Chgairflowlab JOSHI CAREY
Chgairflowlab JOSHI CAREY
Chgairflowlab JOSHI CAREY
Universit dOttawa
Chemical Engineering
Gnie Chimique
Faculty of Engineering
Facult de Gnie
By
David Carey (7195956)
Shail Joshi (7282674)
CHG3122
Jan 29 2016
Cover letter
To: Dr. Lan
From: David Carey & Shail Joshi, Group 4
Date: Jan 29, 2015
Subject: CHG 3122, Air Flow Measurements
This investigation was designed for students to apply their fundamental
knowledge of fluid mechanics and chemical process to measure accurate pressure drop
values in a 30-foot long pipe. These values were then used to calculate local velocities
using fluid mechanics concepts. These local velocities were in turn used to calculate
appropriate Reynolds numbers and friction factors in the pipes. The relationship between
friction factor and Reynolds number was analyzed by measuring the pressure drop in the
pipe and an orifice meter. Thus the velocity profiles were determined and analyzed at
varying Reynolds number.
In order to determine the velocity profile, it was necessary to calculate traverse
points in order to ensure that equal area was maintained. The traverse points were as
follows: 0.6, 1, 1.5, 2.1, 3, 4.2, 5.4, 6.3, 6.7, 7.2, 7.8. The pressure drop was recorded at
each of these traverse points, and was used to estimate the velocity profile at varying
airflow.
Airflow under contracted conditions was also investigated using the orifice meter.
The differential pressure was measured across the orifce and these values were used to
calculate the discharge coeffients at varying flow rates of air. All the measurements were
made with the aid of an electronic pressure transducer
The local velocities yielded turbulent flow regime in the pipe and the friction
factors suggested that the inside of the pipe was rough. Furthermore the discharge
coefficients calculated in orifice matched literature values.
Table of Contents
Cover letter ................................................................................................................. 1
Equipment .................................................................................................................. 3
Procedure ................................................................................................................... 4
Summary Of Results .................................................................................................... 5
Discussion ................................................................................................................... 7
Conclusions And Recommendations .......................................................................... 10
Appendix ................................................................................................................... 11
Tables ........................................................................................................................................................................... 11
Supporting Figures ................................................................................................................................................. 14
Sample Calculations ............................................................................................................................................... 16
References .................................................................................................................................................................. 19
Equipment
Figure 1: Experimental Setup. Not to scale
The experiment consists of 4 critical
components. The first of which is the
long cylindrical pipe through which the
air flows.
serves
as
an
airflow
Procedure
1. Ambient Conditions Laboratory conditions were recorded
2. The positions for the 10-point traverse of the Pitot tube were calculated and the
appropriate pencil marks were made on the Pitot tube ruler.
3. The motor and blower were turned on to start delivering air to the pipe
4. The dampener was opened so as to let the maximum airflow into the pipe
5. The bulk air pressure just above the blower, the pressure drop across the orifice
and the pressure drop along the pipe were all recorded.
6. The pressure drop across the Pitot tube was measured along the 10 point traverse.
7. The dampener was closed progressively until the reading on the Pitot tube was 80%
of the original maximum value.
8. Step 6 is now repeated and the pressure drop is measure along the traverse.
9. The dampener is now closed till the reading reaches 60 percent of the maximum
value.
10. The above steps are then repeated until the dampener is fully closed.
Summary Of Results
1. A pitot tube operates by measuring the stagnation pressure at the entrance of the
tube and comparing it to the static pressure that surrounds the tube in the flow
stream.
Advantages:
-cost effective pressure measurements
-no moving parts
-simple to use and install
-low pressure drop
-works in high temperatures
Disadvantages:
-can be clogged if the flow stream includes particulates
-if the flow rate is too low or too high the measurements on the transducer
could be incorrect
2. The averages velocity and Reynolds numbers calculations can be found on the Excel
sheet and in the Appendix. The centre point measurement was not included, as the
velocity measurements are found inbetween the radii of equal area cirlces. This
makes each measured velocity an average of the the velocity found at each radius,
meaning that the velocity on either side of the centre point includes the centre point
velocity in its calculation. Including the centre point in a total average calculation
would raise the velocity to a larger value than it should be.
3. If equal areas were not used to find the transverse points, then any distances used
would have to be referenced to the centre to create an average velocity profile.
Otherwise the weighting of the velocities in the average equation would be
incorrect.
Discussion
The experiment began by determining the locations that would be needed for an
equal-area ten point traverse of the tube. With these 5 circles extending from the center,
the locations could be determined for the points of measurement. The radii of these
circles on either side of center became the target points where the Pitot tube was set to
record the pressure drop across the pipe. Since these velocities are actually the average of
velocity at the radii on either side of it, the center point velocity can be ignored for the
calculation of average velocity as it is included in the measurement on either side of it.
These was held true for each of the flow rates throughout the experiment. The largest
change made to the procedure was that the blower was unable to reach 10% of its
pressure drop before the dampener was fully closed. This made the final step closer to
18% instead of 10%.
While making measurements of the pressure differences recorded by the Pitot
tube during the experiment, it was often very difficult to get a constant readout from the
pressure transducer. Quite often, guesswork and approximated averaging was used to
determine the value to be recorded. This lead to a good deal of potential error in the final
data points and calculated velocities. A way to improve this may be to have software
recording the fluctuations at each point. Once a desired time is reached, the operators can
take the mean and standard deviation to find a far more accurate value than what is taken
by guessing where the display is most constant.
When all the pressure differences were recorded, the local velocities could be
determined by using equation 1 in the lab manual, which assumes the fluid is isothermal
and incompressible. The velocities were also calculated using equation 2, which is used
when the fluid is compressible and adiabatic, gave nearly identical results to equation 1,
with an error on the scale of 0.0006% and 0.005%. This verified the assumption that the
gas was incompressible and isothermal. This assumption also lead to density being
assumed to be constant throughout the pipe at 1.225kg/m3, based on the conditions. The
Reynolds numbers were calculated from these assumptions, with the average velocities
coming from equation 1, and the density being assumed to be constant. The numbers
varied from 1.3*104 to 2.9*104, making the flow regime at all flow rates turbulent. This
turbulence was very important for the other calculations, and may be a slight source of
error for other aspects.
The orifice plate installed in the pipe plays an important role in determining
velocity of the air when it is contracted. This velocity was determined by examining the
incoming velocity and adjusting according to the size of the contraction of the orifice
plate. The Reynolds numbers for these new velocities were also calculated, using the
diameter of the orifice to determine the values instead of the diameter of the pipe. For this
case, the discharge coefficient was calculated and plotted on a graph against the
Reynolds numbers for each flow percentage. Compared to literature values and charts of
the same type (figure 4), the results matched almost perfectly, apart from the points from
the 20% and 18% flow rate trials, which are believed to be outliers. They may be
explained by low velocity having a skewing effect on the transducer readout from the
orifice plate, but the other data points follow the established literature very well, and
approach the ideal value, which is around 0.62, as the flow rate increases.
The friction factors of this system were calculated in a number of different ways,
resulting in different numbers each time. This is most likely due to the difference in how
the equations were created analytically, and what different variables they depend on. The
main friction factor was calculated using equation 11 in the lab handout, which, unlike
the other methods, does not use the Reynolds number to calculate friction factor. By
analyzing the pressure drop across the entire length of the tube, an approximation for
friction can be found. This method was compared to equation 15, which is based on the
assumption of smooth piping and a Reynolds number in the range of 3*103 to 3*106,
which fit the previously calculated Reynolds numbers of the average velocities perfectly.
The smooth pipe approximations vary from the calculated friction factors by an average
of 97.64%, which is incredibly significant. This would mean that the inside of the pipe
cannot be assumed to be smooth, as roughness causes the friction factor to become much
larger. This is also shown on a log friction factor vs. log Reynolds number graph, on
which both calculations are present. The expected smooth straight line is seen by the
smooth pipe approximation, but as the flow rates increase as used in equation 11, the
friction factors increase as well, indicating more surface disruption.
When the friction factor data was analyzed, it was then modeled in the form of the
Van Karman equation, using a y-axis of
1
f /2
resulted in a much different form, with a slope of -1.419 and a y-intercept of 14.475. The
difference from the Van Karman equation shown in equation 13 of the lab handout and
this empirically found equation is most likely explained again by the surface roughness of
the pipe, bringing about much different friction factors than were used to originally find
the given equation.
calculations were all done under the assumption that the density of air stays constant at
1.225kg/m3 . This value was verified by literature, however instead of relying on literature
values, the density of air could have been measured using an aerometer.
The friction factors within the pipe were calculated using numerous equations and
methods. However the main friction factor was calculated by analyzing the pipe as
whole and using the pressure drop from the bottom to the top. This method relied
primarily on the assumption that the inside of the pipe was smooth and did not affect the
calculations. However when this assumption was compared with friction factors
calculated using equation 15 it was found that the both values varied considerably. This
meant that the inside of the pipe was rough and the smooth pipe assumption was false.
Thus in order to generate better friction factor values the roughness of the pipe could be
estimated using moody diagrams.
In depth analysis of the system was done and appropriate charts and figures were
plotted; thus the investigation was a success and generated acceptable data.
Appendix
Tables
Table 1 - Operation Parameters
Parameters
Patm (Pa)
Tatm (K)
air (Pas)*
Do (m)
Dc (m)
L (m)
103591.48
295.95
0.000018
0.04572
0.0762
6.26
*Evaluated at 20 oC
Table 2 - Local pressure at each radial position and percentage flow rate
%
maximum
flow rate
0.6
1.5
2.1
3.0
4.2
5.4
6.3
6.7
7.2
7.8
Kpa gauge
100%
0.0135
0.018
0.02
0.024
0.026
0.028
0.026
0.023
0.021
0.02
0.016
80%
0.012
0.015
0.017
0.019
0.021
0.022
0.022
0.019
0.018
0.016
0.013
60%
0.009
0.011
0.013
0.014
0.016
0.017
0.016
0.015
0.014
0.012
0.01
40%
0.006
0.008
0.009
0.009
0.010
0.011
0.011
0.010
0.009
0.0085
0.0075
20%
0.004
0.0045
.0055
.0055
.0055
0.0055
0.005
0.005
0.005
0.045
0.004
10%
0.0035
0.004
0.004
0.0045
0.0045
0.005
0.005
0.0045
0.0045
0.004
0.004
P Top (KPa)
Orifice
(bulk)
Discharge
(Torr)
.21kpa
0.04
2.1
0.031
1.7
100
80
0.165
60
0.122
0.023
1.1
40
0.066
0.015
06
20
0.017
0.006
0.2
18
0.015
0.004
0.1
(m/s)
(m/s)
Re
Re (orifice)
flow rate
(Orifice)
100
80
60
4.649622728
40
3.820459847 19276.80132
20
18
23460.4883
39100.81383 12.91561869
32128.0022
10.61238846
Supporting Figures
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
10000
20000
30000
40000
50000
60000
Reynolds Number
9.5
9.6
9.7
9.8
9.9
10
-2
-4
-6
log smooth
10.1
10.2
10.3
10.4
y = -1.4187x + 14.475
0
0
10
12
Sample Calculations
Any parameters dependent on flow profile were calculated using the centre point
maximum flow rate and 100% flow rate.
Local velocity at the traverse points
u=
2PP
2 * 0.028kPa * 1000
u=
Pa
kPa
1.225kg / m 3
u = 6.76 m / s
u is the local velocity
Pp is the pressure difference measured by the Pitot tube
Average velocity through pipe
1 n
ui
n i =1
1
u = * (4.69 m / s + 5.42 m / s + 5.71m / s + 6.26 m / s + 6.52 m / s + 6.52 m / s + 6.13m / s
10
+ 5.86 m / s + 5.71m / s + 5.11m / s)
u = 5.79 m / s
u=
Re =
DP u
u A p = u o Ao
uo = u (
DP 2
)
Do
u o = 5.79 m / s * (
0.0762m 2
)
0.04572 m
u o = 16.09 m / s
1.225kg / m 3 * 0.04572 m * 16.09 m / s
Re o =
1.8 * 10 5 m 2 / s
Re o = 48715.11
u o is the average velocity of the orifice
Reo is the Reynolds number of the orifice
Ap and Ao are the cross sectional area of the pipe and orifice
Do is the diameter of the orifice
Orifice discharge coefficient
uo =
Co
2Po
1 4
Co = uo 1 4
2Po
C o = 16.09 m / s 1 0.6 4
1.225kg / m 3
2 * 266.65 Pa
C o = 0.72
Co is the discharge coefficient of the orifice plate
is the ratio of orifice diameter to pipe diameter
Po is the pressure drop across the orifice
Friction Factor of flow in pipe (equation 11)
f =
Ps D p
2 Lu 2
40 Pa * 0.0762m
2 * 9.144m *1.225kg / m 3 * 5.79 m / s
f = 4.57
f =
0.125
Re 0.32
0.125
f = 0.0014 +
29229.07 0.32
f = 0.006
f = 0.0014 +
ueq1 = 6.76 m / s
ueq 2
2 Po Ps
=
( )[( )
1 o Po
ueq 2 =
1]
2 *1.4 101535Pa
28 Pa + 101535Pa
(
)[(
)
3
1.4 1 1.225kg / m
101535Pa
1.4 1
1.4
1]
ueq 2 = 6.76 m / s
is the ratio of the pressure and volumetric heat capacities, in this case of an ideal gas
Po is the bulk pressure through the pipe
Ps is the stagnation pressure as measured by the Pitot tube
References
Lan, C., Air Flow Measurements Chemical Engineering Practice, Ottawa ON. (2016)
McCabe, W.L., Smith, J.C. and Harriott, P., Unit Operations of Chemical Engineering
Sixth ed., McGraw-Hill, New York NY, 2001.
Nevers, N. d. Fluid Mechanics for Chemical Engineers, McGraw-Hill.(2004)
Holinsgard, C., Discharge Coefficient Performance of Venturi, Standard Concentric Orifice Plate, VCone, and Wedge Flow Meters at Small Reynolds Numbers (2011)