Materials and Structures (2009) 42:301–320

DOI 10.1617/s11527-008-9463-x

ORIGINAL ARTICLE

Analysis of fully anchored light-frame timber

shear walls—elastic model
Bo Källsner Æ Ulf Arne Girhammar

Received: 12 February 2008 / Accepted: 23 December 2008 / Published online: 22 January 2009
Ó RILEM 2009

Abstract In order to stabilize light-framed timber members and shear deformations in the sheets, and
buildings against lateral loads, the diaphragm action also the effect of vertical loads on the shear wall, both
of roofs, floors and walls is often used. This paper with respect to tilting and second order effects, on the
deals with an elastic analysis model for fully horizontal load-bearing capacity and displacement
anchored sheathed wood frame shear walls. The are evaluated. The stress distribution and the reaction
model is based on the assumption of a linear elastic forces at the ends of the different framing members
load-slip relation for the sheathing-to-framing joints. are derived. The elastic model is experimentally
Only static loads are considered. The basic structural verified and an illustrative example is given.
behaviour and assumptions for the elastic model are
elucidated. Formulas for the load-bearing capacity Keywords Shear walls
Wall diaphragms

and the deformation of the shear walls in the ultimate Elastic model
Full anchorage
Racking
and serviceability limit states, respectively, are load capacity
Wall displacement

derived. Both a discrete point description and a Fastener displacement
continuous flow per unit length modelling of the
fasteners are discussed. Also, the forces and dis-
placements of the fasteners and sheathing are derived.
The effect of different patterns and spacing of the
fasteners on the capacity and displacement of the wall 1 Introduction
is illustrated. The influence of flexible framing
In order to stabilize light-framed timber buildings
against lateral loads, the diaphragm action of roofs,
B. Källsner
School of Technology and Design, Växjö University, floors and walls is often used. This paper is mainly
Växjö, Sweden focusing on the structural behaviour of wall dia-
phragms or shear walls. In a typical timber frame
B. Källsner
building subjected to wind, the wall diaphragms
SP Wood Technology – Technical Research Institute
of Sweden, Box 5609, 11486 Stockholm, Sweden undergo racking loads. These loads act in the plane of
the wall and cause shear of the wall. A typical shear
U. A. Girhammar (&) wall consists of a wood frame that is composed of
Department of TFE – Civil Engineering, Faculty of
studs, top and bottom rails, and sheathing panels, see
Science and Technology, Umeå University, 90187 Umeå,
Sweden Fig. 1. The definitions of the details of the wall panel
e-mail: [email protected] used in the figure are as given by the European
302 Materials and Structures (2009) 42:301–320

Fig. 1 A shear wall unit Top rail Framing joint

built up of a timber frame
and a sheet. The sheet is
connected to the timber
frame by means of Sheathing-to-framing Stud Sheathing
joint
mechanical fasteners. The
centre distances of the sps
sis h
fasteners can vary in the
general case. Here they are
assumed to be sr along the
bottom and top rail, sps
along the perimeter studs,
and sis along the
intermediate stud,
respectively
Bottom
sr rail

Standard EN 594 [1]. The top and bottom rails are rail is anchored to the floor/foundation. Due to
connected to the studs to form the frame and the economic reasons the building industry, at least in
sheathing is connected to the frame by nails or other the Nordic countries, desires to reduce the number of
discrete fasteners. tie-downs, nails and screws to a minimum.
In order to explain the structural behaviour of wall By a fully anchored shear wall is meant a wall
diaphragms some theoretical background is needed. where the vertical stud on the tension side is fully
The purpose of this paper is to present the basic anchored to the floor or foundation. In a partially
theories for fully anchored wall diaphragms in order anchored shear wall, the vertical stud is only partially
to elucidate the fundamental behaviour of and anchored to the floor or foundation. In the latter case,
assumptions for shear walls and to calculate the the bottom rail is usually assumed to be anchored to
capacity and stiffness of the walls. the floor/foundation.
Even though the elastic theory for fully anchored
shear walls without imperfections, as presented in the 1.1 Previous research
present paper, has been used by the authors for many
years it has not yet been published for the interna- Extensive research work concerning sheathed light-
tional research community in a refereed journal. A framed timber shear walls and diaphragms has been
companion paper [2], based on the elastic theory carried out since the late 1920’s. The activities have
presented in this paper, evaluates the influence of focused on both experimental and theoretical model-
imperfections on the horizontal displacements in the ling approaches, including testing methods. Most
serviceability limit state. research has been conducted for shear walls with
Shear walls are constructed in different ways mechanical sheathing-to-framing joints, but work has
depending on local traditions. They can be prefabri- also been performed on shear walls where the sheets
cated in a factory or built on site. The prefabricated have been adhesively jointed to the frame. Both
shear walls can consist of one or several wall analytical models for hand calculation and computer-
assemblies of different sizes. The structural behav- based numerical models, usually based on the finite
iour of wall diaphragms is to a large extent element method, have been developed. Formulations
determined by the sheathing-to-framing joints and based on both linear elastic and non-linear elastic-
how the diaphragms are connected to the surrounding plastic properties have been proposed. Most models
structure. Of particular importance is the anchoring of discussed are applicable only to shear walls fully
the shear wall to the floor or foundation. Sometimes anchored to the floor/foundation. But a few models
tie-downs are used for anchorage of the end studs of applicable to shear walls allowing for uplift of the
the shear wall. On other occasions only the bottom studs and/or of the bottom rail have also been
Materials and Structures (2009) 42:301–320 303

discussed. Shear walls with or without openings have perimeter studs are equal ðb=sr ¼ h=sps Þ. For exam-
been studied, though there are very few models that ple, for a shear wall where the height of the sheathing
account for the effect of openings. Static as well as is equal to twice the width, the spacing of the
dynamic loadings, usually earthquake type of load- fasteners along the perimeter stud will then need to be
ings, have been considered. Formulas have been twice the spacing along the rail (these conditions are
developed or analyses have been carried out to reflect exactly valid in cases where there are nailing only
the influence vertical loads have on the shear walls. In along the perimeter of the sheathing). To determine
rare cases, the tying down effect of transverse walls the load-carrying capacity of the wall, the equiva-
inter-connected to the shear walls has been investi- lence of the external and internal energy is used.
gated or modelled. Some of those models have been Another experimentally based model was that by
presented to the international research community Easley et al. [5], where the fasteners along the vertical
and some have not. studs were assumed to be loaded only in the vertical
Many models have been proposed in order to direction. This assumption is an approximation.
analyze and predict the performance of wood based The present method is general in nature and is
shear walls and diaphragms subjected to static lateral based on a consistent elastic model, where the same
loads, eg. Foschi [3], Tuomi and McCutcheon [4], fundamental assumptions as above are used, i.e.
Easley et al. [5], Itani and Cheung [6], Källsner [7], hinged connected framing members, and rigid fram-
McCutcheon [8], Gupta and Kuo [9, 10], Schmidt and ing members and sheathing, but with no other
Moody [11], Falk and Itani [12], Filiatrault [13], simplifying or approximating conditions.
Dolan and Foschi [14], Kasal and Leichti [15], Gupta and Kuo [9] and Filiatrault [13] developed
Murakami et al. [16], Foschi [17], and Judd and numerical solutions to the problem using non-linear
Fonseca [18]. Various approaches were adopted in properties of the sheathing-to-framing joints. Later,
these models, which ranged from linear to nonlinear Gupta and Kuo [10] included the effect of the uplift of
analyses and from strain energy approaches to finite the studs in their model (assuming a fully anchored
element analyses. sill). This uplift was counteracted, besides by the
In the literature, there are some simple hand sheathing-to-framing joints along the bottom rail, also
calculation models for the evaluation of the horizontal by the framing joints. Yoon and Gupta [20] generalized
stability of sheathed timber-framed houses. However, this model and presented a closed analytical solution
they are often based on semi-empirical assumptions. for the analysis of shear walls, with and without uplift,
Most models are of an elastic kind presupposing full assuming linear load-slip behaviour of the sheathing-
anchorage of the wall to the floor/foundation; the to-framing joints, and a numerical solution in case of
studs being pinned connected to the top and bottom non-linear joint behaviour. Salenikovich [21] pre-
rails, and the framing members and sheathing being sented a mechanical model of sheathed shear walls
assumed rigid, see e.g. Tuomi and McCutcheon [4], without hold-downs. He analyzed both elastic and
Källsner [7] och Åkerlund [19]. Easley et al. [5] and plastic conditions for the sheathing joints. Ni and
McCutcheon [8] extended these models to include Karacabeyli [22, 23] developed a mechanics-based
non-linear characteristics of the sheathing-to-framing method to account for the effects of vertical loads and
connections. However, the models of Easley and perpendicular walls on the performance of shear walls
Tuomi et al. were not fully correct or based on without tie-down connections.
assumptions limiting their applicability. For the sake of completeness, early models and
Tuomi and McCutcheon [4] assumed, in addition methods developed and used specifically in Germany
to the assumptions mentioned above, that the four is also discussed. According to Henrici [24], the
fasteners in the corners displace along the diagonals German models for shear walls can basically be
of the sheathing. This assumption is reasonable for divided into two overall types of structural systems:
common designs of shear walls, but principally The shear wall is modelled either (1) as a truss system
incorrect; the direction of the fastener displacements or (2) as a composite cantilever with partial interac-
depend on the configuration of the wall. It is possible tion. In the truss model, where the framing joints
to show that this assumption is fulfilled only if the along the perimeter of each wall segment are
number of fastener spacings along the rails and the modelled as hinges, the top and bottom rails act as
304 Materials and Structures (2009) 42:301–320

upper and lower chords, the leading and trailing studs 1.3 Aim and scope
of each wall segment as verticals, and the sheathing
as a diagonal tension field (with a certain width) in In this paper we focus on analytical models, suitable
each wall segment. In the cantilever model, where the for hand calculations that have been developed in
bottom of the wall is built-in to the floor or Sweden. These methods are based on linear elastic
foundation, the composite horizontal cross-section properties of the mechanical sheathing-to-framing
of the wall (with interlayer slip between framing and connections of the shear wall. This paper is con-
sheathing) is subjected to a maximum bending cerned with statically loaded fully anchored shear
moment at the built-in end. walls without openings. Expressions for both the
The works of Cziesielski [25] and Henrici [26] are horizontal load-bearing capacity and the horizontal
examples of works, where the composite cantilever displacement of sheathed timber frame shear walls
model has been used. This model was also incorpo- are derived.
rated in the German code DIN 1052 (Teil 3, Ausgabe The purpose of presenting the basic elastic method
02/79), cf. [27, 28]. Henrici [29, 30] supported the for fully anchored shear walls in this paper is to
model with experimental results. He also worked out contribute to a better understanding of the structural
a model for calculating the support reactions, the behaviour of such walls and derive the expressions
uplifting force to design the tie-downs and the for their load-carrying capacity and stiffness.
compression force on the bottom rail [31, 32]. In
the latter model he assumed a linear relationship for
the compressive strains in the bottom rail according 2 Elastic model
to the Bernoulli hypothesis, elastic spring forces in
the studs and, except from that, the shear wall was A typical part of a fully anchored shear wall is shown
assumed to behave as a rigid body. However, it was in Figs. 1 and 2. This fundamental unit, consisting of
later noted that for the cantilever model there was no a sheet fastened to a timber frame, will be called a
satisfying agreement found with the real behaviour of wall unit or wall segment. The width of the segment
shear walls [33]. Therefore, the truss model was is, b, and the height is, h. The spacing of the fasteners
introduced in a later version of the code, DIN 1052 (si) along the top and bottom rail is denoted sr, along
(Teil 3, Ausgabe 04/88) [34], see also [33]. That the perimeter studs, sps, and along the intermediate
model was discussed and applied by e.g. Steinmetz stud, sis. An elastic model for the analysis of such
[35]. walls has earlier been developed by Källsner and
others [7, 19, 36]. Here this elastic model is
1.2 Influencing factors for the behaviour of shear generalized and extended. The model is based on
walls the following assumptions:
(1) Framing members and sheets are rigid. How-
For sheathed timber frame shear walls with mechan-
ever, the influence of assuming flexible framing
ical sheathing-to-framing connections, the following
members and shear deformations in the sheets is
influences are of most importance for their horizontal
also evaluated.
load-bearing capacity and structural behaviour:
(2) No contact between adjacent sheets or between
(a) Geometrical configuration—influence of aspect sheets and surrounding structure (sheets are free
ratio and openings. to rotate).
(b) Loading configuration—influence of magnitude (3) Framing joints act as hinges.
and distribution of vertical loads. (4) Sheathing-to-framing joints have linear elastic
(c) Boundary conditions—influence of anchoring of load-slip characteristics up to failure. Slip
studs and bottom rail, anchoring to transverse modulus is constant and the same in all joints.
walls and to floor and roof structures. Joint stiffness is independent of the force
(d) Fastener characteristics—influence of sheath- direction and of the mutual orientation of the
ing-to-framing connections and framing joints. sheets and framing members.
Materials and Structures (2009) 42:301–320 305

Fig. 2 Static model of a uframe =γ h

fully anchored shear wall ucorner
unit or segment in unloaded
and loaded states. The left vcorner
H γ s/2
part shows an undeformed δcorner
hinged timber frame and the ϕ
right a deformed hinged
timber frame and a rotated
sheet in an undeformed and γ
deformed state, ŷ
respectively. The bottom
rail is assumed not to be
continuously supported, but CG h
only simply supported at the x̂
ends. At these ends, tension
(Rt) and compression (Rc)
reaction forces develop. The
global coordinate system
(x,y) has its origin in the Deformed
lower left corner, while the y,v sheet
local coordinate system Undeformed
ð^
x; y^Þin the centre of gravity γ s/2 ϕ sheet
v0
of the fasteners (CG), x,u
usually in the middle of the
sheet Rt u0 Rc
b

(5) Displacements of the wall are small compared

usheet ¼ u0 þ uy ð1aÞ
to the width and height of the sheets.
(6) Edge distances of sheathing-to-framing joints vsheet ¼ v0  ux ð1bÞ
are small compared to the width and height of
For an arbitrary point (x,y) on the timber frame the
the sheets, i.e. the fasteners are approximately
displacement in the x and y directions is
located along the edges of the sheets.
These assumptions are commented on in Sect. 3. uframe ¼ cy ð2aÞ
vframe ¼ 0 ð2bÞ
2.1 Horizontal load-bearing capacity—discretely
located fastener shear forces For a fastener with coordinates, (x, y) = (xi, yi), the
displacement of the sheet in relation to the timber
In order to determine the forces acting upon the frame is
individual fasteners, the displacement of the sheet ui ¼ usheet  uframe ¼ u0 þ ðu  cÞyi ð3aÞ
relative to the timber frame must be formulated. In
the right-hand part of Fig. 2, the wall unit is shown in vi ¼ vsheet  vframe ¼ v0  uxi ð3bÞ
the loaded state. The vertical frame members (studs) The corresponding components of the fastener force
have been given the rotation c and the sheet the are
rotation u in relation to their original positions. The
origin of the coordinates x and y is placed in the lower Fx;i ¼ kui ¼ k½u0 þ ðu  cÞyi  ð4aÞ
left-hand corner of the timber frame. The lower left-
Fy;i ¼ kvi ¼ kðv0  uxi Þ ð4bÞ
hand corner of the sheet is given the displacements u0
and v0 in the positive x- and y-directions, respec- where k [N/m] denotes the slip modulus for the
tively. For an arbitrary point (x,y) on the sheet the fastener. The minus signs indicate that Fx,i and Fy,i
displacement is given by are force components acting on the sheet.
306 Materials and Structures (2009) 42:301–320

Four unknown quantities have been introduced, u, c,


Hh 1 1
u0 and v0. To find these unknown quantities, the method c¼ P n P
þ n 2 ð10Þ
of minimum of the potential energy (U) with respect to k ^2i
i¼1 x ^i
i¼1 y
these quantities will be used (i.e. oU=ou ¼ oU=oc ¼ Insertion of Eqs. 9 and 10 into Eq. 4 gives the
oU=ou0 ¼ oU=ov0 ¼ 0). The strain energy due to con- components of the fastener forces as
nector deformations of all the n fasteners is given by
X n
1 y^i
U1 ¼ kðu2i þ v2i Þ ð5Þ Fx;i ¼ Hh Pn ð11aÞ
2 ^2i
i¼1 y
i¼1
x^i
The potential energy due to the horizontal load is Fy;i ¼ Hh Pn ð11bÞ
^2i
i¼1 x
given by
U2 ¼ Hch ð6Þ Maximum fastener force, Fmax, occurs in that fastener
which is most remote from the centre of gravity of the
Using Eq. 3, the energy function of the problem can fasteners (i.e. in the corners), i.e. Fcorner = Fmax.
then be written as Denoting the coordinates of this fastener by
U ¼ U1 þ U2 ð^
xi ¼ b=2 ¼ x^corner ; y^i ¼ h=2 ¼ y^corner Þ, it follows
n h i
1 X that
¼ k ½u0 þ ðu  cÞyi 2 þ ðv0  uxi Þ2  Hch
2 i¼1 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Fcorner ¼ 2
Fx;corner 2
þ Fy;corner
ð7Þ
After partial derivation, the following relations are sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi


ffi
obtained x^corner 2 y^corner 2
" # ¼ Hh Pn 2 þ Pn 2 ð12Þ
Xn Xn Xn X n i¼1 x ^i i¼1 y ^i
2 2
k u0 yi þ ðu  cÞ yi  v0 xi þ u xi
i¼1 i¼1 i¼1 i¼1
Denoting the shear capacity of the fastener, Fv, the
horizontal load-bearing capacity of the wall unit can
¼0 ð8aÞ
" # be written as
X
n X
n
k u0 yi  ðu  cÞ y2i  Hh ¼ 0 ð8bÞ Fv
H¼ vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2ffi : ð13Þ
i¼1 i¼1 u
2

" # u
X t P x^corner P y^corner
n
h n þ n
k nu0 þ ðu  cÞ yi ¼ 0 ð8cÞ x^2 y^2
i¼1 i i¼1 i
i¼1
" #
X
n As an example, for the shear wall in Fig. 1 the
k nv0  u xi ¼ 0 ð8dÞ calculated distribution and direction of the fastener
i¼1
forces according to the elastic theory is shown in
If the origin of the coordinates is moved from the lower Fig. 3.
left-hand corner of the timber frame to the centre of The expressions for the moments of inertia of the
gravity of the fasteners, then R xi = 0 and R yi = 0. fasteners as given above are general in nature. For
Eqs. 8c and 8d then imply that also u0 = 0 and v0 = 0. ordinary designs of shear walls according to Fig. 1
The new coordinate axes referring to the centre of with centre distances sps, sis and sr between fasteners
gravity of the fasteners are denoted by x^ and y^. along the perimeter studs, intermediate stud and top
The two angles of rotation, u, and, c, are then and bottom rails, respectively, the parameters of the
obtained from Eqs. 8a and 8b as inertia moment are given by the following integer
Hh 1 series representing the different contributions from
u¼ Pn 2 ð9Þ the fasteners along the framing members,
k ^i
i¼1 x
Materials and Structures (2009) 42:301–320 307

Fig. 3 Force distribution

on the sheet according to a
linear elastic model. The
forces in the corners
perpendicular to the
perimeter are given by
Fx;corner  0:545ðsr =bÞH
and Fy;corner 
0:857ðsr =bÞH according to
Eqs. 11 and 15

Fx,corner
Fy,corner

X
n   2 X b=sr
  b2  b   
h b sr h 2 b2 1 sr h b 2
x^2i ¼ 2 1 þ2 2 i1 ¼ þ3 þ  1þ3 b
i¼1
sps 2 i¼0
b 2 sr sps b=sr 6 6 sps b sr
X n   2 h=s X

ps 1   h2 h=s X

is 1   h2
2 b h sps sis ð14a; bÞ
y^i ¼ 2 þ1 þ2 2 i1 þ 2 i1
i¼1
sr 2 i¼1
h 2 i¼1
h 2
  2  
b h h 4 2 h 1 sr h sr h b 2
¼ 6 þ2 þ 3þ þ  6þ2 þ h
sr sps sis h=sps h=sis 12 12 sps b sis b sr

The approximate expressions in Eq. 14 correspond to rffiffiffiffiffiffiffiffiffiffi

the integral of an evenly distributed shear flow of the 1 5929 b b b
H Fv ¼ 0:984 Fv  Fv ð16Þ
fastener forces, cf. Eq. 23. For the special case common 6 170 sr sr sr
in practice where h = 2b and sr = sps = sis/2, and the Based on the Eqs. 15c,d, the maximum fastener force
number of fasteners are large, we arrive at in the corners is given by
Pn   rffiffiffiffiffiffiffiffiffiffi
x^2i  16 1 þ 3hb sb b2 170 sr sr
i¼1 r Fcorner  6 H ¼ 1:02 H: ð17Þ
Pn   5929 b b
y^2i  1 6 þ 5 h b h2
12 2 b sr
ð15a; bÞ
To illustrate the influence of different patterns
i¼1
1 and spacing of the fasteners, the non-dimensional
for sr ¼ sps ¼ sis horizontal load-carrying capacity, H, according to
2
P
n Eq. 13, is for the different designs presented in
x^2i  76 sb b2 Fig. 4. The diagrams are based on the assumption
i¼1 r
that h/b = 2 and sr = sps = sis/2. In order to be able
Pn
y^2i  11 b 2
12 sr h
ð15c; dÞ to compare the capacity of the different wall
i¼1 designs, the scale of the abscissa is chosen in such
1 h a way as to correspond to the same fastener spacing
for sr ¼ sps ¼ sis and ¼ 2
2 b (so the spacing parameter b/sps is used in this figure
Then for this case, the horizontal load-carrying and in Figs. 6–8 instead of the ordinary spacing
capacity is given by parameter h/sps).
308 Materials and Structures (2009) 42:301–320

1,5 In a corresponding way, the non-dimensional

maximum fastener force in the corners, Fcorner,
according to Eq. 12, can for the different designs be
Horizontal load-carrying capacity of shear wall H

illustrated as shown in Fig. 7; only the multiplier

1
needs to be changed to, H(sr/b) or H(sps/b).

1
2.2 Horizontal load-bearing capacity—uniformly
2
distributed fastener shear flow
4
Instead of considering the fasteners as discretely
located points as described above, it is often expe-
0,5
dient for analysis purposes to smear the fasteners
continuously along the framing members, i.e. to
3 model the shear forces of the fasteners as a shear
force per unit length, a shear flow, f [N/m]. Then, the
shear capacity per unit length is given by fv = Fv/sr
[N/m], i.e. the shear capacity of the fastener divided
0 by the spacing, sr, between the fasteners along the top
0 10 20 30 40 50
or bottom rail (the reference spacing). For uniformly
Fastener spacing parameter b /s r (no. 1-3) or b /s ps (no. 4)
distributed shear capacity of the fasteners, the
Fig. 4 Non-dimensional horizontal load-carrying capacity of summations are transformed into line integrals. The
shear wall (H) versus non-dimensional parameter for centre components of the shear flow per unit length (f) can
distances of the fasteners (b=sr or b=sps ) according to Eq. 13. then be expressed as
Aspect ratio, h=b ¼ 2, and centre distances, sr ¼ sps ¼ 12sis . The
values to the right in the diagram are the asymptotic values for k k
b=sr ! 1 or b=sps ! 1 fx;l ð^
x; y^Þ ¼  uð^
x; y^Þ ¼ Kl uð^
x; y^Þ ¼  ðu  cÞ^
y
sl sl
ð18aÞ
Comparing the wall designs 1 and 2 in Fig. 4, it is
noted that the difference in the capacity is marginal k k
fy;l ð^
x; y^Þ ¼  vð^
x; y^Þ ¼ Kl vð^
x; y^Þ ¼ u x^ ð18bÞ
(3%) with respect to using fasteners along the sl sl
intermediate stud or not. However, this does not where k/sl = Kl [N/m2] denotes the slip modulus for
mean that the fasteners in the intermediate stud the shear flow and sl the spacing of the fasteners
should be omitted because that would lead to a great along the rail (l = r), perimeter stud (l = ps), and
risk for buckling of the sheet. Instead, this fact should intermediate stud (l = is), respectively. The minus
be interpreted only to mean that the spacing of the signs indicate that fx and fy are the components of the
fasteners along the intermediate stud can be larger shear flow acting on the sheet. The two angles of
than that of the perimeter framing members (e.g. as in rotation, u, and, c, are then given by
this case, sis = 2sr = 2sps).
Hh 1
The wall designs 3 and 4 are shown to illustrate the u¼ H 2 ð19aÞ
k x^
l sl dl
effect on the horizontal load-carrying capacity of
fastening only the top and bottom rails, and the leading 2 3
and trailing studs, respectively. These designs (along Hh 4 1 1 5
with omitting fastening of the intermediate stud) are c¼ H 2 þ H y^2 ð19bÞ
k x^
not designs that should be used in practice because of l sl dl l sl dl
the buckling risk; they are only shown in order to the components of the shear flow by
elucidate the difference in capacity and behaviour. The
capacity is only about 15% in case of wall design 3 and Hh y^
fx;l ¼ H y^2 ð20aÞ
about 50% for design 4 as compared to wall design 1. sl
l sl dl
Materials and Structures (2009) 42:301–320 309

member is assumed, no forces perpendicular to that

Hh x^
fy;l ¼ H 2 ð20bÞ member develop. The moment of inertia for the shear
sl x^
dl wall changes in such a way that the terms associated
l sl
with the summation signs in Eq. 14a, b or the first
the maximum shear flow of the fastener, if the
integral in each of Eq. 23a and b are excluded. This
spacing of the fasteners along the top or bottom rail is
gives {[(h/sps) - 1]b2/2 & (h/sps)b2/2} in Eq. 14a,
chosen as a reference spacing (l = r),
{[(b/sr) ? 1]h2/2 & (b/sr)h2/2} in Eq. 14b, {(h/
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
fcorner ¼ fx;corner 2
þ fy;corner sps)b2/2} in Eq. 23a and {(b/sr)h2/2 in Eq. 23b}. For
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sr = sps and h/b = 2, the horizontal load-bearing
u" #2 " #2
u capacity is then given by
u
Hh t corner ^
x ^
y corner
¼ H 2 þ H y^2 ð21Þ 1 b b
sr x^
dl s dl H  pffiffiffi Fv ¼ 0:707 Fv ¼ 0:707fv b ð16bÞ
s l l l l
2 sr sr
and the horizontal load-bearing capacity of the wall Note that the forces in the corner fasteners are the
unit by sum of the vector components in the two directions,
fv i.e.
pffiffiffi the corner fasteners are subjected to a load that is
H¼ vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2ffi ð22Þ
u
2
2 greater than the load on the other fasteners.
u
t Hx^corner Hy^corner To illustrate further: In some models it is appro-
h
sr x^2
þ y^2
dl dl priate to neglect the bending stiffness of the top rail,
l sl l sl
cf. Sect. 2.6, i.e. the multiplier ‘2’ in front of the
where x^corner ; y^corner denote the location of the most summation sign in Eq. 14a changes to ‘1’. Then, Eq.
remote shear flow (± b/2, ± h/2), fv = Fv/sr [N/m] 14a becomes (b/sr ? 6h/sps ? 2sr/b - 3)b2/12 &
the ultimate shear flow capacity of the fasteners, and [1? 6(sr/sps)(h/b)](b/sr)(b2/12) and for sr = sps and
the moment of inertia components are given by, h/b = 2 we have (13/12)(b/sr)b2 (corresponding to
respectively, Eq. 15c). The racking capacity is then given by
I 2 Z b Z þh2
x^ 2 þ2 2 b2 143 b b
dl ¼ x^ d^xþ d^y H  pffiffiffiffiffiffiffiffiffiffi Fv ¼ 0:933 Fv ¼ 0:933fv b ð16cÞ
l sl sr b2

2s
 ps 2
h
2 5877 sr sr
1 sr h b 2
¼ 1þ3 b ð23aÞ Note, however, that this result is approximate; the
6 sps b sr
  Z þh change in location of the centre of gravity of the
I 2 Z b
y^ 2 1 2
h2 þ 2 fasteners is neglected in this calculation.
dl ¼ þ y^2 d^
yþ d^
x
l sl sps sis h2

2s
 r 2
b

1 sr h sr h b 2 2.4 Effect of shear deformation of the sheet on

¼ 6þ2 þ h ð23bÞ horizontal load-bearing capacity
12 sps b sis b sr
where the line integration is performed along the The analysis in Sects. 2.1 and 2.2 was based on the
framing members, i.e. l = 2r ? 2 ps ? is. It is noted assumption of rigid sheets. The influence of shear
that these equations are the same as the approximate deformations in the sheathing material can be esti-
expressions in Eq. 14. Then, Eqs. 15–17 still hold, mated by introducing another unknown quantity, the
and the horizontal load-carrying capacity is given by, shear angle cs. The shear angle is assumed to be
H  0:984fv b, and the maximum fastener force in the constant over the entire sheet. Consider Fig. 2: If the
corners by, fcorner  1:02H=b. shear deformations in the sheet are taken into account,
the angle u is changed to u ? cs/2 in Eq. 1a and u is
changed to u - cs/2 in Eq. 1b (and correspondingly in
2.3 Effect of flexible framing members on
Eqs. 3 and 4), and, thereby, the angle u is changed to
horizontal load-bearing capacity
u - cs/2 in Eq. 9 and c is changed to c - cs in Eq. 10.
But the fastener forces according to Eq. 4, adjusted
The analysis above was based on the assumption of
with respect to the changes just made in Eqs. 1, 9, and
rigid framing members. If full flexibility of a framing
10, will finally remain the same as given by Eq. 11.
310 Materials and Structures (2009) 42:301–320

Thus, the fastener forces and, thereby, the horizontal

h
load-bearing capacity are not affected by the shear R t ¼ Rc ¼ H ð27Þ
deformations in the sheet. The only change is in the b
These forces act between the ends of the shear wall
horizontal displacement of the shear wall as shown in
and the floor or foundation. They are composed of
Sect. 2.11.
vertical forces transmitted via the sheathing-to-fram-
ing joints along the stud and along the top and bottom
2.5 Effect of vertical loads on horizontal
rails, respectively, through that stud and top and
load-bearing capacity
bottom rails down to the floor or foundation.
In order to evaluate the proportions between the
The influence of unintentionally tilted vertical studs on
forces transmitted through the stud and rails, respec-
the horizontal load-carrying capacity can be accounted
tively, consider a wall unit with three vertical frame
for by adjusting the capacity with respect to a
members according to Fig. 5. The fasteners are
corresponding fictitious horizontal load given by [37]
assumed uniformly distributed along the frame
  X m
a1 members. The spacing between the fasteners along
Htilt ¼ a0 þ pffiffiffiffi Vi ð24Þ
m i¼1 the perimeter of the wall unit is constant (sps = sr).
Also, the spacing of fasteners along the intermediate
where Vi (i = 1 to m) is the vertical load on the ith stud is constant, but twice as much as that along the
stud, m the number of studs, a0 (= 0.003) the other frame members (sis = 2 sps). According to the
systematic and a1 (= 0.012) the random angle of elastic theory, the stress distribution of the sheet and
tilting of studs, respectively. the framing members is as shown in Fig. 5. Accord-
Second order effects can in a corresponding way ing to Eqs. 20 and 23, we arrive at
be written as
12 H 6H H h
fH ¼ ; ! fH ¼ ¼ 0:545 for ¼2
uframe X
m hb
12 þ 5 b 11 b b b
H2nd ¼ Vi ð25Þ
h i¼1 ð28aÞ
where uframe is the horizontal displacement of the top 3 hb H 6H H h
fV ¼ ; ! fV ¼ ¼ 0:857 for ¼2
of the shear wall, see Sect. 2.7. The effective 1 þ 3 hb b 7b b b
horizontal load-bearing capacity then becomes
ð28bÞ
Heff ¼ H  Htilt  H2nd ð26aÞ
where fH and fV are horizontal and vertical stress per
or in view of Eqs. 24 and 25, unit length in the sheet and framing members,
" pffiffiffiffi m # respectively. The maximum force on the fasteners
uframe X
m
a0 þ a1 = m X qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Heff ¼ 1  Vi  ¼ 2 2
1 þF
Vi H can then be written
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi as F corner Fx¼ 2b y¼12h ¼
Hh i¼1 H i¼1
ðfH sr Þ2 þ ðfV sps Þ2 , which equals Fcorner 
ð26bÞ
1:02 ðsr =bÞH for sr = sps and h/b = 2.
where uframe is given by Eq. 34 or 43, and H by Moment and force equilibrium for the framing
Eq. 13. Usually, the effect of tilted studs and, members give the following reaction forces in the
especially, the second order effects can be neglected. studs in the horizontal direction,
1
2.6 Vertical reactions at ends of wall due RH1 ¼ RH2 ¼ fH h ð29a; bÞ
6
to overturning moment
and in the vertical direction,
If the shear wall is considered as a rigid body on 1
RV1 ¼ fV b
simple supports, the reaction forces related to tension 6 ð30a; bÞ
of the leading stud and compression of the trailing RV2 ¼ RV1 þ fV h
stud, arising due to the horizontal force, are given by The vertical force transmitted through the stud is
(cf. Figs. 2 and 5) given by
Materials and Structures (2009) 42:301–320 311

fV
fH
H
RH1 1R RH1
2 H1
RV1 RV1
RV1 fV RV1
fH 1 f
fH fH fH fH 2 H
1 f
2 H RH1 1R
RH1 2 H1

fV fV h fV fV

RH2 RH2
b
1R
fH fV 2 H2
RV2 RV2 Intermediate stud
RV2
RV2
fH fV

RH2 1R RH2
2 H2 H
Rt Rc

Fig. 5 Stress distribution on the sheet and the framing members in a fully anchored shear wall according to the linear elastic theory
(fV and fH denote shear stress per unit length of the fasteners in the vertical and horizontal direction, respectively)

the bottom rail. Note, however, that the 7% of the

1 1 þ 6 hb h moment taken by the top rail is included in the 93%.
Rstud ¼ RV2 ¼ H ¼ Rt  RV1 ð31Þ
2 1 þ 3 hb b According to Eq. 28b, 86% refers to the vertical
For a wall unit with a height-to-width ratio of 2 (h/ stress, and the rest with 7% to the top rail and 7% to
b = 2), Eq. 31 can be simplified to the bottom rail. [If there is no intermediate stud or if
it is neglected, only the horizontal stress per unit
13 h h length is changed, fH = [3/(3 ? h/b)]H/b (from Eq.
Rstud ¼ H ¼ 0:929 H ð32Þ
14 b b 23b with sis ? ?). The vertical stress fV remains the
This means that 93% of the overturning moment is same and, thus, Rstud also remains the same as given
taken as a force couple in the vertical frame by Eqs. 31 and 32.] According to Eq. 30, the relation
members, while the remaining 7% of the moment is between the vertical forces transmitted through the
taken as fastener forces between the sheet itself and top and bottom rails, respectively, are given by
312 Materials and Structures (2009) 42:301–320

RV1/(RV2 ? RV1) = 1/[2(1 ? 3 h/b)]. For h/b = 2, Eq. 27 (or Eq. 32 or 33b), but also to limit the
this means that less than 7% is transmitted through displacements in the joint. One way of handling the
the top rail and 93% through the bottom rail. problem is to give some limiting value for the displace-
If only the bottom rail is supposed to be rigid ments in this joint, which should not be exceeded before
enough to transmit the vertical forces and, thus, the the maximum racking load is reached. An alternative
bending stiffness ofHthe top rail is neglected, Eq. 14a way is to design the joint for a somewhat higher force
or 23a becomes l ð^ x2 =sl Þdl ¼ ½1 þ 6ðsr =sps Þðh=bÞ than that given by Eq. 27 (or Eq. 32 or 33b).
2
ðb=sr Þb =12, cf. Sect. 2.1. Eq. 28b is then changed In order to limit the horizontal displacements of the
to fV = 6(h/b)/(1 ? 6 h/b)]H/b. The force in the stud wall in the serviceability limit state, a proper anchor-
then becomes (RV1 = 0), ing of the tensile stud (and hence prevent uplift of the
wall) is also very important in this respect. For further
6 hb h 12 h
Rstud ¼ fV h ¼ H ; ! Rstud ¼ H details, see Girhammar and Källsner [2]
1 þ 6 hbÞ b 13 b High external vertical loads and high additional
h h forces arising from the diaphragm action need to be
¼ 0:923 H for ¼ 2 ð33a,bÞ
b b accounted for. Thus, column buckling of the vertical
studs can be decisive for the load-bearing capacity of
This means that 92% of the overturning moment is
the shear wall. Shear stresses acting along the edges of
taken as a force couple in the vertical frame
the sheets introduce compression and bending stresses
members, while the remaining 8% of the moment is
in the studs. These additional forces transferred via the
taken as fastener forces between the sheet itself and
sheathing-to-framing joints, are assumed to vary
the bottom rail. However, note that these results are
linearly along the stud, cf. Fig. 5. Usually the
approximate. For an exact calculation, the change of
sheathing-to-framing joints will brace the stud in the
location of the centre of gravity for the fasteners
weak direction. For out of plane buckling of the wall,
needs to be taken into account.
composite action between the sheet and the studs can
According to Eqs. 32 and 33b, most of the vertical
be taken into account in the elastic stage. The basis for
force is transmitted through the studs and, therefore,
this design problem is summarized in Appendix A.
the contact pressure at the trailing stud can safely be
evaluated for a reaction force resulting from the 2.7 Buckling of wall panels
whole overturning moment according to Eq. 27.
For high strength sheets and dense nailing, the local If the sheets are very thin, there is a risk for local
contact pressure on the bottom rail can be decisive for buckling of the sheets. Based on the elastic model, the
the load-bearing capacity of the shear walls. However, distribution of the forces acting upon a sheet was shown
the local compression strength perpendicular to grain in Fig. 3. To determine the critical load for this force
can be enhanced due to membrane effects in the distribution is not easy. However, some guidelines for
length direction of the deformed wood fibres, see design against buckling of both isotropic and orthotropic
e.g. Eurocode 5 [38]. In case of very high loads, two sheets can be found in Eurocode No. 5, Appendix 7 [39].
or more studs can be used to compose the trailing For the sake of completeness, it is noted that by
stud. assuming plastic conditions for the sheathing-to-fram-
When the elastic solution in Sect. 2.1 was devel- ing joints and a plastic shear flow around the perimeter
oped, one of the basic assumptions was that the of the wall, the buckling load is much easier to evaluate.
members of the timber frame were hinged to each
other. The most critical joint in this respect is the 2.8 Shear walls comprising of more than one unit
connection or anchorage between the vertical stud and and/or of sheets on both sides
the bottom rail or floor/foundation on the tension side
of the wall unit. If large displacements are accepted in The load-bearing capacity of a shear wall comprising
this joint, a different stress distribution occurs where of several wall units (wall segments) is given as the
high stresses are introduced between the sheet and the sum of the load-bearing capacity of the individual
bottom rail. To avoid this problem it is necessary not parts. For the analysis of walls built up of several
only to take care of the vertical force according to different parts, it is convenient to assume plastic
Materials and Structures (2009) 42:301–320 313

conditions for the sheathing-to-framing joints. This rffiffiffiffiffiffiffiffiffiffiffi

model is not discussed in this paper. 170 sr H sr H H
dcorner 6 ¼ 1:02 ¼ 1:02
For shear walls with sheathing of the same type 5929 b k b k Kr b
and thickness on both sides, the load-bearing capacity ð37Þ
may be taken as the sum of the calculated contribu- The relationship between the horizontal displace-
tions. If the sheets or the fasteners are of different ment of the wall and the maximum fastener
types, experiments on wall panels consisting of displacement in the corners can then be written as
normally used combinations of interior and exterior " #
sheathing materials in Sweden show that the maxi-
Pn1 þ Pn1 h
mum load is obtained at about the same displacement x^2i y^2
i¼1 i¼1 i
[7]. This means that the different capacities can be uframe ¼ vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
#2 dcorner ð38Þ
u" #2 "
added. However, a smaller reduction of the capacity u x^ y^
t Pcorner n þ Pcorner n
of the weaker side would be called for before adding 2
x^ y^ 2
i¼1 i i¼1 i
it to the total load-carrying capacity. If fastener sheet
combinations with completely different slip proper- For shear walls of common design where Eq. 15
ties were used an even greater reduction should be apply, this relationship becomes
used. For further details, see Sect. 3. 58
uframe  pffiffiffiffiffiffiffiffi dcorner ¼ 4:45 dcorner ð39Þ
170
2.9 Horizontal displacement of the shear wall
and fastener displacement Some further details concerning the horizontal and
vertical displacements of the fasteners and the angle
The horizontal displacement of the top rail of the of rotation of the frame and sheet are given in
shear wall is, according to Eq. 10, given by Appendix B.

 To illustrate the influence of different patterns and
Hh2 1 1 spacing of the fasteners, the non-dimensional hori-
uframe ¼ ch ¼ Pn 2 þ Pn 2 ð34Þ
k ^i
i¼1 x ^i
i¼1 y zontal displacement of the shear wall, uframe,
This equation can also be obtained by studying the expressed in terms of the horizontal load (H)
geometry of the wall unit according to Fig. 2, i.e. according to Eq. 34, and in terms of the maximum
uframe ¼ 2ucorner þ 2ðh=bÞvcorner , where fastener dis- fastener displacement in the corners according to Eq.
placements in the corners (ucorner, vcorner) are given by 38, is for the different designs presented in Figs. 6
Eq. 51 in Appendix B. For shear walls of common and 8, respectively. The diagrams are based on the
design (h = 2b; sr = sps = sis/2) where Eq. 15 apply, assumption that h/b = 2 and sr = sps = sis/2. The
the horizontal displacement is given by displacement is expressed as a function of the
fastener spacing parameter, b=sr or b=sps , which
348 sr H sr H H corresponds to the number of spaces between the
uframe  ¼ 4:52 ¼ 4:52 ð35Þ
77 b k bk Kr b fasteners along the bottom rail and the vertical stud,
The maximum displacement of the fasteners takes respectively.
place in the corners in the same direction as that of In a corresponding way, the non-dimensional
the corresponding maximum fastener force according maximum displacement of the fasteners in the
to Eq. 12. We then have corners, dcorner, according to Eq. 36, is for the
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi different designs illustrated in Fig. 7.
Fcorner
dcorner ¼ u2corner þ v2corner ¼ It is evident from Figs. 6 and 7 that the
k
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi


ffi displacements are largest for those wall designs
Hh x^corner 2 y^corner 2
¼ Pn 2 þ Pn 2 ð36Þ that had the lowest load-carrying capacity accord-
k i¼1 x ^i i¼1 y ^i ing to Fig. 4. If the horizontal displacement of the
For shear walls of common design (h = 2b and wall, uframe, is expressed as a function of the
sr = sps = sis/2) where Eq. 15 apply, the displace- fastener displacement, dcorner, as in Fig. 8, this
ment of the fasteners in the corners is given by relationship is relatively independent of the actual
wall design.
314 Materials and Structures (2009) 42:301–320

30 6

3
Horizontal displacement of shear wall, u frame

Displacement of fasteners at corners, u corner

3

20 4

4
10 2
2
2

1 1

0 0
0 10 20 30 40 50
0 10 20 30 40 50
Fastener spacing parameter, b /s r (no. 1-3) or b /s ps (no. 4) Fastener spacing parameter, b /s r (no. 1-3) or b /s ps (no. 4)

Fig. 6 Non-dimensional horizontal displacement of shear wall Fig. 7 Non-dimensional displacement of fasteners at the
(uframe) related to the horizontal load (H) versus non-dimen- corners (dcorner) versus non-dimensional parameter for centre
sional parameter for centre distances of the fasteners (b=sr or distances of the fasteners (b=sr or b=sps ) according to Eq. 36.
b=sps ) according to Eq. 34. Aspect ratio, h=b ¼ 2, and centre Aspect ratio, h=b ¼ 2, and centre distances, sr ¼ sps ¼ 12sis . The
distances, sr ¼ sps ¼ 12sis . The values to the right in the diagram values to the right in the diagram are the asymptotic values for
are the asymptotic values for b=sr ! 1 or b=sps ! 1 b=sr ! 1 or b=sps ! 1. This diagram can also be used for the
corresponding maximum fastener force at the corners (Fcorner) if
the multiplier is changed to, H sbr or  H sbps according to Eq. 12
2.10 Effect of flexible framing members
on horizontal displacement pffiffiffiffiffiffiffiffiffiffi
2 5877 sr H sr H
dcorner  ¼ 1:07 ; ð37cÞ
The analysis above was based on the assumption of 143 b k b k
rigid framing members. In case of fully flexible
Note that this result is approximate. The change in
framing members, the parameters of the moment of
P 2 location of the centre of gravity of the fasteners is
inertia were changed to x^i  ðh=sps Þb2 =2 and
P 2 2 neglected in this calculation. In case of a fully
y^i  ðb=sr Þh =2 as shown in Sect. 2.3. For sr = sps
flexible framing member, the direction of the fastener
and h/b = 2, the horizontal displacement is then
displacement is along the framing member, except in
given by
the corners where the direction is governed by the
sr H sum of the two vector components.
uframe  6 ¼ 4:24 dcorner ð35bÞ
b k
If only the bending stiffness of the top rail is
neglected, Eq. 34 for sr = sps and h/b = 2 becomes, 2.11 Effect of shear deformation of the sheet on
horizontal displacement
684 sr H sr H
uframe  ¼ 4:78 ¼ 4:46 dcorner ð35cÞ
143 b k b k The analysis as presented in Sect. 2.9 was based on
The fastener displacement in Eq. 35b is given by the assumption of rigid sheets. The influence of shear
pffiffiffi sr H deformations in the sheathing material can be
sr H
dcorner  2 ¼ 1:41 ; ð37bÞ estimated by introducing another unknown quantity,
b k b k
the shear angle cs. The shear angle is assumed to be
and in Eq. 35c by constant over the entire sheet. The increase in
Materials and Structures (2009) 42:301–320 315

5 For normally used sheathing materials and fastener

distances, the influence of the shear deformation cs on
the displacements is often neglected. However, as is
shown in the illustrative example in Sect. 2.13, the
Horizontal displacement of shear wall, u frame

1 effect can be significant even for ordinary cases.

4,5
2.12 Effect of vertical loads on horizontal
2
displacement
3

The influence of unintentionally tilted vertical studs

and second order effects on the horizontal displace-
4 ment of the shear wall can be accounted for by adding
4
to the horizontal load on the shear wall the corre-
sponding fictitious horizontal loads as given by Eqs.
24 and 25. Thus, the horizontal load (H) in Eq. 34 or
43 is replaced by the total load,
3,5 Htot ¼ H þ Htilt þ H2nd ð44Þ
0 10 20 30 40 50
Fastener spacing parameter, b /s r (no. 1-3) or b /s ps (no. 4) In view of Eqs. 24 and 25, the effective horizontal
displacement can then be written as
Fig. 8 Non-dimensional horizontal displacement of shear wall pffiffiffi
(uframe) related to the fastener displacement (dcorner) versus non- a þa = m Pm
1 þ 0 H1 Vi
dimensional parameter for centre distances of the fasteners uframe;eff ¼ u frame
Pm i¼1 uframe ð45Þ
(b=sr or b=sps ) according to Eq. 38. Aspect ratio, h=b ¼ 2, and 1  Hh i¼1 Vi
centre distances, sr ¼ sps ¼ 12sis . The values to the right in the
diagram are the asymptotic values for b=sr ! 1 or b=sps ! 1 where H is given by Eq. 13 and uframe by Eq. 34 or
43. Usually, the effect of tilted studs and, especially,
potential energy due to the shear deformations in the
the second order effects can be neglected.
sheet can be written as
1
U3 ¼ Gc2s bht ð40Þ 2.13 Illustrative example
2
where G is the shear modulus of the sheet and t is the Consider a typical shear wall as shown in Fig. 1 and
thickness. The change in potential energy, due to the composed of a 2400 9 1200 9 8 mm Masonite fibre
displacement csh of the external force H, is given by hardboard (C40) (mean shear modulus, G =
U4 ¼ Hcs h ð41Þ 2100 MPa) nailed to the frame by using annular ringed
shank nails 50 9 2.1 (mean ultimate limit load, Fv = 1
Adding the contributions U3 and U4 to the potential
kN, according to Girhammar et al. [40]). In the
energy U in Eq. 7 and determining its minimum value
serviceability limit state (considered as a serviceability
by partial derivation (oU=ocs ) then gives
limit load corresponding to 40% of the ultimate limit
H load), the mean secant slip modulus of the sheathing-to-
cs ¼ ð42Þ
Gbt framing joints is k = 800 kN/m (cf. Girhammar et al.
The displacement of the top rail of the frame is [40]. The aspect ratio is h/b = 2 and the spacing
then given by sr = sps = sis/2 = 100 mm. The wall is subjected to a

 vertical load of 6 kN on each stud (3 pieces).
Hh2 1 1
uframe ¼ ðc þ cs Þh ¼ P n P
þ n 2 þ cs h According to Eq. 13, the horizontal load-carrying
k ^2i
i¼1 x y^ capacity of the shear wall is given by H = 0.980 (b/

  i¼1 i
H 2 1 1 h k sr)Fv = 11.755 kN (exact; cf. the exact expression in
¼ h Pn 2 þ Pn 2 þ ð43Þ
k ^
x
i¼1 i y^
i¼1 i b Gt Eq. 14) and H = 0.984 (b/sr)Fv = 11.808 kN
316 Materials and Structures (2009) 42:301–320

(approximate; cf. Eq. 15). The error of the approx- In Table 1, the experimental and theoretical results
imate value is only 0.5%. According to Eq. 34, the are compared with respect to the load-carrying
horizontal displacement of the shear wall is given by capacity. It is evident that the applicability of the
uframe = 4.535 (sr/b)(0.40H/k) = 2.221 mm (exact) proposed analytical procedure is very good. In
and uframe = 4.519 (sr/b)(0.40H/k) = 2.223 mm addition, all values are on the safe side.
(approximate), i.e. an error of 0.1%. Experimental and theoretical results with respect
The influence of the shear modulus is evaluated by to the horizontal displacement of shear walls are
using Eq. 43, i.e. uframe = [4.535 ? (h/b)(k/Gt)(b/sr)] compared in the companion paper [2].
(sr/b)(0.40H/k) = (4.535 ? 1.143)(sr/b)(0.40H/k) =
2.781 mm. Thus, the influence is about 25%.
The effect of the vertical loads on the horizontal
capacity is given by Eq. 26 and on the horizontal 3 Comments on the assumptions of the elastic
displacement by Eq. 45. The effective horizontal model
capacity becomes, Heff = [1 - (2.781/2400)
(18/11.755) - (0.003 ? 0.012/H3)(18/11.755)]H = A common observation in full-scale testing of shear
(1 - 0.0017744 - 0.015203)H = 0.983 H= walls is that the framing members will bend in a
11.555 kN. Then, the effective capacity is 1.7% sinusoidal form. This is especially true when the
lower than the capacity neglecting the influence of spacing between the fasteners is small or when sheets
vertical loads, where 0.2% refers to second order on both sides of the wood frame are used. The
effects and 1.5% to effects of tilted studs. Correspond- consequence of the assumption of rigid framing
ingly, the effective horizontal displacement becomes, members is that the stiffness of the elastic model is
uframe,eff = [(1 ? 0.015203)/(1 - 0.0017744)] uframe = somewhat too high. The assumption of rigid sheets in
1.017 uframe = 2.828 mm, i.e. 1.7% higher than the the elastic model is a good approximation in the
displacement neglecting the influence of vertical ultimate limit state. However, the contribution of
loads, where 0.2% refers to second order effects shear deformations of the sheets in the serviceability
and 1.5% to effects of tilted studs. limit state should be included.
It was assumed that the framing joints act as
2.14 Experimental verification hinges. Significant shear forces must be transmitted
in the framing joints at those vertical studs where two
To validate the proposed theory, it is considered here sheets meet, i.e.
shear walls according to Fig. 1 tested by Källsner [7]. 1
Six full scale wall tests were conducted, five with Fframe joint ¼ RH2 ¼ fH h !
6 ð46Þ
different sheathing materials, of which one with two h
Fframe joint ¼ 0:182H; ¼2
types of fasteners. Also, the sheathing-to-framing b
joints for each wall design were tested.

Table 1 Comparisons between experimental and analytical results for a number of shear wall designs, for details see Källsner [7]
Hexp
Sheathing Full scale Fastener Analytical Fastener spacing parameters Hanal Hanal
wall test, Hexp test, Fv load, Hanal
[N] [N] Fv sbr b
sr
h
sps
h
sis [N]

Gypsum plasterboard (screws) 4555 642 0.988 6 12 8 3806 1.20a

Gypsum plasterboard (nails) 4182 461 1.003 8 16 16 3699 1.13a
Bitumen-impregnated fibreboard (nails) 5571 353 0.981 16 32 16 5541 1.01
Particleboard (nails) 7390 582 0.975 12 24 10 6809 1.09
Medium density fibreboard (nails) 7120 439 0.974 16 32 12 6841 1.04
Plywood (nails) 7304 887 1.004 8 16 16 7124 1.03
a
The lowest values of the ultimate load for the fasteners were used. If the highest values are used, the ration reduces to 1.03 and 1.10,
respectively
Materials and Structures (2009) 42:301–320 317

Especially for double-sided sheathing and doubled Appendix A: Buckling of sheathed walls
number of sheets on each side, these forces become
very significant. Note also that Eurocode 5 [38], for The sheathed wall is a partially composite component
example, stipulates that only 1/3 of the load-carrying consisting of the stud and the sheet mechanically
capacity of nails installed at right angles to the grain connected to each other. The buckling of sheathed
should be used as design values. walls has been studied by Kamiya [41, 42] and of
All studs subjected to tensile forces are assumed composite beam-columns with interlayer slip in
fully anchored to the floor or foundation. Hinges in general by Girhammar and Gopu [43] and Girham-
the model means that there is no displacement in mar and Pan [44] among others.
either the horizontal or the vertical direction. For The effective width of the sheathing is assumed to
these reasons, the stiffness of the elastic model will be constant along the length and equal to that when
become too high. the wall is bent by uniformly distributed loads, and
The behaviour of the sheathing-to-timber joints is buckling of the sheathing is assumed not to occur.
usually non-linear. It is necessary to use some kind of The buckling load according to the one dimensional
secant slip modulus if reasonable values for the composite action theory can then be written as [44]
displacements of the wall are to be obtained. 2 31
EI1
EI0  1 7
In practice, the assumptions of small displace- p2 EI1 6
ments and centre distances of fasteners are good Pcr ¼ 2 4
1þ 5 ð47Þ
ðlLÞ ðlLÞ2
approximations. 1 þ a2
p2
where L is the length of the column, l buckling
length coefficient (the same as for the ordinary Euler
4 Conclusions cases), and
Kr 2
The basic principles for analyzing fully anchored a2 ¼   ð48Þ
EI0
light-frame wood shear walls assuming elastic behav- EI0 1  EI
1
iour of the mechanical sheathing-to-framing joints 9
EI0 ¼ E1 I1 þ E2 I2 > =
have been presented. The models can be applied to
EAp r 2 > ð49a; bÞ
statically loaded shear walls in both the ultimate and EI1 ¼ EI0 þ ;
serviceability limit states to evaluate the horizontal EA0
load-bearing capacity and the horizontal displace- )
EA0 ¼ E1 A1 þ E2 A2
ments of the wall as well as the slip in the sheathing- ð50a; bÞ
to-timber joints. The elastic method is suitable for EAp ¼ E1 A1
E2 A2
hand calculation. where K [N/m2] is the slip modulus of the sheathing-
to-framing joints, r the distance between the centroids
Acknowledgements The authors express sincere
appreciation for the financial support from The Development of the two components (1 and 2; sheet and stud), EA
Fund of the Swedish Construction Industry (SBUF), The the axial stiffness, and EI the bending stiffness.
Swedish Research Council for Environment, Agricultural Subscripts 0 and ? denote the properties for non-
Sciences and Spatial Planning (FORMAS), The County
composite and fully composite walls, respectively.
Administrative Board of Västerbotten, The European Union’s
Structural Funds—The Regional Fund, SP Wood The buckling load is supposed to be applied at the
Technology—The Technical Research Institute of Sweden, center of gravity of the fully composite section.
Umeå University, and Carl Wikström Foundation, Nordmaling, The sheathed wall subjected to both vertical and
Västerbotten, together with the timber and building industry.
transverse loads can then be designed according to
The authors would also like to thank the reviewers for their
constructive comments and suggestions. the principles given in Girhammar [45].
318 Materials and Structures (2009) 42:301–320

Appendix B: Displacements of fasteners and sheet, pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

and angles of rotation of frame and sheet dcorner ¼ u2corner þ v2corner
qffiffiffiffiffiffiffi ffi
170 sr H sr H H
 6 5929 b k ¼ 1:02 b k ¼ 1:02 Kr b
For the sake of completeness, for later comparisons pffiffiffiffiffiffi
170
with analyses of influence of imperfections [2], the ¼ 58 uframe ¼ 0:225 uframe
horizontal and vertical displacements of the individ- ð54Þ
ual fastener and sheet, and the ratio of the two angles
Alternatively, this equation can also be obtained
of rotation of the frame and sheet, respectively, are
by using Eq. 28, i.e.
evaluated.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
dcorner ¼ ðFx¼12b =kÞ2 þ ðFy¼12h =kÞ2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
B.1 Displacements of fasteners ¼ ðfH sr =kÞ2 þ ðfV sps =kÞ2 :

According to Eqs. 3a,b, 9 and 10, the horizontal and

vertical displacement in the new coordinate system B.2 Displacements of sheet
are given by
The displacements of the centre of gravity (CG) of
Hh y^i the fasteners (here assumed in the middle of the
ui ¼ u0 þ ðu  cÞyi ¼ ðu  cÞ^
yi ¼  Pn 2
k ^i
i¼1 y sheet) can, according to Fig. 2, be written as
ð51aÞ  
1 h
usheet;CG ¼ uframe þ yCG  u þ u0
Hh x^i 2 2 ð55Þ
vi ¼ v0  uxi ¼ u^
xi ¼  Pn 2 ð51bÞ
k ^i
i¼1 x vsheet;CG ¼ 0
According to Eq. 3 together with the approximate In case of symmetric conditions, yCG = h/2, the
expressions for ordinary shear walls given in Eqs. 15 middle term on the right hand side of Eq. 53a
and 36, the displacements in the principal directions vanishes. The displacements of the lower left corner
can be written as of the sheet, u0 and v0, can be obtained by using Eq.
51 for yi = 0 and y^i = -h/2 to give
12 y^i sr H 7 y^i y^i
ui  ¼ uframe ¼ 0:241 uframe
11 h b k 29 h h Hh h=2 6 sr H sr H
u0 ¼ Pn 2  ¼ 0:545 ð56Þ
k ^
y
i¼1 i 11 b k bk
24 x^i sr H 22 x^i x^i
vi  ¼ uframe ¼ 0:759 uframe and for xi = 0 and x^i = -b/2 to give
7 hbk 29 h h
ð52a; bÞ Hh b=2 6s H s H
v0 ¼ Pn 2  r ¼ 0:857 r ð57Þ
k ^
x
i¼1 i 7 b k b k
Maximum fastener displacement in the horizontal
and vertical direction (^xi ¼ b=2; y^i ¼ h=2), respec- Eqs. 56 and 57 are in accordance with Eq. 53 as it
tively, is then given by should, cf. Fig. 2.

6 sr H sr H B.3 Ratio of angles of rotation of frame and sheet

ucorner  ¼ 0:545
11 b k b k
7 According to Eqs. 9 and 10, the ratio of the two
¼ uframe ¼ 0:121 uframe
58 ð53a; bÞ angles of rotation of the frame and the sheet,
6 sr H sr H respectively, is given by
vcorner  ¼ 0:857
7 b k b k Pn 2
11 c x^i
¼ uframe ¼ 0:190 uframe ¼ 1 þ Pi¼1n ð58Þ
58 u i¼1 y^2i
The largest displacement takes place in the For a variety of ordinary sheathed wood-framed
fasteners in the corners. The absolute value of this shear walls, the ratio of the two parts of the polar
displacement vector is then given by moment of inertia of the fasteners according to Eq. 14
Materials and Structures (2009) 42:301–320 319

is approximately 1/3, i.e. c=u  1:33. In cases where 18. Judd JP, Fonseca S (2005) Analytical model for sheathing-
Eq. 23 apply, the ratio of the two angles of rotation is to-framing connections in wood shear walls and diaphragms.
J Struct Eng 131(2):345–352. doi:10.1061/(ASCE)0733-
given by 9445(2005)131:2(345)
c 7 19. Åkerlund S (1984) A simple calculation model for
1þ ¼ 1:32 ð59Þ sheathed wood-framed shear walls. Bygg Teknik 1:45–48
u 22 (Sweden)
20. Yoon TY, Gupta AK (1991) Behavior and failure modes of
low-rise wood-framed buildings subjected to seismic and
wind forces. Department Civil Engineering, North Carolina
References State University, Raleigh
21. Salenikovich AJ (2000) The racking performance of light-
1. European Standard (2006) Timber structures – test meth- frame shear walls. Doctoral dissertation, Wood Science
ods – racking strength and stiffness of timber frame wall and Forest Products, Virginia Polytechnic Institute and
panels, prEN 594:2006. European Committee for Stan- State University, VA
dardization, CEN/TC 124. Brussel, Belgium 22. Ni C, Karacabeyli E (2000) Effect of overturning restraint
2. Girhammar UA, Källsner B (2009) Analysis of influence of on performance of shear walls. Proceedings of World
imperfections on stiffness of fully anchored light-frame conference timber engineering, Whistler Resort, British
timber shear walls — elastic model. Mater Struct. doi: Columbia, Canada, July 31–Aug 3, 2000
10.1617/s11527-008-9458-7 23. Ni C, Karacabeyli E (2002) Capacity of shear wall seg-
3. Fosci RO (1977) Analysis of wood diaphragms and trusses ments without hold-downs. Wood Des Focus 12(2):10–17
– part I: diaphragms. Can J Civ Eng ASCE 4(3):345–352 24. Henrici D (1984) Berechnung von Wandscheiben. Kapitel
4. Tuomi RL, McCutcheon WJ (1978) Racking strength of 5 in: Konstruktion und Berechnung von Holzhäusern in
light-frame nailed walls. J Struct Div 104(7):1131–1140 Tafelbauart, hrsg. von E. Cziesielski, Expert Verlag
5. Easley JT, Foomani M, Dodds RH (1982) Formulas for 25. Cziesielski E (1982) Stabilität von Holzhäusern unter
wood shear walls. J Struct Div 108(11):2460–2478 Horizontalbelastung. Bauen mit holz 84(7):446–450
6. Itani RY, Cheung CK (1984) Nonlinear analysis of 26. Henrici D (1984) Zur Bemessung windaussteifender
sheathed wood diaphragms. J Struct Eng 110(9):2137– hölzerner Wandscheiben. Bauen mit holz 86(12):873–877
2147 27. DIN 1052 (1979) Entwurf: Holzbauwerke – Holztafeln –
7. Källsner B (1984) Panels as wind-bracing elements in Berechnung und Ausführung, Teil 3, Ausgabe 02/79. DIN
timber-framed walls. Wood Technology Report 56, The Deutsches Institut für Normung e.V. Cf. also: Holzhaus-
Institute Wood Technology Research, Stockholm, Sweden richtlinien (1979) Richtlinien für Bemessung und
(in Swedish) Ausführung von Holzhäusern in Tafelbauart. Fassung 1979
8. McCutcheon WJ (1985) Racking deformations in wood (Ergänzung zu DIN 1052, Blatt 1 – Holzbauwerke)
shear walls. J Struct Eng 111(2):257–269 28. von Halász R, Scheer C (eds) (1986) Holzbau – Taschen-
9. Gupta AK, Kuo GP (1985) Behavior of wood-framed shear buch, Band 1: Grundlagen, Entwurf und Konstruktionen.
walls. J Struct Eng 111(8):1722–1733 Verlag für Architektur und technische Wissenschaften,
10. Gupta AK, Kuo GP (1987) Wood-framed shear walls with Berlin, Deutschland
uplifting. J Struct Eng 113(2):241–259 29. Henrici D (1983) Das Trag- und Verformungsverhalten
11. Schmidt RJ, Moody RC (1989) Modeling laterally loaded hölzerner Wandscheiben unter Schubbiegung, Teil 1. Holz
light-frame buildings. J Struct Eng 115(1):201–217 Zentralbl 109(131/132):1879–1880
12. Falk RH, Itani RY (1989) Finite element modeling of wood 30. Henrici D (1983) Das Trag- und Verformungsverhalten
diaphragms. J Struct Eng 115(3):543–559 hölzerner Wandscheiben unter Schubbiegung – Teil 2:
13. Filiatrault A (1990) Static and dynamic analysis of timber Versuchergebnisse. Holz Zentralbl 109(147):2105–2107
shear walls. Can J Civ Eng 17:643–651 31. Henrici D (1981) Vereinfachte Theorie der Verankerung
14. Dolan JD, Fosci RO (1991) Structural analysis model for schubbeanspruchter Wandscheiben aus Holz und Hol-
static loads on timber shear walls. J Struct Eng 117(3):851– zwerkstoffen. Bautechnik 58(1):6–10
861. doi:10.1061/(ASCE)0733-9445(1991)117:3(851) 32. Henrici D (1981) Verankerung schubbeanspruchter
15. Kasal B, Leichti RJ (1992) Nonlinear finite-element model Wandscheiben aus Holz und Holzwerkstoffen nach
for light-frame stud walls. J Struct Eng 118(11):3122– Versuchen. Bautechnik 58(5):163–167
3135. doi:10.1061/(ASCE)0733-9445(1992)118:11(3122) 33. Beuth-Kommentare (1989) Holzbauwerke – Eine aus-
16. Murakami M, Moss PJ, Carr AJ, Inayama M (1999) For- führliche Erläuterung zu DIN 1052 Teil 1 bis Teil 3. DIN
mulae to predict non-linear behaviour of sheathed walls Deutsches Institut für Normung e.V., Beuth Verlag GmbH,
with any nailing arrangement pattern. Proceedings of Berlin, Deutschland
Pacific timber engineering conference, Rotorua, New 34. DIN-Taschenbuch 34 (1994) Holzbau – Normen. DIN
Zealand, 14–18, pp 189–196 Deutsches Institut für Normung e.V., Beuth Verlag GmbH,
17. Fosci RO (2000) Modeling the hysteretic response of Berlin, Deutschland
mechanical connections for wood structures. Proceedings 35. Steinmetz D (1988) Die Aussteifung von Holzhäusern am
of World conference timber engineering, Whistler Resort, Beispiel des Holzrahmenbaues. Bauen mit holz 90(12):
British Columbia, Canada, July 31–Aug 3, 2000 842–851
320 Materials and Structures (2009) 42:301–320

36. Källsner B, Lam F (1995) Diaphragms and shear walls. In: 41. Kamiya F (1987) Buckling of sheathed walls: linear anal-
‘‘Holzbauwerke nach Eurocode 5 – Step 3. Informationsd- ysis. J Struct Eng 113(9):2009–2022
ienst Holz, Arbeitsgemeinschaft Holz e.V., Düsseldorf, 42. Kamiya F (1988) Buckling of sheathed walls: nonlinear
Germany, Chapter 15 analysis. J Struct Eng 114(3):625–641
37. Swedish Building Code (1979) Section 2A – load-bearing 43. Girhammar UA, Gopu VKA (1993) Composite beam-col-
structures, Chapter 21A. PFS 1979:7. Swedish National umns with interlayer slip – exact analysis. J Struct Eng
Planning Authority, Stockholm, Sweden 119(4):1265–1282. doi:10.1061/(ASCE)0733-9445(1993)
38. Eurocode No. 5 (2004) Design of timber structures – part 1- 119:4(1265)
1: general rules and rules for buildings. EN 1995-1-1. 44. Girhammar UA, Pan DH (2007) Exact static analysis of
European Committee for Standardization, Brussel, Belgium partially beams and beam-columns. Int J Mech Sci 49:239–
39. Eurocode No. 5 (1987) Common unified rules for timber 255. doi:10.1016/j.ijmecsci.2006.07.005
structures, EUR 9887. Commission of the European 45. Girhammar UA (2008) Composite beam-columns with
Communities, Brussel, Belgium interlayer slip – approximate analysis. Int J Mech Sci 50:
40. Girhammar UA, Bovim NI, Källsner B (2004) Character- 1636–1649. doi:10.1016/j.ijmecsci.2008.09.003
istics of sheathing-to-timber joints in wood shear walls. 8th
World conference on timber engineering, Lahti, Finland