Dimensional Analysis

and Model Similitude

Hari Prasad Neopane, PhD

Associate Professor
Department of Mechanical Engineering
Kathmandu University
 Objectives

• Understand dimensions, units, and dimensional

homogeneity
• Understand benefits of dimensional analysis
• Know how to use the method of repeating
variables
• Understand the concept of similarity and how to
apply it to experimental modeling
 Dimensions and Units
• Dimension: refer to the qualitative characteristics for
physical quantities, e.g., length, time, mass
• Units: are standards of comparison for the quantitative
measure of dimensions, e.g., (m), (sec), (kg)
• The two most common systems of dimensioning a
physical quantity are the Mass-Length-Time (MLT) and
Force-Length-Time ( FLT) system of units
• There is no direct relationship between the quantities
length L, mass M and time T.
• These independent quantities are called fundamental
quantities.
 Dimensions and Units
• Derived or secondary quantities
• Quantities are expressed in terms of the
fundamental quantities
• Examples
• {Velocity} = {Length/Time} = {L/t}
• {Force} = {Mass Length/Time} = {mL/t2}
 Dimensional Homogeneity
• Law of dimensional homogeneity (DH): every additive
term in an equation must have the same dimensions
• Example: Bernoulli equation

• {p} = {force/area}={mass x length/time x 1/length2} = {m/(t2L)}

• {1/2V2} = {mass/length3 x (length/time)2} = {m/(t2L)}
• {gz} = {mass/length3 x length/time2 x length} ={m/(t2L)}
 Applications of DH
• It helps to check whether an equation of any physical
phenomenon is dimensionally homogeneous or not
• It helps to determine the dimensions of a physical
quantity
• Dimensional homogeneity helps to convert the units from
one system to another
• The concept of dimensional homogeneity is a step
towards dimensional analysis which is fruitfully employed
to plan experiments and to present the results
meaningfully
 Dimensional Analysis
• DA deals with the process whereby all the important
factors involved in a physical phenomenon are
systematically organized into dimensionless numbers
• For example, a law expressing the relation between fluid
density, viscosity, velocity and a length parameter ‘D’ is
designated by a dimensionless number ‘Re’
• The number of unknown quantities is thus reduced, the
problem is generalized, and the need for specifying a
particular system of units is eliminated
• The equations are true whether all the terms are
expressed in MKS or SI units or any other system of
units
Dimensional Analysis: Rayleigh’s Method

• The method involves the following steps:

• Gather all the independent variables which are likely to
influence the value of the dependent variable
• Write the functional relationship, i.e., if the dependent
variable ‘y’ is some function of the independent variables
x1, x2,x3,…., then y = f( x1, x2, x3, ….)
• Write the above eq. in the form: y = K(x1a, x2b, x3c,….)….
eq. 1
• Where ‘K’ is a dimensionless coefficient to be
determined either from the physical characteristics of the
problem or from experiments
 Rayleigh’s Method Contd..
• Express each of the quantities on both sides of eq. 1 in
the fundamental units of M, L, and T
• Utilize the principle of dimensional homogeneity to
obtain a set of simultaneous equations involving the
exponents ‘a’, ‘b’, ‘c’,..
• Solve the simultaneous equations to obtain the values of
exponents ‘a’, ‘b’, ‘c’ etc. If the number of exponents
involved is more than 3, then the exponents of the
properties ‘D’, ‘V’ and ‘density’ are evaluated in terms of
other exponents. That helps to group the variables into
recognized dimensionless parameter.
• Substitute the value of exponents in the main equation,
and form the non-dimensional parameters by grouping
the variables with like exponents.
 Example
• Show by dimensional analysis, that the power ‘P’ developed
by a hydraulic turbine is given by
 N 2  D2 
• Solution: P    N3  D  f 
5

 g  H 
• Here functional relationship is P  f ( D, N ,  , H , g )
or P  K (Da  N b   c  H d  g e )
eq. 1
• Where ‘K’ is a dimensionless coefficient. Using M-L-T system, the
corresponding eq. for dimensions is:
• [ML2T-3] = [1] [L]a [T-1]b [ML-3]c [L]d [LT-2]e
• The dimension of ‘K’, being a constant, is taken as unity
• For dimensional homogeneity, we can equate the exponents
of M, L and T on both sides:
M: 1= c; L: 2 = a-3c + d +e; T: -3 = -b-2e
 Exercise Contd..
• We have only three equations to solve for five variables.
Experience shows that recognized dimensionless groups
appear when exponents of ‘D’, ‘N’ and ‘density’ are
evaluated in terms of other exponents. So we express
‘a’, ‘b’ and ‘c’ in terms of ‘d’ and ‘e’. Then by simple
algebraic operations we obtain: c = 1; b = 3 - 2e; a = 5 –
d-e
• Substituting these values of exponents in eq. ;


P  K D5d e  N 32e   1  H d  g e 
 3 5  g e  H  d 
• Collecting like terms: P  K  N D  2    
  N D   D  
 N 2 D   D  3 5 N D 
2 2
 N D f 
3 5
,    N D f  
 g   H    gH 
 Dimensional Analysis: Buckingham’s Pi-theorem

• Rayleigh’s method of dimensional analysis becomes

increasingly laborious and cumbersome when a large
number of physical variables are involved. The difficulty
is then circumvented by using Buckingham’s Pi- theorem
that states:
• “If there are ‘n’ variables in a dimensionally
homogeneous equation and if these variables contain ‘m’
primary dimensions, then the variables can be grouped
in to (m - n) non-dimensional parameters”. The non-
dimensional groups are called pi-terms.
• Mathematically: given a physical equation; f (x1, x2,
x3,…xn) = 0
 Buckingham’s Pi-theorem Contd..
• Where X’s are dimensional physical quantities (such as
velocity, density, viscosity, pressure and area etc.)
pertinent to a physical phenomenon, then the same
phenomenon can be described by (n - m) dimensionless
pi-terms 
   ,  ,  .....
1 2 3 n m 
0

• Where ‘m’ represents the fundamental dimensions (such

as mass, length and time or force, length and time)
• Out of given physical variables, one has to select any
three variables which amongst them contain all the
fundamental units of M, L and T. These variables are not
to form non-dimensional parameters amongst
themselves
 Method of Repeating Variables
• Non-dimensional parameters  can be generated by
several methods.
• We will use the Method of Repeating Variables
• Six steps
• List the parameters in the problem and count their total number
‘n’.
• List the primary dimensions of each of the ‘n’ parameters
• Set the reduction ‘m’ as the number of primary dimensions.
Calculate ‘k’, the expected number of 's, ‘k’ = n - m.
• Choose ‘m’ repeating variables [a geometric property , a fluid
property and a flow property].
• Construct the ‘k’ 's, and manipulate as necessary.
• Write the final functional relationship and check algebra.
 Guidelines for choosing Repeating
parameters
1. Never pick the dependent variable. Otherwise, it may appear in all
the 's.
2. Chosen repeating parameters must not by themselves be able to
form a dimensionless group. Otherwise, it would be impossible to
generate the rest of the 's.
3. Chosen repeating parameters must represent all the primary
dimensions.
4. Never pick parameters that are already dimensionless.
5. Never pick two parameters with the same dimensions or with
dimensions that differ by only an exponent.
6. Choose dimensional constants over dimensional variables so that
only one  contains the dimensional variable.
7. Pick common parameters since they may appear in each of the 's.
8. Pick simple parameters over complex parameters.
 Example
• Step 1: List relevant
Ball Falling in a Vacuum parameters. Z = f(t,w0,z0,g) 
n=5
• Step 2: Primary dimensions of
each parameter

• Step 3: As a first guess,

reduction ‘m’ is set to 2 which
is the number of primary
dimensions (L and t). Number
of expected 's is ‘k’ = n - m =
5-2 = 3
• Step 4: Choose repeating
variables w0 and z0
 Example Contd..
• Step 5: Combine repeating parameters into products
with each of the remaining parameters, one at a time, to
create the ’s.
• 1 = zw0a1z0b1
• a1 and b1 are constant exponents which must be determined.
• Use the primary dimensions identified in Step 2 and solve for a1
and b1.

• Time equation:
• Length equation:

• This results in
 Example Contd..
• Step 5: continued
• Repeat process for 2 by combining repeating parameters with t
• 2 = tw0a2z0b2

• Time equation:

• Length equation:

• This results in
 Example Contd..
• Step 5: continued
• Repeat process for 3 by combining repeating parameters with g
• 3 = gw0a3z0b3

• Time equation:

• Length equation:

• This results in
 Example Contd..
• Step 6:
• Double check that the 's are dimensionless.
• Write the functional relationship between 's

 w0 t g z0 
1  f  2 ,  3 
z
  f  , 2 
z0  z0 w0 
• Overall conclusion:
• Method of repeating variables properly predicts the functional
relationship between dimensionless groups.
• However, the method cannot predict the exact mathematical
form of the equation.
 Exercise
• Using Buckingham’s Pi-theorem, show that the discharge
‘Q’ consumed by an oil ring is given by
3    w 
Q  Nd   , , 2 
 Nd 2
N 2 3
d N d
• Where ‘d’ is the internal diameter of the ring, ‘N’ is rotational
speed, ‘w’ is the specific weight
• Solution clue:
• Total number of variables, ‘n’ = 7
• Total number of fundamental dimensions, ‘m’ = 3
• Total number of pi-terms, ‘K’ = n - m = 7 - 3 = 4
• Choose ‘d’, ‘N’ and ‘density’ as repeating variable
 Dimensionless number
• Reynolds number: ratio of the inertia force to  V  l
Re 
viscous force

• Froude number: Ratio of the square root of the
V
inertia force to the square root of the force due Fr 
to gravity lg
• Mach number: ratio of the square root of the
V
inertia force to the square root of the force M 
stemming from the compressibility of the fluid C
(elastic force)
  l V 2
• Weber number: ratio of the inertia force to the W
force of surface tension 
• Euler number: ratio of the pressure force to the P
E
inertia force  V 2
 Model Studies
• Prototype and Model:
• Prototype is the full size structure employed in the actual
engineering design. The prototype operates under the actual
working conditions
• Model is a system by whose operation the characteristics of
other similar systems can be ascertained
• Experimental observations made on a model bear a definite
relationship to the prototype
• Model a mechanical analog of the prototype, is generally a small
scale replica of the prototype. However, in some cases, a model
may be even larger or of the same size as prototype depending
upon the need and purpose
 Objectives & Importance of Model Studies

• To study the geometrical appearance and relative proportions

of the system
• To examine the flow paths over, through / around the objects
• To determine the pressure distribution and the resulting
forces on the systems and the components
• To estimate the flow capacity and calibration of the various
flow passages
• To ascertain whether the prototype will the intended
performance
• To compute the mechanical energy loss resulting from
pressure and shear drags
• To ascertain the overall performance of a hydraulic machines
 Applications of Model Studies
• Civil engineering: Dams, spillways, cannel etc..
• Mechanical engineering: Pumps, compressors,
turbines
• Aeronautical engineering: Aero-planes, rockets,
missiles
• Naval engineering: ships, submarines
• Architectural engineering: tall building to predict
wind loads, stability characteristics, air flow
patterns etc..
 Exercise
• A pipe of diameter 1.5 m is required to transport an oil of
sp. gravity 0.9 and viscosity 0.003 Ns/m2 at the rate of
3000 l/s. Test were conducted on a 15 cm diameter pipe
using water at 20 0C. Find the velocity an rate of flow in
the model. Viscosity of water at 200C is 0.001 Ns/m2
• Solution Clue & Ans:
• Equate Reynolds number for model and prototype and find ratio
of model and prototype velocity and rate of flow through model
• 89.9 l/s
 Similitude
• Similitude refers to the theory and art of predicting
prototype conditions from model observations
• It prescribes the relationship between a full scale flow
and flow involving smaller but geometrically similar
boundaries
• The result obtained from experiments on models can be
applied to the prototype only if a complete similarity
exists between the model and prototype, and for that the
two systems must be
(i) Geometrically,
(ii) Kinematically, and
(iii) Dynamically similar
 Similarity
• Geometric Similarity - the model must be the
same shape as the prototype. Each dimension
must be scaled by the same factor.
• Kinematic Similarity - velocity as any point in
the model must be proportional
• Dynamic Similarity - all forces in the model flow
scale by a constant factor to corresponding
forces in the prototype flow.
• Complete Similarity is achieved only if all 3
conditions are met. This is not always possible,
e.g., river hydraulics models.
 Dimensional Analysis and Similarity

• Complete similarity is ensured if all independent

 groups are the same between model and
prototype.
• What is ?
• We let uppercase Greek letter  denote a non-
dimensional parameter, e.g., Reynolds number Re,
Froude number Fr, Drag coefficient, CD, etc.
• Consider automobile experiment
• Drag force is F = f(V, , L)
• Through dimensional analysis, we can reduce the
problem to
 Experimental Testing and Incomplete
Similarity
• One of the most useful applications of dimensional
analysis is in designing physical and/or numerical
experiments, and in reporting the results.
• Setup of an experiment and correlation of data.

• Consider a problem with 5 parameters:

one dependent and 4 independent.
• If each independent variables changed 4
times then it would require 54 = 625
experiments !!
• If we can reduce to 2 's, the number of
independent parameters is reduced from
4 to 1, which results in 51=5 experiments
vs. 625 !!
Experimental Testing and Incomplete
Similarity
Wanapum Dam on Columbia River
• Flows with free surfaces present
unique challenges in achieving
complete dynamic similarity.
• For hydraulics applications, depth is
very small in comparison to horizontal
dimensions. If geometric similarity is
used, the model depth would be so
Physical Model at small that other issues would arise
Iowa Institute of Hydraulic Research
• Surface tension effects (Weber
number) would become important.
• Data collection becomes difficult.
• Distorted models are therefore
employed, which requires empirical
corrections/correlations to extrapolate
model data to full scale.
Experimental Testing and Incomplete Similarity
DDG-51 Destroyer
• For ship hydrodynamics, ‘Fr’ similarity is
maintained while Re is allowed to be
different.
• Why? Look at complete similarity:

1/20th scale model

• To match both ‘Re’ and ‘Fr’, viscosity in

the model test is a function of scale ratio !
This is not feasible.
 Classification of Models
• Undistorted models & distorted models
• A model is said to be undistorted if it is geometrically similar
to its prototype or in other words if the scale ratio for the
linear dimensions of the model and its prototype is same
• The behavior of the prototype can be easily predicted from
the results of undistorted model
• A model is said to be distorted if it is not geometrically
similar to its prototype.
• For distorted model different scale ratios for the linear
dimensions are adopted
• E.g. in case of rivers, harbors, reservoirs etc., two different
scale ratios, one for horizontal dimensions and other for
vertical dimensions are taken
 Classification of Models Contd..
• Thus the models of rivers, harbors and reservoirs will
become as distorted models
• If for the river, the horizontal and vertical scale ratios are
taken to be same so that the model is undistorted, then the
depth of water in the model of the river will be very-very
small which may not be measured accurately
• The following are the feature of distorted models:
• The vertical dimensions of the model can be measured accurately
• The cost of the model can be reduced
• Turbulent flow in the model can be maintained
• The result of the distorted model cannot be directly transferred to its
prototype
Scale ratios for distorted models
• (Lr)H = Scale ratio for horizontal Lp Bp
dimension
Lr H  
Lm Bm
hp
Lr V 
• (Lr)V = Scale ratio for vertical hm
dimension
Vp
 Lr V
• Scale ration for velocity, Vp/Vm Vm

Ap
• Scale ratio for area of flow,  Lr H  Lr V
Ap/Am Am

• Scale ratio for discharge, Qp/Qm

 Lr H  Lr V 
Qp 3/ 2

Qm
 Exercise
• The discharge through a weir is 1.5 m3/s. Find the
discharge through the model of the weir if the horizontal
dimension of the model = 1/50 the horizontal dimension
of the prototype and vertical dimension of the model =
1/10 the vertical dimension of the prototype
• Answer:
• (Lr)H = 50
• (Lr )V =10
• Qp/Qm= 0.000948 m3/s