MEC2404

Frictional Flow in Pipe

Lab Report

Done by:

Ramith Bulumulle – 24698512

Don Gayanga Millevithanatchy – 24829781
Zhi Ying Lim – 23725540

26/10/2014
Abstract
The principle examined in this experiment was how the frictional flow of a fluid in a pipe relates to the
relationship with the fluids Reynolds number. Two types of fluid flows were observed: laminar and
turbulent. The Moody chart was used to determine the theoretical friction factor and compared with the
values obtained experimentally. It was found that below 2000 Re the flow was laminar and that the
friction factor corresponds to the theoretical value of 64/Re. Beyond 2000 Re the flow fluctuates. The
maximum calculated values for the friction factors were 0.118 and 0.034 for the laminar and turbulent
flows respectively, while the lowest friction factor calculated for the laminar flow was 0.0368 and the
lowest for the turbulent flow was 0.0265. The main reason for the discrepancies in the results was due to
errors which occurred during the experiment such as parallax errors when reading the measuring scales,
impurities and air bubbles present in the water and tube respectively, inaccuracies in timing, and finally
the fact that the assumptions made during the calculations aren’t completely valid in reality where the
experiment takes place.

Introduction
The key objective of this experiment was to observe and comprehend the nature of a frictional flow of a
fluid in a pipe and the relationship with the fluids’ Reynolds Number. As a fluid flows through a pipe,
energy losses will occur in the form of e.g. heat and sound. These losses in energy are referred to as
head losses, which were derived using Bernoulli’s principle. Head losses can be classified as major head
losses or minor head losses. Minor head losses can be defined as the energy lost due to fixtures such as
bends and valves present in the system, while major head losses are defined as losses of energy due to
the frictional resistance which acts against the flow of the fluid, which can be calculated using the
following formula (which is known as the Darcy-Weisbach equation):

𝑙 𝑉2
ℎ𝐿 𝑚𝑎𝑗𝑜𝑟 = 𝑓
𝐷 2𝑔

Where 𝑓 is the friction factor, 𝑙 is the pipe length, 𝑉 is the average velocity of the fluid, 𝐷 is the pipe diameter and
𝑔 is the acceleration due to gravity.

The Darcy-Weisbach friction factor is the friction factor ( 𝑓) used in the previous equation, which is 4 times
greater than the Fanning friction factor. This friction factor depends on the velocity of the fluid flow and
therefore, is not a constant. For a laminar flow, the relationship between the friction factor (𝑓) and Reynolds
number (𝑅𝑒), is highlighted in the following equation:

64
𝑓=
𝑅𝑒
The Reynolds number of a flow is a ratio of the fluids’ inertia force to the fluids’ viscosity force, which can be
calculated using the formula:

𝜌𝑉𝐷
𝑅𝑒 =
µ
Where 𝜌 is the fluid density, 𝑉 is the velocity of the fluid, 𝐷 is the pipe diameter and µ is the fluid viscosity.
Reynolds number can be used to describe the behavior of the flow, which can be either laminar or turbulent
depending if the Reynolds number is above or below a critical value. (Above the critical value would be a
turbulent flow while below the critical value would be a laminar flow. Around the critical value, properties of
both flows can be observed). For pipe flow, the critical Reynolds number is around 2000. For turbulent flows, it
has been concluded via experiments that the friction factor (𝑓) is dependent on the Reynolds number as well as
 k
relative roughness(D or D); which are outlined on the Moody chart.

The Moody chart/diagram was developed by Lewis Ferry Moody during the 1940’s. The chart displays the
relationship between the Moody friction factor (𝑓) (the Darcy-Weisbach friction factor), Reynolds number (𝑅𝑒)
 k 𝑙 𝑉2
and the relative roughness of the pipe (D or D). The Darcy-Weisbach equation for head loss (ℎ𝐿 = 𝑓 𝐷 2𝑔 ) was
introduced in the mid- 19th century by Henry Darcy and Julius Weisbach. However, this equation failed to
effectively provide data for variations of the friction factor with the velocity. Through data gathered through the
works of other engineers such as Antoine de Chezy and Gaspard de Prony, Henry Darcy eventually implemented
the idea of pipe roughness adjusted to the diameter of the pipe, which is what we now refer to as the relative pipe
 k
roughness ((D or D) . The correlation between these terms were showcased in the Moody chart, developed by L.F.
Moody. The chart was divided into 4 Reynolds number zones, and the curves in each zone were establish using
the data and accepted correlations which were establish through experiments, allowing numerous fluid mechanics
design calculations and problems to be solved more conveniently. As the values in the chart are approximations
for a range of data values, and the fact that for Reynolds numbers between 2000 and 4000, the friction factor is an
indeterminate value, it can be said that the accuracy of the results using the Moody chart are quite acceptable for
simple calculations, but for more critical calculations, another method should be used such as the direct
application of the Colebrook equation, without the use of the Moody Diagram. Assumptions made in order to
develop the chart include that the fluid under observation was incompressible and that the system was operating
under steady state.

Pipe losses can be reduced in several ways, such as by incorporating fewer bends and turns in the pipe
system, which would in turn reduce minor head losses. The inside surface of the pipe can be constructed
with a material of relatively lower roughness, which will result in a lower friction factor and thus head
loss. These materials however, can be expensive; hence making the operation less economical.
Increasing the inner diameter of the pipe and reducing the velocity of the fluid flow can also reduce head
losses, as seen by the first equation. This would lead to more material being used in constructing the
pipe, and therefore increase the cost of construction and environmental impact. A suitable compromise
between the pipe diameter/fluid velocity should be reached in order to ensure that the process is
economical and efficient.
Experimental Procedures
Two experimental rigs were used in this experiment: The Lotus Scientific Rig, which was used to
observe turbulent flow and the Armfield Experimental Rig, which was used to examine laminar flow.

Figure 1.The Armfield Experimental Rig Figure 2.The Lotus Scientific Rig

For the Armfield Experimental Rig, the control valve was first closed and the initial value of the water
manometer reading was recorded. The control valve was then opened, allowing water to flow, and the
difference in height between the two manometers were controlled using the valve and recorded. The
volume of water collected from the rig in 60 seconds as well as its temperature was measured using a
measuring cylinder and thermometer respectively, and recorded. Finally, this process was repeated while
varying the difference between the manometer readings in each trial until 10 sets of data were obtained.

For the Lotus Scientific Rig, the control valve was adjusted until a reference point was obtained on the
manometer scales. Upon completion of the initial setup, the control valve was turned and the difference
between the heights of the manometer readings were manipulated and noted down. The exit outlet of the
water collection tank was plugged in, while simultaneously a stop watch was started, in order to record
the time taken for 10 liters of water to collect in the tank. The temperature of this collected water was
also recorded. This process was repeated while varying the difference in heights of the manometer
readings in each trial until 8 sets of readings were recorded.
Results
This section displays the graphs obtained for this experiment, which were derived using the raw data
obtained.

Graph 1. Graph of plotted results for laminar flow and turbulent flow.

In the turbulent regime, most of the data is plotted approximately at the relative toughness of 0.002.

Pressure vs. Flow Rate for Laminar Flow

3500

3000

2500
Pressure(Pa)

2000

1500

1000

500

0
0 0.000001 0.000002 0.000003 0.000004 0.000005 0.000006 0.000007
Flow Rate (m3/s)

Graph 2. Graph of pressure vs. flow rate for laminar flow.

 Pressure vs. Flow Rate for Turbulent Flow
25000

20000
Pressure(Pa)

15000

10000

5000

0
0 0.00005 0.0001 0.00015 0.0002 0.00025 0.0003
Flow Rate(m3/s)

Graph 3. Graph of pressure vs. flow rate for turbulent flow.

Discussion
By looking at the Moody chart we can observe that the data that has been plotted below 2000 Re was close to 𝑓 =
64/𝑅𝑒 as shown by the red dots marked. It has been plotted close to the line with slight deviations that could have
occurred due to errors present when carrying out the experiment. The water flow appeared to be unstable when
the Re was between 2000 and 4000, the friction factor was also fluctuating. This type of flow region is known to
be the transition region. If the experiment was conducted with only minor errors, the transition region period
could be observed at a higher number. For this plot, the estimated 𝜀/𝐷 was also found to be approximately
between 0.004 and 0.001. These values were estimated since the points marked in green colour correspond well
with the two lines between o.001 and 0.004. Based on the results obtained and plotted on the Moody chart, there
is a clear resemblance between the results obtained and the theoretical values in the Moody chart.

As for the volumetric flow rate, Q vs pressure difference, ∆P, the plots were not accurate for higher values of Q.
hence some data points that shows transition flow was omitted and only those that showed the characteristics of
laminar flow was used. Average viscosity for laminar was calculated to be 0.007998 while the theoretical value is
estimated to be 0.0009586722. The difference is 0.00015887 which is an error of 16.57%. This could be due to
round off errors and change in temperatures.

The water may not have been in the ideal conditions for this experiment to be conducted. Various chemicals may
have been dissolved in the water and dust particles or other substances could also be mixed. This could affect the
viscosity of the water deviate the results from the theoretical data. Furthermore during the experiment it was noted
that the pressure and the fluid in the manometer were fluctuating. As a result, recording the results were found to
be rather difficult and incorrect values may have been taken. This case is more obvious when a water manometer
is used, since water has a much lower density than mercury hence more sensitive to small changes, slight
differences of pressure causes it to change rapidly. There is also a good chance for human errors such as parallax
errors to occur due to the difficulty faced while taking readings of a moving fluid. All types of errors that were
discussed above would surely affect the results significantly since they are used for further calculations. One way
to reduce discrepancies and obtain more accurate results is by repeating some processes that are more likely to
have errors like for example time taken to collect a volume of water, then take the average time. Finally, if the lab
equipment was maintained well, we would be able to reduce the systematic errors caused.

Conclusion
The experiment was conducted to analyze the fluid flow properties of the pipe. The energy equation was used in
order to calculate the friction factor. The values plotted on the Moody chart showed some resemblance to the line
on the chart. It was found the below 2000 Re the graph appears to indicate laminar flow and the friction factor
roughly corresponds to the theoretical 64/Re. Beyond that point the flow was not laminar, it was the transitional
stage, hence the values shown were different. The estimated 𝜀/𝐷 value for 7.74 mm diameter of the pipe is
between 0.001 and 0.004 and for 4.48mm diameter was a smooth pipe. Other than that, the theoretical viscosity
which was found to be 0.007998 had an error 16.57% compared to the theoretical value which is estimated to be
0.0009586722. The deviation caused could be due to the assumptions made, systematic errors and human errors.
This can be reduced by improving the experimental procedure and equipment as discussed in the discussion.

References:
Engineeringtoolbox.com, (2014). Moody Diagram. [online] Available at:
http://www.engineeringtoolbox.com/moody-diagram-d_618.html [Accessed 26 Oct. 2014].

Anon, (2014). [online] Available at: http://www.eng.auburn.edu/~tplacek/courses/3600/colebrook.pdf

[Accessed 25 Oct. 2014].

Munson, B., Okiishi, T., Huebsch, W. and Rothmayer, A. (2013). Fundamentals of fluid mechanics.
Hoboken, NJ: John Wiley & Sons, Inc.

Batchelor, G. (2005). An introduction to fluid dynamics. Cambridge: Cambridge University Press.

Biosystems.okstate.edu, (2014). Darcy-Weisbach Equation History. [online] Available at:

http://biosystems.okstate.edu/darcy/DarcyWeisbach/Darcy-WeisbachHistory.htm [Accessed 24 Oct.
2014].

Engineeringtoolbox.com, (2014). Reynolds Number. [online] Available at:

http://www.engineeringtoolbox.com/reynolds-number-d_237.html [Accessed 26 Oct. 2014].

Adiutori, E. (2002). The new engineering. Naples, Fla.: Ventuno Press.

Appendix
Laminar Flow (Armfield Experimental Rig)
Pipe Length = 500mm, Pipe Diameter 3mm

No Flow Rate Water Water Manometer Density Dynamic

Volume Time Temperature h1 h2 ∆𝐡 (kg/m3) Viscosity
(mL) (s) (C) (mm) (mm) (mm) (kg/m s)
1 60 60.25 29 255 275 20 996.02 0.000815
2 138 60.25 30 238 285 47 995.71 0.000798
3 185 60.31 30 230 295 65 995.71 0.000798
4 225 60.13 30 210 305 95 995.71 0.000798
5 263 60.13 30.5 195 315 120 995.56 0.000790
6 285 60.25 31 185 325 140 995.41 0.000781
7 290 60.13 31 175 335 160 995.41 0.000781
8 305 60.31 31 162 345 183 995.41 0.000781
9 340 60.13 31 135 365 230 995.41 0.000781
10 380 60.25 31 105 385 280 995.41 0.000781

Table 1.Table of results for Laminar Flow.

Turbulent Flow (Lotus Scientific Rig)

Pipe Length = 320mm

No Pipe Flow Rate Water Mercury Manometer Density Dynamic

Diameter Volume Time Temperature h1 h2 ∆𝐡 (kg/m3) Viscosity
(mm) (L) (s) (C) (mm) (mm) (mm) (kg/m s)
1 7.7 10 43 28 245 360 115 996.31 0.000833
2 7.7 10 40.60 28 235 370 135 996.31 0.000833
3 7.7 10 39.13 28 225 380 155 996.31 0.000833
4 7.7 10 47.02 28 255 350 95 996.31 0.000833
5 7.7 10 51.59 28 265 340 75 996.31 0.000833
6 7.7 10 61.00 28 275 330 55 996.31 0.000833
7 7.7 10 85.00 28 285 320 35 996.31 0.000833
8 7.7 10 133.00 28 295 310 15 996.31 0.000833

Table 2.Table of results for Turbulent Flow.

1L=0.001m3
1mL=0.000001m3

Volume
Q= Time

𝑄
v=𝐴,
Where v=velocity, A=Area

𝜋𝐷 2
A= 4 ,
Where D=Diameter (Laminar Flow=0.003m, Turbulent Flow=0.0077m)
ALaminar Flow=7.06858 x 10-7 m2 & ATurbulent Flow=4.65663 x 10-5 m2
𝜌𝑣𝐷
Re= ,
𝜇

P1 v12 P2 v22
+ + z1 = + + z2 + hloss
ρg 2g ρg 2g

∆PLaminar Flow =(ρwater − ρair )g∆h

∆PTurbulent Flow =(ρmercury − ρwater )g∆h
where ρmercury = 13600kg/m3 and ρwater =1.23kg/m3

∆P L v2
=f
ρg D 2g
∆P
ρ
f= 2
Lv
2D
∆P(2D)
=ρ(L)(v2 )

Where L=Length (Laminar Flow=0.5m, Turbulent Flow=0.32m)

No. Q (x10-6 m3/s) v (m/s) Reynolds Number ∆𝐏 (Pa) f

1 0.995851 0.140884 516.527 195.177798 0.118473
2 2.290456 0.324033 1212.943 458.524894 0.052630
3 3.067485 0.433961 1624.432 634.130172 0.040581
4 3.741893 0.529370 1981.574 926.805636 0.039858
5 4.373857 0.618775 2339.344 1170.525276 0.036849
6 4.730290 0.668200 2554.927 1365.406812 0.036866
7 4.822884 0.682299 2608.836 1560.464928 0.040410
8 5.057204 0.715448 2735.587 1784.781761 0.042035
9 5.654415 0.799937 3058.636 2243.168334 0.042260
10 6.307054 0.892266 3416.664 2730.813624 0.041351
Table 3.Table of calculated results for Laminar Flow.

Q (x10-4
No. v (m/s) Reynolds Number ∆𝐏 (Pa) f
m3/s)
1 2.325581 5.0036185 45993.87168 14218.85287 0.027537
2 2.463054 5.289349 48712.72632 16691.69685 0.028819
 3 2.555584 5.488055 50542.72676 19164.54083 0.030735
4 2.126755 4.567154 42061.60776 11746.00890 0.027200
5 1.938360 4.162581 38335.65700 9673.16492 0.026966
6 1.639344 3.520451 32421.90411 6800.32094 0.026504
7 1.176471 2.526443 23267.49973 4237.47696 0.032067
8 0.751880 1.614644 14870.20639 1854.63298 0.034362
Table 4.Table of calculated results for Turbulent Flow.

πD4 ∆P
Q=
128μL

128μL
∆P = Q
πD4
128μL
gradient =
πD4

According to graph, gradient =241.11,

(241.11)(π)(0.0034 )
Therefore μ = (128)(0.5)

= 9.586722 x 10-4kg/m s

𝜇𝑇ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 − 𝜇𝑒𝑥𝑝𝑒𝑟𝑖𝑚𝑒𝑛𝑡
𝑥100% = 16.57%
𝜇𝑇ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙