Similarity and model testing

• In most experiments, to save time and money, tests are performed on

a geometrically scaled model, rather than on the full-scale prototype.

• There are three necessary conditions for complete similarity between

a model and a prototype.

• Geometric similarity: the model must be the same

shape as the prototype, but may be scaled by
some constant scale factor.

• Kinematic similarity: the velocity at any point in

the model flow must be proportional (by a
constant scale factor) to the velocity at the
corresponding point in the prototype flow.
Geometric similarity is a prerequisite for kinematic
similarity.

1
 Similarity and model testing
• Dynamic similarity: when all forces in the model flow scale by a constant factor
to corresponding forces in the prototype flow. Kinematic similarity is a
necessary but insufficient condition for dynamic similarity

• Scale factor for forces can be less than, equal to, or greater than one

• To ensure complete similarity, the model and prototype must be geometrically

similar, and all independent nondimensional groups must match between model
and prototype.

• By making use of principle of similarity, it becomes possible to predict the

performance of one machine from the results of tests on a geometrically similar
machine, and also to predict the performance of the same machine under
conditions different from the test conditions

• As long as corresponding nondimensional parameters are equal, actual values

of dimensional parameters used (density, velocity, etc.) in experiments is
irrelevant. (automobile or aircraft performance can be simulated in a water
tunnel , and the performance of a submarine can be simulated in a wind
tunnel)
2
 Dimensional analysis
• Dimensional analysis is a process of reducing number of variables
effecting physical problems and formulating them in terms of
nondimensional numbers and parameters.

• Primary purposes of dimensional analysis

1. To generate nondimensional parameters that help in the
design of experiments (physical and/or numerical) and in the
reporting of experimental results (e.g. A problem require 5
parameter (one is dependent). For testing every possible
combination of five levels for each parameters, 625 number of
experiments are needed. Assume 3 primary dimensions are
represented in the problem. Only 5 experiments are needed)

2. To obtain scaling laws so that prototype performance can be

predicted from model performance

3. To (sometimes) predict trends in the relationship between

parameters 3
 Nondimensional analysis
for rotating machines
• Dimensional analysis of problems in turbomachines identifies the
variables involved and groups them into non-dimensional quantities,
which are much lesser in number than the variables themselves.

• In a design or performance test of a turbomachine, these non-

dimensional quantities are varied instead of the large number of
parameters forming these groups.

• The design procedure uses these non-dimensional numbers to obtain

optimum performance.

• Some non-dimensional numbers give an idea of the type of machine

and its range of operation. Such information enables us to select the
type of machine best suited to a particular application.

4
 Buckingham PI theorem
• There are several methods of reducing a number of dimensional
variables into a smaller number of dimensionless groups, but the most
popular (and simplest) method is the Buckingham PI theorem

Step-by-step procedure for finding the dimensionless  groups

1. List the parameters (dimensional variables, nondimensional variables

(e.g. angle of rotation), and dimensional constants (e.g. g)) and count
them. Let n be the total number of parameters in the problem,
including the dependent variable. Make sure that any listed
independent parameter is indeed independent of the others, i.e., it
cannot be expressed in terms of them. (E.g., don’t include radius r and
area A=πr2, since r and A are not independent.)

2. List the primary dimensions for each of the n parameters.

5
3. Guess the reduction j. As a first guess, set j equal to the number of primary
dimensions represented in the problem. The expected number of  's is equal
to n minus j, according to the Buckingham Pi theorem. If at this step or during
any subsequent step, the analysis does not work out, verify that you have
included enough parameters in step 1. Otherwise, go back and reduce j by one
and try again.

4. Choose j repeating parameters that will be used to construct each  . Since the
repeating parameters have the potential to appear in each  's , be sure to
choose them wisely.

5. Generate the  's one at a time by grouping the j repeating parameters with one
of the remaining parameters, forcing the product to be dimensionless. In this
way, construct all  's . By convention the first  , designated as 1, is the
dependent. Manipulate the  's as necessary to achieve established
dimensionless groups

6. Check that all the  's are indeed dimensionless.

6
Guidelines for choosing repeating parameters

7
• In turbomachine problems, the four basic dimensions are usually
taken to be mass M, length L, time T and temperature Θ

Units and dimensions of relevant variables for turbomachines

Quantity Unit Dimension

Length m L
Area m2 L2
Volume m3 L3
Velocity m/s LT-1
Mass flow rate kg/s MT-1
Pressure kg/ms2 ML-1T-2
Density kg/m3 ML-3
Rotational speed rev/s T-1
Volume flow rate m3/s L3T-1
Power kg/m2s ML-2T-1
Gravity m/s2 MT-2

8
• The technique of dimensional analysis is used to reduce the number of
variables into a few number of dimensionless groups.

• The dimensionless groups involved in rotating machines useful for plotting

the necessary performance curves

Step 1: Let us express the outlet pressure from the compressor, P2, as
function of other independent variables

P2  f  P1 ,  , D, m, N 

P2 outlet pressure from the compressor

P1 inlet pressure to the compressor
D characteristic linear dimension
 density of the fluid
m mass flow rate of the fluid
N rotational speed 9
 Total number of variables n = 6

Step 2: List primary dimensions of each variable

1
P : ML T 2
D : L  : ML3 m : MT 1 N : T 1

Step 3: As a first guess, take reduction j as number of primary dimensions

represented in the problem, that is 3

• Expected number of nondimensional numbers  are n – j = 6 - 3 = 3

Step 4: Take P1 ,  , D as repeating variables to construct  's

Step 5: Now we combine these repeating parameters into products with each
of the remaining parameters, one at a time, to create the  's

• First  is with dependent variable is

1  P1a  b D c P2
10
• We apply the primary dimensions of step 2 for every variable and force the 
to be dimensionless by setting the exponent of each primary dimension to
zero

1  M 0 L0T 0   M a L aT 2 a   M b L3b   Lc   M 1L1T 2 

 M a b1 L a 3bc 1T 2 a  2

• Equating exponents

• By solving these three equations simultaneously, we get

a   1; b  0; c  0

P2
therefore 1 
P1
11
• In similar fashion we create the  2 by combining the repeating
parameters with independent variable N.
1  P1a  b D c N

 2  M 0 L0T 0   M a L aT 2 a   M b L3b   Lc  T 1 

 M a b L a 3bcT 2 a 1

• Equating exponents

• By solving these three equations simultaneously, we get

1 1
a ; b ; c 1
2 2

ND 
therefore 2 
P1 12
In similar fashion we create the  3 by combining the repeating parameters
with independent variable N.

3  P1a  b D c m

3  M 0 L0T 0   M a L aT 2 a   M b L3b   Lc   MT 1 

 M a b1 L a 3bcT 2 a 1

• Equating exponents

• By solving these three equations simultaneously, we get

1 1
a   ; b   ; c  2
2 2

m
therefore 3 
D 2 P1  13
 • Step 6: We should double-check that the  are indeed dimensionless. We
can write the functional relationship between nondimensional parameters.

1  f   2 ,  3 

P2  ND  m 
 f , 2 
P1  P1 D P1 
 

• Dimensional analysis cannot predict the exact mathematical form of the

equation.

• In the above analysis, viscosity was not considered. Generally, over the
operating range of compressors and turbines the variation of μ has a negligible
effect on their performance

14