Académique Documents
Professionnel Documents
Culture Documents
This program is registered with the AIA/CES for continuing This presentation is protected by US and
professional education. As such, it does not include content that International copyright laws. Reproduction,
distribution, display and use of the
may be deemed or construed to be an approval or endorsement by
presentation without written permission of the
the AIA of any material of construction or any method or manner of speaker is prohibited.
handling, using, distributing, or dealing in any material or product.
© R. Shankar Nair 2007
Questions related to specific materials, methods, and services will
be addressed at the conclusion of this presentation.
Thank you!
Nair 1 Nair 2
Nair 3 Nair 4
Nair 5 Nair 6
1
ØWhat is stability?
ØHow have we been designing for it?
ØWhy do we need to change?
What is stability?
ØThe new approach
Buckling Buckling
P P
No No
P displacement P displacement
y y
P P P P
Nair 9 Nair 10
Buckling Buckling
Pcr Pcr Pcr
Pcr P
y No y
P displacement
…until y
bifurcation Buckling of this type
y is a theoretical
Pcr Pcr Pcr Pcr
y y
abstraction …it does
not happen in real
structures
Nair 11 Nair 12
2
Amplification Amplification
P
F
No y F
P displacement
y F y
P P
y
Nair 13 Nair 14
Amplification F
y0 Amplification
P F
tP
tP
F
hou
hou
y0
P
P
F
ith
ith
Wit
Wit
y F
W
W
P
F
F y y0 y
y
P P
F
y P P
y
Nair 15 Nair 16
F
y0 Amplification F
y0 Amplification
F F While buckling
is a theoretical
y0 y0 abstraction that
P does not happen in
P Constant P real structures…
F
F F
y y y
y0
F F
P P P P
y y
Nair 17 Nair 18
3
F
y0 Amplification F
y0 Amplification
F While buckling F y0 may be the
is a theoretical result of either a
y0 abstraction that y0 lateral load …
does not happen in
P real structures… P
F Amplification F
y y
( y >y0 )
F is real F
P P P P
y y
Nair 19 Nair 20
P
y0 Amplification
Initial imperfection y0 y0 may be the Buckling
result of either a
y0 lateral load … Amplification
P or an initial
imperfection y
y0
y
Due to lateral
P P
load or initial
imperfections
y
Nair 21 Nair 22
P P
Buckling Buckling
Amplification Amplification
y y
y0 y0
Buckling is not a “real” Failure of a frame or member loaded
phenomenon because there is in compression always involves
always a y0 lateral displacement and flexure
Nair 23 Nair 24
4
P P P
Pure
compression;
no flexure
y y
P P P P
AFy π2EI
4L2
Pure Pure
compression; compression;
no flexure no flexure
-- until failure -- or failure
by yield y by buckling y
P P P P
Pure Pure
compression; compression;
no flexure no flexure
-- until failure at AFy -- until failure at AFy
by yield y by yield y
or buckling 2
or buckling 2
at π EI at π EI
4L2 4L2
Perfect column Perfect column
(plumb, straight) (plumb, straight)
Nair 29 Nair 30
5
y0 y0
P y P
P
Load causes
y
compression y
y0 and flexure y0
y0 y0
y P y P
P P
y0
y Failure of a “real” member loaded
Pn in “pure” compression is by:
P
…not crushing in compression (AFy)
Load causes
compression y
and flexure y0
…the flexure increases
more rapidly than the
Real column compression
(out of plumb) …until the strength limit
Nair 35
6
Failure of a “real” member loaded Failure of a “real” member loaded
in “pure” compression is by: in “pure” compression is by:
y0
y
Pn y0 Pn
P
Initial imperfection y0
P
P
Pn
Idealized
7
P Capturing F
“actual” y0 P
response
requires
Idealized second-order
analysis 2nd-order
Actual analysis
P Linear
analysis
y F
y0 y y
y0
Failure of a frame or member (due to
lateral load)
loaded in compression always
involves flexure
Nair 43 Nair 44
y0 P P
y y y
y0 0
(due to *with
initial imperfection
imperfection) included in
the analysis
Nair 45 Nair 46
P P
Nair 47 Nair 48
8
P
For correct prediction of response:
Ø Second-order analysis
Second-order CORRECT Ø Correct y0 (including imperfections)
analysis
Linear analysis
y
0 y0
Correct prediction of response also
requires correct y0
... including initial imperfections
Nair 49 Nair 50
The second-order analysis can be either: The second-order analysis can be either:
Ø Rigorous second-order analysis or Ø Rigorous second-order analysis or
Ø Amplified first-order analysis Ø Amplified first-order analysis
…apply multiplier of the form 1/(1-P/Pcr)
to the results of first-order analysis
Nair 51 Nair 52
9
Second-order analysis Second-order analysis
Typical residual
(either rigorous or by (either rigorous or by
P stress in a rolled
amplification) is not amplification) is not
section = 0.3 Fy
enough enough
Pn = lesser of { π2EI/L2
AFy
Pn
P
L/r
Nair 59 Nair 60
10
Compression Member Strength Compression Member Strength
This is the
“compressive
Pn Pn strength for
flexural
buckling” in the
Specification
(Chapter E)
L/r L/r
Nair 61 Nair 62
Pn
Yielding Adjusted for
imperfections &
AFy Buckling inelasticity
y y0
Load causes
Failure by compression y
Pn compression and flexure y0
and flexure …the flexure increases
NOT more rapidly than the
BUCKLING compression
L/r …until the strength limit
P
Nair 63 Nair 64
11
Conventional design approach (pre-2005) Conventional design approach (pre-2005)
ØFind P & M in members from second- ØFind P & M in members from second-
order analysis* neglecting imperfections order analysis* neglecting imperfections
and inelasticity and inelasticity
ØCheck member capacity using column
curve (strength equation) that includes
effects of imperfections and inelasticity
*The 2nd-order analysis may consist of either rigorous 2nd-order *The 2nd-order analysis may consist of either rigorous 2nd-order
analysis or 1st-order analysis adjusted by B1 and B2 analysis or 1st-order analysis adjusted by B1 and B2
Nair 67 Nair 68
2nd-ORDER
ANALYSIS
ANALYSIS
Nair 71 Nair 72
12
Conventional design approach (pre-2005) Conventional design approach (pre-2005)
2nd-ORDER
ANALYSIS
ANALYSIS
imperfections member imperfections member
& inelasticity capacity & inelasticity capacity
in analysis calculations in analysis calculations
Member Member Member Member
P&M Capacity P&M Capacity
2nd-ORDER
ANALYSIS
ANALYSIS
imperfections member imperfections member
& inelasticity capacity & inelasticity capacity
in analysis calculations in analysis calculations
Member Member Member Member
P&M Capacity P&M Capacity
This error causes This error causes Gives correct
understatement of understatement of member capacity
member demand member demand
Nair 75 Nair 76
Applied Include y0
Loads imperfections & Structure braced
against sidesway
Neglect inelasticity in
2nd-ORDER
Initial imperfection y0
imperfections member affect P & M in frame
& inelasticity capacity members
in analysis calculations
Member Member
P&M Capacity
This error causes Important only in
understatement of structures subject to
member demand sidesway
Nair 77
13
Conventional design approach (pre-2005)
y0 Applied Include
Structure braced Loads imperfections &
against sidesway
Neglect inelasticity in
2nd-ORDER
… y0 will not greatly
ANALYSIS
Initial imperfection y0
affect P & M in frame imperfections member
members & inelasticity capacity
in analysis calculations
Member Member
y0 Structure not braced P&M Capacity
against sidesway
… y0 will significantly …understates member
Initial imperfection y0 affect P & M in frame demand in structures
members subject to sidesway
Nair 80
2nd-ORDER
2nd-ORDER
ANALYSIS
ANALYSIS
P P
Consider a Consider a
cantilever column cantilever column
… apply conventional
analysis/design approach
Nair 83 Nair 84
14
Conventional design approach (pre-2005)
P Second-order analysis indicates:
Axial force = P
Applied Include Base moment = 0
Loads imperfections &
Neglect 2nd-ORDER inelasticity in
imperfections ANALYSIS member
& inelasticity capacity
in analysis calculations
Member Member
P&M Capacity
…understates member
demand in structures Compensate
subject to sidesway by using K>1
Nair 85 Nair 86
Design member for force P and Design member for force P and
zero moment, using column zero moment, using column
strength equation (which includes strength equation (which includes
effect of initial imperfection) effect of initial imperfection)
with K=2
P
Nair 87 Nair 88
Pn Pn
P
L/r L/r
Nair 89 Nair 90
15
P Second-order analysis indicates: P Second-order analysis indicates:
Axial force = P Axial force = P
Base moment = 0 Base moment = 0
Design member for force P and Design member for force P and
zero moment, using column zero moment, using column
strength equation (which includes strength equation (which includes
effect of initial imperfection) effect of initial imperfection)
with K=2 with K=2
y0
initial
P P
Nair 91 Nair 92
L
zero moment, using column zero moment, using column
strength equation (which includes strength equation (which includes
effect of initial imperfection) effect of initial imperfection)
with K=2 … same result as with K=2 … same result as
y0 y0
including y0 in the analysis including y0 in the analysis
initial initial
2L
1000
P P
Nair 93 Nair 94
2L 2L
1000 1000
P P
Nair 95 Nair 96
16
Conventional design approach (pre-2005) Conventional design approach (pre-2005)
Applied Include
Loads
Neglect 2nd-ORDER
imperfections &
inelasticity in … appears to work,
ANALYSIS
imperfections
& inelasticity
member
capacity
but there are problems
in analysis calculations
Member Member
P&M Capacity
…understates member
demand in structures Compensate
subject to sidesway by using K>1
Nair 97 Nair 98
y0
P Second-order analysis indicates: y P
Axial force = P
Base moment = 0 P
Real column
(out of plumb)
Nair 99 Nair 100
imperfections member
… the conventional design & inelasticity capacity
process does not indicate in analysis calculations
the correct moment at the Member Member
base P&M Capacity
Understates Compensate
moment in column design
by using K>1
Nair 101 Nair 102
17
Conventional design approach (pre-2005) Conventional design approach (pre-2005)
2nd-ORDER
ANALYSIS
ANALYSIS
imperfections imperfections under-designed beams
& inelasticity & inelasticity and connections
in analysis in analysis
Member Member Member Member
P&M Capacity P&M Capacity
Understates Compensate Understates Compensate
moment in column design moment in column design
by using K>1 by using K>1
Nair 103 Nair 104
… indicates incorrect
design moments
18
“Sidesway uninhibited” alignment chart “Sidesway uninhibited” alignment chart
for column effective length— for column effective length—
The Commentary to the AISC The Commentary to the AISC
Specification lists nine conditions for the Specification lists nine conditions for the
applicability of this chart. applicability of this chart.
Ø 8. All columns buckle simultaneously.
P/2 P P P/2 P P
NG for NG for
alignment alignment
I/2 I I I/2 I/2 I I I/2
chart chart
19
NG for NG for
Moment-resisting frame alignment Moment-resisting frame alignment
chart chart
Moment-resisting frame
Moment-resisting frame
Moment-resisting frame
Moment-resisting frame
Moment-resisting frame Moment-resisting frame
20
Conventional design approach (pre-2005) Conventional design approach
2nd-ORDER
ANALYSIS
imperfections member
… requires calculation of K-- & inelasticity capacity
how indeed? in analysis calculations
Member Member
P&M Capacity
…understates member
demand in structures Compensate
subject to sidesway by using K>1
Nair 121 Nair 122
imperfections member
& inelasticity capacity … requires calculation of K--
in analysis calculations how?
Member Member
P&M Capacity … sometimes,
…understates member
too conservative
demand in structures Compensate
subject to sidesway by using K>1
Nair 123 Nair 124
imperfections member
… requires calculation of K-- & inelasticity capacity
how? in analysis calculations
Member Member
… sometimes, P&M Capacity
too conservative …understates member
demand in structures Compensate
subject to sidesway by using K>1
Nair 125 Nair 126
21
Conventional design approach (pre-2005)
2nd-ORDER
ANALYSIS
ANALYSIS
imperfections member imperfections member
& inelasticity capacity & inelasticity capacity
in analysis calculations in analysis calculations
Member Member Member Member
P&M Capacity P&M Capacity
…understates member …understates member
demand in structures Compensate demand in structures Compensate
subject to sidesway by using K>1 subject to sidesway by using K>1
K=1 Nair 129
K=1 Nair 130
imperfections member
& inelasticity capacity for Steel Buildings
in analysis calculations
Member Member
P&M Capacity
Use K=1
22
CHAPTER B CHAPTER C
DESIGN REQUIREMENTS STABILITY ANALYSIS AND DESIGN
23
Effects to be considered: Effects to be considered:
Ø Member deformations Ø Member deformations
Ø Other deformations Ø Other deformations
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections
…member shape imperfections
δ (out-of-straightness)
…node position imperfections
(out-of-plumbness)
When the analysis has considered all these When the analysis has considered all these
effects, members can be designed using effects, members can be designed using
the provisions for individual members the provisions for individual members
(Chapters D, E, F, G, H, I) (Chapters D, E, F, G, H, I)
Nair 143 Nair 144
24
Effects to be considered: APPENDIX 7
Ø Member deformations DIRECT ANALYSIS METHOD
Ø Other deformations
Ø P- Δ effects Spec presents
Ø P- δ effects approaches
Ø Geometric imperfections that consider
Ø Residual stresses these effects
APPENDIX 7 Ø P- Δ effects
Ø P- δ effects
DIRECT ANALYSIS METHOD Ø Geometric imperfections
Ø Residual stresses
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
Use second-order elastic analysis that Use second-order elastic analysis that
considers both P-Δ and P-δ effects. considers both P-Δ and P-δ effects. Options:
… any general second-order analysis method
25
DIRECT ANALYSIS METHOD
Ø P- Δ effects
Ø P- δ effects
Ø Geometric imperfections
First-order analysis
Ø Residual stresses
amplified by B1 and B2
Use second-order elastic analysis that is
considers both P-Δ and P-δ effects. Options:
… any general second-order analysis method Second-order analysis
… amplified first-order analysis (B1 & B2 )
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
Use second-order elastic analysis that Use second-order elastic analysis that
considers both P-Δ and P-δ effects. Options: considers both P-Δ and P-δ effects. Options:
…any general second-order analysis method …any general second-order analysis method
…amplified first-order analysis (B1 & B2 ) …amplified first-order analysis (B1 & B2 )
Exception: P-δ effects can be neglected when
member axial loads are below a specified level.
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
26
DIRECT ANALYSIS METHOD
Ø P- Δ effects
Ø P- δ effects
Y3 Ø Geometric imperfections
N3 Ø Residual stresses
Y2
N2
Y1 Apply “notional loads”:
N1 Ni = 0.002 Yi
The 0.002Yi notional load is equivalent to an
initial out-of-plumbness of 1/500
Ni = 0.002 Yi
Ø P- Δ effects
Ø P- δ effects
Y3 Y3 Ø Geometric imperfections
N3 1 Ø Residual stresses
Y2 Y2
N2
500
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
27
DIRECT ANALYSIS METHOD DIRECT ANALYSIS METHOD
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
Use reduced flexural and axial stiffnesses Use reduced flexural and axial stiffnesses
EI* and EA* in the analysis EI* and EA* in the analysis
EI* = 0.8 τb EI
EA* = 0.8 EA
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
Use reduced flexural and axial stiffnesses Use reduced flexural and axial stiffnesses
EI* and EA* in the analysis EI* and EA* in the analysis
EI* = 0.8 τb EI EI* = 0.8 τb EI
EA* = 0.8 EA EA* = 0.8 EA
τb depends on level of axial stress in member; τb can be taken as 1 in all cases if additional
τb = 1 when P < 0.5 Py notional load of 0.001Yi is applied.
Nair 167 Nair 168
28
DIRECT ANALYSIS METHOD DIRECT ANALYSIS METHOD
Ø P- Δ effects Ø P- Δ effects
Ø P- δ effects Ø P- δ effects
Ø Geometric imperfections Ø Geometric imperfections
Ø Residual stresses Ø Residual stresses
Use reduced flexural and axial stiffnesses Use reduced flexural and axial stiffnesses
EI* and EA* in the analysis EI* and EA* in the analysis
EI* = 0.8 τb EI EI* = 0.8 τb EI
EA* = 0.8 EA EA* = 0.8 EA
The EI and EA adjustments also account for
the effects of member out-of-straightness on
the stiffness of the overall structure
Nair 169 Nair 170
Ø P- Δ effects
Ø P- δ effects Applied Include
Ø Geometric imperfections Loads imperfections3
Ø Residual stresses Include & inelasticity3 in
2nd-ORDER
ANALYSIS
or B1 & B2
Rigorous
imperfections1 member
& inelasticity2 capacity
Design members using the provisions for in analysis calculations
individual members (Chapters D, E, F, G, H, I),
with K=1 for computing compression strengths Member Member
P&M Capacity
K=1
1Notional 3The usual column
loads or direct modeling
2Reduced EI & EA strength formulas
APPENDIX 7 APPENDIX 7
DIRECT ANALYSIS METHOD DIRECT ANALYSIS METHOD
29
APPENDIX 7 CHAPTER C
DIRECT ANALYSIS METHOD STABILITY ANALYSIS AND DESIGN
30
EFFECTIVE LENGTH METHOD EFFECTIVE LENGTH METHOD
or B1 & B2
Rigorous
31
EFFECTIVE LENGTH METHOD
CHAPTER C
Applied Include
Loads imperfections3 STABILITY ANALYSIS AND DESIGN
Include & inelasticity3 in
2nd-ORDER
ANALYSIS
or B1 & B2
Rigorous
imperfections1 member
but neglect capacity C2. CACULATION OF REQUIRED STRENGTHS
inelasticity in calculations 2. Design Requirements DIRECT ANALYSIS
analysis METHOD
Member Member Can use Appendix 7 in all cases
P&M Capacity Can use C2.2a or C2.2b if Δ/Δ0 < 1.5
K>1 EFFECTIVE
2a. Design by Second-Order Analysis LENGTH
METHOD
1Notional loads 3The usual column 2b. Design by First-Order Analysis
strength formulas FIRST-ORDER
ANALYSIS METHOD
Nair 187 Nair 188
Can be used only when P < 0.5 P y in all Perform first-order elastic analysis
columns whose flexural stiffness contributes use nominal geometry
to the lateral stability of the structure use nominal stiffness
32
FIRST-ORDER ANALYSIS METHOD FIRST-ORDER ANALYSIS METHOD
33
DIRECT EFFECTIVE FIRST DIRECT EFFECTIVE FIRST
ANALYSIS LENGTH ORDER ANALYSIS LENGTH ORDER
Limitations --- Δ/Δ0 <1.5 Δ/Δ0 <1.5 Limitations --- Δ/Δ0 <1.5 Δ/Δ0 <1.5
P/Py <0.5 P/Py <0.5
Analysis Second- Second- First- Analysis Second- Second- First-
Order Order Order Order Order Order
Geometry Nominal Nominal Nominal Geometry Nominal Nominal Nominal
EI & EA Reduced Nominal Nominal
34
DIRECT ANALYSIS METHOD —the right way!
Applied Include
Loads imperfections3
Include
2nd-ORDER
& inelasticity3 in … summary of how DAM
ANALYSIS
or B1 & B2
Rigorous
imperfections1 member
& inelasticity2 capacity accounts for basic Spec
in analysis calculations
Member Member
requirements for stability
P&M Capacity design
K=1
1Notional 3The usual column
loads or direct modeling
2Reduced EI & EA strength formulas
35
Stability Design Features of Direct Stability Design
Requirements: Analysis Method: Requirements:
Second-order P-Δ and P-δ effects Second-order P-Δ and P-δ effects
analysis* analysis*
Out-of-plumbness Out-of-plumbness
Notional loads or Member out-of-straightness Notional loads or Member out-of-straightness
direct modeling -- effect on member strength direct modeling -- effect on member strength
Member out-of-straightness Member out-of-straightness
Usual column
-- effect on structure stiffness -- effect on structure stiffness
strength formulas
Residual stresses Residual stresses
-- effect on member strength -- effect on member strength
Residual stresses Residual stresses
-- effect on structure stiffness -- effect on structure stiffness
*Rigorous or B1-B2 *Rigorous or B1-B2
Nair 213 Nair 214
Second-order P-Δ and P-δ effects Second-order P-Δ and P-δ effects
analysis* analysis*
Out-of-plumbness Out-of-plumbness
Notional loads or Member out-of-straightness Notional loads or Member out-of-straightness
direct modeling -- effect on member strength direct modeling -- effect on member strength
Member out-of-straightness Member out-of-straightness
Usual column Usual column
-- effect on structure stiffness -- effect on structure stiffness
strength formulas strength formulas
Residual stresses Residual stresses
Stiffness -- effect on member strength Stiffness -- effect on member strength
reductions reductions
Residual stresses Residual stresses
-- effect on structure stiffness -- effect on structure stiffness
*Rigorous or B1-B2 *Rigorous or B1-B2
Nair 215 Nair 216
36
Features of Direct Stability Design
Analysis Method: Requirements:
W2/ W1 =α
W2 W2
Non-linear
W1 response W1
indicated by
Load
Load
2nd-order
analysis
R1 R2 R1 R2
R2/R1 >α
WLRFD WLRFD
Design by Design by
W ASD
LRFD ASD
Load
Load
37
WLRFD WLRFD
Design by Design by
W ASD W ASD
ASD ASD
Load
Load
Design by ASD for this force
Cannot use RASD will yield smaller member
than design by LRFD
Load
R1.6 ASD
Load
Divide Divide
by 1.6 by 1.6
R1.6 ASD R1.6 ASD
RASD RASD
DESIGN DESIGN
38
Members Shankar Nair Chairman Members Shankar Nair Chairman
Greg Deierlein Vice-Chairman Greg Deierlein Vice-Chairman
of Bill Baker of Bill Baker
AISC Reidar Bjorhovde AISC Reidar Bjorhovde
TC-10 Charlie Carter TC-10 Charlie Carter
(Stability) Shu-Jin Fang (Stability) Shu-Jin Fang
Jim Fisher Jim Fisher
Ted Galambos Ted Galambos
Larry Griffis Much of the early Larry Griffis
Jerry Hajjar work on the 2005 Jerry Hajjar
Todd Helwig stability provisions Todd Helwig
Richard Henige was done under Richard Henige
Leroy Lutz the leadership of Leroy Lutz
Clint Rex former TC-10
Clint Rex
Steve Thomas Steve Thomas
Don White chairman Don White
Ron Ziemian Joe Yura Ron Ziemian
Nair 229 Nair 230
The paper in the 2007 NASCC proceedings The paper in the 2007 NASCC proceedings
(for this presentation) includes a model (for this presentation) includes a model
specification written exclusively around the specification written exclusively around the
Direct Analysis Method. Direct Analysis Method.
Thank you
Questions
Nair 233
39