Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

Chapter 2

2.0 DIMENSIONAL ANALYSIS, SIMILITUDE &HYDRAULIC MODELS

2.1. INTRODUCTION

Hydraulics is an experimental science & is also a complex subject. Most phenomena involving
the movement of water are known to be dependent on many variables- geometric characteristics,
fluid properties and flow characteristics. Much of raw experimental data were incomplete,
ambiguous, & incorrectly taken.
It is usually impossible to determine all the essential facts for a given fluid flow by pure theory,
&hence, dependence must often be placed up on experimental investigations. The number of tests
to be made can be greatly reduced by the systematic use of Dimensional Analysis and the laws of
similitude or similarity. Dimensional Analysis is a mathematical technique, which makes use of
the study of dimension as an aid to the solution of several engineering problems.
Dimension less grouping reduces the number of variables that have to be processed.
Dimension analysis can be used to obtain functional relation ship among the variables in terms of
non- dimensional parameter. It helps in obtaining a systematic form of the variables involved in
the problem. However, dimensional analysis doesn’t give complete relation ship; it gives only a
general relation ship.

Some of the uses of dimensional analysis are narrated as follows: -

i. Testing the dimensional homogeneity of any equation of fluid motion
ii Deriving equations expressed in terms of non- dimensional parameters to show the
significance of each parameter.
iii planning model tests & presenting experimental results in a systematic manner in terms
of non- dimensional parameters; thus making it possible to analyse the complex fluid
flow phenomena.
Hydraulic structures or machines can be designed using:
(i) Pure theory,
(ii) Empirical methods,
(iii) Semi-empirical methods which are mathematical formulations based on theoretical
concepts supported by suitably designed experiments,
(iv) Physical models,
(v) Mathematical models.

i) The purely theoretical approach in hydraulic engineering is limited to a few cases of laminar
flow, for example the Hagen Poisseuille equation for the hydraulic gradient in the laminar flow of
an incompressible fluid in a circular pipeline.
ii) Empirical methods are based on correlations between observed variables affecting a
particular physical system. Such relationships should only be used under similar circumstances to
those under which the data were collected. Due to the inability to express the physical interaction
of the parameters involved in mathematical terms some such methods are still in use. One well-
known example is in the relationship between wave heights, wind speed and duration for the
forecasting of ocean wave characteristics.
iii) A good example of a semi-empirical relationship is the Colebrook white equation for the
friction factors in turbulent flow in pipes. This was obtained from theoretical concepts and
experiments designed on the basis of dimensional analysis; it is universally applicable to all
Newtonian fluids.

1
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

iV) Dimensional analysis also forms the basis for the design and operation of physical scale
models, which are used to predict the behavior of their full –sized counterparts called
‘prototypes’. Such models, which are generally geometrically similar to the prototype, are used in
the design of aircraft, ships, submarines, pumps, turbines, harbors, breakwaters, river and estuary
engineering works, spillways, etc.
V) The mathematical modeling techniques have progressed rapidly due to the advance of high-
speed digital computers, enabling the equations of motion coupled with semi-empirical
relationships to be solved for complex flow situations such as pipe network analysis, pressure
transients in pipelines, unsteady flows in rivers and estuaries, etc., there are many cases,
particularly where localized flow patterns can not be mathematically modeled, when physical
models are still needed.
With out the technique of dimensional analysis experimental and computational progress in fluid
mechanics would have been considerably retarded.

2.2. DIMENSIONAL ANALYSIS

Dimensional analysis is a method, which describes a natural phenomenon by dimensionally

correct equation among certain variables, which affect the phenomenon. It is a mathematical
method, which is of considerable value in problems that occur in fluid mechanics.
In dimensional analysis, from a general understanding of fluid phenomenon, one first predicts the
physical parameters that will influence the flow, and then by grouping these parameters in
dimensionless combinations, a better understanding of the flow phenomena is made possible.
Application of dimensional analysis:
 Developing equations –reducing number of variables in an experiment.
 Producing dimensionless parameters – establish the principle of model design.
FUNDAMENTAL DIMENSIONS
In an equation expressing physical relation ship between quantities, absolute numerical and
dimension equality must exist. The magnitude of such quantities as mass, length, force,
acceleration etc are expressed differently in different units of measurement.
The mass, length, and time are independent of each other and their units of measurement are
prescribed by international standards and called fundamental quantities. The units of all other
quantities may be determined either from their definitions or from the physical laws describing
them.
In place of mass, the force is also sometimes considered as fundamental quantity. The force and
the mass are related by the Newton's second law as follows
F = m .a
F=m.a = (M) (LT-2)

DIMENSIONS OF PHYSICAL QUANTITIES USED IN FLUID MECHANICS

Number Quantity Symbol Dimensions (MLT)

1 Length L L
2 Time T T
3 Mass m M
4 Force F M L T -2
5 Velocity V L T -1
6 Acceleration a L T -2
7 Area A L2
8 Discharge Q L T -1
3

9 Pressure P M L-1 T-2

2
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

10 Power P ML2T-3
11 Density  M L-3
12 Unit gravity force γ ML-2 T -2
13 Dynamic viscosity  M L-1T-1
14 Kinematic viscosity L2 T-1
15 Surface Tension  MT-2
16 Bulk modulus of elasticity K ML-1 T-2
17 Shear stress ‫ﺥ‬ ML-1T-2

DIMENSIONAL HOMOGENEITY and DIMENSIONLESS RATIOS

The principle of dimensional homogeneity is greatly useful in establishing the forms of equations,
for solving problems involving physical phenomena, from the knowledge of relevant variables
and their dimensions.
An equation is said to be dimensionally homogeneous if the form of the equation doesn’t depend
up on the units of measurement. All rational equations that relate physical quantities must be
dimensionally homogeneous. I.e., all terms in an equation must have the same dimensions.
As all the terms in the dimensionally homogeneous equation have the same dimensions; the ratio
of these dimensions becomes dimensionless ratios.
Some times non-homogeneous equations are said used in fluid mechanics, like Manning’s
equation, Chezy’s equation. Such equations have resulted from fitting equation to observed data.

METHODS OF DIMENSIONAL ANALYSIS

The dimensional analysis is a power full tool in formulating problems of physical phenomenon. It
must be solved experimentally. The dimensional analysis accomplishes this by formation of
dimension-less groups containing relevant variables. There are several approaches for the
dimensional analysis. However, all the methods are absolutely dependent on the correct
identification of all the factors, which govern the physical event being analyzed.
The application of dimensional analysis to any practical problem is based on the assumption that
certain variable, which affect the phenomenon are independent variables, and all variables other
than these and the dependent variable, are irrelevant and have no bearing on the phenomenon.

Steps in dimensional analysis

☺ To decide which variables enter the problem
☺ Formation of dimensionless groups of the variables
☺ Requires thorough understanding
☺ Judgment and experience

1. BUCKINGHAM -π THEOREM
The Buckingham - pi method is widely used in the dimensional analysis of a problem and
expresses the resulting equation in terms of dimensionless groups ( terms).
☺ It arranges variables in dimensionless groups.
☺ It reduces the number of variable.
Rules:
(a) If a phenomenon involves n variables and these variables are described by m fundamental
dimensions, they will produce n - m groups
f1 (1,2,3... n-m)=constant;
(b) Each  - group should be a function of maximum of m+1 variable

3
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

(c) m repeating variables are selected from amongst the n - variables so that in combination
contain m fundamental dimensions
Selection of repeating variables
1st - from those describing the geometry of flow. (Diameter, length)
2nd - from those representing fluid properties.
(Density, viscosity, surface tension, elasticity, vapor pressure)
3rd-from those characterizing the fluid motion.
(Velocity, acceleration, discharge, pressure, force, power)
It is conventional to select a length dimension for 1st group.
Density dimension for 2nd group. .
Velocity dimension for 3rd group.
(d) The governing quantities must not combine among themselves to form dimensionless groups.
And each -term is dimensionless, the final function must be dimensionless, and therefore
dimensionally homogeneous.
f  1 ,  2 ,  3 ,  ,  n  m   M
0
L 0 T 0

Example
A scale model test have been carried out on a new hydraulic machine, the experiment team has
been presented the following data
Thrust force F M1L1 T -2
The flow velocity V L1 T -1
Viscosity  M L-1 T -1
Density  ML-3
Size of system L L, are given
Produce meaning full dimensionless ratios
Solution:
F, v,, , L
n = 5, m= 3 so n-m=2 groups ( 2 π-groups)
The repeating variables are , L and v.
  L  x1  M  y1 z1  ML 

1 =V  L F  M L T     ,  3  , L ,  2  
x1 y1 z1 0 0 0
 T   L   T  

  L x2  M  y2 z2  M 

2 =V  L   M L T     ,  3  , L 
x2 y2 z2 0 0 0


 T   L   LT 
 
After equating both equations:
F
1 =
V 2 L2

2 =
VL
 F  
 2 2 , 0
 V L VL 
F  VL 
or 2 2
  
V L   

4
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

 VL 
OR F= V 2 L2  
  
Example-
1) Using the variables, Q, D, ΔH/L, ρ, μ, g as pertinent to smooth pipe flow, arrange them in to
dimension less parameters by the Buckingham’s π theorem.
Solution: - The pertinent variables are
Q, D, ΔH/L, ρ, μ, and g
- Functional relation
ƒ (Q, D, ΔH/L, ρ, μ, g)
- Select the repeating variables
D, ρ, Q
- Write the pi – parameters

2.3. SIMILITUDE AND HYDRAULIC MODELS

Hydraulic Models
The design of a major hydraulic system may be approached
☺ By theoretical reasoning and use of numerical models Q =V*A.
☺ On the basis of previous experience of similar system.
☺ By scale model experiments true distorted.
It is known that even with modern computing facilities, many complex problems still challenge
complete theoretical analysis. A combination of past experience, theory and dimensional analysis
will provide partial or complete solution to a number of problems. However, there still remain
many problems, which are tractable only through experimentation. This will be done through
model studies of proposed hydraulic structures and machines.
Models permit visual observation of the flow and make it possible to obtain numerical data.
E.g. Calibrations of weirs and gates, depth of flow, velocity distributions, forces on gates,
efficiencies and capacities of pumps and turbines, pressure distribution and losses.
The main objectives achieved through dimensional analysis and subsequent model testing may
be:

 To obtain a relation ship amongst variables influencing a problem in terms of dimensionless

parameters. The knowledge of this relation ship is essential for planning and conducting the
model tests and there after in presenting the test results.
 To obtain experimental data from model tests for predicting the performance of its prototype
under different conditions of flow it may be subjected to during its operation.
 To check the validity of assumptions made in the design of certain hydraulic structures. The
design of many of hydraulic structure is usually based on certain assumptions, which may not
be fully satisfied. It is always in the best interest, to find out whether the assumptions made
are on the safer side and also to ascertain whether the structures so designed will serve the
desired purpose.

The model may be larger, smaller or even of the same size as the prototype depending on type of
fluid used. The choice of the fluid and the geometrical scale will depend only by the practical
considerations. If complete similarity is to exist between the flow in the prototype and the flow in
its model, every dimensionless parameter referring to the conditions in the models must have the
same numerical value as the corresponding parameters referring to the prototype.
Hydraulic models may be either true or distorted. True models have all significant characteristic
of the prototype reproduced to scale (geometrically similar) and satisfy discharge restrictions
(kinematically and dynamically similar).

5
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

The models in which it is not possible to maintain geometric similarity is known as distorted
models

TYPES OF SIMILARITY
For complete similarity to exist between the model and prototype, they must be geometrically,
kinematically and dynamically similar.

 GEOMETRIC SIMILARITY
It is the similarity of the shape (scale factor). It is obtained when the solid boundaries that control
the follow of fluid are geometrically similar. The model is a geometric reduction of the prototype
and is accomplished by maintaining a fixed ratio of all homologous lengths between the model
and prototype. These physical quantities are length, area, diameter, volume, etc.

HP

Bp MODEL
LP

PROTOTYPE
Lp Bp Hp
Model scale ratio: Lr =   (1)
Lm Bm Hm

Ap L p* * B p
Area Ratio: Ar =   L2r
Am Lm * Bm

Lp * Bp * H p
Volume ratio: Vr = =Lr3
Lm * Bm * H m

 KINEMATIC SIMILARITY

It is the similarity of motion. For kinematic similarity to exist the streamline pattern in the model
must be the same as in its prototype. The ratios of kinematic quantities representing the flow
characteristics such as, time, velocity, acceleration, and discharge must be the same at all
corresponding points.
Vp
The velocity ratio is: Vr =
Vm
Tp Lr
Time scale ratio: Tr =  {b/c T= L/V}
Tm Vr

6
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

ap Lr Vr2
Acceleration scale ratio: ar    {b/c a=[L/T2] }
am Tr2 Lr

L3 p
Qp Tp L3r
Discharge scale ratio: Qr =  
Qm L3m Tr
Tm
 DYNAMIC SIMILARITY

It is the Similarity of forces involved in motion. Dynamic Similarity is attained if the ratio of
homologous forces in the model and prototype are kept constant.
Fp
i.e. = Fr
Fm
The conditions required for complete similarity are developed form the Newton 2 nd law of motion
 F x  ma x
The forces acting may be any one or a combination of several of the following: viscous, pressure,
gravity, elasticity, surface tension, inertia forces etc.
 Forces (viscous  pressure  gravity  elasticity  surface tension)p mpap

 Forces (viscous  pressure gravity  elasticity  surface tension)m mm am

      


F  Fv  Fg  Fs  F p  Fe  RESULTANT   Fi  m. a
NOTE: Newton’s Law: Inertia force (Fi) is equal and opposite to the resultant forces.

 F  p m.a p Fi p
  (1)
 F  m
m.a m Fi m
However, in practice, a mode is designed to study the effects of only a few dominant forces.
Dynamic similarity requires that the ratios of these forces be kept the same between the model
and prototype.
 v2 
Inertia force, Fi= m *a =     p 2 v 2
3

 
du v
Viscous force, Fv = ‫ ح‬A =  A   * 2  . .
dy 
Gravity force, Fg = m. g =  . 3 g
Pressure force, Fp = p. A = . 2
Elastic force, Fe = Ev.A= Ev.  2
Surface tension force, Fs = l

In problems of fluid flow, the inertia force will always exist and hence it is customary to find out
the force ratios with respect to the inertia forces, thus:

7
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

Inertia- to- viscous forces ratio

 Fi  F 
    i  (2)
 Fv  m  Fv  p

Inertia-to gravity forces ratio

 Fi   
    Fi  (3)
F   
 g  m  Fg  p

Inertia to elastic forces ratio

 Fi  F 
    i  (4)
 Fe  m  Fe  p

Inertia- to surface tension forces ratio

 Fi  F 
    i  (5
 Fs  m  Fs  p

Inertia -to pressure forces ratio

 Fi   
    Fi  (6)
F   
 p  m  Fp  p

☺ when the two systems are geometrically, kinematicallyt and dynamically similar, then they
are said to be completely similar or complete similitude exists b/n the two systems.
The above six equations are dimensionless groups. The significance of the dimensionless ratios is
discussed below:

(a) REYNOLDS NUMBER (phenomenon governed by viscous force)

A fluid in motion always involves inertia forces. If the inertial forces and viscous forces (example
pipe flow) can be considered to be the only forces that govern the motion, the ratio of these forces
acting on homologous particles in a model and its prototype is defined by the Reynolds number.
Fi
Re = = (Inertial force)/(Viscous force)
Fv
L2V 2  VL
=  LV  (Non dimensional ratio)
VL  

 This is for flow of fluid in pipe. And also for airplane traveling at speed below that at
which compressibility of the air is appreciable. Further, for a submarine submerged far
enough so as not to produce waves on the surfaces.

8
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

This states that when the inertial force and the viscous force are considered to be the only forces
governing the motion, the Reynolds number of the model and of the prototype or of two pipelines
of different fluids, must be kept constant. Thus:

 LV   LV 
   Re m   Re  p   
  m   p

NOTE: D is taken as L for pipe flow.

(b) FROUDE NUMBER: (phenomenon governed by gravity force)

When inertial forces and gravity forces are considered to be the only dominant forces in the fluid
motion, the ratio of inertia forces to gravity forces acting on the homologous elements of the fluid
in the model and prototype are considered as follows:
 Fi  L2V 2  V 2 
   
 Fg  gL3
   gL 

 V 
The square root of this ratio   is known as Froude number.
 gL 
 
V
Therefore, Fr =
gL
 This is used for the wave action setup by a ship, the flow of water in open channels, the
forces of a stream on a bridge pier, the flow of jet from an orifice, and so on.
In open hydraulic structures
 Spill way
 Weirs
 Channel transitions.
 Sluices etc
For dynamically similitude model and prototype:
 V   
   Fr (mod el )  Fr ( proto)   V 
 gL   gL 
 m  p

N.B For open channel case L is the depth of the channel.

For a ship, L is taken as the length at water line.
 In some flow situations fluid friction is a factor of gravity as well as Inertia. Hence,
Reynolds number & Froude number must both be satisfied simultaneously. To satisfy
this, we have to use fluids of different viscosity for both model and prototype. Thus:
3
2
 m  Lm 

 p  Lp 

 The scale number for Froude number similarity will be:

9
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

Vp
Vr =  Lr (for the same Fr and g).
Vm
Tp Lr
 Time ratio: Tr    Lr
Tm Vr
V
 Acceleration ratio: ar  r  1
Tr
5
 Discharge ratio: Qr  Vr Ar  Lr 2

(c) WEBER NUMBER (phenomenon governed by surface tension)

The surface tension is a measure of energy level on the surface of a liquid body. The force is of
primary importance in hydraulic engineering practice in the study of small surface waves or
control of evaporation from a large body of water, such as water storage tank / reservoir.
In river and harbor models reduction of scale often leads to appreciable viscous and capillary
effects in the shallow regions of flow. The depth of flow in such cases should be sufficiently large
so that capillary effects are negligible.
The ratio of Inertia to Surface tension forces in prototype and model is:
Fi  V 2 L2 V 2
 
Fs L 
L
The square root of this dimensionless ratio is known as WEBER NUMBE (We):
V
We=

L
It is applied at the leading edge of a very thin sheet of liquid flowing over a surface. Like:
 Capillary movement of water in solids.
 Flow of liquid at a very small depth over a surface.
 Flow over weir at very small heads.
 Spray of liquid from the exit of discharging tube resulting in the formation of drops
of liquids.
(d) MACH NUMBER (phenomenon governed by elastic forces)
The Mach number can be regarded as the ratio of inertia and elastic forces. In problems where the
compressibility of the fluid becomes important, the elastic force must be considered. The high-
speed motion through air causes compressibility effect (elastic force). The Mach number is
therefore, measures of the effects of compressibility.
 Aerodynamic testing.
 Flow gases exceeding the velocity of sound.
 Water hammer problems. (Design of surge tanks).
The ratio of Inertia forces& elastic forces in prototype and model is:
Fi L2V 2 V 2
 
Fe EL2 E

The square root of this dimensionless number is known as
MACH NUMBER (M). Thus:

10
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

V V
M  
E C

This is for fluid velocity (or velocity of the body through a stationary fluid) to that of a
sound wave in the same medium.
C is the sonic velocity (or celerity) in the given medium.

(e) EULER NUMBER (Phenomenon governed by pressure forces.)

The ratio of Inertia forces to pressure forces for both prototype and model is given by:
Fi V 2 L2 V 2
 
Fp PL2 P

The square root of this dimensionless number is known as EULER NUMBER (Eu). Thus:

V V V
Eu   
P 2P
  2 g  P 
 

Similarity Laws or Model laws

The results obtained from the model tests may be transferred to the prototype by the use of
model laws, which may be developed, from the principle of dynamic similarity.

a) Reynolds Model laws: for inertia & viscous forces are only predominant forces, the similarity
of flow in the model & its prototype can be established if the Reynolds number is the same
for both systems.

(Re) model = (Re) prototype.

 m vm Lm  p Vp L p  p Vp Lp m
  . . . 1
m p  m Vm Lm  p

 r .Vr .Lr V .L
 1 r r
r r

b) Froude Model Law: when gravitational force is addition to the inertia force, is the only
predominant forces which controls the motion, the similarity of flow in any two such systems
(model &prototype) can be established if the Froude number for both the system is the same.

F r mod el  F r  prototype.
Vm Vp

g m Lm g p .L p

11
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

Vr
 1 or Vr  g r L r  Lr
g r Lr

(For most of the case gr = gp /gm =1)

c) Euler Model Law

(Eu) model =(Eu) proto

Vm Vp Vr
1
 1
 1
1
2 2 2
 Pm   Pp   Pr 
        
 m  p  r

Vr  Pr
r

Types of Models

In general hydraulic models can be classified under two broad categories

I) Undistorted models
II) Distorted models

I) Undistorted Models: - if a model is geometrically similar to its prototype, it is known as

undistorted models. i e., the scale ratios for the corresponding linear
dimension are the same.
II) Distorted models: - if one or more terms of the models are not identical with the prototype it
is known as distorted models. The distortion may be geometrical, or
material or hydraulic quantities or a combination of these.
In geometrical distortion, the distortion can be either of dimension or that of configuration.

When different scale ratios are adopted for the longitudinal, transverse, & vertical dimensions;
then it is known as distortion of dimensions.
It is adopted in river models where a different slope ratio for depth is adopted.

The distortion of configuration results when the general configuration of the model doesn’t have
resemblance with its prototype. If a river model has different bed slope ratio, this is distortion of
configuration.

The material distortion is occurred when the physical properties of the material used in the model
and prototype are different.

The distortion of hydraulic quantities is occurred for certain uncontrollable hydraulic quantities
such as time, discharge etc

Distorted models are required to be prepared for rivers, dams across very wide rivers, harbors,
and estuaries etc. for which the horizontal dimensions are large in proportion to the vertical
ones.

12
 Hydraulics II Chapter 2 Dimensional Analysis, Similitude & Hydraulic Models

The following are some of the reasons fro adopting distorted models:
a) to maintain accuracy in vertical measurements;
b) to maintain turbulent flow;
c) to obtain suitable bed material & its adequate movement;
d) to obtain suitable roughness condition;
e) to accommodate the available facilities such as space, money, water supply
& time.

13