ME 320 Lab 5

Heat Exchanger Investigation

Cole Meyer
Section ABE, Monday 11-1pm
TA: Yu Chen
November 27, 2017
 I. Introduction

Heat exchangers are used in a wide array of real world applications such as cars and

air conditioners. The types of heat exchangers studied in this lab are shell-and-tube,

concentric-tube, and brazed plate. Shell-and-tube heat exchangers are the most common

type and consist of an outer shell that has many smaller tubes. The hot liquid will flow

through the smaller tubes and the cold liquid flows around the smaller tubes within the

shell. This type of heat exchanger is neither parallel or counter flow. Concentric-tube heat

exchangers have the cold and hot fluids flow in either the same direction or opposite

direction, known as parallel flow or counter flow respectively. Brazed plate heat

exchangers consist of thin metal plates that are held together. There are small gaps that

the hot and cold fluids alternate through.

In this lab, the effectiveness of each of the heat exchangers was studied as well as the

differences between parallel and counter flow within concentric-tube heat exchangers.

The effect of changing inlet temperatures and flow rates were also observed in the

different heat exchangers.

During this lab, there are a multitude of assumptions that are made to simplify the

analysis done for this lab. It was assumed that the heat capacity of water is constant with

respect to changing fluid temperature, pressure drop is negligible, potential and kinetic

energy are negligible, and that there is no heat loss to the surroundings.

A multitude of equations were used to help analyze the data in this lab.

𝐶𝑗 = 𝑚̇ 𝑗 𝑐𝑝𝑗 (1)

𝑄̇ = 𝐶𝑗 ∆𝑇 (2)
 Ci is the heat capacity rate, 𝑚̇ 𝑗 is the mass flow rate of the fluid, cp is the constant-

pressure specific heat capacity, 𝑄̇ is the heat transfer rate, and ∆𝑇 is the temperature

difference between the inlet and outlet in the equations above. The log-mean temperature

difference for both parallel and counter flow and the overall heat transfer conductance are

also necessary equations for analysis and can be seen below.

(𝑇ℎ,𝑖 −𝑇𝑐,𝑖 )−(𝑇ℎ,𝑜 −𝑇𝑐,𝑜 )

∆𝑇𝑙𝑚,𝑃𝐹 = 𝑇ℎ,𝑖 −𝑇𝑐,𝑖 (3)
ln
𝑇ℎ,𝑜 −𝑇𝑐,𝑜

(𝑇ℎ,𝑖 −𝑇𝑐,𝑜 )−(𝑇ℎ,𝑜 −𝑇𝑐,𝑖 )

∆𝑇𝑙𝑚,𝐶𝐹 = 𝑇ℎ,𝑖 −𝑇𝑐,𝑜 (4)
ln
𝑇ℎ,𝑜 −𝑇𝑐,𝑖

𝑄̇
𝑈𝐴 = ∆𝑇𝑛𝑒𝑡 (5)
𝑙𝑚

Th,I is the hot fluid inlet temperature, Th,o is the hot fluid outlet temperature, Tc,I is the

cold fluid inlet temperature, Tc,o is the cold fluid outlet temperature and UA is the overall

heat transfer conductance in the equations above. In order to analyze the performance of

the heat exchangers the number of transfer units, the effectiveness of both parallel and

counter flow configurations, and the heat capacity rate ratio are listed below.
𝑈𝐴
𝑁𝑇𝑈 = 𝐶 (5)
𝑚𝑖𝑛

𝐶
𝐶𝑟 = 𝐶 𝑚𝑖𝑛 (6)
𝑚𝑎𝑥

1−exp[−𝑁𝑇𝑈(1+𝐶𝑟 )]
𝜀𝑃𝐹 = (7)
1+𝐶𝑟

1−exp[−𝑁𝑇𝑈(1−𝐶𝑟 )]
𝜀𝐶𝐹 = 1−𝐶 (8)
𝑟 ∗exp[−𝑁𝑇𝑈(1−𝐶𝑟 )]

In equation 5, NTU is the number of transfer units. Cmin and Cmax, shown in equation

6, are the minimum and maximum heat capacity rates of the cold and hot fluids. Lastly,
Cr is the heat capacity rate ratio, 𝜀𝑃𝐹 and 𝜀𝐶𝐹 are the effectiveness’ of the parallel and

counter flow configurations.

II. Experimental Methods

Figure 1. Schematic of experimental equipment setup

This figure shows the setup of the lab. A hot water reservoir was used and it had two

heaters on it. Attached to the reservoir were a concentric-tube, brazed plate, and shell-

and-tube heat exchanger. Hot water was supplied from the reservoir to the hot fluid inlets

of each of the heat exchanger. Coming from the other direction was a cold water supply

that interacted with the hot water and various ways. The flow rates of each of the water

supplies could be controlled and changed. Thermocouples were attached to the various

setups to measure the temperatures of the water throughout the setup. Initially, the water

heater was turned on and heated the water up to the desired temperatures of 40˚C, 50˚C,

and 60˚C. The flow of the water was then turned on and adjusted to values of 1, 2, and 3

L/min. The cold water was then allowed to flow as well and the system was allowed to
reach steady state. After it was reached, the values were then recorded using the

thermocouples. The steps were repeated for different heat exchangers, flow rates, and

temperatures.

III. Results and Discussion

70

60
Local Temperature - T - °C

50
1 L/min C
40 2 L/min C

30 3 L/min C
1 L/min H
20 2 L/min H

10 3 L/min H

0
0 0.5 1 1.5 2
Cumulative Heat Transfer Rate - Q_dot - kW

Figure 2. Load curves for parallel flow in concentric-tube at different flow rates

The figure above shows the relationship between the temperature at a specific

location and the cumulative heat transfer rate at the location. The relationship is a linear

one with hot fluids having a negative linear motion while cold fluids have a positive

linear motion. This makes sense since the heat transferring between the two fluids causes

the hot water temperature to drop and the temperature of the cold water to rise. As the

flow rate of the hot liquid increases, the hot water slope becomes shallower, whereas the

cold liquid’s slope seems unaffected by flow rate. The table below shows the effect of a

changing flow rate in a concentric-tube for parallel flow.

 Table 1. Varying flow rates effect on overall heat transfer conductance and log-

mean temperature difference for parallel flow in a concentric-tube

Vhot [L/min] UA Tlm

1 0.028885 21.7554
2 0.042034 29.89973
3 0 0
It’s difficult to give a definitive trend on the behavior since the final flow rate results

in a zero. However, an intelligent guess could be that with an increasing flow rate, the

overall heat transfer conductance and log-mean temperature increase as well.

70

60
Local Temperature - T - °C

50
60C C
40 50C C

30 40C C
60C H
20
50C H
10 40C H
0
0 0.5 1 1.5 2
Cumulative Heat Transfer Rate - Q_dot - kW

Figure 3. Load Curves for parallel flow in a concentric-tube at different

temperatures

Figure 3 shows the relationship between the local temperature and the cumulative

heat transfer rate at that location while changing the inlet temperature of the hot fluid. As

can be seen above, raising the temperature of the flood causes a vertical shift in the line

but has no noticeable effect on the slope. The cold fluid lines are unaffected by the

different inlet temperatures of the hot fluid. The effects of the different temperatures can

also be seen in the table below.

 Table 2. Varying inlet temperatures effect on overall heat transfer conductance

and log-mean temperature difference for parallel flow in a concentric-tube

T inlet[˚C] UA Tlm
60 0.05681 29.49717
50 0 0
40 0 0
Due to possible experiment error, it is difficult to make an assumption on the trend as

a result of two different data points having zeroes.

60

50
Local Temperature - T - °C

40 1 L/min C
2 L/min C
30
3 L/min C
20 1 L/min H
2 L/min H
10
3 L/min H
0
0 0.5 1 1.5 2
Cumulative Heat Transfer Rate - Q_dot - kW

Figure 4. Load curves for counter flow concentric-tube at different flow rates

The figure above illustrates the relationship between the local temperature and the

cumulative heat transfer rate associated with that location while varying the hot fluid

flow rate. As can be seen in figure 4, increasing the flow rate of the hot fluid causes a

decrease in the slope of the hot fluid line. However, the cold fluid line remains unaffected

by the change in the flow rate of the hot liquid. These effects can be seen below in Table

3.
 Table 3. Varying flow rates effect on overall heat transfer conductance and log-

mean temperature difference for counter flow concentric-tube

Vhot [L/min] UA Tlm

1 0.02949 21.30872
2 0.050025 20.93635
3 0.063495 16.49495
For this specific case, with an increasing flow rate the UA increases while the log-

mean temperature decreases.

60

50
Local Temperature - T - °C

40 60C C
50C C
30
40C C
20 60C H
50C H
10
40C H
0
-0.5 0 0.5 1 1.5 2
Cumulative Heat Transfer Rate - Q_dot - kW

Figure 5. Load curves for counter flow concentric-tube at different temperatures

Figure 5 above demonstrates the relationship between the local temperature of the

flow and the cumulative heat transfer rate at that location while varying the temperature

of hot fluid at its inlet location. If the hot fluid temperature is increased, the whole line is

translated vertically, which is also true for the cold fluid in the counter flow

configuration. For the two of the data points, they are either negative or zero. This error

can be attributed to either a human error or issues with the readings the thermocouples
took. Temperatures could have read lower than the actual temperature, which would

result in the negative or zero points. The effects that changing the temperature in the

counter flow concentric-tube setup can be seen in table 4 below.

Table 4. Varying temperature effect on overall heat transfer conductance and

log-mean temperature difference for counter flow concentric-tube

T inlet[˚C] UA Tlm
60 0 0
50 0.063495 16.49495
40 0.046741 -4.418142
This case almost has a parabola like behavior by hitting a peak then dropping again.

Table 5. Effect of varying flow rates on overall heat transfer conductance and

log-mean temperature difference for brazed plate

Vhot [L/min] UA Tlm

1 0.013828 15.1483
2 0.069325 18.12919
3 0.102128 18.45939
Table 6. Varying temperature effect on overall heat transfer conductance and

log-mean temperature difference for brazed plate

T inlet[˚C] UA Tlm
60 0.102725 24.46936
50 0.102128 18.45939
40 0.104978 11.97217

Table 7. Effect of varying flow rates on overall heat transfer conductance and

log-mean temperature difference for shell-and-tube

 Vhot [L/min] UA Tlm
1 0.066876 15.66092
2 0.061614 20.39797
3 0.066695 21.98484

Table 8. Varying temperature effect on overall heat transfer conductance and

log-mean temperature difference for shell-and-tube

T inlet[˚C] UA Tlm
60 0.067358 27.98809
50 0.066695 21.98484
40 0.057807 14.49425
Errors in this experiment can be attributed most likely to faulty thermocouples. If the

readings are slightly off, it can skew the results in either direction dramatically. In order

to decide upon which setup is the most effective heat exchanger, the largest UA must be

found. In this experiment, the brazed plate with a flow rate of 3 L/min and a temperature

of 40˚C is the most effective.

1.2

0.8 Cr=0
Effectiveness

Cr=0.25
0.6
Cr=0

Cr=0.75
0.4
Cr=0.9

0.2 Cr Exp

0
0 1 2 3 4 5 6 7
NTU
 Figure 6. Plot of effectiveness versus NTU for parallel flow configuration with

varying Cr values

1.2

0.8 Cr=0
Effectiveness

Cr=0.25
0.6
Cr=0.5

0.4 Cr=0.75
Cr=0.9
0.2 Cr Exp

0
0 2 4 6 8
NTU

Figure 7. Plot of effectiveness versus NTU for counter flow configuration with

varying Cr values

The two figures above, figures 6 and 7, are theoretical plots of heat exchangers

effectiveness versus NTU for parallel and counter flow setups. Both of them show the

effect that the value of Cr has on the plots. To compare, the experimental values were also

plotted against these theoretical values. It can be seen that generally the higher the Cr

value, the lower the effectiveness of the configuration for both cases. The experimental

data follows roughly the same initial slope, but there are not enough data points to show

the rest of the slope. Again, the issues can be attributed to thermocouples along with

human error in calculation or messing up the input into an equation. However, the slopes

of the graphs show that the data acquired was on the right track overall.

IV. Conclusions and Recommendations

 This lab looked into three different types of heat exchangers; brazed plate, shell-

and-tube, and concentric-tube. The log-mean temperature difference, and overall heat

transfer product were solved for in order to compare the different setups. This was

done to help determine the most effective heat exchanger, the brazed plate heat

exchanger at a rate of 3 L/min. For the concentric-tube heat exchanger, the parallel

configuration was compared against the counter flow by using plots of the load

curves. The flow rates and temperatures were changed with these in order to properly

analyze the two different setups. Lastly, the theoretical effectives-NTU curves were

created to show how Cr affects the plots as well as to compare against the

experimental data.

The only suggestion for the lab I have is being able to truly understand the

different setups. I would like to have a thorough walkthrough of what they look like

as well as how they function. That would add a lot to the understanding of the lab as

well as something to intrigue the students.