Lecture Notes for ME 313

Day 5 – Fluid Properties

>>FLUID STATICS

There are various kinds of equilibrium that we need to specify in order to verify that a
static model is justified; in particular mechanical equilibrium means that all forces are
balanced so that nothing is accelerating. In fact in this phase of our journey we will
make an even stricter assumption – that the fluid is not even moving, which is what we
mean when we say it is static. In a few lectures we will learn how to apply this model to
fluids undergoing uniform acceleration; but for now our picture is very simple.

So the basic paradigm here is the distribution of pressure in a tank of fluid that is at
rest, or in, say, the atmosphere if we assume that nothing is moving. If we look at a
differential element of fluid; we find that the equations of equilibrium are very simple.

dz

dy

dx

r
Suppose we have a coordinate system oriented with gravity so that g = − gk$ .

Firstly, since there is no velocity, there is no velocity gradient and thus there is no
viscous force. In other words, the shear is zero. This leaves only the pressure on the
six faces of our cubical element.

Here is a mental picture that will get you through 90% of fluid static problems.

First, think of this cube as containing material of uniform density ρ, and sitting in a
vertically oriented gravitational field g. The mass of material in the cube is

dm = ρdxdydz
Its weight is dw = gdm = ρgdxdydz .

Now what force balances that weight force? There is no shear, so the vertical walls
cannot contribute any force. In fact we see that the pressure on the lower surface must
be greater than the pressure on the upper surface. Let the pressure on the lower
surface be P and that on the upper surface be P+dP.
Then the equation of static equilibrium is

Pdxdy − ( P + dP )dxdy − ρgdxdydz = 0

⇒ dP = − ρgdz

We multiply by dx and dy because a pressure times an area gives a force.

Canceling, we find

dP
= − ρg
dz

In words this says that the pressure decreases as z increases and that it decreases at
a rate equal to the weight density (ρ times g).

This is called the hydrostatic equation. “Hydrostatic” is Latin for “still water” but it
applies to any static fluid.
r
More generally, suppose that the gravity vector is g = g x $i + g y $j + g z k$ . There are now 3
equations of equilibrium, viz.

∂P ∂P ∂P
= ρg x = ρg y = ρg z
∂x ∂y ∂z
or more compactly, r
∇ P = ρg

We can see some immediate implications of this relation. First, if the density is “small”
in some sense [if ρgz / P << 1], then the gradient of pressure will be negligible and the
pressure can be considered to be uniform. For example, consider the air in this room.
The pressure at the floor is very nearly equal to the pressure at the ceiling because the
density of air is small.

If this room were filled with water, which has roughly 1000 times the density of air, then
we would find that the pressure at the ceiling would be noticeably higher than the
pressure at the ceiling. Think about what happened to your ears that last time you dove
to the bottom of a relatively deep swimming pool.

We can also integrate the hydrostatic equation for incompressible fluids, to get

P = − ρgz + C
or equivalently
P − P0 = ρg ( z − z 0 )
Lastly today, I must mention that there are two common ways to cite pressures: we call
them absolute and gage pressures.

“Gage pressure” simply means a pressure measurement that is referenced to some

ambient conditions, either the local atmospheric condition or some other pressure in
the vicinity of the gage. This citation is common because many types of pressure
measurement devices such as manometers give readings relative to local conditions.

“Absolute pressures” are simply gage pressures referenced to absolute vacuum. That
is, a perfect vacuum would have zero absolute pressure.

It should be fairly obvious that you can convert from absolute to gage and vice versa if
you know the reference pressure.

Also clearly, it is quite easy to generate negative gage pressures. You do it every time
you suck soda up a straw. However, you cannot create negative absolute pressures,
just as you cannot create negative absolute temperatures.

The only “catch “ I’d like to warn you about is that the pressure in the ideal gas law is
the absolute pressure, just as the temperature is the absolute temperature. If you are
given a gage pressure to use in a problem requiring you to solve for gas properties,
you must must must convert to absolute.

In applications where both types of measurements are common, such as aerospace,

you will see the notations “psia” and “psig” used to denote lbf/in2 absolute and gage,
respectively.
MEASUREMENT OF PRESSURE

Let’s talk about two devices that illustrate practical applications of the hydrostatic
equation.

First, the barometer. The barometer measures atmospheric pressure. Now technically
ANY device which measures atmospheric pressure is called a barometer, but let’s
discuss the simplest and perhaps the oldest such device. Here we fill a tube full of fluid
and invert it in a tank of tank of the same fluid.

It’s crucial for the interpretation of the barometer reading that the height of the fluid
column and tube be sufficient to create a pocket of vacuum at the top. When this
condition is satisfied, the difference in pressure between the bottom of the column and
the top is given by ρgz . Since the pressure in the pocket at the top is basically the
vapor pressure of the barometer fluid, which is usually pretty small compared to
atmospheric pressure, the barometer gives a very close approximation to atmospheric
pressure. Note that the cross-sectional area of the tube is irrelevant so long as it is
sufficient to make surface tension effects negligible.

Now since in hindsight we know that atmospheric pressure is roughly 14.7 psia, we see
that the required height of the barometer is inversely proportional to the density of fluid
used as a medium. A water barometer would require roughly 34 feet to measure
standard atmospheric pressure. Mercury, being substantially denser, requires 29.92
inches – or 760 mm. In fact this is a popular definition of standard atmospheric
pressure: the pressure required to give a reading of 760 mm of mercury at 0 C. This
turns out to be 14.696 psia.

This point should be obvious – don’t get confused about a unit of length (millimeters)
being used to measure pressure. You always have to specify the working fluid, be it
mercury or water or alcohol or whatever, so that the ρ in ρg is specified. It’s actually
ρgz which is a pressure.

Now I hesitate to even bring this up… . Have any of you ever heard the expression,
“Nature Abhors a Vacuum”? For centuries this was the considered opinion of wise men
and was thought to explain why water in an inverted tube would not drain out. It was
thought that water would behave this way because it would seek to expunge whatever
vacuum might occur in the pocket at the top of the barometer. According to this view,
water would climb up a tube, no matter how high, just to get rid of that abhorrent
vacuum at the top.

This view is NONSENSE. “Nature abhors a vacuum” is WRONG. In fact Evangelista

Torricelli [1608-47] showed that water will only climb to a height of 34 feet and no
higher. You could make the tube arbitrarily tall but the water would only climb up as
high as atmospheric pressure demanded.

In words of one syllable, vacuum doesn’t pull the water up – the atmosphere pushes it
up. If you don’t understand this distinction and why it’s important, you should stay away
from designing anything remotely resembling a pump or pipeline.

Well, let’s move on to the measurement device called a manometer. A manometer

resembles a barometer in the sense that it is comprises a tube filled with liquid. Usually
we say the tube is U-shaped.

First let’s suppose that the density of the working fluid in the manometer is much
greater than the density of the fluid whose pressure is being measured. This might be
the case if you were using water or mercury to measure gas pressure. Since one end of
the tube is open to the atmosphere, the manometer measures gage pressure. Clearly
the reading is ρgz . Let me say that again: unlike the barometer, the manometer reads
gage pressure.
If the working fluid were mercury and the reading were, say, 5 cm, then the pressure
reading would be 50 mm of Hg gage or about 1 psig. If the ambient pressure were, say,
770 mm Hg, then the absolute pressure would be 820 mm Hg absolute or 15.9 psia.

Note that the manometer is quite capable of measuring negative as well as positive
gage pressures. What would the fluid in the tube look like for a negative gage
pressure?

Now a very important observation is that a given manometer configuration has a

maximum pressure difference that can be measured. This is dictated by the quantity of
working fluid. Suppose you only put 20 inches of mercury in a manometer, and you
tried to measure a pressure that was 21 inches of mercury gage. What would happen?
… I speak from experience; you do not want to do this.

Now consider the situation where the test fluid’s density is less than that of the working
fluid, but not negligibly so. This might be the case if you were measuring water
pressure with a mercury manometer. You would introduce about 7% error by neglecting
the water density. You have to account for the ρgz of the water as well as that of the
mercury.
Hydrostatic Force on Submerged Areas and Objects

So far the discussion of hydrostatics has been very elementary, basically applying
∆P = ρgz in various contexts. Let’s consider a broader class of problems in fluid statics,
where we try to compute the force on submerged surfaces. You would want to be able
to do this to, say, design a dam, or a water tank, or a boat, or a submarine… .

To simplify this discussion, we will assume that force due to atmospheric pressure is to
be subtracted out on all surfaces. This is equivalent to using gage pressures
everywhere, so that the pressure at a depth z is just ρgz .

The point behind all this is that the force is the surface integral of pressure.
r
F=− ∫Pnda
A
$

if n$ is the outward-pointing unit normal. If the surface under consideration is horizontal,

then the pressure is uniform and the force is just F=P*A. If the surface is vertical or
inclined, the pressure will vary with depth and we have to integrate over an area.

For example, consider the force on a vertical wall.

Z_0

Here I specify that the depth is Z_0 and the width of the wall is w. The total force acting
on the wall is just

z0 w
z 02
F = ∫ ∫ρgzdxdz = ρgw
z =0 x=0
2
Note that it happens that this can be written F = Pc A with A = wz 0 the area of the wall
ρgz0
and Pc = the local pressure at the centroid of the area. On the other hand, for the
2
purposes of computing moments, we cannot assume that the force acts at the centroid.

Instead we have to equate the moment with a force times a lever arm.

We find that it is convenient to introduce the concept of “center of pressure” at this

point. The center of pressure is the hypothetical point at which the force on the wall can
be considered to act for the purposes of computing the moment. [Replacing the
pressure distribution with a point force at the c.p. does not give, e.g., the correct shear
force, bending moment, slope, or displacement of the submerged surface]. Specifically
we mean that the force times the lever arm to the center of pressure equals the integral
of the moment induced by the pressure.

For this case let us denote the vertical location of the center by z p . Then we write

z 03 2 ρgz 0
( )
z0 w
 2 
rF = z p ( Pc A) = ∫ ∫( z )(ρgz )dxdz = ρgw =  z 0  wz 0  =  z 0 ( Pc A)
z =0 x=0
3  3  2  3 

2
So in the case of a rectangular wall we find z p = z0 . That is, the center of pressure is
3
below the centroid of the geometric area of the wall.

Now you have to understand that the nice factor 2/3 comes about only because the wall
geometry is rectangular. If the wall has another shape, then we get a different factor.
The more general formula is

Ic
z p = zc + .
zc A

where zc is the vertical location of the centroid and Ic is the moment of inertia of the
wall area about the centroidal line.

1 1
In the case of a rectangular wall, we have Ic = wz 03 , zc = z0 , and A = wz 0 , which
12 2
gives
1
1 wz 03 2
z p = z0 + 12 = z0
2 1 3
z wz
2 0 0
as we just showed.
Finally we can consider the case of inclined areas. A similar equation applies to
inclined surfaces,

Ic
y p = yc + .
yc A

where the y-coordinate to be the distance along the incline.

Now in the general case of a curved surface, the drill is the same though the execution
is somewhat more complicated. It is usually simplest to work out the pressure forces
component by component. Generally speaking, the horizontal components will cancel if
an object has sides that face in both + and - directions, even if the slopes are different.

In the case of submerged objects, the net pressure force is particularly simple to
calculate. The result is called Archimedes’Principle:

The buoyant force on a submerged object is equal in

magnitude and opposite in direction to the weight of the fluid
displaced by the object.

This also means that the buoyant force acts at the centroid of the displaced volume
(and not at the centroid of the object! The object may well have a non-uniform density
distribution!)

There are two relatively easy ways to demonstrate the truth of this principle. The first is
to integrate the pressure force in the vertical direction. Each element of area dx*dy has
a delta pressure acting in the vertical direction of ∆P = ρg∆z . When we integrate this
pressure-area product, we get F = ∫dxdyρg∆z = ρgV = Mg . The other way to
demonstrate the principle is to imagine that we cut out the volume displaced by the
object. (Or think of putting a dotted line around the object). Replace that volume with
fluid. Now in so doing we haven’t changed the pressure at all. Yet that pressure
distribution is precisely enough to balance the weight of water or whatever and also to
keep it from rotating about its center of gravity. In other words, the pressure force
acting on that volume is equal in magnitude to the weight of the displaced volume of
fluid, and also opposite in direction, in accordance with Newton’s law of action and
reaction.
P

∆z

P+ρg∆z