3

Au: Please check the numbring of figures cited or Display maths also.

Moment and Deflections in Simply Supported Beams

3.1 The Effect of a Concentrated Load on Shear, Moment, and Deflection

Start GOYA-S to find a window showing a simply supported beam (Figure3.1.1). This window shows a beam with its left and right ends supported by a pin and a roller, respectively. A perspective view of the beam is shown in Figure3.1.2. Because the force F is located at midspan, the reactions from the supports are F/2. In the lower right-hand corner of the screen, you will find a window showing the shear-force distribution (Figure3.1.3). The shear force changes abruptly at the load point (midspan). You can see the reason for the change using Figure3.1.4if you cut the beam to the left of the load (Figure3.1.4a), you will obtain Figure3.1.4b, showing that the shear force is clockwise or positive; if you cut it to the right of the load (Figure3.1.4c), you will obtain Figure3.1.4d, showing that the shear force is counterclockwise or negative. Recall that the shear force of a cantilever beam also changed abruptly at the points with concentrated loads (see Section 2.2, Chapter 2). In the lower left-hand corner of the screen, you will find a window showing the bending moment distribution (Figure3.1.5). The moment is zero at both ends and reaches its maximum at the load point (midspan in this case). You can see how the moment distribution is obtained if you cut the beam as shown in Figure3.1.6a to get the free-body diagram in Figure3.1.6b, which indicates that a section at a distance x from the left support is subjected to a bending moment of M= Fx 2 (3.1.1)

Note that M = 0 at x = 0 and that M increases linearly as x increases. The moment is said to be positive because the top of the beam is compressed. We can arrive at the same equation even if we consider the part of the beam to the right of the cut (Figure3.1.6c). Equilibrium of moment around any point on the section gives the following equation: (moment at the section, M ) + (moment by the external force, F ) + (moment by the reaction, F/2) = 0,

163

6861X_C003.indd 163

2/7/08 [Link] PM

164

Understanding Structures: An Introduction to Structural Analysis

10

0.7

Figure3.1.1 Window of a simply supported beam.

100 150 Roller 500 F/2

F/2

Pin

500

Figure3.1.2 Perspective view of beam.

5 5

Figure3.1.3 Shear-force window.

Cut

Cut

F/2 (a) Cut to the left of load F/2 F/2 (b) Equilibrium

F/2

F/2 F/2 (c) Cut to the right of load F/2 F/2 (d) Equilibrium

Figure3.1.4 Cut the beam to see the shear force.

6861X_C003.indd 164

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

250

165

Figure3.1.5 Bending moment.

or L F M + F x ( L x ) = 0 2 2

where clockwise moments are defined as being positive. This again results in M = Fx/2. If you cut the beam to the right of the load as shown in Figure3.1.6d, you will obtain the free-body diagram shown in Figure3.1.6e, which indicates that the
F F

Cut

Cut

F/2

(a) Cut to the left of load L/2 x L/2 F/2 M = F . x/2

F/2

F/2

(d) Cut to the right of load

F/2

x M = F . (L x)/2 F/2

(L x)

F/2

(b) Equilibrium of left part L x 2 (L x) F

(e) Equilibrium

F/2

M = F . x/2

F/2 F/2

(c) Equilibrium of right part

Figure3.1.6 Cut the beam to see bending moment.

6861X_C003.indd 165

2/7/08 [Link] PM

166

Understanding Structures: An Introduction to Structural Analysis

10 5 5 150 5 0.7

Figure3.1.7 Cut the beam in the window.

bending moment is linearly proportional to the distance from the right support (M = F.(L x)/2). The bending moment at the loaded point is, therefore, M= FL 2 (3.1.2)

Click the free-body diagram button to obtain Figure3.1.7, and confirm that the moment varies depending on the position. Example 3.1.1
Evaluate the force that would be required to break a chopstick of length 10 in. and section 0.2 0.2 in2. The chopstick is supported simply and the force is applied at midspan. Assume that the tensile strength of the wood for the chopstick is 6000 psi.

Solution
As we learned in Section 2.4 Chapter 2, the bending moment that causes a tensile stress of 6000 psi is determined as follows. M= ( 0.2 in ) bh 2 = (6, 000 psi) = 8 1bf-in 6 6

Substituting the result for M into Equation 3.1.1, we obtain the force to break the chopstick: P= 4 M 4 (8 lbf-in) = = 3.20 lbf L 10 in

This is a force that you can apply using your thumb. For comparison, the force required to break the chopstick by pulling is P = bh = ( 0.2 in )2 (6, 000 psi) = 240 1bf or 75 times the force required if the chopstick is loaded as a simply supported beam.

6861X_C003.indd 166

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

L/2 FL/(4EI) x (a) Curvature, = d2v/dx2 L/2

167

A (b) Slope, = dv/dx

(c) Deection, v

Figure3.1.8 Deformation.

AU: Insert intext figure 3.1.

Next, we shall evaluate the deflection of a simply supported beam. Dividing the bending-moment distribution of Figure3.1.5 by EI, we obtain the distribution of curvature shown in Figure3.1.8a, or d 2v Fx = dx 2 2 EI Integrating this function, dv Fx 2 = + A dx 4 EI (3.1.3) for 0 x L /2

where qA is a constant of integration representing the slope at the left end. Because the slope should be zero at midspan, we have

A =

FL2 16 EI

(3.1.4)

Integrating this equation with the boundary condition v = 0 at x = 0 leads to v= Fx 3 Fx + A x = ( 4 x 2 3L2 ) for 12 EI 48EI 0 x L /2 (3.1.5)

The deflection in the right half (L/2 x L) can be obtained by replacing x with (L x) as shown in the following equation: v= F (L x) [4( L x )2 3L2 ] for 48EI L /2 x L (3.1.6)

The deflection at midspan (x = L/2) is v= FL3 48EI (3.1.7)

6861X_C003.indd 167

2/7/08 [Link] PM

168

Understanding Structures: An Introduction to Structural Analysis

F

a = L/2 v

R = F/2

R = F/2

Figure3.1.9 Equivalent cantilever beam. where the negative sign represents that the beam deflects downward. Note that the deflection is proportional to the third power of the beam length L. Go back to GOYAS. Push the Detail of beam button and double the beam length; the deflection will be 23 = 8 times. If you substitute a = L/2 and R = F/2 into Equation 3.1.7, you will have v= Ra 3 3EI (3.1.7)

which is identical with the deflection of the cantilever beam (Section 2.8). The reason can be inferred from Figure3.1.9each half of the simply supported beam bends like a cantilever beam.

Example 3.1.2
Evaluate the deflection at rupture for the chopstick discussed in Example 3.1.1. Assume that the Youngs modulus is 1000 ksi.

Solution
As discussed in Section 2.5 Chapter 2, the moment of inertia of the section is I= bh 3 0.2 4 = = 1.3 10 4 ( in.4 ) 12 12

Substituting this into Equation 3.1.7 and noting that the force is 3.2 lbf, we obtain the deflection v= FL3 3.2 10 3 = 0.5 ( in ) 48EI 48 1, 000 10 3 1.3 10 4

The strain when the chopstick breaks under pure tension is

6, 000 = = 0.006 E 1, 000 10 3

The elongation at the break under pure tension is e = L = 0.006 10 = 0.06( in ) which is much smaller than the deflection of the simply supported beam.

6861X_C003.indd 168

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

10

169

0.6

Figure3.1.10 A simply supported beam with a load.

Exercise
Take any two numbers i and j you choose. Consider a timber beam of L = 1 m, b = (5+i) mm, h = (5+j) mm, tensile strength = 60 N/mm2, and Youngs modulus = 10,000 N/mm2. Calculate the force and the deformation at rupture (a) if it is loaded transversely and (b) if it is loaded in tension axially. Go back to GOYA-S and move the load to the right as shown in Figure3.1.10. You will find that the reaction (green arrow and digit) at the right support increases, whereas the one at the left decreases as you move the load. You can see the reason for the change of the reactions with the help of Figure3.1.11aif we consider the equilibrium of moments around the right support, we get the following equation with the clockwise moment defined as positive: which leads to RA = b F L (3.1.8) M = RA L Fb = 0

Similarly, the equilibrium of moments around the left support yields the following equation: RB = a F L (3.1.9)

As you move the load to the right (or increase a), the reaction at the right support RB increases. This is a demonstration of the principle of the lever. In GOYA-S, look at the windows showing the bending moment and the shear force (Figure3.1.12). The moment and shear diagrams also change as you move the force. If you cut the beam to the left of the applied load as shown in Figure3.1.11b, you will find that and M = RA x (3.1.11) V = RA (3.1.10)

If you cut the beam to the right of the applied load as shown in Figure3.1.11c, you will find that

V = RB

(3.1.12)

6861X_C003.indd 169

2/7/08 [Link] PM

170

Understanding Structures: An Introduction to Structural Analysis

L F

RA x

(a) Beam

RB

RA M = RA . x RA (b) Cut to the left of load x M = RB . (L x) RB (c) Cut to the right of load RB (L x)

Figure3.1.11 Equilibrium. and M = RB ( L x ) The maximum moment takes place at the loaded point, and its magnitude is M max = RA a = ab F L (3.1.13)

(3.1.14)

Mmax reaches its highest value if the load is placed at midspan (a = b = L/2). Look at Figure3.1.10 again. You may notice that the maximum deflection takes place not at the loaded point but near the middle of the beam. The deformation of the
210

3 0
Au: Please provide Figure3.1.12 Figure Caption

6861X_C003.indd 170

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

a F vmax A dv/dx = 0 (a) Deection, v B A (b) Slope, = dv/dx B b

171

Mmax/EI (c) Curvature, = d2v/dx2

AU: Caption?

Figure3.1.13 Deformation.

beam is illustrated in Figure3.1.13, where you should note that the maximum deflection occurs at the point of zero slope (dv/dx = 0). For interested readers: We can detect the point of zero slope as follows. In the left part of the beam (0 x a), we have the following equation for curvature: d 2 v M Fbx = = dx 2 EI EIL Integrating this equation leads to dv Fbx 2 = A + dx 2 EIL where qA is the slope at the left end. Noting that v = 0 at x = 0, we obtain v = A x + Fbx 3 6 EIL (3.1.17) (3.1.16) (3.1.15)

For the right segment of the beam (a x L), we obtain the following equation: d 2 v M Fa( L x ) = = dx 2 EI EIL Integrating this equation leads to dv Fa( L x )2 = B dx 2 EIL (3.1.19) (3.1.18)

6861X_C003.indd 171

2/7/08 [Link] PM

172

Understanding Structures: An Introduction to Structural Analysis

where q B is the slope at the right end. Integrating this and noting v = 0 at x = L , we obtain v = B ( L x ) + Fa( L x )3 6 EIL (3.1.20)

The deflection (v) should be continuous at the load point (x = a); i.e., the deflection computed with Equation 3.1.17 should be equal to that computed with Equation 3.1.20. The slope (dv/dx) should also be continuous at the load point. These boundary conditions lead to a set of simultaneous equations about qA and q B:

A = B =

Fb( L2 b 2 ) 6 EIL

(3.1.21)

Fa( L2 a 2 ) 6 EIL

(3.1.22)

If we substitute Equation 3.1.21 into Equation 3.1.16, we get dv Fb = ( L2 b 2 3 x 2 ) dx 2 EIL The point of zero slope (dv/dx = 0) is x= L2 b 2 3 (3.1.24) (3.1.23)

This is the point where the maximum deflection occurs (Figure3.1.13a). Substitute b = L/10 into Equation 3.1.24, for example, and you will find x 0.57 L, indicating that the maximum deflection occurs near midspan. As you increase the load, however, the beam will break at the load point because the maximum bending moment occurs at that point.

Exercise
Assume a simply supported chopstick with L = 10 in., b = h = 0.2 in., tensile strength = 6000 psi, and Youngs modulus = 1000 ksi (Figure3.1.14). Take any number i and apply a force at a distance of i in. from the right end of the chopstick. Determine at what force the chopstick will break.
AU: Figure 3.1.14 not cited in text. Tentatively cited here. OK?

Chopstick

i in.

L = 10 in.

Figure3.1.14 Simply supported chopstick.

 This equation is valid only for the case of a > b because we assumed that the point of zero slope is located at the left of the load. You need to use Equation 3.1.19 to analyze the case of a < b.

6861X_C003.indd 172

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

173

3.2 The Effect of Several Concentrated Loads on Shear, Moment, and Deflection
In GOYA-S, add a load and move it to suit Figure3.2.1. We can evaluate the reactions of this beam superimposing Figures3.1.1 and 3.1.10. We can also evaluate them directly as follows: In accordance with the forces and reactions shown in Figure3.2.2, we consider the equilibrium of moments around the left support to obtain (clockwise moments assumed to be positive) M = F1 L1 + F2 L2 RB L = 0 (3.2.1)

Equation 3.2.1 can be rearranged to determine the reaction at support B. RB = F1 L1 + F2 L2 L (3.2.2)

We can similarly evaluate the reaction at the right support, RA. Knowing the reactions, we can calculate the shear force, V. Noting that dM/dx = V, we obtain the bending moment, M. Exercise
Take any number i you choose. Determine the forces that will produce the bending moments shown in Figures3.2.3a,b. Check your results using GOYA-S and sketch the deformed shape of the beam.

Practice: Construct an interesting bending-moment diagram as you did in Section 2.2, Chapter 2. Start GOYA-C and apply forces as shown in Figure3.2.4 (the deformation is magnified by a factor of four). The force at the free end is the same as the reaction at the right in Figure3.2.1. You will note that the equilibrium conditions for the cantilever beam and the simply supported beam are the sameincluding the bendingmoment and shear-force diagrams. The only difference is provided by the boundary conditions. Let denote the deflection at the free end. Rotate Figure3.2.4 by /L clockwise. You will obtain the deformed shape shown in Figure3.2.1. Figure3.2.5a shows a simply supported beam subjected to a couple at the left end. You should note that the bending-moment diagram (Figure3.2.5b) is the same as that for the cantilever beam depicted in Figure3.2.5c.
10 10

12

1.3

Figure3.2.1 A simply supported beam with two loads.

6861X_C003.indd 173

2/7/08 [Link] PM

174

Understanding Structures: An Introduction to Structural Analysis

L2 F2

L1

F1

RA L

RB

Figure3.2.2 Reactions.
M = 50(i + 10) 1000 30 40 100 (a) 200 M = 50(i + 10) 30

30

40 (b)

30

Figure3.2.3 Bending moment.

Using GOYA-S, we can simulate Figure3.2.5 as shown in Figure3.2.6. In this case, the load of 60 N is applied at a distance of 5 mm from the left support, and the equivalent couple is M = 60 5 = 300 N-mm; the reaction at the right support is M/L = 300/100 = 3 N.
10 10

3.84

8 0

12

Figure3.2.4 Bending-moment diagram of cantilever beam.

6861X_C003.indd 174

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

L M

175

M/L

(a) Couple

M/L

M (b) Bending moment M M/L

M/L (c) Equivalent cantilever beam

Figure3.2.5 Simply supported beam with a couple at the left end.

Next, we calculate the shear force and the bending moment in the simply supported beam with a uniformly distributed load (Figure3.2.7a). Because of symmetry, the reactions are RA = RB = wL 2 (3.2.3)

60 15 3 3 0.7 3 15 60 Force (N) Stress (N/mm ) Strain (10^3) 0 285 0 3 3

2

12 3 12

57

Figure3.2.6 Bending-moment diagram similar to that of a cantilever beam.

6861X_C003.indd 175

2/7/08 [Link] PM

176

Understanding Structures: An Introduction to Structural Analysis

L

+w . L/2 (b) Shear force w.L2/8 w . L/2

RA

(a) Distributed load

RB

(c) Bending moment

Figure3.2.7 Simply supported beam under uniformly distributed load.

As we learned in Chapter 2, we have dV/dx = w (negative because the load w is downward). Noting that the shear force at the left support (x = 0) is V = RA (positive because it is clockwise), we have V = RA wx = wL (L 2 x) 2 (3.2.4)

The result is shown in Figure3.2.7b. From Chapter 2 we know that dM/dx = V. Noting that the bending moment at the left support (x = 0) is zero, we write M = RA x + 1 2 wx wx = (L x) 2 2 (3.2.5)

The result is shown in Figure3.2.7c. Example 3.2.1

Calculate the deflection of the simply supported beam shown in Figure3.2.7a.

Solution
Substituting Equation 3.2.5 into the equation d2v/dx2 = M/EI (Chapter 2), we obtain an expression for curvature: d2v M w = = ( x 2 Lx ) dx 2 EI 2 EI (3.2.6)

Integrating Equation 3.2.6, we obtain an expression for the slope of the beam at any position x: dv w = ( 2 x 3 3Lx 2 ) + C1 dx 12 EI (3.2.7)

where C1 is a constant of integration. Integrating Equation 3.2.7, we arrive at the expression for the deflection v= w ( x 4 2 Lx 3 ) + C1 x + C2 24 EI (3.2.8)

6861X_C003.indd 176

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

177

where C2 is another constant of integration. Knowing the two boundary conditions (the deflections at the left and the right supports are zero) helps solve the constants C1 and C2. Thus, we obtain v= w ( x 4 2 Lx 3 + L3 x ) 24 EI (3.2.9)

The deflection at midspan (x = L/2) is v= 5wL4 384 EI (3.2.10)

We can also derive this equation by considering the cantilever beam as having a length equal to half the length of the simply supported beam (a = L/2), as illustrated in Figure3.2.8. The results in Chapter 2 indicate that the deflection of the cantilever beam of length a subjected to uniformly distributed load w is (see Figure3.2.8b) va = wa 4 8EI (3.2.11)

The deflection caused by an upward (reaction) force of wa is (see Figure3.2.8c) vb = wa 4 3EI (3.2.12)

The deflection caused by both loads is (see Figure3.2.8a) v = va + vb = 5wa 4 24 EI (3.2.13)

Substituting a = L/2 into Equation 3.2.13 leads to the same deflection as Equation 3.2.10 in magnitude.

v = va + vb

wa (a) Total deformation a = L/2 w va

wa

a = L/2 vb

(b) Deection by distributed load

(c) Deection by reaction

wa

Figure3.2.8 Deflection of cantilever beam.

6861X_C003.indd 177

2/7/08 [Link] PM

178

Understanding Structures: An Introduction to Structural Analysis

FL/2 FL/2

Figure3.2.9 Bending-moment distribution.

Example 3.2.2
From among the sets of forces shown in Figure3.2.10(ad), select the correct set that produces the bending-moment distribution shown in Figure3.2.9. Select the correct deflected shape from among those shown in Figure3.2.11(ad) for the bending-moment distribution shown in Figure3.2.9.

Solution
Applying dM/dx = V for the moment distribution in Figure3.2.9, we obtain the shear force shown in Figure3.2.12. Because the changes in the shear force are caused by external forces, the correct force set is in Figure3.2.10a. To find the correct deflection pattern, we need to recall that the bending-moment diagram (because of our sign convention) indicates whether the top or the bottom of the beam is compressed. Because Figure3.2.9 indicates that the top is compressed everywhere in the beam, the correct answer must be that shown in Figure3.2.11a. We can simulate the result using GOYA-S. First, apply two downward forces as shown in Figure3.2.13. Then, apply another force at the middle and increase its magnitude to 10 N, as shown in Figure3.2.14, where the deformation is amplified 16 times. If you increase the force to 13.8 N, you will find that the deflection at the middle will become zero, as shown in Figure3.2.15, where the deformation is amplified 32 times. This is the deflected shape of Figure3.2.11c. The beam does not bend abruptly, as shown schematically in Figure3.2.11d. Note that the deflected shape shown in Figure3.2.15 is symmetric about midspan. In Example 2.5.5, for a cantilever beam with two loads applied at the free end and midspan (P1 and P2), we saw that the deflection at the free end is zero if P1/P2 = 5/16. In Figure3.2.15, we see that the ratio of the reaction (3.1 N) to the downward force (10 N) is 3.110 / 5/16 .

F F (a) F 3F (c)

F 2F (b)

F 4F (d)

Figure3.2.10 Loads.

6861X_C003.indd 178

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

179

(a)

(b)

(c)

(d)

Figure3.2.11 Deflection.
+F/2 F/2 +F/2 F/2

Figure3.2.12 Shear force.

10 10 25 10 1 0 . 25 25 25 . 0

10

Force (N) Strain (10^3) 250 0 250 0 10

Stress (N/mm )

10

Figure3.2.13 Two downward forces (deformation amplified 4 times).

10 10 0 5 5 0.3 5 0 10
2 Force (N) Stress (N/mm ) Strain (10^3)

5 5 5

125

125

0 5 5 5 5

Figure3.2.14 Upward force of 10 N (deformation amplified 16 times).

6861X_C003.indd 179

2/7/08 [Link] PM

180

Understanding Structures: An Introduction to Structural Analysis

10 10 9.5 3.1 0 3.1 0 6.9 + 9.5 13.8 Force (N) Strain (10^3) Stress (N/mm2) 2.6 2.6 + 6.9

78 95

78

0 3 7 7 3

Figure3.2.15 Upward force of 13.8 N (deformation amplified 32 times).

Example 3.2.3
Calculate the shear force and the bending moment at A in the beam shown in Figure3.2.16.

Solution
We can represent the total uniform load of 6wL by a concentrated load located at the centroid of the uniform load. The equilibrium of moment around point C gives RB = 6wL L/4 L = 3wL/2 and the equilibrium of forces in the vertical direction gives RC = 6wL RB = 9wL/2 Next, we shall cut the beam at A as shown in Figure3.2.17b and replace the distributed load by a concentrated load of 2wL. From the equilibrium of forces, we get VA = wL/2
w A 2L 2L 2L
AU: Caption?

Figure3.2.16
 When you want to calculate the reactions, you may replace all the distributed load with an equivalent single load. When you want to calculate the shear force or the bending moment at a point, however, you need to go through the following procedure: (a) Evaluate the reactions (Figure 3.2.17a). (b) Cut the beam and then replace the distributed load with an equivalent single load (Figure 3.2.17b). (c) Consider equilibrium of forces and moments.

6861X_C003.indd 180

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

6wL B A RB = 3wL/2 3L L B 3wL/2 2L L 2wL A MA QA L (a) Reactions C RC = 9wL/2 2L

181

(b) Free body

+3wL/2

wL/2

+2wL 5wL/2

(c) Shear force 2wL2 wL2 (d) Bending moment

Figure3.2.17 and from the equilibrium of moments, we get M A = wL2 . We can similarly calculate the shear force and the bending moment at other sections. The results are shown in Figure3.2.17c, d.

AU: Caption?

Example 3.2.4
Calculate the bending moment at A in the beam shown in Figure3.2.18.

Solution
Because the total beam length is 4L , the total vertical load is 4wL. Recognizing symmetry, we determine each reaction to be 2wL, as shown in Figure3.2.19a. Next, we cut the beam at A as shown in Figure3.2.19b and replace the distributed load by a concentrated load of 2wL. The moment equilibrium gives M A = 0 . We can similarly calculate the shear force and the bending moment at other places. The results are shown in Figure3.2.19c, d.

6861X_C003.indd 181

2/7/08 [Link] PM

182

Understanding Structures: An Introduction to Structural Analysis

w A

L
Au: please provide Figure3.2.18 Figure caption

Minigame Using GOYA-S

The green figures in the GOYA window show the maximum upward and downward deflections of the beam. Try producing the largest possible bending moment using four loads while controlling the maximum deflections so that they do not exceed 1 mm. Use default values for the size of the beam and Youngs modulus.

Design Your Own Beam (Part 6)

We want to design a beam that can carry a mini-elephant whose weight is any number you choose plus 10 lbf. Assume that each leg carries the same amount of gravitational force. The density of the beam is 0.5 lbf/in.3, the beam width is 1 in., and
L L L L

w A

2wL (a) Reactions 2wL A L (b) Free body +wL +wL wL (c) Shear force

2wL

2wL L

wL

wL2/2 (d) Bending moment

wL2/2
AU: Caption?

Figure3.2.19

6861X_C003.indd 182

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

1 mm 1 mm

183

Major Structural League Ranking (N . mm) > 100 Rookie > 2,000 > 200 A > 5,000 All Star > 500 AA > 10,000 MVP > 1,000 3A > 50,000 Hall of Fame

Max. moment
AU: Caption?

Major

Figure3.2.20

Trumpet h 4 in 2 in 4 in
AU: Caption?

Figure3.2.21

Youngs modulus is 10 ksi. The elephant is scared of flexible beams. The maximum deflection of the beam should not exceed 0.1 in. The beam material cannot resist tensile stress exceeding 300 psi. What is the required beam depth, h? Check your results using GOYA-S. (Hint: The beam is symmetric. You may replace it by a cantilever beam as shown in Figure3.2.8.)

3.3 Similarities between Beam and Truss Response

In this section we will investigate the similarities in the ways beams and trusses resist transverse loads. We do this to develop an improved perspective of the relationships between internal and external forces, a perspective that will help improve our understanding of structural response. In Figure3.3.1a, b, we compare the internal and external forces in a beam and a truss, both loaded with vertical forces of magnitude F located symmetrically about midspan. We choose each of the 10 truss panels to have a length of 10 mm and a depth of 10 mm. Because the panels are square, the web members, or diagonals, make an angle of 45 with the horizontal. We choose a beam having the same span and a depth of 15 mm. Because of the symmetry, the magnitude of each reaction of the beam and the truss is determined to be F. Determining the shear-force diagram for the beam is straightforward (Figure3.3.1c). The shear force is constant between the left reaction and the force applied on the left. It is zero between the applied forces. Between the force applied on the right and the right reaction, the shear is again constant. The beam shear is equal in magnitude to the external force F and is shown to be positive on the left and negative on the rightto be consistent with our sign convention.

6861X_C003.indd 183

2/7/08 [Link] PM

184

6861X_C003.indd 184

F A 10 D 30 F (b) Equivalent Truss 2 F F Tension (d) Axial forces in diagonal members 0 E G K L B C F H J F

10

15

30

(a) A beam

Compression 2 F

(c) Shear force

2F

3F Compression (f) Axial forces in top chords 3F Tension

2F

30F

(e) Bending moment 0 F 2F 2F (g) Axial forces in bottom chords F

Understanding Structures: An Introduction to Structural Analysis

AU: Caption?

Figure3.3.1 A beam and an equivalent truss.

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

185

The shear force in the truss is resisted by the web members (Figure3.3.1d). There is a direct relation between the shear in the truss and the force in the web members. Because of the inclination of the web members (45), for a shear force of F, the force in each web member is 2 F . The sign changes from the left to the right end of the truss because the web members work in tension on the left and in compression on the right. Comparing the diagrams for shear distribution in the beam and force distribution in the web members of the truss, we understand that there is a similarity as well as a proportionality between the internal shear distribution in a beam and the distributions of the forces in the web members of a truss. Next, we examine the moment distribution in the beam (Figure3.3.1e). We have studied the relation between shear and change in moment. So, it is not surprising to see there is a steady increase in moment in the left portion of the beam, between the reaction and the force F, where the shear is constant. Between the applied forces F, the moment does not change because there is no shear in this region. In the right portion of the beam, where the shear is constant and negative, the moment decreases at a steady rate from the maximum at the point of application of the force F to zero at the point of reaction. In the truss, the bending moment is resisted by the forces in the top and bottom chords. The moment at any section should be equal to the product of the force in one chord and the distance between the two chords. We expect the chord forces in the truss to vary as the moment varies in the beam. When we look at the distribution of forces in the top and bottom chords of the truss (Figure3.3.1f, g), we notice two surprising features: 1. The variation of the forces in the top and bottom chords differ from one another. 2. They also differ from the distribution of the moment in the beam (there are abrupt changes, and the force distributions are not symmetrical about midspan). In the following text, we try to understand the reason for these apparent inconsistencies. First we look at a simple casethe chord forces in panel HJKL, which is not subjected to shear (Figure3.3.2a). The moment to be resisted is 30F. From this, we deduce that the force in each chord is Chord Force = Moment 30 F = = 3F Truss Height 10 (3.3.1)

The signs for the forces in the top and bottom chords are different. These forces must balance one another. If we consider the internal normal stresses in the beam, we note a similar phenomenon. As shown in Figure3.3.2b, the stress distribution may be assumed to be linear. The internal stresses may be represented by forces at the centroids of the tensile and compressive stresses (Figure3.3.2c). The distance between the two internal forces is 2h/3. Because we assumed h to be 15, we find each force to have a magnitude of 3F to balance the moment of 30F (Figure3.3.1e).

6861X_C003.indd 185

2/7/08 [Link] PM

186

Understanding Structures: An Introduction to Structural Analysis

H M/Z H

3F

3F 2 h = 10 3

10 3F K (a) Forces in the truss K M/Z (b) Stress in the beam

h = 15

3F K (c) Equivalent forces

Figure3.3.2 Cut between HJ.

There appears to be a similarity as well as a proportionality between the internal normal stresses in beams and chord forces in trusses. We now examine the conditions in panel ABED of the truss next to the left reaction. Our first deduction is that the vertical component of the force in the web member must be equal to F (Figure3.3.3). Because the web member makes an angle of 45 with the horizontal, its horizontal component must also be equal to F in tension. The equilibrium of moment around node A requires zero force in the bottom chord; the equilibrium of horizontal forces requires that the top chord must carry a compressive force F. That is why, in this panel, the top chord sustains a force, whereas the bottom chord does not. The moment equilibrium in that section around the top chord gives M = Fx (3.3.2)

which agrees with Figure3.3.1e. Should this We move over to the next panel on the right (Figure3.3.3b). The shear remains AU: be 3.3.4? the same. Therefore, the vertical and horizontal components of the web member are equal to F. Considering the equilibrium at joint B, we decide that the force in the top chord must be equal to 2F in compression. To maintain horizontal equilibrium
x A F x F D F F

Figure3.3.3 Cut between A and B.

 Moment equilibrium around the bottom chord also gives M = Fx.

6861X_C003.indd 186

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

x A B

187

2F x 10

10 D F
Au: please provideFigure3.3.4 Figure citation

F F E F

Cut between B and C.

across any section within the panel, the bottom chord force needs to be F in tension. Now we can consider the moment equilibrium in that section around the top chord: M = (Contribution of web member) + ( Contribution of bottom chord) = F ( x 10 ) + F 10 = Fx

(3.3.3)

This result agrees with the linearly distributed bending moment in the beam shown in Figure3.3.1e. From the foregoing, we deduce that (1) the abrupt changes in chord forces are caused by the condition that forces can change only at the joints of the truss and that (2) the lack of symmetry in the distribution of the chord forces is caused by the orientation of the web members. Otherwise, the distribution of the chord forces along the span of the truss represents a good analogue for the changes in the internal normal forces in a beam. Figures3.3.5 and 3.3.6 compare the results obtained with GOYA-T and GOYAS, respectively. Note that the deformed shapes are also similar. The compressive and tensile forces in the highlighted region agree with the axial forces in the corresponding top and bottom chords. Figure3.3.7a shows the deformation of panel HJKL, where the top chord shortens and the bottom chord lengthens, as in the flexural deformation of a beam. The strains of the top and bottom chords are

P 3F = EA EA

(3.3.4)

10 0

10 14.110 14.1 10 10

20

30 14.1 10 20

30 0 30 0

30 0 30 0

30 0 30 0

30 0 0 30

20

14 0 14 10 14 10 10 20 30

10

Figure3.3.5 Results from GOYA-T.

6861X_C003.indd 187

2/7/08 [Link] PM

188

Understanding Structures: An Introduction to Structural Analysis

10 10

10

10

1.2

Figure3.3.6 Results from GOYA-S.

where P is the axial force, E is the Youngs modulus, and A is the cross-sectional area of each chord. The deformation of each chord is obtained as the product of the strain e and the length of the panel d x = 10. e = x = 30 F EA (3.3.5)

The flexural rotation dq of the panel in Figure3.3.7a (the angle between HK and JL) is

2 | e | 6F = 10 EA

(3.3.6)

According to the definition in Section 2.5 Chapter 2, the curvature f is obtained by dividing dq by the width dx = 10:

6F = x 10 EA

(3.3.7)

3F

3F 5 5

F 5 5

3F

x = 10 (a) Panel HJKL

3F

x = 10

(b) Panel ABDE

Figure3.3.7 Flexural deformation.

6861X_C003.indd 188

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

189

The bending moment around the centerline of the truss (the chained line in Figure3.3.7a) is M = 3F 5 + 3F 5 = 30 F (3.3.8)

Equations 3.3.7 and 3.3.8 lead to

M 50 EA

(3.3.9)

On the other hand, the bending moment at the center of panel ABDE is given by substituting x = 5 into Equation 3.3.2: M = F 5 = 5F (3.3.10)

The curvature f in panel ABDE is obtained in reference to Figure3.3.7b:

F = x 10 EA

(3.3.11)

Equations 3.3.10 and 3.3.11 again lead to Equation 3.3.9. We can discuss the similarities between the truss and the beam further. We regard the truss as a beam having the section shown in Figure3.3.8. As will be discussed in Chapter 4, the moment of inertia of the section is I = 2 Ay 2 (3.3.12)

Substituting y = 5 into Equation 3.3.11, we get I = 50 A . Equations 3.3.9 and 3.3.12 lead to

M EI

(2.4.6)

which we obtained in Section 2.6 Chapter 2 (Equation 3.4.6). In addition to the flexural deformation shown in Figure3.3.7b, panel ABDE is distorted (Figure3.3.7c) because the diagonal member AE is subjected to a tensile force 2 F . You can simulate the distortion if you provide GOYA-T with the boundary condition shown in Figure3.3.7d. A similar distortion also occurs in beams. It is
A y y A

Figure3.3.8 Section.

6861X_C003.indd 189

2/7/08 [Link] PM

190

Understanding Structures: An Introduction to Structural Analysis

2 F A

2 F

2 F (b) Simulation using GOYA-T

(a) Deformation of ABDE

Figure3.3.9 Shear deformation.

called shear deformation or shear distortion (note that the axial force in the diagoFigure nal member is related to the shear force in the corresponding beam) (Figure3.3.9). AU: 3.3.9 not cited text. TenHowever, the effect of the shear deformation (or distortion) on the total deflection is in tatively cited here. OK? small in shallow trusses or beams such as those shown in Figures3.3.1 or 3.3.4. Exercise
Choose any numbers i and j to set the loads shown in Figure3.3.8. Draw figures similar to Figure3.3.1 for the truss shown in Figure3.3.8. Also, draw the deformed shape. Compare your result with those obtained by a friend for the same problem.
10 10 10 + i (N)

10 + j (N)

3.4 Construction and Test of a Timber Beam

You have made many virtual experiments with beams using GOYA. In this section, you will test an actual beam to review Chapters 2 and 3. You need the following: 1. Two pieces of wood 1/8 in. thick and 1/4 in. wide, with length of 2 ft. In this section, we call them beams. You can buy them at a hardware shop. Do not buy balsa. 2. A ruler longer than 8 in. 3. A small plastic bag. 4. Two binder clips. 5. Two desks. 6. A kitchen scale. 7. A friend to help you.

6861X_C003.indd 190

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

191

5 in A B

5 in 10 in C D

Ruler Binder clip 1/8 in 1/4 in (b) Beam for measurement

(a) Beam for test

Figure3.4.1 Mark one of the beams and attach a ruler to the other beam.

Test 1: Measuring Youngs Modulus for Wood

We shall measure Youngs modulus for wood using Equation 3.1.7 developed in Section 3.1. Step 1: Mark one of the beams as shown in Figure3.4.1a and fix the plastic bag at point C using a binder clip. Step 2: Fix the ruler to the other beam using another binder clip as shown in Figure3.4.1b. Step 3: Lay a pencil on each desk at a distance of 20 in. as shown in Figure3.4.2. Place the marked beam on the pencils. Step 4: Take the other beam and place the end without the ruler on the floor. Make sure it is vertical. Measure the initial location of point C. Step 5: Hang the plastic bag at midspan of the beam. Place objects (marbles, sand, etc.) in the bag so that the deflection reaches approximately 2 in. Measure the deflection at point C (vCC in Figure3.4.2) as accurately as you can. Step 6: Weigh the plastic bag with its contents using the kitchen scale. Step 7: Recall the equation for the deflection of a simply supported beam v= FL3 , 48EI (3.1.7)

10 in

10 in vCC

A B Pencil Desk C F

Pencil Desk

Figure3.4.2 Measure deflection.

6861X_C003.indd 191

2/7/08 [Link] PM

192

Understanding Structures: An Introduction to Structural Analysis

5 in 5 in vBC C B Desk F Desk 10 in

Figure3.4.3 Measure deflection between support and load.

and calculate Youngs modulus E by substituting the measured deflection v, the load F, the beam length L, and the moment of inertia I = bh 3 / 12 . The unit of the load should be stated in lbf. Step 8: Check your result noting that Youngs modulus of wood (except balsa) is usually between 1000 and 1800 ksi. If your result is out of this range, reexamine your calculation and/or your measurements.

Test 2: Testing the Equation for the Deflected Shape of a Simply Supported Beam
In this test, we shall check Equation 3.1.5, which determines the deflected shape of the beam, using the same equipment as in Test 1. Step 1: Calculate the deflection vBC shown in Figure3.4.3. The notation vBC indicates that the deflection is measured at point B and the force is applied at point C. Use Equation 3.1.5 and the value for E determined in Test 1. Step 2: Measure the deflection vBC and compare it with the calculation result.

Test 3: Testing the Reciprocity Theorem Using a Simply Supported Beam

In this test, we shall investigate the reciprocity theorem discovered by Scottish physicist James Clerk Maxwell (18311879), who is well known for his contributions to electromagnetism. Step 1: Assume that the weight is at the location shown in Figure3.4.4. Calculate the deflection at point A vCB using Equation 3.1.17. This should be equal to the calculated deflection vBC in Figure3.4.3. This equality is a demonstration of Maxwells reciprocity theorem. Step 2: Do the test shown in Figure3.4.4 to measure the deflection vCB. Compare it with the calculated value.
 Prediction is very important in civil engineering. Car designers can test the safety of their car by testing many prototypes before actually selling the car. On the other hand, civil engineers can rarely test their design using a full-scale model; they need to predict structural behavior through calculation.

6861X_C003.indd 192

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

5 in 5 in 10 in vCB C Desk F Desk

193

Figure3.4.4 Move the load to the left string.

The reciprocity theorem is applicable to any structure. Figure3.4.5 shows two other examples in which vBA (the displacement of point B on a structure caused by a load F acting at point A) is always equal to vAB (the displacement of point A caused by the same amount of load F acting at point B).

Test 4: Testing the Equation for the Deflected Shape of a Cantilever Beam
Use the same beam and the same weight you used for Test 3. Step 1: Place the beam at the edge of a desk and press it down as shown in Figure3.4.6 with a stiff book or pencil case. Press it firmly. Otherwise, the beam would deform, as shown in Figure3.4.7, and you would not obtain the correct boundary condition. (The correct boundary condition is that the beam has zero slope and zero deflection at the edge of the table or that we have a fixed end at C). Measure the deflections vAA and vBA at locations A and B, respectively. Step 2: Calculate the deflections vAA and vBA for this test using the equations in Chapter 2. Do they agree with the measurements?

AU: Insertion OK?

F A F B

vBA B vAB A A

F B F B

vBA

vAB A

(a) Simple beam

(b) Simple beam with projection

Figure3.4.5 The reciprocity theorem.

6861X_C003.indd 193

2/7/08 [Link] PM

194

Understanding Structures: An Introduction to Structural Analysis

5 in

5 in

Press rmly

vAA A

vBA B C

Pencase

Desk F

Figure3.4.6 Deflections at loaded point and between support and load.

5 in

5 in Pencase

C Desk

Figure3.4.7 If you press the pen case insufficiently .

Test 5: Testing the Reciprocity Theorem Using a Cantilever Beam

Step 1: Hang the load at location B as shown in Figure3.4.8 and measure the deflections vAB and vBB at locations A and B, respectively. Step 2: Calculate the deflections vAB and vBB for this test using the equations in Chapter 2. Do they agree with your measurements? Step 3: Study your results. Do they conform to the reciprocity theorem?
5 in 5 in

vAB

vBB

Pencase

Desk

Figure3.4.8 Move the load.

6861X_C003.indd 194

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

195

B A F 10 in

C Desk

Figure3.4.9 Rotate the beam around its longitudinal axis by 90.

Test 6: Investigating the Effects of moment of Inertia

Step 1: Rotate the beam around its longitudinal axis by 90 so that the beam has the shorter dimension of its cross section parallel to the desk surface (Figure3.4.9). Step 2: Measure the deflection at point A. Step 3: Calculate the deflection and compare it with the measurement. Compare it also with the result obtained in Test 4.

Test 7: Sensing a Couple

Step 1: Hold the wood beam with your fingers at midspan. Step 2: Ask a friend to push one end up and the other end down with the same force (Figure3.4.10). You will sense the twist or couple that is required to resist the moment generated by equal and opposite forces applied at the ends of the beam (Figure3.4.11a). You will also see that the deflection of the beam is antisymmetrical about midspan. Step 3: Try holding the beam at different points away from its middle as shown in Figure3.4.11b. Observe the deflected shape.

Figure3.4.10 Sense a couple.

6861X_C003.indd 195

2/7/08 [Link] PM

196

Understanding Structures: An Introduction to Structural Analysis

L/2 M/L M M/2 M/2 L/2 M /L M/L 2L/3 M 2M/3 M/3 L/3 M/L

Figure3.4.11 Reactions.

3.5 Problems
(Neglect self-weight in all the problems. Assume that all the beams are prismatic.) 3.1 Find an incorrect statement among the following five statements concerning the simply supported beam in Figure3.5.1. 1. The shear force at point B is larger than that at point D. 2. The bending moment at point B is the same as that at point D. 3. The bending moment reaches maximum at point C. 4. The deflection reaches maximum at point C. 5. The slope reaches maximum at point A.
F A L B C L 2L 6L 2L D E

AU: Caption?

Figure3.5.1

3.2 Which sets of loads yield the bending-moment diagrams (1)(5) in Figure3.5.2? Select the correct answer from among (a)(e) in Figure3.5.2.

(1)

(2)

(3)

(4)

(5)

(a)

(b)

(c)

(d)

(e)
AU: Caption?

Figure3.5.2

6861X_C003.indd 196

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

197

3.3 The width and the height of the beam in Figure3.5.3 are 100 mm and 180 mm, respectively. What is the maximum bending stress in the beam? Select the correct answer from among (1)(5). (1) 50 N/mm2 (2) 100 N/mm2 (3) 200 N/mm2 2 (4) 300 N/mm (5) 400 N/mm2

w = 12 kN/m

6.0 m

Figure3.5.3

AU: Caption?

3.4 Consider a continuous beam subjected to a uniformly distributed load and supported as shown in Figure3.5.4a. What is the ratio of the reaction at A (RA) to that at B (RB)? Select the correct answer from among (a)(e). (Hint: The deflections due to a uniformly distributed load and a concentric load are shown in Figure3.5.4b.): (a) 1/2 (b) 1/3 (c) 2/5 (d) 3/5 (e) 3/10

w 5wL4 384EI

L/2

L/2

RA L/2

RB L/2

RC = RA L/2

FL3 48EI F L/2

(a) Continuous beam

AU: Caption?

(b) Deections

Figure3.5.4

AU: Provide answers.

3.5 The beam shown in Figure3.5.5 has zero bending moment at point A. Find the correct ratio of P to wL from among (1)(5).

6861X_C003.indd 197

2/7/08 [Link] PM

198

Understanding Structures: An Introduction to Structural Analysis

F : wL (1) (2) (3) (4) (5) 1:3 1:2 1:1 2:1 3:1
AU: Caption?

F A L L L w

Figure3.5.5

3.6 What is the deflection at point A of the beam shown in Figure3.5.6? Select the correct answer from among (1)(5), where I is the moment of inertia of the beam section. (1) FL3 8EI (2) FL3 3EI (3) FL3 2 EI (4) 2 FL3 3EI (5) 5FL3 6 EI

F A L L
AU: Caption?

Figure3.5.6

3.7 What is the deflection at point A of the beam shown in Figure3.5.7? Select the correct answer from among (1)(5), where I is the moment of inertia of the beam section. (Hint: use the result in Problem 3.6) (1) 2 FL3 3EI (2) 5FL3 6 EI (3) FL3 EI (4) 4 FL3 3EI
F A L
AU: Caption?

(5)

5FL3 3EI

Figure3.5.7

3.8 A simply supported beam is subjected to uniformly distributed couples as shown in Figure3.5.8. The beam width and depth are 80 mm each, and Youngs modulus is 1000 N/mm2. Determine the moment diagram, the deflected shape, and the maximum deflection. (Hint: Determine the moment diagram for the cases shown in Figure3.5.9a, b, where the

AU: Corrected as 3.8 OK?

6861X_C003.indd 198

2/7/08 [Link] PM

Moment and Deflections in Simply Supported Beams

10 kN-m/m

199

10 m

Figure3.5.8 Uniformly distributed couples.

uniformly distributed couples are represented by a concentrated couple 10 10 = 100 kN-m and two concentrated couples 5 10 = 50 kN-m, respectively. Recall that these couples are represented by pairs of horizontal forces, as shown in Figure3.5.9 c, d. You can simulate each case using GOYA-S, as shown in Figure3.5.9 e, f.)
100 kN-m 50 kN-m

(a) Concentrated couple 100 kN 1m 100 kN (c) A pair of horizontal forces 50 kN

(b) Two couples 50 kN 1m 50 kN 50 kN (d) Two pairs of horizontal forces

100 kN

50 kN

50 kN

1m (e) A pair of vertical forces

100 kN

1m

1m

(f) Two pairs of vertical forces

Figure3.5.9 Hint.

6861X_C003.indd 199

2/7/08 [Link] PM

6861X_C003.indd 200

2/7/08 [Link] PM

Au: Please check the numbring of figures cited or Display maths also.

Bending and Shear Stresses

4.1 First Moment

In Chapters 2 and 3 we considered beams with rectangular sections and learned that 1. The stress, s, in the extreme fiber of the section, is determined using the section modulus

M . Z

(2.5.7)

2. The curvature of the beam can be determined using the moment of inertia (second moment)

M . EI

(2.6.4)

3. Integrating the curvature obtained by Step 2, we can determine the deflection of the beam. These procedures apply not only for rectangular sections but also for others such as I sections and tubes. In this chapter, we shall consider sections that are not rectangular. First, we need to define the first moment of a section. Start GOYA-I to reach the window in Figure4.1.1a. Each square, surrounded by the grid lines, measures 10 10 mm. The figure shows a rectangular section with b = 30 mm and h = 40 mm. Click the Bending stress button to get the applet shown in Figure4.1.1b. This applet shows the distribution of stress for a bending moment of 10 103 N-mm. You can change the magnitude of the moment using the sliding bar. The red and blue colors indicate compressive and tensile stresses, respectively. Click the six squares indicated by the arrows in Figure4.1.1a. You will get an inverted T-section, as shown in Figure4.1.2a,b. As you click, you will notice that the red line changes its position. The red line indicates the position where the bending stress is zero, as shown in Figure4.1.2c. It is called the neutral axis. To calculate the position of the axis, we shall show that the first moment  defined by the following
 We call this the first moment because Equation 4.1.1 includes the first order of the coordinate y. If it is replaced with the zero order of the coordinate y, i.e., y 0 = 1, we obtain the area of the section A=

dA =

dA

(4.1.2)

The second moment will be defined similarly in the following section.

201

6861X_C004.indd 201

2/7/08 [Link] PM

202

Understanding Structures: An Introduction to Structural Analysis

(a) Section

(b) Bending stress

Figure4.1.1 Initial windows of GOYA-I.

y dA h y0

(a) Beam

(b) Section

(c) Stress distribution

dA h

y y0

(d) Cut the beam at the middle

(e) Dene the coordinate y from the bottom

Figure4.1.2 Beam of inverse-T section.

6861X_C004.indd 202

2/7/08 [Link] PM

Bending and Shear Stresses

203

equation should be zero about the neutral axis. S=

h y0

y0

y dA = 0

(4.1.1)

where y0 is the distance from the bottom of the section to the neutral axis, and y is the distance from the neutral axis to an infinitesimal section dA, as shown in Figure4.1.2b. To derive the previous equation, let us again assume a linear strain distribution over the depth of the section:

= y

(2.5.2)

where f is the curvature at the section. Stress at any level in the section may be expressed using Youngs modulus, E, as

= E = E y

(1.2.3)

or we may relate it to curvature using Equation 2.5.2. (2.5.3)

This equation indicates that the stress is distributed linearly over the depth of the section, as shown in Figure4.1.2c. The axial force on the section is the product of the area dA and the stress s : P = dA = E y dA = E S

(4.1.3)

The axial force in a beam subjected to transverse load is zero (P = E S = 0 ) because of the equilibrium of forces along the beam axis (Figure4.1.2d). We therefore get S = 0 (Equation 4.1.1). If we redefine the y-coordinate as distance from the bottom of the section (Figure4.1.2e), Equation 4.1.1 can be rewritten as S=

( y y0 ) dA = 0

(4.1.4)

We can expand this equation as

( y y0 ) dA =

y dA y0

dA =

y dA y0 A = 0

and obtain the following equation that is more useful for calculating y0 than Equation 4.1.1. y0 = h 0 y dA A (4.1.5)

6861X_C004.indd 203

2/7/08 [Link] PM

204

Understanding Structures: An Introduction to Structural Analysis

a 3a y0 y dy 3a (b) Area of innitesimal sections dA = 3a dy dy dA = a dy

a a a a

(a) Section

Figure4.1.3 Inverted-T section.

Example 4.1.1
Determine the location of the neutral axis y0 for the inverse-T section shown in Figure4.1.3a.

Solution
We shall use Equation 4.1.5 for defining the coordinate y as distance from the bottom (Figure4.1.3). Noting that the infinitesimal area is dA = 3a dy in the wide part of the section (0 y a) and dA = a dy in the narrow part (a y 3 a), we have

y dA =

y 3a dy +

4a

y a dy = 1.5a 3 + 7.5a 3 = 9a 3

The area of the total section is A = 3a a + a 3a = 6a 2 Substituting this in Equation 4.1.5, we have y0 = h 9a 3 0 y dA = 2 = 1.5a A 6a

Example 4.1.2

Calculate the location of the neutral axis y0 for the triangular section shown in Figure4.1.4.

dy

dA f(y) h

Figure4.1.4 Triangular section.

6861X_C004.indd 204

2/7/08 [Link] PM

Bending and Shear Stresses

205

Solution
We use Equation 4.1.5 again. The width of the infinitesimal area f( y) shown in Figure4.1.4 can be defined as

h h

f ( y) h y = b h Because the infinitesimal area is dA = f( y) dy,

We get

y dA =

y f ( y)dy =

b h

y ( h y)dy =

bh 2 6

y0 =

bh 2 /6 h h 0 y dA = = A bh/2 3

indicating that the neutral axis of a triangle crosses its centroid (i.e., center of gravity).

Determining the Centroid of a Section by Experiment

We need the following: 1. A sheet of cardboard (preferably with a grid printed on it) 2. A thread, a needle, a pair of scissors, a ruler, and a calculator In this test, we shall investigate if the neutral axis of any section, as in the case of a triangle, crosses its centroid. 1. Draw any section using GOYA-I. 2. Cut the cardboard to have the same shape as the section. 3. Calculate the location of the neutral axis y0 and check to see if the result agrees with the red line in the screen. Draw the line on the cardboard model (line AB in Figure4.1.5).

Thread

B y0

x0

Figure4.1.5 Hang by thread.

6861X_C004.indd 205

2/7/08 [Link] PM

206

Understanding Structures: An Introduction to Structural Analysis

C y0 B A

D x0
Au: provide Caption

Figure4.1.6

4. Punch a hole on line AB near the edge of the cardboard model. Hang it using the thread as shown in Figure4.1.5. If your calculation of y0 is correct, the line AB will be vertical. This is the proof that the neutral axis AB crosses the centroid of the board. 5. Calculate the location of the neutral axis x0 assuming that the section in Figure4.1.5 is bent around the horizontal axis. Hang the board as shown in Figure4.1.6. Make certain again that the line CD is vertical. This is the proof that the neutral axis CD crosses the centroid of the board. 6. Punch another hole in the cardboard model. The hole must not be on lines AB or CD (Figure4.1.7). Check to see if the extension of the thread (the broken line in Figure4.1.7) crosses the intersection point of the neutral axes. The broken line represents the neutral axis when the section is bent around the broken line.

B A D

Au: provide Caption

Figure4.1.7

6861X_C004.indd 206

2/7/08 [Link] PM

Bending and Shear Stresses

207

Thread Needle

AU: Caption? Figure4.1.8

Hang by needle and thread.

7. Stick the needle and thread into the cardboard model as shown in Figure4.1.8, and make certain that the cardboard is horizontal. This is the proof that the stuck point is the centroid of the board. Let us examine the reason why the line AB in Figure4.1.5 is vertical. We shall use Figure4.1.9, where the y-coordinate is measured from the neutral axis. If t is the thickness of cardboard and r its density, the moment around the hole caused by the slice dA is dM = ( t dA) y

Integrating the moment over the whole body gives M=

dM = t y dA = t S

where S is the first moment defined by Equation 4.1.1. Because S = 0, we get M = 0, which means that the body is in equilibrium in terms of moment and does not rotate.

dA y0 . t . dA y

Figure4.1.9 Moment around the hole.

6861X_C004.indd 207

2/7/08 [Link] PM

208

Understanding Structures: An Introduction to Structural Analysis

If you conduct an integration over both the x and y directions, you can demonstrate the reason why the board is horizontal in Figure4.1.8.

Coffee Break
The first person who discovered the computation method for the center of gravity was a Greek physicist, Archimedes (BC 287212). It is an important extension of his famous principle of the lever. Note that we also used this principle in reference to Figure4.1.9. Archimedes also invented integral calculus, which is indispensable to the computation of the center of gravity. E. T. Bell describes him as the greatest scientist in antiquity in the book Men in Mathematics. The inventors of differential calculus, the counterpart of integral calculus, were Isaac Newton and Gottfried W. von Leibnitz, who lived in the 17th century.

4.2 Second Moment and Section Modulus

In the preceding section, we defined the first moment as S=

y dA

(4.1.1)

where y is the distance between the infinitesimal section dA and the neutral axis (Figure4.2.1). In this section, we shall define the second moment (or the moment of inertia) by replacing y with y2 in Equation 4.1.1: I=

dA

(4.2.1)

In Section 2.5, Chapter 2, we learned that the bending moment is the integral of the axial force of the infinitesimal section [Link] multiplied by the distance from the neutral axis y:
b dA h dy y dA y dy

M = y dA

(2.5.3)

dA y

dy

dy y

dA

ymax

(a) Rectangle

(b) H (vertical)

(c) H (horizontal)

(d) General shape

Figure4.2.1 Various sections and infinitesimal segments.

 Bell, Eric Temple, 1937. Men in Mathematics, Touchstone (Simon & Schuster), New York.

6861X_C004.indd 208

2/7/08 [Link] PM

Bending and Shear Stresses

209

This equation applies to all kinds of sections. In Section 2.6 we learned that the stress is proportional to the curvature f and the distance from the neutral axis y, as expressed in Equation 2.6.3.

= E y

(2.6.3)

Substituting this into Equation 2.5.3, M = E y 2 dA = EI

(4.2.2)

The moment of inertia plays an important role in relating the bending moment to the curvature (curvature is a measure of how or at what rate the beam bends). For the rectangular section in Figure4.2.1a, dA = [Link] and I=

y 2 dA =

h/2

h/2

y 2 b dy =

bh 3 12

(4.2.3)

as we learned in Section 2.6. Figure4.2.2 shows the initial window of GOYA-I. As was stated earlier, each square measures 10 10 mm. The digits in the right-hand column show the contribution of each row to the moment of inertia. For example, the contribution of the uppermost row is I =

20

10

y 2 30 dy = 70 10 3 mm 4

as listed in the column of numbers that appear in Figure4.2.2. If we are interested in obtaining an approximate value, we can state the contribution of these three squares

Figure4.2.2 Window of GOYA-I.

6861X_C004.indd 209

2/7/08 [Link] PM

210

Understanding Structures: An Introduction to Structural Analysis

I = 182 I = 863 y = 35 mm

y = 15 mm

I = 117

I = 163

(a) I = 267

(b) I = 417

(c) I = 577

(d) I = 1907

Figure4.2.3 Moment of inertia (unit: 103 mm4).

to the moment of inertia as I y 2 dA = 152 (30 10 ) = 67.5 10 3 mm 4

where y is the distance from the neutral axis to the centroid of these squares. The total moment of inertia is shown at the bottom of the column (I = 70 + 10 + 10 + 70 = 160 103 mm4). Press Ctrl + N three times to create four windows. In these windows, draw the four sections shown in Figure4.2.3. All the sections have the same area, A = 2000 mm2, but very different moments of inertia, I, ranging from 90 to 1907 103 mm4. The large differences are caused primarily by the different contributions of the extreme rows ( I y 2 dA in Figure4.2.3). For the section in Figure4.2.3a, the average distance to the extreme rows is as small as y = 15 mm, but for the section in Figure4.2.3d the average distance is as large as y = 35 mm. For the section in Figure4.2.3c, the average distance to the extreme rows (or squares) is large but the area dA is small. The expression = M /EI indicates that the beam with the section in Figure4.2.3d will have a smaller curvature and, therefore, smaller deflection for a given load over a given span than that of the other sections. Substituting = M /EI into = E y,

y M I

(4.2.4)

Press the Bending stress button in the windows showing the sections in Figure4.2.3 and obtain the stress distributions shown in Figure4.2.4. Note that the stresses vary linearly with the distance from the neutral axis y. If we define the distance between the edge of the section and the neutral axis ymax, as shown in Figure4.2.4, the maximum stress in the section s max is

max =

ymax M I

(4.2.5)

If the beam is made of brittle material with strength s f, it will fail at the bending moment Mf = I ymax f (4.2.6)

6861X_C004.indd 210

2/7/08 [Link] PM

Bending and Shear Stresses

max ymax max ymax

211

(c) Horizontal H

(d) Vertical H

Figure4.2.4 Stress distribution.

We call the following coefficient as the section modulus. Z= I ymax (4.2.7)

In GOYA-I, Z is indicated at the bottom of the window. The section modulus of a rectangle is Z= I ymax = bh 3 12 h bh 2 = 2 6 (4.2.8)

as we learned in Section 2.5, Chapter 2. Now, we can rewrite Equation 4.2.6 as M f = Z f (4.2.9) In other words, the strength of a beam is proportional to its section modulus. Because the section modulus of the section in Figure4.2.4d is much larger than that of the section in Figure4.2.4c, an I-shaped section should be positioned as shown in Figure4.2.4d to efficiently resist bending. Technical termsflanges and web: Figure4.2.5 shows the typical section of a steel I-beam. Structural engineers call the strips in the top and bottom as flanges, and the vertical plate as a web, which
Flange

Web

Flange

Figure4.2.5 Steel I-beam.

6861X_C004.indd 211

2/7/08 [Link] PM

212

Understanding Structures: An Introduction to Structural Analysis

a 3a X 3a a 2a a 2a
AU: Caption?

Figure4.2.6

may look similar to the skin (web) that joins the toes of swans. Flanges are typically thicker than webs, as shown in the figure to resist bending moment effectively. Example 4.2.1
Calculate the section modulus of the section shown in Figure4.2.6.

Solution
First, we evaluate the moment of inertia as the sum of three parts (top, middle, and bottom): I= = =

y
3a

dA y 2 5a dy +

4a

3a

3 a

y 2 a dy +

3 a

4 a

y 2 5a dy

185 4 54 4 185 4 424 4 a + a + a = a 3 3 3 3

Noting ymax = 4a, we have Z= I ymax =

424 3

a 4 106a 3 = 4a 3

Because the section considered is symmetrical about its neutral axis, we can shorten the calculation process by partitioning the section as shown in Figure4.2.7a rectangular section of 8a 5a minus two sections of 6a 2a. Recalling I = bh 3 /12 for a rectangular section, we obtain the same result. I= (5a ) (8a )3 ( 2 a ) (6a )3 640 4 216 4 424 4 2 = a a = a : OK 12 12 3 3 3

6861X_C004.indd 212

2/7/08 [Link] PM

Bending and Shear Stresses

213

ymax = 4a 6a 8a = 8a 2x

ymax = 3a 6a

5a

5a

2a

Figure4.2.7 Partitioning of the section (not for section modulus).

However, we should not use this technique for calculating the section modulus because ymax of the outer rectangle (4a) is different from that of the inner ones (3a). Z= (5a ) (8a )2 ( 2 a ) (6a )2 160 3 72 3 88 3 106 3 2 = a a = a < a : NG! 6 6 3 3 3 3

Also, we cannot use this shortcut for calculating the moment of inertia of a section that is not symmetrical about the horizontal axis (Figure4.2.8) because the neutral axes of the partitioned sections are different from each other. This technique is valid only for the moment of inertia of a section symmetrical about the bending axis. The correct moment of inertia of the section in Figure4.2.8 is

I=

0.5 a

1.5 a

y 2 3a dy +

2.5 a

0.5 a

y 2 a dy = 3.25a 4 + 5.25a 4 = 8.5a 4

If you use the shortcut, you will get an incorrect answer: I= (3a ) ( 4 a )3 ( a ) (3a )3 2 = 16a 4 4.5a 4 = 11.5a 4 > 8.5a 4 : NG! 12 12

3a

3a

3a

1.5a

4a

2x

3a

Figure4.2.8 Never do this because neutral axes are different.

6861X_C004.indd 213

2/7/08 [Link] PM

214

Understanding Structures: An Introduction to Structural Analysis

2h/3 h h /3

dy

dA f(y)

2h/3 y y

Figure4.2.9 Triangular section.

Example 4.2.2
Calculate the moment of inertia and the section modulus of a triangular section (Figure4.2.9).

Solution
Consider an infinitesimal slice of thickness dy and width f( y), as shown in Figure4.2.9. The dimension f( y) can be expressed as 2 h y : f ( y ) = h : b 3 or 2 y f ( y) = b 3 h

Because the area of the slice is f( y) dy, we have I=

y 2 dA =

2 h/3

h/3

2 y2 3

y bh 3 b dy = h 36

(4.2.10)

Noting that ymax = 2h/3, we obtain Z= I ymax =

bh3 36

2 h/3

bh 2 24

(4.2.11)

Both the moment of inertia and the section modulus for the triangular section are smaller than those of the rectangular section with width b and depth h. That does not surprise us.

Example 4.2.3
Calculate the moment of inertia of a circular section with a radius of R.

Solution
We define the angle between the neutral axis and the edge of the slice, q, as shown in Figure4.2.10a. The width of the slice dA varies, as

f ( y) = 2 R cos Figure4.2.10b shows the segment defined by dq. Noting that dq is so small that the arc length, R dq, approximates the chord length, we obtain Figure4.2.10c, which shows in detail how we express dy in terms of R dq and cosq :

dy = R d cos

6861X_C004.indd 214

2/7/08 [Link] PM

Bending and Shear Stresses

f(y) R . cos R . cos dA R d dy y = R . sin d R Rd dy = Rd . cos

215

Rd

(a) Whole section

(b) Segment dened by d

(c) Detail to express dy

Figure4.2.10 Circular section.

The area of the slice in Figure4.2.10a is

dA = f ( y) dy = 2 R 2 cos 2 d Noting that y = R sin as shown in Figure4.2.10a, the moment of inertia, I, is I=

y 2 dA = 2 R 4

/2

/ 2

sin 2 cos 2 d

R4 = 2

R4 sin 2 d = 4 / 2
2

/2

R4 (1 cos 4 ) d = 4 / 2
/2

(4.2.12)

Let us compare the preceding result with the moment of inertia of a square section having the same area, i.e.,

h2 = R2 where h denotes the side dimension of the square. Substituting the previous equation into Equation 4.2.12, I=

R4 h4 h4 = 4 4 12.56

showing that the moment of inertia of a circular section is similar to that of the square section (I = h 4 /12) having the same area.

Example 4.2.4
Building columns or bridge piers may be subjected to bending moment both in the x- and y-direction by earthquake or storm effects. Assume that the tube section of Figure4.2.11a is subjected to bending moments of M x = M y = 50 10 6 N-mm and compute the maximum stress in the section. This type of column is often used in bridges.
 This is equivalent to a bending moment of about the inclined axis in Figure 4.2.11b. See Figure 4.2.11c showing the vector summation of Mx and My. As we learned in Chapter 2, the bending moment itself is not a vector. However, if we cut the member and consider the forces at the cut, we can treat the moment acting on the cut as a vector. You will remember that we did that for axial forces.

6861X_C004.indd 215

2/7/08 [Link] PM

216

Understanding Structures: An Introduction to Structural Analysis

500 50 500 50 50 400 (a) 50 My Mx 600 My Compression (b) (c) M M = 502 106 N-mm Tension Mx

Figure4.2.11 Tube section.

Solution
The moment of inertia around the x-axis is obtained by subtracting the moment of inertia of the inner rectangle (400 500) from that of the outer one (500 600): Ix = 500 600 3 400 500 3 = 48.3 108 mm 4 12 12

The corresponding section modulus is Zx = Ix 48.3 108 = = 16.1 10 6 mm 3 ymax 300

The maximum stress caused by the bending moment of M x = 50 10 6 N-mm is

x =

Mx 50 10 6 = = 3.11 N/mm 2 Z x 16.1 10 6

The stress distribution is shown in Figure4.2.12a.

= 3.50 N/mm2 Tension

m N/m 11 ssion . 3 e = ompr C (a)

Compression = -3.50 N/mm2

2 sion m Ten N/m 1 1 . =3

My Mx M

= 6.61 N/mm2 Tension

Compression = 6.61 N/mm2 (b) (c)

Figure4.2.12 Stress distribution.

6861X_C004.indd 216

2/7/08 [Link] PM

Bending and Shear Stresses The moment of inertia around the y-axis is Iy = 600 500 3 500 400 3 = 35.8 108 mm 4 12 12

217

Iy is smaller than Ix because of the smaller height (h = 600 mm). The corresponding section modulus is Zy = Iy ymax = 35.8 108 = 14.3 10 6 mm 3 250

The maximum stress caused by the bending moment of M y = 50 10 6 N-mm is

y =

My Zy

50 10 6 = 3.50 N/mm 2 14.3 10 6

The stress distribution is shown in Figure4.2.12b. The stress caused by the simultaneous bending moments of M x = M y = 50 10 6 N-mm is shown in Figure4.2.12c. The maximum stress is

= x + y = 3.11 + 3.50 = 6.61 N/mm 2

and occurs at the corners where the moment Mx and My cause stresses of the same sense (tension or compression).

Determine the Optimum Proportions for a Section Resisting Moment

We wish to design a beam section that can resist a bending moment of +50 103 N-mm (using GOYA-I) with the limitation that neither its height nor its width should exceed 80 mm. Note that the positive sign of the bending moment indicates that the bottom fiber will be in tension. Assume that we can use a material with a compressive strength of 2 N/mm2 and a tensile strength of 1 N/mm2. Design the section so that the area (the number of squares) is minimized. The section should be continuous, as shown in Figure4.2.13a, and should not be discontinuous, as shown in Figure4.2.13b. It is difficult to find the best solution directly. Do it by trial and error.

(a)

(b)

Figure4.2.13 Examples.

6861X_C004.indd 217

2/7/08 [Link] PM

218

Understanding Structures: An Introduction to Structural Analysis

4.3 Construction and Test of a Styrofoam Beam

We need the following: 1. A Styrofoam sheet, 2 2 ft, with a thickness of 1/2 in. 2. A knife 3. Adhesive for the Styrofoam 4. A weight of approximately 5 lb, such as marble or sand 5. A small plastic bag 6. A piece of wood 2 ft long 7. A ruler longer than 8 in. 8. A binder clip 9. A kitchen scale

In this section, we will make a rectangular tube beam (Figure4.3.1) using a Styrofoam sheet. To prevent failure at the middle of the tube beam, the joint shall be strengthened, as shown in Figure4.3.2, using the remaining material. The beam shall be located between desks placed 40 in. apart, as shown in Figure4.3.3, and designed to fail if the load applied at midspan reaches a value of F = 4 lbf. The beam height (h) shall be 2 in. The beam width shall be determined based on the following assumptions. Assumption 1: The tensile strength of Styrofoam is ; s f 30 psi Its compressive strength is higher. Assumption 2: The beam will break at either end of the strengthened part (Figure4.3.3), where the bending moment is M= F L = 2 18 = 36 lbf-in. 2

Therefore, the width of the beam shall be designed to satisfy Mf = Z = s f = 36 1bf-in

Glue

24 in 2 in b 24 in

Figure4.3.1 Rectangular tube beam.

6861X_C004.indd 218

2/7/08 [Link] PM

Bending and Shear Stresses

219

2 in b

4 in

Figure4.3.2 Strengthen the joint.

To make optimum use of the beam, we shall also evaluate the Youngs modulus of the Styrofoam as follows: 1. Fix the ruler to the wood using a binder clip as we did in Section 3.4. 2. Hang the plastic bag on the beam. 3. Place objects (marbles, sand, etc.) into the plastic bag so that the deflection reaches approximately 2 in. 4. Measure the deflection at midspan. 5. Determine the weight. 6. Assume that the strengthened part is rigid. The deflection of the beam shall be similar to that of the cantilever beam shown in Figure4.3.4, from which you can calculate Youngs modulus.

4 in

40 in

4 in Bar with ruler

F = 4 lbf

F/2

Breaks at either end L = 18 in 4 in L = 18 in

F/2

Figure4.3.3 Beam under load.

6861X_C004.indd 219

2/7/08 [Link] PM

220

Understanding Structures: An Introduction to Structural Analysis

Deection V L = 18 in

Figure4.3.4 Equivalent cantilever beam.

After the calculation, increase the weight gradually until the beam fails. Then compute the strength of your Styrofoam from the test data using f = M f /Z (i.e., assuming that Styrofoam is a brittle material).

4.4 Shear Stress

We have considered tensile and compressive stresses (normal stresses) caused by bending moment. In this section, we investigate stresses caused by shear. Figure4.4.1 shows a body on a desk pushed by a force V. If the force is smaller than a certain threshold value, the body will not move because of the friction between
A V V

= V/A

(a) Solid on desk

(b) Friction

Figure4.4.1 Shear stress on desk.

6861X_C004.indd 220

2/7/08 [Link] PM

Bending and Shear Stresses

H y D vx A E xy F x z x y x y y C

221

D xy C

yx y x yx

A xy B

x y B

B (a) Innitesimal cube

(b) Normal stress

(c) Shear stress

Figure4.4.2 Stresses on an infinitesimal cube.

the body and the desk surface. We define the average shear stress on the contact surface between the base and the desk as the ratio of the force, V, to the area of the contact surface, A:

V A

(4.4.1)

We can define the unit shear stress on a beam section similarly by dividing the shear force on a section by the area of the section. However, the actual distribution of shear stress is not necessarily uniform in either case considered. In this section we examine shear-stress distribution on a beam section. Before tackling the problem of shear-stress distribution over a beam section, we need to develop a set of definitions for stresses. Figure4.4.2a shows an infinitesimal cube that is subjected to four stresses acting on the top face ADEH and side face ABEF. Axis x is perpendicular to face ABEF, and axis y is perpendicular to face ADEH. We focus on four stresses (also see Table4.4.1): s : the normal stress acting in the direction of the x-axis on face ABEF sy: the normal stress acting in the direction of the y-axis on face ADEH t xy: the shear stress acting in the direction of the y-axis on face ABEF t yx: the shear stress acting in the direction of the x-axis on face ADEH The stresses s x and sy are called normal stresses because they act in a direction perpendicular to the faces of the cube (Table4.4.1). The stresses t xy and t yx, acting in

Table4.4.1
Direction of Force x Direction of face x y x yx (= xy) y xy y Normal stress Shear stress

AU: Table title not provided.

6861X_C004.indd 221

2/7/08 [Link] PM

222

Understanding Structures: An Introduction to Structural Analysis

V dx

M + dM

Figure4.4.3 Simply supported beam.

directions parallel to the faces of the cube, are called shear stresses. Generally, there are normal as well as shear stresses on the face ABCD; but for simplicity we shall ignore them in order to be able to show them in two dimensions (Figure4.4.2b,c). In Figure4.4.2b, note that the equilibrium of forces requires a stress of s x on face DC and a stress of sy on face BC in the opposite direction to the normal stresses on faces AB and AD. In Figure4.4.2c, similarly, t xy and t yx act on faces DC and BC. The equilibrium condition for moment leads to the following important equation.

xy = yx

(4.4.2)

Let us move to the problem of the beam shown in Figure4.4.3. We assume that the beam has a rectangular section. The left and the right faces of the slice dx are subjected to bending moments M and M + dM, respectively. Figure4.4.4a shows the distribution of normal stresses caused by the bending moment. Normal stresses s t and s t + ds t denote the tensile stresses at the bottom edge of the slice. They are related to bending moments M and M + dM.

c c + dc

t =

M M = Z bh 2 /6

(4.4.3) (4.4.4)

M + dM M + dM t + d t = = Z bh 2 /6

h h/2 t dx t + d t dx (b) Cut the slice at the middle t t + dt b dx h

3 V 2 bh

(a) Stress on slice dx

(c) Shear stress

Figure4.4.4 Equilibrium of slice dx.

6861X_C004.indd 222

2/7/08 [Link] PM

Bending and Shear Stresses

223

Cutting the beam slice in Figure4.4.4a at mid-height results in the free-body diagram shown in Figure4.4.4b. The volumes of the triangular prisms represent the tensile forces acting on the two sides of the beam slice. They are T= T + dT = bh 3 M t = 4 2 h (4.4.5) (4.4.6)

bh 3 M + dM ( t + d t ) = 4 2 h

The difference between the tensile forces, dT, is balanced by the shear force on the upper face of the free body in Figure4.4.4b. The required force is the shear stress, t , multiplied by the area of the face, [Link]. Rearranging,

b dx = dT =

3 dM 2 h

(4.4.7)

3 dM /dx 2 bh

(4.4.8)

Recalling that V = dM/dx, we have the following important equation for shear stress at the neutral axis of a beam with a rectangular section:

3 V 2 bh

(4.4.9)

Equation 4.4.9 indicates that the shear stress at the middle height of the section is 1.5 times the average shear stress, V/bh. Though the shear stress in Figure4.4.4b is in the horizontal direction, the equilibrium of moments in the infinitesimal square, shown in Figure4.4.4c, requires the same shear stress in the vertical direction (Equation 4.4.2). You might be surprised to find shear stresses in the horizontal direction; but if you cut the beam horizontally, as shown in Figure4.4.5, you will see the need for glue (or shear strength) to prevent slip. Cutting the slice dx at a distance y from mid-height, we obtain Figure4.4.6. The volume of the solid body on the left of the slice represents the tensile force on the left face. T=
2 b 2y h 6 M h t + t y = y2 3 2 h 2 h 2

Figure4.4.5 Beam cut horizontally.

6861X_C004.indd 223

2/7/08 [Link] PM

224

Understanding Structures: An Introduction to Structural Analysis

Neutral axis y

b t dx t + dt

Figure4.4.6 Cut the slice at y.

The tensile force on the right side, T + dT, can be obtained similarly. These results and = dT /( b dx ) lead to

2 6V h y2 3 bh 2

(4.4.10)

This equation indicates that the shear stress distribution is parabolic (Figure4.4.7). Shear stress reaches its maximum at mid-height ( y = 0) and is zero at the top and bottom ( y = h/2) of the section. The latter statement is intuitively obvious because the top and bottom faces of the beam are unrestrained. We cannot justify stresses acting on those surfaces. To satisfy equilibrium, the force represented by the volume of the curved solid representing the distribution of shear stress on the section (Figure4.4.7) must be equal to the shear force in the beam. We can similarly calculate the shear stress distribution for sections with nonrectangular shapes. Consider the section in Figure4.4.8a. Let us compute the shear stress at distance y1 below the neutral axis. Recall that the normal stress caused by the bending moment at position y from the neutral axis is

y M I

(4.2.4)

b 3 V 2 bh

= h

Figure4.4.7 Shear stress distribution.

6861X_C004.indd 224

2/7/08 [Link] PM

Bending and Shear Stresses

225

b1 dx Neutral axis b1 dA (a) Section y1 y

y1

T + dT

y0

dA (b) Stress

Figure4.4.8 Shear stress in general section.

where I is the moment of inertia. Integrating this from the bottom ( y0) to the target ( y1), we obtain the tensile force acting on the left face of the body, as shown in Figure4.4.8b. T=

y1

y0

dA =

y1

y0

y M dA I

(4.4.11)

The tensile force on the right face is obtained similarly. T + dT =

y1

y0

( + d ) dA =

y1

y0

y ( M + dM ) dA I

(4.4.12)

From equilibrium in the horizontal direction,

b1 dx = dT =

y1

y1

y0

y dM dA I

(4.4.13)

where b1 is the width of the beam at y1. The preceding equation results in

V b1 I

y0

y dA

(4.4.14)

Note that the integral in this equation is the first moment of the section from the bottom to y1. Thus, we may rewrite it as follows:

=
where S1 is the first moment. Example 4.4.1

S1 V b1 I

(4.4.15)

Compute the shear stress at the neutral axis of the I-section in Figure4.4.9a subjected to a shear force V.

6861X_C004.indd 225

2/7/08 [Link] PM

226

Understanding Structures: An Introduction to Structural Analysis

0.4 24 13 9.6

700

12.3

24 0.4 300 (a) Section (b) Moment of inertia (c) Shear stress 105V 9.6

Figure4.4.9 H-shaped section.

Solution
The moment of inertia of the section is obtained in accordance with Figure4.4.9b: 1 [300 700 3 (300 13) ( 700 2 24 )3 ] 195 10 7 mm 4 12

I=

The first moment from the bottom to the neutral axis is S = 300

326

350

y dy + 13

326

y dy 3.12 10 6 mm 3

Thus, the shear stress at the neutral axis is

S1 3.12 10 6 V= 12.3 10 5 V b1 I 13 195 10 7

We can compute the shear stresses at the other locations similarly as shown in Figure4.4.9c. Note that the shear stress changes abruptly at the border between the flange and the web (about 23 times, in this case) because the width (b1 in Equation 4.4.15) changes abruptly from 300 mm to 13 mm (about 1/23). If we simply divide the shear force V by the area of the web, we get

V V = 11.8 10 5 V Aweb 13 ( 700 2 24 )

Note that this approximation is very close to the maximum shear stress shown in Figure4.4.9c, indicating that the web resists almost all the shear force and that the contribution of the flanges to shear resistance is negligible.
 You may be surprised to see such an abrupt change in stress. In fact, Equation 4.4.15 is an approximate determination of the shear-stress distribution. An improved formation leads to continuous (but rapid) change at the border, though Equation 4.4.15 is sufficiently accurate for structural design.

6861X_C004.indd 226

2/7/08 [Link] PM

Bending and Shear Stresses

227

a a a (a) (b) a

Figure4.5.1 Square section.

4.5 Problems
4.1 A beam with a square section made of a brittle material is to be subjected to a moment acting about an axis parallel to an edge (Figure4.5.1a) or about a diagonal axis (Figure4.5.1b). About which axis is it (1) stronger and (2) stiffer? 4.2 A steel pipe has an outer radius of 500 mm, an inner radius of 480 mm, and a thickness of 20 mm (Figure4.5.2a). Calculate the ratio of the moment of inertia of the pipe to that of the solid circular section with the same cross-sectional area. (Hint: the moment of inertia of a circular section with radius r is I = pr4/4.) Which of the five options listed here is the correct one? 1. Approximately 5 2. Approximately 15 3. Approximately 25 4. Approximately 35 5. Approximately 45 4.3 Calculate the ratio of the moment of inertia of the I-section in Example 4.4.1 (or Figure4.5.3a) to that of the solid square section with the same cross-sectional area. Which of the five options listed here is the correct one?

500

480 20

(a) Pipe

(b) Equivalent circular section

AU: Caption?

Figure4.5.2

6861X_C004.indd 227

2/7/08 [Link] PM

228

Understanding Structures: An Introduction to Structural Analysis

120 kN Table 4-5-1 Options a a 1 2 A B Section a-a 200 200 C 150 150 3 4 5 A 0 0 5 5 10 B 10 15 10 15 10 C 0 0 5 5 10

Figure4.5.3 Cantilever column.

1. Approximately 5 2. Approximately 15 3. Approximately 25 4. Approximately 35 5. Approximately 45 4.4 A force of 1500 N is applied to a cantilever column with a rectangular section as shown in Figure4.5.24. Select the correct set of shear stresses at the points A, B, and C from among the options listed in Table4.5.1. The stress unit is N/mm2.
AU: Should this be Figure 4.5.3?

6861X_C004.indd 228

2/7/08 [Link] PM

Au: Please check the numbring of figures cited or Display maths also.

Buckling

6.1 Simple Models

In Section 1.7 Chapter 1, we developed an expression to determine the buckling load for a wooden column (Equation 1.7.1). In Chapter 6, we shall derive a similar but more general expression based on what we learned in Chapter 2. In this section, we shall approach the general expression by using simple models of the buckling phenomenon. Before we do that, we shall conduct a simple test to obtain a feel for the physical phenomenon of buckling. 1. Take a wooden stick approximately 2-ft long with a 1/8-in. (~3.2 mm) square section. 2. Insert the stick into a small block of Styrofoam as shown in Figure6.1.1a. 3. Attach a binder clip to the top of the stick (Figure6.1.1b). 4. Place several small magnets onto the binder clip carefully until the stick starts to bend as shown in Figure6.1.1c. Bending may occur suddenly without warning, or slowly, depending on the straightness of the stick and the arrangement of the weights. What you have observed is another example of a buckling column albeit on a small scale. Leonhard Euler (17071783), a Swiss mathematician and physicist, derived the following relationship between the buckling load for a concentrically loaded (axis of load aligned with the axis of the column) column and its geometrical and material properties. An example of the Euler equation applicable to a cantilever column subjected to an axial load is reproduced in Equation 6.1.1. It should apply to our experiment. Pcr =

2 EI EI 2.47 2 2 4L L

(6.1.1)

where E is Youngs modulus, I is the moment of inertia, and L is the free length of the stick. Buckling can be very dangerous in a structure. Structural designers need to understand well the buckling mechanism so they can prevent it in structures they design. Figure6.1.2a shows a cantilever column subjected to a horizontal load F at its free end. We ignore the self-weight of the column. An expression for the lateral deflection at the free end of such a column was developed in Section 2.8, Chapter 2. v= FL3 3EI (2.8.14)
293

6861X_C006.indd 293

2/7/08 [Link] PM

294

Understanding Structures: An Introduction to Structural Analysis

Binder clip Wood

Magnets

Push vertically

Styrofoam

(a)

(b)

(c)

Figure6.1.1 Experiment.

In order to help us understand the buckling phenomenon, we set up an analog column (Figure6.1.2a). The analog column is rigid throughout its height. It is supported on a pin at the base and its free end is maintained in position by a horizontal, linearly elastic spring attached to the end of the column and a fixed point. In effect, the flexibility of the entire column is concentrated in the spring. Inspection of Figure6.1.2 will reveal that the spring is analogous to the flexural stiffness of the cantilever column. The column resists the load F because of its flexural stiffness. The analogous column resists the load F with the help of the spring.

k F=k.v

(a) Cantilever column

(b) Rigid column with a spring

Figure6.1.2 System with horizontal spring.

6861X_C006.indd 294

2/7/08 [Link] PM

Buckling

295

Without the spring, it would topple over. To improve the analogy, we make the stiffness of the spring, k: k= 3EI L3 (6.1.2)

Note that F = kv is the same as Equation 2.8.14. Even though the lateral stiffness of the analog column is provided by a different mechanism, it mimics the cantilever column successfully. What would happen if you push the column in Figure6.1.2b at the top? If you could push the column at its exact cross-sectional center with a load that is exactly vertical, and if the column is perfectly straight and isotropic throughout its length, the column would remain as it is except for a small amount of shortening. However, such a setup is virtually impossible under practical conditions. The vertical is almost always eccentric with respect to the resistance axis of the column. At this time, we stop and introduce the definition of eccentricity. It is the perpendicular distance between the axes of column resistance and applied load. We can illustrate it in two dimensions as depicted in Figure6.1.3a. The axis of column resistance may be represented by the centerline of the column in Figure6.1.3a (assuming that the column is straight and the column section and material are uniform throughout). You will note that the axial vertical load P acts at a small distance, e, to the right of the centerline of the column at the point of application. The distance, e, is the eccentricity of the applied vertical load with respect to the center of column resistance. We return to the analogous column (Figure6.1.3a), which is loaded with an axial load P at eccentricity e. Because of the eccentricity, the axial load generates a clockwise moment at the top equal to Pe. In response to the applied moment, the analogous

P k A

v e P k

k.v

k.L

e/L = 0.001

e/L = 0.01 k.L 2

k.v P (a) Light load P

B 0 0.1L 0.2L v

(b) Heavy load

(c) Load-deection relationship

Figure6.1.3 System with horizontal spring.

6861X_C006.indd 295

2/7/08 [Link] PM

296

Understanding Structures: An Introduction to Structural Analysis

column tends to rotate clockwise. This tendency is resisted by the spring. Taking moments about the free end of the analogous column, P ( v + e ) = kv L (6.1.3)

Rearranging Equation 6.1.3 to define the deflection, v, at the top of the analogous column, v= Pe kL P (6.1.4)

The solid line and the broken line in Figure6.1.3c show the relationship between P and v for cases of large eccentricity (e/L = 0.01) and small eccentricity (e/L = 0.001), respectively. For the case of large eccentricity, the displacement v increases gradually as the load P increases. For the case of small eccentricity, the displacement v increases dramatically when the load P approaches the value of kL because the denominator in Equation 6.1.4 approaches zero. This phenomenon is similar to what we observed when we loaded the wooden stick in the experiment (Figure6.1.1). We call Pcr = kL the buckling load. If we substitute Equation 6.1.2, we obtain 3EI (6.1.5) 2 L This equation is similar to Equation 6.1.1 (the exact solution) except that the coefficient 3 is 20% larger than 2.47. Note that this equation is independent of the strength of the spring. The way we have set it up with our assumptions, the spring does not fail. The buckling strength of the analogous column is determined by the stiffness of the spring. If we remove the load, the moment P (v + e) will disappear and the spring will push back. The deflection at the top of the analogous column will return to zero. Therefore, we call this failure mechanism elastic buckling as it is referred to in engineering jargon even though it should be called buckling in the range of linearly elastic response. Pcr = kL = Exercise 6.1.1
In GOYA-U1, you can find a system with k = 0.5 N/mm and L = 200 mm. Fill in Table6.1.1 for the three cases: e = 0.1 mm, e = 1 mm, and e = 1 mm. Check your results using GOYA-U1. Figure6.1.4a shows another simple model with a rotational spring at the bottom of a rigid column. The stiffness of the spring is assumed as K= 3EI L2 (6.1.6)

If q is small enough so that v = L sin L , F = Kq /L is equivalent to Equation 2.8.14. Figure6.1.4b defines the eccentricity. Figure6.1.4c shows the deformation caused by the axial force. Moment equilibrium around the spring leads to P ( L + e) = K (6.1.7)

6861X_C006.indd 296

2/7/08 [Link] PM

Buckling

297

Table6.1.1 Load versus Deflection

Deflection v (mm) Load P (N) 0 20 40 60 80 95 e = 0.1 mm e = 0.1 mm e = 0.1 mm

or

=
from which

K L

P e P L K 3EI = 2 L L

(6.1.8)

Pcr =

(6.1.9)

Exercise 6.1.2
In GOYA-U2, you can find a system with K = 20 103 N-mm and L = 200 mm. Fill in Table6.1.2 for the case of e = 1 mm in terms of radians and degrees. Check your results using GOYA-U2.

v=L. F L

L.q

e P

K M=K. (a) Horizontal force P

K . P M=K (c) Heavy load

(b) Light load

Figure6.1.4 System with rotational spring.

6861X_C006.indd 297

2/7/08 [Link] PM

298

Understanding Structures: An Introduction to Structural Analysis

Table6.1.2 Load versus Rotation

Rotation q (e = 0.1 mm) Load P (N) 0 20 40 60 80 95 Unit: Rad. Unit: Deg.

Now we use a two-spring model that we had considered in Section 2.7, Chapter 2, to evaluate the deflection of a cantilever beam. Figure6.1.5a shows the model. Recall that each spring represents the flexural deformation of a length of beam equal to L/2. The relationship between the bending moments (MA and MB) and the rotations of the springs (aA and a B) are M A = K A and M B = K B where K= 2 EI L (6.1.10)

The free-body diagrams shown in Figure6.1.5b,c lead to P (vA + e) = M A = K A (6.1.11) (6.1.12)

P ( v A + vB + e ) = M B = K B

e P L/4 A L/2 aB B L/4

vB aB

vA e P aA A A

vA e P

vB

vA

e P

P.e MA

A P MA = KaA B P MB = KaB

MB

(a) Model

(b) Deformed shape

(c) Free-body

(d) Free-body

(e) Bending moment

Figure6.1.5 Two-spring model.

6861X_C006.indd 298

2/7/08 [Link] PM

Buckling

299

Figure6.1.5d shows the moment distribution along the height. We assume that aA and a B are small enough (sin A A and sin B B). From Figure6.1.5b, we obtain vA = ( A + B ) vB = B L 4 (6.1.13) (6.1.14)

L 2

Substituting Equations 6.1.13 and 6.1.14 into Equations 6.1.11 ad 6.1.12, L P ( A + B ) + e = K A 4 L L P ( A + B ) + B + e = K B 4 2 If we solve these equations in terms of aA and a B, we obtain (8K 4 PL ) Pe ( PL )2 8KPL + 8K 2 8KPe B = ( PL )2 8KPL + 8K 2 (6.1.15) (6.1.16)

A =

(6.1.17) (6.1.18)

Equations 6.1.17 and 6.1.18 have the same denominators. They will be zero if P= or P= 2 2 K EI 13.7 2 2 1 L L (6.1.20) 2 2 K EI 2.34 2 2 +1 L L (6.1.19)

Figure6.1.6 shows the relationship between the load P and the rotations of the springs (aA and a B) assuming EI = 4 106 N/mm2, L = 200 mm (or K = 40,000 N-mm), and
250 200

A B

Load (N)

150 100 50 0 0 1 2 3 4 5 6 Rotation (deg) 7 8

Figure6.1.6 Load-rotational relationship.

6861X_C006.indd 299

2/7/08 [Link] PM

300

Understanding Structures: An Introduction to Structural Analysis

Table6.1.3 LoadMomentDeflection
Bending Moment Load P (N) 0 50 100 150 200 220 MA (N-m) MB (N-m) Deflection v (mm)

e = 1 mm. We can see that the rotations increase dramatically as the load approaches K = 234 N (the value given by Equation 6.1.19). Note that Equation 6.1.19 is similar to Equation 6.1.1. The error for the approximate solution is only 5%.

Exercise 6.1.3
In GOYA-U3, you can find a two-spring model with L = 200 mm. Fill in Table6.1.3 for the case of EI = 4 106 N/mm2 and e = 1 mm. Check your results using GOYA-U3.

Example 6.1.1
Figure6.1.7a shows the plan of a 110-story skyscraper. Each floor is supported by 76 columns with cross sections shown in Figure6.1.7b. The columns are steel with a Youngs modulus of 30,000 ksi. The weight of floor per unit area is 200 lbf/ft2. The story height is 15 ft (Figure6.1.7c). Assume that the beams are much stiffer than the columns. Estimate the safety factor of the structure against buckling for the following two cases: (In this application, the safety factor is defined as the buckling strength divided by the axial load.) Case 1: Structure as it is (the solid line in Figure6.1.7c). Case 2: All the beams supporting the second and third floors are destroyed because of fire (the solid line in Figure6.1.7d). Furthermore, Youngs modulus of steel is reduced to 7,500 ksi because of the high temperature.

Hint: Calculate the horizontal stiffness against the force F shown in Figure6.1.7c,d. Use the method shown in Figure6.1.3. Solution
The total weight of the building is W = 110 60 2 1000 = 8.80 108 lbf

 We ignore the possibility of rotation of the building about its vertical axis.

6861X_C006.indd 300

2/7/08 [Link] PM

Buckling

301

16 in 200 ft 24 in 24 in

200 ft (a) Floor plan 13 ft 13 ft 13 ft 13 ft F 13 ft 13 ft (c) Before re (d) After re F (b) Section 13 ft 13 ft 13 ft 13 ft 13 ft 13 ft

Figure6.1.7 A skyscraper.

The moment of inertia of the column section is I= 24 4 164 = 22, 200 in 4 12 12

Recalling Equation 5.5.4, the horizontal stiffness for Case 1 is k= 12 EI 76 12 (30, 000 10 3 ) 22, 200 = = 1.60 108 lbf/in H3 (13 12 )3

The critical load for Case 1 is Pcr = kH = 1.60 108 (13 12 ) = 2.50 1010 lbf The safety factor is Pcr 2.50 1010 = 28.4 W 8.80 108

6861X_C006.indd 301

2/7/08 [Link] PM

302

Understanding Structures: An Introduction to Structural Analysis

The structure is quite safe against buckling. On the other hand, the horizontal stiffness for Case 2 is 12 EI 76 12 ( 7, 500 10 3 ) 22, 200 = = 1.48 10 6 lbf/in H3 (3 13 12 )3

k=

The critical load for Case 1 is Pcr = kH = 1.48 10 6 (3 13 12 ) = 6.93 108 lbf The safety factor is Pcr 6.93 108 = = 0.79 W 8.80 108 The structure should collapse as shown by the broken line in Figure6.1.7d. Collapse is caused by the gravity force, not by a horizontal force. Note that Pcr is proportional to EI/H2. Now that E is 1/4 and H is three times their values before the fire, the safety factor is (1/4) (1/3)2 = 1/36 of that in Case 1.

Example 6.1.2
A column is loaded in compression as shown in Figure6.1.8a. Estimate the buckling load assuming that the eccentricity e is small enough. (Hint: the column buckles as shown by the broken lines in Figure6.1.8a. Use the two-spring model shown in Figure6.1.8.)

P L/4 A

e P a

e P

P.e M

A L/2

A P M = Ka

L B

B L/4

a P

(a) Column

(b) Model

(c) Deformed shape (d) Free-body

(e) Bending moment

Figure6.1.8 Simply compressed column.

6861X_C006.indd 302

2/7/08 [Link] PM

Buckling

303

Solution
Figure6.1.8c shows the deformed shape of the column axis, where a denotes the rotation angle of springs A and B. Figure6.1.8d shows free-body diagram, which leads to Noting v = L/4, we get L P + e = K 4 If we solve this equation in terms of a , we obtain P (v + e) = K

e K PL 4

Assuming the denominator is zero, we conclude that Pcr = 4 K 8EI = 2 L L (6.1.21)

If we assume more than two springs, the coefficient 8 in Equation 6.1.21 will be larger and close to Eulers solution, which will appear in Section 6.2. Pcr =

2 EI 10 EI 2 L2 L

(6.1.22)

Note that the value is four times that in Equation 6.1.1 for a cantilever column.

Example 6.1.3
One end of a column is fixed and the other is supported by a vertical roller as shown in Figure6.1.9a. Estimate the buckling load using the two-spring model shown in Figure6.1.9b. (Hint: the column will buckle as depicted by the broken line in Figure6.1.9a.)

e R

v = aL 2 R L/4

e P 2a

v = aL 2

e P R

e P R A 3Ka

P .e

A L B L/4 L/2

3a

R P

A M = 3Ka

a B

R P (d) Free-body M = Ka

Ka

(a) Column

(b) Model

(c) Deformed shape

(e) Free-body

(f) Bending moment

Figure6.1.9 Fixed-end and roller-supported column.

6861X_C006.indd 303

2/7/08 [Link] PM

304

Understanding Structures: An Introduction to Structural Analysis

Solution
Figure6.1.9c shows the deformed shape of the column axis, where a denotes the rotation angle of spring B. The rotation angle of spring A is 3a . The free-body diagrams shown in Figure6.1.9d,e lead to L L P + e R = 3 K 2 4 L L R + P e = K 4 2 Eliminating the horizontal reaction R and solving in terms of a , we obtain

4 Pe 20 K 3PL

Assuming the denominator is zero, we conclude that Pcr = 20 K 40 EI = 3L 3L2 (6.1.23)

If we assume more than two springs, the coefficient 8 in Equation 6.1.21 will be larger and close to Eulers solution. Pcr = 2 2 EI 20 EI 2 L2 L (6.1.24)

Note that the value is eight times that in Equation 6.1.1 for a cantilever column.

6.2 Continuously Deformable Model

Having dealt with the problem of buckling using simple and rigid-discrete models, we are ready to derive Eulers equation in reference to a continuously deformable model. To simplify the notation, we shall refer to a horizontal member (Figure6.2.1a) or a cantilever column rotated through 90. In Figure6.2.1b, v0 denotes the deflection at the free end. Figure6.2.1c shows the deflected shape of the member axis, where v denotes the deflection of the member at a distance x from the fixed end. The freebody diagram shown in Figure6.2.1d determines the bending moment at any section with a known deflection v. M = P ( e + v0 v) (6.2.1)

Figure6.2.1e shows the bending-moment diagram corresponding to Equation 6.2.1. Recall the following equation derived in Section 2.8, Chapter 2, to describe the relationship between curvature and bending moment. d2v M = dx 2 EI (2.8.9)

 Leonhard Euler (17071783) was completely blind for the last seventeen years of his life, during which time he produced almost half of his total work output. He had extraordinary powers of memory and mental calculation.

6861X_C006.indd 304

2/7/08 [Link] PM

Buckling
L

305

e P 1.6 e v0 1.4 1.2 qL (rad)

e = 0.1 mm

(a) Very small load

(b) Large load e + v0 v x v (c) Deected shape e + v0 v v P P M (d) Free-body diagram P . (e + v0 v) e v0 P e v0

1.0 0.8 0.6 0.4 0.2 0 0 2

e = 1 mm

4 6 v0 (mm)

10

(f) Load-deection relationship

P . (e + v0)

P.e

(e) Bending moment

Figure6.2.1 Buckling of cantilever beam.

Replacing M as defined by Equation 6.2.1, we obtain d 2 v P ( e + v0 v) = dx 2 EI (6.2.2)

If we let q = P/EI ,

d2v + q 2 v = q 2 ( e + v0 ) dx 2

(6.2.3)

The solution of the differential equation is of the form v = C1 sin qx + C2 cos qx + e + v0 (6.2.4)

where C1 and C2 are constants of integration.

 The second derivative of Equation 6.2.4 is d2v = C1 q 2 sin qx C2 q 2 cos qx dx 2 (6.2.4)

If we substitute Equations 6.2.4 and 6.2.4 into Equation 6.2.3, we find that Equation 6.2.3 is satisfied. Therefore, we conclude that Equation 6.2.4 is the solution.

6861X_C006.indd 305

2/7/08 [Link] PM

306

Understanding Structures: An Introduction to Structural Analysis

To evaluate the constants C1 and C2, we use the boundary conditions v = 0 and dv/dx = 0 at x = 0 (at the fixed end), from which we get C1 = 0 and C2 = ( e + v0 ). Substituting these values into Equation 6.2.4, we obtain v = ( e + v0 )(1 cos qx ) (6.2.5)

To determine the deflection at the free end, v0, we use the other boundary condition: v = v0 at x = L (at the free end). This leads to 1 v0 = e 1 cos qL (6.2.6)

Figure6.2.1f shows the relationship between qL and v0 for the case of L = 100 mm. The value qL is indicated in terms of radians (no physical unit) because the unit of q = P/EI is a reciprocal of length. Note that v0 increases to infinity as qL approaches /2 1.57 even if the eccentricity e is very small. The reason for this result is that cosqL in Equation 6.2.6 approaches zero. Substituting q = P / EI into qL = /2 and solving for P, we get Eulers equation for the buckling load of a cantilever column. Exercise
In GOYA-U4, you can find the case illustrated in Figure6.2.2a: a 200-mm long column with a cross section of 10 mm 15 mm and a Youngs modulus of 1000 N/mm2.

Pcr =

2 EI 4 L2

(6.2.7)

10 mm 15 mm z y

200 mm x

(a)

(b)

Figure6.2.2 Column.

6861X_C006.indd 306

2/7/08 [Link] PM

Buckling

307

AU: Table title?

Table6.2.1
qL Load P (N) 0 25 50 75 q (/mm) Unit: Rad. Unit: Deg. cosqL v0 (mm) e = 1 mm e = 0.1 mm

Because the moment of inertia is smaller around the y-axis, the column deflects in direction x as shown in Figure6.2.2b. Calculate the moment of inertia around the y-axis and calculate the buckling load. Fill in Table6.2.1 for the two cases: e = 0.1 mm and e = 1 mm. Check your results using GOYA-U4.

Example 6.2.1
Take a spaghetti strand 200 mm (~8 in.) long with a diameter of 1.6 mm (~0.06 in.). Assume that Youngs modulus is 2000 N/mm2, the compressive strength is 200 N/mm2, and the tensile strength is 20 N/mm2. Compute the buckling loads for the spaghetti strand for the three types of loading shown in Figure6.2.3. Check your results in the kitchen using a scale.

100 200 F

100

F (a) Compressive force

F (b) Tensile force (c) Bending force

Figure6.2.3 Loading of spaghetti.

6861X_C006.indd 307

2/7/08 [Link] PM

308

Understanding Structures: An Introduction to Structural Analysis

P P P

v0

2L

v0 P v0

P (a) Deected shape

(b) Bending moment

(c) Cantilever column

P (d) Test apparatus

Figure6.2.4 Pin-Supported column.

Solution
Moment of inertia of the spaghetti strand is I=

r 4 3.14 0.82 = = 0.32 mm 4 4 4

Figure6.2.4a shows the deformed shape of the strand, with length 2L, after buckling. Figure6.2.4b shows the bending-moment diagram, where eccentricity is assumed to be negligible. The condition is equivalent to that of a cantilever column of length L shown in Figure6.2.4c. Therefore, the buckling load is Pcr =

2 EI 3.14 2 2000 0.322 = = 0.16 N 4 L2 4 100 2

(or 0.6 oz)

The cross-sectional area of the spaghetti strand is A = r 2 = 3.14 0.82 = 2.0 mm 2

6861X_C006.indd 308

2/7/08 [Link] PM

Buckling

309

The force required to break the strand by pulling is the product of the cross-sectional area and the tensile strength. F = A = 2.0 20 = 40 N Section modulus of the spaghetti is Z= I r 3 3.14 0.83 = = = 0.4 mm 3 r 4 4 (or 8.9 lb)

Flexural strength is determined using the bending strength ( M = Z ) and the span length L.
AU: Twice what?

F=

2 M 2 Z 2 0.4 20 = = = 0.16 N L L 100

(or 0.6 oz)

If you test a 100-mm-long spaghetti, the flexural strength will be twice, whereas the buckling strength will be four times. In GOYA-U5, you can have a test apparatus shown in Figure6.2.4d. We use such an apparatus when we conduct a test of buckling of an I-shaped steel column. Isaac Newton (16431727) showed that the laws of physics observed on Earth are also observed in space. You should understand that the laws of mechanics observed in the kitchen using a spaghetti strand are also observed in skyscrapers that use steel columns.

Example 6.2.2
If you keep loading the spaghetti strand in Example 6.2.1 after buckling, it will break. Estimate the deflection v0 at fracture. (Hint: The spaghetti strand will break if the maximum tensile stress reaches its tensile strength. Look in GOYA-U5 and see the stress distribution in the section at midspan.)

Solution
Figure6.2.5a shows the stress distribution of a column subjected to pure compression, where P is assumed positive though in compression. If we apply a small bending moment, the stress distribution will be trapezoidal as shown in Figure6.2.5b (Section 5.1, Chapter 5). If we increase the bending moment, the stress distribution will be as shown in Figure6.2.5c. If the maximum tensile stress reaches the strength, the column will fail.

P M + A Z

(6.2.8)

Substituting M = Pv0 and = 20 N/mm 2 into Equation 6.2.8, and solving for v0 , we get P Z 0.159 0.402 v0 = + = 20 + = 50.8 mm A P 2.01 0.1 159 The maximum compressive stress is (or 2.0 in.)

P M P Pv 0.159 0.159 50.8 = 0 = = 0.08 20.1 = 20.2 N/mm 2 A Z A Z 2.01 0.402

6861X_C006.indd 309

2/7/08 [Link] PM

310

Understanding Structures: An Introduction to Structural Analysis

Uniform compression P/A P/A

(a) M = 0 Uneven compression P/A + M/Z P/A M/Z

(b) M is applied Tension

P/A + M/Z

P/A M/Z Compression (c) M is increased

Figure6.2.5 Stress distribution.

Note that the magnitude is almost equal to the maximum tensile stress, = 20 N/mm 2. The stress distribution in a buckled column is quite similar to that in a beam without axial force.

Exercise 6.2.1
Compute the deflection v0 at fracture for a 100-mm-long spaghetti strand. The answer will be about 1/4 of that obtained in Example 6.2.2.

Exercise 6.2.2
Compute the deflection of the spaghetti strand v0 at fracture under the loading shown in Figure6.2.6 for cases of L = 100 mm and 50 mm. The answers will be smaller than those for buckling, but not very different, because the bending-moment distributions are similar.

L F v0

Figure6.2.6 Simply supported beam.

6861X_C006.indd 310

2/7/08 [Link] PM

Buckling

311

20 ft 20 ft A C 20 ft 40 ft E

B D Snow F

40 ft

40 ft

G 20 ft H 20 ft J (a) Perspective view (b) Elevation

Figure6.3.1 Canopy in snowy region.

6.3 Problems
6.1 Figure6.1.8a shows a canopy that extends out from a building. We may assume that the building is stiff and strong enough not to fail under the conditions specified. The canopy is supported by three beams: AB, CD, and EF. You may assume the beams to be simply supported (Figure6.1.8b). Each beam is supported by a steel pipe-column having an outside diameter of 8.0 in. and an inside diameter of 6.0 in. The connections at both ends of each column may be assumed to be pins. Assume that Youngs modulus for steel is 30,000 ksi. The building is in a snowy region, and the maximum possible snow load per unit area of the canopy is estimated to be 1,000 lbf/ft2. Assume that the self weights of the canopy and the beams are negligible, and the tributary area of column CH is 20 20 = 400 ft2 as indicated by the broken lines in Figure6.3.1a. Check if column CH is safe against buckling. 6.2 Assume that we make the connections between the foundation beams and the columns continuous so that the columns may be assumed to be fixed at their bases. Check if the columns are safe against buckling under the specified snow load. 6.3 Assume that the columns are fixed top and bottom. Check if the columns are safe against buckling under the specified snow load.
 The tributary area for a particular column is defined as the area that contributes load to that column. In this case, we assume that the moment restraints on the slab across lines AB, CD, and EF are similar. Accordingly, half the load on the slab goes to beam CD. Because beam CD is supported simply, half of the load it carries goes to support C. The tributary area is one-fourth of the area ABEF or 20 20 = 400 ft2.

6861X_C006.indd 311

2/7/08 [Link] PM

6861X_C006.indd 312

2/7/08 [Link] PM

Au: Please check the numbring of figures cited or Display maths also.

Moment and Deflections in Cantilever Beams

2.1 What is a Bending Moment?

Chapter 1 dealt with trusses. Trusses are used mainly for large structures such as bridges and domes. In this chapter, we shall study beams which are used not only in bridges but also in buildings. A beam is a horizontal member supporting a roof or a floor as shown in Figure 2.1.1a. A beam that projects out from a support and has a free end is called a cantilever beam. To understand the phenomenon of bending of beams in simple steps, we shall start by studying strength and deformation properties of a cantilever beam embedded in a rigid wall (Figure 2.1.1b). Start GOYA-C to reach the window in Figure 2.1.2 showing a cantilever beam with a load of 10 N at its free end. You will find a sliding bar titled Load Magnitude at the bottom of the window. Using your pointer, move the bar to the left to obtain -10 N. The beam will deflect down as shown in Figure 2.1.3. How does the beam deflect if it is subjected to the set of forces with the same magnitude (Figure 2.1.4a)? Which of the four deflected shapes in Figures 2.1.4b through e would you guess is the correct one? To check your guess, do the following. 1. Click the Add Load button to create another load pushing up at the end of the beam. The deflection will disappear (Figure 2.1.5). 2. Click on the downward load and drag it to mid-span (Figure 2.1.6).

The entire beam deflects up! Let us now investigate in detail how the beam bends. Click the Zoom in button and type 4 in the text field titled Amplification to get the image shown in Figure 2.1.7. You will find that the yellow segment rotates and moves up and that the top of the yellow segment shortens while the bottom lengthens. Click the Zoom out button and type 1 in the Amplification text-field. Look at the window in the upper right (Figure 2.1.8) showing the deformation of the highlighted (yellow) segment. This segment has been removed from the beam and rotated so that its y-axis is vertical. Also, the flexural deformations (the shortening of the top and lengthening of the bottom) have been magnified by a factor of ten. Why does the top of the segment shorten while the bottom lengthens? The answer can be obtained if you check the Free-body Diagram item in the bottom right to obtain the image shown in Figure 2.1.9. GOYA-C shows a free-body diagram of the beam to the left of the broken line in Figure 2.1.10a. The two external forces tend to rotate the beam to the left of the cut clockwise (Figure 2.1.10b). To prevent the rotation, there should be a counterclockwise action M to the right of the highlighted segment (Figure 2.1.10c). This action is called a bending moment. The moment is
79

6861X_C002.indd 79

2/7/08 [Link] PM

80

Understanding Structures: An Introduction to Structural Analysis

Roof Beam Floor Column Rigid wall Column Foundation (b) Cantilever beam considered in Chapter 2

Cantilever beam

Beam Floor Foundation beam (a) Section of a building

Figure 2.1.1 Cantilever beam.

11.85

10 1000 10

Figure 2.1.2 Deflected shape of cantilever as seen in first window of GOYA-C.

10
10

11.85
1000

Figure 2.1.3 Deflected shape for a load of -10 N applied at free end of a cantilever beam.

6861X_C002.indd 80

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

e entire beam deects down. Only the center deects down. Left half of beam deects up. e entire beam deects up.

81

(b) a

(c)

F (a) Two forces

(d)

(e)

Figure 2.1.4 Quiz: a cantilever beam with two forces.

FD 10 FU 10 0 0

Figure 2.1.5 Click Add load button.

FD 10 8.15

0 500

FU 10

Figure 2.1.6 Drag the load to mid-span.

6861X_C002.indd 81

2/7/08 [Link] PM

82

Understanding Structures: An Introduction to Structural Analysis

10 y Shorten

0 500 Lengthen

Figure 2.1.7 Click the Zoom in button and type 4 in the Amplification text-field.

y 50 0 50 50 50 0

Figure 2.1.8 Flexural deformation.

FD 10 8.15

0 0 500

FU 10

Figure 2.1.9 Free-body diagram to the left of the highlighted segment.

6861X_C002.indd 82

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

a a F h F a Cut (a) Cantilever beam F F

83

C=T T

2 h 3

(d) Forces acting on the section Compression

F (b) Moment by the external forces a F M F (c) Bending moment in the section M 2 3h C=T T

Tension (e) Deformed shape C=T T M (f )Equilibrium of a segment (g) Bending moment

Figure 2.1.10 Cut the beam.

defined as a force multiplied by a distance (see Section 1.6). The magnitude of the bending moment is given by Equation 2.1.1. M = F.a (2.1.1) In GOYA-C, the width of each segment is 10 mm, so that the length between the forces is a = 50 mm (Figure 2.1.9). From Equation 2.1.1, the moment M is 500 N-mm (M = F.a = 10 50 = 500 N-mm.) Move the location of the forces to the left or to the right to obtain your own perspective of the variations in bending moment and deflected shape of the beam. Recall that the beam deflection increased when you moved the force FD to the right in Figure 2.1.6, suggesting that an increase of the distance a in Equation 2.1.1 results in a larger bending moment and larger deflection. Using GOYA-C, test to find out if this suggestion is correct. Move this force to the left or right. Figure 2.1.10c shows a free-body diagram of the beam with the moment M required to balance the moment generated as the product of F and a. The moment M is essentially a couple as illustrated in Figure 2.1.10d. It comprises two equal and opposite forces at a distance (2/3)h from one another, where h is the beam height.


Strictly speaking, tensile and compressive forces acting on the beam cross section vary linearly over the height of this rectangular cross section starting from zero at mid-height and reaching maxima at the extreme fibers in tension and compression (Figure 2.1.11a). The distribution of the bending stresses will be described in section 2.4. In this section, these distributed forces are lumped in concentrated forces shown in Figure 2.1.11b for simplicity. The lumped forces are assumed to act at the centroids of the distributed forces, so that the distance between the concentrated forces is two thirds of the height of the beam, (2/3)h. The height of the beam is selected to be h = 15 mm for the particular beam we are considering. Substituting F = 10 N, a = 50 mm and h = 15 mm into (2/3)T.h = F.a, we have T = 50 N as indicated in Figure 2.1.8. In a cross section subjected to bending only, if the lumped tensile force is 50 N, the lumped compressive force is also -50 N. One has to balance the other.

6861X_C002.indd 83

2/7/08 [Link] PM

84

Understanding Structures: An Introduction to Structural Analysis

10 mm 10 mm
C=T

h = 15 mm (a) Actual distribution

2 h = 10 mm 3
T (b) Concentrated representation

Figure 2.1.11 Forces acting on the highlighted segment.

Equilibrium in the horizontal direction requires T = C. The product (2/3) hT or (2/3) hC is equal to M. The forces, T and C, cause tensile and compressive deformations of the beam (Figure 2.1.10e). To understand their effect, cut the beam again at the left end of the yellow segment as shown in Figure 2.1.10f, where equilibrium in the horizontal direction requires the same magnitude of forces at both faces of the segment. These pairs of tensile and compressive forces create moments causing the flexural deformation (i.e. the lengthening at the top and the shortening of the bottom). The moments are represented by round (bent) arrows as shown in Figure 2.1.10g. Equilibrium of a beam is similar to that of a truss. Look at the truss in Figure 2.1.12a. It is connected to the wall at its right end and is analogous to the cantilever beam shown in Figure 2.1.12b. Cutting the truss as shown in Figure 2.1.12c
10 10 10 50 50

(a) Truss 10

15 10 (b) Equivalent cantilever beam 10 10 10 (c) Free body and equilibrium 50 50

Figure 2.1.12 A truss similar to a cantilever beam.

6861X_C002.indd 84

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

85

10 20 30 10 14.1 40 10 14.1 50 10 14.1 50 50 10 14.1 0 50 10 14.1 10 10 0 20 0 0 0 0 0 30 40 50 50 50

50 0 50 0

50 0 50 0

Figure 2.1.13 Analysis using GOYA-A.

and considering moment equilibrium, you will find that the top chord carries the tensile force of 50 N and the bottom chord the compressive force of 50 N. This is equivalent to the equilibrium condition depicted in Figure 2.1.10d. Figure 2.1.13 shows the results of the analysis of the truss in Figure 2.1.12a obtained using GOYA-A. The deflected shape of the truss is also similar to that of the cantilever beam. Experiment
Make a cantilever beam using a plastic ruler as shown in Figure 2.1.14, and apply a pair of forces using your thumb and little finger. If you can make the magnitude of the forces equal, the ruler will bend up.

Exercise
Compare Figures 2.1.15a and b. Which beam has larger bending moment at the highlighted segment? Which beam deflects more?

Answer
The bending moments are the same because the distances between the vertical forces are the same. However, because beam (a) has a longer deformed region than beam (b) does, the deflection of beam (a) is more than that of beam (b).

Ruler Push

Hard box or book

Pull

Desk

Figure 2.1.14 Experiment using a plastic ruler.

6861X_C002.indd 85

2/7/08 [Link] PM

86

Understanding Structures: An Introduction to Structural Analysis

F

F (a) F

F (b)

Figure 2.1.15 Bending moment.

Exercise using GOYA-C:

Click the Details of Beam button to obtain the window below specifying the length, width and height of the beam. Change the height to 7.5 mm, click the update button, and click the main window. You will obtain Figure 2.1.17. Note that the tensile and compressive forces on the yellow segment double from 50 to 100 N because the distance between the forces ((2/3)h in Figure 2.1.11) is halved from 10 to 5 mm, while the bending moment M = T.(2/3)h remains the same. The increase of the forces causes larger flexural deformation (8 times). This phenomenon will be discussed in Section 2.8.

Design your own beam (Part 1) We want to design a beam complying with the following conditions. 1. The magnitude of the external forces (F in Figure 2.1.18) shall be any single digit you choose plus 10 lbf.

Figure 2.1.16 Window specifying the details of the beam.

6861X_C002.indd 86

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

87

10 65.19 0 10 500 0 100 100

100 100

Figure 2.1.17 Beam with smaller height.

2. The distance between the forces shall be any single digit you choose plus 30 in. 3. The tensile force in the beam (T in Figure 2.1.18) shall not exceed 50 lbf. What is the required beam height? Check your result using GOYA-C.

Large distance makes large moment.

Fa F a F

Bending moment is an action that bends a beam.

T < 50 lbf T C T

F = any number + 10 lbf h?

a = any number + 30 in

Figure 2.1.18 Your beam.

6861X_C002.indd 87

2/7/08 [Link] PM

88

Understanding Structures: An Introduction to Structural Analysis

Bending moment and deformation Hi, Joan. I still do not understand the action of a bending moment. I know that a tensile axial force elongates each atom in a member. Does a bending moment bend each atom?
Tension Axial force Axial force

Tension

No, it does not. Each atom in a bent member has a pair of tensile or compressive forces so that it elongates or shortens. The uppermost atom has the largest tensile forces, while the lowermost atom has the largest compressive force.
Tension M Compression M

Hmm, if the distances between atoms elongate or shorten, the beam must be a trapezoid, and will not bend, I believe.

Tension M Compression M

Not true. Each atom is so small that you should slice the beam into very thin slices to understand what is happening. Each slice will be a trapezoid, and the beam will bend. Bend your eraser with your fingers. You will observe flexural deformation.
Tension M Compression M

6861X_C002.indd 88

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

89

By the way, I dont like your figure. Atoms do not deform in reality. The distance between atoms elongates or shortens.

Dont worry. I like it.

2.2 What is a Shear Force?

In section 2.1, we studied a case with two loads applied to a beam. In this section, we shall study a case with one load applied as shown in Figure 2.2.1a. Turn on GOYA-C and click Fade to expose the forces within the beam (Figure 2.2.1b). As well as the bending moment
M

V (a) Left of section

(b) Shear

of 800 N-mm, you will find a force of 10 N acting down. This force balances the external load of 10 N at the end of the beam. This type of internal force acting on the section is called a shear force and is usually denoted by the letter V. Note that the distance between the external load and the section we have cut is 80 mm (8 segments). The external load and the shear force cause a clockwise moment of 10 80 = 800 N-mm. This moment, acting on the beam, is balanced by the internal (resisting) bending moment of equal magnitude but opposite sense at the cut.


Similar to axial force or bending moment, you can represent and see a shear force only if you cut the beam and consider equilibrium at the section of the cut. While the part of the beam to the left of the cut is subjected to a downward shear force and counterclockwise bending moment (Figure 2.2.1b), the part of the beam to the right of the cut is subjected to an upward shear force and clockwise bending moment as shown in Sketch (a) on the right. Shear force is not a vector but a set of two forces shown in Sketch (b) on the right. Its effect is similar to that of a pair of shears (or scissors). We shall revisit this topic at the end of this section.

6861X_C002.indd 89

2/7/08 [Link] PM

90

Understanding Structures: An Introduction to Structural Analysis

11.85 10 1000 10 (a) Intial window

10 11.85 10 800 10 (b) Free body

Figure 2.2.1 Beam with load at the free end.

Move your cursor to any grid and click. The shear force remains at 10 N, while the bending moment varies. For example, at a distance of 40 mm from the beam end, the moment acting on the section is 400 N-mm (Figure 2.2.2). Let the distance from the free end to the section be x (Figure 2.2.3a). Equilibrium of moments of the free body shown in Figure 2.2.3b leads to M = V.x (2.2.1)

Bending moment M is distributed as shown in Figure 2.2.3c: it is zero at the free end (M = 0 at x = 0) and increases linearly to the fixed end (M = V.L at x = L). The symbol in the diagram is used to indicate that the bending moment compresses the top fiber of the beam.

10 11.85 10 400 10

Figure 2.2.2 Internal Shear Force and Bending Moment at A Section.

6861X_C002.indd 90

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

L

91

F x V=F M F (b) Free body

(a) Cantilever beam

Vx

VL C V (c) Bending moment

0 A M V B
Au: please provide Figure missing Captions

M + dM (d) Equilibrium

dx

2.2.3

Differentiating Equation 2.2.1 with respect to x, we obtain dM/dx = V (2.2.2) This relationship is very important because it is valid for any beam or column subjected to any set of loads. Equation 2.2.2 states that the rate of change of moment with distance is the shear force. Conversely, the integration of shear over a distance results in moment. We can also derive Equation 2.2.2 considering equilibrium of a short length (length = dx) shown in Figure 2.2.3d which corresponds to the highlighted (yellow in GOYA) segment in Figure 2.2.3b. Equilibrium of the forces in the vertical direction requires the existence of a downward shear force V in section CD (Figure 2.2.3d). Call the bending moment in section CD (M + dM) and consider the equilibrium of moments as follows. Bending moment at section AB: M, clockwise; Bending moment at section CD: M + dM, counterclockwise; and


Bending moment is an action that bends beam and does not have a sense (direction). However, if you cut the beam and look at the section at the cut, the bending moment has a sense (clockwise or counterclockwise). Recall that an internal axial force also has a sense only when you cut the member and consider equilibrium at the section of the cut.

6861X_C002.indd 91

2/7/08 [Link] PM

92

Understanding Structures: An Introduction to Structural Analysis

1000

(a) Bending moment

10

(b) Shear Force

Figure 2.2.4 Bending moment and shear force diagram.

Moment caused by the shear forces in sections AB and CD: [Link], clock wise. Defining clockwise moments to be positive, we have the following equation: M - (M + dM ) + [Link] = 0 (2.2.3)

which again leads to dM/dx = V. In the lower part of the window, you can find the distribution of the bending moment (linear) and the shear force (constant). Because the length of the beam is 100 mm, the slope of the line denoting the variation of moment along the span of the beam is 1000/100 = 10 N and is equal to the shear force. Click at various positions along the beam to see the flexural deformation of each grid as shown in Figure 2.2.5. You will find that the flexural deformation is larger in segments closer to the wall, because the bending moment is larger there. If you apply an upward force at the end of a wood beam as shown in Figure 2.2.6, it will break at the fixed end because the bending moment is largest there. If you increase the length L, the breaking force required will be smaller because, for the same force, the bending moment at the fixed end will be larger. This phenomenon is equivalent to the action of a crow-bar (Figure 2.2.7). If you increase L, the force pulling the nail T will increase. Note that the sense or direction of action of the tensile force T is shown as being to the right because the force

30 40 10 30 40 F 10 10

70 80 10 70 80

Figure 2.2.5 Deformation of each grid.

6861X_C002.indd 92

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

93

Wood beam F Breaks at xed end L

Pen case Desk

Figure 2.2.6 Breaking a wood beam.

represents the resistance of the nail. The wall pushes against the heel of the crow-bar with a force having a magnitude of C (= T ). The wall also applies a downward force equal to F on the crow-bar. Move the load to the middle of the beam as shown in Figure 2.2.8. You will notice that the bending moment becomes zero to the left of the load. Figure 2.2.9 provides the reason. If you cut the beam to the left of the force as shown in Figure 2.2.9a, you will find that the part to the left of the cut is free of any force: it moves upward but does not deform. If you cut the beam at a distance of x to the right of the force as shown in Figure 2.2.9b, you will find that the section resists a shear force of V = F and a bending moment of M = V.x. Click Add load. As shown in Figure 2.2.10, you will find that the shear force distribution (bottom right of the window) changes abruptly at the loading point. You will also find that the slope of the bending moment changes at the loading point (bottom left of the window). We can explain this change by superposing the effect of F1 on that of F2 (Figures 2.2.11a and b). Note that the slope of the bending moment between points A and B is dM/dx = 500/50 = 10 N and the slope between points B and C is dM/dx = (1500500)/50 = 20 N. Each slope agrees with the shear force in each region. Click load F2 and change it to -10 N as shown in Figure 2.2.12 using the bar in the bottom right. This is the loading we studied in Section 2.1. Note that the shear is

C=T F F L C=T F F T T

2 3h
Ouch! Craw-bar

Dont put your nger between the wall and the craw-bar.

2 3h

Compression

Figure 2.2.7 Equilibrium of crow-bar.

6861X_C002.indd 93

2/7/08 [Link] PM

94

Understanding Structures: An Introduction to Structural Analysis

3.7

10 500

10 500 0 0 0 10

Figure 2.2.8 Move the load to the middle.

F x V=F F M=Vx

(a) Cut to the left of the load

(b) Cut to the right of the load

Figure 2.2.9 The beam cut at two sections.

15.56 20 F1 10 1500 0 500 10 20 F2 10 1500

Figure 2.2.10 Beam with two loads.

6861X_C002.indd 94

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

Eect of F1 Eect of F1

95

1000 N-mm

10 N

+
Eect of F2 500 N-mm

+
Eect of F2 10 N

500 N-mm A 50 mm B 50 mm C

1500 N-mm

50 mm

50 mm

20 N

(a) Bending moment diagram

(b) Shear force diagram

Figure 2.2.11 Superposition.

zero in the right half of the beam. Also note that the bending moment is constant in the right half of the beam because dM/dx = V = 0. Click at various points along the right half of the beam and find that each segment deforms the same amount because the bending moment is constant. The bending moment and the shear force diagrams in Figure 2.2.12 can again be explained by the superposition shown in Figure 2.2.13. Note that the slope of the bending moment between points between points B and C is dM/dx = (500 -500)/50 = 0 N, which agrees with the shear force in the region. Look at the truss in Figure 2.2.14. Between points A and B, the top and bottom chords resist axial forces of -50 N (compression) and +50 N (tension), respectively. This corresponds to uniform bending moment. Between points A and B, the inclined members do not resist any axial load because the shear force is zero. The axial forces in the top and bottom chords decrease from a maximum at B to a minimum at C, while the axial forces in the inclined members remain constant a 10 2 N, balancing to the uniform shear force of 10 N. Next, go back to the beam and change force F2 to -20 N as shown in Figure 2.2.15. The bending moment at the wall becomes zero, and the shear force in the right half

F2 10 8.15 0 500

F1 10 500 0 500 10 0

Figure 2.2.12 Beam with middle load of -10 N.

6861X_C002.indd 95

2/7/08 [Link] PM

96

Understanding Structures: An Introduction to Structural Analysis

Eect of F1 Eect of F1

1000 N-mm

10 N

+
Eect of F2 500 N-mm

+
Eect of F2 10 N

50 mm

50 mm

C 500 N-mm

50 mm

50 mm

10 N

(a) Bending moment diagram

(b) Shear force diagram

Figure 2.2.13 Superposition.

10

10 C 10 B A

50

50

10 20 30 10 14.1 40 10 14.1 50 10 14.1 50 50 10 14.1 0 50 10 14.1 10 10 0 0 0 20 0 0 0 30 40 50 50 50

50 0 50 0

50 0 50 0

Figure 2.2.14 Cantilever truss.

F2 20

4.44

10 F1 10 500 10 0 10 0

Figure 2.2.15 Change the middle load to -20 N.

6861X_C002.indd 96

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

L/2 2F Pen case F Breaks at the center Desk L/2

97

Figure 2.2.16 Break a wood beam by forces of 2 to 1.

of the beam becomes -10 N (negative). Recall the equation, dM/dx = V, again. The bending moment decreases at the rate of -10 N in the right half of the beam. If you apply a set of forces to a wood beam as shown in Figure 2.2.16, the bending moment has the largest value at mid-span so that the beam will break at the middle as a result of the fracture of the bottom fiber. In general, we can predict where a beam will break using the bending moment diagram. In Chapter 1, we defined the axial force to be positive if tensile and negative if compressive. Here, we shall define the shear force to be positive if it acts clockwise as shown below. Also, we shall define the bending moment to be positive if it compresses the top fiber of the beam. Click and change the force F2 to - 40 N as shown in Figure 2.2.17. The shear force in the right half of the beam becomes -30 N. Because dM/dx = -30, the bending moment at the face of the wall becomes negative, indicating that the bottom fiber is in compression. You should remember that the position of the bending moment diagram with respect to the axis of zero moment indicates which face of the beam is compressed edge. If the moment diagram is below the line of zero moment, the bottom edge of the beam is in compression. If the moment diagram is above the line of zero moment, the top edge of the beam is in compression. If you apply a set of forces on a wood beam as shown in Figure 2.2.18 so that the top force is four times as large as the bottom force, the bending moment will have

Table 2.1 Notations (in art review)

Axial force Tension + N N V Shear force Clockwise V Bending moment Top compression

V Bottom compression

Compression

Counter-clockwise

6861X_C002.indd 97

2/7/08 [Link] PM

98

Understanding Structures: An Introduction to Structural Analysis

F2 2.96 F1 10 500 40 30
1000

0 1000

10 30

Figure 2.2.17 Change the middle load to - 40 N.

Breaks at 4F the xed end

L/2

L/2

Figure 2.2.18 Break timber by forces of 4 to 1.

F2 7.78 F1 10 750 0 0 10

40 30 0

30

Figure 2.2.19 Move the load to the right.

6861X_C002.indd 98

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

99

the largest value at the fixed end and the beam will break at that point as a result of tensile failure of the top fiber. Click and move the force F2 to the right as shown in Figure 2.2.19. When it reaches a point 25 mm (two and a half grids) from the fixed end, the bending moment at the wall becomes zero. Note that dM/dx = 750/75 = 10 and dM/dx = -750/25 = -30 to satisfy dM/dx = V again.
M = 50(i + 10)

1000 50 (a) M = 50(i + 10) 200 50

50 (b)

50

Exercise 2.2.1 using GOYA-C

Use any number i you may choose to define the moment as shown below. Find the loads that cause the two bending-moment diagrams below. And sketch the deformed shapes for the two cases.

Exercise 2.2.2 using GOYA-C

If you select the locations and the magnitude of the forces carefully, you may create interesting bending moment diagrams as those illustrated below. Make up your own diagram, and report the corresponding loads and the deformations. If the deformation of your beam becomes too large, you can adjust it decreasing the number in the Amplification text-field in the bottom left.

(a) Mount Fuji

(b) Seismic wave

(c) Thaw

(d) Pyramids in Giza

6861X_C002.indd 99

2/7/08 [Link] PM

100

Understanding Structures: An Introduction to Structural Analysis

10 N D 2m C 2m 10 N A B 2m 10 N

Figure 2.2.20

Example 2.2.1
Calculate the shear force and the bending moment diagrams for the loads shown in Figure 2.2.20, and sketch the deformation of the beam.

Solution
Cut the beam between the points A and B. From the conditions of equilibrium, you will find that the shear force is -10 N at any section between A and B (negative because it is counter-clockwise). Cuts between B-C and C-D yield the shear force diagram in Figure 2.2.21a. Noting that the bending moment at the free end is zero and the slope is dM/dx = -10 between A and B, we get M = -10 (-2) = 20 N-m at B. The bending moment between B and C should be constant because the shear force is zero from B to C. Using the same logic we get M = 0 at D. The bending moment diagram in Figure 2.2.21b indicates that the top fiber of the beam is in compression between A and D. Noting that the fixed end of the beam

10 N +10 N (a) Shear force Slope = 10 N +20 N m D C B Slope = 10 N A

(b) Bending moment Right angle

(c) Deformation Acute angle

(d) Wrong answer

Figure 2.2.21

6861X_C002.indd 100

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

101

B C 2m

+20 N D 3m

1m

Figure 2.2.22 is perpendicular to the wall, we can reason that the beam will deform as shown in Figure 2.2.21c. The deflected shape shown in Figure 2.2.21d is not correct because the deflected beam makes an acute angle with the wall.

Example 2.2.2
Calculate the loads and the bending moment diagrams corresponding to the shear force diagram shown in Figure 2.2.22, and sketch the deflected shape of the beam.

Solution
Because the shear force changes at load points, there should be applied loads of 20 N at points B and C (and nowhere else). Noting that the shear force is positive (or clockwise), we conclude that the direction of the loads should be as shown in Figure 2.2.23a. Because there is no load between A and B, the bending moment is zero from A to B. It increases at the rate of dM/dx = 20 between B and C to reach 20 N-m at C. Between C and D, it is constant because V = 0 from B to C. Noting that the bending moment is positive between B and D,we reason the beam will deform as shown in Figure 2.2.23c. Note that the base of the beam is

20 N

20 N (a) Loads +20 N m A B C D

(b) Bending moment Straight Right angle

(c) Deformation

Figure 2.2.23

6861X_C002.indd 101

2/7/08 [Link] PM

102

Understanding Structures: An Introduction to Structural Analysis

A 3m 10 N B 3m

Figure 2.2.24

perpendicular to the wall. The beam does not deform (remains straight) between A and B because the bending moment is zero in that portion of the beam.

Example 2.2.3
Calculate the shear force and the bending moment corresponding to the loads shown in Figure 2.2.24, and sketch the deflected shape of the beam. (Hint: Recall that the sense of the bending moment diagram gives us a clue as to which faceleft or right of the beam is compressed.)

Solution
The portion between A and B is free of shear force and bending moment. Between B and C, the load induces a clockwise shear force (Figure 2.2.25a). The beam bends between B and C. The deformed shape is shown in Figure 2.2.25c. The bending moment at C is 10 3 = 30 N.m. Noting that the sense of the bending-moment diagram indcates which side of the beam is in compression, we have Figure 2.2.25b. If we define the positive values of the coordinates of x and M as shown by the arrows in Figure 2.2.25b, the figure satisfies the relation dM/dx = V.

A x

A Straight B

B +10 N C (a) Shear force

30 N m (b) Bending moment

C (c) Deformation

Figure 2.2.25

6861X_C002.indd 102

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

A B 20 N m C 20 N m F C

103

1m 2m

D E

2m 1m

Figure 2.2.26

Example 2.2.4
Calculate the loads and the shear force diagrams corresponding to the bending moment diagram shown in Figure 2.2.26, and sketch the deflected shape of the beam.

Solution
Defining the positive values of x and M as we did in Figure 2.2.25b, we have V = dM/dx = 10 in B- C, V = dM/dx = -20 in C-E, and V = 0 in E-F as shown in Figure 2.2.26a. Recalling that the positive shear force is clockwise, we have the loads as shown in Figure 2.2.26b. Because the right edge of the beam is compressed between B and D and the left edge between D and F, we reason the deflected shape to be as shown in Figure 2.2.26c.

Design your own beam (Part 2) We want to design a beam that can support a mini-elephant. The weight of the mini-elephant is (any one digit you choose) plus 10 lbf. Assume that each leg of the mini-elephant carries the same amount of gravity force. The weight of the beam is negligible. The tensile force on the wall (T in Figure 2.2.7) must be smaller than

A B +10 N C 20 N E F (a) Shear force

10 N

A B

Straight

30 N 20 N

F (b) Loads (c) Deformation

Figure 2.2.27

6861X_C002.indd 103

2/7/08 [Link] PM

104

Understanding Structures: An Introduction to Structural Analysis

Trumpet T h 2 in 2 in 6 in C

Figure 2.2.28

100 lbf because of the strength of the wall. What is the required beam depth, h? (Check your result using GOYA-C.) What is an internal shear force? I could not understand the definition of shear force in todays lecture. Is it an action that shears a beam as shown in Sketch (a)?

Shear

Beam

Shear (a)

No, it is different. A pair of shears acts to cut the beam from outside, whereas the shear force is not an action from outside. It acts inside the beam as shown in Sketch (b). You cannot see the forces unless you cut the beam over a length dx. Similarly, the shear force refers to a force pair acting perpendicularly to the axis of a structural member as shown in Sketch (c). The force pair is internal. It is not applied externally.
N N

dx (b)

V dx (c)

6861X_C002.indd 104

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

105

You mean that shear force is similar to axial force. But it looks different. In the case of tensile axial force, the pair of forces balance each other by pulling the portion dx in opposite directions. In the case of shear force, however, the two forces balance each other in the vertical direction and they create a moment [Link] that tends to rotate the beam portion dx. You have got it. The product of V and dx results in an incremental bending moment, dM = [Link] that is, a shear force always leads to a slope in the bending-moment diagram dM/dx. I know that an axial force elongates or shortens a member and a bending moment bends a member. How does a shear force deform a member? Good question. Assume that you cut the beam and glue the sections using a very soft adhesive. Then, apply shear force. The adhesive will deform as shown in Sketch (e) because the atoms in the adhesive are subjected to the forces shown in Sketch (f). We call the deformation so caused shear deformation. It is also called shear distortion or, simply, distortion.

Adhesive V V

Shear force V

(e) Deformation

(f ) Forces on atoms

Note that this is a special case occurring for very small ratios of moment to shear (M/V). As you can see in Sketch (g), tensile and compressive forces caused by bending moment are usually much larger than shear force. So, shear deformation is usually much smaller than flexural deformation. But shear force is important in determining the bending moment. Also, shear force sometimes induces a brittle failure. You should never ignore it.

C V T dx

C + dC V T + dT V (g) V T

6861X_C002.indd 105

2/7/08 [Link] PM

106

Understanding Structures: An Introduction to Structural Analysis

2.3 What is a Distributed Load?

WINTER SNOW

A cantilever beam may deflect because use of a heavy snow load in winter. In summer, strong winds may push the beam up from below. We call such loads distributed loads. Log into GOYA-C. Type 0.5 in the text field of for Amplification to reduce the deformation scale. Click Add load nine times and move each load to develop the distributed load as shown in Figure 2.3.1a. Note that the width of each element is 10 mm and the magnitude of each load is 10 N. This represents a uniform distributed load of (10 N)/(10 mm) = 1 N/mm. Though Tthe shear-force diagram (Figure 2.3.1c) is shown in steps at intervals of 10 mm. You will have smoother diagrams if you take the time to specify distribute 100 loads of 1 N at intervals of 1 mm. Figure 2.3.2 shows the vertical forces acting on a segment of length of dx of the beam where we assume that the distributed force w acts on every inch (or mm) along

y 10 mm 44.37 100 5000 10 10 10 10 10 10 10 10 10 10 (a) Deection 5000 4500 3600 2800 2100 1500 1000 100 x

0 0

600 100 300

10

20

30

40

50

60

70

80

90

(b) Bending moment diagram

(c) Shear force diagram

Figure 2.3.1 Simulation of distributed load.

6861X_C002.indd 106

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

dx

107

V + dV

Figure 2.3.2 Equilibrium of dx.

the beam. Equilibrium in the vertical direction requires that dV = [Link] or dV/dx = w (2.3.1)

If w is constant (uniformly distributed load), integration of the above equation leads to V = w.x + C, where C is constant. Because the shear force is zero at the free end as shown in Figure 2.3.1, the constant C should be zero and we have V = w.x (2.3.2)

Substituting this into dM/dx = V and noting that M = 0 at the free end (x = 0), we have M = w.x2/2 (2.3.3)

which describes a parabola. Note that the bending moment diagram in Figure 2.3.1 is almost parabolic. If we substitute w = 1 N/mm and x = 100 mm (the beam length), we have V = 100 N and M = 5000 N-mm, which agree with the numbers indicated in Figure 2.3.1 at the fixed end. We may also obtain these equations without integration. Cut the beam as shown in Figures 2.3.3a and b, and represent the distributed load by an equivalent concentrated force as shown in Figure 2.3.3c.; then, equilibrium in the horizontal vertical direction leads to Equation 2.3.2 and equilibrium of moment equilibrium leads to Equation 2.3.3. We can approximate the distributed force as by a few concentrated forces as shown in Figure 2.3.3d. The broken and solid lines in Figure 2.3.2e show the exact and approximated shear force diagrams, respectively. They agree at the free end, mid-span, and at the fixed end. The bending moment diagrams also agree at the same points as shown in Figure 2.3.3f. Simulate the concentrated forces shown in Figure 2.3.3d on GOYA-C. You will find that the deflections is are also similar to those shown in Figure 2.3.1.

In Figure 2.3.2, the symbol

is meant to indicate that the forces are distributed continuously.

6861X_C002.indd 107

2/7/08 [Link] PM

108

Understanding Structures: An Introduction to Structural Analysis

L l/2 l/2

w (a) Cantilever beam x V w (b) Equilibrium x/2 V wx (c) Equivalent concentrated load M M

wl/2

wl/2

(d) Concentrated forces

wl/2 (e) Shear force

wl

wl 2/8

wl 2/2

(f ) Bending moment

Figure 2.3.3 Beam with uniform ly distributed load.

Example 2.3.1
Assume that you have a beam with the same magnitude of distributed load as the beam in Figure 2.3.3a but double the lengthspan L. Determine the shear force and the bending moment at the fixed end.

Solution
According to Equations 2.3.2 and 2.3.3, the shear force will double and the bending moment will be four timesquadruple that in the previous example.

Example 2.3.2
Draw shear-force and bending-moment diagrams caused by the load shown in Figure 2.3.4.

50

50

x 0.2 N/mm

Figure 2.3.4 Load on left half.

6861X_C002.indd 108

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

Linear Constant 10 N 50 (a) Shear force 50 10 N 10 N (c) Concentrated forces 75

109

Linear Parabola 250 N mm 50 50 (b) Bending moment 750 N mm

10 N (d) Approximated shear force 250 N mm

750 N mm

(e) Approximated bending moment

Figure 2.3.5 Solution.

Solution
We have dV/dx = 0.2 in the left half (0 < x < 50) and V = 0 at x = 0. So, V = 0.2 x in 0 < x < 50 (2.3.4) as shown in Figure 2.3.5a. For 50 < x < 100, dV/dx = 0 yields V = constant. Substituting Equation 2.3.4 into dM/dx = V and noting M = 0 at the free end (x = 0) M = 0.1 x2 in 0 < x < 50. as shown in Figure 2.3.5b. Because V = 10 in 50 < x < 100, the slope of the bendingmoment diagram should be 10. Figure 2.3.5c shows the concentrated force equivalent to of the distributed load. Figures 2.3.5d and e show the approximated shear force and bending moment diagrams. Again, Note again that they agree with the exact values at the free end, mid-span, and at the fixed end.

Example 2.3.3
Draw shear-force and bending-moment diagrams caused by the distributed force in Figure 2.3.6. Note that w is negative because the force is downward.

0.3 N/mm x 100

Figure 2.3.6 Linearly varying distributed load.

6861X_C002.indd 109

2/7/08 [Link] PM

110

Understanding Structures: An Introduction to Structural Analysis

67 Parabola 15 N (a) Shear force 15 N (d) Approximated shear force (c) Concentrated forces

15 N

1,000 N mm (b) Bending moment

1,000 N mm (e) Appr. bending moment

Figure 2.3.7 Solution.

Answer
Because the load is w = - (0.3/100)x, we integrate dV/dx = - (0.3/100)x noting V = 0 at the free end (x = 100) and obtain

V = -15 x 2 10 -4 + 15 Substituting this into dM/dx = V and noting M = 0 at x = 100 yield

M = -5 x 3 10 -4 + 15 x - 1000

These results are shown in Figures 2.3.7a and b. Figure 2.3.7c shows the concentrated force equivalent to the distributed load. The force has the magnitude equal to the triangle in Figure 2.3.6, F= 1 ( 0.3 N/mm) (100 mm) = 15 N 2

and is located at the centroid of the triangle. Figures 2.3.7d and e show the approximated shear-force and bending-moment diagrams. They agree with the exact values at the free end and at the fixed end.

Example 2.3.4
Draw the shear-force and bending-moment diagrams caused by the loads shown in Figure 2.3.8. (Hint: you may draw the diagrams for the distributed and concentrated loads separately and add them.)

6861X_C002.indd 110

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

40 N

111

1 N/mm 100

Figure 2.3.8 Distributed load plus concentrated load.

Solution
As we noted in Section 2.2, the shear- force and bending- moment diagrams caused by the concentrated load (- 40 N) are given plotted as shown in Figure 2.3.9. Those related to the distributed load are shown in Figure 2.3.10. The areas in the shearforce diagrams equal the bending moments ( V dx = M ). Adding (superposing) these diagrams, we obtain Figures 2.3.11a and b. Note that the shear force and the slope of the moment diagram are zero at x = 40 mm. Figure 2.3.5c 11c shows the concentrated forces equivalent to the distributed load. Figures 2.3.5d 11d and e show the approximate d shear- force and bending- moment diagrams. They agree with the exact values at the free end, mid-span, and at the fixed end. Figure 2.3.12 shows the results obtained by GOYA-C. Note that the deflection at the beam end is only 3 mm, which is less than 1/10 of the deflection caused by the distributed load shown in Figure 2.3.1 (44 mm). The clue to explain this result may be found is in the bending-moment diagram, which shows that the beam deflects up near the fixed end and down near the middle. The upward deflection tends to reduce the downward deflection.

40 40 N 40 N (a) Shear force

40

60 4,000 N mm

1,600 N mm

(b) Bending moment

Figure 2.3.9 Contribution of concentrated load.

6861X_C002.indd 111

2/7/08 [Link] PM

112

Understanding Structures: An Introduction to Structural Analysis

40 40 N 60 (a) Shear force

100 N

5,000 N mm 800 N mm 40 60 (b) Bending moment

Figure 2.3.10 Contribution of distributed load.

40 N

40 N

60 N 50 N 40 60 (a) Shear force 10 N 1,000 N mm 800 N mm 40 N (d) Appr. shear force 1,000 N mm 60 (b) Bending moment 1000 N mm 60 N 50 N

(c) Concentrated forces

40

500 N mm

(e) Appr. bending moment

Figure 2.3.11 Solution.

40

0.94 3.04

60 1000 10 10 10 10 10 10 10 10 10 10

Figure 2.3.12 Simulation using GOYA-C.

6861X_C002.indd 112

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

113

40 N

1 N/mm 40 60

Figure 2.3.13

Example 2.3.5
Plot shear-force and bending-moment diagrams caused by the load in Figure 2.3.13. (Do not use GOYA-C before making your own calculations.)

Solution
The shear- force and bending- moment diagrams caused by the concentrated load (- 40 N) are given as shown in Figure 2.3.14. Adding this diagram to the one in Figure 2.3.10, we obtain Figure 2.3.15. Note that the shear force is discontinuous at x = 40 mm. Figure 2.3.16 shows the results simulated by GOYA-C. Note that the deflection at the beam end is 24 mm, which is eight times that obtained for the loading condition in Figure 2.3.12 (3 mm). The reason lies again in the bending moment diagram, which shows that the beam deflects upward throughout its length. Approximate the bending moment diagram using concentrated forces equivalent to the distributed load and compare with the exact diagram. We have distinguished the effects of a concentrated load from those of a distributed load. However, if we look at the experiment shown in Figure 2.3.17 closely, we may notice that the finger applying the load has a finite length in the direction of the span. The applied load should be modeled as a distributed load as shown in Figure 2.3.18a. Such modeling leads to the shear-force diagram shown in Figure 2.3.18b. In other words, the equation dV/dx = w also applies to concentrated loads.

40 N 40 60 (a) Shear force 60

40

2,400 N mm (b) Bending moment

Figure 2.3.14 Contribution of concentrated load.

6861X_C002.indd 113

2/7/08 [Link] PM

114

Understanding Structures: An Introduction to Structural Analysis

40 N 60 N

40

60 (a) Shear force

2,600 N mm 400 N mm 40 60 (b) Bending moment

Figure 2.3.15 Solution.

40

23.89 60 2600 10 10 10 10 10 10 10 10 10 10

Figure 2.3.16 Deformation.

Pencase

Desk

Push

Figure 2.3.17 Push a cantilever beam.

6861X_C002.indd 114

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

dx

115

(a) Distributed load dx V (b) Shear force

Figure 2.3.18 Distributed load.

Example 2.3.6
Draw shear-force and bending-moment diagrams caused by the distributed load shown in Figure 2.3.19.

Solution
Integrating the distributed load leads to the shear force. V = w dx = 20 cos x + C1 2

where C1 is an integral constant. Because V = 0 at the free end (x = 0), we have

C1 = -

20

Figure 2.3.20a shows the shear force diagram. Integrating the shear force leads to the bending moment. M = V dx =

40 sin 2 2

20 x x + C2

where C2 is an integral constant. Because M = 0 at the free end (x = 0), we have C2 = 0. Figure 2.3.20b shows the bending-moment diagram. Because the diagram
[N/m] 2

w = 10sin

x 2m 2m

Figure 2.3.19

6861X_C002.indd 115

2/7/08 [Link] PM

116

Understanding Structures: An Introduction to Structural Analysis

40/ [N] (a) Shear force

40/ [[N m]

80/ [N m]

(b) Bending moment

(c) Deformation

Figure 2.3.20 indicates that the bottom of the beam is compressed, the beam deflects downward as shown in Figure 2.3.20c. Let us represent the distributed load by a pair of concentrated loads as shown in Figure 2.3.21a, noting that

40 sin x dx = 2

The shear-force and bending-moment diagrams for these loads are shown in Figure 2.3.21b and c. They agree with the exact values at the free end, at mid-span and at the fixed end.
40/ [N]

2m

40/ [N]

(a) Equivalent concentrated load

40/ [N] (b) Shear force

40/ [N]

80/ [N m]

(c) Bending moment

Figure 2.3.21

6861X_C002.indd 116

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

117

x 2m 2m

20 [N m]

Figure 2.3.22

Example 2.3.7
Construct the shear-force diagrams and evaluate the loads corresponding to the bending moment in Figure 2.3.22, where M = 20 cos 2 x [N.m]

Solution

Differentiating the bending moment leads to the shear force. V= dM = -10 sin x 2 dx

This is shown in Figure 2.3.23a. Differentiating the shear force leads to the distributed load. w= dV = -5 2 cos x 2 dx

Because the bending moment at the free end is 20 N-m, there should be a couple at the end, and the load should be as shown in Figure 2.3.23b. The bending- moment diagram indicates that the beam deflects as shown in Figure 2.3.23c.

Design your own beam (Part3): We want to design a beam that can carry a mini elephant, whose weight is (the last two figures of your ID#)/any twoone digits you select)/100 plus 10 kglbf. Assume that the gravity acceleration is 10 m/s2 and each leg carries the same amount of gravity force. The density of the beam is 0.5 10-5 kglbf/mm3in3. The tensile force on the wall (T in Figure 2.3.22a the figure) must be smaller than 100 90 Nlbf. What is the required beam depth if the beam is prismatic (its section remains the same along its span) has constant depth (Figure 2.3.24a) or if it has varying depth of as specified in Figure 2.3.24b? Check your results using GOYA-C. Hint: If the depth is constant, the distributed load caused by the self-weight is w = b h,

6861X_C002.indd 117

2/7/08 [Link] PM

118

Understanding Structures: An Introduction to Structural Analysis

x 10 [N] 2m 2m

(a) Shear force

20 [N m] (b) Load Top compression

Bottom compression (c) Deformation

Figure 2.3.23

w 2 in 2 in 6 in

T C

h/3 2 in 2 in 6 in

T C h

(a) Constant depth Unit length w=bh h b = 1 in (c) Uniform load 3 4 10 4 5 10 6 6 h/3 b = 1 in 7

(b) Varying depth Unit length w h

(d) Varying load 8 9

80 33.35
3860

(e) Simulation by GOYA-C

Figure 2.3.24 Design your beam.

6861X_C002.indd 118

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

119

where r: density (0.5 10-5 kglbf/mm3in.3), b: beam width (10 mmin.), and h: beam depth (unknown) , and g: acceleration of gravity (10 m/s2) as shown in Figure 2.3.24c. The load w will vary if the beam depth varies. In both cases, you will derive equations in terms of h and solve them. The depth required for case b (varying) is smaller than that required for case a (uniform). You can find beams with varying depths in actual structures such as those supporting balconies or elevated highways. GOYA-C cannot simulate a beam with varying depth. You can check your result by using a varying load as shown in Figure 2.3.24e with the depth at the fixed end, h.

2.4 What is a Couple?

Apply a pair of equal and opposite forces of 100 N close to one another as shown in Figure 2.4.1. Note that a shear force exists only between the points of application of the forces. The bending moment increases at a high rate in that region. We call the pair a couple. If the distance (a in Figure 2.4.2a) is small enough, we represent the couple using a curved arrow as shown in Figure 2.4.2b. A couple may be applied through a bolt embedded in the side of the beam and twisted by a wrench. The bending moment has a sudden change or a jump at the point of application of the couple as shown in Figure 2.4.2c. We ignore the shear force associated with a couple though in fact a very large shear force might exist over a very small region (See the shearforce diagram inFigure 2.4.1). Figure 2.4.3 shows a beam subjected to a couple and a concentrated load. The shear-force diagram is constructed ignoring the couple because it is assumed that the couple is applied at a point. The bending moment diagram has a jump at the point of application of the couple. The bending-moment diagram in Figure 2.4.3c is obtained

100

14.16 0 1000
100

100

1000 0 0

1000 0 0

Figure 2.4.1 A pair of close loads.

6861X_C002.indd 119

2/7/08 [Link] PM

120

Understanding Structures: An Introduction to Structural Analysis

a F

(a) Pair of forces

C = Fa

(b) External moment

M = Fa (c) Bending moment

Figure 2.4.2 Beam with a couple.

combining the moment diagram in Figure 2.4.2c with that for load F (the broken line in Figure 2.4.3c). Figure 2.4.4 shows the simulation by GOYA-C for M = 500 N-mm and F = 10 N. If you use forces of +1000 N with a distance of 1 mm apart, you will have a moment diagram very similar to Figure 2.4.3c.
F L

C (a) External forces

V = F (b) Shear force

C (c) Bending moment

FL

Figure 2.4.3 Beam with a couple and a load.

6861X_C002.indd 120

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

20
100

121

6
100

20
800

80 0 0 200 600 800 0 20 20

Figure 2.4.4 Simulation by GOYA-C.

Example 2.4.1
Construct the shear-force diagrams and evaluate the loads corresponding to the bending moment in Figure 2.4.5, where M = 20 cos x [N.m] 2

Solution

Differentiate the expression for bending moment to obtain the shear force at any point x. V= dM = -10 sin 2 dx x

The resulting expression is plotted in Figure 2.4.6a. Differentiate the expression for the shear force to obtain the distributed load. w= dV = -5 2 cos 2 dx x

Because the bending moment at the free end is 20 N-m, there should be a couple at the end, and the load should be as shown in Figure 2.4.6b. The bending-moment diagram indicates that the beam deflects as shown in Figure 2.4.6c.

x 2m 2m

20 [N m]

Figure 2.4.5

6861X_C002.indd 121

2/7/08 [Link] PM

122

Understanding Structures: An Introduction to Structural Analysis

x 10 [N] 2m 2m (a) Shear force

20 [N m] (b) Load Top compression

Bottom compression (c) Deformation

Figure 2.4.6

The most important concepts in the sections 2.1 through 2.4 are: 1. Shear force at a section is determined by cutting the beam at that section and considering force equilibrium in the direction perpendicular to the beam axis. 2. Bending moment at a section is determined by cutting the beam at that section and considering the moment equilibrium. It is useful to remember that: 1. The slope of the bending-moment diagram along a member is equal to the shear force (dM/dx = V ). A couple applied on the beam at a point makes the bending moment diagram discontinuous (its magnitude changes abruptly) at that point. 2. The slope of the shear-force diagram along a member is equal to the distributed load (dV/dx = w). A concentrated force applied on the beam at a point makes the shear force diagram discontinuous (a jump occurs) at that point.

Review: A couple is to bending moment as a force is to shear force. In Section 1.2, we concluded that external force, a vector, is different from internal axial force, a scalar. If you apply forces parallel to the member axis (Figure 2.4.7a),

6861X_C002.indd 122

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

3F 2F

123

(a) External force 3F F

(b) Axial-force diagram

Figure 2.4.7 Axial force.

the axial force changes where the forces are applied (Figure 2.4.7b). In the preceding section, we concluded that external force is different from shear force. If you impose forces perpendicular to the member axis as shown in Figure 2.4.8a, the shear force changes where the external forces are applied (Figure 2.4.8b). Similarly, a couple is a vector having a sense or direction with respect to the axis of rotation, while bending moment is a scalar. Bending moment changes abruptly where a couple is applied as shown in Figure 2.4.9.
C

aF

2F

3F (a) External force 3F F

(b) Shear-force diagram

Figure 2.4.8 Shear force

6861X_C002.indd 123

2/7/08 [Link] PM

124

Understanding Structures: An Introduction to Structural Analysis

3C 2C

(a) External force 3C C

(b) Bending-moment diagram

Figure 2.4.9 Bending moment.

2.5 The Effects of Moment on Stresses in a Beam

In this section, we shall focus on bending moment based on the following definitions. (Force) = (Stress) (Area) (Moment) = (Force) (Distance) (2.5.1) (2.5.2)

Using GOYA-C, apply forces as shown in Figure 2.5.1 to have a bending moment of 600 N-mm at the fixed end. Click Stress in the middle right to obtain Figure 2.5.2a. In Figure 2.5.2a we see that stress is distributed linearly over the depth of the section. The width, b, and the depth, h, of the beam section are 10 mm and 15 mm, respectively. The stress at the extreme fiber in compression (top fiber) is -1.6 N/mm2 and the stress at the extreme fiber in tension (bottom fiber) is +1.6 N/mm2.. The objective of this section is to study how the assumed stress distribution on the section develops a bending moment of 600 N-mm. The bending moment generated by the stress distribution can be determined if we partition the section into many layers of thin horizontal slices (Figure 2.5.3a). Because the thickness of each slice (dy) is small enough, we may assume that the stress in each slice is uniform. Consider the shaded slice in Figure 2.5.3a. It has an area, [Link]. Assuming a constant stress acting on the slice, we determine the total force on the slice as the product of the stress, s, and the area of the slice [Link]: dF = [Link]. The contribution of the force dF = [Link] to the bending moment is dM = - [Link] = -[Link] with a negative sign because the stress is compressive (a < 0) in the upper half of the section ( y > 0). The bending moment of the section is the sum of dM over the section or M = - y dF = -

h/2

-h / 2

y b dy

(2.5.3)

Strictly, the stress distribution is linear if the beam material remains in its range of linear response where unit stress is proportional to unit strain. In this text we shall consider only response in the linear range.

6861X_C002.indd 124

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

10

125

9.39

0 600 10

Figure 2.5.1 Forces making 600 N-mm.

Calling the stress at the bottom of the section s f and s f at the top (Figure 2.5.3b), we describe the stress distribution as follows.

=-

y f h/2

(2.5.4)

1.6

1.6

(a) Window b

h/2

h/2 f = 1.6 N/mm2 (b) Isometric view

Figure 2.5.2 Stress distribution.

6861X_C002.indd 125

2/7/08 [Link] PM

126

Understanding Structures: An Introduction to Structural Analysis

Area dA=b dy h/2 dy y

b Force dA Distance y h/2

f dy y

h/2 (a) Partitioned section

h/2

(b) Stress distribution

Figure 2.5.3 Partitioned section and stress distribution.

Substituting this into Equation 2.5.3 leads to the resisting moment in terms of the stress in the extreme fiber, s f. M=

2 b f y f y b dy = h - h / 2 h/2
h/2

h/2

- h/2

y 2 dy =

bh 2 f 6

(2.5.5)

The term bh2/6 modifying s f is called the section modulus of a rectangular section. We shall denote it by the symbol Z. Z= bh 2 6 (2.5.6)

For a given moment and section, we express the maximum flexural stress in a rectangular section simply as:

f =

M Z

(2.5.7)

The initial setting of GOYA-C is b = 10 mm and h = 15 mm, which leads to the following maximum bending stress for Figure 2.5.2.

f =

M 600 = = 1.6 N/mm 2 Z 10 152 /6

Click Force to get Figure 2.5.4a. The compressive force shown here represents the total stress in the upper half of the section (Figure 2.5.4b), C=

h/2

dA =

h/2

b dy =

h/2

y bh b dy = f h/2 4

The force is located at the centroid of the compressive stress block that is h/3 from the mid-height of the section. The tensile force is also located h/3 from the

6861X_C002.indd 126

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

127

b f 60 0 60 60 (b) Isometric view 60 0 h/2 C= bh 4 f

h/2

T=C

3h/2

(a) Wi ndow

Figure 2.5.4 Concentrated force.

mid-height. The distance between the compressive and tensile forces is 2h/3, which leads to M =C 2 h bh 2 h bh 2 = f = = Z f 3 4 3 6 f

Next, click on the Beam Detail button and change the beam depth to 7.5 mm to get Figures 2.5.5 and 2.5.6. Figure 2.5.6 shows that the stress is 6.4 N/mm2, which is 4 times that for h = 15 mm. Note that the stress is inversely proportional to the second power of the beam depth as indicated in Equation 2.5.6 and 2.5.7. Generally, the depth of a beam is more than its width because an increase in depth reduces the stress, for a given moment, more effectively than an increase in width. Example 2.5.1
A cantilever beam of width b = 1 in, depth h = 1.5 in, and length l = 15 in. is subjected to a force as shown in Figure 2.5.7a. Note that the selected dimensions correspond roughly to those of the bone in your forearm. Assume the material of the beam to be

75.09

10

0 600

10

Figure 2.5.5 Beam depth is halved.

6861X_C002.indd 127

2/7/08 [Link] PM

128

Understanding Structures: An Introduction to Structural Analysis

6.4 6.4

Figure 2.5.6 Stress distribution.

linearly elastic. Calculate the maximum load that can be resisted for the four cases of tensile strength listed below. Compare it in each case with the axial force that the beam can carry if it is loaded in axial tension as shown in Figure 2.5.7b. (a)s = 15,000 psi (human bone) (c)s = 60,000 psi (steel) (b)s = 10,000 psi (pine) (d)s = 400 psi (concrete)

Solution
The section modulus is Z = bh2/6 = 0.45 in2. Substituting the given material strengths into the equation M = Zs, we obtain the allowable bending moments listed below for cases (a) through (d). (a)M = 6,750 lbf-in (human bone) (c)M = 27,000 lbf-in (steel) (b)M = 4,500 lbf-in (pine) (d)M = 180 lbf-in (concrete)

The allowable forces are obtained by dividing the moments by the length of the beam, L = 15 in. (a)F = 450 lbf (human bone) (c)F = 1,800 lbf (steel) (b)F = 300 lbf (pine) (d)F = 12 lbf (concrete)

Dumbbell (450 lbf )

1.5 in 15 in 1.2 in (a) Cantilever beam

1.2 in

1.5 in

27,000 lbf

(b) Pure tension

Figure 2.5.7 Cantilever beam and pure tension.

6861X_C002.indd 128

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

1 h 1 in b Distributed load w=.b.h h = 1.5 in b = 1.2 in

129

L?

Figure 2.5.8 Limiting length of span under self-weight.

The result indicates that the bone in your arm can resist a weight comparable to your own weight as described in Figure 2.5.7a. Using the approach described above, you can also calculate the maximum wind force that a pine tree can resist. The force that the beam can resist in axial tension can be obtained from T = As, where A = bh = 1.8 in.2 (a)T = 27,000 lbf (human bone) (c)T = 108,000 lbf (steel) (b)T = 18,000 lbf (pine) (d)T = 720 lbf (concrete)

Note that the ratio of the force that can be re sisted in direct tension to that in flexure is AL/Z = 60 for the beam considered.

Example 2.5.2
Calculate the limiting length of a beam having the same section and strength as in the previous example. In this case assume that the only load is the self-weight of the beam. Assume the specific weight to be: (a)0.04 lbf/in3 (human bone) (c)0.3 lbf/in3 (steel) (b)0.015lbf/in3 (pine) (d)0.08 lbf/in3 (concrete)

Solution
The distributed load for each material is: (a)w = 0.072 lbf/in (human bone) (c)w = 0.54 lbf/in (steel) (b)w = 0.027 lbf/in (pine) (d)w = 0.144 lbf/in (concrete)

On the other hand, the bending moment caused by the distributed load, w, is M = wL2/2 at the fixed end. Using the allowable bending moments obtained in the previous example, we have the following results.


The actual bone is not as simple as assumed: it is a pipe structure that is soft inside. As we have seen in this section, the center of the beam carries smaller stress and therefore need not be strong.

6861X_C002.indd 129

2/7/08 [Link] PM

130

Understanding Structures: An Introduction to Structural Analysis

2M w 2M w 2M w 2M w 2 6,750 0.072

(a) L = (b) L = (c) L = (d) L =

= 433 in (human bone) = 577 in (pine) = 316 in (steel)

= = =

2 4 ,500 0.027 2 27,000 0.54 2 180 0.144

= 50 in (concrete)

You may note the impressive strength of bio-materials. Reviewing the list above, you may consider concrete to be inferior to steel or timber, but concrete is in fact a good structural material if reinforced with steel. Example 2.5.3
Repeat Example 2.5.1 assuming that both the width and the depth of the section are doubled. Will the limiting length of span also be doubled?

Solution
The section modulus Z = bh2/6 will be 23 = 8 times of that for the section shown in Figure 2.5.9a. Therefore, the allowable moment also will be 8 times as much. On the other hand, the distributed load will be 22 = 4 times as much. Because L is proportional to M /w , L shall be 8/4 = 2 times the length of a beam with half the width and depth. Figure 2.5.9 shows the results for concrete. Note that the (length)/(depth) ratio decreases as the beam depth increases. This is the reason why the legs of elephants are so stocky compared with those of mice.

Design your own beam (Part 4) We want to design a beam to carry a mini elephant, with weight equal to (any one digit number you choose) plus 10 lbf. Assume that each of the four legs carries the same amount of gravity force. The density of the beam is 0.5 lbf/in3. The beam width

L = 50 in (a) Small section

h = 1.5 in b = 1.2 in L = 70 in (b) Large section

h = 3 in

b = 2.4 in

Figure 2.5.9 Beams with allowable length.

6861X_C002.indd 130

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

131

w 2 in 2 in 6 in

h/3 2 in 2 in 6 in

(a) Uniform height

(b) Varying height

Figure 2.5.10 Beams to carry elephant.

is 1 in. The tensile strength of the material is 500 psi. What is the required beam depth (a) if the beam is prismatic (its section remains the same along its span) (Figure 2.5.10a), and (b) if it has varying depth as specified in Figure 2.5.10b? Check your results using GOYA-C. Hint: The stress s = M/Z , will be a maximum at the fixed end. The condition M/Z = 500 psi
will lead to quadratic equations for the depth h.

Coffee break In this section, you have learned that bendig stress varies linearly over the depth of a beam in the linear range of response. You also know that the section subjected to bending only has an axis (neutral axis) where the stress is zero and that this axis is at mid-depth of a rectangular section. Recognizing this simple fact, however, has required many centuries of thinking. The first scientist who studied the strength of a cantilever beam was Galileo Galilei (15641642) (Figure 2.5.11a). He assumed that the bending stress was constant over the depth of the section as shown in Figure2.5.11b. After him, the stress distribution in a bending beam was
L h A B D
A B

F=

M bh2 = L 2L

(b) Galileos assumption h M bh2 = L 2L (c) Naviers assumption F= B A F

C E

(a) Original gure

Figure 2.5.11 Galileos cantilever beam.

6861X_C002.indd 131

2/7/08 [Link] PM

132

Understanding Structures: An Introduction to Structural Analysis

investigated by a lot of prominent scientists and mathematicians such as Bernoulli, Euler and Coulomb in the 17th and 18th centuries. Finally, a book published in 1826 by Navier (17851836) put an end to the controversy about the stress distribution (Figure2.5.11c). [See History of Strength of Materials by S.P. Timoshenko.] Galileo Galilei, 1638. Dialogues Concerning Two New Sciences, Translated by H. Crew and A. de Salvio, 1954. Dover Publications, New York.

2.6 The Effects of Moment on Strains in a Beam

Consider aA cantilever beam is subjected to a couple M at its free endas shown in (Figure 2.6.1a). We shall investigate the portion of the beam between consider two planes (P1 and P2 in Figure 2.6.1a). Before the application of the moment, the two planes are perpendicular to the beam axis, which is straight. We may assume the two planes to remain plane and perpendicular to the deflected beam axis after the beam is deformed in bending (Figure 2.6.1b). The portion between the two planes is shown in detail in Figure 2.6.2. To highlight the deformation, we shall redraw Figure 2.6.2 to make the y-axis vertical as shown in Figure 2.6.3a. Given the angle q between P1 and P2, we infer that the top of the beam shortens by q. h/2 and the bottom lengthens by the same amount as shown in Figure 2.6.3b. As defined in Chapter 1, the lengthening and shortening changes in length shown in Figure 2.6.3b divided by the original length a lead togives the strain distribution shown in Figure 2.6.3c, where the term

a
P2 h a

(2.6.1)

P1

(a) Before deformation Beam axis M P1 P2 h a (b) After deformation

Figure 2.6.1 Planes P1 and P2.



Rigorously, this assumption is valid only for pure bending but can be used to arrive at plausible results except for beams with span to depth ratios larger than two.

6861X_C002.indd 132

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

y

133

P1 P2 M

Beam

axis

Figure 2.6.2 Detail of Region between Planes P1 and P2.

is introduced. The term f is called the unit curvature because it represents the amount that the beam axis curve bends is curved in over a unit length. Noting that the strain is negative (compression) at the top and positive (tension) at the bottom, the strain distribution can be illustrated as shown in Figure 2.6.3d as a linear function of y: e = - f y (2.6.2)

For a material that responds linearly, the relationship between stress and strain is written as s = E e (1.2.3)

where E is Youngs modulus. Substituting e = - f y into Equation 1.2.3, we have s = -E f y (2.6.3)

As discussed in section 2.5, bending moment is M=-

h/2

- h/2

y b dy

(2.5.3)

Substituting Equation 2.6.3 into Equation 2.5.3, we can develop a relationship between bending moment and unit curvature for a rectangular section: M = E b

or


h/2

- h/2

y 2 dy = E

bh 3 12

M = EI

(2.6.4)

In Figure 2.6.3a, r denotes the radius of the beam axis in its bent form. Because r.q = a and f = q/a, r is a reciprocal of the unit curvature (r = 1/f), and is called the radius of curvature.

6861X_C002.indd 133

2/7/08 [Link] PM

134

Understanding Structures: An Introduction to Structural Analysis

Shortening

Shortening h/2
M h/2 h/2 h/2 Lengthening (b) Deformation

Compression h/2

y E h/2 h/2 = E y E h/2

h/2

/2 Beam axis

h/2

0 h/2

0 h/2

Original a Lengthening (a) Upright position

= y h/2

h/2 Tension (c) Strain

(d) -y relation

(e) -y relation

Figure 2.6.3 Deformation and strain between P1 and P2.

where I= bh 3 12 (2.6.5)

The term I is called the moment of inertia. Although Equation 2.6.4 was derived for the special case of uniform bending moment, it is also applicable if the bending moment varies with x. In such cases, we obtain the same equation assuming the bending moment to be constant within an infinitesimal length dx. The initial setting of GOYA-C is b = 10 mm and h = 15 mm, which leads to the following moment of inertia: I= bh 3 10 153 = = 2810 mm 4 12 12

Apply forces as shown in Figure 2.6.6a to have a bending moment of 600 N-mm and click Strain to obtain the strain distribution. Youngs modulus is assumed to be 100 N/mm2 in GOYA-C. The unit curvature caused by the bending moment of 600 N-mm is

M 600 = = 2.13 10 -3 /mm EI 100 2810

Note that the unit curvature has the unit of (1/length), because it is the inverse of the approximated radius. Substituting the unit curvature determined above into


The reason for the designation moment of inertia will be discussed in Chapter 4.

6861X_C002.indd 134

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

16

135
16

10 9.39 (a) 0 10 600 (b) 5.33 8

(c)

Figure 2.6.4 Strain distribution ( 10-3).

e = - f y, we obtain the strain at the top of the beam ( y = 7.5 mm) as

= -( 2.13 10 -3 /mm) ( 7.5mm) = -16 10 -3

(Compression)

and that at the bottom (y = -7.5 mm) as

= - ( 2.13 10 -3 /mm) ( -7.5mm) = + 16 10 -3

(tension)

as shown in Figure 2.6.6b. These strains imply that each rectangle in the grid having a dimension of 10 mm shortens e 10 mm = 0.16 mm at the top and lengthens 0.16 mm at the bottom. Next, click the grids between the two forces, where the bending moment is not constant (Figure 2.6.6c). You will find that the strain gets smaller as the bending moment gets smaller. Thus, the strain varies not only in the y-direction y but also in the x-direction x if the moment varies along the span of the beam. Example 2.6.1
Assume that L = 100 mm, b = 10 mm, h = 15 mm, and E =100 N/mm2, as in GOYA-C. Apply a point load of 10 N at the free end (Figure 2.6.7). Evaluate the strain distribution at three sections: (1) the fixed end, (2) the mid-span and (3) the free end.
100 mm 15 mm 10 N 10 mm

E = 100 N/mm2

Strain distribution

Figure 2.6.5 A force at the free end.

6861X_C002.indd 135

2/7/08 [Link] PM

136

Understanding Structures: An Introduction to Structural Analysis 100 mm

1000 N mm 500 N mm

(a) Bending moment 1.78 103/mm 3.55 103/mm (b) Curvature 26.6 103 Compression Tension 26.6 103 (c) Fixed end 13.3 103 (d) Mid-span 0 (e) Free end 13.3 103 0

Figure 2.6.6 Strain distribution.

Shortening

y M a/2 Beam axis Original a Lengthening (a) Continuously deformable model M M

Shortening h/2
M

h/2

h/2 h/2

Lengthening

(b) Rigid body & spring model M (c) Bar & spring model M

Figure 2.6.7 Spring model.

6861X_C002.indd 136

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

L L

137

L/10

L/2

F v F FL (a) Ten-spring model

L/4

L/2 1

L/4 2

F L 4 3L 2 4 2

M1

F M2

(b) Two-spring model L

L/2 1

L/2

F 1 F L 2

FL

M1 (c) One-spring model

Figure 2.6.8 Spring model for a cantilever beam.

Solution
The moment of inertia is I = 2810 mm4 as determined above. The bending moment is distributed as shown in Figure 2.6.8a. Dividing each ordinate of the moment distribution by EI = 281 103 N/mm2, we have the unit curvature distribution shown in Figure2.6.8b. Using the expression e = -f . y, we get the strain distributions of shown in Figures 2.6.8c through e.

Exercise
Use the last two digits of your ID ij any two digits i and j to change the load to F = (i + 2) N and the depth to h = ( j + 5) mm, and solve the example above. Check your answer using GOYA-C.

2.7 Deformation of a BeamSpring Model

In Section 2.6, we studied the strains in a beam using the continuously deformable model shown in Figure 2.7.1a. In this section, we shall simplify the model as shown in Figure 2.7.1b, where all the deformation is assumed to be concentrated

6861X_C002.indd 137

2/7/08 [Link] PM

138

Understanding Structures: An Introduction to Structural Analysis

L/2 L/2

Figure 2.7.1 A load at the mid-span.

(or lumped) at the middle of the segment considered. In Figure 2.7.1b, a rotation occurs at the middle of the segment considered but the two half segments remain rigid (un-deformed). Note that the total shortening and the lengthening are the same as those in the continuously deformable model. Using f = a/a and M = ELf, we obtain

M a EI

(2.7.1)

We do this to develop a simple method to determine deflection as well as to understand the relationship between moment and deflected shape of a structural member. The representation in Figure 2.7.1b is essentially identical to the one shown in Figure 2.7.1c. We shall use this approach to model a cantilever beam subjected to a concentrated force. Intuitively, we understand that the more springs we use, the better will be the result. If we assume ten springs as shown in Figure 2.7.2a, we may obtain a satisfactory result. However, try the two-spring model shown in Figure 2.7.2b to investigate if that will also give an acceptable result. The bending moments and the rotations at the springs are: M1 = F M2 = F 3 L 4 1 L 4

1 = 2 =

M1 L 3FL2 = EI 2 8EI M 2 L FL2 = EI 2 8EI

(2.7.2) (2.7.3)

The deflection of the free end is: v = 1 3L L 9 FL3 FL3 5FL3 + 2 = + = 4 4 32 EI 32 EI 16 EI (2.7.4)

This result is only 6% smaller than that given by the ten-spring model or by GOYA-C. The exact result is FL3/3EI.


GOYA-C is based on the continuously deformable model which assumes infinite number of springs. This will be discussed in Section 2.8.

6861X_C002.indd 138

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

L/2 FL/2 FL/4 (a) Bending moment 1 M1 F 2 M2 v L/2

139

(b) Two-spring model

Figure 2.7.2 Spring model.

What if we assume only one spring as shown in Figure 2.7.2c? The results are: M1 = FL 2 v = 1

1 =

M1 FL2 L= EI 2 EI

(2.7.5) (2.7.6)

L FL3 = 2 4 EI

The computed deflection is 30% smaller than that given by the continuous model. This is not too bad. Exercise
Use any two digits i and j to change the force to F = (i + 2) N and the depth to h = ( j + 5) mm. Calculate the deflection at the free end using the two-spring model. Compare the result with that obtained by GOYA-C. It is important to note that the deflection is inversely proportional to the moment of inertia I. If you change b = 10 mm and h = 15 mm to b = 15 mm and h = 10 mm (i.e. rotate the section by 90 degrees), I = bh3 /12 will be reduced by 1/1.52 (approximately 1/2) and the deflection will be approximately twice what it was. A shallow beam deflects more than a deep beam of comparable width supporting the same load.

Example 2.7.1
A force F is applied at mid-span of a cantilever beam. Calculate the deflection at the free end using the two-spring model.

Solution
According to Figure 2.7.4a, the bending moments and the rotacions of the springs are: M1 = FL 4 M2 = 0

1 =

M1 L FL2 = EI 2 8EI

(2.7.7) (2.7.8)

2 = 0

6861X_C002.indd 139

2/7/08 [Link] PM

140

Understanding Structures: An Introduction to Structural Analysis

L/4 FL/4 FL/8 F 3L/4

Figure 2.7.3 Move the force to the left.

The deflection of the free end is: v = 1 3L L 3FL3 3FL3 + 2 = +0= 4 4 32 EI 32 EI (2.7.9)

This result is 10% smaller than that given by the continuous model. Note that the deflection for this case is only 30% of that for the case of a load at the free end (Equation 2.7.4). Check this result using GOYA-C. Exercise
What would the deflection be if we move the force to the left as shown in Figure 2.7.5?

Hint: Use of the two-spring model in Figure 2.7.4b results in zero deflection. You
need to have the springs at different locations.

Example 2.7.2
Two loads, F1 and F2, are to be applied at the free end and at mid-span of a cantilever beam as shown in Figure 2.7.6. Calculate the deflection at the free end using the twospring model for F1 = 20 N and F2 = 10 N. (Use the default values in GOYA-C for Youngs modulus and the moment of inertia.)

Solution
According to Figure 2.7.7a, the bending moments and the rotations of the springs are:

1 = 2 =

M1 = M 2 = 250 N mm 250 50 = 4.45 10 -2 rad 100 2810

50 F1 = 20 N E = 100 N/mm2, I = 2810 mm4 F2 = 10 N

100

Figure 2.7.4 Loads at mid-span and at the free end.

6861X_C002.indd 140

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

25 250 25 500 25 250 25

141

(a) Bending moment 2 1 M1 M2 v

(b) Two-spring Model

Figure 2.7.5 Solution.

The deflection of the free end is:

v = 1 75 + 2 25 = 4.45 mm Simulate this using GOYA-C and compare the results.

Exercise
Take any two digits i and j. Assume F1 = (i + 2) N and F2 = ( j + 5) N in Figure 2.7.6. Calculate the deflection at the free end. (Use the default values in GOYA-C for the size of the section and Youngs modulus.)

Example 2.7.3
Two loads, F1 and F2, are to be applied at the free end and at mid-span of a cantilever beam (Figure 2.7.8) so that the deflection at the free end is zero. Determine the ratio F1/F2 using the two-spring model.

Solution
Because the deflection at the free end must be zero, the beam shall deflect as shown in Figure 2.7.9b. This implies that the bending moment M1 should be negative (concave downward) and M2 should be positive (concave upward) as shown in Figure 2.7.9a.

L/2 F1 EI

L/2

F2

Figure 2.7.6 Loads at mid-span and at the free end.

6861X_C002.indd 141

2/7/08 [Link] PM

142

Understanding Structures: An Introduction to Structural Analysis

L/4 L/4 L/4 L/4

M1

M2

(a) Bending moment L/2 L/4 3 v

(b) Two-spring model

Figure 2.7.7 Solution.

The moment equilibrium requires: M1 = F1 L 3L L - F2 = ( F1 - 3F2 ) 4 4 4 M 2 = F2 The rotations of the springs are: L 4 (2.7.10) (2.7.11)

1 =

M1 L L2 = ( F - 3F2 ) EI 2 8EI 1 M2 L L2 = F EI 2 8EI 2

(2.7.12) (2.7.13)

2 =

Click

100

3.9

70 2000

30

Deection

1.48

Figure 2.7.8 Simulation.

6861X_C002.indd 142

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

L wL2/2 M = wx2/2

143

(a) Bending moment 1 M1 M2

(b) Two-spring model

Figure 2.7.9 Distributed load.

On the other hand, Figure 2.7.9b requires:

and

1 = 3 =

v 2v = , L/2 L v 4v = , L/4 L

2 = 1 + 3 =

6v = 31 L

Substituting Equations 2.7.12 and 2.7.13 into a 2 = 3a1 leads to:

or

F2 = 3( F1 - 3F2 )

F1 10 = F2 3

(2.7.14)

Figure 2.7.10 shows the result obtained from GOYA-C, where the green numeral (-3.9) indicates the maximum deflection in the span. If you click the segment at the free end, the deflection of the free end (-1.48) will be indicated. If you increase F1 from 30 to 31.25, the deflection of the free end will be exactly zero, as will be discussed in the next section. Example 2.7.4
Calculate the deflection of a cantilever beam with a length L subjected to a uniformly distributed load w pushing the beam up.

6861X_C002.indd 143

2/7/08 [Link] PM

144

Understanding Structures: An Introduction to Structural Analysis

y dy dy/dx

0 (a) Slope of a curve y Large a

dx

Small a 0 Negative a (b) Various curves x

Figure 2.7.10 Quadratic functions.

Solution
As we studied in Section 2.3 (see Figure 2.7.11a), the bending moment over the span is: M= w 2 x 2 (2.3.3)

+d
P2 F dx

P1 x

(a) Deection of a beam v dv 0 dx (b) Deection and slope = dv/dx x

Figure 2.7.11 Cantilever beam.

6861X_C002.indd 144

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

145

where x is a coordinate indicating the distance from the fixed end. The bending moments and the rotations of the springs are: M1 = 9wL2 32 wL2 32

1 = 2 =

M1 L 9wL3 = EI 2 64 EI

(2.7.15) (2.7.16)

M2 =

M2 L wL3 = EI 2 64 EI

The deflection of the free end is: v = 1 3L L 27wL4 wL4 7wL4 + 2 = + = 4 4 256 EI 256 EI 64 EI

(2.7.17)

The result is 12 % smaller than that given by the continuously deformable model. Note that the deflection for uniform load is proportional to the fourth power of the beam length. The two-spring model is accurate enough for most practical uses.

2.8 Deformation of a BeamContinuously Deformable Model

If you we differentiate a quadratic function such as you will obtain we get dy/dx = 2ax + b (2.8.2) y = ax2 + bx + c (2.8.1)

As shown in Figure 2.8.1a, dy/dx represents the slope of the curve defined by the function. If we you differentiate it again, we obtain you will obtain d2y/dx2 = 2a (2.8.3)

As shown in Figure 2.8.1b, the radius of the curve corresponding to Equation 2.8.3 is reduced as the constant a increases. If a is negative, the curve is bent in the opposite direction. If the curve is defined by trigonometric ( y = sin x) or exponential ( y = ex) functions, the term d2y/dx2 also indicates how the curve is bent. The broken line in Figure 2.8.2a shows the deflected shape of the beam axis. The deflection at a distance x from the fixed end is denoted as vDv. Figure 2.8.2b shows
L

EI M

Figure 2.8.1 A couple at the free end.

6861X_C002.indd 145

2/7/08 [Link] PM

146

Understanding Structures: An Introduction to Structural Analysis

v d dv 0 1

ds =

d/

Figure 2.8.2 Exact definition of curvature.

the deflection vDv as a function of x, the distance from the fixed end. The first-order differential derivative of the deflection,

dv dx

(2.8.4)

In US usage, is the correct symbol. represents the slope of the beam axis. As we did in Section 2.6, we shall again assume that the planes P1 and P2 in Figure 2.8.2a remains perpendicular to the beam axis, and define the unit curvature as by the following equation.

d dx

(2.8.5)

Note that Equation 2.8.5 is equivalent to Equation 2.6.1 (f = a /a ) except that the curvature is defined in anfor infinitesimal region length dx rather than a finite length region a. Substituting Equation 2.8.4 into Equation 2.8.5 leads to:

d2v dx 2

(2.8.6)

In other words, the second-order differential of the deflection indicates how the beam bends or how the slope changes over a very small distance (dq /dx).


Equations 2.8.5 and 2.8.6 are approximate. The exact definition of unit curvature is

d ds

(2.8.7)

where ds is the length of the curve shown in Figure 2.8.3. The curve with a constant curvature is a circle with a radius 1/f. Using calculus, it can be shown that:

d2v dx 2

dv 2 1 + dx

3/ 2

(2.8.8)

In the case of beam deflection, dv/dx is very small. Therefore, (dv/dx)2 is negligible in beam, and thus we can assume f = d2v/dx2.

6861X_C002.indd 146

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

147

M (a) Bending moment M/EI (b) Curvature, = d2v/dx2

Linear

ML/EI

(c) Slope, = dv/dx

tic Quadra
(d) Deection, v

Figure 2.8.3

Recalling Equation 2.6.4 (M = EIf), we have: d2v M = dx 2 EI (2.8.9)

This result should be compared with that of Equation 2.7.1, where the deformation of the beam is evaluated within a finite region a. On the other hand, Equation. 2.8.9 deals with the bending deformation within an infinitesimal region dx. Solving this differential equation, we can determine evaluate the deflection formation of the beam. This is equivalent to assuming ane infinite number of springs representing the deformation within an infinitesimal region dx. Example 2.8.1
Assume that a cantilever beam is subjected to a couple M at the its free end (x = l). Determine the maximum slope and the maximum deflection of the beam at the free end.

Solution
First, we attempt to visualize how the beam would bend. The bending moment is distributed uniformly as shown in Figure 2.8.5a. The section does not change along the beam. We conclude that the curvature is uniform along the beam. Inspecting the sense of the applied moment, we conclude that the top fiber will be in compression and that the beam will bend into a shape concave -upward shape. From this information, we infer that the beam will bend as sketched in broken lines in Figure 2.8.4 and that the maximum slope and deflection will be at the free end.

6861X_C002.indd 147

2/7/08 [Link] PM

148

Understanding Structures: An Introduction to Structural Analysis

L

FL (a) Bending moment

FL/EI

Linear
(b) Curvature, = d2v/dx2

tic Quadra

(c) Slope, = dv/dx

Cubic

(d) Deection, v

Figure 2.8.4

We obtain the unit-curvature distribution using the expression d2vD/dx2 = M/EI (Figure 2.8.5b). Integrating this expression yields the slope

dv M = x + C1 dx EI

where C1 is a constant of integration. Because the slope is zero at the fixed end (dv/ dx = 0 at x = 0), we have C1 = 0. Thus, the slope q = dvD/dx is distributed linearly along the span (Figure 2.8.5c). Integrating the expression for the slope and noting

EI F

Figure 2.8.5 A force at the free end.

6861X_C002.indd 148

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

149

that the deflection is zero at the fixed end (v = 0 at x = 0), we get the following result (Figure 2.8.5d). v= M x2 EI

The slope and the deflection at the free end (x = L) are and D = v( L ) = ML2 EI (2.8.10)

(L) =

ML EI

(2.8.9)

Exercise

Using two randomly selected numbers Ii and j change the couple in the previous example to C = (i + 2) 100 N-mm and the depth to h = ( j + 5) mm, to and solve the example above again.

Example 2.8.2
Assume that a cantilever beam is subjected to a force F at the free end (x = L). Determine the slope and the deflection of the beam at the free end.

Solution
The bending moment varies linearly along the span as shown in Figure 2.8.7a:

M = F (L - x)

Use of q = M/EI leads to the unit-curvature distribution shown in Figure 2.6810b7b:

d2v F = (L - x) dx 2 EI

L/2

L/2

Figure 2.8.6 A load at mid-span.

6861X_C002.indd 149

2/7/08 [Link] PM

150

Understanding Structures: An Introduction to Structural Analysis

L/2 FL/2 (a) Bending moment L/2

FL/2EI (b) Curvature, = d2v/dx2 Constant

(c) Slope, = dv/dx

Constant

slope

(d) Deection, v

Figure 2.8.7

Integrating this equation and noting that the slope is zero at the fixed end (dv/dx = 0 at x = 0), we have

dv F = dx EI

x2 Lx 2

(2.8.11)

The slope at the free end (x = L) is given by the following equation.

(L) =

FL2 2 EI

(2.8.12)

Integrating Equation 2.8.12 and noting that the deflection is zero at the fixed end (v = 0 at x = 0), we have v= F Lx 2 x 3 - EI 6 2 (2.8.13)

The deflection at the free end (x = L) is given by the following equation. D = v( L ) = FL3 3EI (2.8.14)

6861X_C002.indd 150

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

L/2 F2

151

F1

Figure 2.8.8 Loads at the free end and the mid-span.

If you substitute the default values used in GOYA-C (F = 10 N, L = 100 mm, b = 10 mm, h = 15 mm, and E =100 N/mm2), you should get Dv = 11.9 mm. Check it the results using GOYA-C. Exercise
Use the last two digits of your ID ijtwo numbers iI and j you selected randomly t to change the force to F = (i + 2) N and the depth to h = ( j + 5) mm. Calculate the slope and deflection at mid-span and at the free end.

Example 2.8.3
A force F is applied at mid--span of a cantilever beam. Calculate the slope and deflection at the free end.

Solution
The bending moment varies linearly between 0 < x < lL/2 as shown in Figure 2.8.9a: L M = F - x 2 (2.8.15)

Figure 2.8.9b shows the unit curvature, q = M/EI. Integrating this curvature and noting that the slope is zero at the fixed end (dvVD/dx = 0 at x = 0), we have

dv F Lx x 2 = - dx EI 2 2
16

(2.8.16)

0.52 5

11 300

Figure 2.8.9 Two forces making the tip deflection at the free end zero (deformation magnified 8 times).

6861X_C002.indd 151

2/7/08 [Link] PM

152

Understanding Structures: An Introduction to Structural Analysis

Integrating the expression for the slope and noting that the deflection is zero at the fixed end (VD v = 0 at x = 0), we have v= F Lx 2 x 3 - EI 6 4 (2.8.17)

The slope and the deflection at the mid-span of the beam (x = L/2) are L FL2 = 2 8EI and FL3 L v = 2 24 EI (2.8.20) (2.8.19)

Note that we can obtain these equations substituting L /2 for L in Equations 2.8.12 and 2.8.14 with L/2. Also note that the bending moment is zero in the right half of the beam (L/2 < x < L) as shown in Figure 2.8.9a. The unit curvature d2v/dx2 is also zero there, which leads to the conclusion that the slope dv/dx is constant in that region as shown in Figure 2.8.9c. Integrating Equation 2.8.19, we have v= FL2 x +C 8EI (2.8.21)

where C is a constant of integration. Substituting x = L/2 and Equation 2.8.20 into Equation 2.8.21, we can determine the constant C. Fl 3 Fl 2 L = +C 24 EI 8EI 2 or C=FL3 48EI (2.8.22)

Substituting x = l L and Equation 2.8.22 into Equation 2.8.21 leads to the deflection at the free end (x = lL). D = v( L ) = FL2 L FL3 5FL3 = 8EI 2 48EI 48EI (2.8.23)

Example 2.8.4
Two loads, F1 and F2, are to be applied at the free end and at mid-span of a cantilever beam as shown in Figure 2.8.10. Determine the ratio F1/F2 required to make the deflection at the free end zero.

Solution
Though we can solve the problem by integrating the moment distribution created by F1 and F2, it is easier to use the results of the previous examples (Equations 2.8.14

6861X_C002.indd 152

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

50 mm F1 F2 50 mm

153

(i + 1) mm

(i + 1) mm

Figure 2.8.10 and 2.8.23). Denoting the deflection at the free end created by F1 and F2 as v1 and v2, respectively, the deflection caused by the two forces is given by the following equation using Equations 2.8.14 and 2.8.23. D = D1 - D 2 = Setting Dv = 0 leads to the following result F1 5 5 = 3 = F2 48 16 F1 L3 5F2 L3 3EI 48EI

Figure 2.8.11 shows the result obtained from GOYA-C.

L x wL2/2 (a) Bending moment

wL2/2EI
(b) Curvature, = d2v/dx2

(c) Slope, = dv/dx

(d) Deection, v

Figure 2.8.11

6861X_C002.indd 153

2/7/08 [Link] PM

154

Understanding Structures: An Introduction to Structural Analysis

Exercise
Take two numbers the last two digits of your ID ii and j you may choose. Assume F1 = (i + 2) N and F2 = ( j + 5) N. Calculate the slope and deflection at mid-span and at the free end for the previous example. (Use the default values in GOYA-C for the size of the section and Youngs modulus.) shall be the default values used in GOYA-B.

Exercise
Take the last two digits of your ID ij. two numbers i and j you may choose. Assume the beam depth to be h = ( j + 10) mm. Calculate the set of forces that yields a maximum deflection of (i + 1) mm as shown in the figure below. (Use the default values in GOYA-C for beam width and Youngs modulus.)

Hint: Apply F1 = 10 N in GOYA-C and increase F2 from zero until the upward deflection equals the downward deflection.

Example 2.8.5
Calculate the deflection of a cantilever beam with a length L subjected to a uniformly distributed load w pushing the beam up.

Solution
As we studied learned in Section 2.83, we determine the shear force distribution can be determined by integrating the uniform load w over the span.

V = wx + C1 If we define a coordinate indicating the distance from the fixed end, we get C1 = w.l L because V = 0 at the free end (x = lL). Also, Integrating the expression for shear, V, we obtain integrating V gives the bending moment distribution.

M=

w 2 ( x - 2 Lx ) + C2 2

We get obtain C2 = w.L2/2 because M = 0 at the free end (x = L). The unit curvature (q = M/EI ) is

d2v w = ( x 2 - 2 Lx + L2 ) dx 2 2 EI Integrating the expression above and using the boundary conditions, we get dv w x3 2 2 = - Lx + L x dx 2 EI 3  and w x 4 Lx 3 L2 x 2 + 2 EI 12 3 2

v= The deflection at the free end is

D = v( L ) =

wL4 8EI

(2.8.24)

6861X_C002.indd 154

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

< 1 mm < 1 mm Max. moment

155

Major Structural League Ranking (Unit: N-mm) > 200 Rookie > 5,000 Major > 500 A > 10,000 All Star > 1,000 AA > 20,000 MVP > 2,000 3A > 200,000 Hall of Fame

Figure 2.8.12

Exercise
Take two number i and j you may choose. the last two digits of your ID ij. Assume the distributed force as w = (i + 2) N/mm and the depth to as h = ( j + 5) mm. Calculate the slope and deflection at mid-span and at the free end.

Mini-game using GOYA-C The green figures in the window show the maximum upward and downward deflections of the beam. Obtain the largest possible bending moment using three loads while the maximum deflections does not exceed 1 mm. Keep the size of the beam and Youngs modulus at the default values.

Design your own beam (Part 5) We want to design a beam that can carry a nervous mini-elephant, whose weight of the elephant is (any one digit you choose the last two digits of your ID#/100) plus 101 lbfkg. Assume that the acceleration of gravity is 10 m/s2 and each leg carries the same amount of gravity gravitational force. The density of the beam is 0.5 lbf/in35 x 10-5 kg/mm3. The beam width is 10 [Link] and the Youngs modulus is 15,000 psi100N/mm2. Because the elephant is nervous, we are asked to design the beam so that its free end does not deflect more than 10 inmm. What is the required beam depth, h? Check your results using GOYA-C.
x

h Deection 1 in 2 in 2 in 6 in

Figure 2.8.13

6861X_C002.indd 155

2/7/08 [Link] PM

156

Understanding Structures: An Introduction to Structural Analysis

Hint: You may use the equation obtained in Example 2.8.5. You will have to solve a cubic equation for the beam depth, h. Because the equation (h3 + ah2 + bh + c = 0) is difficult to solve directly, you may resort to the following scheme: input y = h3 + ah2 + bh + c into a spreadsheet and gradually increase h until you determine a deflection not exceeding 10 inmm. What is unit curvature? SirMaster, I could not understand the idea concept of unit curvature.

Large radius Small radius Shorten much Shorten a little

Beam axis Lengthen much Large curvature

Beam axis Lengthen a little Small curvature

OK. Imagine a circle which approximates the beam axis. Unit curvature is the inverse of the radius of the circle. A lLarge curvature means a small radius, which means that the top of the beam shortens and the bottom lengthens considerably. Another important point is M = EIf , which indicates that the unit curvature is proportional to the bending moment. You need more moment to bend more. The equation, f = M/EI, also has an important meaning. An oak tree is hard to bend because it has a high value of Youngs modulus. A thick board is also hard to bend because it has large moment of inertia. Yes, the wood I used in Chapter 1 bent very easily. But why are you so stiff (or not flexible) whenwhile you are so slim? Cough! Anyhow, unit curvature has another face: it is the second derivative of the deflection with respect to distance, f = d2v/dx2, which leads to the important equation. d2v M = dx 2 EI

6861X_C002.indd 156

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

157

Integrating the bending moment, you can predict the beam deflection can be determined by calculation. Unit curvature represents how a beam curves, and the beam deflects because of unit curvature. It sounds so obvious circular. By the way, I dont like calculus. Its boring. Could you tell me how to understand the idea intuitively? Good question. In fact, calculus is rarely required in most structural design. Engineering intuition is much more important than calculus. The point is that a beam will be concave upward under positive bending moment and vice versa. You also need to train yourself so that you can image how the beam will deflect. Solve the problems in the next section and look at the graphics carefully in order to develop a good intuition.

(a) Force and deection Negative moment Positive moment Concave down Concave up +

(b) Bending moment and curvature

2.9 Problems
(Neglect self-weight in all the problems. Assume that all beams are prismatic.) 2.1 A cantilever beam is subjected to a couple M at its free end as shown in Figure 2.9.1. Select the correct pair of bending-moment distribution and free-end deflection from Table 2.9.1.
Table 2-9-1 Bending Free-end moment deection (a) 0 (a) ML2/(6EI) (a) ML2/(3EI) (b) 0 (b) ML2/(2EI)

M M (a) M (b)

1 2 3 4 5

Figure 2.9.1

6861X_C002.indd 157

2/7/08 [Link] PM

158

Understanding Structures: An Introduction to Structural Analysis

2.2 Two cantilever beams, one twice as long as the other, with identical sections are connected by a hinge at their free ends A force F is applied at the hinge as shown in Figure 2.9.2. Select the correct ratio of the vertical reactions RA and RC from Table 2.9.2.

Table 2-9-2 F RA L 2L RC 1 2 3 4 5 2 3 4 6 8

RA/RC

Figure 2.9.2

2.3 Two cantilever beams A and B, made of the same material, are loaded uniformly as shown in Figure 2.9.3. Select the correct ratio of their deflections at their free ends from Table 2.9.3.

w A L w B 2L d 2d d d Table 2-9-3 A/B 1 2 3 4 5 1/4 1/2 1 2 4

Figure 2.9.3

2.4 A load of 16 N is applied on the propped cantilever beam shown in Figure 2.9.4. Select the correct bending-moment diagram from Table 2.9.4.

6861X_C002.indd 158

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

Table 2.5 (1) 6 N-m 10 N-m

159

16 N

(2) 6 N-m

16 N-m 6 N-m 10 N-m

(3) 2m 2m (4)

6 N-m 10 N-m 12 N-m 10 N-m

(5)

Figure 2.9.4

2.5 A couple of 10 N-m is applied on the pin-supported end of the propped cantilever beam shown in Figure 2.9.5. Select the correct bendingmoment diagram for the beam from Table 2.9.5.
Table 2.6 (1) (2) 10 N-m (3) 5 N-m 4m (4) 10 N-m 10 N-m 15 N-m 10 N-m 10 N-m 10 N-m

10 N-m

(5)

Figure 2.9.5

2.6 A distributed load of 4 N/m is applied on the propped cantilever beam shown in Figure 2.9.6. Select the correct shear-force diagram for the beam from Table 2.9.6.

6861X_C002.indd 159

2/7/08 [Link] PM

160

Understanding Structures: An Introduction to Structural Analysis

Table 2.7 (1) (2) 4 N/m (3) 4m (4) (5) +16 N 4 N +8 N +10 N 6 N 8 N +8 N +8 N

Figure 2.9.6

2.7 Two columns made of the same linearly elastic material are connected with a rigid bar and a horizontal force is applied as shown in Figure 2.9.7. Select the correct ratio of the stresses at points a and b from Table 2.9.7.
Rigid bar Table 2.8 A a b Elevation D D a A Section b 2D B 2D B a/b 1 2 3 4 5 1/4 1/2 2 4 8

Figure 2.9.7

2.8 Assume that a cantilever beam with constant width but linearly varying depth is subjected to its self weight (Figure 2.9.8). Select the correct ratio of the bending moment at the fixed end MA and that at mid-span MB from Table 2.9.8.

6861X_C002.indd 160

2/7/08 [Link] PM

Moment and Deflections in Cantilever Beams

161

MA

Table 2.9 MB 1 2 3 MA/MB 2 4 6 8 12

L/2 L/2

4 5

Figure 2.9.8

6861X_C002.indd 161

2/7/08 [Link] PM

6861X_C002.indd 162

2/7/08 [Link] PM

Au: Please check the numbring of figures cited or Display maths also.

Frames

5.1 Introductory Concepts

The load-resisting capability of most modern buildings is provided by a structural system called a frame. It consists of beams and columns. In this section, we shall examine a beam subjected to bending moment and axial force as an introduction to understanding the response of a frame. Start GOYA-N to reach the window shown in Figure5.1.1. Reduce the vertical force to zero using the sliding bar in the upper right corner. You will see AU: Parts ac Figure5.1.2a on your screen. The axial load is constant along the span (P = F ) as x not explained in figure caption. shown in Figure5.1.2b. The axial force causes a uniformly distributed stress in the section (Figure5.1.2c). Recall that

P A

(A: sectional area)

(1.2.1)

Reduce the horizontal force to zero and increase the vertical load to develop Parts ad the condition in Figure5.1.3a. The shear force will be constant along the span AU: not explained figure cap(P = Fx) as shown in Figure5.1.3b, and the bending moment is distributed linearly in tion. (Figure5.1.3c). As discussed in Section 1.2, the normal stress on the section at the fixed end varies linearly with height of section (Figure5.1.3d). Recall that

M Z

(Z: section modulus)

(2.4.7)

If you apply horizontal and vertical forces simultaneously (Figure5.1.4a), the beam will elongate and deflect. The stress distribution will represent the sum or superposition of Figures5.1.2c and 5.1.3d (Figure5.1.4b). Note that the distribution is not symmetrical about the center of the section. In GOYA-N, it is assumed that the beam has a width of 10 mm, a depth of 15 mm, and a length of 100 mm (These dimensions were also used in GOYA-B). The axial stress caused by the horizontal force 10 N is

P Fx 10 = = = 0.07 N/mm 2 A A 10 15

whereas the bending stress caused by the vertical force 10 N is

M Fy L 10 100 = = = 2.7 N/mm 2 Z Z 10 152 / 6

229

6861X_C005.indd 229

2/7/08 [Link] PM

230

Understanding Structures: An Introduction to Structural Analysis

0.07 10 10 11.85 P/A + M/Z = 2.73

P/A M/Z = 2.6 10

1000

10

FIGure5.1.1 Initial window of GOYA-N.

Note that the stress caused by the bending moment is much larger than that caused by the axial force. The asymmetry of Figure5.1.4b is exaggerated. How about deformations? In GOYA-N, Youngs modulus is assumed to be 100 N/ mm2. Elongation caused by the horizontal tensile force is obtained using Equation 1.2.2. ux = L = Fx 10 L= 100 = 0.07 mm EA 100 10 15

The deflection caused by the vertical force is obtained using Equation 2.8.14. uy = Fy L3 3EI = 10 100 3 = 11.9 mm 3 100 (10 153 / 12 )
Fx ux

(a) Deformation P = Fx (b) Axial force

= P/A

(c) Stress

Figure5.1.2 Beam under Fx only.

6861X_C005.indd 230

2/7/08 [Link] PM

Frames

231

uy (a) Deformation V = Fy (b) Shear force (d) Stress M = Fy L L (c) Bending moment Fy = M/Z

Figure5.1.3 Beam under Fy only.

Note that the horizontal elongation is much smaller than the vertical deflection. Structural members are typically stiffer if loaded axially than if loaded transversely. You can move the location of the force using the sliding bar at the upper right of the window. Move the force to midspan as shown in Figure5.1.5a. The horizontal and vertical deformations reduce to 1/2 and 1/3, respectively. Note that the axial force, shear force, and bending moment are zero to the right of the load (Figure5.1.6). Example 5.1.1
Construct the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.7.

Solution
We can decompose the applied forces into horizontal and vertical components as shown in Figure5.1.8 to obtain the axial-force, shear-force, and bending-moment diagrams in Figure5.1.9.

Example 5.1.2
Construct the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.10.
P/A + M/Z = 2.74 N/mm2 Center P/A M/Z = 2.60 N/mm2 (b) Stress

ux = 0.07 mm Fx = 10 N uy = 11.85 mm Fy = 10 N (a) Deformation

Figure5.1.4 Beam under Fx and Fy.

6861X_C005.indd 231

2/7/08 [Link] PM

232

Understanding Structures: An Introduction to Structural Analysis

ux uy

Fx Fy (a) Deformation by Fx and Fy Fx

ux (b) Deformation by Fx

uy Fy (c) Deformation by Fy

Figure5.1.5 Move the force to the middle.

Solution
We can decompose the applied forces into their horizontal and vertical components (Figure5.1.11a,b) to obtain the axial-force, shear-force, and bending-moment diagrams in Figure5.1.11ce. The axial forces are shown as negative because they are compressive.

Example 5.1.3
Construct the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.12.

P = Fx (a) Axial force V = Fy (b) Shear force M = Fy a a (c) Bending moment

Figure5.1.6 Internal forces.

6861X_C005.indd 232

2/7/08 [Link] PM

Frames

233

10 N

10 N

10 N 1m 1m

10 N

Figure5.1.7 Beam under two forces.

10 N 10 N

(a) Horizontal forces 10 N 10 N

(b) Vertical forces

Figure5.1.8 Decomposed forces.

20 N

10 N

(a) Axial force 20 N 10 N

(b) Shear force 30 N m

10 N m

(c) Bending moment

Figure5.1.9 Internal forces.

10 N 10 N 1m 1m 10 N

10 N

Figure5.1.10 Beam under two forces.

6861X_C005.indd 233

2/7/08 [Link] PM

234

Understanding Structures: An Introduction to Structural Analysis

10 N

10 N

20 N

10 N

(c) Axial force (a) Horizontal forces 10 N (d) Shear force 10 N m (e) Bending moment

10 N 10 N (b) Vertical forces

Figure5.1.11 Solution.

Solution
The horizontal and vertical components of the applied force induce the reactions shown in Figure5.1.13a,b, leading to the axial-force, shear-force, and bending-moment diagrams in Figure5.1.13ce. Note that the axial force is zero in the right half of the beam.

Example 5.1.4
Construct the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.14.

Solution
The reactions are shown in Figure5.1.15a. Please note that there are no horizontal components because the roller support cannot resist any horizontal force. Figures5.1.15b,c show the axial and transverse components of the applied force and the reactions. Thus, we get the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.15df. This example is similar to the truss of Problem 1.12 in Section 1.11, Chapter 1.

Example 5.1.5
Construct the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.16.

Solution
Let us denote the reactions as shown in Figure5.1.17. Note that there is no vertical reaction at roller support A.
L/2 L/2

Fx Fy

Figure5.1.12 Simple beam.

6861X_C005.indd 234

2/7/08 [Link] PM

Frames

235

Fx

Fx (a) Reaction by Fx

P = Fx (c) Axial force V = Fy/2 V = Fy/2 (d) Shear force

Fy 2

Fy (b) Reaction by Fy

Fy 2 M = Fy L/4 (e) Bending moment

Figure5.1.13 Solution.

1m 1m 30 20 N

Figure5.1.14 Inclined simple beam (please call it simply supported).

5N 10 N 20 N 10 N 5N 10 N (b) Axial component 5N 5 3N (e) Shear force 5 3N 10 3N 5 3N (c) Transverse component (f) Bending moment 5 3N (a) Reactions 5 N (d) Axial force

5 3N m

Figure5.1.15 Internal forces.

6861X_C005.indd 235

2/7/08 [Link] PM

236

Understanding Structures: An Introduction to Structural Analysis

1m 1m 30 20 N

Figure5.1.16 Inclined simple beam.

Force equilibrium, SX = 0 and SY = 0, leads to the following equations, where the plus sign represents the force to the right or up. RAx + RBx = 0 -20 + RBy = 0 Moment equilibrium, SM = 0, around point B, leads to the following equation, where the minus sign represents counter-clockwise moment. -20 3 - RAx 1 = 0 2

Solving these equations, we get the reactions shown in Figure5.1.18a. Figure5.1.18b,c show the axial and transverse components of the applied force and the reactions, leading to the axial-force, shear-force, and bending-moment diagrams for the beam shown in Figure5.1.15df. This example is similar to the truss of Problem 1.13 in Section 1.11, Chapter 1. Imagine how the beam deforms: Point A moves down and the beam elongates. This causes the tensile axial force shown in Figure5.1.15d.

Example 5.1.6
Three forces are applied at the top of a cantilever column as shown in Figure5.1.19. Calculate the normal stress distribution at the fixed end. Hint: Evaluate the stresses caused by the axial force (Figure5.1.20a) and the bending moments around the x- and y-axes (Figure5.1.20b,c).

1m 1m RAx A 20 N RBy 30 3 m 2

RBx B 1m

Figure5.1.17 Reactions.

6861X_C005.indd 236

2/7/08 [Link] PM

Frames

237

10 3N 20 N 10 3N 20 N (a) Reactions 25 N 10 N (b) Axial component 15 N 5 3N 10 3N 5 3N (c) Transverse component 5 3N m 15 N

25 N

(d) Axial force

5 3N 5 3N (e) Shear force

(f) Bending moment

Figure5.1.18 Internal forces.

30 N 10 mm 10 N 10 N 30 mm 60 mm Fixed end z x y

Figure5.1.19 Cantilever column.

P = 30 N

Mx = 600 N-m

My = 600 N-m

y (a) Axial force (b) Moment around x-axis (c) Moment around y-axis

Figure5.1.20 Hint.

6861X_C005.indd 237

2/7/08 [Link] PM

238

Understanding Structures: An Introduction to Structural Analysis

0.9 N/mm2

= P/A = 0.1 N/mm2

= Mx/Zx = 1.2 N/mm2

+
1.2 N/mm
2

= My/Zy = 0.4 N/mm2 0.4 N/mm2

1.7 N/mm2

=
0.7 N/mm2 1.5 N/mm2

(a) Axial force

(b) Moment around x-axis

(c) Moment around y-axis

(d) Total

Figure5.1.21 Solution.
100 mm Fx = (10 + i) N 10 mm h

Fy = (10 + j) N
Au: provide Figure Caption.

Figure5.1.22

Solution
The stresses caused by the axial force and the bending moments are shown in Figure5.1.21ac, where Z x and Zy represent the section moduli with respect to x- and y-axes: Zx = 30 10 2 = 500 mm 3 6 and Zy = 10 30 2 = 1500 mm 3 6

Superposing these results leads to the solution shown in Figure5.1.21d.

Design Your Beam

Using single digits i and j of your choice, define the forces shown in Figure5.1.22. The material used for the beam has tensile and compressive strengths of 3 and 2 N/ mm2, respectively. The beam width is 10 mm. Determine the required beam depth ignoring the effect of shear stress.

5.2 A Simple Bent

Start GOYA-L to find the simple bent shown in Figure5.2.1. Vary the vertical force Fy using the sliding bar in the lower left corner. You will notice see that the free end C displaces not only vertically but also horizontally whether the force Fy acts up or down. Looking at the deflected shape, we can understand this intuitively. The horizontal member AB bends. This results in deflection and slope changes at corner B. The vertical member BC does not bend, but the vertical position of the free end C
 The horizontal displacement, ux , is assumed to be small enough compared with the length, L.

6861X_C005.indd 238

2/7/08 [Link] PM

Frames
ux Fy C uy Fy C ux

239

uy

L (a) Fy > 0

L (b) Fy < 0

Figure5.2.1 A simple bent

changes because of the deflection at B, and its horizontal position changes because as a result of the slope at B. In this section, we shall learn how to determine the deflected shape of the bent analytically. Cut the vertical member between B and C (Figure5.2.2a). You will find that a tensile axial force P = Fy acts at the cut to maintain equilibrium. Now, cut the bent at the horizontal member between A and B (Figure5.2.2b). You will find a negative (counter-clockwise) shear force V = Fy and a positive (concave upward) bending moment M = Fy ( L - x ). The distributions of the axial force, the shear force, and the bending moment are illustrated, plotted in Figure5.2.3. Inspecting from the moment diagram in Figure5.2.3c, we understand that the horizontal member deforms like a cantilever beam as shown in Figure5.2.4a, where,

Fy C

Fy C

x P = Fy A B M = Fy (L x) A

(L x)

B V = Fy

(a) Cut vertical member

(b) Cut horizontal member

Figure5.2.2 Free body.

6861X_C005.indd 239

2/7/08 [Link] PM

240

Understanding Structures: An Introduction to Structural Analysis

C P = Fy A B (a) Axial force A V = Fy (b) Shear force B A C x C

M = Fy (L x)

(c) Bending moment

Figure5.2.3 P, V, M diagrams.
AU: OK?

AU: Should this be Section 2.6 of Chapter 2?

the slope, and the deflection of point B are obtained using the equations for a cantilever beam subjected to a force applied atop the free end (Section 2.68, Chapter 2).

Fy L2 2 EI Fy L3 3EI

([Link])

v=

([Link])

Note that the vertical member remains straight as shown in (Figure5.2.4b) because there is no bending moment acting on it. Also note that member BC rotates by q because it needs to remain perpendicular to member AB at point B. We conclude that the horizontal displacement of end C is ux = - L = Fy L3 2 EI (5.2.1)

where the minus sign indicates that the top end displaces to the left. How about the vertical displacement? Because the axial deformation of the vertical member caused by the axial force P = Fy is negligible in comparison with the upward deflection of

uy Fy L2 2EI

ux

B A Fy L (a) Horizontal member v=

Straight Right angle A (b) Vertical member

B Fy L 3 v= 3EI

Fy L3 3EI

Figure5.2.4 Deformation.

6861X_C005.indd 240

2/7/08 [Link] PM

Frames
ux Fx ux

241

uy

uy

Fx

L (a) Fx > 0

L (b) Fx < 0
AU: Parts a, b not explained here.

Figure5.2.5 Horizontal force.

member AB (Section 5.1), we conclude that the vertical displacement of end C is equal to the deflection of the horizontal member at B. (5.2.2) 3EI Reduce the vertical force to zero and apply a horizontal force as shown in Figure5.2.5. You see that the horizontal and vertical displacements are much larger than they were when the vertical force was applied. Why? Cut the vertical member between B and C (Figure5.2.6a). You will find a shear force V = Fx and a bending moment M = Fx ( L - y) at the cut (the coordinate y defines the height of the cut). If you cut the horizontal member (Figure5.2.6b), you will find a tensile axial force P = Fx and a bending moment M = Fx L . The distributions of the
Fx C (L y) V = Fx M = Fx (L y) B y A M = Fx L (a) Cut vertical member (b) Cut horizontal member P = Fx L Fx C

uy = v =

Fy L3

Figure5.2.6 Free body.

 According to the definition in Chapter 2, the bending moment is negative because it is concave downward. However, this sign system is confusing when dealing with sophisticated structures. From here, we shall ignore the sign of the bending moment. As long as you know which side is compressed, you can ignore the sign.

6861X_C005.indd 241

2/7/08 [Link] PM

242

Understanding Structures: An Introduction to Structural Analysis

C C V = Fx P = Fx A B A B (b) Shear force (a) Axial force A M = Fx L B (c) Bending moment C

Figure5.2.7 P, V, M diagrams.

axial force, the shear force, and the bending moment are illustrated, plotted in Figure5.2.7. The moment does not change with distance from point A (Figure5.2.7c) because there is no shear in the horizontal member (Figure5.2.7b). Note that the bending moment is continuous at point B (Figure5.2.7c): The moments at point B acting on the horizontal member and vertical member are the same. The joint AU: Sentence between the two members, considered ideally to be a point in this case, has to be in incomplete. equilibrium. This is important, and is an important feature of frames with continuous members. From the sense (sign) of the moment diagram in Figure5.2.7c, we understand that the horizontal member deforms as shown in Figure5.2.8a. The slope and deflection at point B are obtained using the equations for a cantilever beam subjected to AU: Should uniform bending moment M = Fx L (Section 2.68, Chapter 2). this be Section
2.6 or 2.8?

AU: Check equation numbers.

Fx L2 EI Fx L3 2 EI

([Link]) ([Link])

v=-

The vertical member does not remain straight as in the previous case; it bends as shown in (Figure5.2.8b). Also note that the vertical member rotates by q at its base because it is connected to remain perpendicular to the horizontal member at point B.
FxL3 3EI C FxL3 3EI C

Fx M = Fx L A L B v= = Fx L3 2EI Fx L2 EI

uy L

ux L

(a) Horizontal member

B (b) Vertical member

A (c) Total B

v=

Fx L3 2EI F L2 = x EI

Figure5.2.8 Deformation.

6861X_C005.indd 242

2/7/08 [Link] PM

Frames

243

We conclude that the horizontal displacement of the top end is the sum of the deflections related to the base rotation and bending of the vertical member: ux = L + Fx L3 4 Fx L3 = 3EI 3EI (5.2.3)

The vertical displacement of the top end is the same as that at end B of the horizontal member. uy = v = Fx L3 2 EI (5.2.4)

Compare these results with Equation 5.2.1 and 5.2.2; the horizontal and vertical displacements caused by Fx are 2.7 and 1.5 times those caused by Fy (= Fx).

For Interested Readers: Joint

An interested reader may ask the following questions: 1. In the previous discussion, we assumed that the lengths of the horizontal and vertical members are L. The actual lengths are, however, L h/2 as shown in Figure5.2.9a. If we recognize that, would the displacement be determined to be smaller than the results discussed? 2. Does recognition of member thickness affect the axial-force, shear-force, and bending-moment diagrams? 3. What happens within the joint enclosed by the broken lines in Figure5.2.9a? Does it deform? The answer to the first question is yes. But the effect is small if L is much larger than h. The answer to the second question is no. You can check it by cutting the bent at any place: h does not affect equilibrium. The third question is not so simple.

Tension h Lh 2 Compression Zero L 2 h 3 2 h 3 (b) Truss model (c) Deformation of the joint

h Lh 2 L (a) Original

Figure5.2.9 Beam-column joint.

6861X_C005.indd 243

2/7/08 [Link] PM

244

Understanding Structures: An Introduction to Structural Analysis

To answer it, we may analyze a truss model as shown in Figure5.2.9b, where the solid and broken lines indicate tensile and compressive members, respectively. Note that the joint is stressed diagonally so that the joint deforms as shown in Figure5.2.9c. This deformation increases the displacement at the top end of the bent, but not significantly in most cases. When you design an actual bent, however, you should pay attention to the effects of stresses in the joint to make it safe. Example 5.2.1
Construct the axial-force, shear-force, and bending-moment diagrams for the bent in Figure5.2.10. Also, calculate the displacement at point C.

Solution
Cut members AB and BC as shown in Figure5.2.11 to obtain the axial-force, shear-force, and moment diagrams shown in Figure5.2.12ac. We can also obtain Figure5.2.12c by superposing the moment diagrams related to the horizontal (F) and vertical (2F) two force diagrams as illustrated in Figure5.2.12d. The moment diagram in Figure5.2.12c is the algebraic sum of those in Figure5.2.12d. The displacements can be obtained by either one of two methods. One is to use Equations 5.2.1 through 5.2.4, substituting Fy = 2F and Fx = F. Another is to use the following equation for the deflection of a cantilever beam, which is useful in many cases:
AU: Check equation number.

v=

FL3 ? 3EI

([Link])

The deflection at midspan of the horizontal member is obtained substituting 2F F for F 2F and L/2 L for L L/2 because the bending moment is zero at midspan (Figure5.2.13a): vm = 2F L FL3 = 3EI 2 12 EI
3

AU: Check equation number.

([Link])

The deflected shape of the horizontal member is symmetrically about its midspan; thus, the vertical displacement of the top end is uy = 2vm as shown in (Figure5.2.13b). Next,

2F F C L

Figure5.2.10 Bent with two forces.

6861X_C005.indd 244

2/7/08 [Link] PM

Frames

245

2F F C

2F C

V M A P B A P M B V

(a) Cut vertical member

(b) Cut horizontal member

Figure5.2.11 Free body. note that the slope of the horizontal member is zero at B. Therefore, the horizontal displacement ux is obtained using Equation 2.6.17 again as shown in Figure5.2.13b.

Example 5.2.2
Construct the axial-force, shear-force, and bending-moment diagrams for the bent in Figure5.2.14. Also calculate the displacement at point C.

Solution
Considering the free-body diagrams of the horizontal and vertical members as shown in Figure5.2.11 leads to the diagrams shown in Figure5.2.15ac. We can also obtain Figure5.2.15c by superposing the effects of the vertical and horizontal forces as we did in Figure5.2.12d. The horizontal member deforms in the same way as in the previous example because the bending moment is the same. Therefore, the vertical displacement uy is equal to that in the previous example, as shown in Figure5.2.16c. To evaluate the deformation of the vertical member, imagine a cantilever beam shown in (Figure5.2.16a). Because the member is straight where the bending moment is zero, as shown in Figure5.2.16b, the deflection of the free end is v = L 5FL3 + vm = 2 24 EI (2.6.25)

2F

2F
F 2F

F
FL FL (b) Shear force (c) Bending moment

=
FL

+ 2FL

(a) Axial force

(d) Components

Figure5.2.12 P, V, M diagrams.

6861X_C005.indd 245

2/7/08 [Link] PM

246

Understanding Structures: An Introduction to Structural Analysis

FL3 ux = 3EI C

uy =

FL3 6EI

L/2 2F vm A

L/2 vm (b) Bent

L/2 v = 2vm = Zero slope FL3 6EI

(a) Cantilever beam

Figure5.2.13 Deformation. as we found in Example 2.6.4, Chapter 2. Recall that the slope of the lower right corner is zero as shown in (Figure5.2.16c). Therefore, the horizontal displacement ux is equal to v in Equation 2.6.25, Chapter 2.

Example 5.2.3
Construct the axial-force, shear-force, and bending-moment diagrams for the bent in Figure5.2.17. Also, calculate the displacement at point E.

Solution
Consider the free-body diagrams as shown in Figure5.2.18. You will note that the shear force is positive (clockwise) in Figure5.2.18a, zero in Figure5.2.18b, and negative (counterclockwise) in Figure5.2.18c. This leads to the diagrams shown in Figure5.2.19. To calculate the displacement, we use the deflection of a cantilever beam shown in Figure5.2.20a, again. Because the moment diagram is symmetric about midspan (point B), the displacement of point C is twice the deflection of point B. v= 2 FL3 3EI (5.2.5)

2F C

L/2 2F L/2

Figure5.2.14 Bent with two forces.

6861X_C005.indd 246

2/7/08 [Link] PM

Frames

247

2F 2F 2F (a) Axial force

2F FL

FL

(b) Shear force

(c) Bending moment

Figure5.2.15 Inner forces.

v= FL3 vm = 12EI

5FL3 24EI uy = FL3 6EI

ux =

5FL3 24EI

ght Strai

2F
= FL2 4EI

vm

L/2 2F L/2

L/2 B

Zero slope A (c) Total B

v= FL3 6EI

B (b) Vertical member

(a) Cantilever beam

Figure5.2.16 Deformation.

2L

C L

Figure5.2.17 Bent with two joint.

6861X_C005.indd 247

2/7/08 [Link] PM

248

Understanding Structures: An Introduction to Structural Analysis

(a)

(b)

(c)

Figure5.2.18 Free body.

V = F P=F (a) Axial F. (b) Shear F. (c) Bending M. V=F M=FL

Figure5.2.19 Inner forces.

L ux F
FL 3EI (a) Cantilever beam vm =
3

vm

uy

Zero slope C 2FL3 v= 3EI

D (b) Total

Figure5.2.20 Deformation.

6861X_C005.indd 248

2/7/08 [Link] PM

Frames

249

Noting that the slope at point C is zero, we can conclude that the displacement of the loaded end is obtained as the displacement of point C plus the result obtained in Figure5.2.8. ux = 2 FL3 4 FL3 2 FL3 + = 3EI 3EI EI uy = FL3 2 EI (5.2.6) (5.2.7)

Example 5.2.4
Construct the bending-moment diagram of the structure in Figure5.2.21a. Select the correct deformed shape among those in Figure5.2.21bd. (Hint: look at the deformation of the supporting beam, AB.)

Solution
The reaction at support A (RA in Figure5.2.22a) is obtained by moment equilibrium around support B ( RA L - F c = 0 ): RA = c F L (5.2.8)

The reaction at support B (RB in Figure5.2.22a) is obtained by moment equilibrium around support A: RB = a+b F L (5.2.9)

A a b L (a) Structure F c

A (b) Deformation

A (c) Deformation

A (d) Deformation

Figure5.2.21 Which is correct?

6861X_C005.indd 249

2/7/08 [Link] PM

250

Understanding Structures: An Introduction to Structural Analysis

F

A RA a

C L (a) Reactions F RA RB (b) Shear force Fb RA a C (c) Bending moment RB (b + c) b c

B RB

Fb

RA a

RB (b + c)

(d) Equilibrium

Figure5.2.22 Solution.

The shear-force diagram is obtained as shown in Figure5.2.22c. This leads to the bending-moment diagram shown in Figure5.2.22c. At joint C, bending moments balance as shown in Figure5.2.22d, or Fb + RA a - RB ( b + c) = 0 (5.2.10)

You can confirm Equation 5.2.10 if you substitute Equations 5.2.8 and 5.2.9 into Equation 5.2.10: c a+b Fb + F a - F (b + c) = 0 L L (5.2.11)

Note that the bending moment in beam AB is upward (concave) everywhere. Therefore, we conclude that Figure5.2.21b is correct. You should check the results using GOYA-D.

Exercise
Take two numbers i and j (you may choose the last two digits of your ID, ij) to determine the location of the vertical member in Figure5.2.23. Draw the bending-moment diagram of the structure shown in Figure5.2.23 and sketch the deformations.

6861X_C005.indd 250

2/7/08 [Link] PM

Frames
50 50 N 50 N 50 50 N

251

50 + ij

150 ij 200 (a)

50 + ij

150 ij 200 (b)

50 + ij

150 ij 200 (c)

Figure5.2.23 Structure.

Example 5.2.4
Figure5.2.24 shows a bent, of which point C is supported by a vertical roller at point C. Assume that the bent is subjected to a vertical force Fy. Determine the reaction R, the bending-moment diagram, and the deflection of point C. Hint: Use the superposition technique in Example 2.8.4, Chapter 2, for a beam with a fixed end and a roller support.

Solution
Recall that, in Figure5.2.1a, the downward force on a simple bent (Fy < 0) causes rightward movement of point C to the right (Equation 5.2.1). On the other hand, a leftward force to the left (Fx < 0 in Figure5.2.5b) causes leftward movement of point C to the left (Equation 5.2.3). To make the lateral movement zero, we set the right-hand terms of the two equations equal to each other. Fy L3 2 EI = 4 Fx L3 3EI (5.2.12)

In this case, the horizontal force Fx is equivalent to the reaction R. Thus, we obtain R = Fx = 3 Fy 8
Fy uy C L R

(5.2.13)

A B L
Ai: provide Caption.

Figure5.2.24

6861X_C005.indd 251

2/7/08 [Link] PM

252

Understanding Structures: An Introduction to Structural Analysis

Fy Fy (3/8)Fy

(3/8)Fy

+
Fy L (a) Vertical force
AU: Caption?

(3/8)Fy L (b) Horizontal force

=
(5/8)Fy L

(3/8)Fy L (c) Total

Figure5.2.25 Bending-moment diagrams. The bending-moment diagram is obtained by superposing those given by Fy, and Fx = (3 / 8) Fy , as shown in Figure5.2.25. The deflection is obtained similarly superposing the values given by Equation 5.2.2 and 3/8 times that given by Equation 5.2.4 (see Figure5.2.26). uy = Fy L3 3EI (3 / 8) Fy L3 2 EI = 7 Fy L3 48EI (5.2.14)

Note that the deflection is less than one half of that given by Fy only. In other words, the vertical roller makes the structure stiff. Check this conclusion using GOYA-L.

Design Your Structure

Take the last digit of your ID number. Assume any one-digit number i. We want to design a structure shown in Figure5.2.25 that can carry a force F = 50 N. Assume that the material strength is 3 N/mm2 and Youngs modulus is 100 N/mm2. You can neglect its self-weight. The cross section of the member shall be square. What is the required cross section? How much is the deflection? Hint: Reexamine the process leading to Equations 5.2.1 and 5.2.3 very carefully and derive equations that are applicable to a bent in which the length of horizontal member is different from that of the vertical member. Then, follow the process in Example 5.2.4. Use M = Z to determine the cross section. Check your result using GOYA-L.
Fy
uy = Fy L3 3EI ux = Fy L
3

ux = uy = 3Fy L3 16EI

Fy L3 2EI uy = 7Fy L
3

Fy
48EI

2EI

(3/8)Fy

(3/8)Fy

+
(a) Vertical force (b) Horizontal force

=
(c) Total

Figure5.2.26 Deflected shape.

6861X_C005.indd 252

2/7/08 [Link] PM

Frames
Fy = 50 N Deection C 100 R

253

A (11 + i) 10
AU: Caption & citation?

Figure5.2.27

5.3 A Portal Frame

Start GOYA-P to find the portal frame shown in Figure5.3.1. Apply a horizontal force at the support. You will find diagrams showing distributions of axial force, shear force, and bending moment (Figure5.3.2). Figure5.3.3a shows possible reactions for the frame as well as the applied force. The equilibrium of horizontal forces requires RAx = FD. The equilibrium of vertical forces and moments requires RAy = RBy = 0. The reactions should be as shown in Figure5.3.3b. As we have done in the previous section, we consider the free-body diagrams shown in Figure5.3.4 to obtain the axial-force, shear-force, and bending-moment diagrams shown in Figure5.3.2. Note that the free-body diagrams in Figure5.3.4b,c yield axial forces and bending moments having the same magnitudes.
L

FD

Figure5.3.1 Portal frame with horizontal force at support.

 A frame is a structural system that may include many beams and columns. In this section, we consider a portal frame that includes only three members: two columns and a beam.

6861X_C005.indd 253

2/7/08 [Link] PM

254

Understanding Structures: An Introduction to Structural Analysis

FDL FD +FD FDL

FD

(a) Axial force

(b) Shear force

(c) Bending moment

Figure5.3.2 P, V, and M diagrams.

Figure5.3.5 shows the deflected shape of the portal frame. Make a special effort to relate the deflected shape of the frame to its bending-moment diagram (Figure5.3.2c). Note that the moment causes tension on the outside and compression on the inside for all three members of the frame. That condition suggests that all three members should be bending concave out. The deflected shape confirms this expectation. To become a competent structural engineer, one should always go through this type of reasoning and relate bending-moment distribution to deflected shape. Because the bending moment in the beam is uniform and the slope at midspan is known to be zero in this symmetric case, the slope at beam end (q in Figure5.3.5b) is determined to be

L/2

M ML FD L2 dx = = EI 2 EI 2 EI

(5.3.1)

The displacement of joint B is obtained by superposing the contributions of the slope of the beam (q.L) and the flexural deformation of column AB given by Equation 2.8.14, Chapter 2. uB = L + FD L3 5FD L3 = 3EI 6 EI (5.3.2)

Note that the deformation is symmetrical: i.e., column CD deforms in the same way as column AB. Therefore, the displacement of support D is u x = 2uB = 5FD L3 3EI (5.3.3)

RAx RAy

FD RDy

FD

FD

(a) Possible reactions

(b) Reactions

Figure5.3.3 Reactions.

6861X_C005.indd 254

2/7/08 [Link] PM

Frames
P M P M M

255

V FD

M FD (a) Cut left column

V FD (c) Cut beam and see right (d) Cut right column FD

(b) Cut beam and see left

Figure5.3.4 Cut the portal frame.

The length L is assumed to be 100 mm in GOYA-P. Change the height of the member section to 22.9 mm so that EI will be 1 106 N/mm2 and apply a force of F = 6 N. You will obtain ux = 5 6 10 6 / (3 10 6 ) = 10 mm . Assume a vertical load on the portal frame as shown in Figure5.3.6a. Select the plausible deflected shape among those shown in Figure5.3.6be. To solve the problem, we need to determine the reactions. See Figure5.3.7. Equilibrium of horizontal forces yields RAx = 0. Equilibrium of vertical forces and moments requires RAy = RBy = Fy / 2 . Reactions are shown in Figure5.3.7b. Considering the free-body diagrams in Figure5.3.7c,d, we obtain the axial-force, shearforce, and bending-moment diagrams in Figure5.3.8. Because the bending moment is zero in each column, we reject Figure5.3.6c,d with bent columns. We reject Figure5.3.6e, in which the beam is concave down at its ends. Figure5.3.6b is the plausible deflected shape because the beam bends concave up, consistently with the moment diagram, and the columns remain straight because there is no moment acting on them. Let us determine the magnitude of the deflections and rotations in reference to Figure5.3.9. For the loads applied, the bending moment in the beam is same as that
uB B uB C M = FDL L uB M = FDL

x L

ux L FD L3 3EI A D FD

(b) Beam

(a) Total

Figure5.3.5 Deflected shape.

 Note that we could have inferred this result directly from symmetry.

6861X_C005.indd 255

2/7/08 [Link] PM

256

Understanding Structures: An Introduction to Structural Analysis

Fy

Straight

Straight

Curved

(b)

(c)

A (a) Force

D (d) (e)

Figure5.3.6 Portal frame with vertical force on beam.

of a simply supported beam (Figure5.3.9b). Expressions for the deflection and the end slope of a simply supported prismatic beam with a concentrated load at midspan were given in Section 3.1, Chapter 3: v= Fy L3 48EI Fy L2 16 EI (3.1.7) (3.1.4)

Because the columns are continuous with the beams at the joints, the slopes of the columns are also q as shown in Figure5.3.9a. With the pin support at A not being free to move horizontally, the horizontal displacement of the roller support D is ux = 2 L = Fy L3 8EI (5.3.4)

Check the result using GOYA-P. To have a lateral movement of zero at the support, the movement in caused by the lateral force FD (Equation 5.3.3) should be offset by the movement out caused by the
Fy Fy P = Fy /2 M V

RAx RAy RDy Fy /2 (b) Reactions Fy /2 Fy /2 (c) Cut right column (d) Cut beam Fy /2

(a) Possible reactions

Figure5.3.7 Reactions.

6861X_C005.indd 256

Curved
2/7/08 [Link] PM

L/2

L/2

Frames
Fy/2 Fy/2 Fy /2 Fy /2 FyL/4

257

(a) Axial force

(b) Shear force

(c) Bending moment

Figure5.3.8 P, V, and M diagrams.

vertical force Fy (Equation 5.3.4). Therefore, we set the right-hand terms of the two equations equal to one another.
3 5FD L3 Fy L = 3EI 8EI

(5.3.5)

from which we obtain the required proportion of FD and Fy. FD 3 = Fy 40 (5.3.6)

The axial-force, shear-force, and bending-moment diagrams are obtained by superposing 3/40 times the values in Figure5.3.2 on those in Figure5.3.8 as shown in Figure5.3.10bd. The conditions depicted in Figure5.3.10a are equivalent to those in Figure5.3.11a where a portal frame is supported by two pins. The frame deforms (Figure5.3.11b) in accordance with the bending-moment diagram in Figure5.3.10d. Assume that a horizontal load Fx is applied on the portal frame (Figure5.3.12a). Select the correct deflected shape among those in Figure5.3.12be. The first step is to determine the reactions shown in Figure5.3.13a. The given support conditions will allow horizontal and vertical reactions at A and a vertical
B v B Fy/2 (b) Beam D (a) Total v Fy/2 C Fy C

ux A

Figure5.3.9 Deflected shape.

6861X_C005.indd 257

2/7/08 [Link] PM

258

Understanding Structures: An Introduction to Structural Analysis

3 F 40 y

Fy/2

Fy/2

3 FL 40 y

FyL/4

Fy Fy/2 (a) Force FD = 3 F 40 y 3 F 40 y Fy/2 Fy/2

Fy/2 3 F 40 y

3 FL 40 y

7 FL 40 y

3 F 40 y

(b) Axial force

(c) Shear force

(d) Bending moment

Figure5.3.10 Reduce the horizontal displacement of the roller to zero.

Fy Right angle

Concave down Right angle

3 F 40 y Fy/2 Fy/2

3 F 40 y

Concave left

Concave up (b) Deected shape

Concave right

(a) Reactions

Figure5.3.11 Portal frame with pin supports.

Fx

L B C

(b) Vertical

(c)

A (a) Force

D (d)

(e)

Figure5.3.12 Portal frame with horizontal force.

6861X_C005.indd 258

2/7/08 [Link] PM

Frames
L L

259

Fx

Fx

L RAx RAy RDy Fx Fx (b) Reactions Fx

(a) Possible reactions

Figure5.3.13 Reactions.

reaction at D. We note that the height of the frame is equal to its span. The equilibrium of horizontal forces yields RAx = - Fx . The equilibrium of vertical forces and moments requires RAy = - Fx and RAy = Fx . We conclude that the reactions are as shown in Figure5.3.13b. Considering the free-body diagrams (Figure5.3.14ac), we obtain the axialforce, shear-force, and bending-moment diagrams in Figure5.3.15. We look at Figure5.3.15c carefully. The moment diagram indicates that the left column and the beam should have compressive strain on their outside faces. They should bend so that they are concave out. The right column, not subjected to moment, should remain straight. We decide that the correct deflected shape is the one shown in Figure5.3.12e. (Check using GOYA-P.) Let us compute the deformations in reference to Figure5.3.16a. Because the two ends of the beam are connected to the columns, vertical displacements at the two ends of the beam are considered to be negligible. Therefore, the beam bends as shown in Figure5.3.16b. Recall the following equation in Chapter 2: d2v M = ? dx 2 EI (2.8.9)

P M V Fx Fx (a) Cut left column

Fx V M

Fx Fx (b) Cut beam Fx (c) Cut right column

Figure5.3.14 Free-body diagrams.

6861X_C005.indd 259

2/7/08 [Link] PM

260

Understanding Structures: An Introduction to Structural Analysis

Fx L Fx +Fx Fx +Fx

(a) Axial force

(b) Shear force

(c) Bending moment

Figure5.3.15 P, V, and M diagrams.

Substituting the bending moment shown in Figure5.3.16b, M = Fx ( L - x ) , and integrating, we obtain and v= Fx x 2 L x 3 - + C1 x + C2 EI 6 2 dv Fx = dx EI x2 xL + C1 2 (5.3.7)

(5.3.8)

The boundary condition that v = 0 at x = 0 and x = L leads to C1 = Fx L2 3EI and C2 = 0 (5.3.9)

uB B B L uB ux BL FxL3 3EI (a) Portal C

M = Fx (L x) C CL M B (b) Beam C x

Figure5.3.16 Deflected shape.

6861X_C005.indd 260

2/7/08 [Link] PM

Frames

261

Substituting this into Equation 5.3.6, we obtain the slopes at beam ends:

B = -

dv dx

=
x=0

Fx L2 3EI

and

B =

dv dx

=
x= L

Fx L2 6 EI

(5.3.10)

The horizontal displacement of the beam (uA in Figure5.3.16a) is obtained by adding the deformation of the left column and the contribution of the rotation: uB = Fx L3 2 F L3 + B L = x 3EI 3EI (5.3.11)

The displacement of the roller (ux in Figure5.3.16a) is obtained by adding the displacement of the beam and the contribution of the rotation of the right column: u x = uB + C L = 5Fx L3 6 EI (5.3.12)

Figure5.3.17a shows the deformation of the portal frame subjected to a horizontal force Fx (Figure5.3.13). Figure5.3.17b shows the deformation of the portal subjected to a horizontal force Fx/2 on the roller support (see Equations 5.3.2 and 5.3.3). Adding the deflections in the two figures, we obtain Figure5.3.17c, which refers to a portal frame supported by two pins and subjected to a horizontal force at the top. Check this result using GOYA-P. Superposing one half of Figure5.3.2 and Figure5.3.15, we obtain the axial-force, shear-force, and bending-moment diagrams in Figs. 5.3.18ac. The bending-moment diagram indicates that the portal frame deforms as shown in Figure5.3.18d. Example 5.3.1
Construct the bending-moment diagram for the portal frame shown in Figure5.3.19.

Fx

2Fx L3 3EI

5Fx L3 12 EI

Fx

3Fx L3 12 EI

+
(a) (b)

Fx/2

=
Fx

Fx/2 (c)

Fx/2 Fx

5Fx L3 6 EI

5Fx L3 6 EI

Figure5.3.17 Reduce the horizontal displacement of the roller to zero.

6861X_C005.indd 261

2/7/08 [Link] PM

262

Understanding Structures: An Introduction to Structural Analysis

Concave up Concave down Fx/2 Fx L/2 Fx Fx/2 Fx Fx/2 Fx L/2 Right angle Right angle

Fx

Concave left (c) Bending moment

Concave left (d) Deected shape

(a) Axial force

(b) Shear force

Figure5.3.18 Portal frame with pin supports.

2F F L L

Figure5.3.19 Portal frame.

2F F FL FL

F F (a) Vertical force F F/2 (b) Horizontal force F/2

2F

FL

3FL/2

F/2 (c) Both

3F/2

Figure5.3.20 Bending moment.

6861X_C005.indd 262

2/7/08 [Link] PM

Frames
Fy = 10 70 + i 70 + i Fx = 5 140 + 2i

263

100

100

(a) Vertical force Fx = 5 70 + i

Fy = 10 70 + i

(b) Horizontal force

100

(c) Both
AU: Figurecaption?

Figure5.3.21

Solution
The reactions and the bending moment caused by the vertical force are shown in Figure5.3.20a. Those caused by the horizontal force are shown in Figure5.3.20b. Superposing these figures, we obtain Figure5.3.20c. (Determine the deflected shape using GOYA-P.)

Exercise
Select any one-digit number i. Construct the bending moment diagram for the portal frame shown in Figure5.3.19 and sketch the deflected shape.

Hint: Reexamine the process leading to Equation 5.3.5 very carefully and project it to the case where the length of the beam is different from that of the columns (Figure5.3.21).

AU: Figure 5.3.21 not cited in text. Tentatively cited here. OK?

5.4 Statically Indeterminate Frame

In Section 1.5, Chapter 1, we distinguished between statically determinate and statically indeterminate trusses: the axial forces in the statically determinate trusses can be calculated considering the equilibrium of forces only, but the calculation of the axial forces in the statically indeterminate trusses requires, in addition to statics, consideration of deformation compatibility. In this section, we shall study a statically indeterminate frame, for which the calculation of the internal forces (axial forces, shear forces, and bending moment) requires consideration of both force equilibrium

6861X_C005.indd 263

2/7/08 [Link] PM

264

Understanding Structures: An Introduction to Structural Analysis

and deformation compatibility. Many frames are statically indeterminate. We need to learn to determine their response to load. Compared with the calculation process for a determinate structure, that for an indeterminate structure is longer yet straightforward. Recall Figure1.5.9, Chapter 1, where the axial forces in statically indeterminate trusses changed when we changed the axial stiffness EA for the section. Similarly, the internal forces in an indeterminate frame changes if we change the bending stiffness EI (Youngs modulus multiplied by the cross-sectional area) for the section. In Figure5.3.10d, we obtained a bending-moment diagram for a portal frame with all of its members having the same stiffness. If we assume that the stiffness of column CD is extremely large as shown in Figure5.4.1, which bending-moment diagram included in Figure5.4.2ad is the correct one? We note that the moment diagram in Figure5.4.2a is identical with the one obtained for a frame with all of its members having the same stiffness and, in Figure5.4.2b,c, the columns resist larger moments than those in Figure5.4.2a. The portal frame in Figure5.4.1 is an indeterminate frame. We cannot determine the internal forces using the conditions of equilibrium alone. So we decide to go through a simple four-step procedure. Step 1: We release the horizontal-force restraint on reaction D (Figure5.4.3a). Now the portal frame, with the degrees of freedom increased, is determinate. We can determine the reactions and the internal forces on the basis of equilibrium conditions alone. Step 2: Given the distributions of internal forces, we can determine the horizontal displacement, uFD, at reaction D caused by the applied vertical load F (Figure5.4.3c). Step 3: Again referring to the frame with the released horizontal restraint, we apply a horizontal force of a given magnitude, R, at reaction D (Figure5.4.4a) and determine the horizontal displacement uRD it causes (Figure5.4.4c). (The value of R is not important but it may simplify arithmetic if it is chosen to be unity.)

L/2

F EI

L/2

EI

Extremely Large EI D

Figure5.4.1 A portal frame.

6861X_C005.indd 264

2/7/08 [Link] PM

Frames
3 F 40 3 F 40 3 F 32 3 F 32

265

7 F 40

5 F 32

(a) FL/8 FL/8 FL/6

(b)

FL/8

FL/6

(c)

(d)

Figure5.4.2 Options.
F FL/4 L B
Straight

C
t Straigh

uright F/2 (a) Force F/2 L/2 L/2 uleft uFD

(b) Bending moment

(c) Deformed shape

Figure5.4.3 Vertical force on portal frame.

B C RL RL B C

R A (a) Force D

Almost straight

(b) Bending moment

BL

RL3 3EI

CL uRD

(c) Deformed shape

Figure5.4.4 Horizontal force on portal frame.

Step 4: The condition we must satisfy is that the horizontal displacement at reaction D must be zero under the influence of the vertical load F and the horizontal load R. We determine the horizontal reaction at D (which must be equal to the horizontal reaction at A) from uFD = uRD (5.4.1)

6861X_C005.indd 265

2/7/08 [Link] PM

266

Understanding Structures: An Introduction to Structural Analysis

We shall go through the previously described process in detail in the following paragraphs. Step 1: If the horizontal restraint is released at support B, the portal frame has no horizontal reaction. As we have observed earlier (Section 5.3), the beam responds as a simply supported beam. Given that the slope is zero at midspan (deduced from symmetry), the rotation at each end is

B = C =

FL2 16 EI

(5.4.2)

The columns are not subjected to bending moment. They do not bend. Therefore, the contributions of the column rotations to the horizontal deflections are uleft = B L and uright = C L (because the column height is equal to the beam span L). The total displacement is uFD = uleft + uright = B L + C L = FL3 8EI (5.4.3)

Step 2: We apply a horizontal force R at support B. The resulting bending-moment distribution is shown in Figure5.4.4b. We note that the bending deformation of the column on the right (with an extremely large bending stiffness EI ) is negligible. The slopes at the two ends of the beam are the same:

B = C =

RL2 2 EI

(5.4.4)

Therefore, the total horizontal displacement is the sum of the contribution of the column on the left uleft = B L + RL3 5 RL3 = 3EI 6 EI (5.4.5)

and that of the very stiff column on the right resulting in uRD = uleft + uright = 4 RL3 3EI (5.4.7) uright = C L = RL3 2 EI (5.4.6)

To get the proper horizontal reaction at B, we equate the displacements FL3 4 RL3 = 8EI 3EI (5.4.8)

6861X_C005.indd 266

2/7/08 [Link] PM

Frames

267

leading to R= 3 F 32 (5.4.9)

Now we have the horizontal reaction for the indeterminate frame in terms of the force F. Because we know the reactions (or external forces), we determine the bending moment at the top of the column (which is equal to the moment at the end of the beam) using statics M = RL = 3 FL 32 (5.4.10)

We conclude that solution 5.4.2b is the correct diagram. The process can be repeated in GOYA-P as follows: 1. Make the vertical force zero and apply a horizontal force of 3N at the roller support (support B). 2. Click the Setting button and increase E for the right column to at least 100 times the default value. Note that the horizontal displacement of the roller has now been reduced to 4/5 of that in the previous stage. 3. Apply a vertical force of 32 N at midspan of the beam. You will find that Equation 5.4.9 is satisfied and that the horizontal displacement of the roller support is zero. What would happen if you increase EI for both columns equally to an extremely large value? In this case, the displacement of the roller caused by the horizontal force R (Figure5.4.4a) is uRD = B L + C L = RL3 EI (5.4.11)

To get the proper horizontal reaction at D, we equate this displacement to that caused by the vertical force (Equation 5.4.3). RL3 FL3 = EI 8EI

which leads to

(5.4.12)

R=

1 F 8

(5.4.13)

Therefore, the bending moment is as shown in Figure5.4.2c. The deflected shape is shown in Figure5.4.5. Note that the columns do not deflect. The deflected shape and bending moment of the beam are, therefore, the same as those of a beam with both ends fixed as shown in Figure5.4.6.

6861X_C005.indd 267

2/7/08 [Link] PM

268

Understanding Structures: An Introduction to Structural Analysis

F =0 Rigid column =0 Rigid column R = F/8

Figure5.4.5 Two rigid columns.

Example 5.4.1
Construct the bending-moment diagram of the structure shown in Figure5.4.7a and calculate the horizontal displacement of the beam, ux.

Solution
Because the beam is rigid and does not deform, the top of each column cannot rotate (q = 0) (Figure5.4.7b). Therefore, the columns are deformed as if they are cantilever beams of length L with their free ends at the supports. Because the stiffness (EI ) and the lateral displacement (ux) of the two columns are the same, the shear forces in the two columns are the same. We conclude that the shear force in each column is F/2 and the bending moment diagram is as shown in Figure5.4.7c. Note that the deflection of a cantilever beam with a load of F/2 is ux = ( F / 2 ) L3 FL3 = 3EI 6 EI (5.4.13)

This is the lateral displacement of the beam shown in Figure5.4.7b. In GOYA-P, modify Youngs modulus of the columns to 178 N/mm 2 so that EI = 106 N.mm2, and make Youngs modulus of the beam very large (at least 100 times the default value). Make F = 6 N. The lateral displacement of the beam indicated on the screen should be ux = 6 100 3 / (6 10 6 ) = 1 mm.

Example 5.4.2
Draw the bending-moment diagram of the structure shown in Figure5.4.8a and calculate the horizontal displacement of the beam, ux.

L FL/8 FL/8 FL/8

Figure5.4.6 Beam of which ends are fixed.

6861X_C005.indd 268

2/7/08 [Link] PM

Frames
F EI Rigid beam =0 L =0 FL/2 FL/2

269

EI

ux

(a) Question

(b) Deformation

(c) Bending moment

Figure5.4.7 A symmetric frame with a rigid beam.

Solution
Because the beam is rigid, the columns may be treated as cantilever beams with their fixed ends at the beam level. Let R1 and R2 denote the reactions of the supports as shown in Figure5.4.8b and note that the deformation (ux) of each column is the same ux = R1 L3 R2 L3 = 3EI 3 ( 2 EI ) (5.4.14)

or R2 = 2 R1 . Noting that F = R1 + R2 , we obtain R1 = F / 3 and R2 = 2 F / 3. The shear force in the right column is twice that in the left column because the bending stiffness of the right column is twice that of the left column. If we substitute R1 = F / 3 into Equation 5.4.14, we get ux = ( F / 3) L3 FL3 = 3EI 9 EI

(5.4.15)

You should check this result using GOYA-P.

Example 5.4.3
Construct the bending-moment diagram of the structure with distributed load shown in Figure5.4.9.

Rigid beam

=0 L

=0

FL/3 2FL/3

EI

2EI

ux R1

ux R2

(a) Question

(b) Deformation and reactions

(c) Bending moment

Figure5.4.8 An unsymmetrical frame with a rigid beam.

6861X_C005.indd 269

2/7/08 [Link] PM

270

Understanding Structures: An Introduction to Structural Analysis

w

Rigid column

Figure5.4.9 A frame with distributed load.

Solution
We can solve this example replacing one of the pin supports by a roller support as we did before. However, we note that the columns do not deform. That observation leads to a faster solution. Because the ends of the beam do not rotate, the beam may be considered a beam with fixed ends (Figure5.4.10a). Note that the vertical reaction at each support is wL/2 because of symmetry. Integrating dV/dx = w, we obtain Figure5.4.10b or V= wL - wx 2 (5.4.16)

The bending moment is obtained integrating dM/dx = V: M = M0 + wL w x - x2 2 2 (5.4.17)

where M0 denotes the bending moment at x = 0 (the left end). We substitute this equation into d 2 v / dx 2 = M / EI and integrate it, noting that the inclination at x = 0 (the left end)

wL/2 wL/2 wL/2 x (b) Shear force

Rigid column wL2 24

wL/2

L (a) Beam

wL2 M0 = 24

x (c) Bending moment

wL2 8

wL2 24 wL2 12

wL2 8

wL2 8

wL2 8

(d) Answer

(e) Flexible column

(f) Same stiness

Figure5.4.10 Solution.

6861X_C005.indd 270

2/7/08 [Link] PM

Frames is zero. We obtain dv 1 wL 2 w 3 = x - x M0 x + dx EI 4 6

271

(5.4.18)

Because the deflected shape is symmetrical about midspan, the slope at x = L/2 (midspan) must be also zero. Therefore, wL2 12

M0 = -

(5.4.19)

Figure5.4.10c shows the bending-moment diagram. If the columns are much more flexible than the beam (contrary to Figure5.4.9), the bending moments at the ends of the beam M0 approach zero and we obtain the bending-moment diagram as shown in Figure5.4.10e that is equivalent to that of a simple beam. If the stiffness of the beam is similar to those of the columns, the bending-moment diagram is between those shown in Figure5.4.10d and e, as shown in Figure5.4.10f. In Figure5.4.10df, you should note that the difference between the moment at the ends and that at midspan is always wL2 / 8 . This agrees with the bending moment at midspan of a simple beam under a uniform load of w (see Section 3.2, Chapter 3). Such an agreement is also observed for the case of a concentrated load. The bending moment in an equivalent simple-beam is called static moment. Figure5.4.11a shows a frame subjected to a uniform load of w and a horizontal load of FH. If you call the positive moment at midspan M1 and the two negative moments at the ends of the beam M0 and M2, you will find M1 M 0 + M 2 wL2 = 2 8 or M1 + | M 0 + M 2 | wL2 = 2 8

because the moment distribution in Figure5.4.11a can be decomposed into those in Figure5.4.11b,c. For a beam with a concentrated load at the middle (Figure5.4.12), M1 M 0 + M 2 FL = 2 4 or M1 + | M 0 + M 2 | FL = 2 4

w M0 M2

w WL2 8 L M0 + M2 2

FH

M1

=
L/2

M0

M2

L/2

(a) Frame

(b) Static moment

(c) End moments

Figure5.4.11 Frame subjected to distributed load and horizontal load.

6861X_C005.indd 271

2/7/08 [Link] PM

272

Understanding Structures: An Introduction to Structural Analysis

F F M2 M1 FL 4 L (b) Static moment M0 M0 + M2 2 M2

FH

M0

=
L/2

L/2

(c) End moments

(a) Frame

Figure5.4.12 Frame subjected to vertical and horizontal loads.

The left-hand term, for any loading distribution, is equal to the static moment or the moment at midspan of a simply supported beam for the same loading distribution. This result is useful to remember because it can be used to check the reliability of solutions for statically indeterminate beams.

Exercise 1
Choose any one-digit number i to determine the length of the beam in Intext Figure5.1. Fill out the table using GOYA-P while keeping Youngs modulus of the other members as the default value (E = 100 N/mm2). Also, sketch the deflected shape of the frame for each case.
F = 100 N MA E 70 + i MB 70 + i MA 100 R E of Left Column 500 N/mm2 100 N/mm2 20 N/mm2 R MA MB

Exercise 2
Choose any one-digit number i to determine the length of the beam in Intext Figure5.2. Fill out the table using GOYA-P but keep Youngs modulus of the other members equal to the default value (E = 100 N/mm2). Sketch the deflected shape of the frame for each case.
F = 20 N MA E RA 100 RB 140 + 2i E of Left Column 500 N/mm2 100 N/mm2 20 N/mm2 RA RB MA MB

MB

6861X_C005.indd 272

2/7/08 [Link] PM

Frames

273

Beam Column

Beam C Column

B Foundation Foundation Beam

Column Foundation Beam D Foundation

F Foundation

Figure5.5.1 Structural concrete building.

w C Column Mc = M b wL/2 (b) Vertical load wL/2 (c) Bending moment (d) Equilibrium wL2/8

A Column

L Beam

Mb

Foundation B D Beam Foundation Foundation (a) Model

Figure5.5.2 One-story one-span frame under vertical load.

5.5 Multistory Frame Goals of Structural Design

In the First century b.c., a Roman builder, Vitruvius, wrote that a building should be strong, that it should function properly, and that it should please. We translate those maxims into the engineers current work environment as follows: 1. Safety: A building structure must have the strength to resist permanent and transient loads with appropriate factors of safety. It must not endanger the lives of its occupants.
 The main source of permanent loads is gravity. Transient loads are caused by such effects as wind and earthquakes.

6861X_C005.indd 273

2/7/08 [Link] PM

274

Understanding Structures: An Introduction to Structural Analysis

2. Serviceability: The deformations of a building structure, in the short or long term, must be controlled so that its functionality is not impaired. The engineer needs to consider not only stresses and strains but also durability. 3. Beauty: A building structure should be planned and proportioned to please the eye. Except for transportation and industrial structures, this task is seldom within the purview of the structural designer. Nevertheless, the structural designer must always consider this important aspect of the design challenge. Structures, as well as being right, need to look right. Structural design is essentially an art. In the practice of engineering, one rarely obtains a unique answer from a crisp design procedure. Even in the simplest of cases, structural design requires trial and error and involves a two-step process: Step 1: Determine the framing and dimensions of the structure. This step requires not only theoretical knowledge but also the ability to observe and weigh relevant evidence. An engineer should never accept an answer from a calculation that the engineer could not have guessed at to within +/ 20%. Before the design calculations are initiated, the engineer should be able to make good estimates of the sizes of the structural elements. The student may well ask, How does one do that if one has never designed a structure? The student can compensate for his or her lack of experience by looking at structures critically and developing a sensitivity to ratios such as beam span to depth and floor to column-section area. Step 2: Check if the selected framing and dimensions satisfy the requirement of serviceability and safety. The most effective way to satisfy this requirement would be to build the structure and observe its behavior under service or test loads. For most civil-engineering structures, such an approach would be prohibitively expensive. So, we test the structure by calculation, and if the calculation requires much labor, we use computers. Numerical models of the structure, based on methods we have been discussing and have implemented using powerful computers, give us seven-league boots to help decide whether the sizes we have selected are satisfactory. However, in doing that, we must be very careful to make certain that the conclusions fit into our sense of proportion. Every time we obtain an answer through a sophisticated and complicated procedure, it is our professional responsibility to try to obtain the same answer, if approximately, through simple thinking. The simplest way to think is to know the proportions of the answer from previous observation. That is why experience is important in engineering.

Illustrative Design Examples

The designing of a structure requires extensive work and great care for detail. We cannot undertake that in the classroom. The following examples will illustrate certain basic points of the dimensioning process for a frame. In that respect, they are not

6861X_C005.indd 274

2/7/08 [Link] PM

Frames

275

design examples but illustrations. Our goal is to understand how the internal forces, which will subsequently be used in determining the properties of the elements, are determined. Most modern buildings in developed countries are made of concrete, steel, or timber. Figure5.5.1 shows a very simple example of the framing of a structural reinforced concrete building. The gravity load on the floor (not shown) is transmitted to beams. The beams transmit the load to the columns. The columns rest on the foundations, and the foundations rest on the ground. The sequence of the gravity load going from the slab to the girders to the columns to the foundations is called the load path. It is important in design to have a good grasp of the load path. We can model a portion of this building (frame ABCD in Figure5.5.2a). The foundations are modeled as pin supports because the ground under the foundations is usually soft and its resistance to rotation is much smaller than that of the foundation beams. The gravity load on the slab is transferred to the beam almost uniformly as shown in Figure5.5.2b. This load induces the bending moments shown in Figure5.5.2c. The moment diagrams in Figure5.5.2c are similar to those in Figure5.4.10f, except that the bending moment at the bottom of each column is not zero because of the resistance of the foundation beam. Note that bending moments are in equilibrium at each joint (or node) as shown in Figure5.5.2d. The other frames (such as CDEF) can be modeled similarly. As a result, each column is bent in two directions. Figure5.5.3a shows a symmetrical frame with two spans subjected to uniform vertical load. Because of the symmetry, the column in the middle does not bend. It is not subjected to moment because the end moments of the beams balance one another (Figure5.5.3b). The bending moment in the foundation beams is negative near the middle because the foundations are assumed to be pin supports that do not move. (In reality, the foundations move as the ground deforms, but the deformation of the ground is usually very small.) Figure5.5.4a shows a two-story frame with a span subjected to uniform vertical load. Figure5.5.4b shows the bending-moment diagram. Note that at each end node of the beam CD (Figure5.5.4c), equilibrium of moment requires M b = M c1 + M c 2 . The bending moment in the beam at the node is about twice that of each column.
L w L L/2 wL2/8 L/2 L/2 wL2/8 L/2

Appr. wL/2

Appr. wL (a) Vertical load

Appr. wL/2 (b) Bending moment

Figure5.5.3 One-story two-bay frame under vertical load.

6861X_C005.indd 275

2/7/08 [Link] PM

276
L

Understanding Structures: An Introduction to Structural Analysis

w A

B w

wL2/8

Mc2 Mb

wL2/8

E wL (a) Vertical load

G wL (b) Bending moment

Mc1

(c) Equilibrium

Figure5.5.4 Two-story one-bay frame under vertical load.

Figure5.5.5a shows a one-story one-bay frame subjected to a horizontal load. If the stiffnesses of the columns are the same, the shear forces in the columns are also the same (V = F/2) as discussed in Section 5.4. The vertical reactions, (H/L)F, are determined by equilibrium conditions and are equal to those of a portal frame without a foundation beam (Figure5.4.7). Because foundation beam CD is designed to be stiffer than beam AB, the bending moment at the bottom of the column is larger than the moment at the top of the column (Figure5.5.5b). Note again that bending moments are in equilibrium at each node (Figure5.5.5c). Figure5.5.6a shows a one-story two-bay frame subjected to a horizontal load. The bending-moment diagram (Figure5.5.6b) is similar to that of Figure5.5.5b, except that the column in the center resists a larger bending moment than the other columns because the column in the center is connected to two beams as shown in Figure5.5.6c. Equilibrium requires M c = M b1 + M b 2 . Figure5.5.7a shows a two-story one-bay frame subjected to horizontal load. The bending moment diagram is shown in Figure5.5.7b. As required by equilibrium of moments shown in Figure5.5.7c ( M b = M c1 + M c 2 ), the bending moment in beam CD is larger than that in roof beam AB. If the flexural stiffness of the left column is

F A

L B H C F/2 (H/L)F (a) Horizontal load D F/2 (H/L)F (b) Bending moment (c) Equilibrium Mc = Mb Mb

Figure5.5.5 One-story one-bay frame under horizontal load.

6861X_C005.indd 276

2/7/08 [Link] PM

Frames
F

277

(a) Horizontal load

Mb1 Mc (c) Equilibrium

Mb2

(b) Bending moment

Figure5.5.6 One-story two-bay frame under horizontal load.

the same as that of the right column, the shear forces of the columns in the second story are V = F2 / 2 as shown in Figure5.5.7d. On the other hand, the shear forces of the columns in the first story are V = ( F1 + F2 ) / 2 (Figure5.5.7e). Example 5.5.1
A two-story frame is subjected to horizontal loads, and the columns resist bending moments as shown in Figure5.5.8. 1. Compute the shear forces in the beams. 2. Compute the applied horizontal forces and the reactions. 3. Compute the axial forces in the columns.

Solution
1. If we consider equilibrium at each node, we obtain the bending-moment diagram for each beam as shown in Figure5.5.9. Because the slope of the bendingmoment diagram is equal to the shear force, we conclude that the shear force in

F2

B Mc2 Mb

F2 F2/2 F2 Mc1 (b) Bending moment (c) Equilibrium F1 (F1 + F2)/2 (F1 + F2)/2 (e) Cut at 1st story F2/2

F1

(d) Cut at 2nd story

(a) Horizontal load

Figure5.5.7 Two-story one-bay frame under horizontal load.

6861X_C005.indd 277

2/7/08 [Link] PM

278

Understanding Structures: An Introduction to Structural Analysis

100 kN-m A 100 kN-m

B 4m

120 kN-m

C 80 kN-m

120 kN-m D

80 kN-m 4m

160 kN-m 8m

160 kN-m

Figure5.5.8 Bending moment of columns.

each beam is 2 100 = 25 kN 8 2 200 = 50 kN 8 2 160 = 40 kN 8

Beam AB

V= V= V=

Beam CD

Beam EF

2. If we cut the frame at the second story and at midspan of roof beam AB as shown in Figure5.5.10a, we can conclude that the axial force in column CE is P = 25 kN in tension. If we cut the frame at the first story as shown in Figure5.5.10b, we can conclude that the axial force in column CE is P = 25 + 50 = 75 kN in tension. The axial forces in columns BD and DF have the same magnitude as columns AC and CE, respectively, but they are in compression.

100 kN-m A 200 kN-m C 160 kN-m E 8m

100 kN-m 4m D 200 kN-m F 160 kN-m 4m

Figure5.5.9 Bending moment of beams.

6861X_C005.indd 278

2/7/08 [Link] PM

Frames
100 kN-m A 25 kN 4m (a) Cut at 2nd story 100 kN-m A 200 kN-m C 75 kN (b) Cut at 1st story 100 kN-m A 200 kN-m C 160 kN-m E 115 kN

279

25 kN

25 kN

25 kN

50 kN

50 kN

40 kN

(c) Cut all beams

Figure5.5.10 Cut the frame.

3. If we cut all the beams as shown in Figure5.5.10b, we can determine that the vertical reaction at support E is R = 25 + 50 + 40 = 115 kN The vertical reaction at support F has the same magnitude but is in the opposite direction (Figure5.5.11). We can compute the shear forces in the columns using the slope of the bending-moment diagram shown in Figure5.5.8:

Columns AC and BD

V= V=

100 + 80 = 45 kN 4

Columns CE and DF

120 + 160 = 70 kN 4

Recalling Figs. 5.5.7d,e, we conclude that the horizontal force on the roof is 45 2 = 90 kN. On the second floor, it is 70 2 90 = 50 kN (Figure5.5.10). The magnitudes of the horizontal reactions are same as the shear forces in the columns in the first story.

90 kN

50 kN

E 115 kN 70 kN

70 kN 115 kN

Figure5.5.11 Forces and reactions.

6861X_C005.indd 279

2/7/08 [Link] PM

280

Understanding Structures: An Introduction to Structural Analysis

d

F EI1

V1H/2 EI2 H V1H/2

V1 v H/2 EI1

(a) Deformation

(b) Moment

(c) Cantilever beam

Figure5.5.12 One-story frame with rigid beams. Often, we use computers to calculate deformations of frames subjected to horizontal forces. If we assume that the flexural deformation of beams and the axial deformation of columns are negligible, we can approximate the deformation of frames without computers. Consider the frame shown in Figure5.5.12a. Because the beams are assumed to be rigid, the bending moment at midspan of each column is zero as shown in Figure5.5.12b. Figure5.5.12c shows an equivalent cantilever beam with the flexural rigidity of the left column (EI1) subjected to the shear force in the left column (V1). Recalling section 2.8 in Chapter 2, we can calculate the deflection of the beam as v= ( H / 2 )3 V1 3EI1 (5.5.1)

Note that the deflection of the frame is twice that of the equivalent beam ( = 2 v ), which leads to V1 = 12 EI1 H3 (5.5.2)

We can develop a similar equation for the right column. V2 = 12 EI 2 H3 (5.5.3)

Because the sum of the shear forces in the columns is equal to the external horizontal force ( F = V1 + V2 ), we get F= 12 ( EI1 + EI 2 ) H3 or

H3 F 12( EI1 + EI 2 )

(5.5.4)

Note that the deformation d determined by Equation 5.5.4 is smaller than the actual deformation. Figure5.5.13 shows a two-story frame subjected to horizontal loads. The deformation of each story (d1 and d 2) is called story drift. Using the procedure similar to that for one-story frame, we get

2 =

H3 F2 12( EI 21 + EI 22 )

or

1 =

H3 ( F1 + F2 ) 12( EI11 + EI12 )

(5.5.5)

Note that the horizontal displacement of the roof is equal to the sum of the story drifts (d1 + d 2).

6861X_C005.indd 280

2/7/08 [Link] PM

Frames
F2 EI21 1 2 EI22 H

281

F1

EI11

EI12

Figure5.5.13 Two-story frame.

150 kN 2 1 4m

100 kN

50 kN

4m

E = 20 kN/mm2
8m

4m

Figure5.5.14 Three-story frame.

Exercise Using GOYA-M

Assume a three-story frame shown in Figure5.5.14, where the cross sections of the columns in the second and third story measure 500 500 mm, and Youngs modulus is 20 kN/mm2. You may neglect the flexural deformation of beams and the axial deformation of columns. 1. Construct the bending-moment diagram of beams and columns. 2. Estimate the drift of the second story. 3. Determine the cross-sectional dimensions of the columns in the first story so that the drift of the first story is the same as that of the second story. 4. Check your results using GOYA-M.

5.6 Three-hinged Frame

Suppose we construct a large portal frame in October (Figure5.6.1a). In winter, the roof tends to shorten as shown in Figure5.6.1b. As a result, the beam is bent down and subjected to a tensile axial force (Figure5.6.1d,e). In summer, the roof is heated and tends to elongate as shown in Figure5.6.1c. As a result, the beam is bent up and

6861X_C005.indd 281

2/7/08 [Link] PM

282

Understanding Structures: An Introduction to Structural Analysis

(a) Fall Tension

(b) Winter

(c) Summer

(d) Axial force in winter

(e) Bending moment in winter

Figure5.6.1 A portal frame.

subjected to a compressive axial force. If we build a hinge into the beam as shown in Figure5.6.2a, such internal forces do not develop because the frame can deform freely as shown in Figs. 5.6.2b,c. Such a structure is called a three-hinged frame (one hinge in the beam and two at the supports). The three-hinged frame in Figure5.6.2 is statically determinate. Start GOYA-H and apply a vertical load at midspan. You will get the results shown in Figure5.6.3. The shear-force and bending-moment diagrams are obtained considering equilibrium alone. Figure5.6.4a shows the reactions at the supports. We assume that the span length is 2L, the column height is L, and the frame is symmetrical. Because of symmetry, the magnitude of each vertical reaction (RAy and RCy) is F/2. Figure5.6.4b shows the free-body diagram for the left part of the frame. Moment equilibrium at support A leads to RBx = F/2. Moment equilibrium at hinge B leads to RA x = F/2. The shear-force and bending-moment diagrams are shown in Figure5.6.4c,d. What if we move the load to the right as shown in Figure5.6.5a? Moment equilibrium at support A leads to RCx = 3F/4. Moment equilibrium at support C leads to RAy = F/4. Figure5.6.5b shows the free-body diagram for the left half of the frame. Force equilibrium in the vertical direction leads to RBy = F/4. Moment equilibrium at hinge B leads to RA x = F/4. Moment equilibrium at support A leads to RBx = F/4. The shear-force and bending-moment diagrams are shown in Figure5.6.5c,d. Figure5.6.6a shows the free-body diagram of the right half of the frame. Moment equilibrium at hinge B leads to RCx = F/4. The shear-force and

(a) Fall

(b) Winter

(c) Summer

Figure5.6.2 Three-hinged frame.

6861X_C005.indd 282

2/7/08 [Link] PM

Frames

283

+ _

(a) Deformation

(d) Bending moment

(c) Shear force

Figure5.6.3 Vertical load at midspan.

F B

L L B RAx F/2 A

F/2 RBx

+F/2 F/2

FL/2

RAx RAy

RCx RCy

(a) Reactions

(b) Free body

(c) Shear force

(d) Bending moment

Figure5.6.4 Equilibrium.

3L/2 B A

L/2

L RBy B RAx A RBx +F/4 F/4 (c) Shear force (d) Bending moment FL/4

RAx

C RCx RCy

RAy (a) Reactions

F/4 (b) Free body

Figure5.6.5 Move the load to the right.

L/2 L/2 RBx B RBy C F +F/4 L +F/4 3F/4 FL/8 FL/4 RCx 3F/4 (a) Free body (b) Shear force (c) Bending moment

Figure5.6.6 Equilibrium of right half.

6861X_C005.indd 283

2/7/08 [Link] PM

284
L/2 L/2 F B RAx RAy A

Understanding Structures: An Introduction to Structural Analysis

L

L/2 L

RBy RBx B

+F/2 F/4

FL/4

RCx RCy

RAx

(a) Reactions

F/2 (b) Free body

(c) Shear force

(d) Bending moment

Figure5.6.7 Move the hinge to the right.

bending-moment diagrams are in Figure5.6.6b,c. As we move the location of the vertical load to the right, the bending moment in the beam decreases as does the deformation of the frame. If we move the location of the load so that it coincides with the column axis, both the bending moment and the beam deflection become zero. Move the location of the load back to midspan and move the location of the hinge to the left as shown in Figure5.6.7a. This move also changes the shear-force and bending-moment diagrams. Moment equilibrium of the structure in Figure5.6.7a at support A or C leads to RAy = RCy = F/2 as they were before we moved hinge B. Figure5.6.7b shows the free-body diagram for the left part of the frame. Force equilibrium in the vertical direction leads to RBy = F/2. Moment equilibrium at hinge B leads to RA x = F/4. Moment equilibrium at support A leads to RBx = F/4. We obtain the shear-force and bending-moment diagrams in Figure5.6.7c,d. Figure5.6.8a shows the free-body diagram for the portion of the frame to the right of the hinge. Moment equilibrium at hinge B leads to RCx = F/4. Thus, we get the diagrams in Figure5.6.8b,c. If we move the location of the hinge to the left end, the bending moments in the columns diminish. The beam is bent in the same way as a cantilever beam. Because the flexural deformation in the columns is zero, the beam displaces to the left. We can also notice similar displacement in Figure5.6.7a. Using GOYA-H, apply a horizontal load Fx on the frame as shown in Figure5.6.9a. Moment equilibrium at support A or C leads to RAy = RCy = Fx/2. Figure5.6.9b shows the free-body diagram for the left part of the frame. Force equilibrium in the vertical direction leads to RBy = Fx/2. Moment equilibrium at hinge B leads to RA x = Fx/4. Force equilibrium in the horizontal direction leads to RBx = Fx RA x = 3Fx/4. These results

L/2 RBx RBy

+F/2 L

F/2 +F/4

FL/4 FL/4

RCx F/2

(a) Free body

(b) Shear force

(c) Bending moment

Figure5.6.8 Equilibrium of right part.

6861X_C005.indd 284

2/7/08 [Link] PM

Frames
L/2 3L/2 B D L/2 Fx D B L A F/2 RBy

285

Fx

E C RCx RCy

RBx

Fx/2 +Fx/4

FxL/4

RAy

RAx

RAx

(a) Deformation

(b) Free body

(c) Shear force

(d) Bending moment

Figure5.6.9 Horizontal load.

help us obtain the diagrams in Figure5.6.9c,d. Figure5.6.10a shows the free-body diagram for the right part of the frame. Force equilibrium in the horizontal direction leads to RCx = RBx = 3Fx/4 to obtain the diagrams in Figure5.6.10b,c. Note that the bending moment in column CE is larger than that in column AD. If we move the location of the hinge to the left end, the bending moments of column AD are reduced. Example 5.6.1
Figure5.6.11 shows a three-hinged frame subjected to horizontal and vertical loads, Fx and Fy. Assume that Fy = 30 kN and the bending moment at node E is zero. Compute the required magnitude of Fx and construct the moment diagram.

Solution
Using the results obtained in Figures5.6.7d and 5.6.8c, we determine the moment diagram caused by Fy = 30 kN (Figure5.6.12a). Using the results from Figures5.6.9d and 5.6.10c, we determine the moment diagram caused by Fx (Figure5.6.12b). Because the bending moment at node E is zero, we have the condition 60 + 6Fx = 0 or Fx = 10 kN. Figure5.6.12c shows the moment diagram. Check the result using GOYA-H. You will find that the horizontal displacement of the beam is zero. We can obtain the same solution without using the superposition technique. Figure5.6.13a shows the possible reactions. Because the bending moment at node E was set to zero by choosing Fx, we conclude that the shear force in column EC and the reaction RCx are zero. Force equilibrium in the horizontal direction leads to RBx = Fx -RCx = Fx. Moment equilibrium at support C leads to
3L/2 E C (a) Free body RCx Fx/2 (b) Shear force (c) Bending moment L

8Fx + 16 RAy - 120 = 0

or

RAy = 15 -

Fx 2

RBx

B RBy

Fx/2 +3Fx/4

3FxL/4

Figure5.6.10 Equilibrium of right part.

6861X_C005.indd 285

2/7/08 [Link] PM

286

Understanding Structures: An Introduction to Structural Analysis

4m 4m D B A 8m Fy = 30 kN E C 8m

Fx

AU: Caption?

Figure5.6.11
Fy = 30 kN 60 kN-m 60 kN-m 60 kN-m 15 kN 15 kN (a) Vertical load 15 kN 15 kN Fx/2 Fx 30 kN 2Fx Fx/4 6Fx 3Fx/4 80 kN-m 10 kN 10 kN 20 kN 10 kN 80 kN-m

2Fx

Fx/2

(b) Horizontal load

(c) Solution

Figure5.6.12 Solution using the previous results.

4m 4m Fx

8m Fy = 30 kN E 8m Fx

4m RBy B Fx 15 Fx/2 (b) Free body 10 kN

4m 80 kN-m 20 kN

A RAy

RAx

RCx RCy

10 kN 20 kN (c) Bending moment

(a) Reactions

Figure5.6.13 Another solution.

Figure5.6.13b shows the free-body-diagram for the left part of the frame. Moment equilibrium at hinge B leads to 8Fx + 4 30 - Fx =0 2 or Fx = -10 kN

Force equilibrium in the horizontal direction leads to RBy = 30 - Fx = 20 kN 2

Thus, we obtain the moment diagram shown in Figure5.6.13c. The moment diagram for the right part of the frame is obtained similarly.

6861X_C005.indd 286

2/7/08 [Link] PM

Frames

287

15 kN

2m 4m

4m

4m
AU: Caption?

Figure5.6.14

Example 5.6.2
Construct the moment diagram for the frame shown in Figure5.6.14.

Solution
Figure5.6.15a shows possible reactions. Moment equilibrium at support A or E leads to RAy = REy = 7.5 kN. Figure5.6.15b shows the free-body diagram for the right part of the frame. Moment equilibrium at hinge C leads to 7.5 4 - REx 6 = 0 or REx = 5 kN Thus, we obtain the moment diagram shown in Figure5.6.15c. The moment diagram for the left part is obtained similarly. Figure5.6.15d shows the moment diagram for the entire frame.

Exercise
Use a one-digit number i to define the vertical load Fy shown in Figure5.6.16. Assume that the beams and columns can resist bending moment up to 500 kN-m. Compute the maximum (positive) and minimum (negative) horizontal loads that the frame can resist. Check the results using GOYA-H. Sketch the moment diagram and the deflected shape of the frame.
RCx 15 kN B A RAy RAx 8m (a) Reactions C D E RDx RDy C 4m RCy A D 4m RDx 7.5 kN 4m 5 kN 6m 7.5 kN 2m 20 kN-m 5 kN 4m

7.5 kN

(b) Free body

(c) Bending moment

15 kN

40 kN-m

20 kN-m

10 kN

5 kN

7.5 kN 7.5 kN (c) Bending moment

Figure5.6.15 Solution.

6861X_C005.indd 287

2/7/08 [Link] PM

288

Understanding Structures: An Introduction to Structural Analysis

5m Fx 8m Fy = 30 + i kN 5m

9m

Au Caption

Figure5.6.16

5.7 Problems
5.1 Figure5.7.1 shows a bent subjected to three forces and an unknown couple, M. Determine the magnitude of the couple M to make the moment at the fixed-base A zero. Check to see if your answer is the same as one of the five options listed in Table5.7.1. Positive sign refers to counterclockwise moment as shown in Figure5.7.1.
M B 4m A 6m 100 kN D 100 kN

C 100 kN

Figure5.7.1 Bent.

Table 5.7.1
M 1. 2. 3. 4. 5. + 1,000 kN-m + 600 kN-m + 400 kN-m - 600 kN-m - 1,000 kN-m

5.2 Figure5.7.2 shows a bent subjected to horizontal and vertical forces, Fx and Fy. The bending moment at point A is to be zero. Select the correct ratio of Fx to Fy among the five options listed in Table5.7.2 to make the moment at point A zero.

6861X_C005.indd 288

2/7/08 [Link] PM

Frames
Fy Fx

289

A 2L L L

Figure5.7.2 Bent.

Table 5.7.2
Fy: FX 1. 2. 3. 4. 5. 1:0 1:1 1:2 3:3 3:4

5.3 Figure5.7.3 shows a portal subjected to horizontal and vertical forces, F. Select the correct bending-moment diagram among the options shown in Figure5.7.4ae.
3L F F 2L 2L 3L

Figure5.7.3 Portal.
2FL 1 FL 2 4 FL 3 3 FL 2 4 FL 3 4FL

7 FL 2

(a) FL 1 FL 2 (d) FL FL

(b) FL

(c)

5 FL 2 (e)

Figure5.7.4 Options.

6861X_C005.indd 289

2/7/08 [Link] PM

290

Understanding Structures: An Introduction to Structural Analysis

5.4 Figure5.7.5 shows a portal frame supported on two hinges and subjected to a vertical load. Assume that the stiffness term, EI, for the beam and the columns is the same. Select the correct bending-moment diagram among the options shown in Figure5.7.6ae. (Hint: Construct a shearforce diagram that corresponds to each bending-moment diagram. Then, think!)

15 m

15 m 10 kN 10 m 10 m

Figure5.7.5 Portal.

30 kN-m

46 kN-m

30 kN-m

28 kN-m

40 kN-m

14 kN-m

(a) 34 kN-m 50 kNm 17 kN-m 50 kN-m

(b) 38 kN-m 25 kN-m

(c)

(d)

Figure5.7.6 Options.

AU: 5.7.4 corrected as 5.7.7. OK?

5.5 Figure5.7.7 shows bending-moment diagrams for columns in a two-story frame subjected to lateral loads. Select the incorrect statement among the following options. 1. The shear force in beam AB is VAB = 35 kN. 2. The shear force in beam CD is VCD = 62.5 kN. 3. The axial force in column DG is PDG = 97.5 kN. 4. The vertical reaction at support G is RGy = 140 kN. 5. The horizontal force on node D is FD = 160 kN.

6861X_C005.indd 290

2/7/08 [Link] PM

Frames
140 kN . m A 150 kN . m C 100 kN . m E 170 kN . m 170 kN . m 8m 140 kN . m B VCD D 100 kN . m 150 kN . m FD PDG G RGy 4m 4m

291

VAB

Figure5.7.7 Two-story frame.

5.6. Figure5.7.8 shows a two-story frame with horizontal loads. Assume that the flexural deformation of the beams and the axial deformation of the columns are negligible. Assume also that the stiffness, EI, of the columns in the first story is twice that in the second story. Select the correct ratio of the drift in the first story, d1, to that in the second story, d 2, from the options listed in Table5.7.2.
2F EI 1 2 EI L

2EI

2EI

Figure5.7.8 Two-story frame.

5.7. Figure5.7.9 shows a three-hinged frame subjected to horizontal and vertical loads. Select the correct bending moment at node D, MD, from the options listed in Table5.7.3.
L 8F B A C 3L

Table 5.7.3
d1: d2 D 2L E 1. 2. 3. 4. 5. 1:1 1:2 1:4 3:2 3:4

4F

Figure5.7.9 Three-hinged frame.

6861X_C005.indd 291

2/7/08 [Link] PM

6861X_C005.indd 292

2/7/08 [Link] PM

1
Au: Please provide missing Captions of figures cited Au: please checkthe numbering of Figures Callout Display maths etc.

Trusses

1.1 What is a truss?

Which of the three methods shown in Figure1.1.1 would you prefer to use if you wanted to snap a chopstick? Probably, you will elect the method shown in Figure1.1.1c. It is difficult to snap the chopstick if you pull or push it. The same rule applies to many things such as a chocolate bar, a cookie, and a plastic stick. Assume that you need to cook a pot of stew in a campground. The structure shown in Figure1.1.2a looks vulnerable because the horizontal bar bends and might break at the center. This type of structural system is called a frame. You will learn why and how the bar bends in Chapter 2. The triangular system shown in Figure1.1.2b is stronger and is called a truss. The reason for the superiority of the truss over the frame in this application can be understood as follows. The force, P, corresponding to the pull of gravity on the pot may be resolved into two components P/ 2 as shown in Figure1.1.3a. Each of these components is collinear with an element of the truss and causes the reaction from the ground shown in Figure1.1.3b. As a result, each element of the truss is pushed as illustrated in Figure1.1.3c or Figure1.1.1b and not bent as illustrated in Figure1.1.1c. The truss is also useful for bridge structures. You may make a bridge simply by laying a log as shown in Figure1.1.4a. Such a bridge is flexible and might break if an elephant crosses it. A truss bridge (Figure1.1.4b), which is made forming triangles, is much stiffer and stronger. Ancient Roman engineers already knew the fact by experience. An Italian architect of the 16th century, Palladio, designed many truss timber bridges, some of which still remain in Venice. After the invention of mass production methods for steel in 19th century England, steel trusses became very popular for bridges. The Eiffel Tower ([Link] built in 1889, is a truss. The Nagoya Dome, an elegant structure, is also a truss (Figure1.1.5). The proposed space structure, Figure1.1.6, is also based on the truss principle. However, any one collection of triangles does not necessarily constitute a safe truss structure. The larger the structure you want to build, the more careful the calculations you should make. In ancient or medieval times, people constructed large structures by experience. That was a difficult and dangerous undertaking. In Ancient Egypt, several pyramids collapsed during construction. The Cathdrale SaintPierre de Beauvais in France experienced two major collapses during its long history of construction: the choir in 1284 and the tower in 1573. Structures accomplished on

 Timoshenko, S. P. 1953. History of Strength of Materials. McGraw-Hill, New York.  Mendelssohn, Kurt. 1974. The Riddle of the Pyramids. Praeger, New York.

6861X_C001.indd 1

2/7/08 [Link] PM

Understanding Structures: An Introduction to Structural Analysis

(a) Pull

(b) Push

(c) Bend

Figure1.1.1 Snap a chopstick.

(a) Frame

(b) Truss

Figure1.1.2 Cook a pot of stew.

P 2 P 2 P P 2 P 2 P 2 Push P 2

2 P

P 2

(a) Resolution of the force

(b) Reactions from the ground

(c) Each element is pushed

Figure1.1.3 Equilibrium.

Danger

Safe

(a) Beam

(b) Truss

Figure1.1.4 Timber bridges.

Figure1.1.5 Truss in Nagoya Dome.

6861X_C001.indd 2

2/7/08 [Link] PM

Trusses

Figure1.1.6 Truss in a space station ([Link]

the basis of experience related to gravity load only were also in danger of collapse if subjected to tornados or earthquakes. After structural mechanics was developed in the 19th century, the situation with respect to understanding structural behavior improved dramatically. Present-day engineers can check whether the designed structure is safe or not before they start construction (though the prediction is not always perfect). Structural mechanics has also reduced waste of materials and workmanship, which had been inevitable in olden times. The investment in constructed facilities includes many components in addition to the structure. In fact, in most applications the cost of the structure is less than one-fourth of the cost of the total investment, and in industrial facilities this ratio may be even less. However, the survival of the entire investment depends on the survival of the structure. That is why the structural engineer has a critical responsibility and needs to be well informed about the behavior of the structure. The first step to develop a fundamental understanding of structural behavior is to understand mechanics, which we shall attempt to do in this book.

1.2 Bar in Tension or Compression

In this section, you will study the behavior of a bar in tension and compression, before learning about a truss made up of many bars (elements.) Access one of the following Web sites to start GOYA-T. [Link] [Link] You will find Figure1.2.1, which shows all types of trusses included in this chapter. Select the upper left figure. The left end of the bar is connected to a wall as shown in Figure1.2.2 so that the end can rotate freely but cannot move away from the pin. Such a connection is called a pin support. The right end of the bar can either rotate as

6861X_C001.indd 3

2/7/08 [Link] PM

Understanding Structures: An Introduction to Structural Analysis

Au: please provide Figure1.2.1 missing figure

GOYA-T window.

Figure1.2.2 Rotation at pin support.

shown in Figure1.2.3a or move in the horizontal direction as shown in Figure1.2.3b but not in the vertical direction. Such a connection is called a roller support. Drag the right end of the bar in the direction of F as shown in Figure1.2.4a. An arrow will appear, and the end will move slightly to the right. The arrow represents the force applied to the end. The vertical component of the force, F sinq, is supported by the floor (Figure1.2.4a) through the rollers. In other words, the floor pushes the bar with a force of F sinq. This force is called a reaction. The horizontal component, F cosq, is transmitted through the bar and supported by the wall. In other words, the wall pulls the bar with a force of F cosq. This force is also a reaction. The external
 The rigorous definition of this support is shown in Figure 1.2.3c: the support has two sets of rollers so that it does not move either in the upward or downward direction. We need such a support to build a large structure. If a very long bar with pin-and-roller supports is subjected to high temperature in summer, the bar will freely elongate. If the bar has pin supports at both ends, the bar cannot elongate, and a large compressive force may destroy the bar or the supports.  The force exerted by the wall on the bar acts from right to left as shown in Intext Figure 1.1(a). The force exerted by the bar on the wall works in the opposite direction (Intext Figure 1.1(b)). This is Newtons third law.

F cos

(a) Force exerted by the wall on the bar. (b) Force exerted by the bar on the wall.

6861X_C001.indd 4

2/7/08 [Link] PM

Trusses

(a) Rotation

(b) Horizontal movement

(c) Rigorous denition

Figure1.2.3 Roller supports.

force F is balanced by the reactions as shown in Figure1.2.4b. On the other hand, the bar is pulled in tension by the force F cosq. This action is called an axial force and we shall use the letter P to represent it. There is another way to think of the meaning of axial force. Assume that we cut the bar and glue the sections as shown in Figure1.2.5a. Then, apply the force, F, again. The glue must be strong enough to resist a force P = F cos q (Figure1.2.5b,c) to satisfy the equilibrium of forces (Figure1.2.5d). This force P is an axial force. We call this procedure an imaginary cut.

The roller support moves to the right because the bar elongates. Assume that the bar is made of iron. Iron is made up of atoms. Each atom deforms as shown in Figure1.2.6. This is the reason for the lengthening of the bar.

F F cos F sin (a) Reactions F F sin F cos (b) Equilibrium

Figure1.2.4 Forces and reactions.

 Strictly, the atom does not deform, but the distance between the atoms changes.

6861X_C001.indd 5

2/7/08 [Link] PM

Understanding Structures: An Introduction to Structural Analysis

Glue

(a) Glued section

F cos F sin F

(b) Tensile force in the glue

(c) Imaginary cut

Figure1.2.5 Equilibrium near the roller support.

To describe the deformation of any material, we will define stress and strain as follows: Stress: s = P/A (1.2.1)

where P = axial force (positive in tension) and A = sectional area Strain: e = e/L (1.2.2)

where e = deformation (positive for lengthening) and L = original length The reason why we divide the force by the area is that the number of atoms per unit section is definite depending on the material. The stress represents the force on each atom as shown in Figure1.2.7a. The reason for dividing the deformation by the length (as it is done in engineering convention) is that the number of atoms per unit length is definite depending on the material. The strain represents the deformation of each atom as shown in Figure1.2.7b. In the SI system, force is usually described using the unit N (newton). It is equal to the force that imparts an acceleration of 1 m/s2 to a mass of 1 kg. A force of 1 N is approximately the force generated by a mass of 0.1 kg because the acceleration of gravity is 9.8 m/s2. The unit stress is often stated in terms of N/mm2, also called a mega-Pascal and abbreviated as MPa.

(a) Tension

(b) Compression

Figure1.2.6 Deformation of atoms.

 We ought to call this quantity unit stress because the word stress is also used in the sense of force. However, engineers understand stress to refer to unit stress. The same explanation applies to strain.

6861X_C001.indd 6

2/7/08 [Link] PM

Trusses
Deformation of an atom

Axial force, P

Axial force, P Sectional area, A Forces on an atom (a) Stress

Initial length, L Deformation, e (b) Strain

Figure1.2.7 Definition of stress and strain.

In the imperial system of units used commonly in the United States, the unit of force is the pound, lbf, or the kip, which refers to 1000 lbf. The unit of stress is pound per square inch, psi, or ksi, which is equal to 1000 psi. The strain, as we use it in mechanics, does not have a unit because of the way it is defined: the deformation in millimeters is divided by the length in millimeters, or a deformation stated in inches is divided by the original length in inches. If a bar of 1000 mm elongates by 0.5 mm, the strain is 0.5 10 3. Most people have an instinctive understanding of (unit) stress because of their own experiences. Consider the task of driving a large log into the ground (Figure1.2.8a). It is difficult because the force one can muster is distributed over a large area, the cross section of the log. But if one reduces the diameter of the log (Figure1.2.8b), which is similar to sharpening the log (Figure1.2.8c), it becomes easier to drive it in because the force is distributed over a limited area and the stress applied on the ground is higher. One can understand the concept of strain by thinking of the difference between pulling a short and a long rubber band. The long band stretches more even if you

POUND POUND
DOWN?

tap tap

tap tap

DOWN

(a) Stress is small if cross-section is large

(b) Stress is large if cross-section is small

Figure1.2.8 A thin stick can be easily driven in.

6861X_C001.indd 7

DOWN

(c) Stress is large if cross-section is small

2/7/08 [Link] PM

Understanding Structures: An Introduction to Structural Analysis

apply the same force (i.e., the same stress). Why? The same stress causes the same strain, e ; therefore, the deformation (e = e .L), which is proportional to the original length L, is larger in the longer band. If you want to break the bands, it would take the same force, but the extension of the longer band at fracture would be again larger.

Practice Using GOYA-T

Drag the right end of the bar to the left or to the right. As you drag the bar, record the extension (increase in length) and the corresponding force. Plot the force against the increase in length, e. Draw a graph of the relationship between the axial force, P, and the increase in length, e. You will obtain a plot as shown in Figure1.2.9a. You will also notice that the bar turns blue or red if the force exceeds 100 N or 100 N. The change in color indicates that the axial force of the bar exceeds the tensile or compressive strength. In GOYA-T, the section area of the bar is assumed to be 100 mm2. Look at the lower left corner showing that the tensile and compressive strengths of the material are 1 N/mm2. This is why the bar appears to break at a force of 100 N or 100 N. You can change the strength of the material using the text box in the lower left corner. The length of the bar is assumed to be 1000 mm. Because the deformation at fracture is 0.5 mm, we can determine that the strain at fracture is 0.5/1000 = 0.5 103 for this particular material as shown in Figure1.2.9b. As you will learn soon, the stressstrain relationships of most materials are linear at small strains as shown in Figure1.2.9b. We describe the relationship as follows. s = E.e (1.2.3)

e P

Tensile fracture

1 N/mm2 0.5 103

Tensile fracture

100 N 0.5 mm 0.5 mm 100 N Compressive fracture (a) Axial force vs. deformation e

0.5 103 1 N/mm2 Compressive fracture (b) Stress vs. strain

Figure1.2.9 Horizontal force.

6861X_C001.indd 8

2/7/08 [Link] PM

Trusses

Fy Compressive fracture Tensile fracture 100 N Fx

100 N

Figure1.2.10 Inclined force.

The coefficient E is called Youngs modulus after an English scientist, Thomas Young (17731859), who also contributed to archeology deciphering ancient Egyptian characters. Youngs modulus is a measure of how stiff the material is, because the strain e is inversely proportional to E. In the case of Figure1.2.9b, E = s /e = 2 103 N/mm2. Note that the unit of Youngs modulus is same as that of stress because strain has no unit. Question: If the bar is made of a material with tensile and compressive strengths equal to those in Figure1.2.9b, but if its Youngs modulus is halved, how would the relationship between axial force and deformation look? Practice using GOYA-T: Apply forces in various directions as shown in Figure1.2.10. You will find that the bar is safe (the color does not change) if the absolute value of Px is less than a threshold of 100 N. The vertical lines in Figure1.2.10 show the thresholds. Change the compressive strength to 2 N/mm2 while keeping the tensile strength at 1 N/mm2. Show the thresholds. We need to distinguish stiffness from strength. As J. E. Gordon described in his book, a biscuit is stiff but weak, whereas nylon is flexible but strong. Figure1.2.11 shows Youngs modulus and tensile strengths of various materials. Tendon (bundled fibers connecting bone and muscle) is as strong as bone but much more flexible. Nylon is more flexible than timber but ten times as strong. The strength of steel
 Gordon, J. E. 1978. Structures: Or Why Things Dont Fall Down, Da Capo Press, New York.  Timber is stronger and stiffer when stressed in the direction of its fibers. The plot in Figure 1.2.11 shows the data for such a condition.

6861X_C001.indd 9

2/7/08 [Link] PM

10

Understanding Structures: An Introduction to Structural Analysis

(ksi) 150 100 50 0

(N/mm2) 1,000 Nylon Steel

500

Timber Glass Bronze Tendon Bone 0 100,000 Concrete 0 10,000 20,000 Youngs Modulus

200,000 (N/mm2) 30,000 (ksi)

Figure1.2.11 Youngs modulus and tensile strength.

varies depending on its ingredients and treatment, but its Youngs modulus is approximately 200 kN/mm2 (29,000 ksi) whatever its strength is. For interested readers: The stressstrain relationships of actual materials are approximately as shown in Figure1.2.12: Most are linear (stress and strain proportional to each other) for small strain, but they may start responding nonlinearly as strain increases. The stressstrain relationship of timber remains essentially linear up to failure, if loaded in a short time. In the case of steel, the stress is proportional to the strain either in tension or compression up to a yield point, beyond which the stress remains almost constant, whereas the strain increases to the so-called strainhardening limit where stress starts to increase again. The stressstrain relationship of concrete is nonlinear in compression. In tension, its strength is approximately one-tenth of its compressive strength and too small to show on the plot. Example 1.2.1
Calculate the stresses, strains, and deformations of the bars shown in Figure1.2.13. Assume that Youngs modulus is 200 N/mm2 in each case.

Solution
Substituting the given values into s = N/A, e = s /E, and e = e .L, we obtain the following table, which tells us that a thinner and longer bar tends to elongate more.
Stress (N/mm2) (a) (b) (c) 4 8 8 Strain (no unit) 0.02 0.04 0.04 Elongation (mm) 20 40 80

6861X_C001.indd 10

2/7/08 [Link] PM

Trusses
Yield point Strain-hardening limit

11

Tension Concrete Failure Timber Compression Steel

Figure1.2.12 Examples of stressstrain relationships.

Example 1.2.2
A force of 500 N is applied to point B as shown in Figure1.2.14. Calculate the stresses and strains in bars AB and BC. Also calculate the movements (or the displacements, to use the technical term) of points B and C.

Solution
Pin support A ca n carry a horizontal reaction but roller support C cannot. Therefore, the force at point B travels through bar AB to support A, resulting in a tensile axial
A = 100 mm2 e 400 N

1000 mm (a) A = 50 mm2 e 400 N

1000 mm (b) A = 50 mm2 e 400 N

2000 mm (c)
Au: please provide Figure1.2.13 caption

6861X_C001.indd 11

2/7/08 [Link] PM

12

Understanding Structures: An Introduction to Structural Analysis

A E = 500 N/mm2 A = 100 mm2 1000 mm B 500 N C

1000 mm

Figure1.2.14

force of 500 N. The stress in bar AB is s = P/A = 500/100 = 5 N/mm2, and the strain is e = s /E = 5/500 = 0.01. The axial force, stress, and strain in bar BC are zero. The deformation of bar AB is e = e .L = 10 mm, and that of bar BC is zero. Because point A cannot move horizontally, points B and C move right by 10 mm (Figure1.2.15).

AU: Not cited in text. Tentatively cited here. OK?

Example 1.2.3
Three forces, F, F, and 2F, are applied to a bar as shown in Figure1.2.16. Find the correct answer among the following values for the displacement of the bottom end of the bar, if A is the section area, E is Youngs modulus, and the self-weight of the bar is negligible.
1 0 2 FL EA 3 2 FL EA 4 3FL EA 5 4 FL EA

Solution
We first need to evaluate the axial force in the bar. Because there are several external forces, the technique of the imaginary cut (or the free-body diagram) is useful. With an imaginary cut in the bar as shown in Figure1.2.17b, we can determine that the axial force between C and D is 2F. Similarly, Figure1.2.17c,d give the axial forces in segments BC and AB: F and zero, respectively. The results are summarized in Figure1.2.17e. Note that the axial force changes at the points where forces are applied. Based on the axial force, we calculate stress, strain, and deformation (or lengthening) as shown in the first row of Table1.2.1. Adding the deformations for segments AB, BC,

500 N

10 mm 500 N

10 mm

1000 mm

1000 mm

Figure1.2.15

6861X_C001.indd 12

2/7/08 [Link] PM

Trusses

13

A L F B L F C L D 2F

Figure1.2.16

A F F L B L C L 2F D 2F D 2F D P = 2F F C P=F

P=0 F F B C D

A B F C 2F D (e) Axial-force diagram

2F

(a) Problem

(b) Cut between (c) Cut between (d) Cut between C and D B and C A and B

Figure1.2.17

Table1.2.1
Axial Force AB BC CD 0 F 2F Stress 0 F/A 2FL/A Strain 0 F/AE 2FL/AE Deformation 0 FL/AE 2FL/AE

6861X_C001.indd 13

2/7/08 [Link] PM

14

Understanding Structures: An Introduction to Structural Analysis

A B 3L C C 2L B L F

(c) Load at B

2F

(b) Load at C

(a) Load at D

Figure1.2.18

and CD, we obtain the total deformation (or the displacement at point D) of 3Fl/AE. Therefore, the fourth answer is correct. This problem can be solved using another approach. The force 2F at point D elongates the bar by 6FL/AE (Figure1.2.18a). The force F at points C and B shorten the bar by 2FL/AE and FL/AE, respectively (Figure1.2.18b,c). Adding these deformations, we obtain the same result. e= 6 FL 2 FL FL 3FL = EA EA EA EA

Problem: Compute the displacements of points B and C in Example 1.2.3. Another question in reference to the bar described in Figure1.2.17: Compute the displacements of points B, C, and D, assuming that an upward force of 2F and a downward force of 3F are applied at points C and D, respectively.

What Is an Axial Force?

Hi, Joan. I still do not fully understand what an axial force is. What is the difference between an axial force and an external force?

Good question, Jack. An external force is a vector represented by an arrow. It has a magnitude as well as a direction. You see, an external force has a direction such as to the right or down. On the other hand, an axial force is not a vector. You need a pair of arrows with opposite directions to describe an axial force. It represents the magnitude of force with which a member is pulled in tension or pushed in compression. Therefore, we should call it a scalar. Do you like my explanation, Sir?

6861X_C001.indd 14

2/7/08 [Link] PM

Trusses

15

External force

Axial force

Let me give you another explanation. If a member is pulled in tension, each atom in the member is also pulled in tension by a pair of forces in opposite directions. The axial force is the total of these actions in a sliced section.
Forces on an atom Axial force Axial force

Oh, I see. That is the reason why the axial force has a pair of arrows. Then, how about a reaction? Is it a vector? Yes, it is, because a reaction has a direction such as from the wall to the bar. A reaction is similar to an external force in the sense that it is applied to a member from outside. OK. I see that the pair of an external force and a reaction makes an axial force. But structural mechanics looks too abstract to me. I am losing my confidence. Do not worry. If you understand the concept of an axial force, the rest is easy. Cheer up.
External force is a vector. Compression

Reaction is a vector, too.

Axial force is a scalar.

6861X_C001.indd 15

2/7/08 [Link] PM

16

Understanding Structures: An Introduction to Structural Analysis

1.3 Symmetrical Truss with Two Elements of Equal Size at 90

Click the truss shown in Figure1.3.1a in GOYA-T. The truss you will see consists of two members connected at the top so that the members can rotate as shown in Figure1.3.1b. This type of connection is called a pin joint, or more often, a joint, and the point at which two members intersect is called a node. The intersecting members are assumed to be connected by a pin perpendicular to the axes of the members, and the contact surfaces of the pin shaft and the members are assumed to have no friction. One of the functions of a pin joint is to prevent forces that may be caused by rotation of the members related to thermal or other effects. The two truss members are connected to the floor using pin supports. In GOYA-T, the mouse is used to apply a vertical force Fy (up) to the top node. The node moves upward and the members elongate as shown in Figure1.3.2. To investigate the cause of the elongation, we shall cut the truss member as shown in Figure1.3.3a. The resulting system, shown in Figure1.3.3a is called a freebody diagram, or simply, a free body. From that, we can easily infer that the truss members are in tension. (We can arrive at the same conclusion by looking at the stretch or elongation of the truss members in Figure 1.3.2.) Because the free body must be in equilibrium, Newtons third law requires that all vectors representing the forces form a closed triangle as shown in Figure1.3.3b. Note that the tail of each arrow goes to the head of the adjacent arrow. The lengths of the sides of the triangle indicate the relative magnitudes of the forces PAB, PAC , and Fy. The vertical components of the forces in the truss members (equal to the projections on the vertical of the sides of the triangle representing the member forces PAB and PAC) must balance the applied external force. We use that condition to

Fy

100 mm2 B 45

100 mm2 C

(a) Entire view A

(b) Rotation at pin joint

Figure1.3.1 Symmetrical truss with two elements.

AU: Caption?

6861X_C001.indd 16

2/7/08 [Link] PM

Trusses

17

Elongation Tension

Elongation Tension

45 B

45.006 C

Figure1.3.2 Deformation of the truss.

develop Equation 1.3.1. PAB = PAC = Fy /2 sin 45 = Fy 2 (1.3.1)

Having determined that the forces in the truss members are tensile and will result in elongation of the members, we understand why the top node moves up. You should also pay attention to the reaction forces at the truss supports. The equilibrium at support C is shown in Figure1.3.4 (and also in the window of GOYA-T). Note that member AC is in tension. This tensile force causes a reaction of the same magnitude but in the opposite direction at the support.

Fy

Fy/2

PAB 45 45 PAC Fy

Fy/2

PAB 45.006 F 45.006 y PAC

PAB (a) Free body

PAC

Fy/2

Fy/2

(b) Equilibrium

(c) Equilibrium

Figure1.3.3 Equilibrium around the node.

 For interested readers: GOYA-T also shows the displacement of the node. For example, if Fy = 40 N, the displacement will be 0.14 mm. This movement widens the angle between the diagonal member and the horizontal from 45 to 45.006 as shown in Figure 1.3.2, slightly changing the equilibrium as shown in Figure 1.3.3c. However, the effect is small enough and can be neglected.

6861X_C001.indd 17

2/7/08 [Link] PM

18

Understanding Structures: An Introduction to Structural Analysis

PAC

Figure1.3.4 Equilibrium at the support.

Next, apply a horizontal force to node A as shown in Figure1.3.5a. The resulting equilibrium condition is shown graphically in Figure1.3.5b. Member AB is pulled in tension by a force PAB = Fx /2 F = x cos 45 2 (1.3.2)

and member AC is compressed by a force PAC = Fx /2 F = x cos 45 2 (1.3.3)

The minus sign in the equation indicates that the axial force is compression. As shown in Figure1.3.6, the elongation of member AB by the tensile force and the shortening of member AC by the compressive force result in the movement of node A to the right.

 If you find it difficult to distinguish whether a member is in tension or compression, assume that both the members are in tension (Intext Figure 1.13(a)). Then, draw a diagram as shown in Intext Figure 1.13(b), where you will find that the tail of the arrow representing PAC touches the tail of Fx. We therefore conclude that the direction of PAC was wrong and that member AC is in compression. A Fx Fx PAC (a) Free body Fx /2 F = x cos 45 2 PAC PAB

PAB

(b) Equilibrium

 You may calculate the absolute value first: | PAC | =

Then, you need to add a minus sign ( PAC = Fx / 2 ) because the axial force pushes the member in the free-body diagram (Figure 1.3.5a).

6861X_C001.indd 18

2/7/08 [Link] PM

Trusses
A Fx 45 PAB PAC (a) Free body PAC Fx/2 Fx 45 PAB Fx/2

19

(b) Equilibrium

Figure1.3.5 Horizontal force.

Elongation Tension B Shortening A Compression

Figure1.3.6 Deformation caused by horizontal force.

Example 1.3.1
Calculate the axial forces and the reactions if the truss in Figure1.3.6 is subjected to an external force of 100 N at node A. The direction of the force is to the right, and makes an angle of 45 with the horizontal.

Solution
Cut the truss (Figure1.3.7a). The equilibrium defined in Figure1.3.7b requires that the axial force in member AB is100 N and that in member AC is zero. Note that node A moves in the direction of the force because member AB elongates and member AC remains at its original length. The tensile axial force of member AB requires a tensile reaction of 100 N at support B. The zero axial force in member AC requires no reaction at support C.

Elongation

F = 100 N

Tension PAB

No deformation

PAB = 100 N (a) Free body (b) Equilibrium

B
(c) Reaction and deformation

Figure1.3.7 Load at 45 with the horizontal force.

6861X_C001.indd 19

2/7/08 [Link] PM

20
F = 100 A PAB

Understanding Structures: An Introduction to Structural Analysis

30 PAB PAC F

PAC

Fy = F sin30 = 50

F = 100

Fx = F cos30 = 50 3 (c) Components of external force PAC sin45

2/7/08 [Link] PM

(a) Free body A

(b) Equilibrium Fy = 50

PAB 256 (d) Fx only 256 252 (e) Fy only 252

PAB sin45

Fx = 503

PAB cos45

PAC

PAC cos45

(f ) Components of axial forces

Figure1.3.8 Load at 30 with the horizontal force.

Example 1.3.2
Calculate the axial forces if node A of the truss is subjected to an external force of 100 N pulling right at an angle of 30 with the horizontal (Figure1.3.8a).

Solution
The force equilibrium is shown graphically in Figure1.3.8b. To obtain the axial forces, we shall resolve the external force into the horizontal and vertical components as shown in Figure1.3.8c. To balance the horizontal component, Fx = 50 3 N, requires the following axial forces in the truss members (Figure1.3.8d): PAB = Fx = 25 6 N 2 PAC = Fx = 25 6 N 2

The vertical component, Fy = 50 N, requires the following axial forces (Figure1.3.8e). PAB = PAC = Fy 2 = 25 2 N

Adding the forces PAB and PAC, the following results are obtained:

PAB = 25( 6 + 2 ) 97 N

PAC = 25( 6 + 2 ) 26 N

 Recall that the minus sign for PAC simply indicates that the axial force is compression. It does not indicate a change in the direction of the arrow of PAC in Figure 1.3.8d.

6861X_C001.indd 20

Trusses
(3) A (5) (6) B

21

Watch the reactions, too.

(4) F

(2) (1) (8) C

ha ng

es

Co lo rc

(7)

(a) Turn round Direction (1) (2) (3) (4) (5) (6) (7) (8) Fx +140 N Fy 0N PAB +99 N PAC 99 N

(b) Apply forces in eight directions Fy 100 Fx

100

100

100

(c) Load at failure

Figure1.3.9 Strength of the truss.

Problem Using GOYA-T

If the axial force exceeds 100 N, the color of each member turns into blue in tension and red in compression, symbolizing tensile or compressive failure of the member. Failure of one of the members means the failure of the truss itself. Click on the node and drag the top node with your mouse as shown in Figure1.3.9a to see how the axial forces and the reactions vary. Next, apply a force in eight directions successively as shown in Figure1.3.9b and record the external forces and the axial forces corresponding to failure of the truss in the table included in Figure1.3.9. Do not expect the axial force to be listed exactly as 100 N. It may be listed, for example, as 99 N because of numerical errors. Plot the results in Figure1.3.9c to find the weakest direction. Example 1.3.3
Assume that each truss member cannot resist more than 100 N in tension or compression. Given that a single external force is applied at angle q with the horizontal at the top node, determine the horizontal, Fx, and the vertical, Fy, components of that force corresponding to the limiting axial force reached in either of the truss members for any q varying from 0 to 360 (not necessarily at intervals of 45 as shown in Fig 1.3.9b). Plot the relationships with Fx on axis x and Fy on axis y.

6861X_C001.indd 21

2/7/08 [Link] PM

22

Understanding Structures: An Introduction to Structural Analysis

Fy (N) 1002 6 Safe 1002 Fx + Fy = 1002 Fx (N) Fx + Fy = 1002 1002 Fy (N) Safe 0 Fx (N)

(a) PAB = 100

100 2 (b) PAB = 100 Fy (N)

Fy (N) 1002 Safe 1002 0 1002

(c) 100 PAB < 100

Fy (N) 1002 1002 0 Safe

Fx + Fy = 1002

1002 Fx (N)

1002

Fx (N)

1002 Strong Weak Strong Fx (N) 1002 0 1002 1002

(e) Safe limit

Fx + Fy = 1002 (d) 100 < PAC < 100

Figure1.3.10 Safe limit of the truss.

Solution
In this example, we generalize the discrete results obtained in the GOYA-T solution in relation to the simple truss shown in Figure1.3.9b. This time, we work algebraically to define a domain in terms of the vertical and horizontal components of the forces Fy and Fx within which the truss does not fail. The result is shown in Figure1.3.10e. We reach it by deriving a series of expressions for combinations of Fx and Fy that lead to failure in one of the truss members. First, we ask what combinations of Fx and Fy would lead to tensile failure. The force in truss member AB is given by adding Equations 1.3.1 and 1.3.2. P1 = Fx + Fy 2

Tensile failure is assumed to occur if the axial force PAB is 100 N or if Fx + Fy = 100

2 We can rewrite the preceding expression as

Fy = 100 2 Fx which plots as a straight line as shown in Figure1.3.10a. It gives us the limiting combinations of Fx and Fy. Any combination of Fx and Fy to the left of the line is safe, that is, it does not lead to a tensile force more than 100 N in truss member AB.

6861X_C001.indd 22

2/7/08 [Link] PM

Trusses Similarly, a compressive failure occurs in truss member AB if PAB = 100 N or Fy = 100 2 Fx which plots as a straight line in Figure1.3.10b. This line identifies the limiting combinations of Fx and Fy that may lead to a compressive force of 100 N. Any combination of Fx and Fy to the right of the line is safe. It does not lead to a compressive force of more than 100 N in truss member AB. We plot the two lines together in Figure1.3.10c. The region between the lines contains combinations of Fx and Fy that do not cause the limiting force of 100 N in member AB either in tension or in compression. We repeat the process for member AC to obtain the two lines in Figure1.3.10d. These lines bound combinations of Fx and Fy that do not cause the limiting force of 100 N in member AC. In Figure1.3.10e, we plot all four lines to define the safe domain. Combinations of Fx and Fy within the shaded area bounded by the four lines do not lead to forces exceeding 100 N in either member AB or AC. Figure1.3.10e shows that the truss is strongest if pulled in the x or y direction, and is weakest if pulled at 45. Advanced exercise using GOYA-T: You can change the tensile or compressive strength of the material using the window at the lower left corner. If you change the compressive strength to 2 N/mm2 while keeping the tensile strength at 1 N/mm2, what will the safe domain be?

23

Example 1.3.4
Show the safe domain of the truss in Figure1.3.11, where each member fails at an axial force of 100 N.

Solution
Denoting the axial forces of the horizontal and vertical members as PAB and PAC, we have Fx = PAB and Fy = PAC. Because the upper limit of axial force is 100 N, the allowable range of the external force is

|Fx | < 100 N

and

|Fy | < 100 N

Fy

Fx

Figure1.3.11 Another truss.

6861X_C001.indd 23

2/7/08 [Link] PM

24

Understanding Structures: An Introduction to Structural Analysis

Fy (N) 100 Fx (N)

This is the strongest direction.

100

Figure1.3.12 Safe domain.

The shaded area in Figure1.3.12 shows the safe domain. As long as the combination of Fx and Fy can be plotted within the shaded area, the truss members are not overloaded. Note that the truss in Figure1.3.11 is identical to that in Figure1.3.1 except that it is rotated by 45 clockwise. The safe domain in Figure1.3.12 can also be obtained by rotating Figure1.3.10e by 45.

Example 1.3.5
We want to design a truss to carry a baby elephant weighing 1000 lbf as shown in Figure1.3.13. Assume that the tensile strength of the material is 5000 psi. What are the required cross-sectional areas of the truss members?

Solution
The gravity force related to the elephant is 1000 lbf. The axial force in each member is 1000/ 2 = 500 2 lbf. Because the stress in the member is given by s = P/A, the required cross-sectional area is A = P/s = 500 2 /5000 = 0.1 2 in.2, or approximately 0.14 in.2.

? in2

? in2 45

Safe

Figure1.3.13

6861X_C001.indd 24

2/7/08 [Link] PM

Trusses

25

Safe Fx

? in2 45

Fy

? in2

AU: Caption? Figure1.3.14

Design Your Own Truss (Part 1)

We want to design another truss to support the baby elephant as shown in Figure1.3.14. Assume that the compressive strength of the material is 1000 psi and the weight of the elephant is 2000 lbf plus any three numbers you may choose (e.g., 2650 lbf). Consider also a lateral seismic force (Fx in the figure) of one half the gravity force. The truss should carry a vertical force of Fy = 2650 lbf and a horizontal force of Fx = 1325 lbf. What are the required cross-sectional areas of the truss members? Hint: The needed area should be between 2 and 4 in.2. (In reality, the truss should be three-dimensional to sustain a wind or seismic force in the direction perpendicular to the paper. We will study such a truss in Section 1.7.)

Coffee Break
Leonardo da Vinci (14521519) was the first person to calculate the axial force in a truss. However, he kept his finding in his private notebook and never let the public know. As a result, engineers in the 15th and 16th centuries determined structural member sizes based only on experience, as did the ancient Romans.

1.4 Symmetrical Truss with Two Elements at Various Orientations

Start GOYA-T. Select a truss with two members of equal length as shown in Figure1.4.1a and apply an upward force to node A. The two members elongate, and the node moves upward as shown in Figure1.4.1b. Cut the truss near node A as shown in Figure1.4.2a to obtain equilibrium in Figure1.4.2b leading to PAB = PAC = Fy /2 sin 60 = Fy 3 (1.4.1)

 Timoshenko, S. P. 1953. History of Strength of Materials. McGraw-Hill, New York.

6861X_C001.indd 25

2/7/08 [Link] PM

26

Understanding Structures: An Introduction to Structural Analysis

Fy A 1000 mm 60 B (a) Entire view C B 1000 mm Tension Elongation

Elongation

Tension

C (b) Deformation

Figure1.4.1 Truss of equilateral triangle.

Next, apply a horizontal force to node A. You will see the deformation truss deflecting as shown in Figure1.4.3, and equilibrium in the free-body diagram (Figure1.4.4). The axial force in each member is given by the following equations, in which the minus sign represents compression. PAB = PAC = Fx /2 = Fx cos 60 Fx /2 = Fx cos 60 (1.4.2) (1.4.3)

Problem Using GOYA-T

As in the previous section, the color of each member changes if the axial force exceeds 100 N, indicating failure of the member. Drag the mouse spirally as shown in Figure1.4.5a to see how the axial forces and the reaction vary. Next, apply a force in eight directions as shown in Figure1.4.5b and record the internal and external forces

Fy

Fy/2 60 60

PAC

PAB (a) Free body

PAC

Fy/2 (b) Equilibrium

PAB

Figure1.4.2 Equilibrium around node A.

6861X_C001.indd 26

2/7/08 [Link] PM

Trusses
Elongation A Shortening

27

Tension

Compression

Figure1.4.3 Deformation caused by the horizontal force.

corresponding to the failure of the truss. When you apply the force in the direction indicated as (2), you will find that the node A does not move in the direction of the force but at an angle of 30 with the horizontal as shown in Figure1.4.5d because the right member AC has no axial force and therefore does not change its length. Thus, the member AC simply rotates around the support C without any change in length.
AU: The term Example was used in the preceding section. The rest of the book (apart from this section and subsequent sections of this chapter) uses Example.

Exercise 1.4.1
Derive the equations to represent the axial forces in the members of the truss shown in Figure1.4.5b if an external force is applied successively on the truss in different directions at node A. Then draw a graph to show the safe domain of the truss in the Fx - Fy plane.

Solution
PAB is given by combining Equation 1.4.1 and 1.4.2. PAB = Fx + Fy 3 (1.4.5)

PAC is given by combining Equation 1.4.1 and 1.4.3. PAC = Fx + Fy 3 (1.4.6)

Fx

Fx/2 60

Fx/2 60 PAB

PAB (a) Free body

PAC

PAC

(b) Equilibrium

Figure1.4.4 Equilibrium at node A.

6861X_C001.indd 27

2/7/08 [Link] PM

28

Understanding Structures: An Introduction to Structural Analysis

Fy (2) A (6) (8) (1) 100 100 C (b) Apply forces in eight directions 100 100 Fx

(4) F (5)

(3)

(7)

(a) Turn round

(c) Load at failure A

Direction (1) (2) (3) (4) (5) (6) (7) (8)

Fx +100 N

Fy 0N

PAB +99 N

PAC 99 N Tension

No deformation

B (d) Displacement

Figure1.4.5 Strength of the truss.

From the conditions 100 PAB 100 and -100 PAC 100, we obtain the shaded region in Figure1.4.6a and b, respectively. Combining these figures, we obtain Figure1.4.6c.

Exercise 1.4.2
Evaluate the most disadvantageous direction in which the truss is most vulnerable to failure and determine the magnitude of the force in that direction at which the truss fails. Then calculate the axial force of the members at the failure.

Solution
Focusing on one quadrant of the plot in Figure1.4.6c, we see in Figure1.4.7a that the weakest direction is represented by the arrow in the figure (30 from the horizontal and perpendicular to the boundary). The magnitude of the force is

F = 100 cos 30 = 50 3 N

6861X_C001.indd 28

2/7/08 [Link] PM

Trusses
Fy (N) 1003 PAB = 100 N Fy (N) 1003 PAC = 100 N 0 100 Safe Fx (N) 0 Safe 100 Fx (N) PAC = 100 N (b) 100 < PAC < 100 (c) Safe limit 0 100 Fy (N) 1003

29

Fx (N)

PAB = 100 N

(a) 100 < PAB < 100

Figure1.4.6 Safe domain of the truss.

The axial forces of the members are determined in Figure1.4.7b,c:

PAB =

F = 100 N cos 30

PAC = F tan 30 = 50 N

Exercise 1.4.3
Draw a graph to show the safe domain for the truss in Figure1.4.8 assuming that the members fail at 100 N both in tension and compression.

Solution
The axial forces are PAB = Fy Fx + 2 cos 2 sin Fy Fx + 2 cos 2 sin (1.4.7) (1.4.8)

PAC =

Fy (N) 1003 30 30 253 0 30 503 75 100 Fx (N) PAB (b) Free body A F 30 60 F PAB PAC PAC (c) Equilibrium 30

(a) Enlarged diagram

Figure1.4.7 Most vulnerable direction.

6861X_C001.indd 29

2/7/08 [Link] PM

30

Understanding Structures: An Introduction to Structural Analysis

Fy A Fx L B C L

Figure1.4.8 Equilateral truss with angle q.

From the conditions 100 PAB 100 and -100 PAC 100, we obtain the shaded region in Figure1.4.9.

Exercise 1.4.4
Draw a graph to represent the safe domain of the truss shown in Figure1.4.8 assuming that the members fail at 100 N in tension and at 200 N in compression.

Solution
From the conditions -200 PAB 100 and -200 PAC 100, we obtain the shaded region in Figure1.4.10.

Design Your Own Truss (Part 2)

We want to design another truss with members at 60 to one another to support an elephant. Assume that the compressive strength of the material is 1000 psi. The total weight of the elephant is 2000 lbf plus any three numbers you may choose. Consider

Fy (N) 200 sin Fx 2 cos + Fy 2 sin = 100

200 cos

Fx (N)

Figure1.4.9 Safe domain.

6861X_C001.indd 30

2/7/08 [Link] PM

Trusses
Fy (N)
200 sin

31

Fx 2 cos

Fy 2 sin Fx (N)

= 100

200 cos

400 sin

Fx 2 cos

Fy 2 sin

= 200

Figure1.4.10 Safe domain.

Safety Fx

Fy ? in2 60 ? in2

Figure1.4.11 Your truss.

AU: Figure 1.4.11 not cited in text. Tentatively cited here. OK?

also a lateral seismic force (Fx in Figure1.4.11) of half the gravity force. How much cross-sectional area is required for each member? (Hint: The needed area will be slightly larger than that in the previous section.)

1.5 Unsymmetrical truss with two elements

Start GOYA-T and click the truss with members having different cross-sectional areas as shown in Figure1.5.1. Apply an upward force at node A. You will see that the node does not move vertically in line with the applied force but moves to the left as it moves up (Figure1.5.2). To understand the reason for this movement, cut the truss as shown in Figure1.5.3a and consider the equilibrium diagram in Figure1.5.3b. The axial forces of the members are given as follows: PAB = PAC = Fy /2 sin 60 = Fy 3 (1.5.1)

6861X_C001.indd 31

2/7/08 [Link] PM

32

Understanding Structures: An Introduction to Structural Analysis

Fy

1000 mm 100 mm2 60 B

1000 mm 50 mm2

Figure1.5.1 Unsymmetrical truss.

Smaller elongation A

Larger elongation

Figure1.5.2 Deformation.

Fy

Fy/2 60 60

PAC

PAB (a) Free body

PAC

Fy/2

PAB

(b) Equilibrium

Figure1.5.3 Equilibrium at the node.

6861X_C001.indd 32

2/7/08 [Link] PM

Trusses
Small elongation A Large shortening

33

Figure1.5.4 Deformation.

Note that the axial forces are same as those in the previous section though the cross section is different. However, the stresses in members AB and AC are different because the stress is the axial force divided by the cross-sectional area. The stress in member AC is twice that in member AB, and member AC elongates twice as much as member AB, so that the node moves left as it moves up. Apply a horizontal force, acting to the right, at node A. The node moves down as it moves to the right (Figure1.5.4). The absolute magnitudes of the axial forces in the two members are the same as shown in Figure1.5.5b, but because the cross section of member AC is half that of member AB, AC shortens twice as much as AB elongates. Exercise 1.5.1
Assume that the strength of the material used for the truss in Figure1.5.1 is 1 N/mm2 both in tension and compression. Show the safe domain of the truss for combinations of forces Fx and Fy.

Solution
The axial forces are given by Equations 1.4.4 and 1.4.5 in the previous section. Because the limits of the axial force of members AB and AC are 100 N and 50 N, respectively, the safe domain for the truss is shown by the shaded area in Figure1.5.6. At point G (Fx = 25 N and Fy = 75 3 N), both members fail in tension simultaneously. If you
A Fx

Fx PAC PAB

PAB (a) Free body

PAC

(b) Equilibrium

Figure1.5.5 Equilibrium.

6861X_C001.indd 33

2/7/08 [Link] PM

34

Understanding Structures: An Introduction to Structural Analysis

Fy (N) Fx + Fy Fx + = 50 0 Fx + G H Fx (N) 2 Fy = 50

Fy

= 100

Figure1.5.6 Safe domain.

F = 3 1 3 , the left and right members elonapply a force in the direction of OG so that Fx y gate equally, and the node moves upward. If you apply a force in the direction of OH so that Fx /Fy = 3 , the node moves to the right. Apply forces as shown in Figure1.5.7 to develop a table listing compatible values of Fx, Fy, PAB, and PAC as you did in the previous section, and plot the results in Figure1.5.6. We could evaluate the axial forces in all the trusses up to this point considering equilibrium. Such a structure is described as statically determinate. Figure1.5.8 shows another example in which the axial force in member AB is the same as that member BC but the elongation of member AB is more than that of BC. In this case, we can determine the forces in each member using the condition of horizontal equilibrium. On the other hand, there are structures for which axial forces cannot be evaluated unless the deformation of each member is considered. Such a structure is described as statically indeterminate. Figure1.5.9 shows an example in which the elongations of the two members are the same but the axial forces are different. Equilibrium requires the following equation:

F = P1 + P2

(1.5.2)

(5) (6) (7)

(4)

(3) (2) (1)

PAB

PAC

Figure1.5.7 Loading in seven directions.

6861X_C001.indd 34

2/7/08 [Link] PM

Trusses
A F Larger elongation B Smaller elongation C F

35

Figure1.5.8 Statically determinate structure.

Where the elongation is e and the length of the members L , the strain in the members are given by

e L

(1.5.3)

Calling the cross-sectional areas of the members A1 and A2, and the Youngs modulus E, the axial forces are P1 = A1 = A1 E = P2 = A2 = A2 E = A1 Ee L A2 Ee L (1.5.4) (1.5.5)

Eliminating the elongation e from these equations, we obtain P1 = P2 = A1 F A1 + A2 A2 F A1 + A2 (1.5.6) (1.5.7)

Note that the thicker member carries the larger axial force. Large structures such as bridges and domes are often designed to be statically determinate to prevent thermal stresses in summer and winter. If the left member in Figure1.5.10 is heated, the member simply elongates without any additional stresses. On the other hand, if the upper member in Figure1.5.11 is heated while the lower member is not, the upper member cannot elongate as much as in Figure1.5.10 (e2 < e1)

L P1 P2 A1 Same elongation A2

Figure1.5.9 Statically indeterminate structure.

6861X_C001.indd 35

2/7/08 [Link] PM

36

Understanding Structures: An Introduction to Structural Analysis

e1

Heat elongate

Figure1.5.10

because of the restraint by the lower member. As a result, a compressive force appears in the upper member, whereas a tensile force appears in the lower member.

The bar elongates when heated.

Engineers should be very careful in constructing a statically determinate structure because even a single error may cause a catastrophic collapse. A bad weld in one of the members of the structure in Figure1.5.8 will cause an immediate collapse, whereas the structure in Figure1.5.9 might survive a weld failure in one of the members because the remaining member may be able to carry the applied load F. This section deals only with statically determinate trusses primarily because they are easier to study and provide a good introduction to understanding structural response. Exercise using GOYA-T: You can find the truss shown in Figure1.5.12 in GOYA-T. Assume that the material strength is 1 N/mm 2 both in tension and compression and draw the safe domain of the truss. After drawing the domain, check it using GOYA-T.

e2 P P Heat Same elongation P P

Figure1.5.11

6861X_C001.indd 36

2/7/08 [Link] PM

Trusses
Fy Fx 100 mm2 45 50 mm2

37

Figure1.5.12

Exercise 1.5.2
You can find the truss shown in Figure1.5.13a in GOYA-T. Apply an upward force to node A. You will see that the node does not move upward but diagonally as shown in Figure1.5.13b. Explain the reason.

Solution
Figure1.5.14 shows equilibrium at node A. All of the external force is carried by the axial force of the vertical member (PAB = Fy). Member AC carries no axial force and does not elongate or shorten. Therefore, the length of CA is the same as that of CA, which is the reason for the diagonal movement of node A.

Exercise 1.5.3
Show the safe domain for the truss in Figure1.5.13a assuming that each member can carry 100 N both in tension and compression.

Fy A

45 (a) Vertical force

B (b) Deformation

Figure1.5.13

6861X_C001.indd 37

2/7/08 [Link] PM

38

Understanding Structures: An Introduction to Structural Analysis

Fy

PAB

Figure1.5.14

AU: Caption?

Solution
Figure1.5.15 shows the equilibrium at node A for a horizontal force, Fx. The resulting axial forces are P1 = Fx P2 = 2 Fx (1.5.8) (1.5.9)

The minus sign in the equation represents that the axial force is compression. The axial forces caused by Fx and Fy are P1 = Fx + Fy P2 = 2 Fx (1.5.10) (1.5.11)

Because the upper limit of the axial force is 100 N, the safe domain is as shown in Figure1.5.16.

Fx

Fx PAC PAB

PAB

PAC (b) Equilibrium

(a) Free body

Figure1.5.15 Horizontal force.

6861X_C001.indd 38

2/7/08 [Link] PM

Trusses
Fy (N) Fx + Fy = 100

39

Fx (N) Fx = 502

Figure1.5.16 Safe domain.

Safety Fx

Fy ? in2

? in2 45

Figure1.5.17

Design Your Own Truss (Part 3)

We want to design another truss to support an elephant as shown in Figure1.5.17. Assume that the tensile and compressive strengths of the material are 500 and 1000 psi, respectively. The weight of the elephant is 2000 lbf plus any three numbers you may choose. Consider also a seismic force (Fx in the figure) of half the gravity force to the left and the right. How much cross-sectional area is needed for each member? (Hint: The needed area will be larger if a seismic force is applied to the left rather than to the right. The area would be between 2.5 and 5 in.2)

1.6 Truss with three members

Figure1.6.1 shows a truss with a pin support at B and a roller support at C. Note that the two supports are connected with a horizontal member. Select this type of truss in GOYA-T and apply an upward force, FAy, to node A. You will find that the reactions at supports B and C (R By and RCy) are FAy/2. You will also find that member BC shortens and support C moves to the left.

6861X_C001.indd 39

2/7/08 [Link] PM

40

Understanding Structures: An Introduction to Structural Analysis

FAy

B RBy

60

C RCy

AU: Caption?

Figure1.6.1 Truss of three members.

Why member BC shortens may be understood as follows. First, look at Figure1.6.2 showing the free-body diagram of node A. Equilibrium conditions and symmetry help determine the axial forces in the inclined members: PAB = PAC = FAy /2 sin 60 = FAy 3 (1.4.1)

Second, look at Figure1.6.3 showing the free-body diagram at support C. Note that the reaction is in the vertical direction. A horizontal component cannot exist because of the roller. The axial force in the bottom member BC is compressive because it has to resist the horizontal component of the axial force in member AC. PBC = PAC cos 60 = FAy 2 3 (1.6.1)

The negative sign represents that the axial force in member BC is compressive. Because it is in compression, member BC shortens. The axial forces, PAC and PBC, as well as the reaction RCy, satisfy the equilibrium shown in Figure1.6.3b.

FAy FAy

PAC

PAB (a) Free body

PAC

PAB (b) Equilibrium

Figure1.6.2 Equilibrium at node A.

6861X_C001.indd 40

2/7/08 [Link] PM

Trusses
PAC PBC PAC RCy

41

PBC

C RCy (a) Free body

(b) Equilibrium

Figure1.6.3 Equilibrium at support C.

Similarly, the equilibrium at support B is shown in Figure1.6.4. Apply a force FAx pulling to the right at node A as shown in Figure1.6.5. You will find that the horizontal member elongates and support C moves to the right. Look at Figure1.6.6 showing equilibrium at node A. It indicates that the axial force in member AB is tensile: PAB = FAx /2 = FAx cos 60 (1.6.2)

The axial force in member AC is compressive: PAC = FAx (1.6.3)

Next, look at Figure1.6.7 showing the free-body diagram at node C. Again, the reaction at C is vertical because of the presence of the roller. The axial force in the bottom member is tensile: PBC = PAC cos 60 = FAx 2 (1.6.4)

PAB B RBy (a) Free body PBC

PAB PBC

RBy

(b) Equilibrium

Figure1.6.4 Equilibrium at support B.

6861X_C001.indd 41

2/7/08 [Link] PM

42

Understanding Structures: An Introduction to Structural Analysis

A

FAx

L sin60 60

RBx

B RBy

RCy C

Figure1.6.5 Horizontal force to the truss.

FAx PAC

FAx PAB

PAB (a) Free body

PAC (b) Equilibrium

Figure1.6.6 Equilibrium at the top node.

PAC

PBC

RCy C (a) Free body

RCy

PAC

PBC (b) Equilibrium

Figure1.6.7 Equilibrium at support C.

6861X_C001.indd 42

2/7/08 [Link] PM

Trusses

43

PAB PBC RBx B RBy (a) Free body PBC PAB RBx (b) Equilibrium RBy

Figure1.6.8 Equilibrium at support B.

This tensile force elongates the bottom member, moving node C to the right. The free-body diagram in Figure1.6.7b can also be used to determine the reaction at C: RCy = PAC sin 60 = 3 F 2 Ax (1.6.5)

The free-body diagram for support B is shown in Figure1.6.8. From that we can determine the x- and y-components of the reaction. RBx = PAB cos 60 + PBC = FAx FAx + = FAx 2 2 3 F 2 Ax (1.6.6) (1.6.7)

RBy = PAB sin 60 =

All reaction forces can be and were determined considering equilibrium at the nodes. However, we can calculate the reactions directly if we consider the equilibrium of the truss as a whole. Note that the number of unknown variables in Figure1.6.5 is three (R Bx, R By, and RCy). Thus, we need three equations. The first two conditions are: 1. The sum of the x-components of the forces acting on the structure is zero (X = 0). This leads to -R Bx + Fx = 0, where R Bx has a negative sign because it goes to the left. 2. The sum of the y-components of the forces acting on the structure is zero (Y = 0). This leads to R By - RCy = 0, where RCy has a negative sign because it goes downward. The third condition may be called the principle of the lever, discovered by Archimedes, an ancient Greek philosopher. In the case of the lever scale of Figure1.6.9, the principle requires F1a1 = F2 a2 . To generalize the principle, we need to introduce a moment, an action that turns an object around a point as shown in Figure1.6.10. The moment is defined as follows: (Moment) = (Force) (Distance) (1.6.8)

6861X_C001.indd 43

2/7/08 [Link] PM

44

Understanding Structures: An Introduction to Structural Analysis

F1 + F2 A B C

F1 a1 a2

F2

Figure1.6.9 Lever scale.

Equation 1.6.8 is very important. You should understand and remember it. In this book, a clockwise moment is defined as positive and an anticlockwise moment is defined as negative. For an object in equilibrium, the lever principle requires the following: 3. The sum of the moments acting on the structure is zero (SM = 0). In the case of the lever shown in Figure1.6.9, the principle leads to

M = F a + F a
1 1

2 2

=0

(1.6.9)

where -F1a1 and F2 a2 represent, respectively, the anticlockwise and clockwise moments around point B. In fact, the reference point can be anywhere other than at point B because the lever would not rotate around any point. If the reference point is taken as shown in Figure1.6.11, the principle leads to

M = F x (F + F )( x + a ) + F ( x + a + a ) = 0
1 1 2 1 2 1 2

(1.6.10)

You may find this equation reduces to Equation 1.6.9.

Force

Reference point

Distance

Figure1.6.10 Definition of moment.

6861X_C001.indd 44

2/7/08 [Link] PM

Trusses

45

F1 + F2 F2 a1 a2

Reference point

F1 x

Figure1.6.11 Moment around an arbitrary point.

We return to Figure1.6.5 to consider that the equilibrium of moments around the left support leads to

M = F

Ax

L sin 60 RCy L = 0

(1.6.11)

This results in RCy = FAx sin 60, which is equivalent to the solution obtained previously (Equation 1.6.5). But this process makes it much easier to calculate the reactions directly rather than considering the equilibrium of each node.

Exercise Using GOYA-T

Assume that each member of the truss shown in Figure1.6.12 fails at the axial force of 100 N both in tension and compression. Apply forces in seven directions, one at a time, as shown in the figure. Develop a table of the loads, axial forces, and reaction corresponding to the failure of the truss. Exercise 1.6.1
Plot the safe domain for the truss shown in Figure1.6.12 assuming that the members fail at 100 N both in tension and compression.

(6) (7)

(5)

(4)

(3)

(2) (1)

Direction (1) (2) (3) (4) (5) (6) (7)

FAx +100 N

FAy 0N

PAB +99 N

PAC 99 N

PBC +50 N

RCy +87 N

PAB PBC

PAC

RCy

Figure1.6.12 Apply force in various directions and check the limit.

6861X_C001.indd 45

2/7/08 [Link] PM

46

Understanding Structures: An Introduction to Structural Analysis

Solution
The axial forces of the left and right members, PAB and PAC, were determined in Section 1.4. PAB = FAx + FAy 3 FAy 3 (1.4.3) again (1.4.4) again

PAC = FAx +

The axial force of the bottom member, PBC, is determined adding Equation 1.6.1 and 1.6.4. PBC = FAy FAx 2 2 3 (1.6.12)

Thus, we obtain the shaded safe domain in Figure1.6.13. Note that the conditions of PBC = 100 (the broken lines in the figure) do not affect the shaded region. In fact, Equations 1.6.12 and 1.4.4 lead to PBC = -PAC/2, which indicates that the axial force in bottom member AC is always half of that in member BC. Therefore, the truss members need not have equal cross-sectional areas; it is more economical if we make the cross-sectional area of member AC half the cross-sectional area of each of members AB and BC as illustrated in Figure1.6.14.

Exercise 1.6.2
Calculate the reactions if external forces FAx and FAy are applied simultaneously to the truss at node A as shown in Figure1.6.15. Assume that the reactions in the figure have positive signs.

Solution
The equilibrium conditions SX = 0 and SY = 0 lead to the following equations. FAx + RBx = 0 and FAy + RBy + RCy = 0

FAy (N) 1003 PAC = 100 PAB = 100

0 PAB = 100 PBC = 100

100

200

FAx (N)

PAC = 100 PBC = 100

Figure1.6.13 Safe domain.

6861X_C001.indd 46

2/7/08 [Link] PM

Trusses

47

FAy FAx

60

Figure1.6.14 Economical design.

Also, the equilibrium condition M = 0 around support B leads to

FAx L RCy L = 0 Thus, we have the following reactions:

RBx = FAx

RBy = ( FAx + FAy )

RCy = FAx

You can find this type of truss in GOYA-T. Develop a table similar to that in Figure1.6.12.

FAy FAx

L RBy B

45 RBx

RCy C

Figure1.6.15 Right-angled truss.

6861X_C001.indd 47

2/7/08 [Link] PM

48

Understanding Structures: An Introduction to Structural Analysis

Design Your Own Truss (Part 4)

We want to design another truss. Assume that the tensile and compressive strengths of the material are 500 and 1000 psi, respectively. The weight of the elephant is 2000 lbf plus any three numbers you may choose. Consider also a lateral seismic force (Fx in the figure) of half the gravity force acting successively in both horizontal direc- AU: What figure? tions. What is the minimum cross-sectional area that would be required for each one of the three truss members?

Why Should We Calculate the Reactions?

Sir, I do not understand why we should calculate the reactions.

There are two reasons. First, we need to check whether the ground can sustain the forces from the truss. It would be dangerous if the support fails.

Oh, yes, it is quite obvious. Thanks. Bye.

Wait. There is another important reason. Do you know the reason for determining the axial forces?

Well, to check the safety of the truss. If an axial force exceeds a limit, the truss will fail.

6861X_C001.indd 48

2/7/08 [Link] PM

Trusses

49

Good. Later on, you will learn to deal with trusses comprising many members. In such trusses, you will have great difficulties if you try to calculate the axial forces using equilibrium conditions at each node. It is much easier to calculate the reactions first. As you know, all the three reactions are obtained from X = Y = M = 0, three simultaneous equations with three unknowns.

You should be rigorous, Sir. For a truss supported by a pin and a roller, the number of the reactions is three, and therefore X = Y = M = 0 is enough. For other cases, however, it may not be enough. For example, a truss supported by two pins has four reactions, and we need another equation.

You are right, Joan. But let us ignore such trusses for a while. We may talk about them next year. The important thing is that external forces and reactions always satisfy the condition X = Y = M = 0.

6861X_C001.indd 49

2/7/08 [Link] PM

50

Understanding Structures: An Introduction to Structural Analysis

Oh, you mean that I can treat the reactions and the external forces equally.

Exactly. Both the reactions and the external forces work from outside the truss. In Section 1.2, you learned that an external force and a reaction work from outside making an axial force in a rod. It is a similar story. (For interested readers: the moment produced by a force is a vector.) Imagine a heavy disc on the x-y plane as shown in Figure1.6.16. The disk can rotate around the z-axis. An external force F (a vector) uuur is applied to the disk at point A, whose location is represented by a vector a = OA . Both a and F are on the x-y plane; q represents the angle between a and F. The moment of the force F around point O is defined by M = a F sin

Note that if the force is in the radial direction (i.e., q = 0), the force will not turn the disk (M = 0). As the angle q increases up to 90, the disk tends to turn more easily ( M = a F ). Also note that the moment has an axis of rotation (the z-axis), as the screw in Figure1.6.17. The moment of a force is a vector having both a magnitude

Safety Fx

Fy ? in2 45

? in2

? in2

Figure1.6.16 Force on a disk.

6861X_C001.indd 50

2/7/08 [Link] PM

Trusses
z M O a x A F y

51

Figure1.6.17 Screw.
M

aF

Figure1.6.18 Work or friction energy.

and a direction. Mathematicians call it a cross-product and express it as follows. M = aF

The other definition of the product of vectors (a dot product) also plays an important role in mechanics, representing energy or work, which is a scalar. In Figure1.6.18, the force F and the displacement u make the following work, W: W = F u cos

1.7 Hands-on Approach to Truss Design

You need: 1. Four pieces of wood, 2 ft. long, with a square section of 1/8 in. Do not buy balsa wood. 2. A string or thread approximately 7 ft. long.

6861X_C001.indd 51

2/7/08 [Link] PM

52

Understanding Structures: An Introduction to Structural Analysis

Figure1.7.1 Truss with four legs.

3. A knife. 4. A scale. 5. A computer or a calculator. 6. Clay to be used as a weight. 7. A plastic bag. 8. A kitchen scale (your instructor may provide one).

In this section, you will make a four-leg truss as shown in Figure1.7.1. The truss should be designed to collapse if you hang a specific weight at point A. You may choose the weight to be from 2 to 3 lbf. The elevation of a truss is conceptualized as shown schematically in Figure1.7.2. The force at the top, F, represents the gravity force caused by the weight. The inclined members AB, AC, AD, and AE resist compressive forces, and the horizontal members CB and DE resist tensile forces. Therefore, we shall use wood for the compressive members and string for the tensile members. AU: Changes There are three possible types of collapse modes for the truss: OK? 1. Failure of a connection because of loosening or slip of the string 2. Tensile failure of the string 3. Buckling of the wood (a failure mode caused by bending of an axially compressed member as shown in Figure1.7.3.) We can design a truss with a desired strength, specified in reference to each of these modes of collapse. The first type of collapse mode is difficult to control; we can only tie the string as tightly as possible. The second type is not feasible because the string, in tension, is usually stronger than the wood is in compression. It should be strong enough to tie the wood or to hang the weight from the top of the truss. The remaining method is to design the wood member, acting in compression, to buckle under the desired load. The theoretical explanation of the buckling load is beyond the scope of this chapter, and will be given in Chapter 6. In this example, we shall use the standard expression
 [Link]

6861X_C001.indd 52

2/7/08 [Link] PM

Trusses

53

A D C

B E

Figure1.7.2 Side view.

engineers employ to determine the buckling load of a concentrically loaded prismatic column; but the following equation indicates the critical load: Pcr =

2 Eh 4 12 L2

(1.7.1)

where E is Youngs modulus, h the size of the section, and L the length of the member. Youngs modulus of wood differs, depending on the kinds of trees, but is approximately 1500 ksi. If the wood member measures 1/8 1/8 24 in., the buckling load is determined to be Pcr =

2 1500 10 3 (1/8)4 0.52 lbf 8.3 oz 12 24 2

A F D, C E, B

Figure1.7.3 Buckling.

 In engineering use, prismatic refers to a member that has uniform dimensions and properties along its length. In our application, we are using a wood element having a square section with the same dimensions along its length, and we hope its Youngs modulus, E, does not change from end to end.

6861X_C001.indd 53

2/7/08 [Link] PM

54

Understanding Structures: An Introduction to Structural Analysis

It is instructive to test whether the calculated value of 8.3 oz is a good measure of the buckling load before we construct the truss. We can check the value using a kitchen scale. Place one end of the wood member, 2 ft. long, on the kitchen scale, and push the other end with your finger slowly until the member buckles. In the case of the truss shown in Figure1.7.1, L represents the distance between the nodes, say, A and B. We do not expect the measured buckling load to be exactly as computed. Youngs modulus, the dimensions, and the end conditions of the actual member may be different from those we assumed. However, our calculated answer should be close to the measured load unless we have made a critical error in our calculations or in the test. Having developed confidence in the computed buckling load, how do we relate it to the applied force F on a three-dimensional truss? How do we predict the axial force in the three-dimensional truss? For this purpose, we use Figure1.7.4, where A, B, C, D, and E are the points at which the wood members are tied. Point O is located directly below A and lies on plane BCDE. The distance between points O and B is OB = b 2 + c 2

and the distance between points A and B is L = AB = OA + OB = a 2 + b 2 + c 2

2 2

Pcr h h

Pcr

Figure1.7.4 Dimensions.

6861X_C001.indd 54

2/7/08 [Link] PM

Trusses

55

A O E

C 2c

B 2b

Figure1.7.5 Free-body-diagram.

Because of symmetry, we conclude that L = AB = AC = AD = AE = a 2 + b 2 + c 2 (1.7.2)

This value shall be used to evaluate the buckling load of the inclined members. Next, we consider the equilibrium at point node A as shown in Figure1.7.5, in which F denotes the external force. We may conclude that the axial forces in the inclined members are the same because of the symmetry of the truss. PAB = PAC = PAD = PAE = P (1.7.3)

The sum of the vertical components of the axial forces, P cos , equals the external force, F. where q denotes the angle OAB: cos = OA a = AB L (1.7.5) F = 4 P cos (1.7.4)

If the axial force is equal to the buckling load, Pcr, we have F= 4a P L cr (1.7.6)

Let us consider the following two cases. The squat truss in Figure1.7.6, where a = b = c = 4 in. The tall truss in Figure1.7.7, where a = 8 in. and b = c = 4 in.

6861X_C001.indd 55

2/7/08 [Link] PM

56

Understanding Structures: An Introduction to Structural Analysis

F A

PAD

PAC PAB PAE

Figure1.7.6 Squat truss.

For both trusses, we use wood members with h = 1/8 in. and E = 1500 ksi. Equations 1.7.2 and 1.7.6 lead to a relationship between the applied vertical force F and the axial force in each truss member: For the squat truss in Figure1.7.6: L = 4 2 + 4 2 + 4 2 = 6.9 in. 4 F = 46 .9 P cr = 2.3P cr gives an axial force of For the tall truss in Figure1.7.7: L = 82 + 4 2 + 4 2 = 9.8 in. 8 F=4 9.8 P cr = 3.3P cr gives an axial force of and and

A C D 2c = 8 in.

a = 4 in.

B E 2b = 8 in.

Figure1.7.7 Tall truss.

 In the case of the 2-D model of Figure 1.7.2, the axial force in the inclined members is P = F / 2 = 0.707 F . Noting that each member in Figure 1.7.2 represents two members in the real 3-D truss of Figure 1.7.3 (e.g., AB and AE), the 2-D model implies an axial force of P = 0.707 F/2 = 0.354F, which is similar to the exact value of 0.375F.  In the case of the 2-D model of Figure 1.7.2, the axial force in the inclined members is P = F / 2 = 0.707 F. Noting that each member in Figure 1.7.2 represents two members in the real 3-D truss of Figure 1.7.3 (e.g., AB and AE), the 2-D model implies an axial force of P = 0.707 F/2 = 0.354F, which is similar to the exact value of 0.375F.

6861X_C001.indd 56

2/7/08 [Link] PM

Trusses
AU: Incomplete text.

57

If the buckling load Pcr was the same for the two trusses, the squat truss would be weaker than the tall truss. However, the buckling load is different. It is determined using Equation 1.7.1. For the squat truss in Figure1.7.6: Pcr =

2 Eh 4 3.14 2 1500 10 3 (1/8)4 = = 6.3 lbf 12 L2 12 6.92

For the tall truss in Figure1.7.7: Pcr =

2 Eh 4 3.14 2 1500 10 3 (1/8)4 = = 3.1 lbf 12 L2 12 9.82

The strengths of the trusses are For the squat truss in Figure1.7.6: F = 2.3 6.3 = 14.5 lbf For the tall truss in Figure1.7.7: F = 3.3 3.1 = 10.2 lbf So, the taller truss is weaker for the properties we have assumed. On the other hand, if we reduce the dimension a to 0.8 in. and maintain b = c = 4 in. (Figure1.7.8), Equations 1.7.1 and 1.7.6 lead to Pcr = 9.2 lbf and F = 0.56 9.2 = 5.2 lbf , indicating that a very shallow truss with the same base dimensions is even weaker than the tall truss. If you choose the sizes a, b, and c properly to satisfy Equation 1.7.6, you can design a truss that would fail at a prescribed load, F. We recommend that you use c = 10 in. and b equal to a dimension of your choice from 6 to 10 in. so that the class produces various shapes of trusses. The height a shall be determined to be more than 8 in., because a shallow truss is difficult to construct. To determine a, you may use a spreadsheet such as Excel for the calculation. An example is shown in Figure1.7.9, where SQRT means a square root and A2^2 indicates the square of the value in cell A2. Increase the value of a from 8 until you get the prescribed load F.

AU: Change OK?

A a = 8 in. C D 2c = 8 in.

B E 2b = 8 in.

Figure1.7.8 Very shallow truss.

6861X_C001.indd 57

2/7/08 [Link] PM

58

Understanding Structures: An Introduction to Structural Analysis

C a = 0.8 in.

D 2c = 8 in.

B E 2b = 8 in.

Figure1.7.9 Spreadsheet data.

Do not use the ties as shown in Figure1.7.10a because the wood members bend inward before buckling occurs. Tie them as close to their ends as possible, as shown in Figure1.7.10b. Small cuts such as those shown in Figure1.7.10c will help you to tie the string firmly. Tie node B to E first and node A last so that the strings stretch naturally. At first, the clay weight should be about half of the target weight. Add to the clay weight until the truss collapses, then weigh the clay. The weight may differ from your expectation because 1. Youngs modulus of the wood differs depending on the kind of wood. 2. If you tie node A too firmly, the wood member may not buckle, as shown in Figure1.7.3, and the buckling load will increase. 3. If the wood member was not straight but curved to start with, the buckling load will decrease.
AU: Shouldnt this be an exercise?

Write a report on your hands-on experiences: (a) State the intended and measured strengths of the truss. (b) State possible reasons for the difference, if any. (c) Determine the axial forces in the inclined members at collapse using L F. 4a (d) Back-figure Youngs modulus of the wood ignoring the reasons stated in (b) and using Equation 1.7.1. P=
=3.14^2*1500*10^3*(1/8)^4/(12*D2^2) 1 2 3 A a 8 9 =A2 + 1.0 B b 8 8 C c 10 10 D L 15.1 15.7 E Pcr 1.32 1.23 F F 2.80 2.82

=SQRT(A2^2 + B2^2 + C2^2)

Figure1.7.10 Tie the wood members near their ends first and the top node last.

6861X_C001.indd 58

2/7/08 [Link] PM

Trusses

59

A Tie here rst C, D B, E C, D

Tie here last

B, E (b) Good example (c) Cut

(a) Bad example

Figure1.8.1 Make nodes.

1.8 Analysis of Statically Determinate Trusses

Start GOYA-A. It will help you build a truss with any arbitrary shape. For example, let us build the one shown in Figure1.6.1. 1. Push the Zoom In button to have a close-up view of the canvas. Push the +node button, and click on the canvas to locate the nodes as shown in Figure1.8.1. If you click in error, push the -> node button and drag the node. You may use the -node button to remove a node. 2. Push the +member button, and drag between nodes to create members as shown in Figure1.8.2. 3. Push the +support button, check Fix Y in a new window, and click the lower left node to make a roller support that restrains movement in direction Y. 4. Check Fix X and click the lower right node to make a pin support that restrains movement in directions X and Y as shown in Figure1.8.3. 5. Push the +Load button, and type selected values (in this case, 0 for X and 100 for Y). Click the upper node to show an arrow indicating the force. 6. Push the Analyze button to get the result shown in Figure1.8.4. The numbers indicate the axial forces, positive in tension. Colors also indicate axial forcesblue for tension and red for compression. 7. The size of each square on the screen is assumed to be 10 10 mm. Each member is assumed to have a cross-sectional area of 9 mm2 and Youngs modulus of 10,000 N/mm2. The deformation is amplified ten times. If you move the Amplification bar, you can change the amplification factor. If you push the *Member button and drag from one end of a member to another, you can see the cross-sectional area of the member and change it if you want. As another example, let us build the truss shown in Figure1.8.5. When you draw the members, do not connect A to D directly, but connect AC and CD separately
 Roller supports are represented by ( ) rather than ( ) for brevity.

6861X_C001.indd 59

2/7/08 [Link] PM

60

Understanding Structures: An Introduction to Structural Analysis

Figure1.8.2 Create members.

Figure1.8.3 Make supports.

Figure1.8.4 Result.

6861X_C001.indd 60

2/7/08 [Link] PM

Trusses

61

Figure1.8.5 Truss with five members.

so that your computer distinguishes AC from CD. Figure1.8.6 shows the resulting truss. Note that the right node (D) moves to the right, not to the left. Can you explain why? To understand why D moves to the right, let us begin with the reactions. The reaction of the pin support, A, may have x and y components, R Ax and R Ay, whereas that of the roller support has only a y component, R Dy. First, we consider equilibrium of moments at node A to obtain the equation 100 L RDy 2 L = 0

which leads to R Dy = 50 N. Second, equilibrium in the vertical direction requires 100 + RAy + RDy = 0

L A RAx RAy L C 100 N L D

RDy

Figure1.8.6 Result of GOYA-A.

6861X_C001.indd 61

2/7/08 [Link] PM

62

Understanding Structures: An Introduction to Structural Analysis

70.7

70.7

50

50

Figure1.8.7 Equilibrium at A.

which leads to R Ay = 50 N. Third, equilibrium in the horizontal direction leads to R Ax = 0. Note that three equilibrium conditions (moment and forces in x- and y-directions) are sufficient to determine the three reactions of a truss supported by a pin and a roller. Next, we calculate the axial forces considering equilibrium of forces at each node. Let us begin with node A, where only two members intersect. Equilibrium conditions are shown in Figure1.8.7, leading to PAB = 50 2 N (compression) and PAC = 50 N (tension). Equilibrium conditions at node C are shown in Figure1.8.8 leading to PBC = 50 N (tension) and PCD = 50 N (tension). We note that both horizontal members AC and CD are in tension. They lengthen. Point A is restrained in the horizontal direction. Point D, restrained in the vertical direction, has to move to the right. Exercise 1.8.1
Calculate the axial force in member BC of the truss shown in Figure1.8.9.

Solution
For this problem, all we need to do is to consider equilibrium at node C as shown in Figure1.8.10a. From this we conclude that the axial force in member BC is zero. Incidentally, the reaction at the node A is 50 N, and equilibrium at the node is the same as in Figure1.8.7. The result from GOYA is shown in Figure1.8.10b, where the horizontal displacement of node D is the same as that in Figure1.8.6 but the vertical displacement of node C is smaller because member BC does not elongate.

PAB PAC 50 N (a) Free body PAB PAC (b) Equilibrium

50 N

Figure1.8.8 Equilibrium at C.

6861X_C001.indd 62

2/7/08 [Link] PM

Trusses

63

PBC

PAC

PCD

100 N

Figure1.8.9 Truss with a load at B.

Exercise 1.8.2
AU: Insertion OK?

Calculate the axial force in member BC (see Figure1.8.11).

Solution
Note that there is no horizontal reaction at either support. We conclude that the horizontal reaction at A must be equal to that at E because there is no horizontal restraint provided by the support at E, and equilibrium of horizontal forces for the entire truss is considered as a free body. Recognizing symmetry, we obtain the vertical reaction at the support A as 50 N. Equilibrium at node A is shown in Figure1.8.12, which enables us to determine the axial force in member AB, PAB = 50 2 N (compression). Using this value, we consider equilibrium at node B (Figure1.8.13) and conclude that PBC = 50 2 N (tension).

AU: Change OK?

Exercise 1.8.3
Calculate the axial force in member CE of the truss shown in Figure1.18.4.

100 N B

L A L C L D

Figure1.8.10 Solution to Exercise 1.8.1.

6861X_C001.indd 63

2/7/08 [Link] PM

64

Understanding Structures: An Introduction to Structural Analysis

70.7 0 PAC C (a) Equilibrium at C PCD 50.0

70.7 50.0

(b) Result of GOYA-A

Figure1.8.11 Truss of seven members.

B C E 100 N RAy L L L L REy D L

Figure1.8.12 Equilibrium at A.

PAB PAC 50 N (a) Free Body PAB PAC (b) Equilibrium

50 N

Figure1.8.13 Equilibrium at B.
PBD

PBD PBC 502 N

PBC 502 N (a) Free body

(b) Equilibrium

Figure1.8.14 Truss with a horizontal load.

6861X_C001.indd 64

2/7/08 [Link] PM

Trusses
F B D C L E L L

65

A L L

Figure1.8.15 Reactions.

Solution
We note the support restraints in Figure1.8.15 and observe that support E can provide a horizontal reaction. Equilibrium in the horizontal direction leads to F + REx = 0, which yields REx = F , where the negative sign indicates that the reaction is to the left (opposite to the direction shown in Figure1.8.15). Equilibrium of moment about support A leads to FL + REy 4 L = 0 , which yields REy = F /4. Using these values, we set up the force equilibrium at node E (Figure1.8.16), which yields PCE = 3P/4 (compression).

AU: Change OK?

Exercise Using GOYA-A

Solve the trusses of Figures1.8.11 and 1.8.14 using GOYA-A. Sketch the shapes of the trusses before and after deformation using black and red pens, respectively. State the reasons for the following results: (a) Node B moves to the lower right for the case described in Figure1.8.11. (b) Node D moves to the upper right for the case described in Figure1.8.14. Up to this point, we isolated each node and considered its equilibrium to determine the forces in the truss members. This procedure, called the method of joints, is inefficient if we want to know the axial forces in only a few members. There is another procedure called the method of sections, which makes use of an important condition: if a structure is in equilibrium, all of its parts must be in equilibrium. The method is simple: divide a structure into two parts and consider equilibrium in one of the parts. An example follows.
AU: Change OK?

B C

D L E REy REx

RAy L L L L

Figure1.8.16 Equilibrium at E.

6861X_C001.indd 65

2/7/08 [Link] PM

66

Understanding Structures: An Introduction to Structural Analysis

PDE F PCE F/4 (a) Free body F PDE PCE F/4 (b) Equilibrium F/4

Figure1.8.17 Truss of 21 members.

Exercise 1.8.4
Calculate the axial forces in members AB and CD of the truss shown in Figure1.18.17.

Solution
Note that, for the loading and support conditions given, there is no horizontal reaction at either support. Considering symmetry, we obtain the vertical reaction at each support as 50 N (see Figure1.8.18). Apply an imaginary cut at the broken line in Figure1.8.18 and consider the free-body diagram shown in Figure1.8.19, which includes three unknown axial forces, PAB, PCB, and PCD. Recall that moments must balance around any point. If we consider equilibrium around point C, the contributions of PCB and PCD disappear, and we obtain

M = 50 2 20 1 + P Y = 50 20 20 + P X = P
1m 1m A

AB

1 = 0

(Positive clockwise)

which leads to PAB = 80 N (compression). Equilibrium in the vertical direction leads to

CB

sin 45 = 0

(Positive upward)

which yields PCB = 10 2 N (compression). Equilibrium in the horizontal direction leads to

AB

+ PCB sin 45 + PCD = 0

1m B 1m 1m

(Positive right)

1m

C 20 N 20 N

D 20 N 20 N 20 N

1m

Figure1.8.18 Reactions.

6861X_C001.indd 66

2/7/08 [Link] PM

Trusses
1m 1m A 1m B 1m C 20 N 50 N 20 N D 20 N 20 N 20 N 50 N 1m 1m 1m

67

Figure1.8.19 Free body.

which yields PCD = 90 N (tension). Results obtained using GOYA-A are shown in Figure1.8.20. Note that all bottom members are in tension and all top members are in compression. The elongation of the bottom chord (bottom members) and the shortening of the top chord result in downward deflection of the truss. Note that we can have only three equations based on the conditions of equilibrium at a section: X = 0, Y = 0, and M = 0. Thus, we should cut a truss so that the section does not intersect more than three members (Figure1.8.21).

1m

1m A PAB

PCB C 20 N 20 N PCD

1m

50 N

Figure1.8.20 Result of GOYA-A.

70.7 50

50 50 42.4 30 80

80

80

50 42.450 80

70.7 50

14.1 20 14.1 130 90 90

Figure1.8.21 Do not cut a section similar to the one shown!

6861X_C001.indd 67

2/7/08 [Link] PM

68

Understanding Structures: An Introduction to Structural Analysis

A C 20 N 50 N 20 N 20 N 20 N 20 N 50 N B

Figure1.8.22

Exercise
Calculate the axial force in member AB in the truss (Figure1.8.18) for the following cases using the method of sections. (a) Load of 100 N is applied at node D (and no load at the other nodes). (b) Load of 100 N is applied at node C (and no load at the other nodes).

Determine the required forces using GOYA-A. Sketch the shapes of the trusses before and after deformation, using black and red pens, respectively. State the differAU: Figure ence between these results and those in Figure1.8.22. 1.8.22 not cited text. TenEven if you cannot find a good node to set up the equation defining equilibrium in tatively cited here. OK? of moments, do not worry. As long as the cut section does not include more than three unknown axial forces, you can write three equations (X = Y = M = 0), which enable you to obtain the solutions.

1.9 Stable trusses

As stated in Section 1.1, trusses were invented to support heavy loads using small amounts of material. However, if its design is incorrect, a truss can fail at a load smaller than that for which it was proportioned. An example is shown in Figure1.9.1a. If you push node A or C, the truss will easily deform to the shape indicated by the broken lines. Such a truss is called an
We know the reactions. So, lets divide the truss.
Sozen Inc.

Find a good point for equilibrium of moments.

Figure1.9.1 Unstable trusses with f = 1.

6861X_C001.indd 68

2/7/08 [Link] PM

Trusses

69

unstable truss. Trusses in Figure1.9.1bd are other examples. We shall propose a truss contest in Section 1.10. When you take on that challenge, you should be careful about the stability of the truss. If you try to analyze an unstable truss in GOYA-A, your computer will sound an alarm. In the 19th century, a German mathematician, A. F. Mobius (well known for the Mobius strip), found that a truss is unstable if the freedom of motion, f in the following equation, is larger than zero. f = 2n m r (1.9.1)

where n is the number of nodes, m is the number of members, and r is the number of restraints imposed by the supports (pins and rollers). For example, r for the truss in Fig, 1.9.1a is 3 because the pin support restrains node B from moving in the x- and y-directions, and the roller support restrains node C from moving in the y-direction. The freedom of motion, determined from Equation 1.9.1 for the truss in Figure1.9.1a, is f = 2 3 2 3 = 1 > 0 . It is unstable. The trusses in Figure1.9.1bd also have f = 1 and are unstable. The degrees of freedom actually represent the possible motions of the truss nodes that do not require deformation (lengthening or shortening) of its members. In Figure1.9.1a, the movement of node A is related to that of node C. In other words, the truss can move in a unique fashion. The degree of freedom would be one. If you change support B from a pin to a roller, the degree of freedom would be two. Then, the truss can move as indicated in Figure1.9.2 without any change in shape. In any case, both trusses (Figures1.9.1a and 1.9.2) cannot resist a load. If you change support C in Figure1.9.1a from a roller to a pin (i.e., increase r from 3 to 4 as shown in Figure1.9.3a), we have f = 0 and the truss is stable. The trusses in Figure1.9.3bd have one more member than those in Figure1.9.2 and are stable. The value of f (Equation 1.9.1) may be obtained in each case as follows: (a) Each node has the freedom to move in directions x and y. Thus, n nodes provide 2n degrees of freedom. The number of degrees of freedom is 2n. (b) Each member maintains the distance between two nodes constant and decreases the degrees of freedom by one. The number of degrees of freedom becomes 2n m. (c) Each restraint decreases the degree of freedom by one. Thus, r restraints decrease r degrees of freedom. The number of degrees of freedom becomes 2n m r as stated in Equation 1.9.1. Although a truss with f > 0 is unstable, a truss with f 0 is not necessarily stable (converse statements are not always true). For example, the truss in Figure1.9.4 has f = 0 but is unstable. Note that the bottom member connecting the pin supports does not work at all.

 A node of a truss is defined as the point where members intersect or are supported.

6861X_C001.indd 69

2/7/08 [Link] PM

70

Understanding Structures: An Introduction to Structural Analysis

A

C (a) n = 3, m = 2, r = 3 (b) n = 4, m = 3, r = 4

(c) n = 4, m = 3, r = 4

(d) n = 7, m = 10, r = 3

Figure1.9.2 Unstable truss with f = 2.

Figure1.9.3 Statically determinate trusses ( f = 0).

C (a) n = 3, m = 2, r = 4 (b) n = 4, m = 4, r = 4

(c) n = 4, m = 4, r = 4

(d) n = 7, m = 11, r = 3

Figure1.9.4 Unstable truss with f = 0.

6861X_C001.indd 70

2/7/08 [Link] PM

Trusses

71

Figure1.9.5 Statically indeterminate trusses ( f = 1).

There is another important rule: the number of degrees of freedom f of a statically determinate truss is zero. Recall the definition of a statically determinate truss: the axial force in each member can be calculated considering equilibrium only. The unknowns in a truss are the axial forces, m, and the reactions, r, adding to (m + r). On the other hand, equilibrium at each node in x- and y-directions leads to two equations or a total of 2n equations. In the case of a statically determinate truss, the number of unknowns (m + r) must be equal to the number of equations (2n), so that f = 2 n m r = 0 . If you increase the number of members of the trusses in Figure 1.9.3 as shown in Figure1.9.5, we have more unknowns than available equations. With the added members, the trusses are indeterminate. For three-dimensional trusses, equilibrium at each node in directions x, y, and z leads to three equations, and the degree of freedom is given by f = 3n m r (1.9.2) For example, the value of r for the truss in Figure1.9.6, supported by three pins is 9, and its degree of freedom is f = 3 4 3 9 = 0 . It is statically determinate.

C (a) n = 4, m = 3, r = 6

(b) n = 4, m = 5, r = 4

(c) n = 4, m = 5, r = 4
Figure1.9.6 Three-dimensional truss ( f = 0).

(d) n = 7, m = 12, r = 3

6861X_C001.indd 71

2/7/08 [Link] PM

72

Understanding Structures: An Introduction to Structural Analysis

Figure1.9.7 Three-dimensional rollers.

Two types of roller supports can be used in a three-dimensional truss: the one shown in Figure1.9.7a provides unrestrained movement in two directions and provides a restraint in the vertical direction only (r = 1), whereas the other roller support in Figure1.9.7b provides unrestrained movement in one direction but restrains movement in two directions (r = 2).

1.10 Building a Truss Contest Using GOYA-A

First, we shall compete for the strongest truss using GOYA-A. Assume the size of each square on the GOYA-A screen to be 10 10 mm. To build your truss, you must satisfy the following five requirements AU: Table 1.10.1 (Table1.10.1). not cited in text. 1. The truss shall be supported by a roller at one end and a pin at the other end over a span of 200 mm. 2. The load shall be applied at the highest point in the center of the span. 3. The truss shall be statically determinate ( f = 2 n m r = 0 ). 4. The sum of the lengths of compressive members shall be 360 mm or less and the number of members shall not exceed 6. 5. Each compressive member shall have a cross-sectional area of 3 mm2 and a Youngs modulus of 10,000 N/mm2. Axial force in each compressive member shall not exceed the buckling load, Pcr. Members shall be assumed not to be vulnerable to failure in tension. Table1.10.1 List of Members in Compression
No. 1 2 3 4 Total Length (mm) 104 67 67 104 342 Axial Force P (N) 57 26 26 57 Strength Pcr (N) 62 148 148 62 P/Pcr 0.92 0.18 0.18 0. 92

Tentatively cited here. OK?

AU: Insertion OK?

6861X_C001.indd 72

2/7/08 [Link] PM

Trusses

73

(a) 2-D Roller

(b) 1-D Roller

Figure1.10.1 Example.

An example of such a truss is shown in Figure1.10.1. You should be able to build a much stronger truss than the one shown. After building a truss, check the box View Buckled Members. Buckled members are shown in red. If all the members are black, the truss is safe.

Contest Using Wood and String

Let us make an actual truss that meets the following five requirements (to date, the record for the maximum load carried has been 16 lbf): 1. The truss shall be placed on two desks at 2 ft. from one another. (To make certain that it does not slip off one of the desks, the truss span should be approximately 2 ft. 2 in.) 2. The material shall comprise four pieces of wood with a cross-sectional area of 1/8 in.2 and a length of 2 ft, and a string. You may use as much string as you wish. You cannot use other materials such as adhesives. 3. The only tool you may use is a knife. 4. Before the test, you shall submit a drawing of the truss and a bill of material (lengths of wood pieces used). 5. The load should be applied at the highest point in the center of the span using a string. The following figure (Figure1.10.2) shows the winner of the contest in 2004.
80 N (1) 57 N 60 N 70 (2) 26 N 67 N 60 200 (3) 26 N (4) 57 N 60 N 70
AU: Added OK?

AU: Insertion OK?

30 30

Figure1.10.2 Winner in year 2004 (M. Hirabayasi).

6861X_C001.indd 73

2/7/08 [Link] PM

74

Understanding Structures: An Introduction to Structural Analysis

1.11Problems
1.1 Calculate the axial forces in members AB and AC in Figure1.11.1 (PAB and PAC), and select the set closest to your answer from Table1.11.1.

Figure1.11.1

1.2 Calculate the elongations of members AB and AC in Figure1.11.1 (eAB and eAC) assuming that the cross-sectional area of each member is 100 mm2 and Youngs modulus is 200 kN/mm2. Select the set closest to your answer from Table1.11.2.

Table1.11.2
eAB (mm) 1 2 3 4 5 2 3 4 5 6 eAC (mm) 6 5 4 3 2

1.3 Calculate the maximum force at node A in Figure1.11.1 assuming that the tensile strength of the material is 200 N/mm2. Select the value closest to your answer from Table1.11.3.

6861X_C001.indd 74

2/7/08 [Link] PM

Trusses

75

Table1.11.3
(kN) 1 2 3 4 5 8 13 18 23 28

1.4 Calculate the axial forces in members AC and CE in Figure1.11.2 (PAC and PCE), and select the set closest to your answer from Table1.11.4.
20 m B 10 m 60 30 C 1 2 A 10 kN 3 4 5 Table 1-11-1 PAC PAB 5 kN 6 kN 7 kN 8 kN 9 kN 9 kN 8 kN 7 kN 6 kN 5 kN

Figure1.11.2

AU: Insertion OK?

1.5 Calculate the horizontal displacement of the roller support in Figure1.11.2 assuming that the cross-sectional area of each member is 100 mm2 and Youngs modulus is 200 kN/mm2. Select the set closest to your answer from Table1.11.5.

Table1.11.5
(mm) 1 2 3 2 4 6 4 5 6 (mm) 8 10 12

1.6 Choose the correct set of axial forces in members AB, BE, DE, and FG in Figure1.11.3 from those listed in Table1.11.6, where T, C, and 0 represent tensile, compressive, and zero axial, forces, respectively.

6861X_C001.indd 75

2/7/08 [Link] PM

76

Understanding Structures: An Introduction to Structural Analysis

20 kN B A 5m 5m C 5m 5m D 5m E 1 2 3 4 5

Table 1-11-4 PAB 15 kN 14 kN 13 kN 12 kN 11 kN PAC 5 kN 6 kN 7 kN 8 kN 9 kN

Figure1.11.3

1.7 Choose the correct set of axial forces in members AB, BE, DE, and FG in Figure1.11.4 from those listed in Table1.11.7, where T, C, and 0 represent tensile, compressive, and zero axial, forces, respectively.
F D B A C E G F H

Table 1-11-6 AB 1 2 3 4 5 T C C C T BE 0 C T 0 C DE 0 C T 0 T FG 0 0 T 0 C

Figure1.11.4

1.8 Calculate the axial force in member AB in Figure1.11.5 and select the correct answer from Table1.11.8.

F F B A D F F H C E G 1 2 3 4 5

Table 1-11-7 AB T C C C T BE C 0 T C C DE T C T T T FG 0 C T 0 C

Figure1.11.5

6861X_C001.indd 76

2/7/08 [Link] PM

Trusses

77

1.9 Calculate the axial force in member AB in Figure1.11.6, and select the correct answer from Table1.11.9.
A B Table 1-11-8 1 2 C 60 60 D F 60 60 E 3 4 5 + 3 F 3 F 1 F 3 1 F 3 + 1 F 23 +

60

60

Figure1.11.6

AU: Should this be correct?

1.10 Calculate the axial forces in members A through E in Figure1.11.7. Which listing for axial force in Table1.11.10 is incorrect?
F F F F F Table 1-11-9 1 2m 2 3 4 A 1m 1m 1m 1m 1m B 1m 1m 1m 5 2F 3F 4F 5F 6F

Figure1.11.7

1.11 Calculate the axial force in member AB in Figure1.11.8 and select the correct answer from Table1.11.11.
F E A B C L D L 1 2 3 L L 4 5 F Table 1-11-10 Member Axial force A B C D E + 2 F F +F 2 F F

Figure1.11.8

6861X_C001.indd 77

2/7/08 [Link] PM

78

Understanding Structures: An Introduction to Structural Analysis

1.12 Calculate the axial force in member AB in Figure1.11.9 and select the correct answer from Table1.11.12.
F A F F F

Table 1-11-11 1 F 2 L 3 4 5 0 2.5 F +0.5 F +2.5 F +3.5 F

Figure1.11.9

1.13 Calculate the axial force in member AB in Figure1.11.10 and select the correct answer from Table1.11.13.
F Table 1-11-12 1 60 60 60 60 2 3 B 2L L 4 5 + + 9 F 53 6 F 53 0 9 F 53

60 2L

60 A

60

60

3 F

Figure1.11.10

1.14 Calculate the axial force in member AB in Figure1.11.11 and select the correct answer from Table1.11.14.
F Table 1-11-13 1 60 60 2L 60 A 2L 60 60 60 60 B L 60 2 3 4 5 + + 9 F 53 6 F 53 0 9 F 53

3 F

Figure1.11.11

6861X_C001.indd 78

2/7/08 [Link] PM