Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Nithin K Rajendran July-November

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

CL 613

Nithin K Rajendran

July-November
Lecture 3 (02-08-2021)

It should be possible to simplify the above expression further by the use of the divergence
theorem

− → − →
−
I Z
F · dA = ∇ · F dV

Note that if the control volume moves, →



v is the relative velocity of the control volume
and the extensive quantity inside the control volume. We now derive the governing
equations explicitly from the RTT.

Mass
The total mass of the system remains a constant. This implies

dM
=0
dt

system

We assume that the control volume is finite, non-deformable and stationary. So, from
RTT, we have
Z I
∂ρ
0= dV + ρ(→

v .n̂)dA
∂t cs
Z  
∂ρ →

=⇒ + ∇ · (ρ v ) dV = 0
∂t
∂ρ
=⇒ + ∇ · (ρ→−v)=0
∂t


The last equation is the Equation of Continuity (replacing ρ→ −v with J gives the
more familiar equation from electrodynamics). Fully developed flow refers to the
situation where the density is a constant throughout the fluid. This implies that

∇·→

v =0

1
Momentum
In the case of momentum, the extensive quantity is momentum itself while the intensive
quantity is velocity
N −→ m→ −
v ; n −→ → −
v
Before moving forward, we define the substantial derivative. This is different from a
simple time derivative. With the substantial derivative, we will be following the motion
of a single infinitesimal element of fluid. Suppose we have a fluid with


v = vx (x, y, z, t)x̂ + vy (x, y, z, t)ŷ + vz (x, y, z, t)ẑ
ρ = ρ(x, y, z, t)

Taking two points 1 and 2, we may expand the density function at point 2 around point
1 using a Taylor Series

ρ1 = ρ(x1 , y1 , z1 , t)
ρ2 = ρ(x2 , y2 , z2 , t)
       
∂ρ ∂ρ ∂ρ ∂ρ
ρ2 = ρ1 + (x2 − x1 ) + (y2 − y1 ) + (z2 − z1 ) + (t2 − t1 )
∂x 1 ∂y 1 ∂z 1 ∂t 1
       
ρ2 − ρ1 Dρ ∂ρ ∂ρ ∂ρ ∂ρ
=⇒ lim = = vx + vy + vz +
t2 →t1 t2 − t1 Dt ∂x 1 ∂y 1 ∂z 1 ∂t 1
The difference between the substantial derivative and the partial derivative with respect
to time is that the substantial derivative is the change in the density of a single volume
element as it moves through space and time changes while the partial derivative is the
variation of density at a fixed point in space. The substantial derivative operator
may be defined as
D ∂
= +→−v ·∇
Dt ∂t
Here, →−
v ·∇ is called the convective derivative and is the change due to motion (flow)
of the fluid through the flow field where the physical properties are varying constantly.

∂t
is called the local derivative and its name is self explanatory.

Now, we may begin the derivation of the governing equation for momentum. New-
ton’s third law states that


F = m→ −
a
which contains three equations of the form

Fi = mai

in Cartesian coordinates. An element of fluid experiences two kinds of forces


1. Body Forces - These forces act on the bulk of the fluid element and include
gravitational or electric forces. These are also called volumetric forces.

2
2. Surface Forces - These forces act on the surface of the fluid. There are two
main categories. The first is the pressure due to the external fluid. Second are
the shear and normal stresses on the surface of the fluid element which are also
caused by the external fluid but by cohesive forces and friction.


Let us denote the body force per unit mass acting on the fluid element by f b . The
total force on an infinitesimal fluid element will be

− →

d F = ρ f b dV

or in one direction,
dFi = ρfbi dV
The stresses are related to the rate of change of strain on the fluid element, rather that
on the strain itself as in the case of solids. The stress along direction j on the plane
perpendicular to direction i on a fluid element is denoted by τij . These are the elements
of the stress tensor which is a second order tensor. Since these depend on the rate
of change of strain, τii is related to the time rate of change of normal strain (change
in volume) while τij is related to the time rate of change of shear strain (deformation
parallel to surface). This means they are also related to the velocity gradients in the
∂vi
fluid. Normal stresses are typically smaller than the shear stresses except when ∂x i
(i = 1,2,3 and x1 = x, x2 = y, x3 = z) is large, such as in shock waves. So, the
stress on a fluid element on the plane faces perpendicular to the three directions may
be represented by the stress tensor and the vector of normal vectors to the surface as
 n̂    
τ1 τ11 τ12 τ13 n̂1
τ2n̂  = τ21 τ22 τ23  n̂2 
τ1n̂ τ31 τ32 τ33 n̂3

Here, the normal directions are nothing but x̂, ŷ, ẑ (we assume the element is cuboidal).
Each of the vectors on the LHS, which describe the total stress on the planes perpen-
dicular to each direction denotes a traction vector. The important thing to notice is
that stress is a tensor, traction vectors are vectors. Using the RTT, we have

d(m→− ∂(m→

I
v ) v ) →

= + v (ρ→−
v · n̂) dA
dt ∂t

cs
system CV
X→ − ∂(ρ→

Z I
v) →

=⇒ F = dV + v (ρ→

v · n̂) dA
CV
∂t cs

Let us take this, one direction at a time


Z I
∂(ρvi )
ρvi (→

X
Fi = dV + v .n̂) dA
CV
∂t cs

3
The LHS may be rewritten in terms of the quantities we defined before

n̂ →

X Z Z
Fi = τi .d A + ρfbi dV
CV
Z  
= (∇ · τin̂ ) + ρfbi dV
CV

In addition to this, we also modify the second term on the RHS to


I Z
ρvi ( v .n̂)dA = ∇ · (ρvi →

− −
v ) dV
cs

Putting all these together, for each direction, we have



∇ · τin̂ + ρfbi = (ρvi ) + ∇ · (ρvi →

v)
∂t
This is called Cauchy’s Momentum Equation. It is also referred to as the mass
flux equation. The first term on the RHS is called local acceleration and the second
term on the RHS is called the convective acceleration.

You might also like