Reynolds Vs Heat Transfer Coeff

Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

Analysis of the Relationship Between the Overall Heat Transfer

Coefficient and the Reynolds Number Using a 1,2 Heat Exchanger


Team 4:
Sabrina Dhanani
Derek Eguae-Obazee
Rabya Gillani
Nathan Shedd
06-363: Transport Processes Laboratory
April 3, 2006
Dr. Powers, Mr. Cline
Abstract
The purpose of this experiment was to determine the relationship between Reynolds number and overall heat transfer
coefficient in a steam-water 1-2 shell and tube heat exchanger. Overall heat transfer coefficient from our data ranged from 550
to 850 BTU/(hr*ft
2
*F). We compared overall heat transfer coefficients to the Levenspiel and Petukhov equations for
transitional flow. The values ranged from 575 to 800 BTU/(hr*ft
2
*F) and 675 to 900 BTU/(hr*ft
2
*F) respectively. This
was over a Reynolds number range from 4500 to 5900, all in transitional flow.
2
Table of Contents
Abstract
Introduction ....1
Background and Theory .1
Equipment List ...2
Procedure ............4
Results ....5
Discussion ......7
Conclusion/Future Recommendations...10
Nomenclature ........11
References .....12
Appendix .......13
A. Sample Calculations .....13
B. Calibration Data14
C. Raw Data ..14
Introduction
The idea is to determine the heat transfer coefficient on the tube side in order to draw a comparison with the Reynolds
3
number, and then compare our data to two equations for transitional flow. A shell and tube heat exchanger is the most common
type of heat exchanger in oil refineries and other large chemical processes. Two fluids of different starting temperatures flow
through the exchanger, one through the tubes and the other through the shell. In this experiment water was on the tube side
while steam was on the shell side. Heat is transferred from one fluid to the other so that waste heat can be put to use; this
method conserves energy, which is always the goal. This experiment was commissioned and supervised by Dr. Powers and
Mr. Cline in the Rothfus Laboratory of the Chemical Engineering Department at Carnegie Mellon University.
Background and Theory
The expected value for the overall heat transfer coefficient (U) for both convection and conduction can be calculated according
to the following equation:
1/U = 1/ho + xw/kw + 1/hi (1)
We neglect xw/kw and 1/ho as we are only interested in the heat transfer that occurs inside the pipe and hi can be calculated
from several correlations. To calculate the value of U from experimental data, we combined equations (2) and (3) to obtain (4):
q = U*A*Tm (2)
q = m*Cp*T (3)

U = (m*Cp* T) / (A* Tm) (4)
Normally, the average change in temperature is calculated as a log-mean temperature:
Tlm = (T2- T1) / ln(T2 / T1) (5)
However, because we are not dealing with co-current flow or counter-current flow but rather a combination of both as well as
cross flow, we must use a correction factor to obtain the true mean temperature.
Tm = Tlm*FT (6)
As long as the vapor is condensing, its temperature stays at the boiling point so F
T
=1 for condensing steam as the hot fluid.

The Reynolds number is the ratio of inertial forces to viscous forces and is used for determining whether a flow will be laminar
or turbulent:
4
Re =*V*D/A
cs
* (7)
Laminar flow occurs at Reynolds numbers below 2100. Reynolds numbers above 6000 are considered turbulent, and
between 2100 and 6000 is considered transitional flow.
We compared the overall heat transfer coefficients from our experimental data to two equations for transitional flow, the
Levenspiel correlation and the Petukhov correlation respectively:
h
k
D
|

.
`

,
0.116 Re
2
3
125
|

.
`

, Pr
1
3
:
(8)

h
k
D
|

.
`

8
Re 1000 ( ) Pr
1 12.7

8
Pr
2
3
1
|

.
`

, +
1
D
L
|

.
`

,
2
3
+

]
]
]
]
]
:
(9)
with

1
1.82 log Re ( ) 1.64 ( )
2
:
(10)
Equipment List
5
heat exchanger

computer software, Lab View


3000 mL graduated cylinder for calibrations
Stopwatch
Computer for data acquisition

Figure 1: 1:2 shell and tube heat exchanger. Shell side fluid steam and tube side fluid cold water.
Procedure
6
We began by calibrating the flow rate and ran the cold water until the temperature was constant. We varied cold-water
flow and recorded hot water temperature and condensate temperature. Flow was varied from about 40 to 55 cubic feet per
hour. Steam pressure was monitored to make sure the steam temperature was constant. For each trial, we recorded flow rates
for cold water and hot water as well as temperatures for the steam.
Results
Two trials were run at two different steam pressures. In each trial we went up in flow rate and then back down. Our
Reynolds number varied from 4500 to 5900 and heat transfer coefficients varied from 550 to 850 Btu/(hr ft
2
F). Our average
Prandtl number was around 5 and our L/D was 133. Here is the graph of overall heat transfer coefficient versus the Reynolds
number compared to the Levenspiel and Petukhov correlations for transitional flow.
500
550
600
650
700
750
800
850
900
4500 4700 4900 5100 5300 5500 5700 5900
Reynolds Number
H
e
a
t

T
r
a
n
s
f
e
r

C
o
e
f
f
i
c
i
e
n
t

(
B
t
u
/
h
r
*
f
t
^
2

*
F
)
Levenspiel T2 Trial 1 Trial 2 Levenspiel T1 Petukhov T2 Petukhov T1
7
Figure 2:Heat Transfer Coefficient vs. the Reynolds Number for the tube side fluid of a 1:2 Shell and Tube Heat Exchanger.
The next graph shows some of the data for trial two with the error bars. We calculated a four percent error in the calibration of
the flowrates. This error effected the calculation of both the overall heat transfer coefficient and the Reynolds number.
Levenspiel appears to approximate our data within the error of our experiment while Petukhov over approximates.
500
550
600
650
700
750
800
850
900
950
4400 4600 4800 5000 5200 5400 5600 5800 6000 6200
Reynolds Number
H
e
a
t

T
r
a
n
s
f
e
r

C
o
e
f
f
i
c
i
e
n
t

(
B
t
u
/
h
r
*
f
t
^
2

*
F
)
Levenspiel T2 Trial 2 Petukhov T2
Figure 3: Heat Transfer Coefficient vs. the Reynolds Number for the tube side fluid of a 1:2 Shell and Tube Heat Exchanger. Only includes
partial data from trial two with error bars.
Discussion
We compared overall heat transfer coefficients from our experimental data to two equations for transitional flow. The
Levenspiel equation fit our data within 12% for trial 1 and 8% for trial 2. The Petukhov-Gnielinski equation did not predict our
8
data as well. It was off by 18% for trial 1 and 12% for trial 2.
We calibrated the flow rate by varying the electrical signal from the Lab View software and recorded flow rate with a
graduated cylinder and stopwatch. We used an average value for the conversion factor, which had an uncertainty of 4%.
We calculated the area of the tubes by estimating the tube diameter as 0.25 inches, wall thickness as 0.035 inches,
tube length as 2 feet, and number of tubes as 48. The overall heat transfer coefficient is dependant on tube area. Without
being able to see the tubes, we could not be certain on the accuracy of this area. Figure 4 shows how even a small error in our
estimation would affect the heat transfer coefficient.

Figure 4: The effect on overall heat transfer coefficient, U, due to uncertainty in the area.

Since the fluid properties are dependant on temperature and temperature is changing, we needed to find an average
temperature to calculate the fluid properties. The arithmetic mean of the outlet and inlet temperatures is the suggested method
for determining the average temperature.
The steam temperature was calculated assuming saturated steam at constant pressure. For trial 1, the steam pressure did
not change, but in trial 2, the steam pressure went from 16.7 psia to 17.2 psia. This increased the log mean driving force for
trial 2 while keeping the temperature difference almost the same as trial one. Figure 5 shows the change in the log mean
temperature difference compared to the temperature difference for each trial.
9
40
60
80
100
120
140
160
2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500
Mass Flowrate
T
e
m
p
e
r
a
t
u
r
e

C
h
a
n
g
e

(
F
)
Delta T T2 Delta T T1 LMTD T2 LMTD T1
Figure 5: The temperature difference and log mean temperature difference (LMTD) vs. mass flow rate for both trials.
This ratio of the temperature difference to the log mean temperature difference for trial one was 1.08 and for trial two it was
1.24. This increase from trial to trial accounts for the difference in heat transfer coefficients. We also had to assume that the
cold water input stream was at a constant temperature since we did not have a thermocouple in that stream. In our previous
experiment we noticed that the water temperature dropped when it was first turned on and eventually became constant once it
was running for awhile.
Conclusion:

Overall heat transfer coefficient from our data ranged from 550 to 800 BTU/(hr*ft
2
*F), while the Levenspiel and
Petukhov equations ranged from 575 to 800 BTU/(hr*ft
2
*F) and 675 to 900 BTU/(hr*ft
2
*F) respectively. This was over a
Reynolds number range from 4500 to 5900, all in transitional flow. We werent able to get data from pure laminar flow, pure
turbulent flow, or even the full range of transitional flow. The flow ranged from 40 to 54 cubic feet per hour. We found that
the Levenspiel equation more accurately modeled our data. There was only a 12% difference in our data from Levenspiels
equation, while there was a 22% difference in our data from Petukhovs equation.
10
Future Recommendations
The data we obtained was all in the transitional flow regime. Due to the constraints of the heat exchanger, we were not
able to gather data on a larger scale. If the flow rate of the cold water in was too low, the water would vaporize. If the steam
flow rate could be turned down, then the minimum cold water in flow rate could be decreased giving a wider range of Reynolds
number. In addition, we could conduct an experiment in which we take into account the shell side heat transfer coefficient and
conduction term to determine the overall heat transfer coefficient.
References:
1) Geankoplis, Chritie. Transport Processes and separation process principles. Prentice Hall, Upper Saddle River, NJ
(2003)
2) Welty, James, Charles E. Wicks, Robert E. Wilson, Gregory L. Rorrer. Fundamentals of Momentum, Heat, and
Mass Transfer. New York: Hamilton Printing, 2001.
3) www.cbu.edu/~rprice/lectures/convection.html
4) http://www.cheresources.com/convection.shtml#tube
5) https://blackboard.andrew.cmu.edu/courses/1/S0606361/content/_179232_1/Notes2006.pdf
11
Nomenclature
Symbol Description
ft foot
lb pound
s second
min minute
gpm gallons per minute
Symbol Description
A The surface area of the heat exchanger, ft
2
A
cs
Cross-sectional area of one tube
BTU British Thermal Units
c
p
specific heat of fluid, (BTU/lbF)
D diameter of tube, ft
F degree Farenheit
h
i
Heat transfer coefficient of the inner wall
h
o
Heat transfer coefficient of the outer wall
k Thermal conductivity
Re the Reynolds number
Pr the Prandtl number
Q Heat Flow
T
in
the inlet temperature of the fluid, F
T
out
the outlet temperature of the fluid, F
T
in
temperature difference at one end of heat exchanger, F
T
out
temperature difference at the other end of heat exchanger, F
T
LM
the log mean temperature difference, F
U Overall heat transfer coefficient (BTU/ft
2
hr
o
F)
V Volumetric flow (gpm)
x
A
Width of the tubing between streams
Symbol Description
Fluid density, lb/ft
3
Fluid viscosity
12
Appendix A - Sample Calculations
k 0.366
BTU
hr ft R
:
T
wi
512 R :
T
wo
624 R :
T
si
679 R :
T
so
650 R :
.000441
lb
ft s
:
D .0179 ft :
62.1
lb
ft
3
:
c
p
1
BTU
lb R
:
L 2 ft :
N 48 :
A
cs
D
2

4
:
A N D L :
A 5.399ft
2

A
cs
2.516 10
4
ft
2

V .0116
ft
3
sec
:
Re

V
N
2
|

.
`

,
A
cs

]
]
]
]
D

:
Re 4.841 10
3

Pr
c
p

k
:
Pr 4.338
Levenspiel Correlation
h
k
D
|

.
`

,
0.116 Re
2
3
125
|

.
`

, Pr
1
3
:
h 623.469
BTU
hr ft
2
R

Petukhov Correlation
13

1
1.82 log Re ( ) 1.64 ( )
2
:
h
k
D
|

.
`

8
Re 1000 ( ) Pr
1 12.7

8
Pr
2
3
1
|

.
`

, +
1
D
L
|

.
`

,
2
3
+

]
]
]
]
]
:
h 700.338
BTU
hr ft
2
R

m V :
m 2.593 10
3

lb
hr

T
lm
90.225R
h
mc
p
T
wo
T
wi
( )

]
]
A T
lm

:
h 596.305
BTU
hr ft
2
R

Experimental
T
lm
T
si
T
wo
( )
T
so
T
wi
( )

]
]
ln
T
si
T
wo
( )
T
so
T
wi

]
]
]
:
Appendix B Calibration Data
14
flow controller
(mA)
volume
(mL) time (s)
flowrate
(mL/s)
conversion factor
(m^3/s/mA)
9 2500 5.89 424.4482173 4.71609E-05
10 2850 6.45 441.8604651 4.4186E-05
10 3000 6.83 439.238653 4.39239E-05
8 2900 7.48 387.7005348 4.84626E-05
7 2450 7.56 324.0740741 4.62963E-05
4.60059E-05
flow controller
(mA)
volume
(mL) time (s)
flowrate
(mL/s)
conversion factor
(m^3/s/mA)
6.75 2850 10.13 281.3425469 4.16804E-05
7 3150 9.75 323.0769231 4.61538E-05
7.25 3100 8.89 348.7064117 4.80974E-05
7.5 3500 9.55 366.4921466 4.88656E-05
7.75 3400 9.17 370.7742639 4.78418E-05
8 3600 9.24 389.6103896 4.87013E-05
8.25 3450 8.84 390.2714932 4.73056E-05
8.5 3325 8.19 405.982906 4.77627E-05
8.75 3450 8.48 406.8396226 4.6496E-05
9 3500 8.5 411.7647059 4.57516E-05
4.68656E-05
Appendix C Raw Data
See Attached
15

You might also like