STRAIN ENERGY

1
When a structure is subjected to loads it will deflect and the points of application of the
loads will be displaced. This means that the loads will do work and it has already been
seen that the work done by a gradually applied load W which is displaced through a
deflection  is given by the expression ½ W.

If the loads are removed and the structure remains elastic the points of application of the
loads will return to their original positions, which means that work is now done on the
applied loads. This implies that the original work done by the loads is stored in the
structure and is available to restore the loading points to their original positions.

The work or energy stored in the structure is known as STRAIN ENERGY and is given
the symbol U.

It is now necessary to develop expressions for this strain energy in the case of both a
truss or pin-jointed frame, and a beam.

EXPRESSIONS FOR STRAIN ENERGY

Strain energy due to direct stress

A bar cross-sectional area A length I extends l on application of gradually applied force

F (i.e. average force while bar is extending = F/2) see Fig 1.

Cross sectional
Area A
ℓ

dℓ

Fig. 1

load change in length

Modulus of elasticity E = 
area original length
F l
 
A 

Fl
 l 
AE

2
Now:

strain energy stored = work done

= average force x distance moved

F
=  l
2

F Fl
 strain energy U = 
2 AE

F 2l

2 AE

In a framed structure containing many “bars” (struts and ties) total strain energy due to
direct stress is

F 2l
U 
2 AE

Strain Energy due to bending

An element of beam length x rotates through angle  on application of gradually applied

moment M: from Fig. 2.

M f E
 x  R and  
I y R

x M x
  
R El

3
Now:

strain energy stored = work done

= average moment x angle turned through

M
= 
2

dx 

R
M

Fig. 2
W
Example: Deflection of a beam
A B C
ℓ/2 ℓ/2

Find the mid-span deflection of the constant El beam carrying a central point load of W

M 2 dx
1
2 W  
o 2 El

As the expression for the bending moment is different for A – B and B – C it will be
necessary to deal with each of these sections separately and carry out the integration in
two steps:-
Wx
M W x  
A B A C 2  2
B
w/ x w /2 x Wx W
2
  Wx 
2 2
W
 (  x)
2

4
Section M M2  M 2 dx

/2
Wx W 2 x2 /2 W 2 x2 W 2  x2  W2  3  W2 3
AB
2 4 
0 4
dx 
4  3 

4  24   96
 
0
2
W2  2 x3 
 
2
BC
W
2
  x
W
4
2
2 x x 2 W
 /2 4  2
 2 x  x  dx 
2

4 
x x  
2

3 /2

W 
2 3
  3 3 3
 
3
 3
      
4  3   2 4 24  

W2  3 7 3  w2 3 w2 3
  3  24   4  24  96
4  

M2 2W 2 3 W2 3
 for whole beam U   dx  
0 2 EI 2  96 EI 96 EI
2
1 W W2 3
But W    dx 
2 0 2 EI 96 EI
2 3 3
W W
W  or    a standard result
48EI 48EI

Note: In the above calculation the strain energy stored in both AD and E proved to be
WI 3
. As the beam is symmetrical in both properties and loading it is pretty
2  96 EI
obvious that the strain energy stored in each half must be the same, therefore it is not
necessary to carry out a separate integration for BC, merely to say that the total strain
energy stored in the beam is twice that stored in AB, i.e.

/2 M2 W2 3
U  2 dx 
o 2 EI 96 EI

Note that the integration for BC would also have bean simpler if X was measured from
support B, as this would give an identical expression for the bending moment as in AB.

M M x
Strain energy U  
2 EI

M 2 x

2 EI

5
 L M2
Total strain energy due to bending in a beam is U  x
0 2 EI

Wd M 2 dx
and for beam  
2 0 2 EI

Example: Deflection of a pin-jointed truss

If Youngs Modulus for all members of the pin-

30 A B
jointed truss shown is 200 GPa and all members
have a cross section 20 mm square, find the
vertical deflection of point C.

1 F2
4
Solution WD 
2
 2 AE

E  200 GPa  200 106 KN / m 2

30 C A  20  20  400 mm2  400 106 m2
D
AE  400 106  200 106  80 103 KN
40 kN

member 𝓵m F F2𝓵
AB 3 0 0
BC 4 0 0
CD 3 -30 2700
AC 5 50 2500

F 2 15200 W
 F 2  15200  
2 AE 2  80 10 3
 0, 095 
2

0, 095  0, 095  2
    4, 75 mm
W 40 103

In this case A and E was the same for all members so the division by AE can be done
after the summation. However, if say A had been different for each member then an extra
F2
column for A would have been necessary in the table plus another for and the
A
summation of this.

6
In both of the examples dealt with there was just a single vertical load acting on the
structure and the vertical deflection of the point of application of the load was calculated.

W1 Point C in the truss will also have a horizontal deflection, but it

appears it is not possible to calculate this at the present state of
knowledge of Strain Energy. Also if two loads acted on the truss
as shown in the sketch, then while the forces F in all the
members can quite readily be found and the Strain Energy in
F2
the structure  can also be calculated, when the energy
W2 2 AE
1 1 F2
equation is written, i.e. W11  W2  2   it is found that
W1 W2 2 2 2 AE
there are two unknowns 1 & 2 and only one equation and so 1
& 2 cannot be found.

Similarly if two loads act on the beam, while the bending moment M can be written for the
M 2 dx
three sections and the Strain Energy  calculated, once again there will be two
0 2 EI

unknowns in the equation which therefore cannot be solved.

Thus it appears that the Theory of Strain Energy requires further refinement in order to
deal with multiple loads and/or deflections in any direction.

CASTIGLIANO’S FIRST THEOREM

Consider any structure acted upon by say three point loads, W 1, W 2 and a general load
W.

The vertical deflections at the points of application of these three loads are Y1, Y2 & Y3
respectively.

W1 W W2

Y1 Y Y2

The work done by these three loads and hence the Strain Energy stored in the structure
will be given by the expression:-

U  12 W1 y1  12 W2 y2  12 Wy  1

7
It is now assumed that load W increases to W + SW. The deflections under all three loads
will therefore increase as follows:-

y1 becomes y1 + Sy1
y2 becomes y2 + Sy2
y becomes y + Sy

Loads W 1 W 2 and W are already in position when this increase in deflection occurs and
the work done by these loads during the increase in deflection will therefore be the load
times the increase in deflection, NOT half the load times the increase in deflection as is
the case with SW only.

Thus the increase in strain energy during the application of SW will be

Su  W1Sy1  W2 Sy2  WSy  12 SW Sy

1
Sw Sy is a second order differential and can be ignored; thus:-
2
Su  W1Sy1  W2 Sy2  WSy  2

Presuming now that the three loads W 1, W 2 and W + SW were put in position right from
the start, then the strain energy in that case would be:-

u  Su  12 W1 y1  12 W1Sy1  12 W2 y2  12 W2 Sy2  12 Wy  12 Swy  12 WSy  12 SwSy

1
Once again 2 SwSy is a second order differential and will be ignored, giving:-

u  Su  12 W1 y1  12 W1Sy1  12 W2 y2  12 W2 Sy2  12 Wy  12 WSy  12 SWy  3

Subtracting 1 from 3 gives:

u  Su  12 W1 y1  12 W1Sy1  12 W2 y2  12 W2 Sy2  12 Wy  12 WSy  12 SWy  3

u  12 W1 y1  12 W2 y2  12 Wy

Su  12 W1Sy1  12 W2 Sy2  12 WSy  12 Swy  4

Dividing Equation 2 by 2 gives:-

Sv 1 1 1
 W1Sy1  W2 Sy2  WSy  5
2 2 2 2

8
and subtracting this from Equation 4 gives:=

Su  12 W1Sy1  12 W2 Sy2  12 WSy  12 Swy  4

Sv 1 1 1
 W1Sy1  W2 Sy2  WSy  5
2 2 2 2

Su Swy Su
 or y
2 2 SW

u
And in the limit as SW  0 y
w

This is known as Castigliano’s First Theorem (generally abbreviated to Cast I) and when
put into words is as follows:-

The displacement of the point of application of any load, along the line of action of the
load, equals the partial differential coefficient of the total strain energy with respect to
that load.

It should be noted that in the derivation of the Theorem no restriction whatsoever on W,

which can be in any direction and can have any value including zero. Thus the theorem
can be used to find the deflection of a point where no load acts, or to find the deflection
at any angle to a load which does act.

For example consider the framework already studied:-

The 40 kN load at C is now replaced by W

A B Member ℓ F F2ℓ
AB 3 0 0
BC 4 0 0
CD 3 3W 2YW 2
4 16
AC 5 5W 125 w2
4 16

152w2
F 2 
16
C 152W 2
D U 
16  2 AE
W

u 304W
Cast I   But W  40
w 16  2 AE
304  40 103
   4.75 mm as previously.
16  2  80 103

9
This does not save much work as compared with the previous direct application of strain
energy, but it is now possible to find the horizontal deflection of point C.

The 40 kN vertical load at C is replaced, but now an imaginary horizontal load W is also
placed at C and the forces in the members found once more:-

A
B
Member ℓ F F2ℓ
AB 3 0 0
BC 4 0 0
CD 3 -(30 + W) 2700 + 180W + 3W 2
50 AC 5 50 12500

 F2ℓ = 15200 + 180W +3W 2

-(30 + W)
F 2 15200  180W  3W 2
D C
W U  
2 AE 2 AE
40

u 180  6W
Cast I   But W  0
w 2 AE

180 180 103 180

     1.13 mm
2 AE 2  80 10 160
3

The deflection of, say, point B in any direction could have been found by placing an
imaginary load W in that direction at B, calculating the forces in the members due to all
loads including W, calculating the strain energy, differentiating it with respect to W and
putting W in the final expression equal to zero.

This technique is known as the Partial Differential Method.

W2
Consider the truss shown in the sketch.

It is required to find the deflection of point vertically W1 W3

downwards under the loading shown.

All the members of the frame will have a force F

produced by the actual loading W 1, W 2 and W3 and
these can be found by normal methods of truss
analysis. x

10
An imaginary vertical load W is now placed at point X and this will also produce a force in
each member of the truss. This force will be designated fW, which means of course that
f would be the force in that particular member is a unit load was placed at X.

The total force in any member now = F + fW

( F  fW ) H 2
u
2 AE

v 2( F  fW ) f
Cast I   
w 2 AE

Ff f 2w
 
AE AE

Ff
But W=0  
AE

Where it will be remembered F is the force in any member due to the actual real life
loading acting on the truss and f is the force in any member due to a Unit Load acting at
the point where the deflection is required in the direction of that deflection.

For example using this method to calculate the horizontal deflection of C in the frame
already dealt with:-
a) Calculate forces F due to actual loading
3

A
B
50

4
-30

40

b) Calculate unit load at point C & calculate f

D C

40 kN

-1 1

11
 ℓ Ffℓ
c) Member F F
90 103
AB 3 0 0 0 
BC 4 0 0 0 AE
CD 3 -30 -1 90
90 103
AC 5 50 0 0 
Ffℓ = 90 80 103
 1.13 mm as previously.

A simply supported frame AB carries a single concentrated load of 160 kN as shown in

Figu. Calculate the horizontal displacement of point C. (E = 210 kN/mm2. Cross-sectional
area of all members 5,000 mm2.)

3m 3m 3m 3m

A 1 2 3 4 B

1.5 m
8 9 10 11 12 13 14 15
Roller
5 6 7

1
1.5 m 3m 3m 3m 1.5 m

12 m

Figure 2
SOLUTION
1. Find the forces, F, in the members due to the loading.

160

-40 -120 -200 -120

-402
-402 402 402

-402 -402
402 1202

80 160 240

12
2. Apply unit load at point C acting horizontally and calculate the forces, , in the
members due to this unit load.

-7/ -5/ -3/ -1/

8 8 8 8

-2/8 -2/8 -2/8 -2/8

2/8 2/8 2/8 2/8

3/ 1/ 1/
4 2 4
1

3. Calculate FL/AE for each member and sum up. (Since A and E are constant in
this question, they may be applied to the sum of FL.)

In problems involving a summation a tabular method is generally simplest.

Compression  -Ve
Tension  + Vc

Member L(m) F  E L
1 3 -40 -7/8 105
2 3 -120 -5/8 225
3 3 -200 -3/8 225
4 3 -120 -1/8 75
5 3 +80 +3/4 180
6 3 +160 +1/2 240
7 3 +240 +1/4 180
8 3/2 +402 -(2)/8 -152
9 3/2 -402 +(2)/8 -152
10 3/2 +402 -(2)/8 -152
11 3/2 -402 +(2)/8 -152
12 3/2 +402 -(2)/8 -152
13 3/2 -402 +(2)/8 -152
14 3/2 -1202 -(2)/8 452
15 3/2 +1202 +(2)/8 452
 + 1330
 + 1330
E = 210 kN/mm2 A = 5000 mm2

1,330 103
Horizontal displacement of . C   1.27 mm to the left.
210  5, 000

13
(Note: If the result had been negative it would mean deflection was to the right.)

Application of Castigliano I to beams

If the benching moment at a section in a beam is M under the _____ load, and the moment
at the same section is mw under a dummy load of w magnitude. Then, moment in the
section will be M + mw for an entire number.

( M  mw)2
U  dx
0 2 EI

U 2( M  mw) Mm m2 w
  dx   dx   dx
W 0 2 EI 0 2 EI 0 EI

Mm
As before    dx where  is the _______ at the point of application of the
 EI
dummy ( w  o)

Example : Deflection of a beam:

Draw the bending moment diagram for the beam ABCD and calculate the vertical and
horizontal deflection of point D. IAD = 96 x 106 mm4, IBC = ICD = 48 x 106 mm4. E for
allmembers = 210 kPa.

Solution

(a) Bending Moment Diagram (drawn on tension side of each member)

4
150
A B
B
A
90

90
D C
D C
1.5
90

60 kN

14
 Expressions for
B bending moment M
D C M = 60  M = 90 due to 60 kN load


C
60 
90

60

D C M1 =  M1 = 1.5 Expressions for

bending moment M1
due to unit vertical
C
1 load at D
1.5

A B
1.5
 M1 = -1.5 + 
1

B
1 D C M2 = 0 M2 = -


C
1
1
Expressions for bending moment M2 due to unit
2 M2 = 2 horizontal load at D

15
 u Mm1dx
Vertical Deflection v 
W1
  EI

Let 48 10 m4  I then 96 106 mm4  2I

DC M  60 x m1  x

u 1.5 60 x
2
1 1.5 67.5
 dx   20 x3  
 
W1 0 EI EI 0 EI
u 2 135 1 270
CB M  90 m1  1.5  dx  135x  0 
2

w1 0 EI EI EI

u 4  90  60 x  1.5  x 
BA M  90  60 x m1  1.5  x  dx
w1 0 2 EI

0 135  180 x  60 x  dx
1 4 1 4
 2
 135 x  90 x 2  20 x3 
2 EI 2 EI   0

190

EI
u 67.5  270  190 527.5
 Total  
w1 EI EI
527.5 103
v   52.3 mm
210 106  48 106

u Mm2 dx
Horizontal Deflection h  
W2 EI
Mm dx
DC M  60 x m2  0  EI2  0
Mm2 dx 1 2 1 2 2
CB M  90 m2   x  EI

EI 0 90x dx  EI 45x 
 0

180

EI

16
 u 1
0  180  120 x  dx
4
BA M  90  60 x m2  2 
w2 2 EI
4
1  120 x 2  240 120
  180 x    
2 EI  2  0
2 EI EI

u 120  180 60 103

    5,95 mm
w2 EI 210  48
The Application of Strain Energy to Redundant Structures

Consider the following pin-jointed truss:-

Members AC and BC are independent of each other and are not connected at their
intersection point.

This structure has one more member than is required for structural stability and is
therefore one-degree redundant.
A A I
t
can
be
C C
(i) (ii)

made statically determinate by removing, in this case, any one or the members as
follows:-
Here member AC will be taken as the redundant member and it will further
be assumed that member AC was fabricated  too short, i.e. the initial lack of fit of AC = .

The structure is now presumed to consist of two separate parts, (i) the statically
determinate frame with AC removed and (ii) the member AC itself.

17
When AC is stretched to make it fit in position a tensile force will be developed in it and
this tensile force will be called W. The same force W will obviously act at points A and C
on the statically-determinate frame.

A If U is the strain energy in the whole frame including AC, U1 the

strain energy

in the statically-determinate frame without AC, and U11 the strain
energy in
C
member AC on its own, then:-
C
U = U1 + U11

and partially differentiating w.r.t. W:-

v v1 v11
  1
w w w

Points C1 (at the end of member AC) and C (on the statically-determinate frame) will have
to move towards each other by an amount  in order to fit the frame together, and this
closing movement will be made up both of the movement of C inwards () and the
movement of C1 outwards due ‘to the stretch in member AC ().

thus:- =+  2

But, by Castigliano’s First Theorem:

v1 v11
  
ow w

v1 v11
  
w w

v v1 v1
w But from 1  
A w w w
w A
v
 
w w
C
C

U1 w
U11

18
This is known as Castigliano’s Second Theorem, commonly referred to as Castiqliano II,
and which con be put into words as follows:-

P1
the partial differential coefficient of the total strain energy of
a redundant structure with respect to the force in the
redundant member is equal to the initial lack of fit in that
member.

P2
A special case of Castigliano’s Second Theorem arises
where there is no initial lack of fit of the redundant
member(s), i.e.  = 0.

v
P1 The above equation then becomes:- 0
w

and this is known as Castigliano’s Theorem of Least Work.

The Practical Application of Gastigliano’s Second Theorem

P2

Forces F
Consider the frame previously dealt with, which is now carrying
external loads P1 and P2. It is required to find the force in all the
members.

Forces fW
The redundant member, which has an initial lack of fit , is removed
to make the frame statically determinate. The frame can now be
analysed by ordinary statics to find the forces in all the members.

Let the force in any typical member be designated F.

The real loads on the structure are now imagined to be removed and a load W is placed
on the frame at the points where the force in the redundant member would act, and once

19
again the frame is analysed by static to find the forces in all the members under this
loading.

The force in the same typical member under this loading condition is designated fW,
where f would be the force in the member if a unit load acted in place of W.

The total force in the typical member therefore:-

F1 = F + fW

and the strain energy in the typical member:-

( F  fW )2
2
F1
 
2 AE 2 AE

The total strain energy in all the members of the structure (including the redundant
member) will therefore be given by:-

( F  fw)2
u
2 AE

v 2( F  fw) f Ff f2 w
and   
w 2 AE AE AE

v
But by Castigliano II  
w

Ff f2 w
  
AE AE

   FfAE
or w
 f2
AE

and having found W from the above expression the forces in all the members can be
found from F1 = F + fW.

Member ℓ F F Ffℓ f2ℓ w = 25 w = 52.2

F + fW(a) F + fW(ℓ)
(kN) (kN)
AB 3 30 -0.6 -54 1.08 15 -1.32
BC 4 20 -0.8 -64 2.56 0 -21.8
CD 3 0 -0.6 0 1.08 -15 -31.3
BD 5 -50 1 -250 5 -25 2.2
AC 5 0 1 0 5 25 52.2

20
  = -368 14.72

   FfAC
W f2
(a) if   0 and AE the same for all members.
 AE

Ff 368
then W    25 kN (Tension)
f 2 14.72

(b) Where there is a lack of fit of the redundant member it is necessary to select a sign
for this. If the force in the redundant member is assumed to be tensile, as is
generally done and has been here, then a lack of fit where the member is too short
is taken as positive and where a member is too long is taken as negative. If for any
reason it is assumed initially that the force in the redundant member is
compressive, then where the member is too long would be a positive lack of fit and
too short a negative one.

The final sign of W obtained from the above equation will show if the force in the
redundant member is in the sense assumed or not.

It must also be remembered that the lack of fit in the above equation must be
expressed in meters in order to be consistent with the units of the equation.

Therefore the force in AC was assumed to be tensile and the member is 5 mm too
short. Therefore  = + .005 or 5 x 10-3.

   FfAC 5 103  368

80103
W f2


14.72
AE 80103

400  368
 by 80 103 :  W  52.2 kN (Tension)
1472

21
Therefore the procedure for determining the forces in a singly redundant plane frame is
as follows:

1. Remove any one member and determine the forces in

20 kN the statically determinate frame F.
A B
2. Remove external loading apply unit load in place of the
member removed and determine the forces in the frame
4
.
C
D
3. Calculate F L/AE and 2L/AE for each member and
20 kN sum up in tabular form. A sign convention for tension and
3 compression must be used.

4. Calculate the force in the selected member

(a) (b) 1.32

15

25
52.2 21.8
2.2

15 31.3

  ( F fL / AE )
W
( f 2 L / AE )

5. Find the force in all other members from (F + w) for each member.

Example: Find the forces in all members of the frame

A = 400 mm2 & E = 200 GPa for all member.

a) If  = 0
b) If AC is 5 mm too short

22
 AE = 400 x 10-6 x 200 x 106
= 80 x 103 kNM2 for all members

Step 1 Step 2

The plane frame shown in Fig. is pin-jointed throughout. All bars are of the same material
and have a constant ratio of length to area of cross-section. Determine the loads in the
members of the frame due to the given loading.

20
30 A -0.6 B
A B
0.8 1
20 1 52.2 -0.8 Example
-50
0.8 1
D C
D -0.6 C
20
Forces F Forces f

23
 1,5 m 1,5 m 1,5 m

A B C D

1,5 m
50 kN 100 kN

E F

1,5 m
G H

1,5 m
J K
o o

1. Remove redundant member BF and calculate forces in the frame as shown.

A B C D

502 502 1002

50 kN -50 100 kN
100
E F

150

G H

150
J K
o o

0 150 kN

2. Remove external loading and replace BF with unit load. Calculate forces in this
frame as shown.

24
 A B C D

1 1/
1/ 2
2 1
1

E F
1/
2

G H

J K
o o
3. Summation

Note: is constant for all members.

AE

 only introduced at end of calculation.

Member F  F 2  F0 + W
(kN)
AB 50 0 -50
BC 50 -1/2 -50/2 1/2 -100/4 -75
CD 100 0 -100
EF -100 -1/2 +100/2 1/2 -100/4 +75
GH 0 0 0
BE 0 -1/2 1/2 -100/4 -25
EG 0 0 0
GJ 0 0 0
CF -50 -1/2 +50/2 1/2 -100/4 +25
FH -150 0 +150
HK -150 0 +150
AE -502 0 +502
BF 0 1 1 +35.35 +3,535
EC 502 1 +502 1 +35.35 -252
GF 0 0 0
JH 0 0 0
FD -1002 0 +1002
 + 100 2 4

4. W = -(1002)/4 = -3535 kN.

25
5. The calculation of the loads in all members is shown in the last column of the table.

Example

Fig. shows a two-hinged arch structure with a horizontal span of 12 m, between hinges
which are not at the same level. Determine the horizontal thrust at the hinges when the
arch carries a vertical central load of 100 kN. The cross-sectional area and the material
of each member of the structure is constant.

100 kN

3m 3m 3m 3m

2 3

8 9 10 11

3m
1 6
4
5
7
3m
1,5 m

12 m

Solution

This is a similar type of problem, but instead of taking out a redundant member the
horizontal component of reaction H is removed to find F and then unit load applied
horizontally at the supports to find .

Then as before

F0 fL / AE
H as  0
f 2 L / AE

1. Forces F

26
 100 kN

50 50
50/
50/
2 2

50/ 50/
2 2
50
50 100

H (removed)

50 kN
H (removed)

50 kN

2. Forces 

1 3 /8 1 1 /8
1/
8 /2
1/
8 /2
1 3 /8 1 1 /8
2 1 /4

2
13/2

1/
1 8

1/
8

3. Summation

27
 AE Constant
Member L (m) Fo  F0L 2 2L
1 7.5 -50 11/8 -515.6 1.89 14.2
2 6 -50 11/8 -412.5 1.89 11.3
3 6 -50 9/8 -337.5 1.26 7.6
4 6 -50 9/8 -337.5 1.26 7.6
5 32 0 -2 0 2 8.5
6 6 100 -9/4 -1350 5.07 30.4
7 1.513 0 -(13)/2 0 3.25 17.6
8 32 502 -11(2)/8 -583 3.78 16.0
9 32 -502 -(2)/8 +53 0.03 0.1
10 3(2) -502 (2)/8 -53 0.03 0.1
11 32 502 -9(2)/8 -447 2.52 10.6
 -4013.1 124.0

4013 1
Horizontal Reaction H    32  4 kN
124

The application of Castigliano’s Second Theorem to Statically-indeterminate Beams

v
Castigliano’s Second Theorem   can also be applied to means where R may be a
R
redundant moment or reaction and  will be a support settlement in the case of a
redundant reaction or a support rotation in the case of a redundant support moment.
When dealing with a beam as distinct from a framework, the strain energy will be
expressed in terms of the bending moment, i.e:-

M2
U  2 EI
dx

Consider a beam with a redundant action R of any sort. If R is removed to

make the beam statically determinate, then an expression can be written for the bending
moment in the statically determinate beam due to the loading and this will be called M.

If R is now replaced as an external load then the expression for the bending moment due
to R can be written, i.e. mR, where m would be the bending moment due to a unit load
placed at R.

Thus the total bending moment M1 = M + mR

M 12 dx ( M  mR)2 dx
U  2 EI
  2 EI

28
 v 2( M  mR)m dx Mmdx m2
R
  2 EI
  EI

EI


m2 Mmdx
R  dx    
EI EI

Mmdx

OR R EI
2
m dx
 EI

For example consider the following propped cantilever. It is required to calculate the
redundant reaction R and draw the BM and SF diagrams for the beam.

W
EI is constant
=0
Step 1 Remove the
redundant reaction R and find M for the statically-determinate cantilever.

Step 2 Place a unit load at

R and find m.

Step 3
   Mmdx    wx 3
 wx 
4
0
  Mmdx
0
dx  8 
EI 2   3w
R  0
  0

m2 dx
 EI  m2 dx 0 x dx  x3  8
2
0
 0 3 
  0

29
BM diagrams
ℓ
1) Free

2) Reactant

3) Combined

SFD

Find the reaction at C in the beam shown in Fig. assuming constant cross-section and
material throughout its length.

Example
60 kN/m
A
B C

1 2

2m 2m

Solution

Remove the redundant prop at C and obtain an expression for bending moment M (see
Fig. 38).

From C to B: M 0

60 x12
From B to A: M  30 x12
2

Remove the load and apply unit force in place of redundant prop at C to obtain an
expression for bending moment m.

30
From C to B: m   x2

From B to A: m  (2  x1 )

B
From C to B: C
Mm x  0
2
B 2 x 2 
 m x   x2  x   2   2  67
2 2
C 0
 3 0

2
A 2  60 x13 30 x14 
 Mm x    30 x (2  x1 ) x       280
2
From B to A: 1
B 0
 3 4 0
2
A 2  x13 4 x12 
 m  x   (2  x1 )  x   4 x1     18  67
2 2
B 0
 3 2 0

A
For whole beam: C
Mm x  0  280  280

A
C
m2 x  2  67  18  67  21 34

280
 Rc    13  kN
21 34

Example

31
A rectangular portal frame ABCD is shown loaded in Fig. If AD increases by 10 mm on
application of the loading and E is 200 kN/mm2, draw the bending moment diagram for
the loaded frame.

2 3

60 kN/m
E C

Iac = 44 x 106 mm4

IAB = 22 x 106 mm4

ICD = 22 x 106 mm4

2m 2m

5m

1 4
H A O H

VA VD

Solution

Take moments about A to find VD.

4 x VD = 60 x 2 x 1
VD = 30 kN
VA = 60 x 2 – 30
= 90 kN

Take the horizontal reaction H as redundant and consider frame in sections.

32
 M 0 
 limits 0 to 5
AB: m   x1 
I  22 106 m4

B Mm x

A EI
0

5
B m2 x 106 5 2 106  x13  106 125

A EI

22 E 0
x1  x     x
22 E  3  0 22 E 3

60 x2 2 
M  90 x2  
2  limits 0 to 2
BE: 
m  5 
6
I  44 10 m 4

E Mm x 106 2

D EI

44 E 0
5(90 x2  30 x2 2 ) x

2
106  3x 2 x 3 
 150  2  2 
44 E  2 3  0

106
  250
22E

E m2 x 106 2 106

 D EI

44 E 0
25 x 
22 E
 25

M  30 x3 
 limits 0 to 2
m  5 
CE:
I  44 106 m4

33
 2
C Mm x 106 2 106  x32 

E EI

44 E 0
5  30 x3 x  
22 E
 75  
 2  0

106
 150
22E

C m2 x 106 2 106


E EI

44 E 0
25 x 
22 E
 25

M 0 
 limits 0 to 5
m   x2 
DC:
I  22 106 m4
C Mm x
D EI  0
2
C m 106 5 2 106 125
D EI  x  22E 0 x4  x  22E  3
For whole frame:

DMm x 106 106

A EI  22E [250  150]  22E  400
Dm x 106 125 125  106 400
2

A EI  22E  3  25  25  3   22E  3

 10
10  10 400
6

H 
3 22200106
106400
222001063

 2, 67 kN

34
 60 kN/M

2.67 kN 2.67 kN

90 kN 30 kN

13.4 13.4

54
BMD (kN/M)

1.5 m Draw on Tension

Side

35
Question 3

In the pin-jointed truss shown in Figure 3 the cross-sectional area of all horizontal
members is 6000 mm2 of all vertical members 5000 mm2 and of all diagonal members
4000 mm2. Young’s Modulus for all members is 200 GPa.

Using Strain Energy methods calculate the vertical deflection of point G and the horizontal
deflection of point I.

Answer  G  11 3 mm   I  2  65 mm 

100 kN 100 kN 100 kN

A B C D

E F G H I 100 kN

3 3 3 3

Fig. 3

36
Question 4

In the pin-jointed truss shown in Fig. 4 all horizontal members have a cross-sectional area
of 300 mm2, all vertical members an area of 400 mm2 and all diagonal members a cross-
sectional area of 500 mm2.

Young’s Modulus for all members is 200 GPa.

Using Strain Energy methods calculate the vertical deflection of point B and the horizontal
movement of the roller support at D.

Answer  B  18  7 mm   D  15 mm 

E F G H

A B C D
300 kN

3 3 3

Fig. 4

37
Question 2

In the pin-jointed steel framework shown in Figure 2, all members are square, solid
sections, those in compression under the loading shown having dimensions 10 mm x 10
mm, while those in tension or having zero load have dimensions 7,5 mm x 7,5 mm.

Young’s Modulus for all members is 200 GPa.

Calculate using Strain Energy methods, the vertical and horizontal deflections of point F
of the frame.

Answer 19.6 mm  down

3 mm  left

A B C
10 kN

D E F

2 2

Fig. 2

38
Question 3

In the pin-jointed girder shown in Fig. 3 the various members have the following cross-
sectional areas:-

All vertical members: 600 mm2

All horizontal members: 800 mm2
All diagonal members: 1000 mm2

During erection of the girder it was discovered that member FC had been fabricated 6,5
mm too long.

Using STRAIN ENERGY techniques calculate the forces in all the members of the girder
under the loading shown.

180 kN 360 kN

E F G H

3m

A B C D

4m 4m 4m

Fig. 3

Answer

39