CHAPTER 10.

SIMILITUDE AND PHYSICAL MODELING

10.1 Introduction

Dimensional analysis is a technique to establish a relationship between variables by using

dimensions. Application of fluid mechanics in design makes use of experimental results. The
results are often difficult to interpret. Dimensional analysis provides a strategy for choosing
relevant data and helps to analyze fluid flow especially when fluid flow is too complex for
mathematical analysis. Dimensional analysis gives a single equation, which relates all the physical
factors of a problem to each other. The result of this analysis depends on the correct identification
of variables and their relationship. It doesn’t give the complete answer. So experiments are
necessary to complete the solution.

Dimension

Dimension is a property of physical quantity expressed in terms of mass, length and time.
Fundamental dimension: Mass (M), Length (L), time (T)
Derived/Secondary quantity: Quantity containing more than one fundamental dimension, e.g.
velocity = distance/time = LT-1

Dimension of most widely used physical quantities

a. fundamental: Mass = M, Length = L, Time = T
b. Geometric: Area = L2, Volume = L3
c. Kinematic quantities
Velocity = LT-1
Angular velocity = T-1
Acceleration = LT-2
Angular acceleration = T-2
Discharge = L3T-1
Kinematic viscosity = L2T-1
d. Dynamic quantities
Force, Weight = MLT-2
Density = ML-3
Specific weight = ML-2T-2
Dynamic viscosity = ML-1T-1
Pressure = ML-1T-2
Surface tension = MT-2
Work, energy = ML2T-2
Shear stress = ML-1T-2
Power = ML2T-3
Torque = ML2T-2
Momentum = MLT-1

Dimensional homogeneity

Dimensional homogeneity means dimensions of each terms in an equation on both sides are equal.
Dimensionally homogeneous equations are independent of the system of units.
e.g. Discharge through rectangular orifice is
2
𝑄 = √2𝑔𝐿𝐻 3/2
3
Dimension of Left hand side = 𝐿3 𝑇 −1

Dimension of right hand side = (𝐿𝑇 −2 )1/2 𝐿𝐿3/2 = 𝐿3 𝑇 −1

Dimensional homogeneity can be useful for the following purposes.

 Checking units of equations
 Converting between two sets of units
 Defining dimensionless relationships

10.2 Methods Of Dimensional Method

10.2.1 Rayleigh’s Method

In this method, a functional relationship of variables is expressed in terms of one exponential

equation which must be dimensionally homogeneous.

Steps
  Write down the functional relationship between dependent and independent variables.
 Change the functional relationship to the equation of exponential form.
 Write down the dimension of variables.
 Equate the powers of dimensions and find the constants.

X = dependent variable
X1, X2, X3,………,Xn = independent variables
Functional relationship: X =f(X1, X2, X3,………,Xn)
Equation in exponential form: 𝑋 = 𝐾𝑋1𝑎 𝑋2𝑏 𝑋3𝑐 … … … 𝑋𝑛𝑍
where a, b, c…….Z are constants
After equating the power of dimensions, three equations are formed. If there are more than three
constants, then three appropriate constants are expressed in terms of other constants. The variables
of the expression to be derived are examined, and the powers of these three variables are usually
expressed in terms of other constants.

10.2.2 Buckingham’s π theorem

It states that if there are n variables (dependent and independent) in a physical phenomenon and if
these variables contain m fundamental dimensions, then the variables are arranged into (n-m)
dimensionless terms. Each term is called π term.

Rayleigh method is not practicable for large number of variables due to the difficulty in finding
the constants. This is overcome by Buckingham’s π theorem, which requires lesser number of
dimensional groups of variables.

Steps
 Write down the functional relationship between dependent and independent variables.
X1 = dependent variable
X2, X3, X4,………,Xn = independent variables
X1 =f(X2, X3, X4,………,Xn)
f1(X1, X2, X3, X4,………,Xn) = 0 (I)

 Count total number of variables (n) and fundamental dimensions (m)

No. of π terms = n-m
 Write down the functional relationship in π terms.
φ (π1, π2, π3,…………, πn-m) = 0 (II)
 For each π term, write down equation in exponential form with (m+1) variables, where
m=3 is called fundamental dimension and it is also called repeating variables.
 Each π group is a function of n repeating variables plus one of the remaining variables. The
π term is dimensionless.
Let X2, X3, X4 be the repeating variables.
𝜋1 = 𝑋2𝑎1 𝑋3𝑏1 𝑋4𝑐1 𝑋1
𝜋2 = 𝑋2𝑎2 𝑋3𝑏2 𝑋4𝑐2 𝑋5
……………………………..
𝑎(𝑛−𝑚) 𝑏(𝑛−𝑚) 𝑐(𝑛−𝑚)
𝜋𝑛−𝑚 = 𝑋2 𝑋3 𝑋4 𝑋𝑛
 Solve each equation by the principle of dimensional homogeneity (writing dimensions and
equating the powers of M, L and T to get constants. )
 Obtain π1, π2, π3,…………, πn-m and substitute in eqn. (II). Establish functional relationship
between π terms.

Criteria For Selecting Repeating Variables

Repeating variables appear in most of the π groups. They have a large influence on the problem.
There is great freedom in choosing these.
Some rules which should be followed are
 There are n ( = 3) repeating variables.
 In combination they must contain all of dimensions (M, L, T)
 The repeating variables must not form a dimensionless group.
 Dependent variable should not be selected as repeating variable.
 No two repeating variables should have the same dimensions.
 They do not have to appear in all π groups.
 They should be measurable in an experiment.
 They should be of major interest to the designer.

In repeating variables, it is usually possible to take one variable representing fluid property (e.g.
density or viscosity), next variable representing flow property (e.g. velocity) and the last variable
representing geometry (e.g. diameter or length).

Manipulation of the π groups

Once identified the π groups can be changed to derive the required equation. The number of groups
does not change, but their appearance may change drastically.
Taking the defining equation as:
φ (π1, π2, π3,…………, πn-m) = 0
The following changes are permitted:
I. Combination of existing groups by multiplication or division to form a new group to replaces
one of the existing.
E.g. π1 and π2 may be combined to form π1a = π1 / π2. So the defining equation becomes
φ(π1a, π2, π3,…………, πn-m) = 0
II. Reciprocal of any group is valid.
φ (1/π1, π2, π3,…………, 1/πn-m) = 0
III. A group may be raised to any power.
φ (π12, (π2)1/2, π3,…………, πn-m) = 0
iv. Any groups are multiplied by a constant.
φ (kπ1, π2, π3,…………, πn-m) = 0
v. A group is expressed as a function of other groups
π2 = f (π1, π3,…………, πn-m) = 0

10.3 Model And Prototype

The small scale replica of actual structure or the machine is known as its model. The actual
structure or the machine is known as prototype.

Before the construction of hydraulic structure such as dam or hydraulic machine such as pump,
turbines, it is necessary to know how the structure or machine would behave when it is actually
constructed. For this purpose, experimental investigation is needed, which cannot be carried out
on the full size of structure or machine. Hence it is necessary to construct a model (replica) on
which the tests are performed to obtain the desired information.
10.3.1 Similitude

The similarity between the prototype and model in every aspect is known as similitude. There are
in general three types of similarities to be established for complete similarity to exist between the
model and its prototype: geometric, kinematic and dynamic

a. Geometric similarity

Geometric similarity is the similarity of shape between the model and the prototype. Geometric
similarity exists between the model and the prototype if the ratios of corresponding length
dimensions in the model and the prototype are equal. Such a ratio is called scale ratio. e.g.
𝐿𝑚 𝑏𝑚 𝑑𝑚
= = = 𝐿𝑟
𝐿𝑝 𝑏𝑝 𝑑𝑝
subscript m: model, subscript p: prototype, l= Length, b= Breadth, d = height, Lr = Scale ratio
For area and volume ratio, the relationships are
𝐴𝑚 𝐿𝑚 𝑏𝑚
= = 𝐿2𝑟
𝐴𝑝 𝐿𝑝 𝑏𝑝
 𝑉𝑜𝑙𝑚
= 𝐿3𝑟
𝑉𝑜𝑙𝑝

b. Kinematic similarity

Kinematic similarity is the similarity of motion. Geometric similarity implies that in addition to
geometric similarity, the ratios of velocities as well as acceleration of fluid particle at a certain
point in the model and at the corresponding point of prototype are equal. e.g.
𝑉𝑚1 𝑉𝑚2
=
𝑉𝑝1 𝑉𝑝2
𝑎𝑚1 𝑎𝑚2
=
𝑎𝑝1 𝑎𝑝2
where V = velocity, a = acceleration, subscript m = model, subscript p = prototype

c. Dynamic similarity

Dynamic similarity is the similarity of forces. Dynamic similarity implies that in addition to
geometric and kinematic similarity, the ratios of all the forces acting at a certain point in the model
and at the corresponding point of prototype are equal. e.g.
(𝐹𝑖𝑛𝑒𝑟𝑡𝑖𝑎 )𝑚 (𝐹𝑣𝑖𝑠𝑐𝑜𝑢𝑠 )𝑚 (𝐹𝑔𝑟𝑎𝑣𝑖𝑡𝑦 )𝑚
(𝐹𝑖𝑛𝑒𝑟𝑡𝑖𝑎 )𝑝
= (𝐹𝑣𝑖𝑠𝑐𝑜𝑢𝑠 )𝑝
= (𝑔𝑟𝑎𝑣𝑖𝑡𝑦)𝑝

Forces acting on fluid particle

 Inertia force: resistance by inert mass to acceleration
 Friction or viscous force
 Gravity force
 Pressure force
 Elastic force
 Surface tension force

10.4 Dimensionless Numbers Based On Force Ratios

Inetia force
1. Reynolds number (Re) = Viscous for𝑐𝑒
Inetia force 𝑚𝑎𝑠𝑠 𝑥 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛
=
Viscous for𝑐𝑒 𝑠ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 𝑥 𝑎𝑟𝑒𝑎
 𝑉 𝑉𝑜𝑙
𝜌𝑉𝑜𝑙 𝑡 𝜌 𝑡 𝑉 𝜌𝑄𝑉 𝜌𝐴𝑉 2 𝜌𝑉𝐿
𝑅𝑒 = = = = =
𝑑𝑉 𝑉 𝑉 𝑉 𝜇
𝜇 𝐴 𝜇𝐿𝐴 𝜇𝐿𝐴 𝜇𝐿𝐴
𝑑𝑦
𝑉𝐿
or 𝑅𝑒 = 𝜈
In case of pipe flow, linear dimension diameter (D) is taken as linear dimension L.
𝑉𝐷
𝑅𝑒 =
𝜈
Inetia force
2. Froude number (Fr) = Gravity for𝑐𝑒
Inetia force 𝜌𝐴𝑉 2 𝜌𝐴𝑉 2 𝜌𝐴𝑉 2 𝑉 2
= = = =
Gravity for𝑐𝑒 𝑚𝑎𝑠𝑠𝑥𝑔 𝜌𝑉𝑜𝑙𝑔 𝜌𝐴𝐿𝑔 𝐿𝑔
Taking square root
𝑉
Fr =
√𝐿𝑔

Inetia force 𝑉
3. Euler number (Eu) = pressure for𝑐𝑒 =
√𝑃/𝜌

Inetia force 𝜌𝐴𝑉 2 𝑉2

= =
pressure for𝑐𝑒 𝑃𝐴 𝑃/𝜌
Taking square root
𝑉
Eu =
√𝑃/𝜌

Inetia force 𝑉
4. Mach number (Ma) = Elastic for𝑐𝑒 =
√𝐾/𝜌

Inetia force 𝜌𝐴𝑉 2 𝜌𝐴𝑉 2 𝑉2

= = =
Elastic for𝑐𝑒 𝐸𝑙𝑎𝑠𝑡𝑖𝑐 𝑠𝑡𝑟𝑒𝑠𝑠 𝑥 𝐴 𝐾𝐴 𝐾/𝜌
Taking square root
𝑉
Ma =
√𝐾/𝜌

Inetia force 𝑉
5. Weber number (Wb) = Surface tension for𝑐𝑒 =
√𝜎/𝜌𝐿

Inetia force 𝜌𝐴𝑉 2 𝜌𝐿2 𝑉 2 𝑉2

= == =
Surface tension for𝑐𝑒 𝜎𝐿 𝜎𝐿 𝜎/𝜌𝐿
Taking square root
𝑉
Wb =
√𝜎/𝜌𝐿
V = Velocity, L = Length, P = Pressure, ρ = Density, K = Bulk modulus, µ = Dynamic viscosity,
υ = Kinematic viscosity, σ = Surface tension

10.5 Similarity Laws Or Model Laws

a. Reynolds model law

In addition to inertia force, if viscous force is the only predominant force, then the similarity of
flow in the model and its prototype can be established if the Re is same for both systems. This is
known as Reynolds model law.
(Re) model = (Re) prototype
𝑉𝑚 𝐿𝑚 𝑉𝑝 𝐿𝑝
=
𝜐𝑚 𝜐𝑝
Application of Reynolds model law: pipe flow, motion of air planes, motion of submarine
completely under water, flow around structure under moving fluid

Scale ratios using Reynold’s law

𝑉𝑟 𝐿𝑟
= 1 where subscript r represents respective scale ratio.
𝜐𝑟
𝜐
𝑉𝑟 = 𝐿𝑟
𝑟
𝐿
Time scale ratio (𝑇𝑟 ) = 𝑉𝑟
𝑟
𝑉
Acceleration scale ratio (𝑎𝑟 ) = 𝑇𝑟
𝑟

Force scale ratio (𝐹𝑟 ) = 𝑚𝑟 𝑎𝑟 = 𝜌𝑟 𝐴𝑟 𝑉𝑟 𝑎𝑟 = 𝜌𝑟 𝐿2𝑟 𝑉𝑟 𝑎𝑟

Discharge scale ratio (𝑄𝑟 ) = 𝐴𝑟 𝑉𝑟 = 𝐿2𝑟 𝑉𝑟
Work scale ratio (𝑊𝑟 ) = 𝐹𝑟 𝐿𝑟
𝐹𝑟 𝐿𝑟
Power scale ratio (𝑃𝑟 ) = 𝑇𝑟

b. Froude model law

In addition to inertia force, if gravity force is the only predominant force, then the similarity of
flow in the model and its prototype can be established if the Fr is same for both systems. This is
known as Froude model law.
(Fr) model = (Fr) prototype
𝑉𝑚 𝑉𝑝
=
√𝑔𝑚 𝐿𝑚 √𝑔𝑝 𝐿𝑝
Application of Froude model law: free surface flow such as flow over spillways, sluices, flow jet
from an orifice or nozzle

Scale ratios using Froude’s law

𝑉𝑚 𝑉𝑝
=
√𝑔𝑚 𝐿𝑚 √𝑔𝑝 𝐿𝑝
𝑉𝑟
= 1 where subscript r represents respective scale ratio. (gm = gp)
√𝐿𝑟

𝑉𝑟 = √𝐿𝑟
𝐿 𝐿𝑟
Time scale ratio (𝑇𝑟 ) = 𝑉𝑟 = = √𝐿𝑟
𝑟 √𝐿𝑟
𝑉 √𝐿𝑟
Acceleration scale ratio (𝑎𝑟 ) = 𝑇𝑟 = =1
𝑟 √𝐿𝑟

Force scale ratio (𝐹𝑟 ) = 𝑚𝑟 𝑎𝑟 = 𝜌𝑟 𝐿3𝑟 𝑥1 = 𝜌𝑟 𝐿3𝑟

Discharge scale ratio (𝑄𝑟 ) = 𝐴𝑟 𝑉𝑟 = 𝐿2𝑟 𝑉𝑟 = 𝐿2𝑟 √𝐿𝑟 = 𝐿2.5
𝑟
Work scale ratio (𝑊𝑟 ) = 𝐹𝑟 𝐿𝑟 = 𝜌𝑟 𝐿3𝑟 𝐿𝑟 = 𝜌𝑟 𝐿4𝑟
𝐹𝑟 𝐿𝑟 𝜌𝑟 𝐿4𝑟
Power scale ratio (𝑃𝑟 ) = = = 𝜌𝑟 𝐿3.5
𝑟
𝑇𝑟 √𝐿𝑟

c. Euler model law

In addition to inertia force, if pressure force is the only predominant force, then the similarity of
flow in the model and its prototype can be established if the Eu is same for both systems. This is
known as Euler model law.
(Eu) model = (Eu) prototype
𝑉𝑚 𝑉𝑝
=
√𝑃𝑚 /𝜌𝑚 √𝑃𝑝 /𝜌𝑝
Application: pressure rise due to sudden closure and opening of valve

d. Mach model law

In addition to inertia force, if elastic force is the only predominant force, then the similarity of flow
in the model and its prototype can be established if the Ma is same for both systems. This is known
as Mach model law.
(Ma) model = (Ma) prototype
𝑉𝑚 𝑉𝑝
=
√𝐾𝑚 /𝜌𝑚 √𝐾𝑝 /𝜌𝑝
Application: water hammer problems, aerodynamic testing
e. Weber model law

In addition to inertia force, if surface tension force is the only predominant force, then the similarity
of flow in the model and its prototype can be established if the Wb is same for both systems. This
is known as Weber model law.
(Wb) model = (Wb) prototype
𝑉𝑚 𝑉𝑝
=
√𝜎𝑚 /𝜌𝑚 𝐿𝑚 √𝜎𝑝 /𝜌𝑝 𝐿𝑝
Application: Flow over a weir involving very low heads, very thin sheet of liquid flowing over a
surface

10.6 Types of models

a. Undistorted model

An undistorted model is that model which is geometrically similar to its prototype. That means the
scale ratios for corresponding linear dimensions of the model and its prototype are same. Result
from such model can be directly applied to the prototype.

b. Distorted model

A distorted model is that model in which one or more terms of the model are not identical with
their counter-parts in the prototype. A distorted model may have either geometrical distortion, or
material distortion, or distortion of hydraulic quantities or a combination of these. Results obtained
from the distorted models cannot be directly applied. These models can be applied to modeling of
rivers, dams etc.

10.7 Scale effects in models, advantages and limitations of model

If complete similarity does not exist between a model and its prototype, there will be some
discrepancy between the results obtained from the model tests and those which will be indicated
by the prototype after construction. This discrepancy is called scale effect.

Advantages
I. The behavior and working of a structure or machine can be predicted.
II. A number of alternatives can be worked out.
III. Safety and reliability can be analyzed for a particular problem for which analytical method
cannot be used.
IV. Defects can be detected and rectified in case when existing structure is not functioning
properly.

Limitations

I. No any model is perfectly similar to its prototype. Therefore, experience and judgment is
required to analyze the results based on partial similarity.
II. Bigger model is expensive.
III. It is more difficult to predict the performance of actual body from its model of distorted type.
IV. In general the model results are qualitative but not quantitative.

References;

1. K.N Dulal, Hydraulics Notes for Civil Engineering, Kathmandu Nepal 2012.
2. D.P. Sangraula & P. Bhattarai, A Text Book of Hydraulics, Kathmandu Nepal, 2012
3. Arora K.R. Fluid Mechanics Hydraulics and Hydraulic Machines, Standard publishers,
New Delhi, 2005
4. R.K Bansal, A Text Book of Fluid Mechanics & Hydraulic Machines, Laxmi Publication,
New Delhi, 2010
5. Zoeb Hussain et al, Basic Fluid Mechanics & Hydraulic Machines, BS Publication,
Hyderabad, 2008
6. S.K Mondal, Fluid Mechanics & Hydraulic Machine, 2008
7. Hubert Chanson, The Hydraulics of Open Channel Flow: An Introduction, Elsevier, 2004
8. P.N Modi & S.M. Seth, Hydraulics and Fluid Mechanics, New Delhi,