[go: up one dir, main page]

0% found this document useful (0 votes)
19 views7 pages

Lecture 6

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 7

1

Fluid Mechanics (ME 201)


Lecture 6

1 Flow Patterns
The picture below shows flow of water. Can you guess the direction of flow? Is it from top
to bottom or bottom to top? From the given picture it is just anybody’s guess. Clearly,

we need more visual cues to establish the direction of the flow. And if we cannot even
say anything for sure about the direction of the flow, one can imagine what we would do
about the strength of the flow, nature of the flow, etc.
In order to study the nature of the flow, we need to generate flow patterns to indicate
the direction of flow, strength of the flow, variations in the flow over time, etc. Flow
patterns can be generated in different ways. Of course, all these different ways, at some
basic level, would involve determination of the motion of the underlying fluid particles.
Thus, flow patterns are drawn using lines that indicate motion of an individual fluid
particle or that of a collection of them. In what follows, we discuss the different flow lines
used to study fluid mechanics.

1.1 Path lines


Pathlines essentially indicate the trajectory of a certain marked material (fluid) particle.
It is the path traced by a fluid particle.
The motion of the particle can be described by the variation in the position vector of the
particle using its current velocity as following:
dri
= ui (t)
dt
If we choose to define the velocity of the fluid particle using a field variable u(r, t) which
represents the velocity at a location r at time t, then the path line can be obtained by
integrating:
dr
= u(r, t)
dt
1
©Gaurav Tomar(2022)

1
Pathline of the Ball

with the initial condition r(t0 ) = r0 , which simply indicates that the curve so obtained
from the integration is essentially for the particle that was at the location r0 at time t0 .
This is just one way of obtaining the pathline mathematically. If we could simply make
the particle leave a mark behind as it moves, we could obtain the pathline directly. Some
examples of pathlines (star trails, fire spark, car tail light trails) that all of us must have
seen are shown below in different photographs below. (These images are from the web.)

Particle image velocimetry


Shiny particles are released into the flow and then photographs are taken at a high frame
rate (at regular short intervals). These shiny particles appear as white spots on the image
and are tracked with the help of image processing softwares to yield position vectors at
short intervals to obtain the velocity field:
r(t + ∆t) − r(t)
V(r, t) =
∆t
The shiny particles are chosen to be neutrally buoyant (well more or less) and are small in
size so that they follow the fluid particles. The trace of these particles essentially results
in pathline of the fluid particles in their neighborhood. Of course, in practice keeping
track of a zillion identical particle is not straighforward and requires one to look at a
cluster at a time. For more information, see the presentation in the following link: PIV.

Example (from IG Currie):


Lets consider the two-dimensional velocity field u = x(1+2t)î+y ĵ +0k̂. We are interested
in the pathline of a particle that is at (x, y) = (0, 0) at t = 0. Equation of the pathline
can be obtained by integrating the following equation:
dx
= u(x, y, t) = x(1 + 2t)
dt
dy
= v(x, y, t) = y
dt

2
=⇒
x = C1 et(1+t)
y = C2 e t
Using the initial condition, we obtain C1 = C2 = 1. Further, we can relate x and y by
eliminating t:
x = y (1+lny)
or
lnx = (1 + lny)lny.
Figure below shows the pathline on an x − y plane.
y

7 Pathline

1 3 5 7 9
x

1.2 Streaklines
Consider a situation where a colored dye is leaking into the fluid. The dye molecules would
certainly follow the fluid particles and collectively they would form a pattern (see figure
below [NASA Photo ECN 33298-36]). The pattern essentially indicates the current location of the

set of particles that, over time, have passed through the location at which dye is being in-
jected.

3
5
6 4 3
The flow lines thus formed are called 2
7
Streaklines. In other words, a streakline 1
is the locus of all the particles that have 8

passed through a given location in space.


The schematic shown in the adjacent fig-
ure depicts an illustration of a streakline.
In this example, a bowling machine is con-
tinuously throwing balls with different ve-
locities and at different angles. A high-speed photo would be able to capture the balls in
action and it would look something like that shown here in the figure below. Clearly, at a
different time instant (t2 ) the balls would have moved and the new balls would have been
shot in a different direction with varying velocities and thus, the streakline would look
different. Clearly, streaklines vary with time. We should also note that, if the balls were
shot at the same angle with the same velocity and encounter the same wind direction,
drag, etc., they would all follow the same contour i.e. streakline would be exactly the
same as the pathline of the first ball.
6

5 5
7
6 4 3
8 4 2 @t1
7 3 1
9
8 2

@t 2 1

Mathematically, we can write an equation for the streakline by generating the locus
of all the points passing through a given location in space:

{ri (t) : ri (t = τ ) = r0 ; 0 ≤ τ ≤ t}

The above expression can be read as the set of all particles i such that at time t = τ the
particle i is at the location r0 ; where 0 ≤ τ ≤ t. Thus, for a given velocity field, if we can
write an equation for the motion of a particle with the initial condition ri (t = τ ) = r0 ,
then we can eliminate the parameter τ to obtain the locus of all the particles.

Example:
Lets consider the same velocity field chosen for the example of pathlines. Pathline of any
particle in the given velocity field is given by:

x = C1 et(1+t)
y = C2 e t

To compute C1 and C2 , we use the initial condition at t = τ :

C1 = e−τ (1+τ ) , C2 = e−τ .

Thus,
x = e−τ (1+τ )+t(1+t) , y = et−τ .

4
Eliminating τ from the above relation, we can relate x and y:

ln(x) = (1 − ln(y) + 2t)ln(y)

or
x = y 1+2t−ln(y)
As indicated by the above equation, streakline would vary with time. In the figure
below, we sketch the streaklines at t = 0 and t = 1s. The solid blue curve indicates the
streakline at t = 0 and the dashed blue curve indicates the streakline at t = 1 which is
quite different.

Pathline
y

7 Streakline (t = 0)
Streakline (t = 1)

1 3 5 7 9
x
Streaklines are quite extensively used by experimentalists. The famous Reynolds
experiment that demonstrated the instability in a pipe flow employed dye to show the
irregular motion of the fluid particles.

1.3 Streamlines
Pathlines and streaklines are related to the motion of a single or a collection of fluid
particles. Although useful in some capacity, these flowlines can also be quite deceptive
at times. For instance, consider a dye that has gone through a vortical motion and looks
a bit spirally now and unfortunately would remain so even though the flow has become
parallel (for example, after vortex shedding). Thus, although the snapshot of the dye
would suggest vortical motion in a region, it may actually be a reminiscence of past
history of the motion of particles. In such situations, it would be more useful to generate
flowlines using the velocity field that indicate the tendency of motion of particles at a
given instant of time.
Streamlines are flowlines that are tangent to the velocity (u = uî + v ĵ + wk̂) all along
them:
dx dy dz
= = .
u v w
By choosing a parameter λ, we can write:

dx = u(x, y, z, t)dλ
dy = v(x, y, z, t)dλ
dz = w(x, y, z, t)dλ

5
Clearly, streamlines are defined for a given velocity field i.e. at a given time instant =⇒
Streamlines give only the instantaneous picture.

Example:
We will take up the same velocity field that we had chosen for pathlines and streaklines.
Thus, for u = x(1 + 2t)î + y ĵ, we can write:
dx
= x(1 + 2t)

dy
= y

Integrating the above equations, we get:

x(λ, t) = C1 eλ(1+2t) , y(λ, t) = C2 eλ

We can obtain the streamline passing through (1,1) [a particular streamline], by elimi-
nating λ:
x = y (1+2t)
Figure below shows the streamlines drawn at t = 0 and t = 1s.

Pathline
Streakline (t = 0)
y

7
Streakline (t = 1)
Streamline (t = 0)
Streamline ( t = 1)
5

1 3 5 7 9
x

1.3.1 Streamtube
000000
111111
111111
000000
000000
111111
000000
111111
If we consider a collection of stream- 000000
111111
000000
111111
000000
111111
000000
111111
000000
111111
000000
111111
000000
111111
lines originating from a closed contour,
we would get a surface through which no-
flow occurs, since V is tangent everywhere
along all the streamlines constituting the
surface. The tube type structure so ob-
tained is called the streamtube. See figure 1111111
0000000
0000000
1111111
0000000
1111111
0000000
1111111
below for illustration. For an incompress- 0000000
1111111
0000000
1111111
0000000
1111111
0000000
1111111
0000000
1111111
0000000
1111111
ible flow, clearly, massflux entering from
one end of the streamtube would be the
same as that exiting from the other end.

6
1.3.2 Frame of reference
Interestingly, for the same flow different streamline patterns can be obtained just by
changing the frame of reference. To illustrate this, lets consider flow around a particle
moving with a constant velocity. In the lab reference frame, particle moves right to left
and displaces the fluid infront sidewards and fluid in the rear moves towards the particle
to fill in the space generated due to its motion. Thus, the schematic shows streamlines
that emerge from the front surface of the particle and then curve around to join at its
back surface.

On the other hand, in the reference frame of the particle, the particle itself is station-
ary, and therefore the velocity field is left to right towards the particle.

The two representations are for the same physical scenario, however, look completely
different. Therefore, we should be careful in interpreting the streamlines.

By the way, the image shown to motivate the discussion on flow lines was a cropped
portion of the one shown below. Save water!

You might also like