supply before ensuring that all the
ON/OFF switches given on the panel are at OFF position.
● Never run the apparatus if the power supply is less than 200 volts or
more than 230 volts .
● Operate the selector switch of the temperature indicator gently.