1 FLUID STATICS

When considering a stationary fluid there are a number of problems we are interested in, most of these
revolve around the pressure on a surface of a body. If we are not considering heat then pressure is the
only way a stationary fluid interacts with an impermeable surface. However, as we are engineers it is
often the impact of that pressure on the submerged surface that we are interested in. So, in this section
we will look at the definitions of pressure and its variation. Then we will investigate its effect on bodies.
What is pressure and how does it act? To answer this, it’s easier to think of a gas
but the same is true of any fluid. A static gas does not appear to move, even p
though individual molecules are in constant (random) motion. If we enclose the gas
within a container, we can detect the pressure as a fixed force per unit area acting
on the walls. This force is due to the molecules colliding with the walls of our
container and its fixed value is due to the huge numbers of molecules involved. If
we don’t approach the molecular scale, we can make the container as small as
possible and the pressure doesn’t change. From this argument, we see that
pressure is a scalar. It has magnitude but no direction associated with it. Pressure is p
in fact a normal stress that acts in all directions at a point in any fluid. At the surface
of a fluid, such as a boundary with a solid surface, the pressure acts normal to the
surface. Pressure is defined as acting towards the surface (away from the fluid).

1.1 Pressure Variation in a Static Fluid

We will start by looking at the derivation of pressure variation with depth/height of a stationary fluid. To
do this look at an element of stationary fluid that has notionally been isolated from the fluid around it
(the surrounding fluid is still there but we have drawn an imaginary boundary between it and the
element of fluid we are interested in). We consider a cylindrical fluid element noted in the diagram
below by its circular ends A − B.
The fluid is at rest so we know 2 things: firstly that only pressure and gravity can be acting on the
element and secondly, the net resulting force must be zero. Consider if this second point wasn’t true, a
finite net force would mean that the fluid could no longer remain stationary. Before we start the
derivation proper we must remember the way pressure acts. Pressure from the surrounding fluid acts
inwards at the surface, it doesn’t matter that we have created an imaginary surface between the fluid
inside and outside of the cylinder.
The areas of the circular ends, δ𝐴, are small enough so that
pressure can be considered constant on each end face.
However, there is a difference in pressure of δ𝑝 between A
and B. Resolving forces in the direction 𝐵𝐴 gives
(𝑝 + δ𝑝)δ𝐴 − pδ𝐴 + ρ𝑔δ𝐴δ𝑙𝑐𝑜𝑠θ = 0
Where we see contributions from pressure at the ends plus
the component of weight in the direction 𝐵𝐴. Note that the
pressure acting around the rest of the fluid is perpendicular to 𝐵𝐴, and so does not contribute. The
weight is calculated as the volume, δ𝐴δ𝑙, multiplied by the density of the fluid, ρ, and the gravity, 𝑔.
Dividing throughout by δ𝐴, simplifying and rewriting the change in vertical position in terms of length
and angle of inclination, δ𝑧 = δ𝑙𝑐𝑜𝑠θ,
gives the relation
δ𝑝
δ𝑝 + ρ𝑔δ𝑧 = 0 or δ𝑧
= −ρ𝑔

Taking the limit as δ𝑧 → 0 means that the relationship becomes formally differential and gives the
“hydrostatic equation”

∂𝑝
= −ρ𝑔
∂𝑧
There are a number of important points about this relationship:
1. The usual definitions of calculus notations, which you may or may not already be familiar with,
have been used. Here, δ, has been used to mean “a small but finite amount of”. While, ∂
represents the partial differential. At this stage, it is easy to think of the partial differential as
being the same as the ordinary differential (denoted by 𝑑) except that the object being
differentiated is a function of more than one variable. In this case pressure is a function of both
depth and density. We will use these notations again so please remember their meanings.
2. This is the first time we have used “control volume analysis” in the fluids side of the unit and we
will see it many times again. It is used to show were commonly used relationships come from
however you will not be expected to reproduce these derivations, only apply the resulting
equations.
3. We have introduced a “datum plane” which is just the position from where we have decided to
calculate the vertical position, 𝑧. This is important as, each time we apply a hydrostatic relation,
we can reset the datum plane (𝑧 = 0).
4. This relationship is valid for any static fluid, both compressible and incompressible (constant
density) fluids.
This relationship can be used in a number of interesting engineering application but for now, let us
consider the hydrostatic equation’s most important outcome:
The pressure at any two points on a horizontal plane is the same provided they are connected by a line wholly in
the static fluid.
We will understand more about what this means as we see some examples, although they are for
incompressible fluids the same applies to all fluids,

1.2 Equilibrium of Constant Density Fluid

In this side of the unit we will concentrate mainly on Constant density fluids. These are referred to as
incompressible. For this year, we will consider light gases (usually air) and liquids (usually water).
Water can be considered incompressible for all cases this year whereas air needs a little more caution.
We will consider air to be incompressible if: it is moving slowly (we delay this discussion while we look
at static fluids) and if the problem is at a “human scale”. The density of air can be considered constant
if we are only looking at a variation of a few metres in height but, over several hundred metres the
density can change quite significantly. For this year, if the object is at a human scale, such as a
building, the density is considered constant. However, when we are looking at columns of air the height
of a mountain, then we must consider density variation. Obviously intermediate heights, such as
around sky scrapers or dams, need to be dealt with carefully but for this year we will usually treat them
as having constant density.
If we apply the assumption of a constant density to our previous hydrostatic equation then we have an
ordinary differential relationship where pressure is only a function of vertical position,

d𝑝
= −ρ𝑔
d𝑧

This equation can be integrated with respect to z from a datum (z = 0) to ℎ, to give

𝑝(𝑧=ℎ) 𝑧=ℎ 𝑝(𝑧 = ℎ) + ρ𝑔ℎ = 𝑝(𝑧 = 0)

∫ 𝑑𝑝 = −ρ𝑔 ∫ 𝑑𝑧 or
𝑝(𝑧=0) 𝑧=0

Which is expressed as the “constant density hydrostatic equation”

p + ρ𝑔ℎ = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡

To understand what this equation means we will look at a couple of examples, where at the same time
we will introduce the idea of the “guage pressure”
Three differently shaped columns, all open to the 𝑝𝑎
atmosphere, are connected by a base. Note that each
column is wide enough that capillary effects are not 𝑧=0 𝑝 = 𝑝1 = 𝑝2 = 𝑝3
present (these are not covered in this unit). The liquid,
of density ρ, has an atmospheric pressure, 𝑝𝑎 , acting 𝑝 − ρ𝑔ℎ = 𝑝𝑎
𝑧 = −ℎ
on the surface. Taking the datum at the surface of the 𝑝 − 𝑝𝑎 = 𝑝𝐺 = ρ𝑔ℎ
first column so that p(𝑧 = 0) = 𝑝𝑎 . The first thing we
see by looking at the hydrostatic equation, is that if
each surface is at the same pressure then it must be at
the same height. Then, every position (labelled 1,2,3
on the right) is at the same depth (z = −ℎ) and so is also at the same pressure. So if we look specifically
at our constant density hydrostatic equation, and apply it for positions 1,2,3, relative to a datum at the
surface, then
𝑝1,2,3 + ρ𝑔(−ℎ) = 𝑝𝑎 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡

Note that ℎ is negative, as the vertical coordinate is defined positive upwards. Also that the container
shape is irrelevant. Depth of fluid is the only parameter that changes the pressure.
We can see that pressure at a constant depth stays fixed relative to the atmospheric pressure. This
then shows why the gauge pressure, 𝑝𝐺 , defined as the difference from ambient atmospheric pressure:

𝑝 − 𝑝𝑎 = 𝑝𝐺 = ρ𝑔ℎ
is a useful definition.
In this discussion we have two fluids, the liquid and the surrounding air. The impact of the surrounding
air has been condensed into the definition of an atmospheric pressure. But what if we introduce a
second liquid? Consider the new diagram on
the right. Here, a “U”-shaped cylinder is 𝑝(𝑁1 ) + ρ𝐴 𝑔Δ𝑧 = 𝑝(𝑀1 ) 𝑝𝑎
partially filled with two liquids with densities
ρ𝐴 and ρ𝐵 . Firstly, consider the horizontal line 𝑝(𝑁2 ) + ρ𝐵 𝑔Δ𝑧 = 𝑝(𝑀2 )
marked M. The pressure in the left and right
arms of the U-shaped tube (M1 and M2
respectively) must be the same as they are at 𝑝(𝑀1 ) = 𝑝(𝑀2 )
the same depth and wholly within the same
fluid, ρ𝐵 . However, what if we now consider
the horizontal line marked N. Taking the
datum at the position M and then applying the
constant density hydrostatic equation in each fluid we find

𝑝(𝑁1 ) + ρ𝐴 𝑔Δ𝑧 = 𝑝(𝑀1 ) in the left arm, and 𝑝(𝑁2 ) + ρ𝐵 𝑔Δ𝑧 = 𝑝(𝑀2 )

in the right. So, as the pressure at M1 and M2 is the same but the density of the fluids is different, the
pressure at N1 and N2 must be different. Hopefully at this stage we now understand our previous
statement
The pressure at any two points on a horizontal plane is the same provided they are connected by a line wholly in
the static fluid.
 1.3 Pressure Measurement
Pressure is a measure of force per unit area but can often be defined with different units, depending on
the area of application. The table below gives the more commonly used units, though we will restrict
ourselves to: N/m2 , Pascals and Bar.

Unit Symbol Conversion

Newtons per metre squared N/m2
Pascals Pa 1 Pa=1 N/m2
Bar bar 1 bar =100,000 Pa
standard atmosphere atm 1 atm=101,295 Pa ≈ 1.013 bar
pounds per square inch psi 1 psi =6892.7 Pa
mm of mercury mmHg 1 mmHg =1 torr =133.28 Pa
Meters of Water mH2O 1 mH2O =9807 Pa

Pressure is measured relative to another pressure, usually atmospheric (the gauge pressure). It is
seldom measured relative to a vacuum (absolute pressure). We will now look at a few common
measurement devices:
The Barometer.
A long closed cylindrical glass tube is filled with the barometer fluid and places
in a bath of the same fluid. The cylinder is then turned vertically with the open
end continuously submerged in the bath. The low pressure at the top of the
cylinder causes some gas, previously dissolved in the liquid, to come out of
solution. Gas will continue to come out of solution until the pressure at the top of
the column is equal to the “vapour pressure” of the liquid. This is the limit of
pressure, above which, dissolved gas will not be drawn out of solution.
Applying our constant density hydrostatic equation to the top of the column of barometer fluid, with a
datum at the surface in the bath of fluid gives:

𝑝𝑎 = 𝑝𝑣 + ρ(𝐻𝑔 ) gh

Hence, the height of the barometer fluid is directly proportional to the atmospheric pressure minus the
vapour pressure. Any fluid can be used for a barometer but usually it is Mercury. Mercury’s density
(𝜌𝐻𝑔 = 13560 Kg/m3 ) means that the cylinder can be short and the vapour pressure ( 𝑝𝑣 = 0.16 N/m2) is
small relative to atmospheric (𝑝𝑎 ≈ 105 N/m2 ) so the scale is can be approximated as linear.
U-tube manometer.
The U-tube manometer measures the difference in pressure between 𝑝
the two surfaces of the manometer fluid held in a clear U-shaped
cylinder. Normally (but not always) used for measuring the gauge
pressure in gases. One end is attached to the fluid being measured and
the other is open to the atmosphere. In the figure to the right, a
pressure vessel containing fluid A is attached to the manometer 𝑧=0
containing fluid B (always a liquid). We need to ensure that fluids A & B
are immiscible (don’t mix) and then we can apply the constant density
hydrostatic equation with the datum at the interface of fluids A&B to give
p + ρ𝐴 gy = 𝑝𝑎 + ρ𝑔x
Then, if the A is a light gas, the density of B is much greater than A and so we can make the accurate
approximation
p = 𝑝𝑎 + ρ𝐵 gx ρ𝐵 ≫ ρ𝐴
Hence, to measure the difference in pressure we just need a scale on each tube to calculate x, and a
value for the manometer density, ρ𝐵 .
Generally, the Interface of fluid A & B has increased area so
x stays approximately fixed. For a more accurate
measurement, when the pressure change is small, we can
use a lower density of liquid B and incline the measuring
sections of the tube. The picture on the right shows an
inclined manometer where x must be calculated using the
length of the liquid in the inclined tube multiplied by the sine
of the angle from the horizontal.

U-tube manometers can also be used to measure the pressure

difference in flowing fluids. In the figure on the right a manometer is
being used to measure the pressure difference across a sudden
constriction in a pipe. Although the fluid in the pipe is moving, the fluid in 𝑦
the smaller pipes connecting the U-tube manometer is stationary. We 𝑧=0
will discuss this more in the following sections, but the main practical 𝑝𝐼1
point is that the manometer must be connected perpendicular to the 𝑝𝐼2
flow direction. The junctions with the flow being measured are known as
static pressure tappings. Applying the constant density hydrostatic
equation with the datum at the interface of the higher manometer fluid
we see that
𝑝𝐼1 − ρ𝐴 gx = 𝑝1 + ρ𝐴 gy and 𝑝𝐼2 = 𝑝2 + ρ𝐴 gy
for the fluid between the interfaces and the static pressure tappings. While within the manometer fluid
we have
𝑝𝐼1 − ρ𝐵 gx = 𝑝𝐼2
Substituting in relations for the interface pressure gives
𝑝1 + ρ𝐴 gy + ρ𝐴 gx − ρ𝐵 gx = 𝑝2 + ρ𝐴 gy or 𝑝1 − 𝑝2 = (ρ𝐵 − ρ𝐴 )gx
Again, using the assumption that fluid A is a light gas we have
𝑝1 − 𝑝2 = ρ𝐵 gx ρ𝐵 ≫ ρ𝐴
Generally, when we look at fluid mechanics problems, it is vital to identify important points to use for the
analysis and the interfaces between fluids are just such points.
Other Devices
There are many other forms of pressure measurement device, but common ones
that you may experience as undergraduates include the Bourdon and
Piezoresistive strain gauges. Bourdon pressure gauges use the idea that a
flattened tube tends to regain its circular cross-section under pressure. This tiny
strain is magnified by curving the flattened tube so that the entire tube straightens
elastically under pressure. The motion of the tube is turned into the rotation of a
pointer around a dial using geared connection links.
There are many forms of electronic pressure transducers but most use a change in the electrical
resistance of a diaphragm as its strain changes due to the applied pressure difference across it. The
main practical point is that they all need periodic recalibration to remain accurate.
Example: Determine Gauge Pressure at A and a Closer Look at Density

In this problem, we will be repeatedly applying the constant

density hydrostatic equation to find the pressure at A in the
diagram on the right. But, before we tackle the question we
need to know a little more about density, its usual values and
how it is written. In this example the density is given as a
Specific Gravity (S.G), which is defined as the density as a
proportion of the density of water. The density of water is 1000
kg/m3. This is constant, even for reasonable changes in
temperature and significant variation in depth. So, for the oil in
the diagram, with a specific gravity of 0.8, the density is
0.8x1000 kg/m3=800 kg/m3. The density of water can be
compared to the density of the air at standard atmospheric
conditions which is 1.2 kg/m3. So, we can see immediately why
we can often ignore hydrostatic effects in gases over small distances but never for liquids.
As we look again at the problem of finding the pressure at A, the interesting points are at the interfaces
of the different fluids and so applying the constant density hydrostatic equation at the interfaces is the
best technique for this type of problem. As a reminder
p + ρ𝑔h = constant
Numbering the interfaces as 1,2,3 and noticing we have atmospheric pressure at the open end, we can
write the pressure at each successive interface in terms of the pressure at the previous.

Remember to be consistent with the direction (h is measured upwards). Take particular note of the plus
and minus signs above, these are related to h being positive and negative respectively. Then by
repeated substitution we get

𝑝𝐴 = 4414.5 − 2.35 − 19953.6 + 3531.6 + 𝑝𝑎 ⟶ 𝑝𝐴 − 𝑝𝑎 = −12009.9N/m2

In this question, we have assumed the air to be incompressible, this is a good working assumption and
one which works because of the short length of tubing. However, what would happen in the real tubing
and how would you attempt a similar question if this assumption were not made? In the
thermodynamics section of the unit we will introduce the equation of state of a perfect gas:
p = ρ𝑅𝑇
where R is the gas constant and T is the temperature in oKelvin. This represents how the pressure
changes with temperature and density in a gas. However, what if the temperature is constant? Then we
have a direct relationship between pressure and density.
p
= constant
ρ
We can take this a stage further and consider an enclosed gas whose mass must remain constant as
the volume, V, of that gas changes. In this case we have
 pV = constant
Returning to the problem, we can see that the average gauge pressure in the air is
𝑝𝐴𝑖𝑟 = −1.175 − 19953.6 + 3531.6 + 𝑝𝑎 ⟶ 𝑝𝐴𝑖𝑟 − 𝑝𝑎 = −16423.2N/m2
So, using our previous relationships, the air in the tubing should approximately have a density of 0.84
times the density of air and is taking up a volume 1.2 times greater than it would at standard
atmospheric conditions.

1.4 Forces on Submerged Surfaces

𝑝𝑎
Consider the pressure at a point on a submerged surface. We know
how to work out the pressure at that point. However, we often need to
know the force exerted by the pressure on the surface. In this section
we will look formally at how to define the force in terms of an integral of ℎ 𝑝 = 𝑝𝑎 + 𝜌𝑔ℎ
pressure. However, the important result is that this complicated
integration can be replaced by using the equivalent weight of fluid
bove.
Consider each point on the surface to be represented by an
infinitesimal area, 𝑑𝐴. This is a vector area with components in each of the three coordinate directions,

𝑑𝐴 = (𝑑𝐴𝑥 , 𝑑𝐴𝑦 , 𝑑𝐴𝑧 )

This can be written as the magnitude of the area, ⌊𝑑𝐴⌋ = 𝑑𝐴 ,

multiplied by the unit outward normal, 𝑛 , i.e.

𝑑𝐴 = (𝑑𝐴𝑥 , 𝑑𝐴𝑦 , 𝑑𝐴𝑧 ) = 𝑛⌊𝑑𝐴⌋ = 𝑛𝑑𝐴 = (𝑛𝑥 , 𝑛𝑦 , 𝑛𝑧 )𝑑𝐴

The force exerted on this infinitesimal area is just 𝑑𝐴 multiplied

by the pressure and it acts in the opposite direction to the
outward normal, i.e.

𝑑𝐹 = (𝑑𝐹𝑥 , 𝑑𝐹𝑦 , 𝑑𝐹𝑧 ) = −𝑛 𝑝𝑑𝐴

Note that the forces are defined by the coordinate system, so are positive in the coordinate direction.
Coordinate systems will become important through this unit.
The total force given by summing all of the forces on the infinitesimal areas that make up the surface of
the submerged object. In the limit as these areas tend to zero we have:

𝐹 = (𝐹𝑥 , 𝐹𝑦 , 𝐹𝑧 ) = − ∬ 𝑛 𝑝𝑑𝐴 = − ∬ 𝑛 (𝑝𝑎 + 𝜌𝑔ℎ)𝑑𝐴

𝐴 𝐴

Note that the area A is the entire area that represents the interface between the fluid and the
submerged object and the double integral is intended to remind us that the area is always made up
from two coordinate directions that define the surface area.
We also want to know where the line of action of the overall force acts. To do this we must calculate
the moment about a defined point for each infinitesimal surface area and integrate that to find the
overall moment. If we divide this by the force, then we can find the line of action relative to the defined
point.
If we had to do these integral calculations for every surface it would be extremely cumbersome.
However, we will now look at how we can simplify the problem.
1.4.1 Vertical Hydrostatic Thrust on Submerged Surfaces

The vertical force (or thrust), which we will take to be acting

downwards , can be calculated as the integral of the vertical
component of pressure force over the area. If that surface is planar
and horizontal, then the pressure is constant over the area and so
the total vertical force is
𝐹𝑣 = −𝐹𝑧 = ∬ (𝑝𝑎 + 𝜌𝑔ℎ)𝑑𝐴 = (𝑝𝑎 + 𝜌𝑔ℎ)𝐴
𝐴

Where 𝐴 is the area surface and ℎ is the depth of the fluid.

Rewriting 𝐴ℎ as the volume of fluid above, 𝑉 , we have
𝐹𝑣 = 𝜌𝑔𝑉 + 𝑝𝑎 𝐴
So, we can see that the vertical force is given by the weight of the liquid plus a component from the
atmospheric pressure. In fact, this atmospheric pressure components represents the weight of all the
air above the area 𝐴 . Hence, the total vertical force seems to be the weight of fluid above the surface.
This simple definition means that the problem changes from a complicated integral over a surface to a
much simpler problem of finding a volume.
Considering that any surface can be made up from elemental
areas, 𝑑𝐴, shows that this can be generalised to any shape; the
vertical force is given by the weight of fluid above. For example,
consider a domed cylinder of radius, 𝑟, submerged under ℎ metres
of water. The weight of water above is still
𝐹𝑣 = 𝜌𝑔𝑉 + 𝑝𝑎 𝐴
where the area is that of a circle of radius 𝑟 and the volume is
given by a cylinder of height ℎ minus a hemisphere of radius 𝑟, i.e.

𝐴 = 𝜋𝑟 2 𝑉 = 𝜋𝑟 2 ℎ − 23𝜋𝑟3

There are other interesting aspects that need to be considered for the evaluation of vertical forces. The
first of these is the problem of other solid surfaces.
Obstructions
The presence of other objects does not change the pressure on the
submerged surface so cannot change the vertical force. Consider the
force on the top surface of a shape shown in the figure on the right. If
we know the area of the top surface, we can work out the volumes and
therefore the total vertical force. However, then consider the vertical
force on the same object but with a second submerged object
occupying some of the volume of the fluid
above the original surface (see left). The
first thing to realise is that the depth of fluid is the same and so the
pressure on the surface is unchanged. If the pressure remains the same,
then the vertical force must also be unchanged. So, to use the idea of
the weight of fluid above we must always ignore any obstructions and
construct a volume from the surface of interest up to the level the fluid
would take. So we can now rephrase our definition and say that:
The vertical thrust on a submerged surface is equal to the weight of the effective fluid above.
The second interesting point about using the idea of the weight of fluid above is to consider the force
on the underside of objects.
Upward Vertical Force
Consider the force on the bottom surface of a shape shown in the figure
on the right. The pressure on a surface is only dependent on depth and is
independent of the orientation of the surface. The force on a surface is
therefore still the weight of the effective fluid above, but the direction can
now be different. In the figure on the right, the force is now acting
upwards. Again, remember that obstructions are still ignored. Note in this
figure that there is also a vertical thrust on the upper surface as well (acting downwards), and this is
something we will return to when considering buoyancy.
Net Force
When looking at the forces on a surface we often need to find
the net force from both faces. This sometimes (but not always)
means that the role of atmospheric pressure cancels as it acts
equally on both faces. As an example, look at the net force on
the bottom of a container which is placed in a bath of one liquid
(referenced as 1) but contains a different liquid (referenced as
2). Looking at the adjacent diagram we can see that the force
acting downwards is
𝐹𝑣 𝑑𝑜𝑤𝑛 = 𝜌2 𝑔ℎ2 𝐴 + 𝑝𝑎 𝐴 = 𝜌2 𝑔𝑉2 + 𝑝𝑎 𝐴
Notice that we have ignored the change in atmospheric pressure between the surfaces of the two
liquids so this must be at a “human” scale. Similarly, the force acting upwards is
𝐹𝑣 𝑢𝑝 = 𝜌1 𝑔ℎ1 𝐴 + 𝑝𝑎 𝐴 = 𝜌1 𝑔𝑉1 + 𝑝𝑎 𝐴

We should notice that the volume used to work out the “weight of the effective fluid above”, when
pictured in the diagram, contains fluid 2 and air. Think on this for a little until you are happy, return to
the previous notes if not.
So, the net force (acting in the 𝑧 direction i.e. upwards) is
𝐹𝑧 = 𝜌1 𝑔𝑉1 − 𝜌2 𝑔𝑉2
and we see that the atmospheric pressure has no effect as it acts equally on both faces.
Line of action of vertical forces
So far, we have calculated the magnitude of the vertical force and its sense (up or down). However, we
also wish to know its line of action. Simply stated, the vertical force runs through the centre of gravity
(C.G) of the effective fluid above the surface. For a simple column of fluid above a horizontal surface,
the vertical force acts through the centre of area (or centroid) of that surface.
What about more complex surfaces? Usually we can split the volume into
simple shapes whose C.G is known but we also should consider the
atmosphere. To understand this, we will look at a simple example. In the
figure on the right, a triangular prism with width, 𝑤, height, ℎ, and depth, 𝑑,
sits in water, also of height ℎ. The total vertical force downwards on the
prism is made up of the weight of the water, 𝐹𝑤 , and the weight of the air,
𝐹𝑎 = 𝑝𝑎 𝑤𝑑. The C.G of the triangular prism that makes up the water above
the surface acts through the middle of the width and 1⁄3rd of the depth of the
base. The C.G of the rectangular column that makes up the weight of the air
acts through the centre of the rectangle on the surface of the water.
Just as with solid mechanics, the total force acts
through the point about which there is no moment (i.e.
the moments balance). Drawing the moment diagram,
left, we can find the distances 𝑎 and 𝑏 by solving:
𝑑 𝑑 𝑑
𝐹𝑤 𝑎 = 𝐹𝑎 𝑏 and 𝑎+𝑏 = 2−3 = 6
1.4.2 Horizontal Hydrostatic Thrust on Vertical Plane Submerged Surfaces

The horizontal force on a surface can be

calculated as the integral of the
horizontal component of pressure force
over the area. If that surface is planar
and vertical, then the horizontal force is
equal to the average pressure, 𝑝𝑎𝑣𝑒 , on
the surface multiplied by the area of that
surface, 𝐴. As the pressure varies
linearly with depth of fluid, it is easy to
see that the average pressure is given by the value of the pressure “in the
middle” of the vertical surface. Specifically, “the middle” of the surface is
formally defined as the centroid, and we will return to this in a second. Given
the surface shown in the diagram on above, the horizontal force, 𝐹ℎ , is therefore given as
𝐹ℎ = −𝐹𝑦 = ∬ (𝑝𝑎 + 𝜌𝑔ℎ)𝑑𝐴 = 𝑝𝑎𝑣𝑒 𝐴 = (𝑝𝑎 + 𝜌𝑔ℎ𝐶 )𝐴
𝐴

or in words: The horizontal force on a plane vertical surface equals the pressure at the centroid times the area
Just as for the vertical forces, considerations of obstructions do not change the pressure or, therefore,
the horizontal forces. Similarly, the net horizontal forces are calculated by considering all faces of the
surface.
The position of the centroid of a planar area is covered in the Mechanics element of
Engineering Science. It can be intuitively defined as the average position of all the points
that make up the area. In the fluids element of this unit, you will be given the centroid
(and C.G), other than for very simple shapes. i.e. The centroid of circles squares and
rectangles is the geometric centre, while for right angled triangles it is 1⁄3rd of the base
and 1⁄3rd of the height from the right angle. This should also help with understanding how
the C.G of the triangular prism of fluid in the previous example was defined.
Line of action of horizontal forces
We now have a simple method to calculate the magnitude of the horizontal force, However, while we
can see that the average pressure will be at the centroid, we can also see that less force is being
applied above than below it. This means that the line of action will not be through the centroid. Instead,
the line of action of the horizontal force is through the Centre of Pressure (C.P) as shown in the
diagrams below.
As for the vertical force, the line of action (i.e. the C.P.) is once again found by considering moments.
However, to make the mathematics simpler, we take moments about the centroid of the area. This
means moving the origin of our coordinate system to the centroid. We can then equate the moment
from the horizontal force to the moment induced
by the pressure distribution. This is found as the
summation (integration) of moments from the
pressure acting on the elemental areas, 𝑑𝐴, that
make up the surface. Therefore, the moment
(𝑀𝑥 around the 𝑥 axis as drawn) of the horizontal
force about the centroid is given by

𝐹ℎ (ℎ𝑐 − ℎ𝐶.𝑃 ) = ∬ 𝑧 𝑝𝑑𝐴

𝐴

Writing the pressure in terms of the new coordinate system is

𝑝𝑎 + ρ𝑔(ℎ𝑐 − 𝑧)
Substituting into the integral and expanding gives
 ∬ 𝑧 𝑝𝑑𝐴 = ∬ 𝑧[𝑝𝑎 + ρ𝑔(ℎ𝑐 − 𝑧)]𝑑𝐴 = [𝑝𝑎 + ρ𝑔ℎ𝑐 ] ∬ 𝑧 𝑑𝐴 − ρ𝑔 ∬ 𝑧 2 𝑑𝐴
𝐴 𝐴 𝐴 𝐴

As the origin of the 𝑧 coordinate is now at the centroid, be definition

∬ 𝑧 𝑑𝐴 = 0
𝐴

So

∬ 𝑧 𝑝𝑑𝐴 = −ρ𝑔 ∬ 𝑧 2 𝑑𝐴
𝐴 𝐴

However, the integral of 𝑧 2 over the area is defined as the second moment of area (once again see the
Mechanics section of Engineering Science)

𝐼𝑥𝑥 = ∬ 𝑧 2 𝑑𝐴
𝐴

and the formula for its calculation will always be given in the fluids element of this unit. The change in
depth between the centroid and the C.P can therefore be found using
𝐹ℎ (ℎ𝑐 − ℎ𝐶.𝑃 ) = −ρ𝑔𝐼𝑥𝑥
Also, we can see that the C.P. has a shift in the 𝑥, as well as the 𝑧 coordinates, with respect to the
centroid. Although you will not be asked to calculate this shift this year, for reference, it can be show in
the same way as above, that

𝐹𝐻 𝑥𝐶.𝑃 = ρ𝑔𝐼𝑥𝑧

Where 𝐼𝑥𝑧 is the cross product moment of area

1.4.3 Total Hydrostatic Thrust on Submerged Surfaces

Any submerged surface will have a pressure distribution that will produce a force with components in
three coordinate directions. One vertical and two horizontal. This year, we will usually look at problems
where one of the horizontal components can be neglected, leaving us with one vertical and one
horizontal direction to consider. Everything we do can be extended easily for problems where the other
horizontal direction cannot be neglected.
We have worked out the vertical component of the force already. It has been defined for a general
surface as:
The vertical thrust on a submerged surface is equal to the weight of effective fluid above, and its line of action is
through the C.G. of the effective fluid above.
But, how do we calculate the horizontal force on a general surface which is neither vertical nor planar?
The answer is very simple. Consider a general surface, as an
example see the figure to the right. We now construct a vertical
planar projection of the surface normal to the horizontal force. In the
figure below, this projection of the surface is onto an 𝑥 = constant
plane, normal to the horizontal
force required, 𝐹ℎ . The planar
region is shown with a dotted
outer edge and will be referred to
as the projected surface with a
projected area, 𝐴′,. The horizontal force, 𝐹ℎ′ , on this projected
surface is exactly equal to the force on the general surface, 𝐹ℎ . It is
equal in both magnitude and line of action. We could prove this
equivalence mathematically but is simpler to consider the fluid bounded by the general surface and the
projected surface. This fluid will experience the horizontal force 𝐹ℎ′ from the fluid outside the domain
and a horizontal force equal to −𝐹ℎ as a reaction from the general surface. These 2 forces must
balance each other in magnitude or the fluid would have to move horizontally. As the fluid is static we
know they must balance. Similarly, the forces must be along the same line of action or the fluid would
have to rotate. We can therefore take everything we already know about forces on vertical planar
surfaces and apply it to the projected area of the general surface.
𝐹ℎ = 𝐹ℎ′ = 𝑝𝑎𝑣𝑒 𝐴′ = (𝑝𝑎 + 𝜌𝑔ℎ′𝑐 )𝐴′

where ℎ′𝑐 is the centroid of the projected surface, or in words:

The horizontal force on a surface equals the pressure at the centroid of the projected surface times the area of
that projected surface.
Similarly, for the line of action
𝐼′𝑦𝑦
𝐹ℎ (ℎ′𝑐 − ℎ𝐶.𝑃 ) = −ρ𝑔𝐼′𝑦𝑦 or ℎ𝐶.𝑃 = ℎ′𝑐 + ρ𝑔 𝐹ℎ

where the second moment of area, 𝐼′𝑦𝑦 , is again calculated on the projected surface. Note that we are
now using 𝐼𝑦𝑦 rather that 𝐼𝑥𝑥 as the projected surface is aligned with the 𝑦 − 𝑧 rather than 𝑥 − 𝑧 plane.

The total force, 𝐹𝑇 , is simply a vector whose components are (𝐹ℎ , 𝐹𝑣 ), so its magnitude is simply

𝐹𝑇2 = 𝐹ℎ2 +𝐹𝑣2

Its line of action can be drawn through the intersection of the lines of action
of the vertical and horizontal forces. The intersection of the line of action of
the total force with the surface is referred to as the point of action.

1.4.4 Example: Hydrostatic Thrust on a Dam

Find the magnitude and point of action of the hydrostatic force

exerted on the structure of a dam. The height of the water and the
dam is 80𝑚, the width of the dam is 30𝑚 and the base has a depth
of 60𝑚.
Let us consider the vertical and horizontal components of the force
(𝐹_𝑣 & 𝐹_ℎ) separately

𝐹_𝑣 equals the weight of effective fluid above.

60
For the dam in our case 𝐹𝑣 = 1000 × 9.81 × 2
× 80 × 30 = 7.0632 × 108 𝑁

Why have we ignored the atmosphere in this case? The atmosphere is often
neglected as either a small term, or more correctly, assumed to have the same
pressure acting all around the body with no net effect. The weight due to
atmospheric pressure in this case is not small: approximately 1.8 × 108 𝑁 on the
water surface.
We have considered that the ground surrounding the dam and its construction is
open to the atmosphere (not airtight). Therefore, the atmospheric pressure acts all
around the outer boundary. What’s more, it is (approximately) constant over the
height of the dam. In this case the vertical force from atmospheric pressure
variation over the 80𝑚 variation in height is small: 0.017 × 108 𝑁.
For the calculation of the vertical force we first construct our projected surface. The formal definition of
the projected surface may seem a little daunting but the simplest way to think
about it is; the shape of the object as seen when looking along the line of the
horizontal force. Hence for the dam, the projected surface is a rectangle of
80𝑚 by 30𝑚 whose centroid will be at a depth of 40𝑚. Again, noting that we
will assume constant atmospheric pressure acting on all faces of the dam, we
have
𝐹ℎ = ρ𝑔ℎ𝑐 A = 1000 × 9.81 × 40 × (80 × 30) = 9.4176 × 108 𝑁
Here, we have dropped the ‘ which we used for reference to the projected surface, from now on we will
assume everything is on the projected area. The total force F𝑇 is then given by

F𝑇 = √(𝐹ℎ )2 + (𝐹𝑣 )2 = 1.1772 × 109 𝑁

The total force acts in the vector direction (𝐹ℎ ,𝐹𝑣 ) which must be normal to the plane for flat surfaces as
all the pressures that make up the force are in the same normal direction. This gives a useful trick for
flat surfaces which we will come back to at the end of this example.
However, the horizontal force acts through the centre of pressure (C.P.). The centre of pressure is
given by
𝐹𝐻 (ℎ𝑐 − ℎ𝐶.𝑃 ) = −ρ𝑔𝐼𝑥𝑥
using our previous formulae for 𝐹ℎ and rearranging
ρ𝑔𝐼𝑥𝑥 ρ𝑔𝐼𝑥𝑥 𝐼𝑥𝑥
ℎ𝐶.𝑃 − ℎ𝑐 = = =
𝐹𝐻 ρ𝑔ℎ𝑐 𝐴 ℎ𝑐 𝐴
3
For a rectangle of height 𝐻 and base 𝑏, 𝐼𝑥𝑥 = 𝑏𝐻 ⁄12
30×803
Hence the line of action for the horizontal force is ℎ𝐶.𝑃. − ℎ𝑐 = 12×40×30×80 = 13.333m

So, the horizontal force acts a depth of 53.333𝑚 (as does the total force)
Flat surface, quick calculation
We have previously stated that the total force must be normal to the plane for flat surfaces, as all the
pressures that make up the force are in the same normal direction. We have also stated that the total
force must pass through the intersection of the lines of action of the vertical and
horizontal forces. Therefore, the vertical and horizontal forces must intersect at a flat
surface. From our previous calculations, we know that

𝐹𝑣 = 7.0632 × 108 𝑁
and its line of action is 20𝑚 from the base of the dam (i.e. 1⁄3rd of 60𝑚). Using similar
triangles, we see that the C.P must be at a depth of
60 80 2
= → ℎ𝐶.𝑃. = 80 × = 53.333𝑚
60 − 20 ℎ𝐶,𝑃, 3

Similarly, as 𝐹𝑣 is the vertical component of the total force

√802 + 602 5
𝐹𝑣 = 𝐹𝑇 cos 𝜃 → 𝐹𝑇 = 7.0632 × 108 × = 7.0632 × 108 × = 1.1772 × 109 𝑁
60 3
and
4
𝐹ℎ = 𝐹𝑇 sin 𝜃 → 𝐹𝑇 = 1.1772 × 109 × = 9.4176 × 108 𝑁
5
We can therefore see that if a surface is flat, but not vertical and the volume & C.G of the effective fluid
above are easily calculated then we can just work out the vertical force and calculate everything else
from geometry.
 1.5 Buoyancy
Archimedes principle: The upward buoyant force on an object, wholly or partially immersed in a fluid, is equal
to the weight of the fluid displaced by the object.
This commonly cited law is due to the pressure difference around an object,
and is easily proven given what we already know about the vertical force on
submerged surfaces. If we split the object into surfaces such that one has
faces pointing downward and the other has
faces pointing upwards, then the force on each
is 𝐹𝑣 𝑑 and 𝐹𝑣 𝑢 respectively. The buoyancy
force is the difference between the two i.e.
𝐹𝐵 = 𝐹𝑣 𝑢 − 𝐹𝑣 𝑑
We also know that the forces on each surface are given by the weight of
effective fluid above. This means that the downward force is given by the
weight of fluid that takes up the volume directly above the object. Also, the
upward force is given by the weight of fluid that takes up the volume
directly above the object plus the volume of the object. The difference
between the two forces, the buoyancy, is therefore the weight of fluid that
takes up the volume of the object. This is just another way of saying the
weight of the fluid displaced by the object.
If we look at the line of action of our buoyancy force, we also see that it
passes through the CG of the object IF the object had the density of the
surrounding fluid, this we will refer to as the centre of buoyancy (C.B.). As we usually deal with
constant density fluids in this type of problem we can see that the C.B. is the centroid of the volume of
the object.

1.5.1 Example: Terminal Velocity of a Rain Drop

Assume that the drop is a sphere of radius 1mm, the air and water have
densities of 1.2 kg/m3 and 1000 kg/m3.
The drag on the spherical droplet is given by the formulae
1
D = 𝐶𝐷 2 ρ𝑎 𝑉 2 𝐴× 𝐶𝐷 = 0.45

where V is the speed of the drop, 𝐴× is the x-sectional area, ρ𝑎 is the density of
the air and 𝐶𝐷 is the drag coefficient (taken here as constant). We will come
back to all these terms in the next section of the unit.
The buoyancy force is given by the weight of displaced fluid,
4
B = π𝑟 3 ρ𝑎 g
3
where 𝑟 is the radius of the spherical drop.
This is not a true static problem, but we shall use it to demonstrate buoyancy. We are replacing the
relative motion of air and water by a simple force relation. At terminal velocity there is no acceleration,
so all forces are in balance and
D=W−B
1 4
𝐶𝐷 ρ𝑎 𝑉 2 πr 2 = πr 3 g(ρ𝑤 − ρ𝑎 )
2 3
where, ρ𝑤 is the density of the water. The relative density of water to air means the buoyancy is small.
Rearranging the equation algebraically before substituting in the number we have

8rg 8 × (1 × 10−3 ) × 9.81 × (1000 − 1.2)

𝑉2 = (𝜌 − 𝜌𝑎 ) → 𝑉=√ = 6.96m/s
3𝐶𝐷 𝜌𝑎 𝑤 3 × 0.45 × 1.2

Keeping the algebra as long as possible is best practise in real life and exams because it allows us to
ask the question. Does the equation give the correct units? This makes a good check and we will return
to this when we look at dimensional analysis:

𝑚 × 𝑚/𝑠2 × 𝑘𝑔/𝑚3 √𝑚2

√ = 2 = m/s
𝑘𝑔/𝑚3 𝑠

What if we had made a typical engineering assumption and neglected the buoyancy of the air
completely, what difference would there have been in the calculated terminal velocity? To the two
decimal figures given above the answer would have been unchanged.
Note: Buoyancy problems in air often include the use of the equation of state i.e. p = 𝜌𝑅T. or for a
fixed temperature, pV = 𝑐onstant. In air we have to remember that the density changes with height,
whereas for water, the density is fairly constant at all depths.

1.6 Learning Outcomes

◼ The hydrostatic equation and the constant density hydrostatic equation
◼ How these relations apply to simple pressure measurement devises and the definition of gauge
pressure.
◼ The definition of standard pressure units
◼ How to solve thrust problems on simple submerged surfaces
◼ How to solve buoyancy problems.