11/11/2025

CE 231 – FLUID MECHANICS

Differential Analysis of Fluid Flow
Selva Balaji M
BITS Pilani Department of Civil Engineering
[Link]@[Link]
Pilani Campus

Learning Objectives

• apply the Buckingham pi theorem.

• develop a set of dimensionless variables for a given flow situation.
• discuss the use of dimensionless variables in data analysis.
• apply the concepts of modeling and similitude to develop prediction
equations.

2 BITS Pilani, Pilani Campus

1
 11/11/2025

Dimensional Analysis

• To understand the functional relationships between dependent variable and independent

variables  Helps in experimental studies with models and similitude
• Pressure drop in unit length Δ𝑝ℓ = 𝑓(𝐷, 𝜌, 𝜇, 𝑉)
• If experiments done for different values of variables, through data the relationship can be
derived. However, in model experiments some variables can’t be changed. For example,
viscosity of the fluid cannot be changed to many options
• Also number of combinations would be too high. Dimensional analysis helps to reduce the
variables to lesser number of non-dimensional groups
𝐷Δ𝑝ℓ 𝜌𝑉𝐷
=𝜙
𝜌𝑉 𝜇
• Number of experiments  based on Reynolds number

3 BITS Pilani, Pilani Campus

Dimensional Analysis

• To understand the functional relationships between dependent variable and independent

variables  Helps in experimental studies with models and similitude
• Pressure drop in unit length
Δ𝑝ℓ = 𝑓(𝐷, 𝜌, 𝜇, 𝑉)
• If experiments done for different values of variables, through data the relationship can be
derived. However, in model experiments some variables can’t be changed. For example,
viscosity of the fluid cannot be changed to many options
• Also number of combinations would be too high. Dimensional analysis helps to reduce the
variables to lesser number of non-dimensional groups
𝐷Δ𝑝ℓ 𝜌𝑉𝐷
=𝜙
𝜌𝑉 𝜇

• Number of experiments  based on Reynolds number

4 BITS Pilani, Pilani Campus

2
 11/11/2025

Dimensional Analysis

• Basic dimensions: M, L, T
𝐹 ≐ 𝑀𝐿𝑇
• Dimensionless products/non-dimensional numbers
𝐷Δ𝑝ℓ 𝐿 𝐹/𝐿
≐ ≐𝐹 𝐿 𝑇
𝜌𝑉 𝐹𝐿 𝑇 𝐿𝑇
𝜌𝑉𝐷 𝐹𝐿 𝑇 𝐿𝑇 (𝐿)
≐ ≐𝐹 𝐿 𝑇
𝜇 𝐹𝐿 𝑇

5 BITS Pilani, Pilani Campus

Buckingham Pi Theorem (Edgar

Buckingham in 1914)
• If an equation involving k variables is dimensionally homogeneous, it can be reduced to a
relationship among k-r independent dimensionless products, where r is the minimum number
of reference dimensions required to describe the variables.
• Dimensionless products/numbers are called as Pi terms ∏ terms
• any physically meaningful equation involving k variables
𝑢 = 𝑓 𝑢 ,𝑢 ,…,𝑢
• the dimensions of the variable on the left side of the equal sign must be equal to the
dimensions of any term that stands by itself on the right side of the equal sign
• The relationship can be written as k − r pi terms
Π = 𝜙 Π ,Π ,…,Π
• r is determined by the minimum number of reference dimensions required to describe the
original list of variables: M, L, T (r=3) or F, L, T (r=3). Sometimes r can be 1 or 2 as well

6 BITS Pilani, Pilani Campus

3
 11/11/2025

Determination of Pi Terms
Method of Repeating Variables
• Step 1 List all the variables that are involved in the problem ‒ Based on knowledge based on
experience
• Geometric variables: Diameter of pipe, length of pipe, angle of approach
• Fluid property variables: density, viscosity, specific weight ‒ If density and gravitational
acceleration are used no need to include specific weight since specific weight = density *
gravitational acceleration
• External effects variables
• Step 2 Express each of the variables in terms of basic dimensions: 𝛥𝑝ℓ ≐ 𝐹𝐿 , 𝜌 ≐
𝐹𝐿 𝑇 and so on
• Step 3 Determine the required number of pi terms: 𝑘 − 𝑟 pi terms
• Step 4 Select a number of repeating variables, where the number required is equal to the
number of reference dimensions (𝑟)
• Step 5 Form a pi term by multiplying one of the nonrepeating variables by the product of the
repeating variables, each raised to an exponent that will make the combination
dimensionless
7 BITS Pilani, Pilani Campus

Determination of Pi Terms
Method of Repeating Variables
• Step 5 Form a pi term by multiplying one of the nonrepeating variables by the product of the
repeating variables, each raised to an exponent that will make the combination
dimensionless
• Step 6 Repeat Step 5 for each of the remaining nonrepeating variables: The resulting set of
pi terms will correspond to the required number obtained from Step 3.
• Step 7 Check all the resulting pi terms to make sure they are dimensionless
• Step 8 Express the final form as a relationship among the pi terms, and think about what it
means
Π = 𝜙 Π ,Π ,…,Π
where would contain the Π dependent variable in the numerator.
• if started out with the correct list of variables, then the relationship in terms of the pi terms
can be used to describe the problem.
• However, the actual functional relationship among the pi terms must be determined by
experiment.

8 BITS Pilani, Pilani Campus

4
 11/11/2025

Pressure Drop Per Unit Length

• Step 1 List all the variables that are involved in the problem ‒ Based on knowledge based on
experience
Δ𝑝ℓ = 𝑓(𝐷, 𝜌, 𝜇, 𝑉)
𝑘 → 5 (number of variables)
𝐷 → 𝑑𝑖𝑎𝑚𝑒𝑡𝑒𝑟 𝑜𝑓 𝑡ℎ𝑒 𝑝𝑖𝑝𝑒, 𝜌 → 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑙𝑢𝑖𝑑, 𝜇 → 𝑑𝑦𝑛𝑎𝑚𝑖𝑐 𝑣𝑖𝑠𝑐𝑜𝑠𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑙𝑢𝑖𝑑, 𝑎𝑛𝑑
𝑉 → average velocity of flow
• Geometric variables: Diameter of pipe, length of pipe, angle of approach
• Fluid property variables: density, viscosity, specific weight ‒ If density and gravitational
acceleration are used no need to include specific weight since specific weight = density *
gravitational acceleration
• External effects variables
• Step 2 Express each of the variables in terms of basic dimensions
Δ𝑝ℓ ≐ 𝐹𝐿 ; D ≐ 𝐿; 𝜌 ≐ 𝐹𝐿 𝑇 ; 𝜇 ≐ 𝐹𝐿 𝑇; 𝑉 ≐ 𝐿𝑇

9 BITS Pilani, Pilani Campus

Pressure Drop Per Unit Length

• Step 3 Determine the required number of pi terms: 𝑘 − 𝑟 pi terms

𝑘 − 𝑟 = 5 − 3 = 2 pi terms
• Step 4 Select a number of repeating variables, where the number required is equal to the
number of reference dimensions (𝑟):
• Repeating variable cannot be the dependent variable Δ𝑝ℓ
• Among 𝐷, 𝜌, 𝜇, 𝑉 one geometric variable (𝐷), one fluid property variable (either 𝜌 or 𝜇),
one flow property variable (𝑉) can be chosen
• Repeating variables should not have same dimensions and also together cover all
reference dimensions (r dimensions: M, L, T)

• Step 5 Form a pi term by multiplying one of the nonrepeating variables by the product of the
repeating variables, each raised to an exponent that will make the combination
dimensionless: start with the dependent variable and combine it with the repeating variables
to form the first pi term

10 BITS Pilani, Pilani Campus

10

5
 11/11/2025

Pressure Drop Per Unit Length

• Step 5 Form a pi term by multiplying one of Π = 𝜇𝐷 𝑉 𝜌

the nonrepeating variables by the product of 𝐹𝐿 𝑇 (𝐿) 𝐿𝑇 𝐹𝐿 𝑇 ≐𝐹 𝐿 𝑇
the repeating variables, each raised to an 1 + 𝑐 = 0 ( for F)
exponent that will make the combination −2 + 𝑎 + 𝑏 − 4𝑐 = 0 (for L)
dimensionless: start with the dependent 1 − 𝑏 + 2𝑐 = 0 (for 𝑇)
variable and combine it with the repeating 𝑎 = −1; 𝑏 = −1; 𝑐 = −1
variables to form the first pi term Step 6:
Π = Δ𝑝ℓ 𝐷 𝑉 𝜌 Δ𝑝ℓ 𝐷
Π =
𝐹𝐿 𝑇 (𝐿) 𝐿𝑇 𝐹𝐿 𝑇 ≐𝐹 𝐿 𝑇 𝜌𝑉
1 + 𝑐 = 0 ( for F) 𝜇
Π =
−3 + 𝑎 + 𝑏 − 4𝑐 = 0 (for L) 𝐷𝑉𝜌
−𝑏 + 2𝑐 = 0 (for 𝑇) Step 7:
𝑎 = 1; 𝑏 = −2; 𝑐 = −1 Δ𝑝ℓ 𝐷 𝐹𝐿 (𝐿)
Δ𝑝ℓ 𝐷 Π = ≐ ≐𝐹 𝐿 𝑇
Π = 𝜌𝑉 𝐹𝐿 𝑇 𝐿𝑇
𝜌𝑉 𝜇 𝐹𝐿 𝑇
Π = ≐ ≐𝐹 𝐿 𝑇
𝐷𝑉𝜌 (𝐿) 𝐿𝑇 𝐹𝐿 𝑇
11 BITS Pilani, Pilani Campus

11

Pressure Drop Per Unit Length

Step 7: Step 8: Express the result

Δ𝑝ℓ 𝐷 𝐹𝐿 (𝐿) Π = 𝜙 Π ,Π ,…,Π
Π = ≐ ≐𝐹 𝐿 𝑇 Δ𝑝ℓ 𝐷 𝜇
𝜌𝑉 𝐹𝐿 𝑇 𝐿𝑇 = 𝜙˜
𝜇 𝐹𝐿 𝑇 𝜌𝑉 𝐷𝑉𝜌
Π = ≐ ≐𝐹 𝐿 𝑇
𝐷𝑉𝜌 (𝐿) 𝐿𝑇 𝐹𝐿 𝑇 Or if Π =
Alternatively 𝐷Δ𝑝ℓ 𝜌𝑉𝐷
Δ𝑝ℓ 𝐷 𝑀𝐿 𝑇 (𝐿) =𝜙
Π = ≐ ≐𝑀 𝐿 𝑇 𝜌𝑉 𝜇
𝜌𝑉 𝑀𝐿 𝐿𝑇
Both are same. For exact relationship,
𝜇 𝑀𝐿 𝑇
Π = ≐ ≐𝑀 𝐿 𝑇 experimental data required.
𝐷𝑉𝜌 (𝐿) 𝐿𝑇 𝑀𝐿

12 BITS Pilani, Pilani Campus

12

6
 11/11/2025

A thin rectangular plate having a width w and a height h is located so that it is normal to a moving
stream of fluid as shown in Fig.. Assume the drag, 𝒟, that the fluid exerts on the plate is a function of
w and h, the fluid viscosity and density and the velocity V of the fluid approaching the plate.
Determine a suitable set of pi terms to study this problem experimentally
ℎ
𝒟 = 𝑓(𝑤, ℎ, 𝜇, 𝜌, 𝑉) Π =
𝒟 ≐ 𝑀𝐿𝑇 ; 𝑤 ≐ 𝐿; ℎ ≐ 𝐿; 𝜇 ≐ 𝑀𝐿 𝑇 ; 𝑤
Π = 𝜇𝑤 𝑉 𝜌
𝜌 ≐ 𝑀𝐿 ; 𝑉 ≐ 𝐿𝑇
𝑀𝐿 𝑇 (𝐿) 𝐿𝑇 𝑀𝐿 ≐𝑀 𝐿 𝑇
• k = 6; r = 3 1 + 𝑐 = 0 for 𝑀
• Number of pi terms: 3 (6-3=3) −1 + 𝑎 + 𝑏 − 3𝑐 = 0 for 𝐿
Π = 𝒟𝑤 𝑉 𝜌 −1 − 𝑏 = 0 ( for 𝑇)
𝑀𝐿𝑇 (𝐿) 𝐿𝑇 𝑀𝐿 ≐𝑀 𝐿 𝑇 𝜇
Π =
𝑎 = −2; 𝑏 = −2; 𝑐 = −1 𝑤𝑉𝜌
𝒟
Π =
𝑤 𝑉 𝜌 𝒟
Π =
Π = ℎ𝑤 𝑉 𝜌 𝑤 𝑉 𝜌
(𝐿)(𝐿) 𝐿𝑇 𝑀𝐿 ≐𝑀 𝐿 𝑇 (𝐹)
𝑐 = 0 ( for 𝑀) ≐ ≐𝐹 𝐿 𝑇
(𝐿) 𝐿𝑇 𝐹𝐿 𝑇
1 + 𝑎 + 𝑏 − 3𝑐 = 0 ( for 𝐿) ℎ (𝐿)
𝑏 = 0 ( for 𝑇) Π = ≐ ≐𝐹 𝐿 𝑇
𝑤 (𝐿)

13 BITS Pilani, Pilani Campus

13

A thin rectangular plate having a width w and a height h is located so that it is normal to a moving
stream of fluid as shown in Fig.. Assume the drag, 𝒟, that the fluid exerts on the plate is a function of
w and h, the fluid viscosity and density and the velocity V of the fluid approaching the plate.
Determine a suitable set of pi terms to study this problem experimentally
𝜇 𝐹𝐿 𝑇
Π = ≐ ≐𝐹 𝐿 𝑇
𝑤𝑉𝜌 (𝐿) 𝐿𝑇 𝐹𝐿 𝑇

𝒟 ℎ 𝜇
= 𝜙˜ ,
𝑤 𝑉 𝜌 𝑤 𝑤𝑉𝜌
Or
𝒟 𝑤 𝜌𝑉𝑤
=𝜙 ,
𝑤 𝜌𝑉 ℎ 𝜇

14 BITS Pilani, Pilani Campus

14

7
 11/11/2025

An open, cylindrical paint can having a diameter D is filled to a depth h with paint having a specific
weight 𝛾. The vertical deflection, 𝛿, of the center of the bottom is a function of D, h, d, and E, where
d is the thickness of the bottom and E is the modulus of elasticity of the bottom material.
Determine the functional relationship between the vertical deflection, and the independent
variables using dimensional analysis.
𝛿
𝛿 = 𝑓(𝐷, ℎ, 𝑑, 𝛾, 𝐸) Π =
𝐷
𝛿 ≐ 𝐿; 𝐷 ≐ 𝐿; ℎ ≐ 𝐿; 𝑑 ≐ 𝐿; 𝛾 Π = ℎ𝐷 𝛾
≐ 𝐹𝐿 ≐ 𝑀𝐿 𝑇 ;
E ≐ 𝐹𝐿 ≐ 𝑀𝐿 𝑇 Π = ;Π = ;Π = ;Π = ;
k = 5; r = 2 (F & L) 𝛿 ℎ 𝑑 𝐸
=𝜙 , ,
𝐷 𝐷 𝐷 𝐷𝛾
Number of pi terms: 4 (6-
2=4)
Π = 𝛿𝐷 𝛾
(𝐿)(𝐿) 𝐹𝐿 ≐𝐹 𝐿
1 + 𝑎 − 3𝑏 = 0 (for L)
𝑏 = 0 (for F)
𝑎 = −1; 𝑏 = 0;
𝛿
Π =
𝐷

15 BITS Pilani, Pilani Campus

15

Drag, d, acting on a spherical particle that falls very slowly through a viscous fluid, is a function of the
particle diameter, D, the particle velocity, V, and the fluid viscosity. Determine, with the aid of
dimensional analysis, how the drag depends on the particle velocity

𝒟 = 𝑓(𝐷, 𝑉, 𝜇)
𝒟 ≐ 𝐹; 𝐷 ≐ 𝐿; 𝑉 ≐ 𝐿𝑇 ; 𝜇 ≐ 𝐹𝐿 𝑇 𝑊𝑒𝑖𝑔ℎ𝑡 𝑎𝑛𝑑 𝐵𝑢𝑜𝑦𝑎𝑛𝑐𝑦 𝑛𝑒𝑒𝑑𝑒𝑑 𝑡𝑜 𝑏𝑒
𝒟 𝑎𝑑𝑑𝑒𝑑
Π =
𝜇𝑉𝐷 𝑊𝛼𝐷 ; 𝐹𝑩 𝛼𝐷 ;
𝒟 = 𝑊 − 𝐹𝑩
Since only one pi term. It can
Earlier 𝒟 = 3𝜋𝜇𝑉𝐷
be written with a constant.
𝒟 Hence, 𝑉𝐷 ∝ 𝐷
=𝐶
𝜇𝑉𝐷
𝒟 = 𝐶𝜇𝑉𝐷 Hence, 𝑉 ∝ 𝐷
𝒟 𝛼 𝑉
Stokes Law:
𝒟 = 3𝜋𝜇𝑉𝐷
Reynolds numbers are small
(Re << 1).
– neglected inertial effects
(fluid density is not
included as a variable)
16 BITS Pilani, Pilani Campus

16

8
 11/11/2025

Drag, d, acting on a spherical particle that falls very slowly through a viscous fluid, is a function of the
particle diameter, D, the particle velocity, V, and the fluid viscosity. Determine, with the aid of
dimensional analysis, how the drag depends on the particle velocity We can use the principles of Stokes flow to analyze
sedimentation (fall velocity) of small particles as
𝛿 long as the Reynolds numbers are small (Re << 1).
𝛿 = 𝑓(𝐷, ℎ, 𝑑, 𝛾, 𝐸) Π = Stokes flow covers dust and dirt settling out of the
𝐷 air or sand particles settling back to the river
bottom.
A sphere slowly falling in a fluid has three forces
acting on it: drag, buoyant force, and weight.
Assuming the particle to be approximately
spherical, the drag force is proportional to the
product of the diameter and fall velocity. The
weight and buoyant force are proportional to the
diameter raised to the third power. Therefore in
equilibrium conditions, the fall velocity is
proportional to the diameter squared. The video
shows two spheres, one having twice the diameter
of the other, falling through syrup. Based on the
Munson, Okiishi, Huebsch, Rothmayer:Fluid Mechanics, 7th fact that velocity is proportional to the diameter
Edition SI VersionVideo Library squared, the larger sphere falls four times faster
than the smaller sphere
17 BITS Pilani, Pilani Campus

17

Common Dimensionless Groups

Read Section 7.6 in Munson

Fluid Mechanics book

18 BITS Pilani, Pilani Campus

18

9
 11/11/2025

Modeling and Similitude

• A model is a representation of a physical system that may be used to predict the behavior of the
system in some desired respect.
• The physical system for which the predictions are to be made is called the prototype.
• Physical models made of different sizes compared to prototype.
• If prototype is larger size, models can be smaller size to reduce the cost.
• Sometimes models can be larges than prototype: models to study the motion red blood cells
• to predict the behavior of the prototype under a certain set of conditions.
• Possible design changes for a hydraulic structure
• Aircraft design for drag forces

19 BITS Pilani, Pilani Campus

19

Theory of Models

• Using dimensional analysis

Π = 𝜙 Π ,Π ,…,Π
Π = 𝜙 Π ,Π ,…,Π
• If
Π =Π
Π =Π
Π =Π

• Then dependent pi term Π =Π

• Prediction equation should remain same for prototype as well as models: model design
conditions or similarity requirements or modeling laws.

20 BITS Pilani, Pilani Campus

20

10
 11/11/2025

Theory of Models: Drag on a Thin

Rectangular Plate
𝒟 = 𝑓(𝑤, ℎ, 𝜇, 𝜌, 𝑉)
𝒟 𝑤 𝜌𝑉𝑤
=𝜙 ,
𝑤 𝜌𝑉 ℎ 𝜇
• For similar relationship between pi terms for model
𝒟 𝑤 𝜌 𝑉 𝑤
=𝜙 ,
𝑤 𝜌 𝑉 ℎ 𝜇
𝑤 𝑤 𝜌 𝑉 𝑤 𝜌𝑉𝑤
= ; =
ℎ ℎ 𝜇 𝜇
• First similarity requirement
ℎ
𝑤 = 𝑤 → (size of the model: dimensional similarity)
ℎ
• Second requirement
𝜇 𝜌 𝑤
𝑉 = 𝑉 → (velocity of the model: kinematic similarity)
𝜇 𝜌 𝑤

21 BITS Pilani, Pilani Campus

21

Theory of Models: Drag on a Thin

Rectangular Plate
• Second requirement
𝒟 = 𝑓(𝑤, ℎ, 𝜇, 𝜌, 𝑉) 𝜇 𝜌 𝑤
𝒟 𝑤 𝜌𝑉𝑤 𝑉 = 𝑉
=𝜙 , 𝜇 𝜌 𝑤
𝑤 𝜌𝑉 ℎ 𝜇 → (velocity of the model: kinematic
• For similar relationship between pi similarity)
terms for model 𝒟 𝒟
=
𝒟 𝑤 𝜌 𝑉 𝑤 𝑤 𝜌𝑉 𝑤 𝜌 𝑉
=𝜙 , 𝑤 𝜌 𝑉
𝑤 𝜌 𝑉 ℎ 𝜇
𝒟= 𝒟
𝑤 𝑤 𝜌 𝑉 𝑤 𝜌𝑉𝑤 𝑤 𝜌 𝑉
= ; =
ℎ ℎ 𝜇 𝜇
• First similarity requirement • to achieve similarity between model and
ℎ
𝑤 = 𝑤 prototype behavior, all the corresponding pi terms
ℎ must be equated between model and prototype
→ (size of the model: dimensional
similarity)

22 BITS Pilani, Pilani Campus

22

11
 11/11/2025

Theory of Models: Similarity

• to achieve similarity between model and prototype behavior, all the corresponding pi terms
must be equated between model and prototype
• Geometric similarity: roughness, diameter of pipe etc.,
• Kinematic similarity: velocity ratios, accelerations ratios between model and prototype
• Dynamic similarity: force ratios

23 BITS Pilani, Pilani Campus

23

Flow Past an Ellipse

• A long structural component of a bridge has an
elliptical cross section shown in Fig. It is known
that when a steady wind blows past this type of
bluff body, vortices may develop on the
downwind side that are shed in a regular
fashion at some definite frequency. Since these
vortices can create harmful periodic forces
acting on the structure, it is important to
determine the shedding frequency. For the
specific structure of interest, D = 0.1 m, H = 0.3
m, and a representative wind velocity is 50
km/hr. Standard air can be assumed. The
shedding frequency is to be determined through
the use of a small-scale model that is to be
tested in a water tunnel. For the model and the
water temperature is 20 degrees Celsius.
Determine the model dimension, and the
velocity at which the test should be performed.
If the shedding frequency for the model is found
to be 49.9 Hz, what is the corresponding
frequency for the prototype?

24 BITS Pilani, Pilani Campus

24

12
 11/11/2025

A long structural component of a bridge has an elliptical cross section shown in Fig. It is known that
when a steady wind blows past this type of bluff body, vortices may develop on the downwind side that
are shed in a regular fashion at some definite frequency. Since these vortices can create harmful
periodic forces acting on the structure, it is important to determine the shedding frequency. For the
specific structure of interest, D = 0.1 m, H = 0.3 m, and a representative wind velocity is 50 km/hr.
Standard air can be assumed. The shedding frequency is to be determined through the use of a small-
scale model that is to be tested in a water tunnel. For the model and the water temperature is 20
degrees Celsius. Determine the model dimension, and the velocity at which the test should be
performed. If the shedding frequency for the model is found to be 49.9 Hz, what is the corresponding
frequency for the prototype?
𝜇 𝜌 𝐷
Shedding frequency, 𝜔 = 𝑓 𝐷, 𝐻, 𝑉, 𝜌, 𝜇 𝑉 = 𝑉
𝜇 𝜌 𝐷
𝜔 ≐ 𝑇 ; 𝐷 ≐ 𝐿; 𝐻 ≐ 𝐿; 𝑉 ≐ 𝐿𝑇 ; 𝜌 ≐ 𝑀𝐿 ;
50 × 10 m/hr
𝜇 ≐ 𝑀𝐿 𝑇 𝑉= = 13.9 m/s
𝜔𝐷 𝐷 𝜌𝑉𝐷 (3600 s/hr)
=𝜙 , 𝑉
𝑉 𝐻 𝜇
1.00 × 10 𝑘𝑔/ 𝑚 ⋅ 𝑠 1.23 𝑘𝑔/m
𝐷 𝐷 =
= 1.79 × 10 𝑘𝑔/ 𝑚 ⋅ 𝑠 998 𝑘𝑔/m
𝐻 𝐻
𝜌 𝑉 𝐷 𝜌𝑉𝐷 0.1 𝑚
= × 13.9 𝑚/𝑠
𝜇 𝜇 20 × 10 m
𝑉 = 4.79 m/s
𝐷 20 × 10 m
𝐻 = 𝐻= 0.3 𝑚
𝐷 0.1 𝑚
𝐻 = 60 × 10 m = 60 mm

25 BITS Pilani, Pilani Campus

25

A long structural component of a bridge has an elliptical cross section shown in Fig. It is known that
when a steady wind blows past this type of bluff body, vortices may develop on the downwind side that
are shed in a regular fashion at some definite frequency. Since these vortices can create harmful
periodic forces acting on the structure, it is important to determine the shedding frequency. For the
specific structure of interest, D = 0.1 m, H = 0.3 m, and a representative wind velocity is 50 km/hr.
Standard air can be assumed. The shedding frequency is to be determined through the use of a small-
scale model that is to be tested in a water tunnel. For the model and the water temperature is 20
degrees Celsius. Determine the model dimension, and the velocity at which the test should be
performed. If the shedding frequency for the model is found to be 49.9 Hz, what is the corresponding
frequency for the prototype?
𝒟ℓ
𝜔𝐷 𝜔 𝐷 If Drag force needed, similarity should be satisfied
=
𝑉 𝑉
𝑉 𝐷
𝜔= 𝜔 𝒟ℓ = 𝒟ℓ
𝑉 𝐷
(13.9 m/s) 20 × 10 m
= (49.9 Hz)
(4.79 m/s) (0.1 m)
𝜔 = 29.0 Hz

26 BITS Pilani, Pilani Campus

26

13
 11/11/2025

Model Scales

• Two length variables

ℓ ℓ
=
ℓ ℓ
ℓ ℓ
= = λℓ
ℓ ℓ
Length scales: λℓ (can be represented as 1:10)
Similarly velocity scale, density scale, viscosity scales are possible

27 BITS Pilani, Pilani Campus

27

Model Scales: Distorted Models

• If more than one pi term is there, the model scale may not satisfy all pi terms. They are called as
distorted models
• In Open channel flow or free surface flows both Reynolds number and Froude number (Fr) are
involved
• For Froude number similarity, =
ℓ ℓ
ℓ
• If gravitational field is same for both model and prototype 𝑔 = 𝑔, = = 𝜆ℓ
ℓ
ℓ ℓ ℓ
• For Re similarity, = → = ℓ
ℓ ℓ ℓ
• From Fr similarity, = 𝜆ℓ and substituting it in Re similarity, =
ℓ ℓ ℓ

ℓ 𝜇 𝜌 ℓ 𝜇 /𝜌 𝜈 /
= → = = 𝜆ℓ
ℓ 𝜇 𝜌 ℓ 𝜇/𝜌 𝜈
/
• Model satisfying condition with fluid properties = 𝜆ℓ is not always possible. So only one of Re
and Fr is satisfied with distorted model.

28 BITS Pilani, Pilani Campus

28

14
 11/11/2025

Flow Through Closed Conduits

• Pipe flow and flow through valves, fittings, and metering devices with conduit can be circular or other
shapes
• there are no fluid interfaces or free surfaces, the dominant forces are inertial and viscous so that the
Reynolds number is an important similarity parameter.
• If Mach number Ma <0.3, compressibility effects can be ignored.
• Geometric similarities between model and prototype should be maintained. Different dimensions can
be scaled with same length scale.
ℓ
Π =
ℓ
• ℓ , ℓ , ⋯ , ℓ and so on are different length terms: diameter, length of pipe, pipe roughness etc,
• Pipe roughness 𝜀, can be scaled to
ℓ
• If model is scaled down the surface roughness might end up smooth and shape of roughness may
not be maintained. But roughness is not important in all problems & it can be ignored sometimes.
ℓ ℓ
• If Ma< 0.3, Dependent pi term = 𝜙 , ,
ℓ ℓ

29 BITS Pilani, Pilani Campus

29

Flow Through Closed Conduits

𝑉 ℓ ℓ 𝑉
• If Ma< 0.3, Dependent pi term = = →𝑉 =𝑉 =
ℓ ℓ 𝑉 ℓ ℓ 𝜆ℓ
𝜙 , ,
ℓ ℓ ℓ
ℓ ℓ 𝜀 𝜀 If model is smaller than prototype, = 𝜆ℓ < 1, 𝑉 is
ℓ
= = higher than velocity of the prototype𝑉 .
ℓ ℓ ℓ ℓ
ℓ 𝜀 ℓ • If model satisfied geometric similarity and Re
= = = 𝜆ℓ
ℓ 𝜀 ℓ similarity, and If dependent pi term is Π = ,
• Once length scale 𝜆ℓ is chosen, it should
ℓ 𝜌 𝑉
satisfy similarity for 𝑅𝑒 = Δ𝑝 = Δ𝑝
𝜌 𝑉
𝜌 𝑉 ℓ 𝜌𝑉ℓ
= • If same fluid for model and prototype = 1,
𝜇 𝜇
𝑉 𝜇 𝜌 ℓ Δ𝑝 = 𝜆ℓ Δ𝑝
=
𝑉 𝜇 𝜌 ℓ If model is smaller than prototype, pressure drop in
• Different fluids can be used. If same fluid model is,
used, = 1& =1 Δ𝑝
Δ𝑝 =
𝜆ℓ
30 BITS Pilani, Pilani Campus

30

15
 11/11/2025

Flow Through Closed Conduits

• If model satisfied geometric similarity and Re similarity, and If dependent pi term is Π = ,
𝜌 𝑉
Δ𝑝 = Δ𝑝
𝜌 𝑉
• If same fluid for model and prototype = 1,
Δ𝑝 = 𝜆ℓ Δ𝑝
If model is smaller than prototype, pressure drop in model is higher than prototype by 𝜆ℓ times.
Δ𝑝
Δ𝑝 =
𝜆ℓ

31 BITS Pilani, Pilani Campus

31

Model tests are to be performed to study the flow through a large check valve
having a 2-ft-diameter inlet and carrying water at a flowrate of 30 cfs as shown
in Fig. The working fluid in the model is water at the same temperature as that
in the prototype. Complete geometric similarity exists between model and
prototype, and the model inlet diameter is 3 in. Determine the required
flowrate in the model. 𝜇 𝜌 𝐷
𝑉 = 𝑉
𝜇 𝜌 𝐷
To ensure dynamic similarity, the model tests should 50 × 10 m/hr
be run so that, Re = Re 𝑉=
(3600 s/hr)
= 13.9 m/s
𝑉 𝐷 𝑉𝐷
= 𝑉
𝜈 𝜈 1.00 × 10 𝑘𝑔/ 𝑚 ⋅ 𝑠 1.23 𝑘𝑔/m
Same fluid, 𝜈 = 𝜈 , =
1.79 × 10 𝑘𝑔/ 𝑚 ⋅ 𝑠 998 𝑘𝑔/m
𝑉 𝐷 0.1 𝑚
= × 13.9 𝑚/𝑠
𝑉 𝐷 20 × 10 m
Discharge, 𝑉 = 4.79 m/s
𝑄 𝑉 𝐴 𝐷 (𝜋/4)𝐷 𝐷
= = =
𝑄 𝑉𝐴 𝐷 (𝜋/4)𝐷 𝐷
(3/12ft)
𝑄 = 30ft /s = 3.75 cfs
(2ft)

32 BITS Pilani, Pilani Campus

32

16
 11/11/2025

Model tests are to be performed to study the flow through a large check valve
having a 2-ft-diameter inlet and carrying water at a flowrate of 30 cfs as shown
in Fig. The working fluid in the model is water at the same temperature as that
in the prototype. Complete geometric similarity exists between model and
prototype, and the model inlet diameter is 3 in. Determine the required
flowrate in the model.
Discussion: =
To ensure dynamic similarity, the model tests should In this case, = → 𝑉 = 8𝑉 = 8 = 76.4 ft/s
be run so that, Re = Re
𝑉 𝐷 𝑉𝐷 • We should achieve the velocity 76.4 ft/s. If the model is
= further smaller diameter, model velocity will be even
𝜈 𝜈
higher which cannot be achieved always without
Same fluid, 𝜈 = 𝜈 , cavitation.
𝑉 𝐷
=
𝑉 𝐷
Discharge,
𝑄 𝑉 𝐴 𝐷 (𝜋/4)𝐷 𝐷
= = =
𝑄 𝑉𝐴 𝐷 (𝜋/4)𝐷 𝐷
(3/12ft)
𝑄 = 30ft /s = 3.75 cfs
(2ft)

33 BITS Pilani, Pilani Campus

33

Flow Around Immersed Bodies

• Flow around aircraft, automobiles, golf balls, If geometric and Re similarity are met
𝒟 𝒟
and buildings =
1 1
• Reynolds number similarity, geometric 2
𝜌𝑉 ℓ
2
𝜌 𝑉 ℓ
similarity required. 𝜌 𝑉 ℓ
• If Ma < 0.3, compressibility effects can be 𝒟= 𝒟
𝜌 𝑉 ℓ
ignored.
ℓ 𝜀 𝜌𝑉ℓ If same fluid used for model and prototype
𝐷𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡 𝑝𝑖 𝑡𝑒𝑟𝑚 = 𝜙 , , ℓ
ℓ ℓ 𝜇 𝑉 =𝑉
ℓ
𝒟 ℓ ℓ
𝐶 = =𝜙 , ,
ℓ ℓ
𝑉 will be higher than prototype velocity. For a
ℓ
1:10 scale model, if prototype velocity is 50 mph,
ℓ ℓ 𝜀 𝜀
= ; = Model velocity should be 500 mph which is not
ℓ ℓ ℓ ℓ
𝜌 𝑉 ℓ 𝜌𝑉ℓ attainable for liquids.
= For gas with 500 mph, compressibility will be
𝜇 𝜇
important.

34 BITS Pilani, Pilani Campus

34

17
 11/11/2025

Flow Around Immersed Bodies

If geometric and Re similarity are met Options: use fluid < 1using different fluids
𝒟 𝒟
= Or increase 𝜌 > 𝜌 by increasing the pressure
1 1
𝜌𝑉 ℓ 𝜌 𝑉 ℓ when air is used as a fluid.
2 2
𝜌 𝑉 ℓ
𝒟= 𝒟
𝜌 𝑉 ℓ
If same fluid used for model and prototype
ℓ
𝑉 =𝑉
ℓ
𝑉 will be higher than prototype velocity. For a
1:10 scale model, if prototype velocity is 50 mph,
Model velocity should be 500 mph which is not
attainable for liquids.
For gas with 500 mph, compressibility will be
important.
35 BITS Pilani, Pilani Campus

35

The drag on the airplane shown in Fig. cruising at 240 mph in standard air is to
be determined from tests on a 1:10 scale model placed in a pressurized wind
tunnel. To minimize compressibility effects, the air speed in the wind tunnel is
also to be 240 mph. Determine (a) the required air pressure in the tunnel
(assuming the same air temperature for model and prototype) and (b) the
drag on the prototype corresponding to a measured force of 1 lb on the
model.
To ensure dynamic similarity, the model tests should
be run so that, Re = Re
𝜌 𝑉 ℓ 𝜌𝑉ℓ 𝜌 𝜇 𝑉 ℓ 𝜇
= → = == (1)(10)
𝜇 𝜇 𝜌 𝜇 𝑉 ℓ 𝜇
𝜌 𝜇
= 10
𝜌 𝜇
It is not possible for same fluid. One option is to 𝑝 = 10(14.7psia) = 147psia
change the density by increasing pressure. Increase
in pressure does not change viscosity much. Hence,
𝜌
= 10
𝜌
p = 𝜌RT
𝑝 𝜌
=
𝑝 𝜌
At constant temperature, = 10
36 BITS Pilani, Pilani Campus

36

18
 11/11/2025

The drag on the airplane shown in Fig. cruising at 240 mph in standard air is to
be determined from tests on a 1:10 scale model placed in a pressurized wind
tunnel. To minimize compressibility effects, the air speed in the wind tunnel is
also to be 240 mph. Determine (a) the required air pressure in the tunnel
(assuming the same air temperature for model and prototype) and (b) the
drag on the prototype corresponding to a measured force of 1 lb on the
model.

For Drag,
𝒟 ℐ
=
1 1
𝜌𝑉 ℓ 𝜌 𝑉 ℓ
2 2
𝜌 𝑉 ℓ
𝒟= 𝒟
𝜌 𝑉 ℓ If Mach number becomes important?
1 𝑉 𝑉
= (1) (10) 𝒟 =
10 𝑐 𝑐
ℓ
𝒟 = 10𝒟 With Re similarity condition, 𝑉 = 𝑉
ℓ
𝑐 𝜈 ℓ
=
𝑐 𝜈 ℓ
Same fluid cannot be used unless length scale is unity
which not possible in high speed aerodynamics.

37 BITS Pilani, Pilani Campus

37

The drag on the airplane shown in Fig. cruising at 240 mph in standard air is to
be determined from tests on a 1:10 scale model placed in a pressurized wind
tunnel. To minimize compressibility effects, the air speed in the wind tunnel is
also to be 240 mph. Determine (a) the required air pressure in the tunnel
(assuming the same air temperature for model and prototype) and (b) the
drag on the prototype corresponding to a measured force of 1 lb on the
model.
If Mach number becomes important?
𝑉 𝑉
=
𝑐 𝑐
ℓ
With Re similarity condition, 𝑉 = 𝑉
ℓ
𝑐 𝜈 ℓ
=
𝑐 𝜈 ℓ
Same fluid cannot be used unless length scale is unity
which not possible in high speed aerodynamics.

Hence, Mach number similarity maintained while Re has

distorted scale.

38 BITS Pilani, Pilani Campus

38

19
 11/11/2025

Read

39 BITS Pilani, Pilani Campus

39

20