th

The 14 World Conference on Earthquake Engineering

October 12-17, 2008, Beijing, China

EFFECT OF MOMENTS DUE TO LATERAL DEFORMATION IN GRAVITY

COLUMNS
1 2
Dane E. Huggard and Paul W. Richards
1
Masters Student, Dept. of Civil and Environmental Engineering , Brigham Young University, Utah, USA
2
Assistant Professor, Dept. of Civil and Environmental Engineering, Brigham Young University, Utah, USA
Email: [email protected], [email protected]

ABSTRACT :

In steel buildings, all columns of a story experience similar drifts under seismic loads when floor diaphragms
are rigid. Columns in lateral-force-resisting frames are designed with these deformations in mind, but gravity
columns often are not. When gravity columns are continuous, moments develop when interstory drifts occur.
Using nonlinear dynamic analysis software, moments due to lateral deformation from seismic forces were
obtained for lateral-force-resisting columns within 3- and 9-story buckling-restrained-brace frames (BRBFs).
Moments in gravity columns were then determined. Gravity columns were evaluated with interaction
equations to determine whether load demands exceeded column capacity. For most gravity columns, column
capacity was exceeded. Recommendations for improved gravity column design are given.

KEYWORDS: Gravity column design, lateral deformation, moment demand, BRBF

1. INTRODUCTION

In steel buildings, all columns of a story experience similar drifts under seismic loads when floor diaphragms
are rigid. Columns in lateral-force-resisting frames are designed with these deformations in mind, but gravity
columns are often not. When gravity columns are continuous, moments develop when interstory drifts occur.

Recent experiments have indicated that columns in lateral force resisting frames can carry large axial loads after
flexural hinges form at each end. Newell and Uang tested nine W14 specimens under combined axial and
lateral loads (Newell and Uang 2006). Columns achieved drifts of 7 to 9% while carrying significant axial
loads. The columns in that study were relatively heavy sections with seismically compact flanges. It is
unclear if columns with non-compact flanges can perform acceptably at large drifts.

This paper discusses analyses performed to determine whether the combination of axial loads and moments due
to lateral deformation exceeds the elastic capacity of gravity columns designed according to U.S. provisions.
From the results, possible recommendations for modifications of gravity column design will be discussed.

2. METHODS

2.1. Lateral-Force-Resisting Frame Design

Eight buildings were designed representing one system (BRBFs), two heights (3-story, 9-story), and four
strength levels (discussed later). Building plan dimensions and floor masses matched those used in moment
frame studies (Gupta and Krawinkler 1999). Seismic weights for the 3- and 9- story buildings were 31.8 and
97.3 MN (1790 and 5470 kips). Braced bays were located around the perimeter of the buildings. BRBFs had
braces in the two-story-X configuration (Fig. 1).

Braced frames were designed according to the 2006 International Building Code (ICC 2006) equivalent lateral
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

force procedure and Seismic Provisions (AISC 2005a). A Los Angeles, California site was used for design
with SDS=1.11 and SD1=0.61, where SDS and SD1 are the site design spectral accelerations at 0.2 and 1.0
seconds in terms of gravity. Base shear coefficients, Cs, for an importance factor, I, of 1, for 3- and 9- story
frames are 0.16 and 0.08 respectively (ICC 2006).

The importance factor in the base shear equations introduces the possibility of different design base shears for
buildings of similar height and system at the same site. Four strength levels (design base shears) were
considered for each building height (Cs=0.15, 0.20, 0.25, 0.30 for 3-story buildings and Cs= 0.06, 0.09, 0.12,
0.15 for 9-story buildings). Buildings with Cs values lower than those corresponding to an I of 1 were
included in the study to investigate the effects of low lateral strength on column demands.

Brace areas and column shapes for the frames are given in Tables 2.1 and 2.2.

[email protected] (30ft) [email protected] (30ft)

[email protected] (30ft)
[email protected] (30ft)

All story heights

3.98m (13ft) unless
otherwise noted
5.49m

(a) 3-story (b) 9-story

Figure 1 Plan and elevation view of BRBF buildings

Table 1 3-Story BRBF Designs

Shape (U.S. Designation) or BRBF Brace Area (cm2)
BRBF BRBF BRBF BRBF
Membera Cs=0.15 Cs=0.20 Cs=0.25 Cs=0.30
BR1b 29 39 52 58
BR2 26 32 39 52
BR3 16 19 26 29
C1-C3c W10x39 W10x49 W10x54 W10x68
a. Brace and column sizes are indicated in the table; beam sizes
governed by gravity with all beams W16x40
b. BR1 is first story brace, BR2 second story…
c. C1 is first story column, C2 second story…
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

Table 2 9-Story BRBF Designs

Shape (U.S. Designation) or BRBF Brace Area (cm2)a
BRBF BRBF BRBF BRBF
Member Cs=0.06 Cs=0.09 Cs=0.12 Cs=0.15
BR1 42 65 84 100
BR2 35 52 65 84
BR3 35 52 65 84
BR4 32 45 65 77
BR5 29 42 58 71
BR6 26 35 52 65
BR7 19 29 42 52
BR8 16 23 29 35
BR9 9.7 13 16 19
C1-C2 W14x159 W14X233 W14X311 W14X370
C3-C4 W14x 90 W14x132 W14x176 W14x233
C5-C6 W14x61 W14x90 W14x99 W14x120
C7-C9 W14x48 W14x48 W14x48 W14x61
a. See all notes from Table 2.1.

2.2. Gravity Column Design

Critical interior gravity columns for the buildings were designed according to current U.S. provisions (ICC
2006). All gravity loads used for the analysis were taken from Gupta and Krawinkler (1999). Load values
are shown in Table 2.3. The column sizes picked for the critical interior columns are shown in Table 2.4.
These were the most efficient columns, by weight; the majority do not have seismically compact flanges.

Table 3 Gravity loads for 3- and 9-story buildings

Floor dead load: 96 psf
Roof dead load excluding penthouse: 83 psf
Penthouse dead load: 116 psf
Reduced live load per floor and for roof: 20 psf

Table 4 Critical Interior Column Design

Stories in Frame Stories Shape
3 All W10x49
9 1-2 W14x120
3-4 W14x90
5-6 W14x74
7-9 W14x53

2.3. Obtaining Moments in Lateral-Force-Resisting Columns

Individual frames were modeled as two dimensional systems using the nonlinear dynamic analysis program
Ruaumoko (Carr 2004). Standard beam elements with bi-linear flexural-axial hinges at each end were used to
represent beams and columns. Buckling-restrained braces were modeled with truss elements that had
multi-linear kinematic-type hardening. This approach has been used by Sabelli et al. (2003) and Coy (2007).
Experimental data for calibration was taken from Merritt et al. (2003) and Reavely et al. (2004).

Columns at the base of the frames were considered fixed. Beam-column connections were considered rigid
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

when a gusset plate was present, and pinned when not.

Each frame was analyzed under ten earthquakes. See Richards (2008) for details on the ground motions.
Moments due to lateral deformation from seismic forces were obtained for the right column within individual
frames. Maximum moments for each earthquake were obtained at each floor level for each column.
Maximum moments from all ten earthquakes were averaged at each location.

2.4. Obtaining Moments in Gravity Columns

Moments in the lateral-force-resisting columns were used to obtain moments in gravity columns that experience
equal lateral deformation, but have different cross-sectional properties. Moments in the lateral-force-resisting
columns were normalized by dividing them by the respective moments of inertia of the column at each story.
This normalized moment was then multiplied by the moment of inertia of a gravity column to obtain the
moment due to lateral deformation.

A derivation of the normalized moment formula may be helpful. Moments can be expressed as a function of
the deflection (Δ) of a member, the member’s length (L), modulus of elasticity (E), and moment of inertia (I):

CEI x Δ
M = (2.1)
L2

Where M is the moment in the member, kip-in.; E is the modulus of elasticity of the member, ksi; Ix is the
moment of inertia of the member with respect to the x-axis, in.4; Δ is the member deformation, in.; L is the
unbraced member length, in.4; and C is a constant dependent upon boundary conditions of the member.

Eqn. 2.1 may be divided by Ix to obtain the normalized moment, as shown in Eqn. 2.2.

M CEΔ
= 2 (2.2)
Ix L

The formula for obtaining moments in the gravity column is:

M lat
M grav = × I grav (2.3)
I lat

where Mlat is the moment in the corresponding lateral-force-resisting column; Ilat is the moment of inertia in the
lateral-force-resisting column; and Igrav is the moment of inertia in the gravity column.

2.5. Column Elastic Moment Capacity

In computing the bending capacity of a wide flange section, Cb accounts for the effect of moment gradient
(AISC 2005b). A Cb value of 1 can be used when computing the moment capacity in a column, but this is
likely conservative, since double curvature often occurs during seismic loading corresponding to a Cb value of 2
or greater. Computing a column capacity with Cb this large causes the capacity value to exceed the plastic
moment capacity value, which is the limiting capacity of a member. For the columns considered (Table 4), any
Cb value higher than 1.18 would cause the moment capacity of a column to reach the plastic moment capacity.
Therefore, in the interaction equation the plastic moment capacity value was always used for the moment
capacity of the column.
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

2.6. Interaction Equation

An interaction equation was used to analyze the gravity columns accounts for combined bending and axial loads
on a member. For all columns, the axial demand was greater than 20% of capacity. Therefore, the governing
interaction equation was:

Pu 8 ⎛ Mu ⎞
+ ⎜⎜ ⎟⎟ < 1.0 (2.4)
ϕ c Pn 9 ⎝ ϕ b M n ⎠

where Pu is the axial demand; φcPn is the axial capacity; Mu is the moment demand; and φbMn is the moment
capacity (AISC 2005b). Pu assumed that 1.2D+0.5L gravity loads were present during the earthquake (ICC
2006).

If the left side of Eqn. 2.3 exceeds a value of 1.0, demand exceeds capacity.

3. RESULTS

3.1. Interaction

Resulting interaction equation values for the 3- and 9-story gravity columns are shown in Figs. 2 and 3. For all
3-story buildings (Fig. 2), the load demand on the gravity columns exceeded column capacity. Capacity was
not exceeded for any second or third stories within the 3-story buildings. This is because the same column size
was used throughout, so in the second and third stories the columns were oversized for axial loads. For the
9-story buildings, capacity was exceeded for the majority of the columns (Fig. 3). Capacity was exceeded for
all columns on the 1st and 7th stories, two columns on the 5th story, and none on the third story.

3.5
Cs = 0.15
3 Cs = 0.20
Cs = 0.25
2.5
Story Level

Cs = 0.30

2 Limit

1.5

0.5
0.00 1.00 2.00
Interaction Equation Value

Figure 2 Interaction equation values for 3-story gravity columns

 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

9
Cs = 0.06
Cs = 0.09
8
Cs = 0.12
Cs = 0.15
7 Limit

6
Story Level

0
0.00 1.00 2.00
Interaction Equation Value

Figure 3 Interaction equation values for 9-story gravity columns

4. CONCLUSION

In this study, moment demands in continuous gravity columns were determined from time history analyses of
eight BRBF buildings. When combined with axial loads, these moments are sufficient to cause yielding for the
majority of the columns considered. The results of this small study indicate a potential problem, and may
justify a more thorough investigation. Currently non-compact columns can be used for gravity columns in
high seismic areas (AISC 2006). It may be good practice to select gravity columns with compact, or
seismically compact flanges if there is a possibility that flexural hinges will form.

REFERENCES

ASCE. (2005). ASCE/SEI 7-05, American Society of Civil Engineers, Reston, VA

American Institute of Steel Construction (AISC). (2005a). Seismic Provisions for Structural Steel Buildings,
AISC, Chicago.
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

American Institute of Steel Construction (AISC). (2005b). Steel Construction Manual, 13th ed. AISC, Chicago.

Carr, A. (2004). Ruaumoko Users Manual, Univ. of Canterbury, Christchurch, New Zealand.

Coy, B. (2007). Experimental Testing of Pinned Beam Connections for Buckling Restrained Braced Frames.
MS Thesis, Brigham Young Univ., Provo, Utah.

Gupta, A., and Krawinkler, H. (1999). Prediction of seismic demands for SMRFs with ductile connections and
elements. Report SAC/BD-99/06, SAC Joint Venture, Sacramento, Calif.

International Code Council (ICC). (2006). International Building Code. International Code Council, Inc.,
Whittier, Calif.

Merritt, S., Uang, C.-M., and Benzoni, G. (2003). Subassemblage testing of Corebrace buckling-restrained
braces. TR-2003/01, Dept. of Struct. Eng., Univ. of California at San Diego, La Jolla, Calif.

Reaveley, L., Okahashi, T., and Fatt, C. (2004). Corebrace Series E Buckling Restrained Brace Test Results,
Dept. of Civil and Env. Eng., Univ. of Utah, Salt Lake City, Utah.

Richards, P. (2008). Seismic column demands in ductile braced frames. J. of Struct. Eng. in press.

Sabelli, R., Mahin, S.A., and Chang, C. (2003). Seismic demands on steel braced-frame buildings with
buckling restrained braces. Eng. Structures, 25, 655-666.