ADVANCED FLUID MECHANICS

CHAPTER 8: SIMILITUDE AND DIMENSIONAL ANALYSIS

SIMILITUDE

Similitude or Similarity Law

Similarity laws play an important role in conducting the model studies. In the field of
hydraulics and aeronautics, the performance of the prototype (full size object, structure or device)
is predicted by conducting tests on the small scale models of the prototype based on similitude or
similarity laws.
Some examples of modeling studies include:
- Hydraulic turbines and pumps (centrifugal pumps)
- Spillways of dams
- River channels
- Ships
- Aero-planes in wind tunnels

There are three types of similarity laws which are used in the modeling studies. These laws include:
1. Geometric similarity
2. Kinematic similarity
3. Dynamic similarity

1. Geometric Similarity
In model studies, geometric similarity means that model and its prototype be identical in shape
but differ only in size. The ratio of the linear dimensions of the prototype to corresponding
dimensions in the model is called as scale ratio and is denoted by Lr. The scale ratio for area and
volume is denoted by Lr2 and Lr3 respectively. On the other hand, The ratio of the linear dimensions
of the model to corresponding dimensions in the prototype is called as model ratio and is denoted
by r.

2. Kinematic Similarity
Kinematic similarity implies that the ratio of the velocities at all corresponding points in the
flow of the prototype and model is the same.
- The velocity ratio (Vr) is given by:
Vp
Vr =
Vm
where subscripts p & m denote prototype and model respectively
- The time scale is given by:
Lr L
Tr = (∵ T = )
Vr V
- The acceleration scale is given by:
Lr Vr2
ar = 2 =
Tr Lr

Page 1 of 13
 3. Dynamic Similarity
A model and its prototype are dynamically similar if the ratio of various
corresponding forces in the model and the prototype is the same.
For example if the forces acting on a fluid element in the model and its prototype
include force due to gravity (FG), pressure force (Fp), viscous force (Fv) and inertia force (FI),
dynamic similarity between model and prototype will be achieved provided:

FGp FPp FVp FIp

= = =
FGm FPm FVm FIm

These relations can be expressed as:

All the below are dimensionless ratios
FIp FGp FI FI
= ⟹ ( ) = ( )
FIm FGm FG p FG m
FIp FIp F F
= ⟹ ( I) = ( I)
FIm FPm FP p FP m
FIp FVp FI FI
= ⟹ ( ) = ( )
FIm FVm FV p FV m

Importance of Dimensionless Ratios

The dimensionless ratios result is very important parameters which play an important role in
describing /characterizing fluid flow patterns in a completely filled conduit pipe flow or partially
filled conduit (open channel flow). Various important parameters which result from various
dimensionless ratios of forces are explained below:
Reynold Number (R)
In case of pipe flow inertia force and viscous force play a significant role in describing
flow pattern. The ratio of inertia force to viscous force results in Reynold number (R) as mentioned
below:
FI L2 V 2 ρ LVρ LV μ
R = = = = ∵ ν=
Fv LVμ μ ν ρ
VD
R =
ν
If a model and its prototype are to be dynamically similar so far as inertia and viscous forces are
concerned, they must have the same value of R. Thus, for such a case, dynamic similarity is
achieved where
VD VD
R𝑚 = ( ) = R𝑝 = ( )
ν 𝑚 ν 𝑝
Example Problem 1: If the Reynold number (R) of a model and its prototype are the same, find
an expression for Vr, Tr and ar.
Solution of example problem 1
As by definition:
Lm Vm Lp Vp
Rm = Rp ⇒ =
νm νp

Page 2 of 13
 Vp Lm νp νr ν
Vr = = = = ( )
Vm Lp νm Lr L r
ν
Vr = ( )
L r
Lr Lr L2r L2 ν
Tr = = ν = ν = (ν) ∵ Vr = (L)
Vr r/Lr r r r
L2
Tr = ( )
ν r
Or
νr
Vr Lr ν2r ν2
ar = = L2r
= = ( )
Tr L3r L3 r
νr
Froude Number (F)
Froude number (F) is the ratio of inertia force (FI) to gravity force (FG). Mathematically,
FI ρ L2 V 2 V2
F = = =
FG ρ g L3 gL
Or
V
F = gL
√

Inertia and gravity forces play an important role in describing flow pattern in case of:
- Wave action set up by a ship
- Flow of water in open channels
- Forces of a stream on a bridge pier
- Flow over a spillway
- Flow of a jet from an orifice
Thus, F is important for above mentioned cases.
V
F =
√g L
Where
L = some linear dimension increase of an open channel, L = depth of flow i.e. d
V
So, F = for open channel flow
√g d
In the cases, where FI, FG are dominant, dynamic similarity is achieved when:
V V
( ) = Fm = Fp = ( )
√g L m √g L p
In case of open channel flow if:

F<1 - flow in the channel is subcritical

F=1 - flow in the channel is critical
F>1 - flow in the channel is super critical

Mach Number (M)

Mach number is the ratio of inertia to the elastic forces i.e., (where compressibility is
important):

Page 3 of 13
 FI ρ L2 V 2
M = = ∵ FE = Ev A = Ev L2
FE E𝑣 L2
Or
V2 V V
M = Ev = =
E C
ρ √ ρv
Where
C = sonic velocity or celerity in the given medium
Thus M is the ratio of the fluid velocity (or the velocity of the body through a stationary fluid) to
that of a sound wave in the same medium.
Where:
M<1 - flow is sub sonic
M=1 - flow is sonic
F>1 - flow is super sonic
Where M value is extremely high – flow is hypersonic

Web Number (W)

Web number is the ratio of inertia force to surface tension i.e.
FI ρ L2 V 2
W = = ∵ FT = σ L
FT σL
Or
V2 V
𝑊 = σ ⟹
σ
ρL √ρ L

In case of flow where surface tension is important, W plays an important role to describe flow
pattern. For example W is applied to describe the flow at the leading edge of a very thin sheet of
liquid flowing over a surface.

Euler Number (F)

Where only FI & FP influence the flow. Euler number (E) is the ratio of inertia force to the pressure
force i.e.
FI ρ L2 V 2 𝑉
= = ∵ FP = (∆p) A
FP (∆p) L2 ∆p
√( ρ )
Or
V V
E = =
∆p ∆p
√2 ( ρ ) √2 g ( γ )

Example Problem 2
A certain submerged body is to move horizontally through oil ( = 8170 N/m2,  = 0.0287
N.s /m2) at a velocity of 15 m/s. To study the characteristics of this motion, a model of the body is
tested in 15oC water. The model ratio () is 8:1. Determine the velocity at which the model should

Page 4 of 13
be pulled through the water to achieve dynamic similarity. If the drag force on the model is 3.5 N,
predict the drag force on the prototype.

Solution (Example Problem 2)

Given:
p = 8170 N/m3
p = 0.0287 N.S/m2
Vp = 15 m/s
Lm Dm 8
model ratio = λ = = =
Lp Dp 1
Dm = 8 Dp

μp 0.0287 0.0287
νp = = =
ρp γp /g 81770/9.81
2
m
νp = 3.45 × 10−5 ∵ ρp g = γp
s
γ
ρg = gp
Liquid to be used in the model is water at 15oC.
So, Appendix 3, Table A-1 gives ν m = 1.14 x 10-6 m2/s
In the given problem, FI & Fv are dominant means they affect the flow pattern in the given case,
so, R is important in the given case:
DV DV
Thus, ( ν ) = ( ν ) will be used to calcualted
p m

i) Vm
Dp (15) (8 Dp ). Vm
= ⇒ Vm = 0.062 m/s
3.45 × 10−5 1.14 × 10−6
Vm = 0.062 m/s

ii) Drag Force on the Prototype

Fp ρp Vp2 L2p
F α ρ V 2 L2 ⟹ = 2 L2
∵ Lm = 8 Lp
Fm ρm Vm m

8170
Fp ( 9.81 ) (15)2 L2p
= = 761.7 ρm = ρw = 1000 Kg/m3
Fm 1000 (0.062)2 (82 Lp 2 )

Fp = 761.7 x Fm = 761.7 x 3.5 = 2666 N

Fp = 2666 N

Example Problem 3

Page 5 of 13
 A 1:50 model of a boat has a wave resistance of 0.02 N when operating in water at 1.0 m/s.
Find the corresponding prototype wave resistance. Find also the power requirement for the
prototype. What velocity does this test represent in the prototype?
Solution
Given:
Resistance = Fm = 0.02 N
Scale ratio = Lp/Lm = 50 = Lr
Vm = 1.0 m/s
Find
i) Fp = ?
ii) Power for prototype = ?
iii) Vp = ?

In the given problem, FI and FG are important means affect the flow pattern so F criteria will be
applied to solve the problem.
V V
Fp = Fm = ( ) =( )
√g L p √g L m
Vp Vm Vp2 2
Vm Vp 2 Lp
= ⟹ = ⟹ ( ) = ⟹ Vr2 = Lr = 50
√Lp √Lm Lp Lm Vm Lm

Fp ρ L2p Vp2 2 3
Vp 2
= = L L
r r = L r ∵ ( ) = Lr
Fm ρ L2m Vm2 Vm
Fp
= L3r
Fm
Fp = L3r Fm = (50)3 (0.02) = 2500 N

Fp = 2500 N

Vp2
2
= Lr ⇒ Vp2 = Lr Vm
2
Vm

Vp = Vm √Lr where Lr = 50

= 1.0 √50 = 7.1 m/s

Vp = 7.1 m/s
ii) Power required by the prototype = Pp
Fp Vp 2500 × 7.1
Pp = = = 17.75 kW
1000 1000
Pp = 17.75 kW

Page 6 of 13
DIMENSIONAL ANALYSIS

Concept of Dimensional Analysis

Dimensional analysis is a mathematical technique which deals with (make use) of the physical
quantities involved in the fluid flow phenomenon. It is based on the assumption that the
phenomenon can be expressed by a dimensionally homogeneous equation with certain variables.
Dimensional analysis is widely used for conducting model studies and in research work for
developing design criteria. Thus, dimensional analysis is related to similitude.
In dimensional analysis, first the physical parameters (quantities) which affect the fluid
flow phenomenon are predicted/indicated, and then these parameters are grouped into
dimensionless combinations. These combinations provide a better understanding of the flow
phenomenon. Dimensional analysis is commonly used to determine the relationship between
various variables affecting the flow phenomenon.
Fundamental Dimensions
There are two systems for fundamental dimensions, namely, MLT (i.e. mass, length and time) and
FLT (i.e. force, length and time). These two systems are inter-related through Newton’s Law which
states that force equals mass time acceleration, F = ma or F = M L/T2.
In dimensional analysis, all physical parameters (quantities) are expressed in either MLT system
or FLT system.

Dimensions as per MLT Dimensions as per FLT

Quantity
System System
Length (l) L L
Area (A) L2 L2
Volume (Δ) L3 L3
Time (t) T T
Velocity (V) LT-1 LT-1
Acceleration (a) LT-2 LT-2
Discharge (Q) L3T-1 L3T-1
Force or weight (F, W) MLT-2 F

Page 7 of 13
 Power (P) ML2T-3 FLT-1
Work Energy (E) ML2T-2 FL
Pressure (p) ML T -1 -2
FL-2
Mass (m) M FT2L-1
Mass density () ML-3 FT2L-4
-2 -2
Specific weight () ML T FL-3
Dynamic viscosity (µ) ML-1T-1 FTL-2
2 -1
Kinematic viscosity (ν) LT L2T-1
-2
Surface tension () MT FL-1
Shear stress () ML-1T-2 FL-2
-1 2
Bulk modulus (Ev) ML T FL-2
Following table provides dimensions of some important parameters (quantities) expressed
in terms of MLT & FLT system.

Page 8 of 13
Dimensional Homogeneity
An equation is called dimensionally homogeneous, at it has same fundamental dimensions
and the identical or the same powers of these dimensions (M-L-T or F-L-T) on both sides of the
equation. Let us consider discharge equation of the hydraulics.
Q=AV
Put this equation in dimensions. i.e.
Q=AV
L3T-1 = L2 . LT-1= L3T-1
So, both the right of left sides of the above equations have the same power of these dimensions
(L&T). Thus the equation is dimensionally homogeneous.

Example Problem 1:
Check the dimensional homogeneity of the following equations in the field of hydraulics:
i) Q = Cd a √2 g H
ii) V = C √Rh S
Solution:

Q = Cd a √2 g H in terms of dimensions:
L3T-1 = 1 x L2 x (LT-2 x L)1/2 = L2 x L1/2 x T-2 x 1/2 x L1/2
L3T-1 = L3 T-1 (dimension of Cd = constant is taken as 1)
As both sides of equation have putting of these dimensions, so given the same dimension and
equation is dimensionally homogeneous
ii) V = C √Rh S
As per dimensions LT-1 = 1 x (L x 1)1/2 = L1/2
LT-1 = L1/2
C = Constant so, its dimension is taken as 1
S = Slope which is dimensionless. So, its dimension is to be as 1
As both sides of above equation not have the same dimensions, so given equation is not
dimensionally homogeneous.

Methods of Dimensional Analysis:

1. Rayleigh’s method
2. Buckingham’s  theorem method

1) Rayleigh’s Method
The Rayleigh’s method of dimensional analysis expresses a functional relationship of same
variables in the form of an exponential equation. For example if y is same function of independent
variables X1, X2, X3 then functional relationship may be written as:
Y = f (X1, X2, X3 ………Xn)
In the above equation, Y is a dependent variable and X1, X2, X3….Xn are independent variables
which govern the variation of the dependent variable, Y.
The Rayleigh’s method consists of following steps:
1. Gather all the independent variables which will affect the dependent.

Page 9 of 13
 2. Write the functional equation considering the independent and dependent variables. For
example, if R is a dependent variable and R1, R2, R3.. Rn are independent variables, then
functional relationship can be written as R = f (R1, R2, R3.. Rn)
3. Write the above equation with a dimensionless constant (C) in the form of arbitrary
exponents (a, b, c,…m).
4. Express each of the quantities in some fundamental dimensions according to the MLT
system or FLT system whatever is required.
5. Solve these equations to obtain the value of exponents a, b, c,… m.
6. Substitute the values of exponents in the main equation and group the variables with similar
exponents.

Example Problem 2
When a sphere is moved through a viscous fluid, drag force (FD) is exerted by the fluid on the
sphere. This drag force is influenced by the diameter of the sphere (D) and velocity of the sphere
(V). Fluid mass density () and viscosity (µ) also affect the drag force. Using the Rayleigh’s
method of dimensional analysis, establish an equation to determine the drag force (F D) on the
sphere.
Solution

Given:
Sphere diameter = D
Sphere Velocity = V
Fluid density = 
Fluid viscosity = µ
Find: Establish an equation for drag force (FD)
Here,
FD = dependent variable
D, V,, µ = independent variables
Based on Rayleigh’s method;
1. Collect data of independent variables which are D, V,, µ
2. Write functional equation:
FD = f (D, V,, µ)
3. Write above equation with a constant (C) and in the form of exponents:
FD = C. Da. Vb. c. µd Main Equation (1)
4. Express each of quantities in fundamental dimensions using MLT system
MLT-2 = CLa (LT-1)b (ML-3)c (ML-1T-1)d
5. Using dimensional homogeneity (powers of dimensions on both sides of the equation
should be the same), obtain a set of simaltaneous equations involving exponents, a, b, c, d:
For M: 1=c+d Eq. (2)
For L: 1 = a + b – 3c – d Eq. (3)
For T: -2 = - b – d Eq. (4)

6. Solve these equations to obtain the value of exponents. As there are 3 equations and 4
unknowns, express 3 of the unknown in terms of fourth one. So solve for a, b, & c in terms of d:
c = 1–d from Eq. (2)
b = 2–d from Eq. (4)

Page 10 of 13
a = 2–d by putting above c & b values in Eq. (3)
Substitute exponent values in Main Equation (1) and group variable with similar exponents.
FD = C. D2-d. V2-d. 1-d. µd
= C. D2.D-d. V2.V-d.  -d .µ-d
= C  D2 V2 (µ/ D V)d
So,
FD =  D2 V2 C (µ/ D V)d =  D2 V2 C ( D V/µ)-d
Here
d = unknown
if d = -1 then above equation becomes:
ρDV
FD = ρ D2 V2 C ( )
μ
Where
ρDV
( μ ) = Reynolds Number = R
FD = C (R)  D2 V2
Or
FD = f’(R)  D2 V2
The above equation indicates that FD on a sphere is equal to some coefficient times  D2 V2, where
coefficient is a function of R.

2) Buckingham’s  - Theorem
The Buckingham’s  theorem states that if these are n variables in a dimensionally homogeneous
equation, and if these variables contain m fundamental dimensions such as (MLT) they may be
grouped into (n – m) non-dimensional independent -terms.
Mathematically, if a variable X1 depends upon independent variables X2, X3, X4,…Xn, the
dimensional equation can be written as;
X1 = f (X2, X3, X4,…Xn)
The equation can be written in its general form as:
f (X2, X3, X4,…Xn) = C
Where C is a constant and f represents some function. In this equation there are n variables. If there
are m fundamental dimensions, then according to Buckingham’s -theorem:
f1 = (1, 2, 3,…n-m) = Constant

Buckingham’s -method has two advantages compared to Rayleigh’s method.

1. When there is a large number of variables involved in the given fluid flow problem,
Buckingham’s -method can be used more conveniently compared to Rayleigh’s method.
2. Buckingham’s method tells one ahead of time how many dimensionless groups are to be
expected. Rayleigh’s method does not provide this information ahead of time.

The Buckingham’s -method consists of following steps:

1. First of all, write the functional relationship with the given data.
2. Then write the equation in its general form.
3. Now choose m repeating / primary variables and write separate expressions for each -term.
Every -term will contain repeating variables All the variables are written in exponential form.

Page 11 of 13
4. Using dimensional homogeneity rule, determine the values of exponents a, b, c,… by obtaining
simultaneous equations.
5. Now substitute the values of these exponents in -terms.
6. After the -terms are determined, write the functional relation in the required form.

Selection of Repeating/Primary Variables

Following points should be considered for the selection of repeating variables:
- Dimensionless variable should not be selected as a repeating variable
- No two variables should have the same dimensions
- Independent variables should be selected as repeating variables
In general the selected repeating variables include the first variable representing the fluid property
like , etc., the second variable representing the flow characteristics (V, a, g, Q, etc) and the third
variable representing the geometric characteristics of the body like diameter (D), length (L), etc.

Example Problem 3
Solve example problem 2 using Buckingham’s -method
Given Data
Sphere diameter = D
Sphere velocity = V
Fluid density = 
Fluid viscosity = µ
Determine: Establish an equation for drag force (FD) acting on the sphere.

Solution
1. According to Buckingham’s -method, write functional relationship between FD and
independent variables;
(D, V, , µ)….i.e. FD = f (D, V, , µ)
2. Write the above equation in its general form:

f (FD, D, V, , µ) = 0
3. Determine number of -terms as: no. of -terms = number of total variables (n) – number of
fundamental dimensional variable as per MLT system (m)
So, no. of -terms = n – m = 5 – 3 = 2
 = (1, 2) = 0
Take , D & V as the primary/repeating variables and write separate expressions for two -terms
in exponential form:
π1 = ρa1 Db1 Vc1 μd1
π2 = ρa2 Db2 Vc2 Fd2 D
4. Determine the values of exponents (a1, b1, c1, d1, & a2, b2, c2, d2) by expressing -terms (1
& 2) and all variables in terms of fundamental dimensions as per MLT System and using
dimensional homogeneity:
M a L c1 M d1
π = ρa1 Db1 Vc1 μd1 ⇒ Mo Lo To = (L3 ) Lb1 (T) (LT) as s are dimensionless, they
can be replaced by Mo Lo To .

Page 12 of 13
M: 0 = a1 + d1 Eq. (1)
L: 0 = -3a1 + b1 + c1 – d1 Eq. (2)
T: 0 = - c1 – d1 Eq. (3)
Solving for a1, b1 & c1 in terms of d1:
a1 = - d1
b1 = - d1 by putting Eq. (1) and Eq. (3) in Eq. (2)
c1 = - d1
5. Put the values of exponents (a1, b1, c1, d1) into -terms, 1 and get
-d1 -d1 -d1 d1
μ d1 ρ D V -d1
π1 = ρ D V μ = ( ) = ( )
ρDV μ
ρDV
if d1 =-1, = Reynold Number = R
μ
So, (R)-d1 = R or 1 = R

Similarly for 2:

a2 c2
M b2 L M d2
π2 = ρa2 Db2 Vc2 Fd2 o o o
D ⇒ M L T = ( ) L ( ) ( )
L3 T 𝑇2
M: 0 = a2 + d2 Eq. (1)
L: 0 = -3a2 + b2 + c2 + d2 Eq. (2)
T: 0 = - c2 – 2d2 Eq. (3)
Or
a2 = -d2
c2 = -2d2
b2 = -2d2
Putting exponents (a1, b1, c1, d1) in π2 , we get;
π2 = ρa2 Db2 Vc2 Fd2 D = π2 = ρ
-d2 -2d2 -2d2 d2
D V FD
FD
π2 =( 2 2 ) or d2
𝜌𝐷 𝑉
𝐹𝐷
π2 = 𝜌 𝐷2 𝑉 2 it is dimensionless group if d2 = 1
So,
FD
π2 =
ρ D2 V2
7. After establishing -terms, write the functional relation in the required form such as:
𝜙 (π1, π2 ) = 0 ⇒ π2 = 𝜙 ′′ (π1 )
So,
FD
2 2 = 𝜙 R ⇒ FD = 𝜙 ′′ (R) ρ D2 V2
′′ ( )
ρD V
′′ ( ) 2 2
FD = 𝜙 R ρ D V Required equation

Page 13 of 13