You are on page 1of 39

A New Approach

to

A New Approach
to

Design for Stability


R. Shankar Nair
Teng & Associates, Inc.

Design for Stability


as reflected in the Stability and Analysis Provisions
of the

2005 AISC Specification

What is stability? How have we been designing for it? Why do we need to change? The new approach

What is stability?

Nair

Nair

Buckling
P P

Buckling

No displacement y

No displacement y

Nair

Nair

Buckling
Pcr Pcr y P No displacement until bifurcation y Pcr

Buckling
Pcr P y

Pcr

Pcr

Pcr

Pcr

Buckling of this type is a theoretical abstraction it does not happen in real structures
Nair

Nair

Amplification
P F P No displacement y F y
Nair

Amplification

Nair

10

Amplification
P
hou tP

y0
F

Amplification
P W ith

Wit

ith

P F

F P y P

y
F P P

Wit

y0

y
Nair

11

hou tP

y0 y

Nair

12

y0
F

Amplification

y0
F

Amplification
While buckling is a theoretical abstraction that does not happen in real structures

y0 P
P F

y0
Constant
P F

F y
F P P P

y0

y
F P

y
Nair

y
13
Nair

14

y0
F

Amplification
While buckling is a theoretical abstraction that does not happen in real structures

y0
F

Amplification
y0 may be the result of either a lateral load

y0
P F

y0
P F

y
F

Amplification ( y >y0 ) is real


Nair

y
F

y
15

y
Nair

16

y0
Initial imperfection y0

Amplification
y0 may be the result of either a lateral load or an initial imperfection Buckling

y0
P

Amplification
y0 y

y
Nair

Due to lateral load or initial imperfections


17
Nair

18

Buckling Amplification
y0 y

Buckling Amplification
y0 y

Buckling is not a real phenomenon because there is always a y0


Nair

Failure of a frame or member loaded in compression always involves lateral displacement and flexure
19
Nair

20

Pure compression; no flexure


y y

Perfect column (plumb, straight)


Nair

Perfect column (plumb, straight)


21
Nair

22

P AFy

P 2EI 4L2

Pure compression; no flexure -- until failure by yield

Pure compression; no flexure -- or failure by buckling

Perfect column (plumb, straight)


Nair

Perfect column (plumb, straight)


23
Nair

24

Pure compression; no flexure -- until failure by yield or buckling Perfect column (plumb, straight)

at AFy y
2 at EI 4L2

Pure compression; no flexure -- until failure by yield or buckling Perfect column (plumb, straight)
Nair

at AFy y
2 at EI 4L2

25

Nair

26

y0 P

y0 y P P

y0

Load causes compression and flexure y0

Real column (out of plumb)


Nair

Real column (out of plumb)


27
Nair

28

y0

y P

y0 P

y P

Load causes compression and flexure y0

Real column (out of plumb)


Nair

Load causes compression y and flexure y0 the flexure increases more rapidly than the compression Real column (out of plumb)
29
Nair

30

y0

y P Pn

Failure of a real member loaded in pure compression is by: not crushing in compression (AFy)

Load causes compression y and flexure y0 the flexure increases more rapidly than the compression Real column (out of plumb) until the strength limit
Nair

31

Failure of a real member loaded in pure compression is by: not crushing in compression (AFy) not buckling (2EI/(KL)2)

Failure of a real member loaded in pure compression is by: not crushing in compression (AFy) not buckling (2EI/(KL)2) but compression & flexure

y0

y P Pn y0
Initial imperfection y0

Pn

Load causes compression y and flexure y0 the flexure increases more rapidly than the compression Real column until the strength limit (out of plumb)
Nair

Load causes compression y and flexure y0 the flexure increases more rapidly than the y compression until the strength limit
Nair

35

36

P Pn

Idealized Actual
y0 y

y0

Initial condition

Load causes compression y and flexure y0 the flexure increases more rapidly than the compression until the strength limit
Nair

Failure of a frame or member loaded in compression always involves flexure


37
Nair

38

Idealized Actual

Capturing actual response requires second-order analysis

y0

2nd-order analysis
P

Linear analysis

y0

Failure of a frame or member loaded in compression always involves flexure


Nair

y0
(due to lateral load)

39

Nair

40

y0
Initial imperfection y0

P
No lateral load Neglect initial imperfections

2nd-order analysis*
P

2nd-order analysis
P

Linear analysis*

Linear analysis 0

y0
(due to initial imperfection)

y
*with imperfection included in the analysis
Nair

41

Nair

42

Second-order analysis Linear analysis


0 y0 y 0 y0

Second-order CORRECT analysis Linear analysis


y

Nair

43

Nair

44

For correct prediction of response:


Second-order CORRECT analysis Linear analysis
0 y0 y

Second-order analysis Correct y0 (including imperfections)

Correct prediction of response also requires correct y0 ... including initial imperfections
Nair

45

Nair

46

For correct prediction of response: Second-order analysis Correct y0 (including imperfections)


Pcr

Useful even if notreal

Pcr y P y

The second-order analysis can be either: Direct second-order analysis or Amplified first-order analysis
Nair

Pcr

Pcr

47

Nair

48

y0
F

y0
F

y0
P F

2nd-order analysis 1st-order analysis

y0
P

2nd-order analysis 1st-order analysis

y0 y
F

y0 y
F

y = y0 /(1 P/Pcr)
P

y
Nair

y
49
Nair

50

y0
F

Amplified first-order analysis


P
2nd-order analysis 1st-order analysis

y0
F

Amplified first-order analysis This is the familiar B1- B2 method of 2nd-order analysis

y0
P F

y0
P

y0 y
F

y
F

y = y0 /(1 P/Pcr)
P P

y = y0 /(1 P/Pcr)
P

y
Nair

y
51
Nair

52

y0
F

Amplified first-order analysis This is the familiar B1- B2 method of 2nd-order analysis B1member B2story of structure

Second-order analysis (direct or indirect) is not enough

y0
P F

y0

y
F

y = y0 /(1 P/Pcr)
P

Due to lateral load or initial imperfections


53
Nair

y
Nair

54

Second-order analysis (direct or indirect) is not enough

Second-order analysis (direct or indirect) is not enough

must also have correct y0


y0 y

must also have correct y0 must also adjust for stiffness reduction due to inelasticity
Nair

y0

Due to lateral load or initial imperfections


55

Due to lateral load or initial imperfections


Nair

56

Second-order analysis (direct or indirect) is not enough

Typical residual stress in a rolled section = 0.3 Fy

Second-order analysis (direct or indirect) is not enough

must also have correct y0 must also adjust for stiffness reduction due to inelasticity
Nair

must also have correct y0 must also adjust for stiffness reduction due to inelasticity
57

Typical residual stress in a rolled section = 0.3 Fy part of section yields when stress due axial force & flexure reaches 0.7 Fy

Nair

58

Second-order analysis (direct or indirect) is not enough

must also have correct y0 must also adjust for stiffness reduction due to inelasticity

Typical residual stress in a rolled section = 0.3 Fy part of section yields when stress due axial force & flexure reaches 0.7 Fy then stiffness decreases

Compression Member Strength


P

For perfect member (no imperfection, no residual stress):

Pn = lesser of

2EI/L2 AFy

P
Nair

59

Nair

60

10

Compression Member Strength


Yielding Perfect member no imperfections or residual stresses

Compression Member Strength


Yielding Adjusted for imperfections & inelasticity

AFy Pn

Buckling

AFy Pn

Buckling

L/r
Nair

L/r
61
Nair

62

Compression Member Strength


Yielding Adjusted for imperfections & inelasticity This is the compressive strength for flexural buckling in the Specification (Chapter E)

Compression Member Strength


Yielding Adjusted for imperfections & inelasticity

AFy Pn

Buckling

AFy Pn

Buckling

Failure by compression and flexure NOT BUCKLING L/r


Nair

L/r
Nair

63

64

P Pn

Compression Member Strength


Yielding Adjusted for imperfections & inelasticity

AFy
y y0

Buckling

Load causes compression y and flexure y0 the flexure increases more rapidly than the compression until the strength limit
P
Nair

Pn

Failure by compression and flexure NOT BUCKLING L/r


Nair

65

66

11

Conventional design approach (pre-2005)

How have we been designing for stability?

Find P & M in members from secondorder analysis* neglecting imperfections and inelasticity

*The 2nd-order analysis may consist of either direct 2nd-order analysis or 1st-order analysis adjusted by B1 and B2

Nair

67

Nair

68

Conventional design approach (pre-2005)


Find P & M in members from secondorder analysis* neglecting imperfections and inelasticity Check member capacity using column curve (strength equation) that includes effects of imperfections and inelasticity
*The 2nd-order analysis may consist of either direct 2nd-order analysis or 1st-order analysis adjusted by B1 and B2

Compression Member Strength


Yielding

AFy Pn

Buckling

Adjusted for imperfections & inelasticity (in Specification)

L/r
Nair

69

Nair

70

Conventional design approach (pre-2005)


Find P & M in members from secondorder analysis* neglecting imperfections and inelasticity Check member capacity using column curve (strength equation) that includes effects of imperfections and inelasticity
*The 2nd-order analysis may consist of either direct 2nd-order analysis or 1st-order analysis adjusted by B1 and B2

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Member P&M

Nair

71

Nair

72

12

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Include imperfections & inelasticity in member capacity calculations

Member P&M

Member Capacity

Member P&M

Member Capacity

Do these compensate?
Nair

73

Nair

74

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Include imperfections & inelasticity in member capacity calculations

Member P&M

Member Capacity

Do these compensate? They do not.


Nair

Member P&M This error causes understatement of member demand


75

Member Capacity

Nair

76

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Include imperfections & inelasticity in member capacity calculations

Member P&M This error causes understatement of member demand

Member Capacity Gives correct member capacity


Nair

Member Member P&M Capacity This error causes Important only in understatement of structures subject to member demand sidesway
77
Nair

78

13

y0
Initial imperfection y0

Structure braced against sidesway y0 will not greatly affect P & M in frame members

y0
Initial imperfection y0

Structure braced against sidesway y0 will not greatly affect P & M in frame members

y0
Initial imperfection y0

Structure not braced against sidesway y0 will significantly affect P & M in frame members

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway

Member Capacity

Member P&M
understates member demand in structures subject to sidesway
Nair

Member Capacity Compensate by using K>1


Nair

81

82

Conventional design approach (pre-2005)


P

Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Consider a cantilever column

Member P&M
understates member demand in structures subject to sidesway

Member Capacity Compensate by using K>1


Nair

83

Nair

84

14

Conventional design approach (pre-2005)


P

Consider a cantilever column apply conventional analysis/design approach

Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway
Nair

Member Capacity Compensate by using K>1


Nair

85

86

Second-order analysis indicates: Axial force = P Base moment = 0

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection)

Nair

87

Nair

88

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2

Compression Member Strength


Yielding

AFy Pn

Buckling

Adjusted for imperfections & inelasticity (in Specification)

P
Nair

L/r
89
Nair

90

15

Compression Member Strength


Yielding

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2

AFy Pn

Buckling

y0

initial

L/r
Nair

P
91
Nair

92

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2 initial

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2 same result as including y0 in the analysis

y0

initial

y0

initial

y0

P
Nair

P
93
Nair

94

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2 same result as including y0 in the analysis initial L 500

Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2 same result as including y0 in the analysis

initial

y0

y0

initial 2L 1000

y0

initial 2L 1000

y0

P
Nair

P
95
Nair

96

16

Conventional design approach (pre-2005)


P P
Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equation (which includes effect of initial imperfection) with K=2 same result as including y0 in the analysis

initial L 500

initial 2L 1000

y0

Member P&M
understates member demand in structures subject to sidesway
Nair

2nd-ORDER ANALYSIS

y0

Applied Loads
Neglect imperfections & inelasticity in analysis

Include imperfections & inelasticity in member capacity calculations

Member Capacity Compensate by using K>1


Nair

e sam

P
97

98

Conventional design approach (pre-2005)


P
Second-order analysis indicates: Axial force = P Base moment = 0 Design member for force P and zero moment, using column strength equations that include effect of initial imperfection, with K=2

appears to work, but there are problems

Nair

99

Nair

100

y0

y P

Second-order analysis indicates: Axial force = P Base moment = 0

Fails by compression and flexure Real column (out of plumb)

y0

the conventional design process does not indicate the correct moment at the base

Nair

101

Nair

102

17

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

This process gives correct column sizes

Member P&M
Understates moment

Member Capacity Compensate in column design by using K>1


Nair

Member P&M
Understates moment
103

Member Capacity Compensate in column design by using K>1


Nair

104

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)

This process gives correct column sizes but under-designed beams and connections

indicates incorrect design moments

Member P&M
Understates moment

Member Capacity Compensate in column design by using K>1


Nair

105

Nair

106

Conventional design approach (pre-2005)

Conventional design approach (pre-2005)

indicates incorrect design moments requires calculation of K--

indicates incorrect design moments requires calculation of K-how??

Nair

107

Nair

108

18

Sidesway uninhibited alignment chart for column effective length

Sidesway uninhibited alignment chart for column effective length The Commentary to the AISC Specification lists nine conditions for the applicability of this chart.

Nair

109

Nair

110

Sidesway uninhibited alignment chart for column effective length The Commentary to the AISC Specification lists nine conditions for the applicability of this chart. 8. All columns buckle simultaneously.

Sidesway uninhibited alignment chart for column effective length The Commentary to the AISC Specification lists nine conditions for the applicability of this chart. 8. All columns buckle simultaneously.
Ratio of column force P to

2EI (KL)2

must be same for all columns in a story


Nair

111

Nair

112

P/2

P/2

I/2

I/2

OK for alignment chart

P/2

P/2

I/2

I/2

OK for alignment chart

P/2

P/2

I/2

I/2

NG for alignment chart

Nair

113

Nair

114

19

P/2

P/2

I/2

I/2

I/2

I/2

NG for alignment chart

Moment-resisting frame

Moment-resisting frame
Nair

115

Moment-resisting frame

OK for alignment chart

Moment-resisting frame

NG for alignment chart

Nair

116

Moment-resisting frame Moment-resisting frame Moment-resisting frame

Moment-resisting frame
Nair

Brace

117

Moment-resisting frame

NG for alignment chart

Moment-resisting frame

Forget about it !!

Brace

Nair

118

Sidesway uninhibited alignment chart for column effective length The Commentary to the AISC Specification lists conditions required for the applicability of these charts. Few real-world buildings meet the conditions for applicability of the alignment chart.

Sidesway uninhibited alignment chart for column effective length

Nair

119

Nair

120

20

Conventional design approach (pre-2005) indicates incorrect design moments requires calculation of K-how ?

Conventional design approach (pre-2005) indicates incorrect design moments requires calculation of K-how indeed?

Nair

121

Nair

122

Conventional design approach


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway

Member Capacity Compensate by using K>1


Nair

Member P&M
understates member demand in structures subject to sidesway
123

Member Capacity Compensate by using K>1


Nair

124

Conventional design approach (pre-2005) indicates incorrect design moments requires calculation of K-how? sometimes, too conservative

Conventional design approach (pre-2005) indicates incorrect design moments requires calculation of K-how? sometimes, too conservative

Nair

125

Nair

126

21

Conventional design approach (pre-2005)


Applied Loads
Neglect imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Conventional design approach (pre-2005)


Applied Loads
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Include

Neglect imperfections & inelasticity in analysis

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway

Member Capacity Compensate by using K>1


Nair

Member P&M
understates member demand in structures subject to sidesway
127

Member Capacity Compensate by using K>1

K=1

Nair

128

Conventional design approach (pre-2005) indicates incorrect design moments requires calculation of K-how? sometimes, too conservative
Applied Loads
2nd-ORDER ANALYSIS

Include

Neglect imperfections & inelasticity in analysis

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway
Nair

Member Capacity Compensate by using K>1

129

K=1

Nair

130

Conventional design approach (pre-2005) Direct Analysis Method Include


Applied Loads
2nd-ORDER ANALYSIS

Direct Analysis Method


Applied Loads
Include imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Neglect imperfections & inelasticity in analysis

Include imperfections & inelasticity in member capacity calculations

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway

Member Capacity Compensate by using K>1

Member P&M

Member Capacity
Use K=1

K=1

Nair

131

Nair

132

22

Stability and Analysis Provisions


of the

CHAPTER B DESIGN REQUIREMENTS B1. GENERAL PROVISIONS The design of members and connections shall be consistent with the intended behavior and the assumptions made in the structural analysis.

2005 AISC Specification for Steel Buildings

Nair

133

Nair

134

CHAPTER C STABILITY ANALYSIS AND DESIGN

CHAPTER C STABILITY ANALYSIS AND DESIGN

specifies that the design of the structure for stability must consider various effects.

Nair

135

Nair

136

Effects to be considered: Flexural, shear and axial deformations of members

Effects to be considered: Member deformations All component and connection deformations that contribute to lateral displacement

Nair

137

Nair

138

23

Effects to be considered: Member deformations Other deformations P- effects

Effects to be considered: Member deformations Other deformations P- effects P- effects


Nair

139

Nair

140

Effects to be considered: Member deformations Other deformations P- effects P- effects Geometric imperfections member shape imperfections (out-of-straightness) node position imperfections (out-of-plumbness)
Nair

Effects to be considered: Member deformations Other deformations P- effects P- effects Geometric imperfections Stiffness reductions due to residual stresses

141

Nair

142

Effects to be considered: Member deformations Other deformations P- effects P- effects Geometric imperfections Residual stresses When the analysis has considered all these effects, members can be designed using the provisions for individual members (Chapters D, E, F, G, H, I)
Nair

Effects to be considered: Member deformations Other deformations P- effects P- effects Geometric imperfections Residual stresses Any method of analysis and design that considers all these effects is permissible

When the analysis has considered all these effects, members can be designed using the provisions for individual members (Chapters D, E, F, G, H, I)
143
Nair

144

24

Effects to be considered: Member deformations Other deformations P- effects P- effects Geometric imperfections Residual stresses General analysis requirements

Effects to be considered: Member deformations Other deformations P- effects P- effects Geometric imperfections Residual stresses

Spec presents approaches that consider these effects

When the analysis has considered all these effects, members can be designed using the provisions for individual members (Chapters D, E, F, G, H, I)
Nair

When the analysis has considered all these effects, members can be designed using the provisions for individual members (Chapters D, E, F, G, H, I)
145
Nair

146

APPENDIX 7 DIRECT ANALYSIS METHOD

APPENDIX 7 DIRECT ANALYSIS METHOD


Applicable to all types of structures Does not distinguish between Braced frames Moment frames Shear wall systems Combinations
147 148

Nair

Nair

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses

P- effects P- effects Geometric imperfections Residual stresses Use second-order elastic analysis that considers both P- and P- effects.

Nair

149

Nair

150

25

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Use second-order elastic analysis that considers both P- and P- effects. Options: any general second-order analysis method

P- effects P- effects Geometric imperfections Residual stresses Use second-order elastic analysis that considers both P- and P- effects. Options: any general second-order analysis method amplified first-order analysis (B1 & B2 )

Nair

151

Nair

152

DIRECT ANALYSIS METHOD

First-order analysis amplified by B1 and B2 is Second-order analysis

P- effects P- effects Geometric imperfections Residual stresses Use second-order elastic analysis that considers both P- and P- effects. Options: any general second-order analysis method amplified first-order analysis (B1 & B2 )

Nair

153

Nair

154

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Use second-order elastic analysis that considers both P- and P- effects. Options: any general second-order analysis method amplified first-order analysis (B1 & B2 ) Exception: P- effects can be neglected when member axial loads are below a specified level.
Nair

P- effects P- effects Geometric imperfections Residual stresses

155

Nair

156

26

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Apply notional loads: Ni = 0.002 Yi

N3 N2 N1

Y3 Y2 Y1

Ni = 0.002 Yi

Nair

157

Nair

158

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Apply notional loads: Ni = 0.002 Yi The 0.002Yi notional load is equivalent to an initial out-of-plumbness of 1/500

N3 N2 N1 Ni = 0.002 Yi

Y3 Y2 Y1

1 500

Y3 Y2 Y1 Equivalent structure

Nair

159

Nair

160

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Apply notional loads: Ni = 0.002 Yi The 0.002 coefficient corresponds to an initial out-of-plumbness of 1/500 lower value can be used if justified by the actual anticipated out-of-plumbness
Nair

P- effects P- effects Geometric imperfections Residual stresses Apply notional loads Logically, these notional loads should be additive to other lateral loads in all cases.

161

Nair

162

27

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Apply notional loads Logically, these notional loads should be additive to other lateral loads in all cases. But in a concession to past practice, the notional loads can be treated as a minimum lateral load when /0 < 1.5
Nair

P- effects P- effects Geometric imperfections Residual stresses Alternative: Instead of applying notional loads to account for geometric imperfections the designer may model imperfections directly in the analysis.

163

Nair

164

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses

P- effects P- effects Geometric imperfections Residual stresses Use reduced flexural and axial stiffnesses EI* and EA* in the analysis

Nair

165

Nair

166

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Use reduced flexural and axial stiffnesses EI* and EA* in the analysis EI* = 0.8 b EI EA* = 0.8 EA

P- effects P- effects Geometric imperfections Residual stresses Use reduced flexural and axial stiffnesses EI* and EA* in the analysis EI* = 0.8 b EI EA* = 0.8 EA

b depends on level of axial stress in member; b = 1 when P < 0.5 Py


Nair

167

Nair

168

28

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

P- effects P- effects Geometric imperfections Residual stresses Use reduced flexural and axial stiffnesses EI* and EA* in the analysis EI* = 0.8 b EI EA* = 0.8 EA

P- effects P- effects Geometric imperfections The EI and EA adjustments also Residual stresses account for the effects of member out-of-straightness on the response of the overall structure

b can be taken as 1 in all cases if additional


notional load of 0.001Yi is applied.
Nair

169

Nair

170

DIRECT ANALYSIS METHOD

DIRECT ANALYSIS METHOD

Design members using the provisions for individual members (Chapters D, E, F, G, H, I), with K=1 for computing compression strengths

Member P&M
1Notional

2nd-ORDER ANALYSIS Rigorous or B1 & B2

P- effects P- effects Geometric imperfections Residual stresses

Applied Loads
Include imperfections1 & inelasticity2 in analysis

Include imperfections3 & inelasticity3 in member capacity calculations

Member Capacity
3The

K=1

2Reduced
Nair

loads or direct modeling EI & EA

usual column strength formulas


Nair

171

172

APPENDIX 7 DIRECT ANALYSIS METHOD


Applicable to all types of structures Does not distinguish between Braced frames Moment frames Shear wall systems Combinations
173

APPENDIX 7 DIRECT ANALYSIS METHOD

MAY be used in all cases MUST be used when /0 > 1.5

Nair

Nair

174

29

APPENDIX 7 DIRECT ANALYSIS METHOD


C2. MAY be used in all cases MUST be used when /0 > 1.5 When /0 < 1.5 there are other options 2a. 2b.
Nair

CHAPTER C STABILITY ANALYSIS AND DESIGN


CACULATION OF REQUIRED STRENGTHS Design Requirements DIRECT ANALYSIS METHOD Can use Appendix 7 in all cases Can use C2.2a or C2.2b if /0 < 1.5 EFFECTIVE Design by Second-Order Analysis LENGTH METHOD Design by First-Order Analysis FIRST-ORDER ANALYSIS METHOD
Nair

2.

175

176

EFFECTIVE LENGTH METHOD

EFFECTIVE LENGTH METHOD

Generally similar to the method in the 1999 Specification

Perform second-order elastic analysis use nominal geometry use nominal stiffness

Nair

177

Nair

178

EFFECTIVE LENGTH METHOD

EFFECTIVE LENGTH METHOD

First-order analysis amplified by B1 and B2 is Second-order analysis

Perform second-order elastic analysis use nominal geometry use nominal stiffness

Nair

179

Nair

180

30

EFFECTIVE LENGTH METHOD

EFFECTIVE LENGTH METHOD

Perform second-order elastic analysis use nominal geometry use nominal stiffness Apply notional loads, Ni = 0.002 Yi as a minimum lateral load

Perform second-order elastic analysis use nominal geometry use nominal stiffness Apply notional loads, Ni = 0.002 Yi as a minimum lateral load (NEW)

Nair

181

Nair

182

EFFECTIVE LENGTH METHOD

EFFECTIVE LENGTH METHOD

Perform second-order elastic analysis use nominal geometry use nominal stiffness Apply notional loads, Ni = 0.002 Yi as a minimum lateral load Determine K factors for columns in moment frames by sidesway buckling analysis

Perform second-order elastic analysis use nominal geometry use nominal stiffness Apply notional loads, Ni = 0.002 Yi as a minimum lateral load Determine K factors for columns in moment frames by sidesway buckling analysis. (Exception: Use K=1 if /0 < 1.1)

Nair

183

Nair

184

EFFECTIVE LENGTH METHOD

EFFECTIVE LENGTH METHOD

Apply notional loads, Ni = 0.002 Yi as a minimum lateral load Determine K factors for columns in moment frames by sidesway buckling analysis Design individual members using K from step 3 for computing compression strengths

Member P&M

2nd-ORDER ANALYSIS Rigorous or B1 & B2

Perform second-order elastic analysis use nominal geometry use nominal stiffness

Applied Loads
Include imperfections1 but neglect inelasticity in analysis

Include imperfections3 & inelasticity3 in member capacity calculations

Member Capacity
3The

K>1

1Notional

loads

usual column strength formulas


Nair

Nair

185

186

31

EFFECTIVE LENGTH METHOD

Applied Loads
Include imperfections1 but neglect inelasticity in analysis
2nd-ORDER ANALYSIS Rigorous or B1 & B2

CHAPTER C
Include imperfections3 & inelasticity3 in member capacity calculations

STABILITY ANALYSIS AND DESIGN


C2. 2. CACULATION OF REQUIRED STRENGTHS Design Requirements DIRECT ANALYSIS METHOD Can use Appendix 7 in all cases Can use C2.2a or C2.2b if /0 < 1.5 EFFECTIVE Design by Second-Order Analysis LENGTH METHOD Design by First-Order Analysis FIRST-ORDER ANALYSIS METHOD
Nair

Member P&M

Member Capacity
3The

K>1

2a. 2b.

1Notional

loads

usual column strength formulas


Nair

187

188

FIRST-ORDER ANALYSIS METHOD

FIRST-ORDER ANALYSIS METHOD

Can be used only when P < 0.5 Py in all columns whose flexural stiffness contributes to the lateral stability of the structure

Perform first-order elastic analysis use nominal geometry use nominal stiffness

Nair

189

Nair

190

FIRST-ORDER ANALYSIS METHOD

FIRST-ORDER ANALYSIS METHOD

Perform first-order elastic analysis use nominal geometry use nominal stiffness Apply additional lateral loads: Ni = 2.1(/L) Yi > 0.0042 Yi

Perform first-order elastic analysis use nominal geometry use nominal stiffness Apply additional lateral loads: Ni = 2.1(/L) Yi > 0.0042 Yi Apply B1 multiplier to full moment in all beam-columns

Nair

191

Nair

192

32

FIRST-ORDER ANALYSIS METHOD

FIRST-ORDER ANALYSIS METHOD

Perform first-order elastic analysis use nominal geometry use nominal stiffness Apply additional lateral loads: Ni = 2.1(/L) Yi > 0.0042 Yi Apply B1 multiplier to full moment in all beam-columns Design individual members using K=1 for computing compression strengths

is the Direct Analysis Method by the back door uses mathematical manipulation to get roughly the same results as the DAM for typical structures

Nair

193

Nair

194

CHAPTER C STABILITY ANALYSIS AND DESIGN


C2. 2. CACULATION OF REQUIRED STRENGTHS Design Requirements DIRECT ANALYSIS METHOD Can use Appendix 7 in all cases Can use C2.2a or C2.2b if /0 < 1.5 EFFECTIVE Design by Second-Order Analysis LENGTH METHOD Design by First-Order Analysis FIRST-ORDER ANALYSIS METHOD
Nair

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

Limitations

---

/0 <1.5 /0 <1.5 P/Py <0.5

2a. 2b.

195

Nair

196

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

Limitations Analysis

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order ---

Limitations Analysis

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order --(but apply B1 to moment in beamcolumns)

Nair

197

Nair

198

33

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

Limitations Analysis Geometry

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order Nominal Nominal Nominal ---

Limitations Analysis Geometry EI & EA

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order Nominal Nominal Nominal --Reduced Nominal Nominal

Nair

199

Nair

200

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

Limitations Analysis Geometry EI & EA Notional Load

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order Nominal Nominal Nominal --Reduced Nominal Nominal 0.002 Yi 0.002 Yi >0.0042Yi minimum* minimum additive

Limitations Analysis Geometry EI & EA

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order Nominal Nominal Nominal --Reduced Nominal Nominal

*additive when /0 >1.5

Nair

201

0.002 Yi 0.002 Yi >0.0042Yi Notional minimum* minimum additive Load K for Sidesway member K=1 K=1 buckling Pn analysis *additive when /0 >1.5

Nair

202

Conventional design approach (pre-2005)


DIRECT EFFECTIVE FIRST ANALYSIS LENGTH ORDER

Limitations Analysis Geometry EI & EA

Reduced Nominal

Nominal

0.002 Yi 0.002 Yi >0.0042Yi Notional minimum* minimum additive Load K for Sidesway member K=1 K=1 buckling Pn analysis *additive when /0 >1.5

Member P&M
understates member demand in structures subject to sidesway
Nair

2nd-ORDER ANALYSIS

/0 <1.5 /0 <1.5 P/Py <0.5 Second- SecondFirstOrder Order Order Nominal Nominal Nominal ---

Applied Loads
Neglect imperfections & inelasticity in analysis

Include imperfections & inelasticity in member capacity calculations

Member Capacity Compensate by using K>1


Nair

203

204

34

Conventional design approach (pre-2005) indicates incorrect design moments (unconservative error) requires calculation of K-how? sometimes, too conservative
205

Direct Analysis Method


Applied Loads
Include imperfections & inelasticity in analysis
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Member P&M

Member Capacity
Use K=1

Nair

Nair

206

Conventional design approach (pre-2005) Direct Analysis Method indicates incorrect design moments (unconservative error) requires calculation of K-how? sometimes, too conservative
Nair

Effective Length Method


Applied Loads
Include imperfections, neglect inelasticity
2nd-ORDER ANALYSIS

Include imperfections & inelasticity in member capacity calculations

Member P&M
understates member demand in structures subject to sidesway
207

Member Capacity Compensate by using K>1


Nair

208

Conventional design approach (pre-2005) Effective Length Method indicates incorrect design moments (unconservative error) requires calculation of K-how? sometimes, too conservative
Nair

First-Order Analysis Method is the Direct Analysis Method by the back door uses mathematical manipulation to get roughly the same results as the DAM for typical structures

209

35

First-Order Analysis Method is the Direct Analysis Method by the back door uses mathematical manipulation to get roughly the same results as the DAM for typical structures generally same benefits as DAM

First-Order Analysis Method is the Direct Analysis Method by the back door uses mathematical manipulation to get roughly the same results as the DAM for typical structures generally same benefits as DAM but do not apply by rote; learn DAM first

DIRECT ANALYSIS METHOD the

right way!

Applied Loads
Include imperfections1 & inelasticity2 in analysis
2nd-ORDER ANALYSIS Rigorous or B1 & B2

Include imperfections3 & inelasticity3 in member capacity calculations

a word about loading

Member P&M
1Notional

Member Capacity
3The

K=1

2Reduced

loads or direct modeling EI & EA

usual column strength formulas


Nair

213

W2 W1 Non-linear response indicated by 2nd-order analysis R1 R2

W2/ W1 =

W2 W1

Load

Load

R1

R2

R2/R1 > Member force or moment


Nair

Member force or moment


215
Nair

216

36

WLRFD

WLRFD

Load

Load

Design by LRFD

WASD

Design by ASD Cannot use RASD


RASD RLRFD

RLRFD

Member force or moment


Nair

Member force or moment


217
Nair

218

WLRFD WASD

WLRFD

Load

Load

Design by ASD Cannot use RASD


RASD RLRFD

WASD

Design by ASD
Design by ASD for this force will yield smaller member than design by LRFD

RASD

RLRFD

ASD process assumes that ratio of ASD & LRFD member forces is the same as ratio of ASD & LRFD applied loads

ASD process assumes that ratio of ASD & LRFD member forces is the same as ratio of ASD & LRFD applied loads

Member force or moment


Nair

Member force or moment


219
Nair

220

1.6 WASD WASD Load Multiply by 1.6

1.6 WASD

Analyze Multiply by 1.6

Load

Design by ASD

WASD

Design by ASD

R1.6 ASD

Member force or moment


Nair

Member force or moment


221
Nair

222

37

1.6 WASD WASD Load

Analyze Multiply by 1.6 Divide by 1.6 RASD R1.6 ASD


DESIGN

1.6 WASD

Analyze Multiply by 1.6 Divide by 1.6 RASD R1.6 ASD


DESIGN

Member force or moment


Nair

Load

Design by ASD

WASD

Design by ASD

Member force or moment


223
Nair

224

Members of AISC TC-10 (Stability)

Shankar Nair Chairman Greg Deierlein Vice-Chairman Bill Baker Reidar Bjorhovde Charlie Carter Shu-Jin Fang Jim Fisher Ted Galambos Larry Griffis Jerry Hajjar Todd Helwig Richard Henige Leroy Lutz Clint Rex Steve Thomas Don White Ron Ziemian
Nair

Members of AISC TC-10 (Stability)


Much of the early work on the 2005 stability provisions was done under the leadership of former TC-10 chairman Joe Yura
225

Shankar Nair Chairman Greg Deierlein Vice-Chairman Bill Baker Reidar Bjorhovde Charlie Carter Shu-Jin Fang Jim Fisher Ted Galambos Larry Griffis Jerry Hajjar Todd Helwig Richard Henige Leroy Lutz Clint Rex Steve Thomas Don White Ron Ziemian
Nair

226

Looking ahead
The paper in the 2007 NASCC proceedings (for this presentation) includes a model specification written exclusively around the Direct Analysis Method.

Looking ahead
The paper in the 2007 NASCC proceedings (for this presentation) includes a model specification written exclusively around the Direct Analysis Method. That model specification is a tentative preview of the stability section of the 2010 AISC Specification.

Nair

227

Nair

228

38

Thank you

Photo credit: Patrick Krohn

39

You might also like