BENDING – SLOPE AND DEFLECTION

The bending equation revisited

Recall the bending equation

M E σ
= = .
I R y

M E
We can use the first equation, = (1)
I R
to find slopes and deflections. First,
express R in relation to the slope θ. The
diagram is of a section of beam bent into a
circular arc. The angle between the radii,
dθ, is equal to the change in the angle with
the horizontal θ of a tangent to the beam.
For an arc length s, we can use the relation
‘s = rθ’ to suggest an expression

‘ds = R dθ’

which relates to the diagram. However,

note that θ is decreasing, and s increasing,
dθ
so that the derivative is negative.
ds
Therefore, when we re-arrange the above expression to get the derivative, we must
take account of the sign and write

1 dθ
=− (2).
R ds

The deflection v and the slope θ are shown relative to the standard axis set in the
x diagram. The slope of the beam is
v equal to the gradient
dv
dx
. The gradient

θ is also equal to the tangent of the angle

dv
tan θ = . For a small angle,
dx
tan θ ≅ θ and we may write
dv
θ= (3).
y dx
The assumed small angle of deflection
also allows us to equate the arc length s with the co-ordinate x: s = x. We can
therefore rewrite (2) as

1 dθ
=−
R dx

and now use (3) to conclude that

1 d2 v
=− 2 (4).
R dx

Now from equation (1),

1 M
=
R EI

and we may rewrite (4) as

d 2v M
2
=− (5).
dx EI

Equation (5) is a differential equation in the deflection v, which can be solved to give
dv
both v and the slope .
dx
Example

Determine the deflection at every point of the cantilever beam subject to the single
concentrated force P, with section modulus EI.

Solution

We need to use equation (5), so must first find the bending moment M. At the left-
hand supported end, there is both a reaction P and a moment acting. It is therefore
simpler to look at the moment produced by the force P at the right-hand end. For the
point x, the moment M produced by the force P is given by

M = P(L-x).

We can now use (5):

2
 d 2v P( L − x ) P( x − L )
2
=− = .
dx EI EI
Integrating the above gives

dv P  x 2 
=  − Lx  + C1 (E1)
dx EI  2 

where C1 is a constant of integration. The equation above gives the slope of the beam.
To obtain the deflection, integrate again:

P  x3 x2 
v=   − L  + C1x + C 2 (E2)
EI  6 2

where C2 is another constant of integration. To complete the solution, we need to find

the unknown constants C1 and C2 . These are obtained from the boundary conditions.

In this case, there are two boundary conditions needed to find the two constants.
They both apply to the left-hand end of the beam. Firstly, the left-hand end is built-in,
so the slope is constrained to be zero. This gives us a condition involving equation
(E1):

P  x2 
0=  − Lx  + C1 .
EI  2 

At that left hand end, x = 0, and so the condition reduces to

C1 = 0.

The other condition is that the deflection is zero at x = 0. This gives us, using
equation (E2),

P  x3 x2 
0=  − L  + C1x + C 2
EI  6 2 

which when x = 0 gives C2 = 0. Putting these values back into (E2), the final solution
is
P  x3 x2 
v =  − L 
EI  6 2 

i.e.

Px 2  x L 
v=  − .
EI  6 2 

3
Complex loading

The above method is called the ‘double integration method’. It is adequate for
relatively simple problems, but is clumsy for anything other than simple loading. The
following example illustrates the difficulty – finding the deflection at all points along
this ‘overhanging beam’.

First we must find the bending moment along AB. For the point x, this is given by

wx 2
M AB = − (M1).
2

Using (5) it follows that, on AB,

d 2 v wx 2
EI = .
dx 2 2

Integrating as before gives

dv wx 3
EI = + C1 (6)
dx 6

and
wx 4
EIv = + C1x + C2 (7).
24

4
For the other region of the beam BC, there is a different equation for the moment.
Assuming a point x now within the region BC, the bending moment is now

wx 2
M BC = − + R 1 (x − a ) (M2)
2

where we have left the reaction as R1 for now. This can now put into (5):
d 2 v wx 2
EI 2 = − R 1 (x − a )
dx 2

and integrated to give

dv wx 3  x2 
EI = − R 1  − ax  + C 3 (8)
dx 6  2 

and again to give

wx 4  x 3 ax 2 
EIv = − −  + C3 x + C 4 (9).
24  6 2 

We have introduced two more constants of integration, making four in all. We

therefore need four boundary conditions to evaluate them. These are
• In overhanging region AB, v = 0 when x = a.
• In region BC, v = 0 when x = a.
• On BC, when x = L, v = 0
dv dv
• At B, slope for AB = slope for BC –i.e. no sudden kink.
dx dx
Though possible, the procedure to find the solution for the two unknowns would be
tedious, and it is clear that with yet more complex loading the method will become
impractical. These difficulties can be avoided by using Macauley’s method.

Step function / Macauley’s method

We must first introduce the step function, denoted by the brackets ‘ ’. For a
variable x - a, we define

n ( x − a ) n for x − a ≥ 0
x−a =
0 for x - a < 0

where n is any real number. This enables us to write down the bending moment for a
beam such as the overhanging beam discussed above in a single equation. Equations
(M1) and (M2) above can be put together into a single equation:

wx 2 1
M=− + R1 x − a (10)
2

5
the inclusion of the exponent 1 being optional. The crucial property of the step
function is that it can be integrated just like an ordinary pair of brackets:

x2
x2  x − a n +1 
∫ x − a dx =  
n

x1  n + 1  x
(11).
1

We now continue to solve the overhanging beam problem. Using equation (10) in (5),
we have

d 2 v wx 2 1
EI = − R1 x − a
dx 2 2

which on integrating using (11) becomes

2
dv wx 3 x−a
EI = − R1 + C1 (12)
dx 6 2

and again

3
wx 4 x −a
EIv = − R1 + C1 x + C2 (13).
24 6

Equations (12) and (13) apply for the whole beam and only involve two constants.
We can find the constants by applying the boundary conditions:
• v = 0 at x = a.
• v = 0 at x = L
Using these values in equation (13) gives

wa 4
0= + C1a + C 2 (14)
24
3
wL4 b
0= − R1 + C1L + C 2 (15)
24 6

We now need to solve these for C1 and C2 . Subtract (14) from (15) to eliminate C2 :

0=
w 4
( )
L − a 4 − R1
b3
+ C1 ( L − a )
24 6

where the step function has been evaluated for positive b. With L – a = b, dividing
through by b gives

b2 w
C1 = R 1 − ( L4 − a 4 ) .
6 24b

6
Now use (14) to get C2 :

wa 4
C2 = − − C1a .
24

Finally we find R1 by taking moments about C to give

wL2
R1 = .
2b

Putting in this value gives

wL2 b w
C1 = − ( L4 − a 4 )
12 24 b
4
.
wa wL2ab wa 4
C2 = − − + (L − a )
4

24 12 24b
To get the slope and deflection, these constants are inserted into equations (12) and
(13).

Example. Use Macaulay's method to determine the equation of the deflection curve
of the simply supported beam as shown.