and HYDROSTATICS

ELEMENTARY STRUCTURES
CIVL1100
Discipline of Civil, Surveying and
Environmental Engineering

2017 Lecture Notes

Lecturer:
Associate Professor Mark Masia
o
Elementary Structures Copyright c Professor R.E. Melchers

o
Hydrostatics Copyright c Associate Professor M. Masia
All rights reserved.
CIVL1100 – Hydrostatics

Fluid Properties
Even though fluid is composed of discrete molecules of matter, for all the scales at
which we will work, it can be effectively treated as continuous. This will allow us to
use calculus in our modelling of fluid behaviour.

Definition of a fluid, viscosity (P&W 1.3 and 1.5)

A fluid is a substance that deforms continuously when subjected to a shear stress (a
force per unit area applied tangentially to a surface).

y
F
U
a aa b bb
u
t
y
x
d c Stationary Base

Figure 1. Deformation of a fluid under a shear stress.

Consider the fluid between the stationary base and the plate moving with velocity U
in Fig 1. a -> aa and b -> bb

u, the horizontal component of fluid velocity varies linearly from 0 at the bottom to U
at the top. Experiments show that the force F is proportional to the area of the moving
plate, A, the velocity, U, and inversely proportional to the thickness of the fluid, thus:

AU
F=µ (1)
t

µ is a proportionality factor.

If the shear stress, τ on the fluid immediately under the moving plate is defined as the
force divided by the area, τ = F / A , then we can rewrite (1) as:
U
τ =µ (2)
t
 U / t is the rate of angular deformation of the fluid. We can write it in differential
form as U / t = du / dy and hence (2) becomes:
du
τ =µ (3)
dy

Equation (3) is called Newton’s Law of viscosity, and the proportionality constant, µ ,
is called the viscosity of the fluid.

For water, at a particular temperature the viscosity is constant. In some other fluids,
paint, for example, the viscosity can vary with du / dy .

We say that water is a Newtonian fluid and paint is non-Newtonian.

We often use ideal fluids for analysis of complex situations.

An ideal fluid has zero viscosity and is incompressible.

Dimensions and Units (P&W 1.2 and Masters 1.2)

In Australia (and most of the Western world), the units of fluid mechanics are SI.
With this system, we symbolize and measure as follows:

Mass (M) : kilograms kg

Length (L) : metres m
Time (T) : seconds s

Special (very important!) units

Force (F) : newtons N 1N = 1 kg.m/s2

Energy (E) : joules J 1 J = 1 Nm
Power (P) : watts W 1W = 1 J/s
Pressure (p) : pascals Pa 1 Pa = 1 N/m2
Volume (V) : litre l 1000l = 1m3

Note:
The symbol ‘k’ in front of a quantity multiplies its value by 1000. (1kg = 1000g). We
will frequently use kN, kW, kPa and kJ.

The symbol ‘M’ multiplies a value by 106, e.g. 1MW = 106W.

The symbol ‘G’ multiplies a value by 109, e.g. 1GW = 109W.
Example:
A plate 0.5mm distant from a fixed plate moves at 0.25m/s and requires a force per
unit area of 2Pa to maintain this speed. Determine the fluid viscosity of the substance
between the plates in SI units.

Solution:

du du u
=τ µ= ;
dy dy t
τ =F/A
F t 0.5 /1000
µ = ⋅ = 2.( ) = 0.004 Pa.s
A u 0.25

Mass, weight and concentration variables (P&W 1.2, 1.5)

The density, ρ of a fluid is its mass per unit volume. In order to measure it at a point,
we need to determine the mass, ∆m , contained by a small volume, ∆V , at the point.
We then compute:

∆m
ρ= (4)
∆V

A related term that is widely used is the specific weight of the fluid, γ . This is the
weight of the small volume of fluid (i.e. the vertical force, mg, required to maintain
its mass in equilibrium when its acted upon by gravity) divided by the volume:
mg
= γ = ρg (5)
∆V

The specific gravity, S, of a substance is the ratio of its density or specific weight
compared with that of water.

Temperature and thermodynamic variables (P&W 1.7, Masters 1.4)

The heat content in a volume, ∆V of fluid is given by:

QH ρ c pT ∆V
= (6)
where c p is the specific heat at a constant pressure and T is the absolute temperature.
Heat is measured in joules(J). Temperature is measured in Kelvins, K or degrees
centigrade (Celsius), °C . The relationship between the two is:
K = C ° + 273 (7)
c p and its partner, cv , the specific heat at constant volume are important measures
of the internal energy in gases. For an ideal gas they are related by:
 c=
p cv + R (8)
where R is the universal gas constant, which we’ll come to later.

The ratio of specific heats, k = c p / cv has a value of approximately 1.4 for many gases
(including air).

Pressure and a perfect gas (P&W 1.7, Masters 1.2)

Pressure, p, is defined as a force per unit area (units Pa). We shall discuss the concept
of pressure at a point in the next section when we start our study of hydrostatics.
Gases are compressed by pressure and their volume or pressure increases when they
are heated. The perfect gas law states:
p = ρ RT (9)
using the notation defined above and where R is a constant for the gas. A perfect gas
is one which satisfies this law. Many real gases, with different R, closely approximate
the law. To differentiate between gases, we need to define a mole.
A mole of an element (abbreviated to mol) is the amount of the element contained in
the same number of molecules of the element as there are carbon atoms in 12g of
carbon = 6.0221367 x 1023 (Avogadro’s number).
The molar mass, M, of a substance is the mass in grams of 1 mol of the substance and
its units are g/mol. Notice that the molar mass of carbon is numerically equal to its
molecular weight. It follows from Avogadro’s Law, that all substances have molar
masses which are numerically equal to their molecular weights.

R, the specific gas constant, has SI units of J/(kg.K) where K is the temperature in
degrees Kelvin, but often grams are used because of conventions in chemistry. It turns
out that there is a universal gas constant, Ru, which results when the R for each gas is
multiplied by its molar mass, M.
Thus:
Ru = RM (10)

Ru = 8.314J/( mol.K).
Masters uses units of litres, grams and atm’s, giving R=0.082056L.atm.K-1.mol-1.
Oxygen has an atomic weight of 16, it gets around mostly as a paired atomic molecule
so that its molecular weight is 32. Therefore a mol of oxygen has a mass of 32g and a
kg-mol of oxygen has a mass of 32 kg.

(9) can be rewritten as:

pV = mRT (11)

where V is the volume of the gas and m its mass, and if we assign M as the molar
mass of the gas and n the number of moles of the gas in the volume V, we get:
pV = nMRT (12)
N J
The units for this equation are: 2 ×=m3 ( kg .mol ) × ( kg / kg .mol ) × ×K
m kg .K

Using (10) we can re-write as:

 pV = nRuT (13)
This is the form used in Masters.
Hence, in SI units:
pV = 8312nT (14)
(for p,V, T in SI units) and also:
R = 8312 / M [N.m/(kg.K)] (15)
or:
R = 8.312 / M [N.m/(g.K)]
Hydrostatics 1

Pressure at a point (P&W 2.2)

Pressure, p, is defined as a force per unit area (units Pa).

A fluid at rest has the same pressure in all directions

The proof of this assertion is easy:

Ps, δs
δy
δs
Px, δy
θ x
(x,y) δx
Py, δx
ρg(δxδy)/2
Figure 1. Free-body diagram of wedge-shaped particle

Consider the fluid at rest contained in the unit thickness wedge, located at (x,y) in Fig.
1. Because there is no motion there are no shear forces and hence no shear stresses
acting on the 3 plane surfaces δx, δy and δs. These surfaces are perpendicular to the x-y
plane. We assume that the ends of the wedge are parallel to the x-y plane and have no
shear forces acting on them so that the only forces acting in the x, y directions are
those shown above.

Notice that in addition to the average pressure forces, px, py and ps acting on the sides
of the wedge there is also a ‘body’ force acting downwards which is the gravitational
force given by the mass density of the fluid multiplied by its volume (the area of the
triangle times its unit thickness) multiplied by the gravitational acceleration, g.

Now if we sum the forces in the x and y directions and equate to zero since the fluid is
at rest:
ΣFx = pxδy – psδs sinθ = 0 (16)
and
ΣFy = pyδx – psδs cosθ – γ(δxδy/2) = 0 (17)

We now reduce the size of the wedge so that the sides are infinitesimal, maintaining
the angle θ.
Since δ s sin θ = δ y and δ s cos θ = δ x , Equations (16) and (17) become:
p xδ y − psδ y =
0 (18)
and
pyδx – psδx – γ(δxδy/2) = 0 (19)

Now the last term in Equation (19) is the product of two infinitesimal terms and so is
very much smaller than the other two terms so it can be dropped. Equation (18) can
be divided through by δy and Equation (19) can be divided through by δx resulting in:
p=s p=
x py (20)
θ is an arbitrarily chosen angle, hence ps can be any direction, and the conclusion is
that the pressure in a fluid at rest is the same in all directions.

If the fluid is in motion then there will be shear stresses and the pressure in all
directions is no longer the same.
The pressure is then defined as the average of the three normal compressive stresses,
i.e.

p x + p y + pz
p= (21)
3
Basic Equations of Fluid Statics (P&W 2.3 and 2.4)
Pressure variation in a static fluid

Figure 2 below shows some of the forces acting on a parallelepiped element of a fluid
at rest. Pressure forces similar to those indicated on the δ xδ z face will be exerted on
the δ yδ z face and the δ xδ y face.

Summing the forces in the y direction gives:

∂p
ΣFy =− δ xδ yδ z − γδ xδ yδ z (22)
∂y
For the x and z directions, since there are no body forces:

∂p
ΣFx =− δ xδ yδ z (23)
∂x
and
∂p
ΣFz =− δ xδ yδ z (24)
∂z
∂p δ y
(p + )δ xδ z
∂y 2
y
δz

γδ xδ yδ z
(x,y,z)
δy

δx
∂p δ y
(p− )δ xδ z
∂y 2
x
z 0
Figure 2. Fluid element at rest

Now for a fluid at rest the resultant force in each direction must sum to zero and so
Equations 22, 23 and 24 reduce to:
∂p ∂p ∂p
= 0; = −γ ; = 0 (25)
∂x ∂y ∂z

Note that this means that as we move through the fluid in the horizontal plane the
pressure does not change. Points at the same elevation in the same continuous mass
of fluid have the same pressure.
Pressure variation in an incompressible fluid

If the fluid density can be considered to be constant, then Equation (25) which we
rewrite as:
dp = −γ dy (26)
which is easily integrated:
p= −γ y + c (27)
Here c is an integration constant.
If we assume that the pressure is 0 at the free surface of a body of liquid, then the
pressure at depth, h, in a liquid with constant density is given by :
p γ=
= h ρ gh (28)
If the density varies with depth, then we must use the differential form and integrate
as shown in the example below.

Example:

What is the pressure at a point 12m below the free surface in a fluid that has a variable
density in kg/m3 given by= ρ 500 + 11.1h , where h is the distance in metres measured
downwards from the free surface?

Solution:
dp = −γ dy = ρ gdh = g (500 + 11.1h )dh (29)
Whence, integrating both sides of the equation, from the surface to h, and assuming
that the pressure at the surface is zero and p at depth h:
p h
∫ dp
=
0 ∫0
g (500 + 11.1h )dh
11.1h 2
=p 9.81(500h + ) (30)
2
p= 9.81 × (500 × 12 + 5.55 × 122 )
= 66.7kPa
Units and scales of pressure measurement (P&W 1.4)
Positive pressure is usually measured relative to local atmospheric pressure. This is
often referred to as gauge pressure. (‘gage’ in US) When pressures are negative it is
often more convenient to measure them as absolute pressure, i.e. relative to that of a
perfect vacuum.

The Bourdon gauge is widely used to measure pressures relative to the atmosphere. It
consists of a hollow curved metallic tube, open at one end to the pressure to be
measured. If the pressure inside the tube exceeds that outside it, the tube straightens
and the consequent movement of the closed end is then used to indicate a change in
pressure on a dial.

Movement of
end

Pressure in

Figure 3. Schematic of Bourdon gauge.

Barometers for meteorological use, measure local atmospheric pressure relative to

absolute. A standard atmosphere is 101.325kPa. For elevation z in metres above sea
level, the pressure can be estimated as:
 293 − 0.0065 z 
5.26

P = 101.325  
 293 
Because p = ρ gh , pressures are often indicated as a ‘head’ (the ‘h’ in the equation)
of a particular fluid with standard ρ .
Thus, a standard atmosphere is also equal to 760mm of mercury or 10.34m of water.
Measuring pressure as a head of water is very common in engineering calculations.
When there is no indication whether pressure is gauge or absolute, normally in
engineering calculations it is assumed to be gauge.
Pressures less than atmospheric are often expressed as negative pressures, in which
case their absolute value is p(absolute) = p(gauge)+p(atmosphere)
Manometers (P&W 2.4.3)
Standard manometers
Manometers are devices used for measuring differences in pressure. The most
elementary manometer is a simple open-ended tube inserted into a body of fluid. The
fluid rises in the tube to some height h above the point at which a pressure
measurement is sought. If the pressure at the top (open) end is zero (atmospheric) then
the pressure h below is ρ gh . If a negative pressure is required, then the tube must
loop below the point of measurement, so that a negative h can be measured.

Sometimes it is inconvenient to measure the head in terms of a water pressure because

it might require a very long tube. In this case a mercury manometer might be used
because its density is 13.6 times that of water and so if the water pressure is 13.6m,
that of mercury is only 1m.

The important thing to remember in calculations involving manometers is that if a

fluid is continuous between two points at the same elevation, the pressure at the two
points is the same. Apart from that, as you move up a distance h in the fluid, the
pressure falls by ρ gh and as you move downwards it rises by ρ gh .

The general procedure is as follows:

1. Start at some location with a (preferably) known pressure, at a meniscus (an
interface between two fluids) – possibly air/water.
2. Move some distance say h1 to the next meniscus, adding ρ gh1 to the pressure
with h1 positive if the movement is downwards, negative if upwards.
3. Continue to the other end of the gauge, adding pressures as you go. The result
will be a sum or an equation providing you with a means of calculating the
unknown pressure.
Example:
Assume in figure (c) above that S1=0.86, S2 =1.0, h1=150mm, h2=90mm. What is the
pressure at A in mm of Hg (mercury) gauge. If the barometer reading is 720mm Hg,
what is the pressure in metres of water, absolute?

Solution:
The pressure at the open meniscus is 0 with reference to gauge. The pressure at the
level of the interface with the second fluid is + ρ 2 gh1 =
S2 ρ water gh1 . The point A is h2
above and so the pressure is then reduced to give:
= p A S2 ρ gh1 − S1ρ gh2
=(1 × 1000 × 9.8 × 0.150) − (0.86 × 1000 × 9.8 × 0.090)
= 712 Pa
If we denote the head at A by hA, then we can convert this to a head of water using the
relationship:
p A = ρ ghA
from which hA = 72.6mm (water)

To express this in mm of mercury, note that for mercury, S=13.6.

p A = 13.6ρ ghA
whence hA = 5.34mm Hg.
To determine the pressure in absolute units we must add the atmospheric pressure.
Thus in mercury units the pressure is 725.34mm, which we can convert to head of
water by multiplying by the specific gravity of mercury-> 9.864m.

Application of the procedure above to Figure 2.11(a) above:

p A =pB − γ 3h3 + γ 2h2 + γ 1h1 (31)
Where the units are Pa.

Sometimes, using specific weights, the pressure may be expressed as a head of water
and we could write the equation:
 hA =hB − S3h3 + S2h2 + S1h1 (32)
Notice that if this equation is multiplied by ρ g where ρ is the density of water, the
equation becomes:
ρ ghA =ρ ghB − S3 ρ gh3 + S2 ρ gh2 + S1 ρ gh1 (33)
in which each term reduces to those in Equation (31).

Micromanometers
A general form of micromanometer is shown in Figure 2.12 below. We can compute
the pressure at C from that at D by the same process:
p=
C pD + ( k1 − ∆y )γ 1 + ( k2 − R / 2 + ∆y )γ 2 + Rγ 3
(34)
− ( k2 − ∆y + R / 2)γ 2 − ( k1 + ∆y )γ 1

Since for equal volumes of the incompressible fluids we require that:

∆yA = Ra / 2 (35)
where A is the large cross section of the reservoirs at the top and a is the cross section
of the lower tube. We can substitute for ∆y from this expression into Equation (34)
to give:
a a
pC = pD + R[γ 3 − γ 2 (1 − ) − γ 1 ] (36)
A A
Note that the quantity in square brackets is a constant for any particular gauge and
hence the pressure difference is proportional to R.
Inclined manometers also can be used to produce a large reading for a small change:
Hydrostatics 2

Forces on plane areas

We now determine what the effects of pressure are on plane objects which are
submerged in a fluid – what are the forces and what are their lines of action.

Because the pressure in general will vary and/or be distributed over an area, we will
usually need to integrate (sum) the effects on many small areas over each of which the
pressure is essentially constant. Keep in mind always that integration is simply the
sum of the small bits dx, dV whatever. The limits of the integration just describe
where the bits are located.

Horizontal surfaces
On a horizontal plane surface, we know that the pressure will be constant, so it is easy
to determine what the total force, F, will be – it is simply the pressure multiplied by
the area of the surface. If the area is broken up into many small elemental areas,
designated by dA, their sum will be equal to the total area, so that in mathematical
notation we can write:
A = ∫ dA (37)
A
which says that the total area is equal to the sum of all the elemental areas in the
region A (indicated below the integral sign). So now we can write:

F = p ∫ dA (38)
A

The location of the line of action of the resultant (called the center of pressure) is not
quite so simple. We want to know what moment (torque) is produced by the pressure
forces about some arbitrary axis. We then find a location of the center of pressure,
( x , y ) , for our total force, which produces the same moment.

y x

δA
A
x

Figure 1. Plan view of arbitrarily shaped horizontal plane surface

If we break our area up into a lot of little areas, δA , the force on each of them will be
pδA and if we designate their location as (x,y) their moment about the y-axis will be
the force multiplied by their distance from the y-axis, ie pδAx. If we add up all these
moments we get the total effect which will be ∫ xpdA - which is simply mathematical
A

notation for saying the sum of the elemental moments pδAx over the entire area, A.

Now we require that the moment of our total force, F, [acting at ( x , y ) ] about the y-
axis be equal to this sum of the elemental moments, i.e.
∫ xpdA = Fx
A
(39)
and we’ll get a similar equation for the moments in the y-direction (moments about
the x-axis):
∫ ypdA = Fy
A
(40)
Equations (39) and (40) can be rearranged to give explicit expressions for the
locations of the center of pressure:
∫A xpdA p ∫A xdA ∫A xdA
= x = = (41)
F p ∫ dA ∫ dA
A A

and
∫ ypdA
p ∫ ydA ∫ ydA
=y A
= = A A
(42)
F p ∫ dA ∫ dA
A A

Notice that p can be taken out of the integral – (put at the front) because it is the same
for all the elemental areas. In this case we see that the centre of pressure corresponds
to the centre of area.
If p varies for each dA, then it must be left inside the integral and we have to associate
the correct value of p with each dA.
The right-most expression in Equations (41) and (42) are the standard expressions for
the centroid of the area with respect to the chosen axes.

SO – the centre of pressure on a horizontal submerged plane area is located at the

centroid of the plane area.
Forces on inclined Planes

In Figure 2.15 we are looking horizontally at an irregularly shaped plane submerged

in a fluid at angle θ to the surface. Dotted underneath the plane is shown a set of axes
and a plan view of the plane, as it would be viewed by somebody looking at the plane
from a position above and perpendicular to it. The origin of the axes lies in the fluid
surface and the x-axis lies in the water surface and also in the plane. The y-axis is
perpendicular to the x-axis and in the plane.
We seek the total force on the plane and the line of action of this resultant.
We divide the surface into infinitesimal horizontal strips (areas) on which the pressure
forces will be constant. These are shown dotted on the diagram, denoted δA, with
thickness, δy. Thus:
=δ F p= δ A γ h=δ A γ y sin θδ A (43)
Where δ F is the resultant force on the thin horizontal strip, δ A , and the other
notation is as before.
The total force is the sum of all similar strips which together make up the area, A.
=
F ∫ pdA
A
= γ sin θ ∫ ydA
A
= γ sin θ=
yA γ=
hA pG A (44)
where y is the distance from the surface to the centroid of the plane measured in the
plane and h is the depth from the surface to the centroid of the plane measured
vertically. pG is the pressure at the centroid.

Centre of Pressure
We define the location of the resultant of this force as ( x p , y p ) and it is designated the
center of pressure.

For an inclined plane the center of pressure is not located at the centroid of the
plane.

To locate the center of pressure, we need to find a location through which the
resultant will cause the same moment about the axes as the sum of the moments
resulting from the elementary forces associated with the pressure applied on every
small or elemental area of the plane. Thus:
x p F = ∫ xpdA (45)
A
and
y p F = ∫ ypdA (46)
A
Note that now we are considering small elements of area which are no longer strips
(although we could use strips to find yp).
From Equations (45) and (44), we get:
1 1 1 I xy
= xp =
FA ∫ xpdA
γ yA sin θ A∫ xγ y=
sin θ dA =
Ay ∫ xydA
Ay
(47)

The last integral in Equation (47) is called the product of inertia and is designated as
Ixy. If you are not comfortable with moments of inertia and centroids have another
look at your notes from Structural Mechanics.
The parallel axis theorem:
I=
xy I xy + Axy (48)
substituted into Equation (47) gives:
I xy
=
xp +x (49)
yA
Note that I xy is the product of inertia for axes through the centroid of the area. When
either of the centroidal axes is a center of symmetry, I xy is equal to zero in which case
from Equation (49), the center of pressure is located at the centroid in the x-direction.
More often we will be interested in the location of the center of pressure in the y-
direction, yp. From Equations (46) and (44) we get:
1 1 1 Ix
= yp =
FA∫ ypdA ∫
γ yA sin θ A
yγ y=sin θ dA =
yA A∫ y 2dA
yA
(50)
The parallel axis theorem gives us:
I=
x I x + Ay 2 (51)
which substituted into Equation (50) leads to:
 Ix
=
yp +y (52)
Ay
Now since the first term on the right hand side must be positive, it follows that the
center of pressure must lie below the centroid for all submerged planes inclined to the
horizontal.
Don’t forget that in these formulae y is measured along the plane not vertically!

The pressure prism

The pressure at any location on an inclined submerged plane is given by
p = ρ gh where h is the depth at that location. If the value of the pressure at each
elementary area on the plane is mapped perpendicular to the surface at some scale, a
volume is described which is called a pressure prism. The volume contained within
the pressure prism is a direct measure of the total force acting on the plane and the
centroid of the pressure prism is the location of the center of pressure.

The volume associated with an element δ A is given by= pδ A ρ= ghδ A δ V which if

integrated over the whole surface, provides the volume of the pressure prism and the
resultant force.
The centroid of the pressure prism is found by computing the moments of the volume
about the axes, thus:
1 1
= xp =
VV ∫ xdV and yp
V V∫
ydV (53)
Comparison with Equations (47) and (50) and recalling that pδ A = δ V , it is clear that
the centroid of the pressure prism and the location of the center of pressure on the
plane are coincident.

Pressure prisms are a convenient way of determining centers of pressure and

resultants for some shapes. For rectangular planes, the method is particularly
convenient.
Example 2.8 from SW&B:

A structure is arranged along the side of a channel so that the L-shaped section will
rotate about the hinge and water will spill if a certain height y is reached. The gate is
made of steel plate weighing 2500N/m2. Determine y.
We work with a unit (1m) width of the section.
There is a pressure force downwards on the horizontal plate causing an anti-clockwise
moment about the hinge and a pressure force to the right on the vertical leg of the L
causing a clockwise moment. In addition there is the moment of the weight of the
steel section itself.
Consider firstly the pressure prism associated with the horizontal plate. Because its
horizontal, the pressure will be constant so that the pressure prism is a rectangular
prism of base area 1.2m by the unit thickness by a height = ρgy. The line of action of
the vertical pressure force is through the centroid of this pressure prism which will be
located 0.6m from the hinge.
On the vertical face, the pressure varies with the depth, so that we end up with a
triangular prism, whose base is ρgy and the pressure reduces linearly to zero at the
surface. The centroid of a triangle is located 1/3 of the height from the base, so the
line of action of the resultant horizontal force is located y/3 from the base.
The moments about the hinge must just balance, prior to an increase in depth, which
would see the clockwise moments dominate and a spill occur. The weight of the
horizontal plate is 1.2x1x2500 = 3000N. Hence the moments anti clockwise are:
γ y2 y
M = (3000 × 0.6) + (1.2γ y × 0.6) − ( × )= 0 (54)
2 3
This equation simplifies to:
M =− y 3 4.32 y − 1.1014 (55)

This can be solved by trial and error or more quickly by using the Newton-Raphson
method.
The answer is 2.196m.
Example:
Determine the pivot location y of the rectangular gate in the figure, so that it will open
when the liquid reaches the level shown.

2m

1m

Solution:
We need y such that the pivot point is at the centre of pressure measured from the
surface, i.e.
1
Ix × 1 × 13
2 − y = yp = y + = 1.5 + 12 = 1.556m
Ay 1.5 × 1 × 1
whence: y = 0.444m.
Hydrostatics 3

Force components on curved surfaces

On curved surfaces, the direction of elemental pressure forces is always normal to the
surface and hence changes direction. In this case, it is necessary to add the resultant
pressure forces as vectors. It turns out that we can determine the horizontal and
vertical components of these resultants in a reasonably straightforward way. In this
lecture, we describe how to do that: to determine the horizontal component of the
resultant force on a curved surface and to determine the vertical component of force
on a curved surface.

Horizontal component of force on a curved surface

We shall prove that:

The horizontal component of pressure force on a curved surface is equal to the

pressure force exerted on the projection of the curved surface onto a vertical plane.

This simple statement allows us to readily compute horizontal components of force on

arbitrary curved surfaces.

The force on the elemental area in Figure 2.21 is pδA acting at right angles to the
surface at that location – i.e. at an angle θ to the negative x direction in the coordinate
system defined in the figure. The component of this force acting in the x-direction will
be:
δ Fx = pδ A cosθ (56)
Now δAcosθ is also the area of δA projected on a vertical plane. It follows that the
horizontal component of this elemental force on the curved surface is exactly equal to
the force on the projection of the elemental area onto the vertical plane. If we sum the
elemental areas over the entire surface then we get;
Fx = ∫ p cos θ dA (57)
A
Exactly the same expression applies to the computation of the pressure on the
projected area so our assertion above is proved.
Notice that if we place a closed body such as a sphere in a fluid we would need to
project the areas on either side to a vertical plane on either side and it is intuitive that
the projected areas would be the same and the associated pressure forces would be
equal and opposite. Consequently:

Closed bodies submerged in a fluid with constant density in the horizontal direction
experience no net horizontal force.

(Why is the limitation on constant density required?)

It also follows from this discussion that the line of action of the resultant horizontal
force component on a submerged curved surface is the same as that of the resultant
force on the projection onto a vertical plane.

Thus we can use the methods established in the previous lecture to determine the
location of the center of pressure of the horizontal component of the resultant, by
using the plane projection.

Vertical component of force on a curved surface

The vertical component of force is equal to the weight of liquid vertically above the
curved surface extending up to the free surface.

The pressure on the element δA is = ρgh. The vertical component of the pressure
force on the element is pδAcosθ = ρghδAcosθ, similar to what we saw for the
horizontal component. Now the volume of the prism above the element, δV, is given
by δV =hδAcosθ, (h is the height, δAcosθ is the area of the cross section). Hence the
vertical component of pressure force on the element is = ρgδV. But this is the weight
of the prism above the area.
It follows we can obtain the resultant by integrating the elemental pressure force over
the entire area:
=Fv ∫=
A
p cos θ dA ∫ ρ gh
A
= cos θ dA γ ∫=
h cos θ dA γ ∫ dV
A V
(58)
where the volume, V is defined as the volume contained by the elemental prisms
above the elemental areas of A.

Thus we see that our assertion is proven – the vertical component of the force is equal
to the weight of the water .

When the liquid is below the curved surface, an imaginary free surface may be
constructed above the surface, provided that the pressure, p, is known at some
location. The imaginary surface is located at p/ρg above the location of the known
pressure.

The line of action of the vertical components is computed in the usual way by
equating the moment of the resultant with that of the elemental forces:
Fv x = γ ∫ xdV (59)
V
and hence:
γ 1
=x = ∫
Fv V
xdV
V V∫
=
xdV because Fv γ V (60)
In other words, the line of action is the same as the centroid of the volume of water
vertically above the surface.
Tensile stress in a pipe and spherical shell
(This is similar to the example of pressure inside a droplet of fluid, due to surface
tension.)

Ignoring longitudinal stresses, in a circular pipe containing a fluid under pressure

there will be tension in the pipe walls as indicated in Figure 2.26. We consider a unit
length of pipe for simplicity, and take a vertical diametric slice through the pipe as
shown in the Figure 2.26.
The horizontal component of force due to internal pressures acts on the projected area
and is 2pr for a unit length of pipe, where p is the pressure at the centerline. If the
pressures are very high relative to atmospheric, or if the fluid is of very low density (a
gas), the center of pressure will be very near the centerline (why?) and for that case
the tensile forces in top and bottom of the pipe walls will be similar. Hence T=pr.

If the pressure differences from top to bottom are more substantial, the location of the
pressure center is computed using:
= T1 + T2 2 pr and = 2rT1 − 2 pry 0 (61)
The first of these equations is the force balance in the horizontal direction and the
second is the moment balance about the base of the pipe section.
Solving these two gives:
= T1 py and= T2 p(2r − y ) (62)
Example:
The 2m long gate shown in the figure is hinged at O. What is the horizontal
component of force acting on the gate and where is its line of action? What is the
vertical component of force and where is its line of action? What force, F, is required
to open the gate, neglecting its weight.

Solution:
We consider a projection of the curved surface on a vertical plane this will be equal
to OC. The vertical distance to the centroid of the vertical plane OC is 4m. The area of
the plane is 4m2. Therefore the force on this vertical plane is 9800 × 4 × 4 =157kN and
the location of its centre of pressure is given by:
1
I xx I xx × 2 × 23
yp = y+ =h+ 4 12
=+ =4.0833m
Ay Ay 4×4
Hence, b=1.0833m.

π × 22
Fv= γ V= 9800[(2 × 3 × 2) + × 2]= 179kN (Note this includes an imaginary
4
rectangle from the line AO up to the level of the water surface)
V = 18.283m 3
To get the line of action we need to get the moments of area of the quarter-circle and
the rectangle about O. The formula for the centroid of a quarter-circle is in Appendix
of Potter and Wiggert. The centroid is 4r /(3π ) from the radial side.
Now if ‘a’ is the location of the vertical resultant from the line OC, then the moment
of the resultant ( Fv × a ) must be equal to the sum of the moment of the rectangle 3m
above OA and the moment of the quarter circle, i.e.:
π × 22 4r
179000 × a= γ (2 × 3 × 2 × 1 + × 2 × ) → a= 0.948m
4 3π

We are now in a position to compute the moment on the gate which must be
overcome by F, i.e. Fv × a − FH × b . The result of this computation is zero! This is
surprising at first but if we look at the pressure forces acting perpendicular to the
circular surface we see that they all act radially, i.e. they all act through the hinge and
therefore have no moment about the hinge. Our computation using the components of
the forces is in fact unnecessary and we can see by inspection that there will be zero
net moment due to water pressure. Thus F = 0.
This is an important feature of circular submerged sections and gates are often
designed using this idea so that excessive forces are not necessary to open them.
Example:
The figure below shows a curved gate which is 1.3m wide. The water behind it is
1.1m deep and the shape of the gate is the cubic as shown on the diagram. What is the
clockwise moment about O, applied by the water on the gate?

Solution:
The vertical force on the gate is given by the weight of the fluid vertically above it.
We will need to compute the area between the curve and z = 1.1m and also the
distance of the centroid of the area from the z-axis, x .
We do this by integration and can choose to integrate in either the x or z direction. To
begin we will integrate in the z direction. Firstly, let’s compute the volume of water.
A = ∫ dA
A

We take the horizontal strip, whose area is dA = xdz

Now if we invert the functional relationship - x = z1/ 3 and substitute, including the
limits of z (we sum dz strips starting at z = 0 up to z = 1.1), we get:

1.1
A = ∫ z1/ 3dz
0
1.1
3
= z4/3
4 0

3
= 1.14 / 3
4

When we write the definition:

1
A ∫A
x= xdA
it is important to realize that the x in the integrand is the ‘lever arm’ of the element dA
about the y-axis and so is the distance from the y-axis to the centroid of the element,
which is actually in this case x/2.
Our equation for the centroid can thus be written:
1 x
x = ∫ ( xdz )
AA2
Now substituting x = z1/ 3
1 z1/ 3 1/ 3
x= ∫ z dz
AA 2
and hence:
1 1.1 2 / 3
2 A ∫0
x= z dz
1.1
z5/ 3
=
2 A × (5 / 3) 0
1.15 / 3
=
2 × (3/ 4) × 1.14 / 3 × (5 / 3)
4
= 1.11/ 3
10

Doing the integration in the other direction (as a check):

We now have :
 1.11/ 3
=A ∫0
(1.1 − x 3 )dx
1.11/ 3
= 1.1x − x 4 / 4
0

= 1.1 4/3
− 1.1 4/3
/4
= 0.75 × 1.14 / 3
..as before.

And for the centroid:

1
A ∫A
x= xdA

1 1.11/ 3
A ∫0
= x(1.1 − x3 )dx
1.11/ 3
1 1.1x 2 x5
= ( − )
A 2 5 0
1 1.15/ 3 1.15/ 3
= ( − )
A 2 5
1 3
= ( 1.15/ 3 )
A 10
4
= 1.11/ 3
10
also as before. (Whew!)

The horizontal pressure force will be the volume of the pressure prism located on the
z-axis in our diagram – its base is ρ=
gh 9800 × 1.1 and its height is 1.1m and width
1.3m. The horizontal force will act 1/3 of the way up the side of the triangle, so its
(clockwise) moment will be:
1.13
9800 × × 1.3 = 2826 Nm
6
The vertical force due to the weight, using our calculations above, will be:
4
γ Vx= 9800 × 1.3 × 0.75 × 1.14 / 3 × 1.11/ 3= 4480 Nm
10
The total moment is 7.3kNm.
Hydrostatics 4

BUOYANT FORCE
The resultant force on a submerged or partially submerged body in a static fluid is
called the buoyant force.

Buoyancy is important in the study of floating objects and their stability. Ships are an
obvious example. Their have been some major problems in the last twenty years,
including a New Zealand ferry disaster, arising from instability of the vessel. Often,
during construction, large objects are floated into place and we need to understand the
forces involved.
Buoyancy is also used effectively in determination of the specific gravity of fluids and
it plays a very significant role in the study of air and water pollution when fluids
carrying pollutants are released into the atmosphere or receiving waters with different
density.

We have already seen that horizontal forces are determined by the projection of the
body onto an imaginary vertical plane and that for a solid body the horizontal forces
applied to the projections on either side will cancel out – leaving a zero net horizontal
force.

Consider a fully submerged body.

In the vertical direction, we again ‘separate’ the body into two sides a top surface and
a bottom surface. Remembering that the resultant vertical force is equal to the weight
of the fluid vertically above the surface, we see that the upwards force on the bottom
surface (ABC in Fig 2.27) is equal to the weight of a volume FABCE, which contains
the volume associated with the top surface force (ADCEF) plus the volume of the
body, ADCB.

From this we conclude that the net pressure force upwards is equal to the weight of
water which would be contained in a volume the same as that of the body. Generally,
this force will be opposed by the body’s gravitational force downwards and if the
pressure force exceeds the gravitational force, the body floats, otherwise it sinks.
In the situation where the body is only partially submerged, it is not difficult to show
that the upwards force is equal to the weight of the volume of fluid corresponding to
the volume of the body which lies beneath the plane of the fluid surface. Look at the
floating body in Fig 2.27 for example. Clearly there is no fluid pressure on the dotted
section, so the part of the body which is directly beneath the dotted line, experiences
an upward force equal to the weight of its volume. Outside the dotted section, we have
a similar situation to the fully submerged body and the net result is that the buoyant
force is equal to the weight of the displaced fluid.

In Figure 2.28 the vertical force exerted on the prismatic volume with cross section
δA is:
δ FB = ( p2 − p1 )δ A =
γ hδ A =
γδ V (64)

Integration of these prisms over the entire volume gives:

= ∫ dV γ V
FB γ=
V
(65)
assuming that γ is constant throughout the volume.
To find the line of action of the force, we take moments of the prisms about some
convenient axis and equate them to the moment of the resultant:
γ ∫ xdV = γ Vx (66)
V
which leads to:
1
V V∫
x= xdV (67)
where x is the distance from the axis to the line of action of the resultant. This
equation is also that for the distance to the centroid of the volume, so that the resultant
acts through the centroid of the displaced volume of fluid.

The centroid of the displaced volume is called the center of buoyancy.

The buoyancy force given by (65) will exceed the weight of the object if the average
density of the system (its total mass divided by its total volume) is less than that of the
fluid in which it is submerged.

Weighing an irregularly shaped object suspended in two fluids of known differing

densities provides enough information to calculate the object’s weight, volume,
specific weight and specific gravity.

The equations for the situation in Figure 2.29 can be written mathematically as:

+Vγ1 W
F1= +Vγ 2 W
F2= (68)

which can be rearranged to:

F1 − F2 F1γ 2 − F2γ 1
=V = W (69)
γ 2 − γ1 γ 2 − γ1
Hydrometers use the principle of buoyant force to determine specific gravities of
liquids.

The hydrometer is a carefully weighted hollow glass or plastic bulb with a cylindrical
stem as shown in Figure 2.30. The stem is calibrated and the weights in the
hydrometer adjusted so that in a liquid with S=1.0, the reading ‘1.0’ appears at the
surface. For this situation the force balance is:

V0γ = W (70)
where V0 is the submerged volume in the LH diagram.
In some other liquid of higher density, the hydrometer will float higher as shown in
Figure 2.30 on the right. If the calibration shifts by ∆h, as shown and the area of the
cylindrical section is given by a, then for the RH diagram we would have:
(V0 − a∆h ) Sγ =W (71)
These two equations can be solved to give:

V S −1
∆h = 0 (72)
a S
 STABILITY OF FLOATING AND SUBMERGED BODIES
We have seen that the centre of buoyancy acts at the centroid of the displaced volume
of liquid. This is the line of action of an upwards directed force. If the center of
gravity of the object itself lies below the center of buoyancy, then an angular
displacement of the objects sets up a restoring moment (Figure 2.32).

For a fully submerged object as shown in Figure 2.32, this is an essential condition for
stability. For a floating object, we shall develop a less restrictive criterion.

Figure 2.33(a) shows a floating prism in which the centre of gravity is located above
the centre of buoyancy. The same prism is shown with a rotational displacement of θ
in Figure 2.33(b). Notice that now the displaced volume is no longer prismatic and so
the centre of buoyancy must move to the centroid, B1 of the trapezium ABCD. If
B1lies to the left of the centre of gravity (G), then a restoring moment is set up and the
prism is stable. Geometrically, this means that the vertical line through B1, intersects
the vertical axis of symmetry at M and M must lie above G for stability. M is called
 the metacentre and the distance MG is called the metacentric height. The restoring
couple is W MG sin θ , for a prism of weight W.

Example

Consider a prism with the following dimensions: 6m wide x 20m long, with a gross
weight of 200tonnes. Its centre of gravity is 0.30m above the water surface. Find the
metacentric height and the restoring couple when one side dips by 0.30m (= ∆y).

The depth of submergence in the water is:

200000 × 9.8
=h = 1.667
6 × 20 × 9800
The centroid in the tipped position is located with moments about AB and BC,
1
1.37 × 6 × 3 + 0.6 × × 2 × 6
x = 2 2.82m
1.67 × 6
1.367 1
1.367 × 6 × + 0.6 × 6 × × (0.2 + 1.367)
y = 2 2 0.842m
1.667 × 6

By similar triangles, AEO and B'PM :

∆y PB′
=
b / 2 MP
Hence,
0.18 × 3
=
MP = 1.8m
0.3
G is h + 0.30 above the base = 1.97m; P is 0.842 above the base, hence:

GP = 1.97 - 0.842 = 1.128m

And MG = MP − GP =1.8 − 1.128 =0.672m

So the scow is stable because MG is positive; the righting moment is:

0.3
θ 200000 × 9.8 × 0.672 ×
W × MG × sin= = 131kNm
32 + 0.32
 Further examples of buoyancy related problems

Example:
A timber 600mm*600mm in section, S=0.5, has a concrete 600mm side cube, S=2.5,
attached to one end.
1) What is the length, L, of wooden cylinder for the system to float in water?
2) For a timber section in which L = 4.0m, is the centre of buoyancy above the centre
of gravity?
3) Is the system in stable equilibrium with the axis vertical?

Wood M

B H
LM G
LB LG
600

Concrete
Solution:
Let the length of the wooden cylinder be L. In order for the system to float, the system
will be just submerged and the buoyancy force = weight of the system:
FB =H × 0.6 × 0.6 × 9800 =(2.5 × 9800 × 0.63 ) + (0.5 × 9800 × [ H − 0.6] × 0.62 )
From which H = 2.4m and L = 1.8m.
Alternatively, for the average density of the whole system to be less than the density
of water, L must be such that the average density of the system, i.e. the mass of the
concrete plus the mass of the timber, divided by their combined volume, must be less
than 1 :
(0.5L*A + 2.5*0.6*A)/[A*(L+.6)] < 1. i.e. L > 1.8m.
For L = 4m, the total weight of the timber is 0.5 × 9800 × L × 0.62 = 7056 N .
The weight of the concrete is 2.5 × 9800 × 0.6 = 3
5292N , giving a total weight of
12348N. The buoyancy force, 9800 × H × 0.6 equals this weight, whence, H = 3.5m.
2

The centre of buoyancy is at the centroid of the rectangular prism of displaced water,
i.e. it is located 1.75m form the bottom of the concrete.
The centroidal distance, LG, of the system, measured from the bottom of the concrete
is determined by taking moments about the bottom, thus:
1
=LG [5292 × 0.6 / 2 + 7056=
× 2.6] 1.614m
12348
So the answer to the second question is yes, the centre of buoyancy is above the centre
of gravity of the floating system and furthermore the answer to the third question is
also clear – the system is like Fig 2.32(a) – it is stable.
LM in the figure is the metacentric height and must be computed for stability analysis
in situations where LB is less than LG.