Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
0% found this document useful (0 votes)
54 views

CL336: Advanced Transport Phenomena: Assignment 2

This assignment involves analyzing and scaling several problems involving transport phenomena: 1) Scaling the equations governing the motion of a falling ball to determine when gravity can be treated as a constant. 2) Scaling the diffusion equation for oxygen transport in insects to find a critical size above which surface breathing is infeasible. 3) Scaling the unsteady flow equation in an oscillating pipe to determine limits of quasi-static and plug flow behavior. 4) Scaling the heat diffusion equation between parallel plates to determine when axial conduction and convection can be ignored.

Uploaded by

Likhith
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
54 views

CL336: Advanced Transport Phenomena: Assignment 2

This assignment involves analyzing and scaling several problems involving transport phenomena: 1) Scaling the equations governing the motion of a falling ball to determine when gravity can be treated as a constant. 2) Scaling the diffusion equation for oxygen transport in insects to find a critical size above which surface breathing is infeasible. 3) Scaling the unsteady flow equation in an oscillating pipe to determine limits of quasi-static and plug flow behavior. 4) Scaling the heat diffusion equation between parallel plates to determine when axial conduction and convection can be ignored.

Uploaded by

Likhith
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

CL336 : Advanced Transport Phenomena

Assignment 2 (total points: 31)

This assignment is largely based on Lec 7 and Lec 7b on o(1) scaling.

1. In Lec. 4, we derived Cauchy’s fundamental theorem which expressed the stress vector t(n),
acting on a surface with unit normal n, in terms of the stress Tensor T as t = n · T (row
vector n multiplying the 3 × 3 matrix T). Alternatively, we could have also expressed this
relation as t = Tn (matrix T operating on or multiplying the column vector n). Write out
the 3 × 3 matrix T in terms of its components (Txx , Txy , etc.) for both cases, assuming a
Cartesian coordinate system in which t = {tx , ty , tz } and n = {nx , ny , nz }. (1+3)

2. A ball of mass m, initially at rest, is dropped from a height H above the surface of the Earth.
Taking the mass and radius of the Earth to be M and R, the motion of the vertically falling
ball is governed by Newton’s equation of motion:

d2 y mM G dy
m 2
=− , (0) = 0, y(0) = H,
dt (R + y)2 dt

where y is the distance above the Earth’s surface, and G is Newton’s gravitational constant.
Scale this equation and thereby find the condition under which this equation can be simplified
to d2 y/dt2 = g, where g is the ”near-Earth” acceleration due to gravity g = GM/R2 . Find
the time-scale of the falling motion and thereby estimate the radius of a planet (relative to
the Earth’s) for which the ball’s fall time is reduced by half compared to its value above
Earth (assume the density of the planet is the same as that of the Earth). (2+2)

3. A simple model for the internal oxygen concentration c of a small insect, of volume V , that
breathes through its body surface, of area A, is given by:
dc
V = HA(catm − c) − kV,
dt
where H is the mass transfer coefficient, k is the rate of oxygen consumption due to
metabolism, per unit volume of inset cells, and catm is the concentration of oxygen in the
atmosphere.

(a) Assume a spherically shaped insect for simplicity. Scale the equation and find the
condition on the size of the insect above which breathing through its body surface
becomes infeasible (i.e. when a positive concentration cannot be maintained). (2)
(b) Does the condition derived above change in a fundamental way if the shape of the
insect is non-spherical? Why or why not? (2)
(c) What does this exercise tell you about the plausibility of movies like Starship Troopers,
or characters like Shelob in the Lord of the Rings or Aragog in Harry Potter and the
Chamber of Secrets? (1)

4. Consider a unidirectional (fully developed), but oscillatory, pressure-driven flow in a cylin-


drical tube of radius R. Adopting the cylindrical coordinate system, with the flow aligned
2

with the z axis, we can reduce the Navier-Stokes equations to the following:
  
∂u 1 ∂ ∂u
ρ = G(1 + A sin(ωt)) + µ r
∂t r ∂r ∂r
u = 0 at r = R
u is bounded at r = 0
u(t) = u(t + T ).

Here, G is the constant part of the pressure gradient (and is positive) while GA and ω are
the amplitude and the frequency of the pressure-gradient oscillations (A is non-dimensional).
We are only interested in the periodic behaviour of the flow after transients have died out,
as indicated by the third condition on time.

(a) As the governing equation is linear, we can split it into an equation for the steady com-
ponent of the flow us (r), which is driven by G, and an equation for the oscillatory part
uo (r, t) driven by GA sin(ωt). Write out the two problems along with their boundary
conditions. Note that the first problem is the well-known case of steady Poiseuille flow
through a pipe. (1)
(b) Now, focus on the second problem for uo (r, t). Scale the corresponding equation and
boundary conditions, and thereby obtain a condition that tells us how slow the forc-
ing frequency must be for quasi-static behaviour (wherein the unsteady term can be
neglected). (3)
(c) Next, consider the limit of very fast oscillations (large ω). Rescale the problem for
uo (r, t), with the understanding that now the inertial term must balance the pressure-
gradient term, atleast over the bulk of the channel. Find the condition under which
the viscous terms may be neglected. (2)
(d) Assuming the condition you derived above to be fulfilled, drop the viscous term, and
solve the resulting equation. Show that in this high-frequency limit, the flow has a plug
profile (no r variation) and oscillates with the same frequency as the pressure-gradient,
but with a π/2 phase-shift . (2)
(e) The solution obtained above is clearly only valid in the centre of the channel as it
does not satisfy the no-slip boundary condition at the wall. Therefore, there must be a
region of influence near the wall where the effect of the viscous terms cannot be ignored,
even for high frequencies. Find the length scale δ of this boundary layer by suitably
rescaling the r coordinate. (2)

5. Consider laminar unidirectional flow of a heat-transfer fluid between two hot flat plates. The
fluid enters this planar channel at temperature θ0 , and the walls are at a hotter temperature
θH . The length of the channel L along the flow direction (x − axis) is considerable greater
than the distance between the two plates H (y − axis). The suitably reduced heat equation
for the temperature θ of the fluid is given by:
 2
∂2θ

∂θ ∂ θ
u(y) =α + ,
∂x ∂y 2 ∂x2
θ = θH at y = 0,
θ = θH at y = H,
θ = θin at x = 0,
3

where α is the thermal diffusivity. The laminar flow profile is given by u(y) = 4Um Hy 1 − y
H ,
where Um is the maximum flow speed.

(a) Notice that the boundary condition for θ at the outlet is missing. Physically, it is not
obvious what condition to specify, and often, in the case long channels, this difficulty
is side-stepped by ignoring axial conduction. By scaling the problem, find the condi-
tion under which axial conduction can be ignored. Keep in mind that the dominant
process in this problem is conduction in the transverse (cross-plate) direction due to
the relatively small gap between the plates. (3)
(b) Next, consider the case of a slow flow. Obtain a condition that allows us to neglect axial
convection. What is the name of the well-known non-dimensional group associated with
this condition? Show that in this limit, when the axial convection term is dropped, the
fluid is predicted to have a uniform temperature equal to that of the plates. (1+1+1=3)
(c) The uniform temperature solution obtained above cannot hold right upto the inlet, as
there the entering fluid has temperature θin (< θH ). Find the scale δ of the entrance
zone in which axial convection must play a role even for slow flows. (2)

You might also like