0% found this document useful (0 votes)
78 views81 pages

NS Equation

The document discusses transport phenomena and provides information about the course, including: 1) The course deals with the unified treatment of momentum, heat, and mass transfer processes which are important in industry and nature. 2) It lists recommended textbooks and notes the course will include class tests graded on a relative scale. 3) The course plan outlines topics like the Navier-Stokes equations, boundary layer concepts, and the formulation and solution of momentum, heat, and mass transfer in laminar flow.

Uploaded by

Ritik Chaturvedi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
0% found this document useful (0 votes)
78 views81 pages

NS Equation

The document discusses transport phenomena and provides information about the course, including: 1) The course deals with the unified treatment of momentum, heat, and mass transfer processes which are important in industry and nature. 2) It lists recommended textbooks and notes the course will include class tests graded on a relative scale. 3) The course plan outlines topics like the Navier-Stokes equations, boundary layer concepts, and the formulation and solution of momentum, heat, and mass transfer in laminar flow.

Uploaded by

Ritik Chaturvedi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 81

TRANSPORT PHENOMENA

Prof. Sunando DasGupta (3 classes)


and
Prof. Manish Kaushal (1 class)
TRANSPORT PHENOMENA
Basic Overview
Over the last few decades, the subject has revolutionized the way
Chemical Engineering science is taught. This course deals with the
unified treatment of the different transport processes, ubiquitous in
industry as well as in nature. Momentum, heat and mass transfer are
taught together due to the underlying similarities of the mathematics and
molecular mechanisms describing such processes.

Books

1. Transport Phenomena by Bird, Stewart and Lightfoot (Wiley)


2. Transport Phenomena Fundamentals by J. L. Plawsky (CRC Press)
3. Heat and Mass Transfer – A Transport Phenomena Approach by K. S. Gandhi (New Age)
4. Fluid Mechanics by R. W. Fox, P. J. Pritchard and A. T. McDonald (Wiley)
5. Introduction to Heat and Mass Transfer by F. P. Incropera & D. P. DeWitt (Wiley)

Grading: 2-3 class tests for TA, relative grading


Course Plan
Formulation and solution of momentum transfer in laminar flow.
Navier-Stokes equation and its applications
Boundary Layer concepts, boundary layer thicknesses (disturbance, displacement
and momentum), Blasius solution for flow over a flat plate
Use of momentum integral equation, turbulent boundary layers, fluid flow about
immersed bodies, drag
Formulation and solution of heat transfer in laminar flow
Development and use of energy equation
Transient conduction - lumped capacitance, analytical solutions and other methods.
Formulation and solution of mass transfer in laminar flow. Development and use of
species balance equation
Introduction to convective flow, natural convection, relevant examples from heat
and mass transfer
Mathematical treatment of the similarities between heat, mass and momentum
transfer, similarity parameters, and relevant analogies.
Solution of coupled heat, mass and momentum transfer problems based on analogy.
Phenomenological Relations

Governing Equations

Re Pr Sc

f Nu Sh

f = f (Re) Nu=f (???) Sh = f (???)


TRANSPORT PHENOMENA

Application of Navier Stokes Equations


Continuity Equation Equation of Motion –
Navier Stokes equation
𝜕𝑣𝑥 𝜕𝑣𝑦 𝜕𝑣𝑧
𝜕𝑥
+
𝜕𝑦
+
𝜕𝑧
=0
 Dv / Dt = −p − [. ] +  g
Cartesian Coordinate – z component

𝜕 𝑣𝑧 𝜕 𝑣𝑧 𝜕 𝑣𝑧 𝜕 𝑣𝑧
𝜌 + 𝑣𝑥 + 𝑣𝑦 + 𝑣𝑧
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧
𝜕𝑃 𝜕 2 𝑣𝑧 𝜕 2 𝑣𝑧 𝜕 2 𝑣𝑧
= − + 𝜇 2
+ 2
+ 2
+ 𝜌 𝑔𝑧
𝜕𝑧 𝜕𝑥 𝜕𝑦 𝜕𝑧

Relevant Boundary Conditions

No slip at the liquid-solid interface

No shear at the liquid-vapor interface


6
The Equation of Continuity
The Equation of Motion for a Newtonian Fluid with constant μ, ρ

The Navier Stokes Equation


Transport Phenomena - Momentum Transfer

The record-write head for a computer disk storage system floats


above the spinning disk on a very thin film of air (the film
thickness is 0.5micron). The head location is 150mm from the
disk centre line, the disk spins at 3600rpm. The record-write
head is 10mm x 10mm square. Determine for standard air
(density=1.23kg/m3, viscosity=1.78x10-5 kg/(m.s)) in the gap
between the head and the disk -
(a) the Reynold's number of the flow, (b) the viscous shear stress
and (c) the power required to overcome viscous shear stress.

For such a small gap the flow can be considered as flow


between parallel plates.

10
Flow between parallel plates

U
V dP/dx < 0 V

Couette Flow Pressure gradient Driven Flow

U
dP/dx < 0 V

Couette and Pressure gradient Driven Flow


11
Solution

Since the gap is too small, the situation can be approximated by the
case of flow between two parallel plates with one plate moving and
the other stationary with no applied pressure gradient.

Here,

V=Rω = 0.15 m x 3600 rev/min x 2 π rad/rev x 1/60 min/s=56.5 m/s

Re = (ρ V a)/µ = (V a)/(μ/ρ) = 56.5 m/s x 0.5 x 10 -6 m/(1.45 x 10 -5


m2 /s) = 1.95 (Laminar Flow)

For Couette flow between two parallel plates, with no applied


pressure gradient, the velocity profile will be linear with no-slip
boundary conditions at the bottom (stationary) and top (moving)
plates.
12
Therefore, the shear stress, τ, can be expressed as

τ = μdu/dy = μ V/a (linear velocity profile)

τ=1.78 x 10 – 5 kg /(m.s) x 56.5 m/s x 1/(0.5x10 – 6 m) = 2.01 KN/m2

Therefore,

Force, F = τ A = τ x (W x L), Torque, T = F x R = τ x (WL) x R

Power dissipation rate. P = T x ω = τ WL R ω

P = 2.01x10- 3 x 0.01x 0.01 x 0.15 x 3600 x 2π x 1/60 = 11.4 W

Thus the power required to overcome viscous shear stress is 11.4 W

13
Transport Phenomena - Momentum Transfer
Problem 2

An oil skimmer uses a 5 m


wide x 6 m long moving
belt above a fixed platform
( = 30º) to skim oil off of
rivers (T = 10ºC). The belt
travels at 3 m/s. The
distance between the belt
and the fixed platform is 2
mm.
The belt discharges into an open tank on the ship. The fluid is actually a
mixture of oil and water. To simplify the analysis, assume crude oil
dominates. Find the discharge of oil into the tank on the ship, the force
acting on the belt and the power required (kW) to move the belt. For oil:
 = 860 kg/m3, viscosity,  = 1x10 – 2N.s/m2 14
The x-component of the equation of motion is given as
𝜕𝑢 𝜕𝑢 𝜕𝑣 𝜕𝑤
𝜌 + 𝑢 + 𝑣 + 𝑤
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧
𝜕𝑃 𝜕2 𝑢 𝜕2 𝑢 𝜕2 𝑢
= − + 𝜇 2
+ 2
+ 2
+ 𝜌 𝑔𝑥
𝜕𝑥 𝜕𝑥 𝜕𝑦 𝜕𝑧

15
Considering one-dimensional flow (v = 0 = w), no
applied pressure gradient, u is a function of only y, and a
steady state process, the above equation reduces to the
following governing equation

𝑑2𝑢 𝜌𝑔
= sinθ
𝑑𝑦 2 𝜇

The relevant boundary conditions (no slip) are

BC 1 At y = 0 u = 0
BC 2 At y = h, u = U
Therefore,

𝑑𝑢 𝜌𝑔𝑠𝑖𝑛𝜃 𝜌𝑔𝑠𝑖𝑛𝜃 𝑦 2
= 𝑦+𝐴 𝑎𝑛𝑑 𝑢= + 𝐴𝑦 + 𝐵
𝑑𝑦 𝜇 𝜇 2
𝜌𝑔𝑠𝑖𝑛𝜃 ℎ 𝑢
Applying the BC s ➔ B = 0 and 𝐴 = +
𝜇 2 ℎ

Thus

𝜌𝑔𝑠𝑖𝑛𝜃 ℎ𝑦 𝑦2 𝑈𝑦
𝑢=− − +
𝜇 2 2 ℎ

The volumetric flow rate per unit width of the film is given by

ℎ ℎ
𝜌𝑔𝑠𝑖𝑛𝜃 ℎ𝑦 𝑦2 𝑈𝑦 𝜌𝑔ℎ3 𝑈ℎ
𝑄 = න 𝑢𝑑𝑦 = − න − + 𝑑𝑦 = − 𝑠𝑖𝑛𝜃 +
𝜇 2 2 ℎ 12𝜇 2
0 0

17
860 𝑥 9,81 𝑥 0.002 3 𝑥 0.002 𝑚2
𝑄=− sin 30 + = 0.0027 (𝑝𝑒𝑟 𝑢𝑛𝑖𝑡 𝑤𝑖𝑑𝑡ℎ)
12 𝑥 10−2 2 𝑠

For the width of 5 m, the volumetric flow rate will be equal to 0.0135 m3/s.

Evaluate τ = μ du/dy at the moving belt from the expression for velocity as
𝑑𝑢 𝜌𝑔𝑠𝑖𝑛𝜃 ℎ 𝑈
=− −𝑦 +
𝑑𝑦 𝜇 2 ℎ
And at the moving belt,
𝑑𝑢 𝜌𝑔𝑠𝑖𝑛𝜃 𝜇𝑈
𝜏= 𝜇 𝑦 = ℎ = ℎ+
𝑑𝑦 2 ℎ

Using the values h = 0.002m, μ = 10-2N.s/m2 and U = 3 m/s

τ│at the belt = 19.21 N/m2

Power = (τ x L x W) U = 19.21 x 6 x 5 x 3 = 1.73 KW


18
Transport Phenomena - Momentum Transfer
Problem 3
Consider two concentric
cylinders with a Newtonian
liquid of constant density,
ρ, and viscosity, μ,
contained between them.
The outer pipe, with radius,
Ro, is fixed while the inner
pipe, with radius, Ri, and
mass per unit length, m,
falls under the action of

gravity at a constant speed. There is no pressure gradient within the flow and no
swirl velocity component. Determine the vertical speed, V, of the inner cylinder as a
function of the following parameters: g, Ro, Ri, m, ρ, and μ. The space between the
two cylinders is not ‘too small’ compared to the radii of the cylinders
19
Solution

For the inner cylinder moving at constant velocity, the downward force
is exactly balanced by the viscous force as

𝜏𝑤 𝐴𝑤 │𝐼𝑛𝑛𝑒𝑟 𝐶𝑦𝑙𝑖𝑛𝑑𝑒𝑟 = 𝑚 𝐿 𝑔

The z component of the equation of motion in cylindrical coordinate is

𝜕𝑣𝑧 𝜕𝑣𝑧 𝑣𝜃 𝜕𝑣𝑧 𝜕𝑣𝑧


𝜌 + 𝑣𝑟 + + 𝑣𝑧
𝜕𝑡 𝜕𝑟 𝑟 𝜕𝜃 𝜕𝑧
𝜕𝑝 1𝜕 𝜕𝑣𝑧 1 𝜕 2 𝑣𝑧 𝜕 2 𝑣𝑧
=− + µ 𝑟 + 2 + + 𝜌𝑔𝑧
𝜕𝑧 𝑟 𝜕𝑟 𝜕𝑟 𝑟 𝜕𝜃 2 𝜕𝑧 2

Cancelling the terms with the observations i) steady state, ii) vz is a


function of r only, not of z or θ, iii) no applied pressure gradient

The following simplified form of the NS equation can be obtained.


20
Governing equation

1𝑑 𝑑𝑣𝑧 𝜌𝑔
𝑟 =−
𝑟 𝑑𝑟 𝑑𝑟 𝜇

Upon integration (with C1 and C2 being the constants of integration)


𝜌𝑔𝑟 2
𝑣𝑧 = − + 𝐶1 ln 𝑟 + 𝐶2
4𝜇
Boundary Conditions
𝜌𝑔
vz = V at r = Ri => 𝑉=− 𝑅𝑖 2 + 𝐶1 ln 𝑅𝑖 + 𝐶2
4𝜇

𝜌𝑔
vz = 0 at r = Ro => 0=− 𝑅𝑜2 + 𝐶1 ln 𝑅𝑜 + 𝐶2
4𝜇

21
𝜌𝑔 2 2 𝑅𝑖
Therefore 𝑉= 𝑅𝑜 − 𝑅𝑖 + 𝐶1 ln
4𝜇 𝑅𝑜

1 𝜌𝑔
𝐶1 = 𝑉 − 𝑅𝑜2 − 𝑅𝑖 2
𝑅𝑖 4𝜇
ln
𝑅𝑜
𝑑𝑣𝑧 𝜌𝑔𝑟 𝐶1 𝑑𝑣𝑧 𝜌𝑔𝑟 𝐶1 𝜇
=− + 𝜏𝑟𝑧 = 𝜇 =− +
𝑑𝑟 2𝜇 𝑟 𝑑𝑟 2 𝑟
Since the force on the inner cylinder = force due to gravity

𝜏│𝑟=𝑅𝑖 2𝜋𝑅𝑖 𝐿 = 𝑚 𝐿 𝑔

Upon substitution for the expression of the shear stress

𝑅𝑖 𝜌𝑔𝑅𝑖 𝑚𝑔 𝜌𝑔
𝑉 = 𝑅𝑖 ln − − 𝑅𝑖 2 − 𝑅0 2
𝑅𝑜 2𝜇 2𝜋𝑅𝑖 𝜇 4𝜇
22
Order of magnitude analysis of NS Equation

Is it possible to identify the relative magnitudes of the different terms


(even approximately)?

It may then be possible to neglect the term(s) that may not play a
crucial role in the transport process thereby simplifying NS equations.
External Incompressible Viscous Flow – Boundary Layer

Freestream Velocity, U∞
Approach y Boundary Layer
Velocity Inviscid
Thickness, δ
V Flow
Viscous
x vx = f (x, y)

Flow over a flat plate

vx = f (x, y) Viscous 2D flow inside BL vx = 0.99U∞


vx = U∞ at y = δ
Inviscid flow outside BL

δ - boundary layer thickness


Flow Inside the Boundary Layer

 vx  vx   2 vx  2 vx 
vx + vy = 2 + 2 
x y  x y 

Boundary Layer Approximations

 vx  vx  2 vx  2 vx
vx vy
y x  y2  x2

 vx  vx  2 vx
vx + vy =
x y  y2
Flow between two parallel disks with liquid entry through
a small hole at the centre of the top plate

Flow In
Disk radius = r2
Hole radius = r1
Z=+b

Z=-b

Flow Out
The pressure is p1 at r = r1 and p2 at r = r2
 1  1  
+ (  r vr ) + (  v ) + (  vz ) = 0
t r r r  z

 vr vr v vr v 2 vr 


 + vr + − + vz 
  t r r  r z 

P   1   1  vr 2 v  vr 
2 2
=− +  ( rvr )  + 2 2 − 2 + 2  +  g r
r  r  r r  r  r  z 
The governing equation

 vr P  2 vr  ( z)
 vr =− + vr =
r r  z2 r

Major assumption on the nature of the flow


Assume a constant applied pressure difference
Boundary Conditions

P b
2  z 
2

vr ( r , z ) = 1 −   
r2   b  
2  r ln  
r1
+b +b
4   P b3
Q = 2   r vr dz = 2    ( z ) dz =
r2
−b −b 3  ln
r1
An incompressible fluid flows between two porous, parallel flat plates
as shown in the figure. An identical fluid is injected at a constant
speed V through the bottom plate and simultaneously extracted from
the upper plate at the same velocity. Assume the flow to be steady,
fully-developed, the pressure gradient in the x-direction is a constant,
and neglect body forces.

Determine expressions for the y component of velocity.


Show that the x component of velocity can be expressed as
𝜌𝑉𝑦
ℎ 𝜕𝑝 1 − exp 𝑦
𝜇
𝑢𝑥 = −
𝜌𝑉 𝜕𝑥 𝜌𝑉ℎ ℎ
1 − exp 𝜇
The equation of continuity for fully developed, steady flow in x-direction

𝜕𝑣𝑥 𝜕𝑣𝑦 𝜕𝑣𝑧


+ + =0
𝜕𝑥 𝜕𝑦 𝜕𝑧

The x component of the NS equation

𝜕 𝑣𝑥 𝜕 𝑣𝑥 𝜕 𝑣𝑥 𝜕 𝑣𝑥
𝜌 + 𝑣𝑥 + 𝑣𝑦 + 𝑣𝑧
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧
𝜕𝑃 𝜕 2 𝑣𝑥 𝜕 2 𝑣𝑥 𝜕 2 𝑥
= − + 𝜇 2
+ 2
+ 2
+ 𝜌 𝑔𝑥
𝜕𝑥 𝜕𝑥 𝜕𝑦 𝜕𝑧
The governing equation

 vx P  2 vx
 vy =− +
y x  y2

Assumptions:

Constant pressure gradient in the x direction

Boundary Conditions
A thrust bearing as shown in the figure is lubricated by pumping oil at a high
pressure of p0. The angular velocity is equal to ω. Note that under laminar
conditions, both Vr and Vθ in the thin gap will be non-zero and p is a function of
r only. Neglect convective and body force terms in equations of motion. You
may also assume that the pressure at r = R is equal to patm, whereas the pressure
from r = 0 till r = R0 is equal to p0.

(i) Start with the equation of continuity to obtain the functional form of Vr.

(ii) Show that V =  r 1 + z  satisfies the θ component of the NS equation.


  
2  h
(iii) Write the r component of NS equation to show that r  p is a constant.
r
(iv) Evaluate the pressure distribution using the boundary conditions

(v) Find the vertical load the bearing can support and the flow rate of oil
required.
(i) Start with the equation of continuity to obtain the functional form of Vr.

The Equation of Continuity


(ii)

=
Show that V
+

1


r

2
z
satisfies the θ component of the NS equation.

h
p
(iii) Write the r component of NS equation to show that r is a constant
r
 p
r
 r
(iv) Evaluate the pressure distribution using the boundary conditions
(v) Find the vertical load the bearing can support and the flow rate of
oil required.
Consider the spin-coating process used to coat silicon wafers with photoresist. The
process is designed to produce a very thin, uniform coating by spinning a viscous,
Newtonian, liquid onto a substrate (wafer). The process has angular symmetry, the
rotation rate is constant, and since the film is thin, there are no real pressure gradients or
fluid accelerations and body forces to speak of. The thin film also moves with the
substrate as if it were a rigid body, v  f(z)

a) Show that the continuity and momentum equations reduce to:


v2
− =
d 2 vr
;
1
( r vr ) +  v z = 0
r dz 2
r r z

b) What are the boundary conditions for this problem?


c) Solve the equations for vr and vz.
d) The velocity, vz, at the film/air interface is just the change in film thickness
with time. Use this to obtain a differential equation for h, and integrate this
equation to obtain the solution as:

1  1 1   2 2 
 2 + 2  =   t
2  h ho   3 
Where ω is the rate of rotation (vθ = 2πrω) and h0 is the initial height of the film.
In this case, the following conditions will hold:

v = Cons tan t , g r = g = g z = 0,  ,  constant


  P P P
= 0, = 0, = = =0
 t r  z

The continuity and momentum equations in cylindrical coordinates become

1   vz
( r)
r v + =0
r r z
Since the fluid moves essentially as a rigid body, vr = f (r, z). This film is
very thin with respect to the radial or angular dimensions so that
v , vr vz

Thus since the angular velocity is constant and vz is small, we only need to deal
with the momentum equation for vr
Neglecting all the obvious terms,

  vr v2  vr    1    2 vr 
  vr − + vz  =   ( r vr )  + 2 
  r r  z   r  r r  z 
 
     r   

2 2
1 v v v v
 ( ) + = − + 
r r
r
v  v v
 r    rr 
r 2 r z

rr z 
z

Since the film is very thin, vr changes rapidly with the film thickness, much
more rapidly than it changes with the radial position. vz is quite small as
compared to the other components

Thus 1st term on LHS is neglected w.r.t. the second term.

For the same reason, 1st and 3rd terms on the rhs can be neglected.
 vr
In one case, vz is very small and in the second  r is also very small.

vθ is quite large
Thus
d 2 vr  v2
 2 + =0
dz r
The boundary conditions are

z = 0; vr = 0 The coordinate system is such that z = 0


 vr refers to the bottom plate, while z = h is the
z = h; =0 free surface of the photoresist
z

Noting that v = 2 r , the above equation is solved


4  2 2
 z2 
vr = r hz − 
  2 

Using the continuity equation

1  vz
( r)
r v + =0
r r z

4  2 2  z 3 2 
vz =  −hz 
 3 

BC: vz = 0 at z = 0
1
h
 2 2 h3 d h
vz =  vz dz = − =
h0  dt

At t = 0 h = h0
1 1 2 2  2
− 2 = t
h 2
h0 

One of the possible answers

This can also be solved by assuming vz at z = h to be equal to dh/dt


Supplementary Information – in relation to the query about spreading of a droplet on a solid substrate
and contact line motion – slip velocity

Wetting: Partial or Total Wetting

A liquid spreads on a substrate in a film if the energy of the system is lowered by the
presence of the liquid film .

The surface energy per unit surface of the dry solid surface is γSG; the surface energy of the
wetted solid is γSL+γLG.

The spreading parameter S determines the type of spreading (total or partial)

S = γSG − (γSL + γLG)

If S > 0, the liquid spreads on the solid surface; if S < 0 the liquid forms a droplet.

54
Problem: One disk approaching another displacing a liquid in between

Two parallel, plane circular disks (of radius R) lie one above the other a
small distance apart. The space between them is filled with a liquid. The
upper disk approaches the lower at a constant (small) velocity U,
displacing the liquid. The imposed pressure at r = R is po and pr is not a
function of z. Simplify the basic equations (continuity and motion, r
component) using an analysis based on your understanding and estimate
of the magnitudes of the different terms in the original equations.

Solve the these equations with proper boundary conditions to evaluate a


expression of the resistance to motion on the moving disk in terms of the
separation between the two plates. Take the origin to be at the centre of the
lower (stationary) plate.
Assumptions and special considerations

1. R is very large compared to h

2. Leakage rate is small (Small vr)

3. ∂p/∂z is negligible

4. vz is very small as compared to vr

5. ∂vz/∂z may not be small (as z is very small)

𝜕 𝜕𝑣𝑧
6. ( )
𝜕𝑧 𝜕𝑧
can be appreciable

7. vr is not small as compared to vz

𝜕𝑣𝑟 𝜕𝑣𝑟
8. 𝜕𝑟
is small compared to 𝜕𝑧

9. Vθ = 0, no θ dependence of vr , vz
The equation of continuity

 1  1  v
+ ( r vr ) + (  v  ) + z = 0 ( A)
t r  r r  z

v r  vr  vz
+ + =0
r r z

The terms in the equation

Relative magnitudes of 1,2 w.r.t. 3


The equation of motion (r component)

1 p  1  vr vr  2 vr  2vr 
0=− +  − 2 + 2 + 2
 r  r r r r z 
1 p  1  vr vr  2 vr  2vr 
0=− +  − 2 + 2 + 2
 r  r r r r z 

Compare terms to arrive at the following equation

1 p  2 vr
0=− + ( B)
 r  z2

1 P 2
vr = z + C1 z + C2
2  r
BC : z = 0; vr = 0  C2 = 0
1 dp
z = h; vr = 0  C1 = − h
2 dr
1 P
vr = ( z − h ) z (C )
2  r

1   vz
( r)
r v + =0 ( A)
r r z

Putting the expression of vr from (C ) to (A) and integrating w.r.t. z and as

BC : z = 0; vr = 0 = vz
z = h; vr = 0; vz = − U

1 d  dp 1 h3 1 d  d p 
h h
1 d
U=  r vr dz = r   z ( z − h ) dz = − r 
r dr 0 r dr  dr  2 0 12  r d r  d r 
h3 1 d  d p 
U =− r 
12  r d r  d r 
Integrating w.r.t. r
12  r 2 dp
− 3 =r + C1
h 2 dr

As dp/dr is finite at r = 0 C1= 0 and p = po at r = r

3 U 2
p − po = 3 ( R − r 2 )
h

3 U 2
R R

3 ( )
Force, F =  2 rdr ( p − po ) = 2  R r − r 3
dr
0 0
h

3   U R4
Force, F =
2 h3
A fluid (of constant density ρ) is in
incompressible, laminar flow through a
tube of length L. The radius of the tube
of circular cross section changes linearly
from R0 at the tube entrance (z = 0) to a
slightly smaller value RL at the tube exit
(z = L).

Using the lubrication approximation,


determine the mass flow rate vs. pressure
drop (ΔP) relationship for a Newtonian
fluid (of constant viscosity μ).

The approximation where a flow


between non-parallel surfaces is treated
locally as a flow between parallel
surfaces is commonly called the
lubrication approximation because it is
often employed in the theory of
lubrication.
The mass flow rate versus the pressure drop (W versus ΔP) relationship for a Newtonian
fluid in a circular tube is given by (where ΔP = P0 – PL)

  P R4  P 8  W
W= or = (1)
8 L L   R4

Even for a tapered tue, W does not change with Z

The above equation is assumed to be approximately valid for a differential


length dZ of the tube, whose radius R is slowly changing with Z.

Approximation that a flow between non-parallel surfaces can be treated


locally as a flow between parallel surfaces is know as LUBRICATION
APPROXIMATION.

d P 8W 1
− = (2)
dZ   [ R ( Z ) ]4
d P 8W 1
− = (2)
dZ   [ R ( Z ) ]4
Z
R ( Z ) = Ro + ( RL − Ro )
L
d R RL − Ro
= (3)
dZ L

From Eqns 2 and 3

8W L dR
−d P =
  RL − Ro R 4

Integrating the above equation between Z = 0 (R = Ro) and Z = L (R = RL)

8W
PL RL
L dR
P − d P =
  RL − Ro 
Ro
R4
o
8W
PL RL
L dR
P − d P =
  RL − Ro 
Ro
R4
o

Po − PL 8  W L  1 1 
=  3 − 3
L   3 ( RL − Ro )  Ro RL 

Thus
  P R o4   3 (  − 1)  RL
W=  −3 
;  =
8 L  1 −  
Ro

Taper correction to Hagen-Poiseuille equation


A cell separation (fractionation) system is based on cell density. Cells are
injected at the centre of a tube of radius R, and are carried by fluid flowing at
a flow rate of Q. Dense cells fall quickly under the action of gravity, adhere
to the tube wall and hence do not pass out of the tube. Assume that the
concentration of the cell is low enough that the laminar flow in the tube is
not perturbed by the presence of the cells. Let the cells be spherical, with
radius ‘a’ and let them have a density ρ + Δρ, where ρ is the density of the
flowing fluid.
Find the axial distance (L) that a cell travels
before it hits the bottom wall. You need to use
the fact that, in fully developed flow in the
presence of gravity, the pressure distribution
in the vertical distance is hydrostatic and the
axial velocity profile is the familiar parabolic
shape. You may assume that the cell is
spherical, that it reaches its terminal (falling)
velocity nearly immediately after injection
into the tube, and that it gets carried axially at
the local fluid velocity in the tube. State any
other assumptions you make.
We need to find the axial distance L that the cell travels before it hits the bottom wall.

The expressions of the forces acting on the spherical cell are:

4
Gravitational force F g = m cell g =  a 3 (  +   ) g
3

4
Buoyancy force F b = m fluid g =  a 3  g
3

The cell is moving downward with a


Drag force F d = 6   a vr velocity vr in the r direction

4 4
 a (  +   ) g =  a 3  g + 6   a vr
3

3 3
2 a2 g  
vr =
9
Find the radial location as a function of time as

2 a2 g   t
t
r ( t ) =  vr dt =
0
9

The time T, when the cell will hit the wall

R 9 R
T= = 2
vr 2 a g  

 P R2   r 2 
vx = 1 −   
4 L   R  
P R 2 P  R 4
vx = ; Q =  R 2 vx =
8 L 8 L
2 Q P R 2
=
R 2
4 L
2 a2 g   t
r (t ) =
2 Q   r (t )   9
2

vx = 1 −   
R   R  
2
 

2 Q   r (t )  
T T 2

L =  vx dt = 2 
1−  dt
0
 R 0 
R 2

2Q   2 a2 g    T 3 
2

L= T −   
 R 
2
 9  R  3 

9 R
As T = 2
2a g 
6Q
L=
 R a2 g  
Epithelium is one basic type of animal tissue, which lines the cavities and surfaces
of structures throughout the body. Epithelial layers contain no blood vessels, so they
must receive nourishment via diffusion of substances from the underlying
connective tissue, through the basement membrane.

An apparatus has been built for testing the effect of various drugs on the rate at
which an epithelium can pump fluid from its luminal side ( the side facing the fluid )
to its basal side (which lies on the channel wall). The cells line the top and bottom
surface of a flow channel that has a separation of h (from top plate to bottom plate;
ignore the thickness of cells), a length L, and a depth into the page of W. Each of
these walls is porous so that any fluid pumped by the cells can leave the channel. Let
each cell layer (top and bottom) pump fluid at a rate of q per unit area of the channel
walls (q has thus units of length/time). The height of the channel is much less than its
length (h≪L).
Fluid enters the channel at the left at a flow rate Q0 and a gauge pressure of P0. Because
of the pumping action of the cells, the flow rate through the channel decreases as a
function of x, the distance from the beginning of the channel. To determine the rate at
which the cells are pumping fluid out of channel, the channel is instrumented with
pressure transducers that can measure P(x). We would like to use this information to find
the rate at which the cells pump fluid. The fluid in the channel has a density of ρ and a
viscosity of μ. The flow is dominated by viscous effects and is steady.

(a) Find the pressure distribution, P(x) , in the flow channel if q=0.

(b) Find the pressure distribution P(x) in the channel for q≠0.

(c) Given that P(x=L)=Pe , find q .

(d) Find the criterion necessary for the assumption that viscous flow dominates to be valid.

All answers must be given in terms of the known quantities e.g., x,L,W,h,Q0 , P0, Pe, ρ and
μ (not all of these parameters need necessarily be used).
(a) Find the pressure distribution, P(x) , in the flow channel if q=0

For q = 0, the situation is simply pressure driven flow between parallel plates.

d 2 vx dP
 2
− =0 At y = 0, vx = 0  C2 = 0
dy dx
1 dP
1 dP 2 At y = h, vx = 0  C1 = h
vx = y + C1 y + C2 2 dx
2  dx
1 dP 2  h 
vx = y 1 − 
2  dx  y

h
1 1 dP  y 3 h y 2 
h
1 1 dP 2
vx =  vx dy =  − =−
2 
h
h0 h 2  dx  3 0
12 dx
1 dP 3
Q0 = W h vx = − hW
12 dx Qo is the
dP 12  flow rate
=− 3
Qo with q = 0
dx Wh

Integrating and imposing the B.C. that P ( x = 0) = Po

12  Qo
P ( x ) = Po − 3
x
Wh
Find the pressure distribution P(x) in the channel for q≠0.

The assumption of viscous flow allows all inertial terms


in the 2D NS Eqn to be neglected

So long the trans-cellular velocity at the channel wall is small compared to the axial
velocity, i.e., q << Qo/(Wh), the flow Can be treated as nearly Poiseuille Flow and
There will still be no slip at the channel wall but now with a flow rate that depends on x
Given that P(x=L)=Pe , find q .
(d) Find the criterion necessary for the assumption that viscous flow dominates to be valid.

The inertial terms must be quite small as compared to the viscous terms
The Equation of Continuity
The Equation of Motion for a Newtonian Fluid with constant μ, ρ

The Navier Stokes Equation

You might also like