Customary Light-Gauge Steel Framing Construction W

Customary light-gauge steel framing construction
with at strap bracing. Seismicity limits for short to

mid-rise buildings in Europe
JORDI NAVARRO GRANADOS (  [email protected] )
Universitat Politècnica de Catalunya: Universitat Politecnica de Catalunya https://orcid.org/0000-0003-
1528-3974
Miquel Casafont
Universitat Politècnica de Catalunya: Universitat Politecnica de Catalunya
Oriol Bové
Jordi Bonada
Francisco López-Almansa
Research Article
Keywords: Steel Framing, Cold-Formed Steel, Built-up Sections, Moderate and High Seismicity Areas,
Short to Mid-Rise Buildings
Posted Date: March 16th, 2023
DOI: https://doi.org/10.21203/rs.3.rs-2674919/v1
License:   This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License
Customary light-gauge steel framing construction with flat strap bracing.
Seismicity limits for short to mid-rise buildings in Europe
Jordi Navarro (corresponding author)1, Miquel Casafont2, Oriol Bové3, Jordi Bonada4,
Francisco López-Almansa5
1
Senior Researcher, The Catalan Institute of Construction Technology, ITeC, Barcelona, [email protected]
2
Associate Professor, Universitat Politècnica de Catalunya-BarcelonaTech (UPC), Strength of Materials and
Engineering Structures Department, Barcelona, [email protected]
3
Assistant Professor, Universitat Politècnica de Catalunya-BarcelonaTech (UPC), Strength of Materials and
4
Serra Hunter Fellow, Universitat Politècnica de Catalunya-BarcelonaTech (UPC), Strength of Materials and
5
Emeritus Professor, Universitat Politècnica de Catalunya-BarcelonaTech (UPC), Architecture Technology
Department, Barcelona, [email protected]; currently, Associate Researcher RiNA, Universidad
Austral de Chile, Valdivia
Abstract
Ordinary (customary) light gauge steel framing is a convenient construction technology; however, given
the studs low axial capacity, has been mainly considered for short buildings in low seismicity regions.
This paper investigates the bounds of this type of steel framing (in moderate and high seismicity areas)
by designing the structures of three representative 5, 7 and 10-story buildings in order to examine their
feasibility. The necessity of using built-up columns consisting of two or more sections is observed. The
axial capacity of studs is estimated by code-type analyses. The critical buckling stress is determined
with well-known closed-form expressions; as they do not contemplate the flexibility of the screwed
connections (in built-up sections), the obtained results are complemented with those of Finite Strip
Method and Generalized Beam Theory formulations.
Keywords: Steel Framing, Cold-Formed Steel, Built-up Sections, Moderate and High Seismicity Areas,
Short to Mid-Rise Buildings.
1 Introduction
According to the United Nations [1], the population of cities is growing faster than that of the whole
world, as towns and rural areas are experiencing a slowdown; future projections show that world urban
population will jump from almost 6 billion in 2020 to near 7.5 billion by 2070. Therefore, cities will
tend to grow more vertically than horizontally; this trend highlights the need of developing construction
technologies that are suitable for multistory buildings. In this context, light-gauge steel framing is a
convenient construction solution; it consists basically of vertical wall-like panels with supporting cold-
formed studs, and composite slabs (concrete-steel or timber-steel). The main advantages of steel framing
compared to traditional steel and concrete-based solutions are its fast erection, high construction quality,
rather moderate cost, and lightness; the latter provides relevant benefits in poor foundations soils and in
seismic areas, as the seismic forces are proportional to mass. However, given the rather low axial
capacity of the cold-formed steel studs, steel framing has been mainly considered for short buildings
only; even code limitations on the number of stories apply [ASCE/SEI 7-22 2023]. On the other hand,
horizontal actions generate additional axial forces on the extreme studs (chord studs) of each wall panel;
therefore, wind and earthquakes also prevent increasing the building height. To overcome this limitation,
built-up or tube sections have been proposed for such end studs.
In the context described in the previous paragraph, the objective of this paper is to investigate the bounds
of light-gauge steel framing construction for short to mid-rise buildings located in seismic areas. In order
to maintain the main aforementioned advantages of this construction technology, customary (habitual)
and inexpensive features are not modified; this refers basically to ordinary steel, thin plates, and also
thin wall panels. To perform this study, steel framing structures of three five, seven and ten-story
housing buildings located in moderate and high seismicity areas (PGA = 0.2 g, 0.3 g, 0.4 g Peak Ground
1
Customary light-gauge steel framing construction with flat strap bracing. Seismicity limits for short to mid-rise buildings in Europe
Acceleration) are preliminary designed to analyze their feasibility. The demanding internal axial forces
on the supporting members (steel studs) of these prototype buildings are approximately obtained by
simplified (although reliable and reasonably accurate) hand calculations. Several options are considered
for the abovementioned extreme (chord) studs by combining Cee profiles: single, back-to-back, toe-to-
toe (face-to-face), nested, and a stud pack (4C component section [Phan et al. 2021, 2022]). The axial
capacity of such members is estimated by code-type analyses following the forthcoming European
regulations; all the feasible buckling modes are considered: member (flexural and flexural-torsional),
local and distortional. The slenderness is obtained as the square root of the ratio between the yield and
buckling stresses or loads. The critical stress is determined with well-known closed-form expressions;
as these formulations do not contemplate the flexibility of the screwed connections (in built-up sections),
the obtained results are complemented with those of FSM (Finite Strip Method) [Li et al. 2014] and
GBT (Generalized Beam Theory) [Schardt 1994; Silvestre, Camotim 2002].
2 State-of-the-art of steel framing for mid to high-rise buildings
Cold-Formed Steel (CFS) framing is commonly used as load-bearing structures for 1 to 2 story
buildings, particularly in North America where this technology has benefited from the timber framing
tradition; in Europe, the development of steel framing apparently stems from traditional and composite
constructions. Report [LGSF 2022] discusses the worldwide prospective of this technology.
As horizontal actions are a major limitation for tall buildings, the first CFS framing load-bearing
structures in mid to high-rise buildings are gravity load carrying systems supplemented with additional
elements to resist horizontal actions (wind and seismic); usually these elements are RC stair (and
elevator) cases (shafts). An example is the SFIA Matsen Tower [Ford 2016]; it is a hypothetical 40-
story residence building designed for a non-seismic area. Next steps have focused on structures made
entirely of CFS framing; however, recent examples are also located in non-seismic regions: two 7-story
Hotels in Chicago, and an 8-story (seven stories of cold-formed steel framing over a pre-cast concrete
podium) apartment building in Green Bay (WI).
Horizontal actions on steel framing require the wall panels be strengthened to resist the ensuing shear
forces. An overview of the proposed strategies can be found in [Sharafi et al. 2018; Taranu et al. 2023;
Yilmaz et al. 2023]; those that do not pervert the CFS nature of the steel frames are: timber sheathing
boards, X-bracing steel straps, additional bolting, concrete or mortar filling, among others. Among these
options, strap bracing (Figure 4) is the most convenient and versatile, given its superior strength and
stiffness; however, the subsequent tensile forces in the bracing straps generate over-compression in the
chord (end) studs. This effect is highly limiting, and may require using built-up sections, made by joining
two or more profiles. The influence of the joining technique and design in their performance can be
analyzed from [AISI S100-16 2016] prescriptions (reduced overall slenderness) or, more deeply, using
the aforementioned FSM or GBT approaches.
An early attempt to explore the bounds of CFS constructions under earthquake excitation can be found
in [Nakata, Schafer, Madsen 2012]; a two-story building is preliminarily analyzed. In [Torabian, Nia,
Schafer 2016], archetype mid-rise pure CFS buildings (4-20 stories) located in a high seismic area are
used to assess the height limits of steel frames, and to aid in the development of new CFS technologies;
the main conclusion is that hold-downs and high-capacity studs are needed. A more recent study is the
seismic analysis of a 10-story building [Zhang et al. 2022] to be built and tested (full scale); this research
is aimed to investigate the seismic performance of buildings exceeding the height limitation of 19.8 m
(65 ft) set by [ASCE/SEI 7-22 2022].
The above studies belong to the US context; this paper presents a rather similar study for the European
situation.
3 Seismic performance of steel framing buildings
As discussed in sections 1 and 2, wind and earthquakes are major limitations for the use of steel framing
for mid and high-rise buildings. As in most of the situations seismic effects are more demanding than
wind ones, this section discusses the seismic vulnerability of steel framing buildings.
Most probably, earthquakes are the most damaging environmental effect for any civil construction.
Unless other actions (gravity, wind, thermal, etc.), strength is neither the only, nor the main requirement;
2
Jordi Navarro, Miquel Casafont, Oriol Bové, Jordi Bonada, Francisco López-Almansa 3
in brief, an adequate seismic design of buildings pursues, among others, the following major qualities:
(i) lightweight to mitigate the mass inertial effects, (ii) bilateral stiffness to prevent excessive interstory
drift and collisions with adjoining buildings, (iii) bilateral strength to resist the equivalent seismic
forces, (iv) plan symmetry (this is a mechanical issue, not a geometrical one), and torsional resistance
and stiffness (to avoid highly damaging twisting motion), (v) uniformity along height to put off
damage concentration in soft stories, (vi) rigid diaphragms to guarantee solidarity between the lateral
resisting systems, and providing a regular, simple and predictable seismic behavior, (vii) adequate
foundation to foil uplift, (viii) structural simplicity mainly to facilitate detecting hidden failure modes,
(ix) structural redundancy to maintain a reserve of strength, and (x) ductility to allow for rather light
design (as long as some damage is accepted). Steel framing structures are particularly well suited for
issues i, v, vi (if a topping RC layer is provided), viii and ix; the remaining issues are discussed next.
▪ Lateral stiffness and strength. The rigidity and resistance in both horizontal directions can be
reached with the measures described in section 2, particularly cross-bracing with steel straps. As
discussed in that section and in subsection 6.6, this requires the over-strengthening of the end (chord)
studs; if only CFS profiles are accepted, built-up sections need to be used. Obviously, this poses
relevant limitations in the use of this construction solution is high seismicity areas.
▪ Mechanical plan symmetry. This type of symmetry is not related to any geometrical regularity,
but to small eccentricity between the centers of gravity (G) and rotation (R). As an important number
of braced wall panels can be placed in any horizontal direction, symmetric or near-symmetric
arrangements can be achieved in all the situations.
▪ Uplift prevention. In rather slender buildings located in high seismicity areas, uplift is a serious
risk for any building. Noticeably, lightweight is not a relevant advantage, as both gravity and seismic
forces are proportional to the building weight; however, the uplift forces can be rather easily
compensated with heavy foundations. Hold-down ties need only moderate section (area), and can
be easily embedded in the steel framing structure.
▪ Ductility. Steel framing buildings should be rather ductile, as the only employed material is steel,
being a highly ductile material. Although x-braced wall panels are not as ductile as moment-resisting
frames, a minimum ductility is largely ensured. This is particularly relevant, given that several
buckling instabilities (flexural, torsional-flexural, local and distortional) are possible, mainly in the
chord studs. The dissipative structural elements are the brace straps; as the chord studs are over-
compressed due to the axial tension in such bracing members, the design of these studs shall be
based on conservative capacity equilibrium conditions. This strategy protects against the
abovementioned fragile buckling instabilities.
Other particular seismic design recommendations, such as “no long cantilevers”, “no interrupted
columns”, “no short-columns”, “tying of footings and pile caps”, “strong column-weak beam”,
“separation to adjoining buildings”, “compact plan layout” and “protection of non-structural
components” (among others) are not related to steel framing technology, and do not require particular
discussion.
The paper [Torabian, Nia, Schafer 2016] contains a rather similar discussion.
The main remark of these considerations is that pure steel framing buildings are rather well suited for
seismic resistance; the most critical issue is the high axial compressive demands on the chord studs.
4 5, 7 and 10-story prototype buildings
4.1 Buildings description
Three 5, 7 and 10-story residential buildings are used as case studies representative of common short to
mid-height buildings. They are regular, the story height is 3 m, and the plan layout is square 18 m × 18
m. These buildings have 3 apartments (flats) per floor and a central hall, containing stairs and elevators;
the wall panel distribution is the same in each floor. The building structure consists of prefabricated wall
panels and floor and roof slab panels. In the walls, only CFS profiles (studs, tracks and diagonal straps)
contribute to the load-bearing capacity (“all steel” approach [AISI S240-15 2015]). The wall panels are
completed with insulation, EPS (Expanded PolyStyrene) lightweight mortar, mortar rendering and
gypsum lining. Furthermore, a concrete layer 60 mm thick is poured on site above the metal structure
of floor and roof panels, where the sliding between the metal frame and the concrete layer is prevented
by shear connectors; this layer is reinforced (with a welded wire mesh) to provide some in-plane strength
and prevent shrinkage cracks. Figure 1 displays an elevation of the 10-story building, and a plan view
of a generic story. The global references axes (x, y, z) used in Figure 1 are maintained along this paper.
18 m
30 m
(Hb)
z y
x x
(a) Elevation (front view) of the 10-story

(b) Plan view of any prototype building
building
Figure 1. Prototype buildings
Figure 1.b describes the aforementioned three apartments; the internal partitions (not represented) are
made with dry walls. Figure 1.a depicts the openings of a main façade of the building with 10 levels;
Figure 1.b shows the correspondence with the architectural distribution. In fact, the architectural layouts
of these buildings and their impact in the wall panels distribution has been carefully considered in order
to have realistic and representative examples.
4.2 Steel framing system
As described in section 2 and subsection 4.1, the wall panels are composed of vertical supporting studs
framed by top and bottom tracking profiles; the slab panels have a similar composition, although are
topped with a 60 mm RC layer. Figure 2 displays views of wall (Figure 2.a) and slab (Figure 2.b) panels.
z z
x,y
x,y
(a) Wall panel [Macillo et al.
(b) Slab panel [ETA 2020]
2014]
Figure 2. Steel framing structure
The internal arrangement of the wall panels without the braces represented in Figure 2.a is convenient
for gravity loads, but its lateral strength and stiffness are largely insufficient to resist relevant wind and
4
seismic effects. For that reason, some wall panels (all of them apart from those with openings) are braced
with steel straps located at both wall sides (front and rear, Figure 2.a). Figure 3 presents several bracing
arrangements.
z
x,y
(b) Diagonal bracing straps (c) Diagonal bracing straps
(a) Diagonal bracing
and one horizontal and two horizontal
straps only
bridging members bridging members
Figure 3. Strap-braced wall panels
Figure 3.a shows a basic X (diagonal) bracing, and Figure 3.b and Figure 3.c describe similar
arrangements with one and two horizontal bridging members, respectively. The purpose of these
members is to provide buckling restraint in the wall plane. It is assumed that this bridging system is
capable of sufficiently blocking the intermediate in-plane displacements of the studs, or that it is
adequately reinforced with complementary elements (additional straps or blocking elements) to fulfil
this function.
As the horizontal structural elements (floor and roof slabs, Figure 2.b) do not have a highly relevant
participation in the building lateral seismic capacity, only the vertical members (wall panels, Figure 2.a)
are described herein. All of them are either 100 or 150 mm thick, and the tracks are made of channel
105 × 40 × 1.5 or 155 × 40 × 1.5 profiles, respectively. The internal (middle) studs are Cee single or
built-up sections (pairs) spaced between 400 and 600 mm, the external (chord) studs are Cee built-up
sections (pairs or foursomes) (Figure 6), and the diagonal braces are straps (Figure 2.a and Figure 4).
Figure 4 displays, similarly to Figure 1.b, a building plan view detailing the location and length of the
braced wall panels. Figure 4 is aimed to deliver deeper information about the elements that provide
lateral stiffness and strength.
Figure 4. Braced (red) and unbraced (white) wall panels in both directions
Figure 4 shows that the braced wall panels in x and y direction are similar (in number and length). The
panels in y direction are symmetrically distributed; conversely, the distribution of panels in x direction
is rather asymmetric. This choice is intentional, and aims to reveal circumstances that can occur in real
life (because of architectural requirements), even in externally symmetrical buildings. As mentioned in
section 3, this might impair its seismic performance; this effect is expected to be moderate.
5 Simplified formulation for the global structural analysis of the prototype buildings
5.1 General considerations
This section presents a simplified formulation for the global structural analysis of the prototype buildings
described in section 4. This analysis consists in obtaining approximately the internal forces in the
building structural elements; section 7 contains a local structural analysis providing the strength of the
most critical structural elements, thus allowing to verify whether they exceed or not the demanding
forces determined in this section. In fact, this process is only performed for the most critical members,
namely the chord studs of the 1st story braced wall panels.
Given the particular characteristics of the seismic action, its effects are discussed in an individual section
(section 6).
5.2 Load combinations
These prototype buildings are designed for the major actions, namely: gravity, wind and seismic. The
actions and their combinations are determined according to the European regulations [EN 1990 2005].
Gravity loads are divided in permanent (dead) and variable (live), being commonly represented as G
6
and Q, respectively; wind forces are variable, named as WX and WY (W stands for wind). Finally, seismic
actions are accidental, being ordinarily denoted as EX and EY (E stands for earthquake). The
combinations involve safety and combination factors; for this simplified analysis, the following major
six combinations are considered:
(I) 1.35 G + 1.5 Q + 0.9 WX (II) 1.35 G + 1.5 Q + 0.9 WY
(III) 1.35 G + 1.05 Q + 1.5 WX (IV) 1.35 G + 1.05 Q + 1.5 WY (1)
(V) G + 0.3 Q + EX + 0.3 EY (VI) G + 0.3 Q + EY + 0.3 EX
5.3 Global analysis for gravity loads
The vertical actions are determined according to the current European regulations [EN 1991-1-1 2002].
The permanent loads are distributed between the levels (slabs); the estimated values are 2430 kN for the
stories and 2592 kN for the roof. The variable loads arise from an average distributed force of 2 kN/m2;
the total force for each level is 648 kN. The internal forces in the studs are determined by assuming that
the load on each slab is distributed isostatically (i.e. as if all the joists supports were hinged) between all
the supporting studs; this criterion is common, and leads to small inaccuracies only. The most loaded
studs are the first story ones; given the relevance of the seismic effects, the chord studs in the façade
wall panels are the structural members mostly analyzed in the simplified study performed in this section.
5.4 Global analysis for wind forces
The static wind forces (pressure and suction) in each direction are determined according to the current
European regulations [EN 1991-1-4 2005] by assuming a roughness length for terrain category IV. Then,
the story shear forces are obtained by equilibrium conditions; inside each story, such forces are
distributed between all the wall panels in the corresponding direction (Figure 4) according to their lateral
stiffness. The axial compressive force in the corresponding chord stud of each panel is determined as
the vertical projection of the tensile force in the steel strap brace. In turn, this force is obtained by a
simplified structural analysis of the panel neglecting the contribution of the internal studs to its lateral
stiffness (subsection 6.1). Noticeably, both gravity and wind axial compressive forces on the studs are
assumed to be applied on the chord stud central axis (passing on the section centroids); in other words,
these elements undergo a centered compression only.
6 Seismic analysis of the prototype buildings
6.1 Analysis approach
The seismic design of the prototype building is performed, in a simplified way, according to the
upcoming version of the European seismic design code [EN 1998-1 2023]; whenever required, that
regulation is supplemented with the American document [AISI S400-20 2020]. The vertical seismic
action is not considered.
[EN 1998-1 2023] proposes two 2-D static design strategies: Lateral Force and Response Spectrum
Methods; they consist basically in considering the contribution of one and several modes, respectively.
Several conditions must be met (in each direction) to allow the use of the most simplified approach
(Lateral Force Method): (i) buildings not taller than 30 m, and (ii) fundamental period in the considered
direction (T1) not exceeding neither 1.5 s nor four times the corner period TC (subsection 6.4, Table 2).
Obviously, the first condition is fulfilled (Figure 1); regarding the bounds for the building fundamental
period, the obtained results (Table 11) also fulfill the required conditions. Hence, the Lateral Force
Method is adopted in this research; in this formulation, the base shear force in the considered direction
is given by:
Fb =  𝑚 𝑆r (𝑇1 ) (2)
In equation (2), m is the mass corresponding to the building seismic weight (subsection 6.5), and  is a
dimensionless coefficient being equal to 1 if the building has more than two stories and T1 is greater
than two times the corner period TC, and 0.85 otherwise. Table 11 displays the adopted values of  in
each direction.
In equation (2), Fb represents the equivalent static effect of the design accelerograms in every horizontal
direction; it shall be distributed between all the stories proportionally to their masses and lateral
displacements. Then, the demanding seismic internal forces in the chord studs are determined similarly
than in the wind analysis (subsection 5.4); the main difference is that the seismic shear story forces are
assumed to be applied in the center of gravity of each story (G), and this point exhibit actual and
accidental eccentricity with respect to the corresponding center of rotation (R). This issue is discussed
more deeply next.
Table 1 displays the actual eccentricity in the y direction (Figure 4) between the centers of mass (G) and
rigidity (R) of the first story of the prototype buildings.
Table 1. Eccentricity ey (m) between the centers of mass (G) and rigidity (R)
S,475 5-story building 7-story building 10-story building
0.5 2.35 2.35 2.54
0.75 2.44 2.37 2.53
1 2.43 2.36 -
In Table 1, S,475 refers to the site seismicity (subsection 6.2); the indicated values correspond to the
final situation, once the structure has been finally designed (section 8).
The real eccentricity in Table 1 shall be complemented with the accidental eccentricity; [EN 1998-1
2023] states that  5% must be taken in any direction. By adding the actual and accidental eccentricities,
the forces at each braced wall panel are determined proportionally to their stiffnesses; noticeably, both
translational and rotational geometric compatibility conditions need to be imposed (rigid diaphragm
effect).
On the other hand, although regularity in plan is not specifically required, this is equally considered, as
it prevents twisting (torsional) behavior, having proven to be extremely damaging. Three types of
conditions are basically entailed for each level and direction: geometrical symmetry (including compact
configuration), rigid diaphragm, and torsion stiffness. The first two conditions are largely fulfilled, while
the third one involves the contentment of two inequalities: ey  0.3 ry and ry  ls; the subindexes
correspond to verification for the x direction. In these expressions, ry is the square root of the ratio
between torsional and lateral stiffness (“torsional radius”), and ls is the radius of gyration of the floor
mass in plan (square root of the ratio between the polar mass moment of inertia -with respect to point
G- and the floor mass). Noticeably, condition ey  0.3 ry refers to small eccentricity between points G
and R, and ry  ls is equivalent to the first 3-D mode of the building being translational. These inequalities
are satisfactory fulfilled in all the analyzed situations.
6.2 Site seismicity
As discussed in section 1, the building is located in a seismic area; three levels of PGA (Peak Ground
Acceleration) are considered: 0.2 g (moderate seismicity), 0.3 g and 0.4 g (high seismicity). This
parameter broadly represents the maximum expected acceleration at the engineering bedrock (soil type
A) for 475 years return period; conversely, in [EN 1998-1 2023], the site seismicity is mainly
characterized by S,475, maximum response spectral acceleration (5% damping) corresponding to the
constant acceleration range (plateau) of the horizontal elastic response spectrum. Assuming that,
approximately, the ratio between and PGA and S,475 is equal to 2.5, the considered site seismicity levels
in terms of S,475 are 0.5 g, 0.75 g and 1 g.
The soil type (site category in [EN 1998-1 2023]) is an inherent component of the site seismicity; in this
study it is soil B. In [EN 1998-1 2023] this site category refers to stiff soil, and corresponds to vs,H
ranging between 400 and 800 m/s and H800 deeper than 5 m. Parameters H800 and vs,H are the engineering
bedrock depth (characterized by a shear wave velocity of 800 m/s), and the weighted harmonic average
of the of shear wave velocity in soil depth H; H is equal to the lowest of H800 and 30 m. Noticeably, this
soil would be rated as C in the American documents.
[EN 1998-1 2023] classifies the seismicity in four levels in terms of the seismic action index S: very
low (S < 1.30 m/s2), low (1.30 m/s2  S < 3.25 m/s2), moderate (3.25 m/s2  S < 6.50 m/s2) and high
(6.50 m/s2  S). For soil B, structures of consequence class CC2 (this concept corresponds to importance
class II in the former version of [EN 1998-1 2023]) and sites without relevant topographic effects S =
S,475 1.3 (1 − 0.1 S,475); in this expression, S and S,475 must be in g. Therefore, for S,475 = 0.5 g, 0.75
8
g and 1 g, then S = 0.6175, 0.9019 and 1.17 g, respectively. These results confirm that the S,475 = 0.5
g corresponds to moderate seismicity, while S,475 = 0.75 g and S,475 = 1 g should be referred as high
seismicity.
6.3 Building behavior factor
The behavior factor (q) in the European regulations corresponds to the response modification factor (R)
in the American documents. [EN 1998-1 2023] prescribes that the behavior factor for Structural Type g
(Lightweight steel frame wall systems) with flat strap bracing (Figure 3 and Figure 4) is q = 2.5 if the
ductility class is DC3. As a matter of fact, for S,475 = 0.75 g and S,475 = 1 g, such ductility class is
strictly required (Table 11.3 of [EN 1998-1 2023]). The conditions for obtaining such level of ductility
are fulfilled, and therefore, q = 2.5 is assumed in this study for all the prototype buildings. Noticeably,
in the American documents this ductility class (DC3) corresponds broadly to special buildings.
6.4 Design spectra
The design spectra are based on the forthcoming version of the European seismic design code [EN 1998-
1 2023].
The design spectrum according [EN 1998-1 2023] considers an elastic response spectrum referred as
Se(T); it has five branches:
1. Short horizontal branch (extremely short periods, T  TA) Se(T) = S / FA.
2. Linear growing branch (very short periods, TA  T  TB) Se(T) = [S / (TB − TA)] [ (T − TA) + (TB
− T) / FA].
3. Constant branch (plateau, short periods, TB  T  TC) Se(T) =  S.
4. Hyperbolically decreasing branch (mid periods, TC  T  TD) Se(T) =  S T / T.
5. Faster decreasing branch (long periods, TD  T) Se(T) =  TD S T / T2.
In these expressions, FA = 2.5, T = 1 s, S = FT F S,475 and S = FT F S,475; FT is the topography
amplification factor, and F = 1.3 (1 − 0.1 S,475) and F = 1.6 (1 − 0.2 S,475) are the short and
intermediate period site amplification factors, respectively. It is assumed that FT = 1 (there is no
topography amplification). S,475 can be taken as ag 2.5 in [EN 1998-1 2023]. The corner periods TA, TB,
TC and TD are given by TA = 0.02 s, TC = S T / S, TB = TC / 4, and TD = 1 + S,475. Finally, the damping
correction factor  is taken as 1, as the building damping ratio is assumed to be 5%.
Then, the reduced spectrum Sr includes the influence of the behavior factor q: Sr(T) = Se(T) / Rq(T). In
the first branch Rq(T) = 1.5, in the second branch Rq(T) = 1.5 + (q − 1.5) (T − TA) / (TB − TA), and in the
other branches Rq(T) = q. Finally, the minimum value of Sr(T) is  S,475, where  = 0.08.
Table 2 displays the values of the parameters that characterize the design (reduced) spectra for the three
considered seismicity levels, and Figure 5 displays the corresponding spectra. In Figure 5, q = 2.5 is
assumed.
Table 2. Main parameters of the reduced spectra
S,475 (g) S,475 (g) S (g) S (g) TA (s) TB (s) TC (s) TD (s)
0.5 0.15 0.6175 0.2328 0.02 0.09425 0.377 2.5
0.75 0.225 0.9019 0.3438 0.02 0.0953 0.3812 3.25
1 0.3 1.17 0.4512 0.02 0.0964 0.3856 4
0.5
0.45
0.4
0.35
Reduced spectral ordinate (g)
0.3
Sa,475 = 1 g
Bound (1 g)
0.25
Sa,475 = 0.75 g
Bound (0.75 g)
0.2 Sa,475 = 0.5 g
Bound (0.5 g)
0.15
0.1
0.05
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Period (s)
Figure 5. Reduced spectra (Sr) according to [EN 1998-1 2023]. Behavior factor q = 2.5 and
soil type B
6.5 Natural periods of the buildings
Prior to any structural calculation, the building fundamental periods in the direction under consideration
(T1,est) can be estimated according to the simplified empirical expression for framed wall buildings in
[EN 1998-1 2023]: T1,est = 0.05 Hb¾ (s), where Hb is the building height (m). The obtained results for the
5, 7 and 10-story prototype buildings are:
T1,est = 0.05 𝐻b¾ = 0.05 T1,est = 0.05 𝐻b¾ = 0.05 T1,est = 0.05 𝐻b¾ = 0.05
× 15¾ × 21¾ × 30¾ (3)
= 0.38 s = 0.49 s = 0.64 s
There is a big uncertainty involved in the determination of these values; therefore, it would not be
surprising that further calculations (once the structural parameters are known) provide significantly
different results. In this context, the building modal parameters (natural periods and mode shapes) are
determined by solving the classical eigenvalue problem (K – 2 M)  = 0, where K is the stiffness
matrix (in each horizontal direction), M is the mass matrix, and  and  represent the eigenvalues
(natural frequencies) and eigenvectors (modal shapes), respectively. Matrix M is diagonal, containing
the masses of each story; matrix K is generated as tri-diagonal, by assuming that the steel framing
structure behaves as a shear building. Story masses (subsection 5.3) are determined according to the
permanent load combined with 15% of the variable one: 2.576 × 105 kg; for the roof, the combination
involves 30% of the variable load: 2.840 × 105 kg. These values correspond to the seismic weight. Story
stiffnesses in x and y directions are determined according to [AISI S400-20 2020]; the contribution of
the nonstructural components to the building stiffness is neglected. This last consideration might
introduce important inaccuracies, as the nonstructural elements contribution can be significant; it has
10
been adopted as, for moderate and strong seismic inputs, usually such elements are damaged, thus
leaving the structure alone. On the other hand, the influence of the nonstructural components in the
building fundamental period would be highly difficult to assess, given the large number of cladding and
partitioning elements. The periods determined from the modal analyses are discussed in subsection 6.7.
6.6 Seismic design of the chord studs
The chord studs are non-dissipative elements that support the strap braces, being dissipative elements
as the only source of ductility; therefore, they should be designed with capacity equilibrium conditions
intended to prevent their brittle buckling failure. The internal axial force in the chord studs due to seismic
effects is equal to [EN 19981-1 2023]:
EEd = EEd,G + 1.1 rm ENfy = EEd,G + 1.595 ENfy (4)
In equation (4), EEd is the demand internal force due to combinations V or VI (equation (1)), EEd,G is the
contribution of non-seismic actions, ENfy is the internal force corresponding to the axial yielding capacity
of the brace straps, and rm is the material overstrength factor; for the steels considered in this study,
rm = 1.45 can be assumed.
6.7 Seismic global analysis
The global seismic analysis must be performed iteratively, as prior to the first design of the structural
members it is not possible to calculate the building fundamental period. Given this circumstance, the
seismic design codes (in this case, [EN 1998-1 2023]) provide empirical expressions for such a period
that depend only on the structural configuration and the building height; these expressions have been
utilized to obtain the first estimates (equation (3)). The values of the fundamental period provide the
spectral ordinates (subsection 6.4), in turn, such ordinates allow obtaining the base shear forces and the
demanding axial forces in the chord studs (subsection 6.1) and all the structural members; these forces
are to be combined with those from other actions (equation (1)), thus allowing a proper structural
dimensioning (subsection 6.6). Then, a new value of the fundamental period can be determined
(subsection 6.5), and the iterative process continues. Iterations are stopped when the periods do not
change significantly, when their change has no effect in the spectral ordinates (as they lie in the plateau
range), or when the demanding axial forces in the chord studs exceed their capacities.
The most important results of this iterative process (for the 5, 7 and 10-story buildings) are described in
Table 3. It is shown that iterations do not start from the initially estimated periods (equation (3)), but
from longer ones (typically, 1 s); this is because the periods in equation (3) are too short, thus providing
excessively high spectral ordinates that generate axial forces in the chord studs that cannot be withstood
by them.
Table 3. Main results of the global seismic analyses for the prototype buildings
𝒎∗𝟏𝐱 / 𝒎∗ /
Iteration T1x T1y 𝟏𝐲 Fbx Fby
Building S,475 x y 𝒎
No. (s) (s) 𝒎 (%) (kN) (kN)
(%)
1 1 1 1 1 - - 1183 1183
0.5 2 1.06 1 1.07 1 82.45 82.56 1116 1105
3 0.98 1 1.01 1 82.53 82.51 1207 1171
1 1 1 1 1 - - 1747 1747
0.75 2 0.89 1 0.90 1 82.46 82.70 1962 1941
5-story
3 0.75 0.85 0.77 1 82.09 82.20 1980 2268
1 1 1 1 1 - - 2292 2292
2 0.78 1 0.79 1 82.30 82.54 2939 2902
1
3 0.61 0.85 0.63 0.85 81.74 81.85 3194 3093
4 0.59 0.85 0.61 0.85 81.84 81.89 3302 3194
1 1 1 1 1 - - 1650 1650
0.5 2 1.24 1 1.26 1 80.67 80.86 1330 1309
3 1.23 1 1.27 1 80.62 80.67 1341 1309
1 1 1 1 1 - - 2436 2436
7-story 0.75 2 1.03 1 1.05 1 80.40 80.68 2366 2320
3 0.93 1 0.96 1 80.09 80.26 2620 2538
1 1 1 1 1 - - 3197 3197
1 2 0.90 1 0.92 1 80.05 80.39 3553 3475
3 0.76 0.85 0.78 1 79.98 80.09 3576 4099
1 1 1 1 1 - - 2351 2351
0.5 2 1.46 1 1.49 1 78.95 79.24 1610 1577
3 1.43 1 1.46 1 77.05 76.98 1644 1610
10-story
1 1 1 1 1 - - 3467 3467
0.75 2 1.20 1 1.23 1 78.47 78.52 2892 2822
3 1.07 1 1.09 1 76.52 76.60 3243 3184
In Table 3, 𝑚1∗ /𝑚 represents the 1st mode mass participation factor; obviously, subindexes x and y denote
the horizontal directions (Figure 4). The case for S,475 = 1.0 in the 10-story building is omitted, as the
result is obvious.
6.8 Drift verification
[EN 1998-1 2023] prescribes that drift should be limited for two reasons: to prevent damage of
nonstructural components (mainly brittle cladding and partitioning elements connected to two
consecutive floors), and to investigate the need for performing second-order analyses. The first
limitation is established as dr,SD  0.01 hs, where hs is the story height, and dr,SD is the drift for the
Significant Damage (SD) limit state; this drift is calculated from the average lateral displacements ds
obtained by multiplying the reduced displacements dr by q (for the range of periods of interest in this
study). The second limitation is set as   0.1, where  (interstory drift sensitivity coefficient) is basically
𝑃 𝑑
the ratio between the first and second-order moments, and is precisely defined as  = tot 𝑟,SD ; in this
1.5 𝑉tot ℎ𝑠
expression, Ptot and Vtot are the vertical gravity and horizontal seismic forces on the story under
consideration, respectively. Table 4 displays the obtained values of the first story drift (dr,SD, mm) and
the corresponding interstory drift sensitivity coefficient ().
12
Table 4. First story reduced drifts dr,SD (mm) and interstory drift sensitivity coefficients () in the
prototype buildings
5-story building 7-story building 10-story building
S,475
dr,SDx x dr,SDy y dr,SDx x dr,SDy y dr,SDx x dr,SDy y
0.5 19.53 0.028 19.89 0.029 18.29 0.033 18.94 0.035 14.49 0.030 14.77 0.032
0.75 18.31 0.016 22.29 0.017 19.97 0.018 20.74 0.020 15.76 0.017 16.30 0.018
1 18.29 0.010 18.94 0.010 18.09 0.012 22.35 0.013 - - - -
Table 4 shows that dr,SD does not exceed the bound (30 mm) in any case. Regarding , all its values are
clearly below the bound (0.1). These trends show that drift is not excessive.
7 Member buckling analysis
7.1 Global observations
The objective of this section is to determine the characteristics of the cold-formed structure members
that have to withstand the forces defined in section 6. The study focusses on the panel studs, especially
the chord ones, which are the most critical members.
The structural capacity of the stud configurations that are typically used in low-rise steel framed
buildings is insufficient for mid-rise buildings undergoing seismic action; such buildings would require
innovative configurations of built-up sections. Therefore, this section briefly introduces the behavior
and design of these types of sections. Different design approaches are considered, as well as several
factors that govern their behavior, namely cross-section configuration, fastener spacing, the effect of
end fastener groups, wall thickness, sheet thickness, material properties and intermediate restraints.
7.2 Built-up members
The intermediate studs of the panels are designed with individual profiles, while the chord studs with
built-up sections. Initially, only combinations of two profiles had been investigated (back-to-back, toe-
to-toe and nested), but given the high demand of some chord studs (section 8), stud packs composed of
4 profiles had to be added.
Figure 6 displays the stud cross-sections considered in the study. They are composed of channel 100 ×
50 × 20 mm and 150 × 50 × 20 mm profiles (Figure 6.a). Figure 6.b, Figure 6.c and Figure 6.d exhibit
different combinations of two Cee profiles; Figure 6.e describes a combination of such four profiles.
(a) Cee (b) Back-to-back (c) Nested
(d) Face-to-face (toe-to-toe) (e) Pack 4

Figure 6. Studs cross-sections (dimensions in mm)
All built-up sections are considered to be fastened by means of screws, except for the toe-to-toe (face-
to-face) section (Figure 6.d), which is welded (screwing is not possible). The location of fasteners is
also shown in Figure 6. Two longitudinal spacings between screws are studied: L / 3 and L / 6 (L is the
member length). The screws stiffness is defined from values used in similar studies of the literature: 10
kN/mm for axial direction [Fratamico 2018], and the same value for shear [Rasmussen 2020; Pham,
Moen 2015; Phan et al. 2022]; these directions refer to the screw axis.
The study involves various sheet thickness: 1.0, 1.5, 2.0, 2.5 and 3.0 mm; as well steel grades: S250GD,
S280GD, S320GD and S350GD [EN 1993-1-3 2006,2023].
7.3 Member design procedure
The resistance of the studs is determined following [EN 1993-1-3 2006,2023] accounting for all relevant
buckling phenomena: local, distortional and member. The effects of the sectional buckling are taken
into account through the effective area of the cross-section, which is calculated as follows:
▪ Local buckling. The effective width approach is applied, according to clauses 7.6.1. and 7.6.2 of
[EN 1993-1-3 2023]. Noticeably, the effective width is determined independently for each element
of the cross-section based on its plate buckling stress and associated slenderness. The calculation
procedure is similar to that proposed by other codes, such as [AISI S100-16 2016].
▪ Distortional buckling. A reduced thickness of the edge and web stiffeners undergoing distortional
buckling is determined following clauses 7.6.1 and 7.6.3 of [EN 1993-1-3 2023]. The Code proposes
a stiffener buckling approach, where distortional buckling stress is determined by hand (in principle,
see next section). The calculation utilizes a model of the stiffened element (with its stiffener) where
it is assumed that they behave as a compressed member with continuous partial restraint. The spring
stiffness of the restraint depends on the boundary conditions and the flexural stiffness of the adjacent
14
plane elements. The reduced thickness is computed from the distortional buckling stress, and the
corresponding slenderness, using a specific buckling curve.
Studs are considered to be subjected to pure compression. Since we are dealing with a preliminary
dimensioning of the structure, additional bending moments caused by the shift between the centroidal
axes of the effective cross-section and the gross cross-section, if exist, is not considered. Actually, the
effect of the shift would only be relevant for the intermediate studs, which consist in individual mono-
symmetric profiles. All the investigated built-up cross-sections are doubly symmetric and, consequently,
there is not any centroid’s shift.
Member buckling verification for the case of pure compression involves two checks: (i) flexural
buckling; and (ii) torsional or torsional-flexural buckling. The studs composed of a single profile, i.e.
mainly the intermediate studs, have to be verified against flexural and torsional-flexural buckling;
whereas the governing buckling failure of built-up studs is flexural buckling (torsional buckling is rarely
relevant). All the verifications are carried out according to clauses 8.2.2 and 8.2.3 of [EN 1993-1-3
2023], which use of the following equations of [EN 1993-1-1 2023]:
χ A fyb χ Aeff fyb
Nb,Rd = Nb,Rd = (5)
γM1 γM1
Left and right equations correspond to Class 1, 2 and 3 cross-sections (fully effective), and Class 4 cross-
sections (slender), respectively. In equation (5), A is the gross cross-section area, Aeff is the effective
area of the cross-section, fyb is the basic yield strength of the material, γM1 is the partial safety factor for
resistance of members to instability (the adopted value herein is 1.00), and χ is the reduction factor for
flexural buckling (clauses 8.2.2 of [EN 1993-1-3 2023] and 8.3.1.3 of [EN 1993-1-1 2023]) or torsional-
flexural buckling (clauses 8.2.3 of [EN 1993-1-3 2023] and 8.3.1.4 of [EN 1993-1-1 2023]). There are
two factors that should be specified concerning the calculation of χ: (i) buckling curve b is applied in all
calculations, except for the flexural buckling about the y of the back-to-back section, for which curve a
should be used (Table 8.7 of [EN 1993-1-3 2023]), and (ii) the buckling lengths considered in the
determination of member buckling load and associated member slenderness are displayed in Table 5.
Table 5. Buckling length. L is the physical length of the stud: 2780 mm
Flexural buckling
Intermediate restraint Torsional buckling
Buckling about y-y Buckling about z-z
No restraint L L L
1 restraint L L/2 L/2
2 restraints L L/3 L/3
7.4 Buckling stresses and loads. Model of the composite action
The design procedures introduced in subsection 7.3 require the calculation of a buckling stress or
buckling load to determine the instability reduction factor from a slenderness and a buckling curve.
Three approaches are considered for the calculation of such buckling stresses and buckling loads in the
present study: closed form-hand calculations, the Finite Strip Method (FSM) and the Generalized Beam
Theory (GBT). The composite action is modelled differently in each method, with the hand calculation
being the most basic approach and the GBT the most sophisticated. The Finite Element Method is not
included in the study, as it is intended for practical design methodologies rather than those used in
research.
7.4.1 Hand calculation approach
The initial aim was to assess the performance of many different built-up configurations from a simple
estimation of their ultimate resistance. Consequently, the hand calculation approach was applied first
because it can easily be implemented in a spreadsheet, allowing parametrization and quick assessment.
Local and distortional buckling stresses were determined according to the Eurocode 3 formulations and
simplified analysis introduced in subsection 7.3. The effective cross-section properties of the built-up
member were calculated by multiplying those of the individual effective cross-sections by the number
of profiles. Non-composite action was considered. For typical built-up members with common fastener
arrangements, this approach can be considered acceptable for local buckling and slightly conservative
for distortional buckling [Abbasi et al. 2018; Fratamico et al. 2018; Phan et al. 2021]. On the other hand,
member global buckling loads were determined twice, with non-composite and fully composite action.
The calculations were carried out using the classical formulations for flexural buckling from the
corresponding individual or fully compound cross-section properties of the profiles. It is noted that
considering fully composite action may lead to an overestimation of the member buckling load. The
accuracy of hand calculations for global buckling of built-up columns can be improved by using the
formulation recently developed in [Rasmussen et al. 2020; Phan et al. 2021; Phan et al. 2022]. However,
in the present study it was preferred to move to numerical methodologies, i.e., FSM and GBT (see
subsections 7.4.2 and 7.4.3).
As an example, Table 6 shows the buckling stresses and loads of the 2.5 mm thick channel section. In
this table, the rows called 2 × Single and 4 × Single, correspond to the situation of non-composite action;
and the other rows below correspond to the situation where composite action is considered. At this point,
only preliminary conclusions can be drawn for this section given the approximate nature of the
calculation: (i) flexural-torsional buckling governs when the stud consists of a single section; in built-
up sections (ii) flexural buckling about the z axis governs, if Lbz = L; (iii) when the intermediate restraint
is considered (Lbz = L / 2), flexural buckling about the y axis becomes the governing mode; and (iv) the
second intermediate restraint (Lbz = L / 3) does not seem necessary. Similar conclusions can be derived
for the other analyzed single and built-up cross-sections.
Table 6. Hand calculation buckling stress and loads for single and built-up 100 × 50 × 20 channel
sections with 2.5 mm sheet thickness
Ncr,Z
cr,L* cr,D Ncr,Y** Ncr,Z (kN) Ncr,Z (kN)
Section type (kN)
(N/mm2) (N/mm2) (kN) Lbz = L / 2 Lbz = L / 3
Lbz = L
Single 526 880 54.5 58.1 231 520
2 × Single 526 880 481 116 462 1040
4 × Single 526 880 596 596 2383 5361
Back-to-back 526 880 483 224 896 2015
Nested 526 880 483 128 512 1152
Toe-to-toe 526 880 489 440 1762 3963
Pack 4 526 880 899 2055 8218 18491
*cr,L corresponds to cross-section element with the lowest local buckling stress.
**For the single profile stud, this value corresponds to torsional-flexural buckling load. For the 2 ×
Single and 4 × Single section types, the Ncr,Y value is calculated from the member y pure flexural
buckling of the single section: 241 kN.
The hand calculation approach was applied to the cross-sections in Figure 6, considering sheet
thicknesses from 1 to 2.5 mm. This preliminary study concluded that the thickness needed for the
buildings under study should be 2 or 2.5 mm, or even larger up to 3 mm. Consequently, only these three
values are used in the subsequent calculations and analyses.
7.4.2 FSM approach
In a second phase of the study of the member, the FSM is applied considering the partial composite
action of the built-up sections by smearing the stiffness of the fasteners along the column length
[Fratamico, Schafer 2014; Abbasi et al. 2018]. The CUFSM software is used to model the compound
cross-sections. The effect of fasteners is introduced by means of springs between the individual cross-
sections, whose axial and shear stiffness is defined from the stiffness of the fastener and the distance
between fasteners presented in subsection 7.2: 10 N/mm/mm and 20 N/mm/mm for screw longitudinal
spacing of L / 3 and L / 6, respectively. In the toe-to-toe cross-sections, welding is represented by means
of additional FSM elements of the same thickness as the profile’s walls (construction thickness approach
[Fratamico, Schafer 2014]).
When local and distortional buckling stresses derived from an LBA (Linear Buckling Analysis) of the
member are used, the Eurocode 3 procedure for the calculation of the effective area is slightly different
from the standard effective width procedure described in subsection 7.3; paragraph 7.6.1(7) of [EN
16
1993-1-3 2023] provides guidance in this calculation. In the present study, (i) the minimum local
buckling stress of the signature curve derived with CUFSM is used to define the effective width of all
cross-section elements; and (ii) the minimum distortional buckling stress is used to determine the
reduced thickness of the edge stiffeners and the adjacent effective parts of the flanges, which are the
elements undergoing distortional buckling.
Member global buckling loads are determined considering simple end supports. Consequently, they are
also derived from the CUFSM signature curve. These loads are directly used to determine the member
slenderness in a similar way as it is done in the hand calculation approach. It is worth to note that the
constrained FSM option is not activated in CUFSM; consequently, the member buckling loads do not
correspond to pure member buckling modes. This can result in slightly conservative predictions of the
bucking resistance, but they are considered acceptable in the preliminary design framework of the
present study (see also comments in subsection 7.5).
Table 7 displays the buckling stresses and loads resulting from FSM for the 2.5 mm thick 100 × 50 × 20
channel cross-sections with fastener spacing L / 3 and L / 6.
Table 7. CUFSM buckling stress and loads for single and built-up 100 × 50 × 20 channel sections
with 2.5 mm sheet thickness and L / 3 and L / 6 fastener distance.
Ncr,Z Ncr,Z
Fastener cr,L cr,D Ncr,Y* Ncr,Z (kN)
Section type (kN) (kN)
spacing (N/mm2) (N/mm2) (kN) Lb = L / 3
Lb = L Lb = L / 2
Single - 672 748 54 58 222 434
2 × Single - 672 748 478 115 444 868
4 × Single - 672 748 593 593 1547 1771
L/3 672 748 454 135 466 884
Back-to-back
L/6 672 748 460 148 486 899
L/3 672 845 468 117 454 962
Nested
L/6 672 878 471 120 459 980
Toe-to-toe - 680 1068 493 420 1606 3315
L/3 672 776 566 728 1917 2561
Pack 4
L/6 672 796 599 853 2175 2963
*For the single profile stud, this value corresponds to torsional-flexural buckling load. For the 2 × Single
and 4 × Single section types, the Ncr,Y value is calculated from the member y pure flexural buckling of
the single section: 239 kN.
These results prompt the following remarks concerning the analysis procedures and column behavior:
▪ Local and distortional buckling. As expected, the local and distortional buckling stresses are
different from those obtained by hand calculation (see first columns of Table 6 and Table 7).
Nevertheless, the differences can be considered reasonable given the approximate nature and the
limitations of the methods proposed by [EN 1993-1-1 2023; EN 1993-1-3 2023]. Furthermore, as
observed in the literature [Fratamico et al. 2018], it is verified that the effect of fasteners on the local
buckling loads is null (see third column of Table 7). The effect on distortional buckling is relatively
relevant, but only in those built-up sections where fasteners constrain flange rotations, i.e. nested
and Pack-4 sections. The increase in buckling loads ranges between 3% and 17% (see fourth column
of Table 7). This can be clearly observed in Figure 7, that shows the ratio of the distortional buckling
loads corresponding to the partial and fully composite action with respect to those of non-composite
action. In Figure 7, NC refers to non-composite action, PC L / 3 and PC L / 6 refer to partial
composite actions for fastener distance L / 3 and L / 6, respectively (except for the welded toe-to-
toe section), and FC stands for fully composite action. On the other hand, welding significantly
increases the toe-to-toe section's performance with respect to distortional buckling, but not with
respect to local buckling (see Table 7).
▪ Built-up members consisting of two profiles. The flexural buckling loads about the y axis are
similar to those obtained by hand when fully composite action is considered. Since the fasteners are
not active, it is reasonable to have such similar results. They have just a small influence on the
buckling loads due to the constraining effects on the distortional buckling which is interacting with
the member y buckling. Contrarily, their impact on buckling about the z axis of back-to-back cross-
sections is relevant, especially when there are not intermediate constrains (Lbz = L). The buckling
load increases 17% with respect to non-composite situation (the 2 × Single case of Table 7). This
beneficial effect of the fasteners is offset by the detrimental effect of the interacting distortional
buckling when intermediate constrains are used (Lbz = L / 2 and Lbz = L / 3). It can be seen in Figure
8.a that the interaction of member z buckling, and symmetric distortion can be significant in the
studied built-up sections. Concerning the nested section, the effect of fasteners is small on member
z buckling loads. Its geometric configuration, with the webs close to the centroidal z axis, does not
involve any advantage in this sense. There is only small improvement for Lbz = L / 3 due to the effect
of fasteners in restraining the distortional mode component interacting with the member z buckling
mode. Finally, it is noted that the best option in terms of section efficiency is the toe-to-toe welded
solution. All this comments are summarized in Figure 9, where the ratio of the member z buckling
loads corresponding to the partial and fully composite action with respect to those of non-composite
action are shown.
▪ Pack-4 built-up section. Buckling of the Pack-4 built-up section. For the buckling about the y
axis, CUFSM shows that fasteners cannot fully constrain flexural-torsional buckling of the lateral
channel section (Figure 8.b). Furthermore, fasteners are not able to force a significant contribution
of the central sections. All this effects result in low buckling loads, that are not significantly higher
than those of the 4 × Single case. As for buckling about the z axis, the use of fasteners involves a
significant increase of the buckling loads, from 20% to 60%, but they are still far from those of fully
composite action (see Figure 9).
▪ Fasteners spacing. It is also observed that decreasing the spacing between fasteners from L / 3 to L
/ 6 produces a moderate increase of z axis buckling loads in the case of back-to-back built-up
sections (maximum 10%); and a relatively significant increase in the case of Pack-4 sections
(between 13% and 17%).
Figure 7. Ratios between the distortional buckling loads corresponding to the partial and fully
composite actions, and those of non-composite action (100 × 50 × 20 profiles with 2.5 mm sheet
thickness)
18
(a) Back-to-back section

(b) Pack-4 section (Lby = L)
(Lbz = L / 3)
Figure 8. Interaction between buckling modes
Figure 9. Ratios between the CUFSM member z buckling loads corresponding to the partial and
fully composite actions, and those of non-composite action (100 × 50 × 20 profiles with 2.5 mm
sheet thickness)
7.4.3 GBT approach
The approach presented in subsection 7.4.2 can be improved by introducing the actual distribution of
discrete fasteners in the LBA instead of using smeared properties. The FEM (Finite Element Method) is
probably the most powerful tool for accurately dealing with discrete fasteners. However, it is considered
to be unpractical for daily design of this type of structures. Alternatively, various practical
methodologies for determining the buckling loads have been recently developed, such as the compound
Finite Strip Method [Abbasi 2018; Akshay 2021] or the effective rigidity approach [Rasmussen et al.
2020; Phan et al. 2021; Phan et al. 2022]. In the present study, GBT is applied because it allows the
numerical model to be created as quickly as with the FSM and, at the same time, explicitly account for
discrete fasteners as in FEM. Furthermore, GBT is also used herein to assess the effectiveness of EFGs
(End Fastener Groups), which consist of two columns of 8 screws at each end section. The study of this
effect cannot be tackled with standard FSM [Fratamico, Schafer 2014; Rasmussen et al. 2020].
The GBT linear buckling analyses have been carried out by means of the formulation described in
[Bonada et al. 2021]. All deformation modes (conventional, natural shear, transverse extension and local
shear modes) have been included. Each fastener is directly modeled through three 1-Dimensional linear
springs (x, y and z global directions) which connect the nodal displacement (in global coordinates) of
the two sectional nodes where the screw is located. Therefore, the axial and shear screw stiffness can be
reproduced. The stiffness values are those presented in subsection 7.2.
The GBT buckling loads are used to determine the corresponding slenderness, buckling reduction factors
and final resistances in the same way as in the FSM approach (subsection 7.4.2). Table 8 displays the
GBT buckling stresses and loads for single and built-up 100 × 50 × 20 channel sections with 2.5 mm
sheet thickness, and Table 9 exhibits the ratio between such values and those of FSM.
Table 8. GBT buckling stresses and loads for single and built-up 100 × 50 × 20 channel sections
with 2.5 mm sheet thickness.
Ncr,Z
Ncr,Z
Fastener cr,L cr,D N cr,Y * (kN)
Section type (kN)
spacing/EFG (N/mm2) (N/mm2) (kN) Lb = L /
Lb = L
2
Single - 650 729 57 61 236
2 × Single - 650 729 114 122 472
4 × Single - 650 729 228 244 944
L/3 650 712 451 129 480
Back-to-
L/6 650 712 473 148 504
back
L / 6 + EFG 649 716 474 189 564
Toe-to-toe - 650 993 496 407 1564
L/3 649 712 521 648 1462
Pack 4 L/6 650 712 597 806 2102
L / 6 + EFG 649 ** 703 1258 2529
*For the single profile stud, this value corresponds to torsional-flexural buckling load
** Not detected due to interaction with other modes
Table 9. Ratios between GBT vs FSM buckling stresses and loads for single and built-up 100 × 50
× 20 channel sections with 2.5 mm sheet thickness
Section cr,L cr,D Ncr,Y Ncr,Z (kN) Ncr,Z (kN)
Fastener spacing
type (N/mm2) (N/mm2) (kN) Lb = L Lb = L / 2
Single - 0.97 0.97 1.04 1.05 1.06
Back- L/3 0.97 0.95 0.99 0.96 1.03
to-back L/6 0.97 0.95 1.03 1.00 1.04
Toe-to-
- 0.96 0.93 1.01 0.97 0.97
toe
L/3 0.97 0.92 0.92 0.89 0.76
Pack 4
L/6 0.97 0.89 1.00 0.95 0.97
Figure 10 contain similar results than Figure 9, but from GBT.
20
Figure 10. Ratios between the GBT member z buckling loads corresponding to the partial and fully
composite actions, and those of non-composite action (100 × 50 × 20 profiles with 2.5 mm sheet
thickness)
The observation of Table 8, Table 9, and Figure 10 leads to the following remarks:
▪ The study on the FSM buckling loads in subsection 7.4.2 showed that the performance of the built-
up nested section is rather lower than in the other options; consequently, they were not analyzed
with GBT and their results not included in Table 8.
▪ The GBT member z buckling loads with two intermediate constraints (Ncr,Z with Lb,Z = L / 3) are not
included in Table 8 because the interaction of member and distortional buckling is strong. This
makes identifying the modes and the corresponding buckling loads very difficult. Furthermore, as
observed in subsections 7.3.1 and 7.3.2, the member z buckling mode is not governing the design.
▪ Table 9 shows a reasonable agreement between the FSM and GBT buckling loads: the mean GBT /
FSM ratio is 0.96, and the standard deviation is 0.06. It is observed that in built-up sections, the
GBT results are slightly lower than those of FSM. This is attributed to the fact that modeling discrete
fasteners decreases the degree of restrain of sectional buckling modes. In this sense, it is noted that
the agreement between GBT and FSM is better when fasteners are closer (L / 6) and such sectional
phenomena are more constrained in GBT.
▪ The agreement between FSM and GBT supports the comments regarding the behavior of the built-
up sections based on the FSM results presented in subsection 7.4.2.
▪ Table 8 and Figure 10 show that EFGs have a relevant positive effect on the behavior of built-up
sections [Fratamico et al. 2018; Rasmussen et al. 2020]. The increase in buckling loads can be up to
25% and 50% for the back-to-back and pack-4 configurations, respectively.
7.5 Stud resistance
The buckling resistance of the investigated built-up sections was determined from the buckling stresses
and loads obtained by hand calculation, the FSM and the GBT. However, the study of the feasibility of
the cold-formed chord studs was finally based on the FSM buckling values. The hand calculation
procedure applied does not allow partial composite action to be considered and can only estimate the
studs’ resistance. Furthermore, the GBT results are sufficiently similar to those of the FSM to focus only
on the latter and avoid repetitions. Only the built-up sections with EFGs were calculated from GBT
buckling results.
For the sake of completeness, hand-calculated and FSM resistances are compared in Table 10 as an
example. The table corresponds to 100 × 50 × 20 channel built-up sections with 2.0 mm sheet thickness,
made of S350GD steel, without intermediate restraints and L / 6 fastener spacing. This time, the results
of 2.5 mm channel profiles are not shown because their section is fully effective, which does not allow
assessing the effect of the different buckling stresses on the final effective area. In this sense, it is
observed that the effective area determined from the FSM local and distortional buckling stresses is
slightly lower than that obtained from the buckling stresses calculated with the formulations of [EN
1993-1-3 2023]. On the other hand, the member resistances determined from FSM buckling loads lie
between those corresponding to the hand-calculated non-composite and fully composite action, as
expected. In the case of toe-to-toe, the FSM resistance is almost the same as the FC hand-calculated
value, because welding involves fully-composite action. Contrarily, in the pack-4 option the FSM result
is more similar to the NC hand-calculated value. This is due to the factors discussed in section 7.4.2.
Table 10. Effective areas and buckling resistances determined from hand-calculated and FSM bucking stresses
and loads. Results for 100 × 50 × 20 channel built-up sections with 2.0 mm sheet thickness, made of S350GD
steel, without intermediate restraints and L / 6 fastener spacing.
Effective area (mm2) Buckling resistance (kN)
Hand Hand Hand
FSM FSM
Section calculated Hand/ calculated calculated Hand- Hand-
buckling buckling
buckling FSM buckling buckling NC/FSM FC/FSM
stresses loads
stresses loads-NC loads-FC
Back-to-
784 732 1.07 72.06 122.05 92.79 0.59 1.32
back
Toe-to-toe 784 751 1.04 72.06 166.87 161.47 0.43 1.03
Pack-4 1567 1495 1.05 281.58 368.81 295.79 0.95 1.25
The final buckling resistance of the built-up studs is shown in Figure 11 and Figure 12. Results are
presented for 2.0, 2.5 and 3.0 mm thickness and L / 6 fastener spacing. It is noted that the use of L/3 was
finally dismissed due to its low efficiency. Before going into the discussion of the feasibility of the chord
studs, some preliminary comments on the performance of the built up sections are included first:
▪ Increasing the thickness and quality of the steel has a significant positive effect on the ultimate
buckling resistance of the studs.
▪ In general, strength of 150 × 50 × 20 studs is significantly higher than 100 × 50 × 20 ones: from
15% to 50% for pack-4 sections, and from 5% to 35% for toe-to-toe sections. However, for back-
to-back sections, the strength increase ranges between 0% (or even slightly lower) and 5%; this is
due to its higher tendency to suffer from sectional instabilities (see, for instance, the distortional
buckling stresses of Table 7).
▪ The use of an intermediate restraint (Lbz = L / 2) is relevant and a key factor in the performance of
back-to-back and toe-to toe built-up cross-sections because it constraints member z buckling, which
is the weak buckling axis and the one under the effects of the fasteners’ partial composite action.
This restraint is unnecessary in pack-4 cross-sections where member z buckling is not governing.
▪ The effect of EFGs in back-to-back sections is significant when there are not intermediate restraints
and the member z buckling effects are relevant. For instance, the increase in resistance in
unrestrained channel built-up studs ranges from 11% to 21%, which is an improvement similar to
that observed in experiments [Fratamico et al. 2018].
▪ It was expected that the EFGs would improve the behavior of pack-4 sections with respect to
member y buckling. Intermediate restraints cannot be used to constraint buckling about the y axis
and, as a consequence, the only way to get better resistances is by adding fasteners. In the end,
22
however, the effect of EFGs is limited, with improvements ranging from 6% to 9% in 100 × 50 ×
20 channel pack-4; and from 1% to 3% in 150 × 50 × 20 channel pack-4. In Table 9 it can be seen
that EFGs have a significant effect on member z buckling, but it is lower in the case of member y
buckling.
▪ The stud resistances have been determined from member buckling loads of modes showing high
interaction between sectional and member instabilities, especially in the case of 150 × 50 × 20 studs
with intermediate restraints (Lbz = L / 2). Consequently, the resistance of the built-up member is
doubly penalized: through the reduction in the effective area, and the “member” buckling load
corresponding to buckling modes involving several coupled instabilities. This effect is more
significant when the studs with EFG are calculated, because the interaction between modes higher
in GBT buckling analyses, as discussed in subsection 7.4.3. Consequently, there are some strength
values of EFG studs that resulted to be similar or lower than those of the same studs without EFG,
which have been determined from FSM member buckling loads. These cases are not considered in
the discussion of the two previous comments. This aspect is to be improved in future researches.
▪ Finally, the most important remark is that the maximum axial capacity of the chord studs analyzed
(Figure 6.f) is 747 kN.
Figure 11. Bucking resistance of 100 × 50 × 20 built-up channel sections

Figure 12. Bucking resistance of 150 × 50 × 20 built-up channel sections

8 Main results of the global structural analyses of the prototype buildings
This section presents the main results of the iterative global structural analyses of the prototype buildings
(sections 5 and 6), and compares them with the axial capacity of the most demanded structural elements,
namely the chord studs (subsection 7.5). Table 11 displays the main results of the global structural
analyses of the 5, 7 and 10-story prototype buildings, respectively.
24
Table 11. Main results of the global structural analyses for the prototype buildings
5-story building 7-story building 10-story building
S,475 𝑵𝐦𝐚𝐱
𝐄𝐝𝐱 𝑵𝐦𝐚𝐱
𝐄𝐝𝐲 𝑵𝐦𝐚𝐱
𝐄𝐝𝐲 𝑵𝐦𝐚𝐱
𝐄𝐝𝐲
Comb. Comb. Comb. Comb. Comb. Comb.
(kN) (kN) (kN) (kN) (kN) (kN)
154 V 152 VI 218 V 212 VI 307 V 302 VI
0.5 201 V 253 VI 244 V 297 VI 407 V 430 VI
229 V 273 VI 261 V 306 VI 499 V 480 VI
216 V 213 VI 306 V 297 VI 432 V 423 VI
0.75 338 V 416 VI 411 V 487 VI 699 V 724 VI
338 V 488 VI 482 V 560 VI 946 V 943 VI
277 V 273 VI 391 V 381 VI - - - -
496 V 594 VI 603 V 701 VI - - - -
1
575 V 658 VI 650 V 845 VI - - - -
672 V 660 VI - - - - - - - -
max
In Table 11, 𝑁Ed is the maximum design value of the demanding axial force in the chord studs, and
“Comb.” refers to the load combination to which it corresponds (equation (1)); obviously, subindexes x
and y denote the horizontal directions (Figure 4), respectively. For the 10-story building, the cases for
max max
S,475 = 1.0 are omitted, as the result is obvious. The values of 𝑁Ed and 𝑁Ed that exceed the axial
capacity of the chord studs (subsection 7.5) are highlighted, as they show an insufficient capacity.
Table 11 shows that the 5, 7 and 10-story buildings withstand moderate ground motions (S,475 = 0.5 g),
and 5, and 7-story buildings resist also high ones (S,475 = 0.75 g); finally, 5-story buildings can even
resist earthquakes with S,475 = 1 g. Two component built-up chord studs (Figure 6.c and Figure 6.e)
suffice in the 5-story building (under moderate seismicity); conversely, packs of four (Figure 6.f) are
needed in the 7 and 10-story buildings.
max
In Table 11, most of the axial force 𝑁Ed is contributed by the seismic action (equation (1)) in virtually
all the cases; this indicates that there is a certain risk of uplift, and hold-down anchors might be
necessary.
This paragraph describes the characteristics of the main structural members other than the chord studs.
Intermediate studs are made of single profiles in the upper stories, and nested profiles in the bottom
ones. Strap braces are made of steel grade S 250 GD; for the 5-story building the required area in the
first story ranges between 1000 mm2 (S,475 = 0.5 g) and 3000 mm2 (S,475 = 1.0 g), for the 7-story
building the area is 1100 mm2 (S,475 = 0.5 g) and 2150 mm2 for (S,475 = 0.75 g). For the 10-story
building such value is 2300 mm2 (S,475 = 0.5 g).
9 Conclusions
This paper studies the seismic bounds of structures made of customary light-gauge steel framing panels
with flat strap braces (dissipative elements) and several types of built-up Cold Formed Steel chord studs;
in this context, “customary” refers to ordinary steel (yield point not exceeding 350 MPa), thin plates
(not more than 3 mm), and rather thin walls (slightly above 100 or 150 mm thickness). Three prototype
buildings representative of the short to mid-rise ranges (5, 7 and 10 stories) have been designed
according to the forthcoming European seismic design code, by accounting for Ultimate and
Serviceability Limit States. Three seismicity levels have been considered, from moderate (Peak Ground
Acceleration 0.2 g) to high (0.3 and 0.4 g); in all the cases, soil type is B. The chord studs are designed
according to the European regulations by considering three approaches to determine the buckling loads:
closed-formed expressions based on classical buckling theory, the Finite Strip Method, and the
Generalized Beam Theory.
The major conclusions of this research are:
▪ The most critical structural elements are the chord studs of the lowest levels. Such members are
connected to the dissipative elements (strap braces); therefore, are to be protected against brittle
buckling failure by equilibrium capacity conditions. Being limited to pure Cold Formed Steel
sections, this proves to be a strong limitation for the design of short to mid-rise buildings.
▪ The 5, 7 and 10-story buildings withstand moderate ground motions, 5 and 7-story buildings also
high ones (0.3 g), and 5-story buildings even the highest earthquake (0.4 g).
▪ The performed research corroborates that, for highly demanding situations (mid-rise buildings or
high seismicity), high capacity bracing panels are required: thick walls (150 mm), four component
built-up chord studs, “thick” walled profiles (3 mm), moderate to short fastening spacing (minimum
L / 6), and one intermediate in-plane restraint.
▪ By neglecting the stiffening effect of the nonstructural components (cladding and partitioning
walls), the fundamental periods of the buildings analyzed are significantly longer than those arising
from the empirical expressions in the code. This circumstance makes that the code limitation of the
period to two times the aforementioned one becomes overconservative; it has been ignored in this
study.
▪ Built-up cross-sections undergo interaction between sectional and member instabilities (more than
in single sections). This issue, together with the fasteners partial action, cause a lower performance
than the full composite action.
▪ This research shows that the Generalized Beam Theory is helpful for calculating buckling loads,
allowing efficient modelling, addressing all relevant buckling phenomena and accurately
considering the fasteners' partial composite action (in complex situations, such as the pack-4
sections with End Fastener Groups). It overcomes the limitations of the manual calculation
procedures and the Finite Strip Method. Furthermore, it may also be an excellent tool to deal with
the modes interaction, due to its capacity to isolate modes (or groups of).
Further research includes eliminating the customary condition; this implies higher grade steel, thicker
plates, and larger profiles. As well, solutions to improve the performance of pack-4 section will be
investigated.
Acknowledgement
The PhD of the first author is funded by the Generalitat de Catalunya (Catalonian Government), grant
No. 2020 DI 089. This research has been also partially funded by the Spanish Research Agency (AEI)
of the Ministry of Science and Innovation (MICIN) through project with reference: PID2020-
117374RB-I00 / AEI / 10.13039/501100011033. These supports are gratefully acknowledged.
References
[1] Abbasi M, Khezri M, Rasmussen KJR, Schafer BW. 2018. Elastic buckling analysis of cold-formed steel
built-up sections with discrete fasteners using the compound strip method. Thin-Walled Structures. 124:58-
71.
[2] AISI S100-16. 2016. North American Specification for the Design of Cold-Formed Steel Structural
Members. American Iron and Steel Institute.
[3] AISI S240-15. 2015. North American Standard for Cold-Formed Steel Structural Framing. American Iron
and Steel Institute.
[4] AISI S400-20. 2020. North American Standard for Seismic Design of Cold-Formed Steel Structural
Members. American Iron and Steel Institute.
[5] ASCE/SEI 7-22. 2022. Minimum Design Loads and Associated Criteria for Building and Other Structures.
American Society of Civil Engineers, Structural Engineering Institute.
[6] Bonada J, Casafont M, Roure R, Pastor MM. 2021. Geometrically nonlinear analysis of perforated rack
columns under a pure compression load by means of Generalized Beam Theory. Thin-Walled Structures.
166:108102.
[7] EN 10346 Continuously hot-dip coated steel flat products for cold forming – Technical delivery conditions.
European Committee for Standardization (2015).
[8] EN 1990. 2005. Basis of structural design. European Committee for Standardization.
[9] EN 1991-1-1. 2002. Actions on structures. Part 1-1: General actions. Densities, self-weight, imposed loads
for buildings. European Committee for Standardization.
[10] EN 1991-1-4. 2005. Actions on structures. Part 1-4: General actions. Wind actions. European Committee
for Standardization.
[11] EN 1993-1-1. 2006,2023. Design of steel structures. Part 1-1: General rules and rules for buildings.
European Committee for Standardization.
26
[12] EN 1993-1-3. 2006,2023. Design of steel structures. Part 1-3: General rules. Supplementary rules for cold-
formed members and sheeting. European Committee for Standardization.
[13] EN 1998-1. 2023. Design of structures for earthquake resistance. Part 1: General rules, seismic actions and
rules for buildings. European Committee for Standardization.
[14] ETA (European Technical Assessment). 2020. European Assessment Document 340452-00-02.04 Building
kit based on massive panels with metal frame structure. ITeC (Institut de Tecnologia de la Construcció de
Catalunya).
[15] Ford P. 2016. The Steel Framing Industry Association (SFIA). https://buildsteel.org/architect-
profession/high-can-construct-building-cold-formed-steel-framing-answer-may-surprise/ (last accessed
December 2022).
[16] Fratamico DC, Schafer BW. 2014. Numerical studies on the composite action and buckling behavior of
built-up cold-formed steel columns. 22nd International Specialty Conference on Cold-Formed Steel
Structures.
[17] Fratamico DC, Torabian S, Zhao X, Rasmussen KJ, Schafer BW. 2018. Experimental study on the
composite action in sheathed and bare built-up cold-formed steel columns. Thin-Walled Structures.
127:290-305.
[18] Landolfo R. 2019. Lightweight steel framed systems in seismic areas: Current achievements and future
challenges. Thin-Walled Structures. 140:114-131.
[19] LGSF. 2022. Light Gauge Steel Framing Global Market Report. The Business Research Company.
[20] Li Z, Abreu JCB, Leng J, Ádány S, Schafer BW. 2014. Constrained finite strip method developments and
applications in cold-formed steel design. Thin-Walled Structures. 81:2-18.
[21] Macillo V, Iuorio O, Terracciano MT, Fiorino L, Landolfo R. 2014. Seismic response of Cfs strap-braced
walls: Theoretical study. Thin-Walled Structures. 85:301-312.
[22] Madsen RL, Nakata N, Schafer BW. 2011. CFS-NEES building structural design narrative.
[23] Mahar AM, Jayachandran SA, Mahendran M. 2021. Global buckling strength of discretely fastened back-
to-back built-up cold-formed steel columns. Journal of Constructional Steel Research. 187:106998.
[24] Nakata N, Schafer BW, Madsen RL. 2012. Seismic design of multi-story cold-formed steel buildings: The
CFS-NEES archetype building. Structures Congress.
[25] Pham HS, Moen CD. 2015. Stiffness and Strength of Single Shear Cold-Formed Steel Screw-Fastened
Connections. Virginia Tech.
[26] Phan DK, Rasmussen KJ. 2019. Flexural rigidity of cold-formed steel built-up members. Thin-Walled
Structures. 140:438-449.
[27] Phan DK, Rasmussen KJ, Schafer BW. 2021. Tests and design of built-up section columns. Journal of
Constructional Steel Research. 181:106619.
[28] Phan DK, Rasmussen KJ, Schafer BW. 2022. Numerical investigation of the strength and design of cold-
formed steel built-up columns. Journal of Constructional Steel Research. 193:107276.
[29] Rasmussen KJ, Khezri M, Schafer BW, Zhang H. 2020. The mechanics of built-up cold-formed steel
members. Thin-Walled Structures. 154:106756.
[30] Schafer BW. 2011. Cold-formed steel structures around the world. Steel Construction. 4(3):141-149.
[31] Schardt R. 1994. Generalized beam theory. An adequate method for coupled stability problems. Thin-
walled structures. 19(2-4):161-180.
[32] Sharafi P, Mortazavi M, Usefi N, Kildashti K, Ronagh H, Samali B. 2018. Lateral force resisting systems
in lightweight steel frames: Recent research advances. Thin-Walled Structures. 130:231-253.
[33] Silvestre N, Camotim D. 2002. First-order generalized for arbitrary orthotropic materials. Thin-Walled
Structures. 40(9):755-789.
[34] Staib G, Dörrhöfer A, Rosenthal M. 2008. Components and systems. Modular construction. Design –
Structure - New technologies. Institut für Internationale Architektur-Dokumentation.
[35] Torabian S, Nia ZS, Schafer BW. 2016. An Archetype Mid-Rise Building for Novel Complete Cold-
Formed Steel Buildings. International Specialty Conference on Cold-Formed Steel Structures.
[36] Taranu G, Ungureanu V, Nagy Z, Alexa-Stratulat M-S, Toma I-O, Luca S-G. 2023. Shake table test and
numerical analyses of a thin-walled Cold-Formed Steel structure: Part 1. Investigation of the structural
skeleton without claddings. Thin-Walled Structures. 182:110258.
[37] United Nations Human Settlements Programme (UN-Habitat). 2022. World Cities Report 2022.
https://unhabitat.org/sites/default/files/2022/06/wcr_2022.pdf (last accessed December 2022).
[38] Yilmaz F, Mojtabaei SM, Hajirasouliha I, Becque J. 2023. Behaviour and performance of OSB-sheathed
cold-formed steel stud wall panels under combined vertical and seismic loading. Thin-Walled Structures.
183:110419.
[39] Zhang J, Singh A, Hutchinson TC, Wang X. 2022. Seismic Analysis of the 10-story CFS-NHERI building.
Cold-Formed Steel Research Consortium Colloquium.
List of acronyms
CFS: Cold-Formed Steel

EFG: End Fastener Group
EPS: Expanded PolyStyrene
FC/NC/PC: Fully composite/non-composite/partial composite action
FEM: Finite Element Method
FSM: Finite Strip Method
GBT: Generalized Beam Theory
LBA: Linear Buckling Analysis
PGA: Peak Ground Acceleration
RC: Reinforced Concrete
List of symbols
A/Aeff: Gross/effective cross-section area (equation (5))
dr,SD: Drift for the Significant Damage (SD) limit state (Table 4)
dr/ds: Average/reduced lateral displacements (Table 4)
EEd: Demand internal force for seismic combinations (equation (1))
EEd,G: Contribution of the non-seismic actions to EEd (equation (4))
ENfy: Internal force the brace straps corresponding to the axial yielding capacity of EX/EY: Seismic action
(earthquake) in X/Y directions (equation (1))
ex/ey: Eccentricity (in x/y direction) between the gravity (G) and rotation (R) centers (subsections 4.2 and 6.1)
FA/FT/F/F: Plateau/topography/short/intermediate period site amplification factors (FA = 2.5, FT = 1) (subsection
6.4)
Fb: Base shear force (subsection 6.1, equation (2))
fyb: Basic yield strength of steel (equation (5))
G/Q: Permanent (dead)/variable (live) loads (equation (1))
G/R: Centers of gravity/rotation (section 3, subsections 4.2 and 6.1)
Hb: Building height (Figure 1.b, subsection 6.5)
hs: Story height (Table 4)
L: Member (chord stud) length (section 7)
Lb/Lby/Lbz: Flexural buckling length (with respect to y/z sectional axes) (section 7)
ls: Radius of gyration of the floor mass in plan (square root of the ratio between the polar mass moment of inertia
-with respect to point G- and the floor mass) (subsection 6.1)
M/K: Mass/stiffness matrices in each horizontal direction (subsection 6.5)
m: Building mass corresponding to its building seismic weight (subsection 6.1, equation (2))
Nb,Rd: Axial design strength to flexural buckling (equation (5))
Ncr,Y: Axial critical force for pure flexural buckling with respect to sectional axis Y (section 7)
Ptot/Vtot: Vertical gravity/horizontal seismic forces (Table 4)
q: Behavior factor (equivalent to the response modification factor R in the American documents) (subsection 6.4)
Rq(T): Reduction factor that establishes the reduced design spectrum (Sr) (subsection 6.4)
rx/ry: Torsional radius (square root of the ratio between torsional and lateral stiffness) (subsection 6.1)
Se/Sr: Design/elastic/reduced spectral ordinate (subsection 6.4)
S/S: Ground acceleration for the plateau/period 1 s (subsection 6.4)
S,475/S,475: Ground acceleration on rock for 475 years return period for the plateau/period 1 s (subsection 6.4)
S: Parameter classifying the seismicity in four levels (very low, low, moderate and high) (subsection 6.2)
T: Period (subsections 6.4 and 6.5)
TA, TB, TC, TD: Corner periods of the design spectra (subsection 6.4)
T: Intermediate period (T = 1 s) (subsection 6.4)
T1: First mode period in each horizontal direction (fundamental period) (subsection 6.4 and 6.5)
T1,est: Fundamental period of the building estimated from code empirical expressions (equation (3))
WX/WY: Wind action in X/Y directions (equation (1))
x, y, z: Coordinates for global (Figure 1, Figure 2, Figure 3, Figure 4) and local reference axes; regarding local
axes, x is longitudinal, and y/z are principal sectional axes (Figure 6)
: Dimensionless coefficient limiting the value of the spectral ordinates (subsection 6.4)
χ: Reduction factor for flexural buckling (equation (5))
γM1: Partial safety factor for resistance of members to instability (equation (5))
: Damping correction factor (subsection 6.4)
: Dimensionless coefficient being equal to either 1 or 0.85 (subsection 6.1, equation (2))
: Interstory drift sensitivity coefficient (Table 4)
cr,Dcr,L: Distortional and local buckling stress (section 7)
28
cr,Y: Axial critical stress for pure flexural buckling with respect to sectional axis Y (section 7)
/: Eigenvalue (natural frequency) and eigenvector (modal shape) (subsection 6.5)
rm: Material overstrength factor (equation (4))

Customary Light-Gauge Steel Framing Construction W

Uploaded by

Copyright:

Available Formats

Customary Light-Gauge Steel Framing Construction W

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Customary Light-Gauge Steel Framing Construction W

Uploaded by

Copyright:

Available Formats

Customary light-gauge steel framing construction

with at strap bracing. Seismicity limits for short to

Posted Date: March 16th, 2023

(a) Elevation (front view) of the 10-story

(a) Cee (b) Back-to-back (c) Nested

(d) Face-to-face (toe-to-toe) (e) Pack 4

(a) Back-to-back section

Figure 11. Bucking resistance of 100 × 50 × 20 built-up channel sections

Figure 12. Bucking resistance of 150 × 50 × 20 built-up channel sections

CFS: Cold-Formed Steel

You might also like