CE 16

Fluid Mechanics for Civil Engineering II

Dimensional
Analysis
& Similitude

Dimensional Analysis and Similitude 1

 Dimensional Analysis
• Definition : Dimensional analysis is a process of formulating fluid mechanics problems in
terms of dimensionless parameters or groups by studying the dimensions
of the quantities or variables
• Why is it used :
• Reduction in variables ( If F(A1, A2, … , An) = 0, then f(P1, P2, … Pr < n) = 0,
where, F = functional form, Ai = dimensional variables, Pj = dimensionless parameters, m = number of
important dimensions, n = number of dimensional variables, r= n – m ). Thereby the number of
experiments required to determine f vs. F is reduced.
• Based on the principle that an equation expressing a physical relationship between variables must be
dimensionally homogeneous, the dimensions on each side of the equation must be the same
• Provides qualitative rather than quantitative relationships
• When combined with experimental methods, results in accurate prediction equation

Variables in Fluid Mechanics

Boundary Geometry Fluid Flow Fluid Property

length (L) time (T) velocity (LT-1) mass density (ML-3)
area (L2) acceleration (LT-2) discharge (L3T-1) specific weight (ML-2T-2)
volume (L3) force (MLT-2) pressure (ML-1T-2) absolute viscosity (ML-1T-1)
work (ML2T-2) energy (ML2T-2) elasticity (ML-1T-2)
power (ML2T-3)
Dimensional Analysis and Similitude 2
 Applications Drag = f(V, L, r, m, c, t, e, T, etc.)
From dimensional analysis,

Vortex formed behind a cylinder

Dimensional Analysis and Similitude 3

 Buckingham p Theorem
1. A prediction equation involving n variables in m dimensions may be
reduced to a set of (n-m) dimensionless groups
– Typically: all variables required to describe the problem geometry (D)
or define fluid properties (r, m) and fluid flow (velocity, acceleration)
2. Express each variables in terms of MLT dimensions (m)
3. Determine the required number of dimensionless parameters (n – m)
4. Select a number of repeating variables = number of dimensions
– Each should represent boundary geometry, fluid flow and fluid property
– The most common repeating variables are Velocity, Mass Density, and
a Length Parameter
5. The repeating variables are combined with each of the other nonrepeating
variables and by a proper choice of exponents arrive at the dimensionless
products
6. Repeat for each nonrepeating variable
7. Express result as a relationship among the dimensionless parameters

Dimensional Analysis and Similitude 4

 Manipulation of p Groups
Once identified manipulation of the p groups is permitted. These manipulations do not change the number
of groups involved, but may change their appearance drastically.

Taking the defining equation as: f ( p1 , p2 , p3 ……… pn-m ) = 0

Then the following manipulations are permitted:
•Any number of groups can be combined by multiplication or division to form a new group which
replaces one of the existing. E.g. p1 and p2 may be combined to form p1a = p1 / p2 so the defining
equation becomes
f ( p1a , p2 , p3 ……… pn-m) = 0
•The reciprocal of any dimensionless group is valid. So f ( p1 ,1/ p2 , p3 ……… 1/pn-m ) = 0 is valid.
•Any dimensionless group may be raised to any power. So f ( (p1 )2, (p2 )1/2, (p3 )3……… pn-m ) = 0 is
valid.
•Any dimensionless group may be multiplied by a constant.
•Any group may be expressed as a function of the other groups, e.g.
p2 = f ( p1 , p3 ……… pn-m)
In general the defining equation could look like
f ( p1 , 1/p2 ,( p3 )i……… 0.5pn-m ) = 0
Dimensional Analysis and Similitude 5
 Example 1

• Find: Drag force on rough sphere is function

of D, r, m, V and l. Express in form:

p 3 = f (p 1, p 2 )

FD D r m V l

MLT-2 L ML-3 ML-1T-1 LT-1 L

p 1 = m ( D aV b r c )
M 0 L0T 0 = ( ML−1T −1 )( L) a ( LT −1 )b ( ML−3 ) c
n=6 No. of dimensional parameters M: 0 = 1+ c  c = −1
m=3 No. of dimensions L: 0 = −1 + a + b − 3c  a = −1
r = n - m = 3 No. of dimensionless parameters T: 0 = −1 − b  b = −1

Select “repeating” variables: D, V, and r m rVD

Combine these with nonrepeating variables: F, m & l p1 = or p 1 = Re =
DVr m

Dimensional Analysis and Similitude 6

 Example 1

FD D r m V l

MLT-2 L ML-3 ML-1T-1 LT-1 L

p 3 = FD ( D aV b r c )
Select “repeating” variables: D, V, and r
Combine these with nonrepeating variables: F, m & l M 0 L0T 0 = ( MLT −2 )( L) a ( LT −1 ) b ( ML−3 ) c
p 2 = l ( D aV b r c ) M: 0 = 1+ c  c = −1
M 0 L0T 0 = ( L)( L) a ( LT −1 ) b ( ML−3 ) c L : 0 = 1 + a + b − 3c  a = −2
M: 0=c  c=0 T : 0 = −2 − b  b = −2
L: 0 = 1 + a + b − 3c  a = −1
T: 0 = −b  b=0 FD
p3 =
rV 2 D 2
l
p2 =
D
FD rVD l
= f ( , )
rV 2 D 2 m D

Dimensional Analysis and Similitude 7

 Example 2

Force F on a body immersed in a flowing fluid depends

on: L, V, r, and m
 ( F , L, V , r , m ) = 0
n=5 No. of dimensional parameters p 1 = m ( LaV b r c )
m=3 No. of dimensions M 0 L0T 0 = ( ML−1T −1 )( L) a ( LT −1 )b ( ML−3 ) c
r = n - m = 2 No. of dimensionless parameters
M: 0 = 1+ c  c = −1
F L V r m L: 0 = −1 + a + b − 3c  a = −1
T: 0 = −1 − b  b = −1
MLT-2 L LT-1 ML-3 ML-1T-1

m rVL
Select “repeating” variables: L, V, and r p1 = or p 1 = Re =
LVr m
Combine these with the rest of the variables: F & m
Reynolds number

Dimensional Analysis and Similitude 8

 Example 2

F L V r m

MLT-2 L LT-1 ML-3 ML-1T-1

p 2 = F ( LaV b r c )
M 0 L0T 0 = ( MLT − 2 )( L) a ( LT −1 ) b ( ML−3 ) c
M: 0 = 1+ c  c = −1
L: 0 = 1 + a + b − 3c  a = −2
T: 0 = −2 − b  b = −2

F
p2 = and p 2 = f (p 1 )
L2V 2 r
F
= f (Re) Dimensionless force is a function
rV 2 L2 of the Reynolds number

Dimensional Analysis and Similitude 9

Dimensional Analysis and Similitude 10
Dimensional Analysis and Similitude 11
 Example 3

Dimensional Analysis and Similitude 12

 1 k D 2.51 
= −2 log  +
f
1
2  3.7 Re f12


Dimensional Analysis and Similitude 13

 Common Dimensionless Numbers
rVd
• Reynolds Number (inertial to viscous forces) Re =
– Important in all fluid flow problems
m

• V
Froude Number (inertial to gravitational forces) Fr =
– Important in problems with a free surface gy
• Euler Number (pressure to inertial forces) p
Eu =
– Important in problems with pressure differences rV 2

• V V
Mach Number (inertial to elastic forces) M= =
– Important in problems with compressibility effects K/r c

• Weber Number (inertial to surface tension forces) rLV 2

W=
– Important in problems with surface tension effects 
Dimensional Analysis and Similitude 14
 Similitude: Prototype and Model

• Scales: model and prototype (full-scale)

• Selection of the model scale: governed by dimensional analysis and similitude

Dimensional Analysis and Similitude 15

Physical Models

Dimensional Analysis and Similitude 16

 Similitude
• Definition : Flow conditions for a model test are completely similar if all significant
dimensionless parameters have the same corresponding values for model and prototype.
▪ Pi model = Pi prototype
▪ Enables conversion of measured quantities from model to prototype
values
▪ However, complete similarity usually is not possible. Therefore, often it is
necessary to use Re, or Fr, or M scaling, i.e., select most important P and
accommodate others as best possible

Models in Fluid Mechanics

▪ Mathematical Models
▪ Physical Models – undistorted, distorted

• Similitude between Model and Prototype

▪ Geometric Similarity : all body dimensions in all three coordinates have the
same linear-scale ratios. Ratio of linear dimensions
▪ Kinematic Similarity : homologous (same relative position) particles lie at
homologous points at homologous times. Ratio of velocity components
▪ Dynamic Similarity : in addition to the requirements for kinematic similarity
the model and prototype forces ratio must be in a constant. Ratio of
homologous forces
Dimensional Analysis and Similitude 17
 Similitude

• Similitude
– Predict prototype behavior from
model results
– Models resemble prototype, but are
• Different size (usually smaller)
and may operate in
• Different fluid and under
• Different conditions
– Problem described in terms of
dimensionless parameters which
may apply to the model or the p 1 = f (p 2 , p 3 ,...,p n )
prototype

– Suppose it describes the prototype p 1 p = f (p 2 p , p 3 p ,..., p np )

–
p 1m = f (p 2m , p 3m ,...,p nm)
A similar relationship can be written
for a model of the prototype

Dimensional Analysis and Similitude 18

 Similitude
• If the model is p 2m = p 2 p
designed & operated
under conditions that p 3m = p 3 p Similarity
requirements or
... modeling laws
p nm = p np

• then p 1m = p 1 p Dependent variable

for prototype will be
the same as in the
model

Dimensional Analysis and Similitude 19

Dimensional Analysis and Similitude 20
 rVd
Re =
m

V
M=
K/r

V
Fr =
gy

rLV 2
W=


p
Eu =
rV 2

Dimensional Analysis and Similitude 21

Dimensional Analysis and Similitude 22
 Similitude Ratios for Flow Variables

▪ Having decided on the Model Law, similitude ratios for

the various flow variables are determined in terms of the
Length Ratio LR
▪ Velocity Ratio corresponds to the Model Law
▪ With pressure computations, the Euler criteria is satisfied
Re m = Re p
For example,
using Reynold’s Law Vm Lm Vp Lp
=
m p
Lp  m
Vm = Vp
Lm  p
Dimensional Analysis and Similitude 23
Similitude Ratios Using Froude’s Law

Dimensional Analysis and Similitude 24

 Example
• Consider predicting the drag on a
thin rectangular plate (L*B) placed
FD = f ( r , V , L, m , B )
normal to the flow.
• Drag is a function of: B, L, m, r, V
p 1 = f (p 2 , p 3 )
• Dimensional analysis shows: FD L rVL
• And this applies BOTH to a model = f ( , )
and a prototype
L2 rV 2 B m

• We can design a model to predict the p 1m = f (p 2 m , p 3m )

drag on a prototype. FDm Lm r mVm Lm
• = f ( , )
Model will have:
L2m r mVm2 Bm mm

p 1 p = f (p 2 p , p 3 p )
• And the prototype will have:
FDp L p r pV p L p
= f( , )
L2p r pV p2 Bp mp

Dimensional Analysis and Similitude 25

 Example
•Similarity conditions
Geometric similarity
Lm L p Bm
p 2m = p 2 p =  Lm = L p Gives us the size of the model
Bm B p Bp

Dynamic similarity
r mVm Lm r pV p L p mm r p Lp
p 3m = p 3 p =  Vm = Vp
mm mp m p r m Lm
Then Gives us the velocity in the model

2 2
FDm FDp  Lp  r p  Vp 
p 1m = p 1 p =  FDp =    F
L2m r mVm2 L2p r pV p2  L  r  V  Dm
 m m m

Dimensional Analysis and Similitude 26

 Example
• Given: Submarine moving below surface in
sea water
(r=1015 kg/m3, =m/r=1.4x10-6 m2/s).
Fm Fp
Model is 1/20-th scale in fresh water (20oC). =
• Find: Speed of water in the test for dynamic r mVm2lm2 r pV p2l 2p
similarity and the ratio of drag force on
Fm r mVm2lm2
model to that on prototype. =
• Solution: Reynolds number is significant F p r pV p2l 2p
parameter. 2 2
Re m = Re p 1000  28.6   1 
=    
Vm Lm V p Lp 1015  2   20 
=
m p Fm
= 0.504
Lp  m Fp
Vm = Vp
Lm  p
20 1
= 2m / s
1 1 .4
Vm = 28.6 m / s
Dimensional Analysis and Similitude 27
Sample Problems

Dimensional Analysis and Similitude 28