1

INTRODUCTION ......................................................................................................................... 2
OBJECTIVES ............................................................................................................................... 2
NOMENCLATURE ...................................................................................................................... 2
THEORY....................................................................................................................................... 3
APPARATUS ............................................................................................................................... 6
PROCEDURE ............................................................................................................................... 6
RESULTS ..................................................................................................................................... 7
DISCUSSION ............................................................................................................................. 11
CONCLUSSION ......................................................................................................................... 12
REFERENCES ............................................................................................................................ 12

Appendix A: Tables
Appendix B: Samples of calculation














2


INTRODUCTION

This is the second report of laboratory of Industrial Fluid Mechanics 632 which
is related to packed beds and fluidised beds. They are widely used in the industry for
filtration, pollution control, chemical processing and heat transfer. Along the laboratory
the packed and fluidised apparatus was used to obtain experimental values of pressure
drop through the packed and fluidised bed at various flowrates which are compared to
theoretical values.
OBJECTIVES

- To compare the head loss characteristics of a packed bed with those determined
from theoretical calculations.
- To compare the graphics of head loss of a fluidised bed with the theoretical.
- To compare the point of incipient fluidisation for fluidised bed with that
determined from theoretical calculations.
NOMENCLATURE

P= Pressure Loss (Pa)
f = friction factor for Chilton-Colburn equation
f
c
= friction factor for Carman-Kozeny equation
Re= Reynolds Number
U = superficial velocity, based on empty cross-section (m/s)
=density of fluid (kg/m
3
)

p
=density of particle (kg/m
3
)
h= height of the bed (m)
D
p
= nominal size of packing particle (m)
= dynamic viscosity of fluid (N.s/m
2
)
= voidage fraction
S
p
= Specific area of the single particle(m
2
/m
3
)
3

g= gravity (m/s
2
)
THEORY

a) Fluid drop through a packed bed of particles
A packed bed is a bed of solid particles in one tube, box or cylindrical vessel, through
which a fluid flows.

Fig. 1 Packed bed Tower Scrubber. Extracted from :
http://old.ecologixsystems.com/wet_scrubbers_tower_tm.php

The pressure drop by packed beds was first study by Darcy in 1856. Nowadays, the
pressure drop can be determined by following models:
1) Chilton-Colburn model
This correlation was presented in 1913 and it was based on Darcy equation for flow in
pipes



The Reynolds number was given by:
(Equation 1)
(Equation 2)
(Equation 3)
4


2) Carman-Kozeny model

This model is based on the idea the fluid flows through tubes or channels
which are formed within the bed.


Fig. 2 Kozeny hydraulic model. Extracted from class notes

Following correlations calculate the pressure drop in the packed bed according
to this model.






b) Fluidization

Fluidized bed is a cylindrical column that contains particles and through which
fluid flows with a velocity sufficiently high to suspend the particles. They are
used in chemical processing industries for separation, rapid mas and heat
transfer operations and catalytic reactions. Fig. 3 shows the phenomena.

(Equation 4)
(Equation 5)
(Equation 6)
(Equation 7)
5


Fig.3 (a) packed bed not fluidized; (b) packed bed fluidized. Extracted from: Rhodes, Martin,
ed. Introduction to Particle
The pressure drop through a fluidised bed is given by :

Figure 4(a) is obtained plotting de pressure loss across the bed versus superficial
velocity through the bed. Figure 4(b) is the plot of the bed height versus superficial
velocity through the bed.


Fig. 4 (a) Pressure drop vs superficial velocity; (b) Bed height vs superficial velocity.
The point Umf is the velocity which the packed bed becomes a fluidized bed and
it is known as the incipient fluidisation point.
Assuming laminar flow, Carman-Kozeny equation (eq 5) and the equation for fluidised
bed (eq 8) are used for predicting the value of Umf with the following expression:


(Equation 8)
(Equation 9)
6

APPARATUS

The packed and fluidised bed apparatus is shown in figure 5. It consists of a
column filled with packing material. It has a pump closed-system which supply flow in
either direction: from top to bottom (packed bed experiment) and from bottom to top
(fluidized bed experiment). This is controlled manipulating the three way valves.

Fig. 5 packed and fluidised bed apparatus schema
At the top of the apparatus there is a bleed valve which assists in bleeding
system of air. Along the body of the apparatus there are some pressure measurement
points. Each point is connected to the manometers by one plastic tube. In point F, there
is one rotameter and in point G one digital flow meter.
One important observation is that the bed height is below the point E and it is
not similar to the schema in the laboratory. The measured bed height is 0.835
PROCEDURE

a) Procedure for packed bed

1. The two directional valves were set so that flow was directed downwards
2. We started the pump and the valve at the pump discharge line was slowly
opened.
3. We adjusted the flowrate using the valves at the pump discharge.
7

4. Three people helped to do the laboratory. One person watched the flowrate in
the digital flow meter, other one watched the pressures in the manometers, and
the last one wrote the readings on the board.
5. After the step 4, we adjusted a new flowrate using the valves
6. The steps 4 and 5 were repeated until 9 readings were made.

b) Procedure for fluidized bed

1. The two directional valves were set so that flow was directed upwards.
2. We started the pump and the valve at the pump discharge line was slowly
opened.
3. We adjusted the flowrate using the valves at the pump discharge.
4. Four people helped to do the laboratory. One person watched the flowrate in the
digital flow meter, other one watched the pressures in the manometers, the third
watched the height of the bed and the last one wrote the readings on the board.
5. After the step 4, we adjusted a new flowrate using the valves.
6. The steps 4 and 5 were repeated until 8 readings were made.
RESULTS

a) Packed bed
Fig. 6 shows the values of pressure drop per meter (in Pa/m) versus flowrate (in
litre/min) using the laboratory data and the two models to determine the loss pressure in
packed beds. The dashed line is obtained plotting Chilton-Colburn equation. The
continuous line is obtained plotting Carman-Kozeny equation. Finally the triangle
points are the taken values during the laboratory.
The values used to plot the Fig.6 are in the tables in the Appendix A. The table 3
contains the values for Chilton-Colburn points used to plot the curve. The table 4
contains the values for Carman-Kozeny points. The table 1 contains the values of the
laboratory points. A sample of calculation per each model is in the Appendix B
b) Fluidized bed
The Fig. 7 represents the pressure drop vs flow rate and the Fig.8 is bed height vs
flowrate. Both are plotted using the data of the fluidized bed laboratory. This data is in
the table 6 in the appendix A.
In each graph, it shows the value Umf. This is the flowrate value at the incipient
fluidization point.
Using eq. 9 it was calculated the theoretical value of Umf which is: 7.579 l/min.
Example of calculations for the values is located in Appendix B.
8


0.00
500.00
1000.00
1500.00
2000.00
2500.00
3000.00
3500.00
4000.00
4500.00
5000.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00
P
r
e
s
s
u
r
e

D
r
o
p

p
e
r

m
e
t
r
e

(
P
a
/
m
)

Flow [l/min]
Fig. 6 Pressure drop vs Flowrate
(Packed bed)
Laboratory Chilton-Colburn Carman-Kozeny
9


0
100
200
300
400
500
600
700
0 2 4 6 8 10 12 14 16 18 20
P
r
e
s
s
u
r
e

D
r
o
p

[
P
a
]

Flow rate [l/min]
Fig. 7 Pressure Drop vs Flow rate
(Fluidized bed)
Incipient
Fluidized Point
Umf = 7.92 l/min.
10


60
70
80
90
100
110
120
130
140
0 2 4 6 8 10 12 14 16 18 20
B
e
d

H
e
i
g
h
t

[
c
m
]

Flow rate [l/min]
Fig. 8 Bed height vs Flow rate
(Fluidized bed)
Umf= 7.92 l/min
11

DISCUSSION

a) Packed bed
According to the curve of laboratory data, the pressure drop increases when the flow
also increases. This pattern is the same in the curves of the two models. Both models
have some level of accuracy related to the experimental data depending on the flow.
The model of Carman-Kozeny shows almost a high level of accuracy before 8 l/min.
After this value the difference between the model and the real values increases and it
losses accuracy. The model of Chilton-Colburn is less accurate than the first before
13.67 l/min. We can realise that using Carman-Kozeny model is more accurate than
Chilton-Colburn.
Looking at the three curves in the same graphic, we can see the line by Chilton-
Colburn results is above the curve of the real curve and the curve obtained by Carman-
Kozeny model is below of the real curve. That means the pressure drop calculated by
Chilton-Colburn is greater than the Carman-Kozeny, thus, the first one (Chilton-
Colburn) is more conservative.
b) Fluidized bed
Comparing the fig 7 with fig 4(a) it is possible to see some similarities between
both. From point 1 to 4 the graphic is similar to the behaviour of fixed bed zone in the
fig. 4(a). The pressure drop increases whilst the packed bed is not fluidized. Thus, the
point 4 is possible to be considered as the incipient fluidization point. According to the
theory, after the incipient fluidization point, the fluidized bed might be one horizontal
straight line. Looking the line described from points 4 to 8, this is a line with downward
trend. The explanation is that the particles start being more separated between them and
the loss produced for presence of the particles decreases.
The fig. 8 seems similar to the theoretical graphic Fig. 4(b). The points along the
transition between the fixed bed and fluidized bed, do not describe clearly an inclined
line with a different slope than the fluidized zone. However, it is possible to allocate the
incipient fluidization point using the value obtained in figure 7, which is 7.92 l/min.
Using eq. 9, it was obtained 7.579 l/min (theoretical incipient fluidization point). This
theoretical value is approximately 96% the experimental value, thus it could be
considered a good approximation of the model to the real behaviour.
Along the experience, two probable points of errors are detected. The first is related
to the reading of the value of the bed height, especially when the bed is higher. This
error is due to the person who read the value had difficult to take the last point (it was
too high to see; despite he was a tall man). The using of a ladder during the experience
is recommended to avoid this kind of error. The second is also related to the last taken
point. According to the readings the bed height at this point is 128 cm, however, the
12

last point to take the pressure value is 100 cm (point E). That means it was not possible
to get a reliable reading of pressure loss from 100cm to 128 cm in the fluidized bed. It is
recommended to perform the experiment within the pressure measurement points, or fill
the apparatus with more packing material until a level above the point E.
CONCLUSSION

- The models of Carman-Kozeny and Chilton-Colburn work to predict the real
behaviour of the packed bed with some precision.
- Chilton-Colburn model is more conservative (higher values of pressure drop are
calculated) than Carman-Kozeny model
- The behaviour of the fluidized bed is similar to the theoretical graphic.
- The theoretical incipient fluidisation point is approximately similar to the
incipient fluidisation point obtained by the plotting of the experimental values.
- In a fluidized bed, the pressure drop decreases after the incipient fluidisation
point because the particles begin to be more separated from each.
REFERENCES

Rhodes, Martin. 2008. Introduction to Particle Technology. Chichester, GBR: John
Wiley & Sons.
P.C. Carman. 1997. Fluid flow through granular beds, Chemical Engineering Research
and Design, Volume 75, Pages S32-S48. http://dx.doi.org/10.1016/S0263-
8762(97)80003-2.

J.F. Richardson, W.N. Zaki, 1997. Sedimentation and fluidisation: Part I, Chemical
Engineering Research and Design, Volume 75, Pages S82-S100
http://dx.doi.org/10.1016/S0263-8762(97)80006-8.

Industrial Fluid Mechanics Class notes 2014