CLL231: Fluid Mechanics for Chemical

Engineers

Somnath Ghosh
[email protected]

Video: https://www.youtube.com/
1
Dimensional Analysis
(Munson Ch7; White Ch: 5)

2
Need for dimensional analysis
Assume, we wish to run a test on smooth sphere/cylinder of diameter D and want to
understand the drag force. ρ→ change
ρ, μ → const μ → fixed
FD = f (D,V, ρ, μ)
V

ρ, μ

ρ→ fixed
μ → change

3
Dimensional analysis and Similitude
What is dimensional analysis?
Method for reducing the number of experimental variables that influences a physical
phenomena.

What is Similitude?
Translating model data (dimensional analysis) to a full-scale object based on similarity.
❖ Dimensional analysis helps to reduce the number of variables. If a physical phenomena
depends on n-number of variables; with dimensional analysis, it could be reduced to k-
number of dimensionless variables. It has several advantages:-
a) Reduces number of experiments; thus, reducing time and cost.
b) Provides insight about the relationship about different physical parameters.
c) Helps to understand the scaling law that can convert data from model to prototype.
When scaling law is valid; a condition of similarity exists.
4
Dimensional analysis and Similitude
Purpose:
❖ In case, the force F acting on an object immersed in a stream of fluid, depends on length L,
stream velocity V, fluid density ρ and viscosity μ; then F= f(L,V, ρ, μ). Instead of performing
many experiments to understand F and generalize the physical relationship, one can non-
dimensionalize the equation to the equivalent form;
F𝐷 𝜌𝑉𝐷
2 2 = 𝑓(
𝜌𝑉 𝐷 𝜇
)
𝐶𝐹 = 𝑓(𝑅𝑒)

In simple situations, similarity is obtained if Re is same for model and prototype.

If, Rep=Rem then Cfm = Cfp which means,
𝐹𝑝 𝜌𝑝 𝑉𝑝 2 𝐿𝑝 2
=
𝐹𝑚 𝜌𝑚 𝑉𝑚 𝐿𝑚

This is scaling law.

5
Consideration for Similarity analysis
❖ Geometric similarity
Model and prototype have similar linear scale ratio
in all three co-ordinates.

❖ Kinematic similarity
Similarity of motion.

❖ Dynamic similarity
Similarity in 1) length-scale ratio 2) time-
scale ratio and 3) force-scale (mass) ratio.

Kinematic and dynamic similarity holds only if there is a geometric similarity.

6
Dimensions and Units
Dimensions are the symbolic representation of physical properties, independent of units.
Primary dimensions
Mass → [M] ; Length → [L] ; Time → [T]
Temperature →[ϴ]; Electric current→[A]
Light intensity →[cd]; Charge→ [C]
Other dimensions are generally the product of these dimensions.
List of derived dimensions
Parameters MLTϴ FLTϴ definition
Acceleration LT-2 LT-2 rate of velocity change in time
Force MLT-2 F mass x acceleration
Area L2 L2
Work ML2T-2 FL force x distance
Power ML2T-3 FLT-1 work/time
Density ML-3 FL-4T2 mass/volume
Viscosity ML-1T-1 FL-2T resistance to deformation
Pressure/Stress ML-1T-2 FL-2 force/area
Moment/torque ML2T-2 FL force x distance 7
Dimensionless groups in Fluid mechanics
Several non-dimensional groups are there to characterize flow behaviour
List of non-dimensional groups
1. Reynolds Number (Re)
3 𝑉2
𝑖𝑛𝑒𝑟𝑡𝑖𝑎𝑙 𝑓𝑜𝑟𝑐𝑒 𝑚𝑎𝑠𝑠 ∗ 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (𝜌𝐿 )( ) 𝜌𝑉𝐿 Important for all allow problems
= = 𝐿 =
𝑣𝑖𝑠𝑐𝑜𝑢𝑠 𝑓𝑜𝑟𝑐𝑒 𝑠ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 ∗ 𝑎𝑟𝑒𝑎 𝑉 𝜇
(𝜇 )𝐿2
𝐿
2. Froude Number (Fr)
𝑉2 3
𝑖𝑛𝑒𝑟𝑡𝑖𝑎𝑙 𝑓𝑜𝑟𝑐𝑒 𝑚𝑎𝑠𝑠 ∗ 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (𝜌𝐿 )( ) 𝑉 2 𝑉
= = 𝐿 = ; 𝐹𝑟 =
𝑔𝑟𝑎𝑣𝑖𝑡𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑓𝑜𝑟𝑐𝑒 𝑚𝑎𝑠𝑠 ∗ 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 𝑑𝑢𝑒 𝑡𝑜 𝑔𝑟𝑎𝑣𝑖𝑡𝑦 (𝜌𝐿3 )𝑔 𝐿𝑔 𝐿𝑔
For flow with free surface
3. Weber Number (We)
𝑉 2
𝑖𝑛𝑒𝑟𝑡𝑖𝑎𝑙 𝑓𝑜𝑟𝑐𝑒 𝑚𝑎𝑠𝑠 ∗ 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (𝜌𝐿3 )( ) 𝜌𝑉 2 𝐿
= = 𝐿 = When surface tension is
𝑠𝑢𝑟𝑓𝑎𝑐𝑒 𝑡𝑒𝑛𝑠𝑖𝑜𝑛 𝑓𝑜𝑟𝑐𝑒 𝑠𝑢𝑟𝑓𝑎𝑐𝑒 𝑡𝑒𝑛𝑠𝑖𝑜𝑛 ∗ 𝑙𝑒𝑛𝑔𝑡ℎ 𝜎𝐿 𝜎 important in free surface flow

8
Dimensionless groups in Fluid mechanics
List of non-dimensional groups
4. Euler Number (Eu)
𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 𝑓𝑜𝑟𝑐𝑒 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 ∗ 𝑎𝑟𝑒𝑎 ∆𝑝𝐿2 ∆𝑝 When pressure differences
= = 2 = 𝜌𝑉 2 are of interest
𝑖𝑛𝑒𝑟𝑡𝑖𝑎𝑙 𝑓𝑜𝑟𝑐𝑒 𝑚𝑎𝑠𝑠 ∗ 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 𝑉
(𝜌𝐿3 )( )
𝐿
5. Mach Number (Ma)
𝑉2
3
𝑖𝑛𝑒𝑟𝑡𝑖𝑎𝑙 𝑓𝑜𝑟𝑐𝑒 𝑚𝑎𝑠𝑠 ∗ 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (𝜌𝐿 )( ) 𝑉 2
𝑉 2 When compressibility of
= = 𝐿 = = 2 fluid is important
𝑒𝑙𝑎𝑠𝑡𝑖𝑐 𝑐𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑖𝑏𝑖𝑙𝑖𝑡𝑦 𝑓𝑜𝑟𝑐𝑒 𝑒𝑙𝑎𝑠𝑡𝑖𝑐 𝑚𝑜𝑑𝑢𝑙𝑢𝑠 ∗ 𝑎𝑟𝑒𝑎 𝐸𝐿2 𝐸ൗ
𝜌 𝑎
𝐸ൗ → 𝑠𝑜𝑛𝑖𝑐 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦; 𝑀𝑎 ≡ 𝑉
𝜌 𝑎
6. Capillary Number (Ca)
𝑉
𝑣𝑖𝑠𝑐𝑜𝑢𝑠 𝑓𝑜𝑟𝑐𝑒 𝑠ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 ∗ 𝑎𝑟𝑒𝑎 (𝜇 )𝐿2 𝜇𝑉 When interfacial tension is
= = 𝐿 =
𝑠𝑢𝑟𝑓𝑎𝑐𝑒 𝑡𝑒𝑛𝑠𝑖𝑜𝑛 𝑓𝑜𝑟𝑐𝑒 𝑠𝑢𝑟𝑓𝑎𝑐𝑒 𝑡𝑒𝑛𝑠𝑖𝑜𝑛 ∗ 𝑙𝑒𝑛𝑔𝑡ℎ 𝜎𝐿 𝜎 important

Other important groups: Grashof number (Gr; for natural convection); Prandtl
Number (Pr; for heat convection); friction factor (f; for internal or external flow);
Drag or lift coefficient (CD or CL; for hydrodynamics) 9
Buckingham Pi Theorem
Now we got to know about fundamental units, derivation of other units from them,
reduction of number of variables, dimensionless groups that can appear in a different
relationships of physical parameters or in governing equations. But the question is how
many dimensionless groups we need, to replace the original list of variables? Once we have
that understanding, we can simplify the general relationships or governing equations to
reduce the complexities of the experiments. Answer is the pi-theorem.
If a physical process (or equation) satisfies the principle of dimensional homogeneity and
involves ‘n’ variables; the relationship could be reduced to ‘k’ dimensionless variables or
Π’s. The reduction, j = n-k equals the maximum number of variables that do not form a pi
among themselves and is always less than or equal to the number of dimensions describing
the variables.
So, if there are ‘n’ physical parameters and ‘j’ reference dimensions, then the main
relationship can be reduced to k=n-j number of dimensional parameters or Π’s.
Further simplification: if a process involves n=6 variables and there are j=3 reference dimensions;
then we will have k=n-j i.e. k=6-3=3 pi-groups.
The commonly used method of finding pi-terms is also known as method of repeating variables. 10
Buckingham Pi Theorem-methods
General guidelines to find out pi-terms:-
1. Select relevant independent variables involved in the problem (n) and define general
functional form
Crucial step. If any important variable is missing, dimensional analysis will not be correct.
Typically, the variables will include those that are necessary to describe the geometry of the
system, to define fluid properties, and to indicate external effects that influence the system
2. List the basic dimensions of variables (j)
For general fluid mechanics problems, basic dimensions are M,L,T or F,L,T
3. Calculate the required number of pi-terms (k=n-j)
This could be done by the help of pi-theorem
4. Select the repeating variables; number of repeating variables =j. Remaining variables are
non-repeating variables
Criteria for repeating variables: a) not dimensionally dependent b) not dimensionless c)
cannot themselves be combined to form a dimensionless group d) collectively they should
contain all the dimensions
11
Buckingham Pi Theorem-methods
5. Write pi-terms by combining all the repeating variables with each of the remaining non-
repeating variables.
Take remaining variables one by one and combine with all the repeating variables with
exponents.
Ex: Π𝟏 = 𝑺𝒊 𝑺𝒂𝟏 𝑺𝒃𝟐 𝑺𝒄𝟑 …
𝑺𝒊 → 𝒏𝒐𝒏 𝒓𝒆𝒑𝒆𝒂𝒕𝒊𝒏𝒈 𝒗𝒂𝒓𝒊𝒂𝒃𝒍𝒆𝒔; 𝑺𝟏 𝑺𝟐 𝑺𝟑 𝒂𝒓𝒆 𝒓𝒆𝒑𝒆𝒂𝒕𝒊𝒏𝒈 𝒗𝒂𝒓𝒊𝒂𝒃𝒍𝒆𝒔; 𝒂, 𝒃, 𝒄 𝒂𝒓𝒆 𝒆𝒙𝒑𝒐𝒏𝒆𝒏ts]
6. Solve equations from step 5
Equating the combination of exponents of the dimensions with respective exponents of the
dimensionless form; one can solve for the exponents. Numerical value of the exponents are
determined so that the combination could be dimensionless. Plug-in the values of exponents
in the equations of step 5; all the pi-terms are obtained.

7. Write down the functional relationship in the form of pi’s.

Express the final form combining all the pi-terms.
12
Examples: drag coefficient
Ex1. Determine the drag force FD on a sphere by a flowing fluid.
Step1:
Drag force depends on density ρ and viscosity μ of the fluid medium; diameter D of the
object and fluid stream velocity. General function, FD = f (ρ,μ,D,V)
So, n=5
Step2:
Dimensions of the variables: MLT format
Drag force (FD) ≡ MLT-2
Density (ρ) ≡ ML-3
Viscosity (μ) ≡ ML-1T-1
Diameter (D) ≡ L
velocity (V) =LT-1
Fundamental dimensions j =3
Step3:
Number of pi-terms: k=n-j=5-3=2. There are two independent dimensionless
groups (Π1 and Π2). 13
Examples: drag coefficient
Step4:
Select repeating variables. D,V and ρ are the repeating variables as fulfilling the criteria.
Also, consider at least one variable form all the relevant properties.
a) Geometric properties (D) (could be L, H, etc.)
b) Flow properties (V) (could be acceleration, angular velocity, etc.)
c) Fluid properties (ρ or μ)
Step5:
Π1 = DaVbρcFD and Π2 = DxVyρzμ-1 ;
pi- terms are dimensionless (based on the definition of Re, -1 is taken)
Π1 = DaVbρcFD = M0L0T0 = (L)a(LT-1)b (ML-3)c (MLT-2) ;
Π2 = DxVyρzμ-1 = M0L0T0 = (L)x(LT-1)y (ML-3)z (ML-1T-1)-1
Step6:
For Π1 For Π2
L: a+b-3c+1=0 form these we find L: x+y-3z+1=0 form these we find
M: c+1=0 a = -2; b = -2; c = -1 M: z-1=0 x = 1; y = 1; z= 1
T: -b-2=0 T: -y+1=0 14
Examples: drag coefficient
Step6:
F𝑫
So, Π1 = DaVbρcFD = D-2V-2ρ-1FD = = 𝑪𝑭
𝝆𝑽𝟐 𝑫𝟐
𝝆𝑽𝑫
Π2 = DxVyρzμ -1 = D1V1ρ1μ-1 = = 𝑹𝒆
𝝁
Step7:
F𝑫
So, the functional relationship: Π1 =Π2 => = 𝝃(𝑹𝒆) => 𝑪𝑭 = 𝝃(𝑹𝒆)
𝝆𝑽𝟐 𝑫𝟐

Ex:A spherical object, D=1mm, moving through stagnant water. Determine FD. Scale model 100 times
larger tested in glycerin (μg=1.5kg/m.s; ρg=1263kg/m) at 30 cm/s and measured drag force =1.3N
𝝆𝑽𝑫 𝟏𝟐𝟔𝟑∗𝟎.𝟑∗𝟎.𝟏
Rem=Rep ; => 𝑹𝒆𝒎 = = = 𝟐𝟓. 𝟑 = 𝑹𝒆𝒑 ⇒ 𝑽𝒑 = 𝟐. 𝟓𝟑𝒎/𝒔 ;
𝝁 𝟏.𝟓
F 𝟏.𝟑
Cm= Cp ; => 𝑪𝒎 = 𝟐𝑫 𝟐 = = 𝟏. 𝟏𝟒 = 𝑪𝒑
𝝆𝑽 𝑫 𝟏𝟐𝟔𝟑∗𝟎.𝟑𝟐 ∗𝟎.𝟏𝟐 .
F𝑫 F𝑫 −𝟕 𝑵
𝑪𝒑 = = = 𝟏. 𝟏𝟒 ⇒ F𝑫 = 𝟕. 𝟑 ∗ 𝟏𝟎
𝝆𝑽𝟐 𝑫𝟐 𝟏𝟎𝟎𝟎 ∗ 𝟎. 𝟎𝟐𝟓𝟑𝟐 ∗ 𝟎. 𝟎𝟎𝟏𝟐 . 15
Nondimensionalize Navier-Stokes equation
ഥ
𝐷𝑉
NS equation: ρ = 𝜌𝑔ҧ − ∇𝑝 + 𝜇∇2 𝑉ത
𝐷𝑡
Considerations: a) Newtonian, incompressible fluid b) external body force gravity is acting.
variables: u,v,w,x,y,x,t and p
all variables can be nondimensionalized by using density and two reference parameters that
might be characteristic of the fluid flow. Reference velocity us and Ls. Reference time ts can
be represented by Ls /us. We can also consider reference pressure as ps (≡ρu2)
dimensionless forms: u*=u/us; v*=v/us; w*=w/us; x*=x/Ls; y*=y/Ls; z*=z/Ls; t*=t/ts; p*=p/ps;
ഥ
𝑉
so, velocity vector can be nondimensionalized by 𝑉∗ =
𝑢𝑠
𝐷𝑢 𝜕𝑝 𝜕2 𝑢 𝜕2 𝑢 𝜕2 𝑢
Let’s start with x-component N-S equation: ρ = − + 𝜌𝑔𝑥 +𝜇 + + 2
𝐷𝑡 𝜕𝑥 𝜕𝑥 2 𝜕𝑦 2 𝜕𝑧
𝑢𝑠2 𝜕𝑢∗ ∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗ 𝑝𝑠 𝜕𝑝∗ 𝑢𝑠 𝜕 2 𝑢∗ 𝜕 2 𝑢∗ 𝜕 2 𝑢∗
⇒ρ + 𝑢 + 𝑣 + 𝑤 =− + 𝜌𝑔𝑥 + 𝜇 2 2 + ∗2 + ∗2
𝐿𝑠 𝜕𝑡 ∗ 𝜕𝑥 ∗ 𝜕𝑦 ∗ 𝜕𝑧 ∗ 𝐿𝑠 𝜕𝑥 ∗ 𝐿𝑠 𝜕𝑥 ∗ 𝜕𝑦 𝜕𝑧

16
Nondimensionalize Navier-Stokes equation
𝑢𝑠2 𝜕𝑢∗ ∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗ 𝑝𝑠 𝜕𝑝∗ 𝑢𝑠 𝜕 2 𝑢∗ 𝜕 2 𝑢∗ 𝜕 2 𝑢∗
⇒ρ + 𝑢 + 𝑣 + 𝑤 =− + 𝜌𝑔𝑥 + 𝜇 2 2 + ∗2 + ∗2
𝐿𝑠 𝜕𝑡 ∗ 𝜕𝑥 ∗ 𝜕𝑦 ∗ 𝜕𝑧 ∗ 𝐿𝑠 𝜕𝑥 ∗ 𝐿𝑠 𝜕𝑥 ∗ 𝜕𝑦 𝜕𝑧
𝑢𝑠
now, divide both side with coefficient of the highest order term: 𝜇 2
𝐿𝑠
𝜌𝑢𝑠 𝐿𝑠 𝜕𝑢∗ ∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗ 𝑝𝑠 𝐿𝑠 𝜕𝑝 ∗ 𝜌𝑔 𝑥 𝐿2
𝑠 𝜕 2 𝑢∗ 𝜕 2 𝑢∗ 𝜕 2 𝑢∗
⇒ ∗ + 𝑢 ∗ + 𝑣 ∗ + 𝑤 ∗ =− ∗ + + 2 + ∗2 + ∗2
𝜇 𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜇𝑢𝑠 𝜕𝑥 𝜇𝑢𝑠 𝜕𝑥 ∗ 𝜕𝑦 𝜕𝑧
𝜌𝑢𝑠 𝐿𝑠
𝑅𝑒 =
𝜇
𝜕𝑢∗ ∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗ 𝑝𝑠 𝜕𝑝 ∗ 𝑔𝑥 𝐿 𝑠 𝜇 𝜕 2 𝑢∗ 𝜕 2 𝑢∗ 𝜕 2 𝑢∗
⇒ ∗ + 𝑢 ∗ + 𝑣 ∗ + 𝑤 ∗ =− 2 ∗+ 2 + 2 + ∗2 + ∗2
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜌𝑢𝑠 𝜕𝑥 𝑢𝑠 𝜌𝑢𝑠 𝐿𝑠 𝜕𝑥 ∗ 𝜕𝑦 𝜕𝑧
𝑝𝑠 𝑢𝑠 𝜌𝑢𝑠 𝐿𝑠
𝐸𝑢 = 2 ; 𝐹𝑟 = ; 𝑅𝑒 =
𝜌𝑢𝑠 𝑔𝐿𝑠 𝜇
𝜕𝑢∗ ∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗
∗
𝜕𝑢 ∗
𝜕𝑝 ∗
1 1 𝜕 2 𝑢∗
𝜕 2 𝑢∗
𝜕 2 𝑢∗
⇒ ∗ + 𝑢 ∗ + 𝑣 ∗ + 𝑤 ∗ = −𝐸𝑢 ∗ + 2 + 2 + ∗2 + ∗2
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜕𝑥 𝐹𝑟 𝑅𝑒 𝜕𝑥 ∗ 𝜕𝑦 𝜕𝑧

17
19