Literatureandknowledgereview

literature review for MSc thesis:
MULTI-STOREY
TIMBER-FRAME
BUILDING
modelling the racking stiffness of timber-frame
shear-walls

Version 23-04-2012
Tunis Hoekstra
t.hoekstra@student.tudelft.nl
+31(0)638819574

Multi-storey timber frame building literature review 23-04-2012
page 2 / 49
Introduction
This report presents a literature review concerning stiffness- and force-distribution issues and modelling aspects
of timber-frame buildings. This review is a part of the MSc thesis Multi-storey timber-frame building. The aim of
the project is explained first. Secondly, the aspects governing the shear-wall behaviour are discussed. Thereafter
experimental test-data is presented, which will be used in the next phases of the masters thesis. The report on
the results of the literature review will be finished with a chapter concerning modelling and analysis of the
timber-frame structure. A reference list of the consulted sources is included in the last part of the report.
Table of contents
Introduction .................................................................................................................................................................. 2
Table of contents .......................................................................................................................................................... 2
1 Aim of the literature review .................................................................................................................................... 3
1.1 Problem definition ........................................................................................................................................ 3
1.2 Position of the literature review within in the project ................................................................................. 3
1.3 Intended thesis results .................................................................................................................................. 3
2 Effects on structural behaviour as found in literature ............................................................................................ 6
2.1 Shear-wall level ............................................................................................................................................. 6
2.2 Fastener and hold-down level ..................................................................................................................... 23
3 Experimental results .............................................................................................................................................. 27
3.1 Full-scale shear-wall tests ........................................................................................................................... 27
3.2 Full-scale shear-wall tests with opening ..................................................................................................... 32
3.3 Sheeting-to-timber fastener tests ............................................................................................................... 34
3.4 Hold-down anchorage tests ........................................................................................................................ 37
3.5 Compression perpendicular to grain tests .................................................................................................. 39
4 Issues concerning the modelling approach and code directives ........................................................................... 40
4.1 Modelling aspects ....................................................................................................................................... 40
4.2 Eurocode 5 methods for strength verification ............................................................................................ 44
4.3 NEN-EN 594 Racking test method ............................................................................................................... 44
5 References ............................................................................................................................................................. 46

page 3 / 49
1 Aim of the literature review
1.1 Problem definition
Each structural design needs to fulfil the demands in ultimate limit state (ULS) and serviceability limit state (SLS).
The requirements in ULS have to do with prevention of collapse, failure and rupture. In ULS, the force distribution
within the structure is often most important. Prevention of undesirable deformation or vibration is important
because these can affect the appearance or effective use of the building. Therefore, the stiffness of the structure
is analysed in SLS.
A safe and reliable estimation of the deformations is required before constructing multi-storey timber-frame
buildings. In Eurocode 5 (NEN-EN 1995), a calculation method is given to check the timber-frame wall and floor
panels for the requirements in the ultimate limit state. The force distribution in ULS is dependent of the stiffness
distribution within the structure. In the calculation method, the forces on the wall or floor diaphragm are
assumed to be known. However, the way in which the stiffness of timber-frame diaphragms has to be determined
is not explicit prescribed in the current code. Some directives are given to calculate a joint slip modulus (Kser) for
dowel-type fasteners. However, the applicability of this joint-slip modulus for stiffness estimations of shear-wall
diaphragms is unclear, and additional information about the theoretical and experimental backgrounds of this
modulus is not given. Hence, the information and directives from the code contain gaps on this topic. Several
approaches and assumptions are possible. Aim of this masters thesis is to gain knowledge about the stiffness
characteristics in both SLS and ULS to provide insight in analysis and modelling of the stability-system. This insight
will result in more reliable prediction of deformations, and force-transfer, in timber-frame buildings.
1.2 Position of the literature review within in the project
Information on shear wall, anchorage, and fastener behaviour and experimental tests, have been found in
literature and will be presented in the next chapters. In this paragraph, an overview will be given of the plan of
approach for the next parts of the project. In the remaining chapters of this literature review, the results of the
literature review will be presented. These will be analysed further in the next phases of this masters project. Two
different phases can be distinguished. Discussing these phases will show the application of the literature review.
First, the standard shear-wall will be analysed (phase A). The standard shear-walls will be analysed to get a clear
understanding of the modelling approach and stiffness values for standard walls. With standard walls is meant,
walls without openings (perforation), sheeted on one or both sides, height h = 2,7 m, width b = 0,6 4,8 m.
Secondly, when the analysis of standard walls has been finished, another step will be made to get a clear
understanding of the modelling approach and stiffness values for perforated walls (phase B). Perforated walls are
walls with one or more openings such as windows and doors. These walls can also be a composition of several
standard walls or wall parts. In figure 1 on the next page, phase A and B are explained more in detail. These
phases will not be elaborated in this report.
1.3 Intended thesis results
On the condition that the comparative research (in figure 1) do not give reason to change the intended plan of
approach, it is becoming clear what the result of the masters thesis will consist of. As described very shortly in
figure 1, the result will be a guideline on stiffness parameters and modelling suggestions for timber-frame walls.
For different wall families, or wall configurations, values will be given, which can be used as stiffness values in
modelling. As explained in figure 1, the stiffness values may be based on code directives (Kser). These code
directives will be clarified by comparison with experimental information. Aim is to find a modelling approach that
will be suitable for analysis with the software available in the company of Boorsma Consultants; it has to fit to the
possibilities of common framework structural analysis software in 2D as well in 3D. Two different alternatives for
modelling will be chosen, first a braced framework model, in which the brace stiffness can contain the
information about fastener, sheeting and anchorage properties. Another (more advanced) type of model can
consist of plate element(s) with certain shear modulus G. In this case, G can be seen as a stiffness parameter
containing the information about fastener, sheeting and anchorage properties, while the plate geometry can be a
model of openings in the wall.

page 4 / 49

figure 1: Intended plan of approach for the remaining part of the project.
B
L
i
t
e
r
a
t
u
r
e

r
e
v
i
e
w

A
n
a
l
y
s
i
s

o
f

s
t
a
n
d
a
r
d

w
a
l
l

b
e
h
a
v
i
o
u
r

A
n
a
l
y
s
i
s

o
f

p
e
r
f
o
r
a
t
e
d

w
a
l
l

b
e
h
a
v
i
o
u
r

Comparison of analytical derived
shear-wall stiffness with shear-wall
stiffness determined from shear-wall
test-data
Determine racking stiffness
value for standard shear-walls
from shear-wall test-data using
NEN-EN 594
Eurocode 5:
joint slip modulus Kser
Analytical derivation of stiffness
parameters for modelling equivalent
shear modulus G for modelling
approach with equivalent plate
model
Modelling & Analysis perforated
shear-walls using equivalent shear
modulus G in a plate modelling
approach
Comparison of perforated shear-
wall modelling & analysis results
with:
Determined racking stiffness
values from test-data
Results of analytical methods
(methods found in literature)
Determine stiffness value
ks for fasteners from test-
data using NEN-EN 26891

Comparison of determined fastener
stiffness value ks with Eurocode 5
approach for joint slip modulus Kser
Analytical derivation of racking stiffness value for
shear walls from:
determined fastener stiffness values
determined anchorage stiffness values
structural mechanics
Comparison of determined
anchorage stiffness with Eurocode
5 approach (using Kser)
Test-data of:
fasteners
hold-down anchorage
shear-walls
perforated shear-walls (with
opening)

Determine stiffness
value for hold-down
anchors from test-data
using NEN-EN 594

A
R
e
s
u
l
t
s

Evaluation: If match is proven, between racking- stiffness determined from tests on perforated walls, and racking
stiffness derived by modeling perforated walls (using plate model & parameter G), the results could consist of:
Modeling suggestion: plate model with G and / or braced framework model with parameter k or k
Equivalent stiffness values G , k and/or k : for a certain range of shearwall geometries and configurations (a certain
sheeting material & fastener)
Determine racking
stiffness value for
perforated shear-walls
from perforated shear-wall
test-data using NEN-EN 594
Alternative: equivalent stiffness
parameter k for a braced
framework modelling approach
can also be derived analytical
Alternative: a stiffness
parameter k for a braced
framework modelling approach
can also be derived analytical
page 5 / 49

figure 2: Illustration of the plan of approach.

k
u

F

G

Comparison between stiffness value derived by modelling and analysis, and stiffness
determined from test-data

G

page 6 / 49
2 Effects on structural behaviour as found in literature
Studying the literature it became clear that within the area of timber-frame construction many subjects can be
distinguished. Several interesting subjects were discovered. Research into timber-frame structures is already
done for many years. Next to this, it is obvious that research topics vary from less profound to very detailed
studies. Studying and discussing every paper and piece of literature would not be useful within the scope of this
project. Therefore, the following paragraphs will treat only the most important findings. These are results,
relevant to the goals of the masters thesis. A distinction is made between the effects playing a role in stiffness of
timber-frame structures, and the experimental results that were found. In this chapter the several phenomena
that are of influence on the stiffness are discussed, in the next chapter the experimental information is presented.
To present the findings in a structured way two different levels are distinguished. The effects and influences on
stiffness of the structure will be considered on shear-wall level, and fastener and hold-down level. An overview of
the next paragraphs is presented below.

2.1 Shear-wall level
2.1.1 Force distribution in the shear-wall sheeting
2.1.2 Influence of fasteners on the stiffness of shear-wall panels
2.1.3 Influence of hold-down characteristics on the stiffness of shear-wall panels
2.1.4 Influence of aspect ratio (h/b) on the stiffness of shear-wall panels
2.1.5 Influence of number of shear-wall panels on the stiffness of shear-wall diaphragms
2.1.6 Influence of asymmetric sheeting on the stiffness of shear-wall panels
2.1.7 Influence of openings (perforation) on the stiffness of shear-wall panels
2.1.8 Diaphragm action of floors
2.2 Fastener and hold-down level
2.2.1 Fasteners
2.2.2 Hold-down anchors

2.1 Shear-wall level
In this paragraph, some aspects will be presented which influence the shear-wall behaviour. The discussion of
these aspects is based on what came across in the literature research. The findings are presented in a way of
comparison and discussion, to prevent copying the literature. Therefore, always reference is made to the original
authors of research (see the references section on page 46).

The timber-frame wall is a building element which can be part of the stabilizing structure. In such a case the
timber-frame shear-wall has a load-bearing function, not only for the purpose of transferring vertical loads, but
also for the transfer of the horizontal forces which act on the building. In case of wind for example, the horizontal
wind-load is transferred to the timber-frame shear-wall by diaphragm action of the floors. The racking force on
the shear-wall is divided over the wall length, this results in a distributed load on top of the wall. This distributed
load will be transferred to the foundation or substructure by the shear-wall.

The shear-wall itself is a composite element. A timber-frame wall consists of timber frame elements, connections,
wood-based or gypsum-based panels, and fasteners: screw, nail or staple. The shear-wall is able to transfer the
racking forces thanks to the sheeting which acts as a brace to the timber framing elements. The shear-force on
top of the wall-element is transferred to the sheeting by means of fasteners which are located along the
perimeter of the sheeting.

page 7 / 49

figure 3: Assembly of the timber-frame shear-wall (h/b 1).

2.1.1 Force distribution in the shear-wall sheeting
First of all, clarification is given on the force distribution in the timber-frame sheeting. In the analysis of timber-
frame shear-walls, assumptions have to be made to facilitate the calculations. These assumptions also include the
force distribution in the sheeting. Several approaches can be followed to approximate the force-distribution in
the sheeting. These approaches will be shown below. For the discussion of the different methods, use is made of
the information provided by (Kllsner & Lam, 1995) and (Kllsner & Girhammar, 2009).

figure 4: Forces from the fasteners on the sheeting of a fully anchored timber-frame shear-wall (h/b = 2). (a): elastic
force distribution, (b): plastic force distribution according to the lower bound method as adopted in the Eurocode 5,
(c): plastic force distribution according to the upper bound method. (Kllsner & Girhammar, 2009)

In figure 4 (a), a force distribution according to a linear elastic model is shown. In a linear elastic model, the
sheeting-to-timber joints have linear elastic load-slip characteristics up to failure. The joint stiffness is
independent of force direction and grain orientation of the sheeting and timber framing elements. As can be seen
from the figure, the load on the fasteners is of different magnitude for each position. From comparison with
CR
(a) (b) (c)
CR
CR
CR
CR = center of rotation of stud / rail
F
p

F
p

F
p
/2
F
p
/2
F
p

F
p

timber framing / bottom rail
h
e
i
g
h
t

h

=

2
6
0
0

m
m

hold-down
fasteners (screw / nail / staple)
panel / sheeting (OSB / plywood / particleboard / gypsum (fibre) board)
timber framing / top rail
timber framing / leading stud
width / length b = 2400 mm
distributed load (from floor diaphragm)
page 8 / 49
experimental testing can be concluded that the elastic model generally will lead to a underestimation of the
shear-wall capacity. Nevertheless, the horizontal displacement of the wall calculated according to the linear
elastic model agreed very well with the results of full-scale shear-wall tests.

In figure 4 (c) a plastic force distribution is shown, based on the plastic upper bound method. Each of the five
timber members rotates around its own centre of rotation. Sheeting-to-timber joints are completely plastic, the
properties are independent of the force direction and grain orientation of the sheet and timber members. The
loads on the fasteners are of the same magnitude, because the fasteners have plastic characteristics. By
experimental testing, the use of this method turned out to result in an overestimation of the racking capacity of
the shear wall.

In figure 4 (b), a force distribution is shown according to a lower bound plastic model. In this model, a pure shear
flow along the perimeter of the wall is assumed, on the condition of force and moment equilibrium. The force on
each fastener is at most equal to the plastic capacity of the fastener. The framing members, the timber framing
elements, are considered to be completely flexible. Therefore, the force distribution will become parallel with the
framing members. This in contrast with models mentioned in (a) and (c). Model (b) turned out to correspond very
well with experimental testing, giving values for the racking capacity lower than, or equal to the exact value
obtained by testing. Besides this, the method is suitable for, and very useful to, hand calculation. This may have
been the reason to include the plastic lower bound method in Eurocode 5. The method is based on a static
theorem, while the upper bound method (c) was based on a kinematic theorem. In a kinematic theorem, a
geometrically possible pattern of deformation is assumed. By means of the principle of virtual work, the internal
work of all the fasteners is made equal to the work of the external forces. Therefore, the centre of rotation (CR) of
the timber frame elements is shown in figure 4 (c).

As a consequence of the fastener forces on the sheet, a force distribution in the sheet will develop as shown in
the figure below. The force distribution in the sheeting will be visible when the buckling mode of the timber-
frame shear-wall is observed. In case of buckling, the compressive stress in the sheeting was too high, and caused
the sheeting to deform out of plane. Due to the remaining tensile forces in the diagonal, the sheeting can
maintain its function partly.

figure 5: Development of tension and compression in sheeting diagonals.

figure 6: Example of buckling in case of a thin plate / sheet, without intermediate stud. Test set-up
for experimental verification. (Kessel & Sandauw-Wietfeldt, 2004)

tension
compression
page 9 / 49
2.1.2 Influence of fasteners on the stiffness of shear-wall panels
From the paragraph before became clear how the fasteners along the edge of the panel provide for the transfer
of a distributed load from the timber framing elements to the sheeting. Without fasteners and sheeting the frame
would be an unstable mechanism. Consequently, the stiffness of the wall depends very heavily on the stiffness of
the fasteners. The disproportionate influence of the fasteners was already considered in 1985. According to
William McCutcheon, the racking behavior of a wood shear wall depends primarily upon the strength and
stiffness of the fasteners that attach the sheeting to the frame. It is possible to predict the performance of the
wall, if the load-displacement characteristics of the individual nail is known. The load-slip properties of the
fasteners can be determined by individual nail tests. (McCutcheon, 1985) Similar considerations can be found in
publications of (He et al., 2001) and (Girhammar et al., 2004). The latter published a study on the characteristics
of sheeting-to-timber joints in timber-frame shear walls. From the numerical work of (Conte et al., 2011) a part of
the report is mentioned below:

For the horizontal stability of wood framed buildings, diaphragm action in the walls is of crucial
importance. The structural behavior and capacity of wall diaphragms are primarily dependent on
the sheathing-to-timber joints. (.) Considering the presence of a lateral load F at the top of the
shear walls, the timber frame must be regarded as unstable, and unable to counteract external
force, which is only resisted by the sheathing and metal fasteners. Apart from the possibility to
uplift of the shear wall, the total lateral displacement is function of fastener slip and shear
distortion of the sheathing (which is usually relatively small and often neglected). Also the global
lateral strength capacity of shear diaphragm is often determined by the resistance of the local joint
between timber frame and sheathing panel (). (Conte et al., 2011)

In the figure below, the separate contributions to the horizontal deformation of a shear-wall element made of
two shear-wall panels are compared. The racking deformation of a timber-frame shear-wall consists of: (a) slip of
the fasteners, (b) strain in the hold-down anchorage, (c) shear deformation of the sheeting, and (d) slip between
the bottom rail and substructure. The remaining issues are considered to be insignificant.

Parameters, which can be varied to influence the effect of the fasteners on the shear-wall behaviour, are:
choice of fastener type: screw, nail, staple,
choice of fastener properties: diameter , and length
spacing of the fasteners (s) along the edges of a panel, being the distance between individual fasteners

figure 7: Separate contributions to the total horizontal racking deformation of the timber-
frame shear-wall. (Conte et al., 2011)
(a)
(d)
(c)
(b)
page 10 / 49
The fastener properties themselves are influenced by the thickness and density of the board material, and the
density of the framing timber. What choices can be made to influence the fastener characteristics and what the
influence is of these choices will be explained in 2.2 (page 23).
2.1.3 Influence of hold-down characteristics on the stiffness of shear-wall panels
A wall panel will most often look like, as drawn in figure 8. When a wall panel is subjected to an in-plane racking
force (Fi,v,Ed), the fasteners will destort, and the toprail of the shear-wall will displace horizontally. If no vertical
load is applied, a tensile force (Fi,t,Ed) will develop in the tensile stud. Uplift of this stud needs to be prevented by
the hold-down anchorage.

When doing a racking test, the aplication of sufficient amount of anchorage, can be equivalent with applying a
vertical load. It is also possible to do racking tests by loading the panel with a diagonal load. These three options
are all equivalent if no uplift occurs. In figure 9 is shown what type of force distribution will occur in case of
sufficient anchorage, and what type of force distribution will develop when no hold-down or, insufficient
anchorage is applied. These differences have an enormous influence on the performance of the wall, and on the
utilization of material.

figure 8: Timber-frame shear-wall panel with fastener
spacing (s).
figure 9: Forces on fasteners exerted by the sheeting: (L) in
case of sufficient hold-down capacity, (R) in case of
insufficient hold-down capacity.

As can be seen from the right hand side of figure 9, the tensile forces caused by the horizontal force (Fi,v,Ed) have
to be transferred trough the plate to the fasteners in the bottom rail, because of the lack of tie-down anchorage
or vertical load. Due to this disadvantageous force distribution, the fasteners in the bottom rail are loaded
perpendicular to the edge of the panel, and perpendicular to the grain of the bottom rail. This will cause a very
bad failure mode, which is reason for brittle behaviour of the wall panel.
Often no-hold down is applied in the test set-up. Sometimes the argumentation is that it is not common practice
to provide the hold-down anchorage, or, from the viewpoint of costs, an engineered solution is not preferred
because hold-downs are very costly. Except for the cases where the vertical permanent load can prevent tensile
forces to develop, it may be clear that effective utilization of materials only can be reached when structures are
equipped with hold-down devices. (See figure 10 & figure 11 on the next page.)

Fi,v,E
d
2s
s
fasteners
hold-down
bi
h

Fi,t,Ed
Fi,c,Ed
page 11 / 49
The effect described above is also found in several tests. Among others, Dolan and Heine did experimental testing
to get more insight into the effect of hold-down anchorage on shear-wall behaviour. They conclude the following.
Failure modes for walls with no overturning restraint were obviously different. For all three walls the typical
failure mode was the tension end stud seperating from the bottom rail and the sheeting unzipping from the
bottom rail (Dolan & Heine, 1997).
Insufficient hold-down capacity magnifies other design choices
As could be concluded from the paragraphs above, the prevention of uplift is also very important for other design
choices. Walls provided with openings, like windows and doors, are much more prone to the presence or absence
of hold-down anchorage. In these cases, the effect and influence of anchors on the stiffness of the wall is
relatively higher than for standard non-perforated shear-wall diaphragms. See also 2.1.7 where the influence of
openings is elucidated.
Prevention of uplift is very important when load-sharing between different panel sides is considered, in
asymmetric timber-frame shear-wall panels, with gypsum and OSB sheeting for instance. The anchorage condition
influences the failure mode of asymmetric sheeted shear-wall assemblies. In this case providing enough hold-
down anchorage is required to be able to add the contribution of the gypsum sheet in the panel stiffness to the
contribution of the OSB sheeting. In 2.1.6 this will be explained more in detail.

figure 10: Unzipping of sheeting from the bottom rail
(Ni et al., 2010)
figure 11: Lack of anchorage / vertical load causing the
fasteners to tear through the sheeting edge. (Salenikovich,
2000)

page 12 / 49
2.1.4 Influence of aspect ratio (h/b) on the stiffness of shear-wall panels
A common design of shear walls for timber-frame buildings also consists of space for doors and windows.
Architectural reasons often lead to slender walls. Take for instance the wish for flexibility in plan. In such a case,
minimal wall length is required. Large spaces are kept free for having big windows over the full floor height. Due
to these openings, shear wall diaphragms consist of shear wall panels with different widths. In the literature
review, research was done regarding the effects from height-to-width ratio on stiffness.

figure 12: Example of timber-frame building
with large window spaces. (Arkana, 2011)

From tests on walls with a length in the range of 0,6 until 3,6 m in plan of the building, was found, that walls of
1,2 m and longer tend to develop the same stiffness. Walls 0,6 m long had 50% lower stiffness relative to the long
walls. Due to the low racking stiffness, the sheeting connections in these walls did not fail, which lead to the
extraordinary ductility (Salenikovich, 2000). The large drifts arise due to a greater effect of body rotation, the
uplift displacements of the end studs contribute to horizontal deflections in proportion to the aspect ratio
(Salenikovich & Dolan, 2003). In STEP 3 this effect is described is following. A wall panel with a small width to
height ratio will reach its maximum load at a rather large displacement. A wall panel with a large width to height
ratio will reach its maximum load at a lower displacement. Therefore the full plastic capacity of the wall with
small width to height ratio might never be reached. (Kllsner & Lam, 1995)

Lower values for the stiffness of walls with a smaller width can be explained by comparing a slender wall with a
standard wall. Let the walls be loaded with the same shear load:

figure 13: Wall panels with height-to-width ratio 4:1 & 2:1 (Forces acting on the wall panel, shear forces on the sheeting, magnification
of
v
.

As can be seen from figure 13, the walls are executed in the same way. The walls are loaded to the same extend;
an equal shear force is developing on the edges of the panel.
v
is the total vertical deformation due to anchorage
slip, and lengthening / shortening due to tension and compression in the leading studs. Because of the small
width of the slender wall (4:1), the deformation
v
is magnified more than for the wall with ratio 2:1.

4:1
2:1
page 13 / 49
The influence of the aspect ratio was also considered in an analytical calculation method (Kllsner & Girhammar,
2009). Based on the conditions of full anchorage a plastic model was derived in which: h/b is the height-to-width
ratio of the wall, s
r
is the fastener spacing of the sheet-to-timber connections, b is the width of the wall panel, H is
the horizontal racking load, and k is the stiffness of the fastener sheet-to-timber connections. The stiffness of a
shear wall panel can be computed by:

(1)

When this relation is plotted for different height-to-width ratios, the decreasing stiffness for higher ratios is
evident.

figure 14: k
frame
as function of h/b. (h/b = 2 )

From literature, and simple analysis can be concluded that for walls with slenderness ratio higher than or equal to
4:1, the stiffness properties of the standard wall (2:1) have to be reduced with 50%. Walls with aspect ratios 2:1
developed identical load-deformation patterns (Salenikovich & Dolan, 2003). On first hand, this 2:1 limit might
be strange. An explanation, found in literature, will be given in the next paragraph.
2.1.5 Influence of number of shear-wall panels on the stiffness of shear-wall diaphragms
In paragraph 2.1.4, the increase in stiffness for ratios h/b lower than 2 was shown. For common design, the
stiffness obtained on h/b = 2 is the maximum stiffness which can be reached. This is related with the maximum
available dimensions of sheeting. Because of the standard dimensions in the manufacturing of the sheet
materials, widths of the shear-wall panels up to 1,2 m are possible. Often the height is approximately 2,4 m,
which lead to a ratio h/b = 2. A shear wall diaphragm consists of multiple shear-wall panels when the length
exceeds the 1,2 m limit. The stiffness of the wall diaphragm will not increase by doing this because the sheets
rotate separate when the wall diaphragm is loaded. See also the figures below. Evidence is also found in
experimental testing. Test results revealed that the performance of the shear wall did not depend on the number
of shear-wall panels, where a panel is defined as the elementary timber-frame shear-wall element, with a width
of one single sheeting, usually equal to 1,2 m.

figure 15: Rotation of the sheeting of a shear wall panel
as observed in experimental test. (Kllsner, 1984)
figure 16: Rotation of sheeting of a shear
wall panel. (Ni et al., 2010)
0,00
0,05
0,10
0,15
0,20
0,25
0,30
0,35
0,40
0 1 2 3 4 5
k
f
r
a
m
e
ratio h/b
k
frame
as function of h/b
page 14 / 49
2.1.6 Influence of asymmetric sheeting on the stiffness of shear-wall panels
The timber-frame walls, which are often used in a cavity wall application, can be executed with sheeting at one or
both sides. Often wall panels are made with two types of sheets. For instance: Gypsum (GKB) sheeting at the
interior of the building, and OSB sheeting at the cavity side of the wall. In literature, information is found about
the contribution of both sheet materials in the total wall behaviour. Insufficient hold-down anchorage was found
to give reason for different results. Eventually a conclusion could be drawn from the publications.

For the described walls, the individual load-displacement curves can be superimposed when sufficient tie-down
capacity is provided by hold-down anchorage and / or vertical load. In this way, an equally distributed shearforce
will develop along the edges of the shear wall panel. (See also 2.1.2) Another requirement is that the different
material / fastener combinations that are used have to show enough deformation capacity until the level of
shear-wall diaphragm utilization.

Research performed by Wolfe in 1983 revealed that the law of superposition explained the contribution of
Gypsum to the ultimate shear capacity of OSB/Gypsum sheeted shear-walls. Wolfe tested 30 walls to study the
contribution of GKB to the racking resistance (Sinha & Gupta, 2009). A first comment to the conclusion of Wolfe
could be made after Karacabeyli and Ceccotti did a similar research in 1996. They verified that the law of
superposition was valid up to a drift of approximately 50 mm for monotonic tests (Ceccotti & Karacabeyli, 2002).

figure 17: Plot of envelope of force-displacement curve for
combination of OSB and Gypsum paper board. (Ceccotti & Karacabeyli,
2002)

When these findings are studied further, it appears that the principle of superposition also holds for the stiffness
of the walls. The deformation of both sheets will be the same because they are fastened to the same frame. The
system can be seen as a parallel set of springs:

figure 18: Model of the sheets on both sides
of the wall panel: parallel set of springs.

(2)

k
(3)

(superimposed)
(experimental verification)
page 15 / 49
The possibility of adding the capacity and load-displacement behaviour of gypsum sheeted panel side to the
capacity and load-displacement behaviour of the plywood side of the shear-wall panel was also demonstrated by
Johnson in 1997. He concluded that GKB will help to resist shear at moderate load level, but plywood resists most
of the shear at higher load levels (Sinha & Gupta, 2009).

Comparison of load-displacement curves for wall panels, consisting of normally used combinations of interior and
exterior sheeting materials in Sweden, showed that the maximum load is obtained at about the same
displacement. Considering the ultimate limit state, the following conclusions were drawn. If a wall panel is built
with sheets of different types or thickness on both sides, a reduction of the weaker side has to be calculated with.
Based on experimental data a reduction of the load carrying capacity of the weaker side should be in the range of
20%. Only if fastener sheet combinations with completely different slip properties were used, a reduction of 50%
is justified (Kllsner & Lam, 1995). The same is written in the Eurocode 5, where a reduction of 25% is prescribed.
(See 4.2.) These findings can be explained by studying figure 17 more detailed. When ultimate racking load is
reached for the combined action, the gypsum part of the wall is already over its ultimate capacity, therefore it is
logical to reduce its attribution in the assembly of gypsum and OSB when considering ultimate strength limits.

Not only the load-bearing capacity of the weakest side needs to be reduced when considering combined action,
also the deformation capacity of the wall is reduced. This was already stated by Karacabeyli and Ceccotti in 1996,
and confirmed from research performed in 2003. Uang and Gatto observed a 12% increase in shear wall strength,
and 60% increase in initial stiffness, but also, a decrease in shear-wall deformation capacity of 31%.
Independently, Toothman found comparable results (Sinha & Gupta, 2009). Explanation of the effect is, that the
material, which originally showed the largest deformation capacity, is subjected to a higher load in the combined
situation, than in the situation of single performance. This can be seen from figure 17.

Sinha and Gupta performed an unusual study to get better insight in the combined action and effects of
combined action. Using digital optical imaging techniques, they were able to record the strains on the surface of
the shear-wall panel sheeting. 16 Full-scale timber-frame shear-walls were tested, 11 were sheeted on both sides,
with OSB on one side and GKB on the other. Five walls were tested without GKB. From the tests they derived the
following conclusions. Upon failure of the GKB sheeting, the GKB carries a load four times higher than the OSB
sheeting because of the higher initial stiffness of the gypsum sheeting. However, when GKB starts to fail, the
strain in OSB will increase on a higher rate. Hence, OSB is taking over the load from the gypsum sheeting. This
continues up to complete failure of the GWB sheeting. On first sight, the research of Sinha and Gupta showed
different results than the previous mentioned tests. The elastic shear stiffness is increased by 50%. However, the
contribution of GKB to the strength of the wall is very low, namely, 0,8%. See also figure 19.

figure 19: Load-displacement graph for combination of
OSB and Gypsum paper board. (Sinha & Gupta, 2009)

Sinha and Gupta explained the difference between these results and the test results of Karacabeyli and Ceccotti,
and Toothman, by the variation in the size of walls and different fasteners used for attaching OSB and GKB to the
shear-wall panel framing (Sinha & Gupta, 2009). However, are the different fastener characteristics really the
explanation for the difference in test results?

page 16 / 49
It seems very logical to attribute the brittle behaviour of the gypsum sheeting material to the sheeting-to-timber
fastener behaviour. However, when a wall is sufficiently tied down, a shear force parallel to the framing members
will develop along the edge. From the experimental information (3.2) can be concluded that a dry-wall screw
2,31x41,3 will not show brittle behaviour. The shift from load taken by the gypsum panel to the OSB panel
should develop more gradually. The correct explanation for the low contribution of gypsum in the ultimate
strength is the vertical uplift of the stud. The hold-down capacity of the anchorage provisions was too low. This
results in a force on the fasteners perpendicular to the edge of the gypsum panel in the lower corner near the
hold-down anchor. Fasteners tearing through the gypsum board are the reason for the brittle behaviour. (See also
2.1.2.)
Conclusion
As could be seen in the present discussion concerning load-sharing between gypsum and OSB in a shear-wall
assembly, three conclusions can be drawn.
Load is shared by both OSB and GKB initially in a shear wall assembly. When the gypsum sheet is starting
to fail, the load will shift to the OSB panel.
GKB will contribute to the overall strength of the shear wall, and it will increases the stiffness of the wall,
provided sufficient tie-down capacity or vertical load is applied.
For OSB also Plywood or another similar wood-based panel can be read.

page 17 / 49
2.1.7 Influence of openings (perforation) on the stiffness of shear-wall panels
In this paragraph the influence of openings (perforations) on the racking stiffness of timber-frame shear-walls will
be discussed. Several reports and papers on this topic were found in literature. An abstract of these will be
presented below. Reference is made to 3.2, where test-data is presented regarding timber-frame shear-walls
with opening.
Reduction of stiffness
Doudak and Smith tested seven different configurations of shear-wall panels. Only geometry and boundary
conditions were changed. The fastener and sheeting configuration remained the same for all panels. Wall lay-out:
OSB t = 11,1 mm 1200 x 2400 mm, overall wall dimension: b x h = 2400 x2400 mm, studs spaced 400 mm center
to center, fastener: nail ??x60,fastener spacing s = 150 mm

figure 20: Shear-walls with opening: wall2, wall 4, wall 6 and wall 7. (Doudak & Smith, 2009)

Vertical load was applied on top of the walls respectively 4,16 kN/m on wall 2 and 2,08 kN/m on walls 4, 6 and 7.
No significant uplift was observed. A certain initial stiffness was calculated from the test-data. These stiffness
values are stated in table 1. From the tests can be seen that perforations can cause a reduction in the stiffness of
the wall-diaphragms. Despite the fact that tie-down anchorage is applied in the inside bottom corners and in the
door openings, adding a door (4) to the standard wall (2) is the reason for a reduction of 44 % in stiffness. The
door dimensions were 1938 x 838 mm which is 35% of the wall width. An explanation for the magnification of the
effect, could be the reduced panel width. The standard wall is sheeted with OSB. The width of these sheets, 1200
mm, will decrease to 781 mm because of the door. This is a reduction in width of the wall of 35%. In 2.1.4 was
explained that this will lead to a relatively higher reduction in stiffness (44%).

Wall
test
Initial
stiffness
(N/mm)
Ultimate
racking load
(kN)

figure 21: load-displacement graphs tested walls. (Doudak & Smith, 2009)
Secants are drawn at 10% and 40% of the ultimate load.
2 558 20,8
3 313 9,3
4 380 11,2
5 416 10,9
6 562 12,1
7 603 15,3
table 1: Calculated values for initial stiffness of
tested shear wall panels with perforations.
(Doudak & Smith, 2009)

On first sight the table reveals a strange effect, when adding a window to the standard wall, the initial stiffness of
the wall increases. This effect could happen because of the method to calculate the initial stiffness. Because the
response of the walls is nonlinear, the initial stiffness is taken as a secant stiffness defined using racking loads and
associated racking deformations at 0,1 and 0,4 times the ultimate racking load. This is intended to characterize
the response in the recoverable deformation range. (Doudak & Smith, 2009) Because the ultimate racking load
decrease, the curve length over which the initial stiffness is calculated is much smaller, this results coincidentally
in a higher initial stiffness. Another remark stated in the paper is the sensitivity of the wall behaviour for small
6 2 7 4
page 18 / 49
changes in boundary conditions. Take for instance the difference between wall 6 and 7. Because the tie-down
anchor in wall 7 was mounted to the outside edge of the tensile stud, instead of the inside bottom corner of the
wall, the stiffness increased with 7,3%. According to Doudak and Smith, the wall perforations have a
disproportionate influence on the stiffness of the wall panels when hold-down anchors are omitted. In such a
case, the wall stiffness of wall 2 was reduced with 56% by adding a door (wall 3).
Sugiyama and Yasumura empirical equation
In 2006, Dujic, Klobcar and Zarnic reported about research done by Sugiyama and Yasumura in 1980. (Dujic et al.,
2006) Sugiyama and Yasumura conducted experimental tests on timber-frame shear-walls sheeted with plywood.
These were not only full-scale tests but also reduced scale tests (1:3). Based on the results of their research they
developed a method to reduce the racking stiffness and ultimate racking capacity of timber-frame shear-walls.
When the racking stiffness and ultimate racking capacity of a certain wall without perforations is known, the
racking properties of the same wall with perforations can be calculated using the formulas developed by
Sugiyama and Yasumura. In the Sugiyama and Yasumura method, the panel area ratio r is calculated. Using this
ratio the stiffness and strength of a perforated shear-wall can be determined as following:

(4)
and:

(5)
In which: is the reduced racking stiffness of a perforated shear-wall
is the racking stiffness of the non-perforated shear-wall
is the ultimate racking capacity of the perforated shear-wall
is the ultimate racking capacity of the non-perforated shear-wall

and:

(6)

In which: is panel area ratio
is height of the wall element
is length of the wall element

length of full height wall segments

sum of openings

is ratio of openings in wall element

is ratio of full wall segments

figure 22: Definition of parameters for calculating panel
area ratio.

Results of testing done by Dolan and Johnson in 1996 showed the applicability of the empirical relation
determined by Sugiyama and Yasumura. The results also indicated that the panel area ratio method is slightly
conservative especially at smaller ratio of openings (Dujic et al., 2006). Sugiyama and Yasumura developed the
empirical equation provided hold-down anchorage is only applied on the end studs of the wall. No additional tie-
down anchorage at intermediate locations of individual shearwall segments was provided.

According to the results of another series of tests performed by the American Plywood Association in 1994, the
empirical equation was also proven to provide correct or slightly conservative values. The difference between the
equation and test results does not exceed 10% (Douglas & Sugiyama, 2001).

Douglas and Sugiyama stated that the method is applicable to use in practice. The statements below were
especially introduced to receive wider acceptation of the procedure in the traditional US design methodology:
L
3
L
2
L
1

A
1

A
2

h
L

page 19 / 49
Shearwall segments are considered to be part of the total full-height length, if the length of the segment
is equal to, or larger than, one-half of the wall height.
Shear capacities of the sheeting materials have been limited to a maximum ultimate shear capacity of
1500 plf. (Equal to 21,89 kN/m.)
Hold-downs with the capacity to resist the maximum overturning in a wall segment are required at each
corner and at all discontinuities that occur in the wall line.

figure 23: Graph displaying the relation between shear stiffness ratio
and panel area ratio.

From research of Dolan and Heine can be concluded that the amount of tie-down anchors influences the load-
displacement characteristics and failure mode. The magnitude of the influence of these anchors depends on the
amount of openings in the wall. If there is a higher amount of perforation, every additional tied-down anchor on a
stud near to an opening will be more effective than in a situation with lower amount of perforation. The study
also reveals an interesting issue with respect to the curve determined by Yasumura and Sugiyama (figure 23). The
amount of conservatism of this relation increases as the amount of opening in the wall increase. This can also be
stated as following. The lower the panel area ratio, the larger the underestimation by the Yasumura and Sugiyama
curve of the stiffness of the perforated element. (Dolan & Heine, 1997)

Conclusion
To conclude with, the most important remarks that can be made to the empirical equation are:
only hold-down anchorage was applied on the edges of the wall diaphragm, no tie-down provisions on
studs next to openings was applied
tests which were the basis for the equation, and tests confirming the applicability of the equation, did not
take in account the effects of vertical loading

An analysis will be made if the Sugiyama and Yasumura equations are applicable to the test-data as found in
literature. This analysis will be done in a later phase of the masters project.

0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
Panel area ratio, r
Ratio of shear stiffness as a function of panel area ratio
page 20 / 49
2.1.8 Diaphragm action of floors
Horizontal forces caused by the wind on facade and roofing are transferred to the shear-walls by the horizontal
diaphragms. These are the floor and roof planes of the building. How the horizontal diaphragms distribute the
horizontal loading over the shear-walls is not only dependant of the stiffness of the shear-walls themselves, but
can also be dependant of the stiffness of the horizontal diaphragm. This is explained with the figure below where
the stiffness of the shear-walls themselves is not regarded.

figure 24: Differences in force distribution over the shear-walls in case of a
flexible (A) and rigid (B) horizontal diaphragm. (Banga & Graaf, 2002)

Hence, the in-plane stiffness of the floor and roof diaphragms is influencing the distribution of forces over the
shear-walls. On the question how to determine if timber-frame floor diaphragms can be seen as either rigid or
flexible, information is found in literature.

Horizontal diaphragms can only be considered as rigid, when the configuration of the diaphragm satisfy a number
of requirements. One of the demands can be found in a publication of Kasal, Collins, Paevere, and Foliente from
2004. They state that the flexibility of the diaphragm has to be defined relative to the rest of the structure. The
magnitude of expected deflection of the horizontal diaphragm relative to the shear-wall deformation can be used
to characterize a diaphragm as rigid or flexible. A rule is stated in the International Building Code that a
diaphragm is rigid when the deformation of the diaphragm is less than two times the average racking deformation
of the shear-wall diaphragms. For most wood-frame buildings, the diaphragms can be considered to be rigid, due
to the significant larger shear wall deformations compared with the floor-diaphragm deflection. (Kasal et al.,
2004) Comments on the backgrounds of the IBC statements however are not present in the paper. More research
into this topic would be justified.

A qualitative requirement would be that the sheeting panels are fastened to the framing along the entire
perimeter. Therefore, the floor structure has to be provided with so-called blocking.

figure 25: Blocking timber applied between the main joists. (Falk &
Itani, 1989)

The effect of blocking was demonstrated in experimental testing performed by Bott in 2005, and numerical
modelling done by Falk and Itani in 1989. If blocking is left away, a force distribution will develop different from
what was already explained in 2.2.1 and drawn in figure 9 (page 10). Forces perpendicular to the sheeting edges
and the phenomenon of buckling will result in unfavourable force distribution and failure mechanisms. Falk and
Itani investigated the effect of blocking on the stiffness of the floor diaphragm by analyzing the diaphragm with
blocking, and without blocking. From their numerical work could be concluded that leaving blocking from the
floor structure could reduce the stiffness of the horizontal diaphragm with 35 to 49 % (Falk & Itani, 1989). The
effect of blocking was also shown by experimental testing. Research was done on the effectiveness of measures
(A) (B)
page 21 / 49
to increase diaphragm stiffness by several structural choices. These were the application of blocking, the
application of foam adhesive, the effect of openings in the sheeting (staircases), the influence of walls along the
edge of the diaphragm, and the sheeting nail density. A limiting factor however, is the choice to study all issues
separate. No combination of effects was studied. In conclusion can be said that the most effective measure was
to provide the diaphragm with blocking. It provided the largest individual increase in diaphragm stiffness with an
average of 135% increase in shear stiffness. Only qualitative results were published. These include the following
conclusions.
Using a smaller fastener distance, applying sheeting connected to each other with glued tongue-and-groove
finished edge, or gluing the perimeter of the sheeting to the floor joists, can increase the stiffness of the
diaphragm enormous.
Perimeter walls can effectively stiffen a diaphragm if the bottom plate is continuous and adequately fastened-
down. Due to the resistance of the bottom plate to the tension forces developing in the diaphragm edge through
bending action, the primary effect of walls on diaphragms is increased flexural stiffness. (Bott, 2005)

figure 26: Test set-up of a horizontal floor diaphragm. (Bott, 2005)

A few years ago an extensive research program (NEESWood) was conducted in the United States. In this project,
full-scale buildings were tested to verify their behaviour when supposed to seismic loading. From these tests a lot
of knowledge was gained. In addition, information about the stiffness of the floor-diaphragms was acquired.
Christovasilis, Filiatrault and Wanitkorkul reported about the research on a full-scale two-storey timber-frame
building in 2007. The building is shown in figure 27.

figure 27: View and lower-level plan of the two-storey timber-frame building. (Christovasilis et al., 2007)

Exposed to different magnitudes of seismic loads, the building became damaged. From observations and
interpretation of this damage and the measurements done during testing was concluded that the horizontal
diaphragms of the two main parts of the building acted as a rigid plane. However, because of the atrium and the
staircase, the in-plane stiffness of the floor and roof at this location was lower than the rest of the structure.
Although the application of gypsum sheeting to the ceiling increased significantly the in-plane stiffness of these
diaphragms. Damage was observed in the gypsum sheeting, in the atrium / staircase are, after testing. The
gypsum sheeting in the two main parts of the building had a minimal effect on the floor diaphragm since the
structural floor system alone provided enough in-plane stiffness to act as a rigid diaphragm. (Christovasilis et al.,
2007)

Garage
Atrium
Dining Living
Kitchen Nook Entry
Porch
page 22 / 49

figure 28: The floor diaphragm was made of plywood panels and floor joists connected by
nails. On the right hand side, the gypsum ceiling is visible. (Christovasilis et al., 2007)

During the tests, the deformations were recorded with multiple sensors. The shapes of deformation could be
composed from these measurements. These shapes show the rigid behaviour of the diaphragm in the main parts
of the building, and the flexible part in the atrium / staircase area.

figure 29: Fundamental mode shapes of the building, (a) transverse,
(b) & (c) longitudinal with torsional influences. (Christovasilis et al.,
2007)

Based on the findings from literature as presented above the following conclusion can be drawn. For the
dimensional proportions of the diaphragms treated above, the assumption of rigid diaphragm behaviour can be
justified. The assumption should be critically treated for more slender floor plan proportions, and floor
diaphragms with perforation such as atrium and staircases. For rigid diaphragms, modelling can consist of rigid
bracing, or plate elements. Flexible floor diaphragms can be modelled as if shear-wall diaphragms would be
modelled.

page 23 / 49
2.2 Fastener and hold-down level
In this paragraph shortly will be discussed what the code design rules on determination of the fastener
characteristics consist of. Moreover, the intended approach to determine the characteristics of hold-down
anchorage will be explained.
2.2.1 Fasteners
As will be presented in chapter 3, test-data of fasteners are result of the literature review. In a later phase of the
masters project, these data will be used to determine stiffness values according to NEN-EN 26891. The
determined stiffness values will then be compared with stiffness values derived from Eurocode 5. This process will
result in more insight into the applicability of the Eurocode 5 rules for the fastener slip-modulus (K
ser
) to timber-
frame shear-wall analysis.
Calculation of Kser according to Eurocode 5
In NEN-EN 1995 equations to determine the fastener-slip-modulus K
ser
are given for several fastener types.
Dowels, bolts, screws, nails, and staples are included in the design rules. The equations identify the timber mean
density
m
and fastener diameter d as parameters influencing the stiffness of the connection. Timber-to-timber
connections and timber-to-woodbased panel connections are aimed at. Regarding the steel-to-timber
connections, only the density of the timber member is involved in the equation, and K
ser
may be multiplied by 2.

In Eurocode 5 gypsum-to-timber connections are not mentioned. About this type of connection, some
information was found in the German code DIN 1052 (2008-12, tabelle G.1). There it is stated that the fastener-
slip modulus K
ser
has to be reduced with 40% for connections of timber to gypsum paperboard panels. In the
German code, the characteristic timber density
k
needs to be used instead of
m
.

An example for the equations given can be the formula to calculate the slip-modulus for nail-to-timber
connections with nail fasteners:

(7)
In which:

(8)

The applicability of the design rules on deriving K
ser
, to timber-frame shear-wall analysis, will be studied by
comparing the Eurocode derived stiffness values with the stiffness values determined using NEN-EN 26891 and
the test-data from literature.
Determination of Kser (ks) by experimental testing according to NEN-EN 26891
K
ser
can also be determined through experimental testing. In Eurocode 5 reference is made to NEN-EN 26891 for
calculation rules on how to do this. In the test standard, Kser is denoted as ks. The test standard provides uniform
rules for conditioning of the test specimens, loading procedure, measurement of the slip and load, format of the
test report, and calculation of the slip modulus. ks is in fact a modified stiffness value between 10% and 40% of
the estimated maximum load Fest of the connection. Furthermore, equations are given to calculate the stiffness at
higher load levels (60% and 80%).
Influence of moisture content on fastener stiffness Kser
The stiffness of the sheeting-to-timber joint is dependent on the moisture content. However, the magnitude of
influence is not known exactly. Analogous to the relation between the modulus of elasticity and the moisture
content, a change of 1,5% per 1% change in moisture content could be a reasonable assumption. (Kuilen, 2008)
Additionally, some experimental information can be found in publications of Nakajima. (Nakajima, 2001),
(Nakajima & Okabe, 2004) Tests on plywood shear-walls under conditions with relative humidity (RH) of 65% and
90% revealed a 21% decrease in initial stiffness of the wall panel in the wet (RH 90%) condition, compared with
the wall panel in the dry (RH 65%) condition.
page 24 / 49
Considering European spruce, the increase of the moisture content is about 12%, when changing the RH from
65% to 95% (Banga & Graaf, 2002). Assuming the plywood from Nakjimas tests to behave the same, this will
result in 2,1% change in initial stiffness per 1% change in moisture content.
According to NEN-EN 26891, 20C and 65%RH is the standard climate condition for testing when comparison of
joints under similar conditions is aimed at. In some of the test reports and papers, reference is made to these
circumstances as boundary condition for the shown load-displacement figures. From other sources of load-
displacement graphs no information about the relative humidity is known.
One might say that for shear-walls applied in Serviceclass 2 a reduction in stiffness has to be calculated with. And,
if the situation in timber-frame structures is analyzed, it becomes clear that the circumstances inside a cavity wall
(Serviceclass 2) are different from the atmosphere inside the building (Serviceclass 1). However, the differences
observed in testing at 90%RH and 65%RH influence the stiffness of the sheeting-to-timber connections much
more, than the smaller differences between the interior and cavity side of the shear-wall will do. The moisture
content of the timber on the inside of the building and inside the cavity wall will be always between 10% and
20%. Additionally the stiffness values of the timber based panel applied on the inner side of the building
(Serviceclass 1) would have to be increased compared with values determined from testing at 20C and 65%RH
atmosphere.
Therefore the conclusion can be drawn that both effects could counterbalance each other. Hence, no effect of
moisture content will be considered in the next phases of the project.
Conclusion
In conclusion can be said that the factors influencing the fastener stiffness K
ser
, of the fastener are the following:
Density of the timber (taken into account in Eurocode 5)
Diameter of the fastener (taken into account in Eurocode 5)
Type of sheeting material
Moisture content of the wood (not taken into account by Eurocode 5, will not be considered)
Thickness of the sheeting material
Although the thickness is important for the ultimate capacity (strength) of the connections, it is not taken into
account in Eurocode 5. Confirmation of this can also be seen in test-data. Compare curve 6.2 and 9.1 in figure
45, and curve 2.1 and 2.2 in figure 46.
Yield strength of the fastener (Plastic moment capacity)
Although the yield-moment capacity of the fastener is influencing the ultimate capacity (strength) of the
connections, the influence on the stiffness of timber-to-framing connection is negligible. (Leichti et al., 2006)

figure 30: Experimental test
set-up to derive fastener
characteristics. (Christovasilis
et al., 2007)
figure 31: Idealized load-deformation curve and
measurements.(NEN-EN 26891)

The influencing factors stated above will determine the fastener-slip characteristics and failure mechanism, as
long as enough edge distance is applied. If the distance to the edge is too short, other effects will become
page 25 / 49
important. For example, the way in which the edge is finished can be of influence on the characteristics of the
fastener (manufacturing issue). For gypsum-based panels this can be seen from the test-data found from
literature. Differences between sawn edges and edges finished with a paper cover become visible in the test
results when not enough edge distance is applied. Compare test results 7.3 and 7.4 in figure 47. The shortest edge
of a standard gypsum paperboard is the sawn edge. No finishing is applied on this edge. The longest edge of a
standard gypsum paperboard can be finished in different ways. The reinforcing paper will make this edge a little
bit stronger than the short edge.

figure 32: Sawn edge on the left, finished edge
on the right. (Knauf product information)

page 26 / 49
2.2.2 Hold-down anchors
For the determination of hold-down stiffness, no standard method is available. However, in this paragraph will be
presented which parts of Eurocode 5 on connections and fasteners will be used when a method to derive hold-
down stiffness will be explored in the later phases of the masters project. The method can be verified using the
hold-down test-data presented in chapter 3 and NEN-EN 26891.

The design rules to derive K
ser
according to Eurocode 5 are already discussed above. The fastener-slip modulus will
again be used to derive a stiffness value for the hold-down anchorage.
When the designer is calculating the design strength for hold-down anchorage, the calculation starts with the
design strength for a single fastener (
). A number of fasteners are used. When calculating the combined

action of all fasteners, the number of fasteners has to be reduced. The effective number of fasteners can be
calculated as following:

(9)

In which
is a reduction factor bringing in account the fastener spacing and the option of predrilled holes. The
capacity of the hold-down can now be calculated to be:

(10)

The reduction of the number of fasteners is introduced because of the splitting behaviour visible near failure of a
parallel to grain positioned and loaded row of fasteners. The accumulation of perpendicular to the grain tensile
stresses, gives reason to reduce the capacity of the connection (Jorissen, 1998). However, considering the
serviceability limit state, this reduction is no issue because it is the stress-level in ultimate limit state causing the
splitting of the timber grains. Therefore, the stiffness of the individual fastener does not have to be reduced when
contributing to the total stiffness of the hold-down anchor.

(11)

This is only valid when the fasteners act as a set of parallel springs, which is the fact when the deformation of the
steel section can be neglected. A comparison can be made with the stiffness value determined on basis of the
test-data. For the standard anchorage solutions, manufactured by Rotho Blaas and Simpson, test-data were
provided, and found in literature. Stiffness values for custom made configurations of hold-down anchors can be
estimated when it is possible to derive the stiffness of hold-down anchorage solutions in the way as it is described
above.

figure 33: Custom made tie-down
steel strips.
page 27 / 49
3 Experimental results
As was mentioned in 1.2, the test-data of timber-frame shear-walls, shear-walls with opening, sheeting-to-
timber fasteners, and hold-down anchorage, will be presented. How the information will be used is already
discussed in 1.2 and 1.3. The information will be the experimental basis to determine stiffness values for the wall
assemblies and connections (2.2 & 4.3). These stiffness values will be compared to the present code directives in
a later phase of the project. Validation of the modelling approach with the experimental information will be
another purpose of the test-data.

figure 34: Deformed timber-frame shear-wall after exceeding the ultimate
racking capacity in racking test. (NAHB Research Center Inc., 2005)
3.1 Full-scale shear-wall tests
From several sources, reports on experimental testing were found. In this section, tests on full-scale shear-walls
are presented. The information found in literature is studied and individual curves are combined into one graph
for wall widths 600, 1200, 1800, 2400, 3600 and 4800 mm. Comparison of one curve with another is made
possible by displaying all relevant information next to the graphs. If it is not stated differently, the standard
parameters are:

Dimensions of the wall (b x h) as displayed in the title of the graph.
Shear-wall sheeted on only one side of the shear-wall panel are considered to be standard. If both sides
of the shear-wall are sheeted, or when different fastener or sheeting lay-outs are applied on both sides it
will be explicitly mentioned.
Walls were tested fully anchored, using hold-down provisions or applying a vertical load.

From a wide range of shear-wall configurations test results were found, this can be seen in table 2. References
displayed in the table can be found on page 31 in the reference list.

Wall dimensions (b x h):
[mm]
600 x 2400 1200 x 2400 1800 x 2400 2400 x 2400 3600 x 2400 4800 x 2400
Fastener spacing (s):
[mm] 50 75 150 50 75 100 150 200 100 200 75 100 150 200 75 100 150 200 150 200
Board material:
OSB (3.1) (3.2) (1.2) (7.2) (5.2)
(3.3)
(5.1)
(7.1)
(3.4)
Plywood
(1.3)
(6.1)
(8.1)
(14.21) (2.3) (4.1)
Gypsum paper board (GKB)
(10.1)
(10.3)

(10.2)
(10.4)
(14.1)
(1.1)
(8.2)
(14.27) (14.2)
(2.1)
(2.2)
(14.9)
(14.3)
(14.10)
(4.2)
Gypsum fibre board (11.1)
(12.1)
(13.1)
(11.2) (11.3,5,7)
(12.2,3,4)
(13.2,3,4)
(11.4,6,8)
Particleboard
(14.15)
(14.19)

Hardboard fibre board (9.1) (9.2)
MDF (14.17)
table 2: Overview available test-data of shear-wall diaphragms as found in literature.
page 28 / 49
The information in the legend of the graphs can be read as following: Ply t = 9,5 nail 2,87 x 50, s = 100, b x h =
1820 x 2440, (1.3) means that the tested panel was made using plywood sheeting with thickness 12,0 mm. The
fastener type is stated, as well as the fastener spacing (s) in mm. Thereafter the dimensions of the shear-wall
panel are stated (b x h). For reference (1.3) is added, in which 1 refer to the source in literature: reference 1, and
3 means that this is the third test in the publication referred to. Reference to literature is made on page 31.

figure 35: Load displacement characteristics for timber-frame shear-wall panels 600 x 2400 (b x h).


0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
Load-displacement characteristics for timber-frame shear-wall panels 600 x 2400 (b x h)
OSB, s = 152, t = 11, nail 3,3x65, b x h = 600 x 2400, (3.1)
Gypsum Fiber Board, s = 50, both wall sides sheeted,
no hold-down, t = 12,5, staple 1,53 x 50,
b x h = 625 x 2600, (11.1)
Gypsum Fiber Board, s = 150, both sides sheeted,
bxh = 625 x 2600, (11.2)
Gypsum Fiber Board, s = 75, both sides sheeted, t = 15,
staple 1,53 x 50, b x h = 625 x 2600, (12.1)
nail 2,5 x 65, b x h = 625 x 2600, (13.1)
0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
OSB, s = 152, t = 11, nail 3,3 x 65, (3.2)
Hardboard Fiber Board, s = 100,
nail 2,1 x 50, t = 8, b x h = 1245 x 2310, (9.1)
GKB, s = 50, both sides sheeted, t=12,5,
nail 2,5 x 45, b x h = 1250 x 2500, (10.1)
GKB, s = 150, both sides sheeted, t = 12,5,
nail 2,5 x 45, b x h = 1250 x 2500, (10.2)
GKB, s = 50, both sides sheeted, t = 18,
nail 2,5 x 45, b x h = 1250 x 2500, (10.3)
GKB, s = 150, both sides sheeted, t =18,
nail nail 2,5 x 45, b x h = 1250 x 2500, (10.4)
GKB, s = 200, t = 13, nail 3,0 x 37,6, (14.1)
staple 1,53 x 50, h x b = 1250 x 2600, (12.2)
staple 1,53 x 50, h x b = 1250 x 3000, (12.3)
Gypsum Fiber Board, s = 75, t = 15,
staple 1,53 x 50, h x b = 1250 x 2600, (12.4)
nail 2,5 x 65, h x b = 1250 x 2600, (13.2)
nail 2,5 x 65, h x b = 1250 x 3000, (13.3)
Gypsum Fiber Board, s = 75, t = 15, nail 2,5 x 65,
h x b = 1250 x 2600, (13.4)
page 29 / 49



0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
Gypsum Fiber board, s = 50, both sides sheeted,
h x b = 1250 x 2600, (11.3)
h x b = 1250 x 2600, (11.4)
Gypsum Fiber board, s = 50, no hold-down,
t = 12,5, staple 1,53 x 50, h x b = 1250 x 2600, (11.5)
Gypsum Fiber board, s = 150, no hold-down,
t = 12,5, staple 1,53 x 50, h x b = 1250 x 2600, (11.6)
h x b = 1250 x 3000, (11.7)
h x b = 1250 x 3000, (11.8)
0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
GKB, s = 100, t = 12, nail 2,34 x 40,
b x h = 1820 x 2440, (1.1)
OSB, s = 100, t = 9,5, nail 2,87 x 50,
b x h = 1820 x 2440, (1.2)
Plywood, s = 100, t = 9,5, nail 2,87 x 50,
b x h = 1820 x 2440, (1.3)
Plywood, s = 100, t = 9, nail 2,87 x 50,8,
b x h = 1820 x 2440, (6.1)
Plywood, s = 100, t = 9,5, nail 2,87 x 50,
b x h = 1820 x 2440, (8.1)
GKB, s = 100, t = 12, nail 2,34 x 40,
b x h = 1820 x 2440, (8.2)
GKB, s = 200, t = 13, screw 3,0 x 37,6,
(14.27)
page 30 / 49



0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
OSB, s = 152, t = 11, nail 3,3 x 65, (3.3)
OSB, s = 140, t = 11,11, nail 3,3 x 63,5,
b x h = 2440 x 2450, (5.1)
OSB, s = 100, t = 11,11, nail 3,3 x 63,5,
b x h = 2440 x 2450, (5.2)
OSB, s = 150, t = 9,5, nail 2,5 x 50, (7.1)
OSB, s = 75, t = 9,5, nail 2,5 x 50, (7.2)
GKB, s = 200, t = 13, nail 3,0 x 37,6,
(14.2)
0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
GKB, s = 100, t = 13, screw 3,9 x 41, (2.1)
GKB, s = 100, both sides sheeted, t = 13,
screw 3,9 x 41, (2.2)
Plywood, s = 200, t = 9, screw 4,2 x 35, (2.3)
OSB, s = 152, t = 11, nail 3,3 x 65, (3.4)
Hardboard fiberboard, s = 100, t = 8, nail 2,1 x 50,
b x h = 3645 x 2310, (9.2)
GKB, s = 200, t = 13, nail 3,0 x 37,6, (14.3)
GKB, s = 150, t = 13, nail 2,45 x 34,5, (14.9)
GKB, s = 200, t = 13, screw 3,0 x 37,6,
finished / plastered seams between sheeting, (14.10)
Particleboard, s = 100, t = 12, nail 2,13 x 50,1, (14.15)
MDF, s = 75, t = 12, nail 1,70 x 39,8, (14.17)
Particleboard, s = 100, t = 12, nail 2,13 x 50,1,
glued edges (14.19)
Plywood, s = 150, t = 8, nail 2,13 x 50,1, (14.21)
page 31 / 49

References to literature
In the legend of the graphs, reference is made to the following numbers. These references can be found in the
back of the report on page 46.

(1) (Yasumura & Kawai, 1997)
(2) (Andreasson, 2000)
(3) (Salenikovich, 2000)
(4) (Ni et al., 2010)
(5) (NAHB Research Center Inc., 2005)
(6) (Yasumura & Karacabeyli, 2007)
(7) (He et al., 2001)
(8) (Yasumura, 1991)
(9) (Vessby et al., 2008)
(10) (LHT Labor fr Holztechnik - Fachbereich Bauingenieurswesen Hildesheim, 29-08-2002)
(11) (VHT Versuchsanstalt fur holz- und trockenbau Darmstadt, 16-02-2000)
(12) (VHT Versuchsanstalt fur holz- und trockenbau Darmstadt, 04-10-2001)
(13) (VHT Versuchsanstalt fur holz- und trockenbau Darmstad, 29-05-2002)
(14) (Kllsner, 1984)
(15) (Richard et al., 2002)

0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
Plywood, s = 150, t = 9,5, nail 2,5 x 63,5,
b x h = 4880 x 2440, (4.1)
GKB, s = 200, t = 12,7, nail 2,5 x 31,8,
b x h = 4880 x 2440, (4.2)
page 32 / 49
3.2 Full-scale shear-wall tests with opening
From three sources, reports on experimental testing of perforated shear-walls were found. In this section, the
results of these tests will be presented. Perforated shear-walls are shear-wall diaphragm, provided with openings
(perforation). These can be either windows, doors, or other type of openings. Because the range of ultimate
capacity and deformation capacity of the test series were different, the load-displacement graphs are presented
according to their source. In this way, comparison of the individual curves is easier. All relevant parameters are
presented in the legend, as well as the figures indicating the individual wall layout. Reference to literature is made
in the subtitles, and below the third graph.

figure 42: Load displacement characteristics for timber-frame shear-wall diaphragms with opening. (Kllsner, 1984)

figure 43: Load displacement characteristics for timber-frame shear-wall diaphragms with opening. (Dolan & Heine, 1997), (Dolan &
Johnson, 1996)

0
2
4
6
8
10
12
14
16
18
20
0 10 20 30 40 50
L
o
a
d
[
k
N
]
Displacement [mm]
Load-displacement characteristics for perforated timber-frame shear-walls)
GKB, t = 13, screw 3,0x 37,6s = 200, 2400x 2400,
gap = 1200 x 1200 (bxh), (14.4)
GKB, t = 13, screw 3,0x 37,6s = 200, 2400x 2400,
gap = 1200 x 1200 (bxh), (14.5)
GKB, t = 13, screw 3,0x 37,6s = 200, 3600x 2400,
gap = 2400 x 1200 (bxh), (14.6)
GKB, t = 13, screw 3,0x 37,6s = 200, 3600x 2400,
gap = 2400 x 1200 (bxh), (14.7)
GKB, t = 13, screw 3,0x 37,6s = 200, 3600x 2400,
gap = 1200 x 1200 (bxh), (14.8)
GKB, t = 13, screw 3,0x 37,6s = 200, 2400x 2400, (14.11),
gap = 1200 x 1200 (bxh), finished / plastered sheeting seams
GKB, t = 13, screw 3,0x 37,6s = 200, 3600x 2400, (14.12),
gap = 1200 x 1200 (bxh), finished / plastered sheeting seams
Particleboard, t = 12, nail 2,13x 50,1s = 100, 3600x 2400,
gap = 1200 x 1200 (bxh), gluededge, (14.20)
Particleboard, t = 12, nail 2,13x 50,1s = 100, 3600x 2400,
gap = 1200 x 1200 (bxh), (14.16)
MDF, t = 12, nail 1,70x 39,8 s = 75, 3600 x 2400,
gap = 1200 x 1200 (bxh), (14.18)
0
20
40
60
80
100
120
140
160
180
0 20 40 60 80 100 120 140 160
L
o
a
d
[
k
N
]
Displacement [mm]
Plywood, t = 12, nail 3,3x63,5s = 152,
+ GKB, t = 13, nail 2,41x38,1, s = 178,
12200x2440(bxh), (16.1)
OSB, t = 11,1, nail 3,3x63,5 s = 152,
+ GKB, t = 13, nail 2,41x38,1, s = 178,
12200x2440(bxh), (16.2)
Plywood, t = 12, nail 3,3x63,5s = 152,
+ GKB, t = 13, nail 2,41x38,1, s = 178,
12200x2440(bxh), door: 1220x2032, door:
2400x1220, window:3658x2032, (16.3)
OSB, t = 11,1, nail 3,3x63,5 s = 152,
+ GKB, t = 13, nail 2,41x38,1, s = 178,
12200x2440(bxh), door: 1220x2032, door:
2400x1220, window:3658x2032, (16.4)
Plywood, t = 12, nail 3,3x63,5s = 152,
+ GKB, t = 13, nail 2,41x38,1, s = 178,
12200x2440(bxh), gap: 8540x2440, (16.5)
OSB, t = 11,1, nail 3,3x63,5 s = 152,
+ GKB, t = 13, nail 2,41x38,1, s = 178,
12200x2440(bxh), gap: 8540x2440, (16.6)
page 33 / 49

figure 44: Load displacement characteristics for timber-frame shear-wall diaphragms with opening. (Richard et al., 2002)

References to literature
(14) (Kllsner, 1984)
(16) (Dolan & Johnson, 1996) & (Dolan & Heine, 1997)

0
5
10
15
20
25
30
0 20 40 60 80 100 120 140 160
L
o
a
d

[
k
N
]
Displacement [mm]
OSB, t = 9,5, nail 3,3x63,5 s = 150, 2440x2440,
gap = 1200x2000 (bxh), (15.1)
OSB, t = 9,5, nail 3,3x63,5 s = 50, 2440 x 2440,
gap = 1200 x 2000 (bxh), (15.5)
OSB, t = 9,5, nail 3,8x76 (with washer) s = 150, 2440x2440,
gap = 1200 x 2000 (bxh), (15.7)
page 34 / 49
3.3 Sheeting-to-timber fastener tests
From several sources, reports on experimental testing of fastener connections were found. These tests are often
used to derive stiffness values to be implemented in structural analysis and modelling. In this section, the test-
data on sheeting-to-timber fasteners are presented. The information found in literature is studied and individual
curves are combined into graphs for the different sheeting materials. Comparison of one curve with another is
made possible by displaying all relevant information next to the graphs. All curves display the behaviour for one
single fastener.

From table 3 can be concluded that from a wide range of fastener and sheeting combinations, test results were
found. Reference displayed in the table can be found on page 34 in the reference list.

Staple Nail Screw
Board material
B
o
a
r
d

t
h
i
c
k
n
e
s
s

[
m
m
]

1
,
5
0

(
1
,
5
3
)

x

5
0

1
,
8

x

6
5

1
,
3
4

/

1
,
6
2

x

4
7

1
,
4
/

1
,
6

x

6
0

1
,
8
3

/

2
,
0
6

x

5
0

2
,
3
4

x

4
0

2
,
2

x

4
5

2
,
9

x

4
8

2
,
0

(
2
,
1
)

x

5
0

2
,
5

x

5
0

2
,
8
7

x

5
0

3
,
0

x

5
1

2
,
5

x

5
5

2
,
5

x

6
5

(
6
3
,
5
)

2
,
8
7

x

6
5

(
6
3
,
5
)

3
,
3

(
3
,
1
)

(
3
,
0
)

x

6
5

(
6
3
,
5
)

2
,
8

x

7
5

3
,
8

x

7
6

3
,
1

x

8
0

3
,
3
5

x

4
5

3
,
9

x

4
1

4
,
2

x

3
5

OSB 9,5

(11.1) (1.2) (6.1) (6.2) (6.3)

11

(5.1)
(5.2)
(9) (3.6)
(3.4)
(3.5)

Plywood 9

(2.1) (8)
9,5

(1.3) (4)
12

(2.2)
Gypsum paper board (GKB) 12

(1.1)

12,5
(13)

(7.1)
(7.2)

(7.3)
(7.4)
(8.1)
(8.2)

Gypsum fibre board 15
(12.1) (13.1) (3.1) (14) (13.2) (12.2)
Particle board 10

(15) (3.3)
Cement splinter board 16

(3.2)
Hardboard 8

(10.1)
table 3: Overview of available test-data for sheeting-to-timber fasteners.

References to literature:
(1) (Yasumura & Kawai, 1997)
(2) (Komatsu et al., 1999)
(3) (Dujic & Zarnic, 2003)
(4) (Karacabeyli & Ceccotti, 1996)
(5) (Christovasilis et al., 2007)
(7) (Bouw- en woningtoezicht Rotterdam, 1989)
(8) (Andreasson, 2000)
(9) (Salenikovich, 2000)
(10) (Vessby et al., 2008)
(11) (He et al., 2001)
(14) (Conte et al., 2011)
(15) (Kllsner, 1984)
page 35 / 49

figure 45: Load-displacement characteristics for fastener tests with OSB sheeting.

figure 46: Load-displacement characteristics for fastener tests with Plywood sheeting.

0
200
400
600
800
1000
1200
1400
1600
1800
0 5 10 15 20 25
L
o
a
d

[
N
]
Displacement [mm]
Load-displacement characteristics for sheeting-to-timber fastener test (OSB sheeting)
t = 9,5 mm nail 2,87x50 (1.2)
t = 11,0 mm nail 3,1x80 (3.4)
t = 11,0 mm spiral nail 3,1x80 (3.5)
t = 11,0 mm annular nail 2,8x75 (3.6)
t = 11,1 mm nail 2,87x63,5 framing 51x102 mm (5.1)
t = 11,1 mm nail 2,87x63,5 framing 51x152 mm (5.2)
t = 9,5 mm nail 3,0x51 (6.1)
t = 9,5 mm nail 3,3x63,5 (6.2)
t = 9,5 mm nail withwasher3,8x76 (6.3)
t = 11,0 mm nail 3,3x65 (// to stud grain) (a4 = 51 mm)
(9.1)
t = 11,0 mm nail 3,3x65 (to stud grain) (a4 = 19 mm)
(9.2)
t = 11,0 mm nail 3,3x65 (to stud grain) (a4 = 10 mm)
(9.3)
t = 9,5 mm nail 2,5x50 (11.1)
0
200
400
600
800
1000
1200
1400
1600
1800
0 5 10 15 20 25
L
o
a
d

[
N
]
Displacement [mm]
Load-displacement characteristics for sheeting-to-timber fastener test (Plywood sheeting)
t = 9,5 mm nail 2,87x50 (1.3)
t = 9,0 mm nail 2,9x48 (2.1)
t = 12,0 mm nail 2,9x48 (2.2)
t = 9,5 mm nail 3,1x63,5 (4)
t = 9,0 mm screw4,2x35 (// to edge) (8.3)
t = 9,0 mm screw4,2x35 (to edge) (8.4)
t = 9,0 mm screw4,2x35 (high edge distances) (8.5)
page 36 / 49

figure 47: Load-displacement characteristics for fastener tests with Gypsum based sheeting.

figure 48: Load-displacement characteristics for fastener tests with different types of sheeting.

0
200
400
600
800
1000
1200
1400
1600
1800
0 5 10 15 20 25
L
o
a
d

[
N
]
Displacement [mm]
Load-displacement characteristics for sheeting-to-timber fastener test (Gypsum (fiber) board sheeting)
t = 12,0 mm nail 2,34x40 (1.1)
t = 15,0 mm (gypsumfiber board) staple 1,34x1,62x47(3.1)
t = 12,5 mm nail 2,2x45(sawn edge) (7.1)
t = 12,5 mm nail 2,2x45(7.2)
t = 12,5 mm screw 3,35x45 (sawn edge) (7.3)
t = 12,5 mm screw 3,35x45 (7.4)
t = 13,0 mm screw 3,9x41(// to edge) (8.1)
t = 13,0 mm screw 3,9x41(to edge) (8.2)
t = 15,0 mm (gypsumfiber board)
staple 1,5x50 (12.1)
nail 2,5x65(rillenagel) (12.2)
staple 1,8x65 (13.1)
nail 2,5x55(smooth) (13.2)
t = 12,5,0 mm(gypsum fiber board)
staple 1,4 x 1,6x 60(Fermacell) (14)
0
200
400
600
800
1000
1200
1400
1600
1800
0 5 10 15 20 25
L
o
a
d

[
N
]
Displacement [mm]
Load-displacement characteristics for sheeting-to-timber fastener test (different type of board materials)
Particle Board t = 10,0 mm
nail 2,5x50(3.3)
Particle board t = 12,0 mm
nail 2,0x50(15)
Hardboard t = 8,0mm
nail 2,1x50(10.1)
Cement splinter boardt = 16,0mm
staple 1,83x2,0x50 (3.2)
page 37 / 49
3.4 Hold-down anchorage tests
From two sources, reports on experimental testing of hold-down connections were found. The tests, which were
(on request) provided by Rotho Blaas GmbH, were originally performed to receive European Technical Approval
certificate for the hold-down solutions manufactured by Rotho Blaas. Another purpose is to receive characteristic
values for the ultimate load capacity of the tie-down anchors. Although the test-data can be used to determine
stiffness values for the hold-down, the stiffness aspects are not mentioned in product information. Another well-
known manufacturer of hold-down solutions is Simpson Strong-Tie. Load-displacement information on Simpsons
HTT16 hold-down was found in literature.

The meaning of the information on the hold-down stiffness aspect, can be explained as following. Often hold-
down anchorage is prescribed by engineers based on a ULS approach. These hold-downs can be bought from
stock, or can be designed individually for each single project as metal straps with certain dimensions, fastener,
and fastener spacing. The hold-down solutions can be very stiff, as can be seen from the curves. When the
deformations of the hold-down contribute to the deflections of the shear-wall panel as a whole, there can be
reason to include the hold-down stiffness in the modelling. In such a situation, Eurocode 5 can be used to
calculate a stiffness value for the hold-down anchor. (As explained in 2.2.2.) However, verification of this
approach is only possible by comparison with the stiffness values determined from the load-displacement graphs.

figure 49: Hold-down test-data from testing, mean curves. figure 50: Impression of
situation after failure.
(NAHB Research Center,
2003)

figure 51: Hold-down test-data from testing,
nail 4x40.
nail 4x75.

0
20
40
60
80
100
0 2 4 6 8 10 12 14 16 18
L
o
a
d

[
k
N
]
Displacement [mm]
Experimental results on hold-down testing
18 x nail 4,11x 89, (smooth shank), (2)
mean of (1.1) (1.5):
40 x nail 4,0 x 40 (rillenagel), = 421, (1)
mean of (1.6) (1.10):
mean of (1.11) (1.15):
26 x screw 5,0x 50, = 426, (1)
0
20
40
60
80
100
0 2 4 6 8 10 12 14 16 18
L
o
a
d
[
k
N
]
Displacement [mm]
50 x nail 4,0 x 40(rillenagel), = 421, (1.1)
mean of (1.1) (1.5):
mean of (1.6) (1.10):
mean of (1.11) (1.15):
26 x screw 5,0x 50, = 426, (1)
0
20
40
60
80
100
0 2 4 6 8 10 12 14 16 18
L
o
a
d
[
k
N
]
Displacement [mm]
mean of (1.1) (1.5):
mean of (1.6) (1.10):
mean of (1.11) (1.15):
26 x screw 5,0x 50, = 426, (1)
page 38 / 49


(1) Rotho Blaas Rothofixing WHT 80/620 hold-down
(Rotho Blaas GmbH Rothofixing, 07-04-2010)
(2) Simpson Strong-Tie HTT16 hold-down (NAHB
Research Center, 2003)

screw 5x50.

0
20
40
60
80
100
0 2 4 6 8 10 12 14 16 18
L
o
a
d
[
k
N
]
Displacement [mm]
30 x screw 5,0 x 50, = 423, (1.11)
27 x screw 5,0 x 50, = 387, (1.12)
25 x screw 5,0 x 50, = 411, (1.13)
25 x screw 5,0 x 50, = 418, (1.14)
22 x screw 5,0 x 50, = 434, (1.15)
mean of (1.1) (1.5):
mean of (1.6) (1.10):
mean of (1.11) (1.15):
26 x screw 5,0x 50, = 426
page 39 / 49
3.5 Compression perpendicular to grain tests
In the literature review, also information came across about compression perpendicular to the grain. In standard
walls, the vertical studs distribute the vertical forces to the foundation through the horizontal rails. Subsequently,
a compressive stress perpendicular to grain will develop in the horizontal rail, resulting in additional horizontal
deflections. The information that was found is displayed below, but will not studied further. In common Dutch
practice deformations due to compression perpendicular to grain are prevented by increasing the cross section of
the compressive stud. The contact area will increase, therefore the stress perpendicular to grain will decrease and
deformation is prevented.

figure 54: When vertical forces are becoming
higher, multiple vertical studs can be used, to
reduce compression perpendicular to grain.

figure 55: Load-displacement test results from several compression to grain tests. figure 56: Compression to grain
deformation. (NAHB Research Center, 2003)

(1) (Andreasson, 2000) (also information available concerning tension & shear in framing member connections)
(2) (Vessby et al., 2008) (also information available concerning tension & shear in framing member connections)

F
F F
0
10
20
30
40
50
60
0 2 4 6 8
L
o
a
d

[
k
N
]
Displacement [mm]
Load-displacement test results from several compression to grain tests
45 x 90 mm
(ungraded framing
members) (1.1)
45 x 90 mm
(ungraded framing
members) (1.2)
45 x 120 mm (2.1)
45 x 90 mm (secant)
(1.3)
page 40 / 49
4 Issues concerning the modelling approach and code directives
In this part of the report, four topics on racking strength and stiffness aspects will be discussed. First, the
requirements to a modelling suggestion will be explained. As stated in chapter 1, the modelling suggestion shall
be treated more in detail, in a later phase of the project. Secondly, a modelling approach found in literature will
be presented. Next, the analytical method according to Eurocode 5, for verification of strengths in SLS and ULS,
will be treated. The final issue that will be discussed is the test method NEN-EN 594, about the experimental
determination of racking strength and stiffness of timber-frame shear-walls.
4.1 Modelling aspects
Possibilities for modelling
A suggestion for a modelling approach to support the structural analysis, should fit to the possibilities of the
software available within the company of Boorsma Consultants. As described in chapter 1, the goal of the project
is to derive or determine the required linear-elastic stiffness properties, which are needed to model the timber-
frame structure in a common used structural analysis framework program (like Technosoft Raamwerken). No
plate or shell elements will be used, but cross bracing will contain the equivalent stiffness parameter to represent
plane elements in 2D.

Source for the stiffness values, which will be used as input in the framework modelling, can be either, the
Eurocode 5 design rules, test-data, a combination of both, or could be derived using more detailed modelling. To
perform more detailed modelling, a software package will be used, more advanced than the common framework
structural analysis software. This will be a commercially available FEM structural analysis program with user-
friendly interface (Scia Engineer). No academic methods will be used because in these methods the possibilities
are too extensive (Ansys / Diana). This choice will limit the possibilities of the FE method to the most used
options. However, 3D analysis, plate and shell elements, and non-linear spring elements are among the
possibilities.

figure 57: Example of intended thesis result: modelling a timber-frame structure using experimental input. Figure composed from
several sources: Rothofixing WHT hold-down device picture &graph (Rothoblaas GmbH), photo test set-up (NAHB Research Center Inc.,
2005), graph test result (Salenikovich, 2000).
Modelling approach as found in literature
As is made clear in the paragraph above, and shown in figure 57, an idea what the modelling approach will consist
of, already exists. The feasibility of this type of approach was shown when studying literature. Different
approaches, but also similar ways of modelling were found.
hd

page 41 / 49
Pang and Rosowsky developed a modelling approach to take part in a benchmark study concerning the
NEESWood project. (Some other findings from this project are discussed in 2.1.8) In the picture below the
physical existing shear-wall and floor diaphragms are drawn in grey colour. The representation of the shear-walls
in longitudinal direction (x-axis) is omitted for clarity. As can be seen, the diaphragms are represented by spring
elements, which are connected rigidly to, master and slave, nodes. The springs contain all the stiffness
information needed to get a sufficient accurate model. In fact, the model only consists of a set of equations and
programming, because it is a mathematical numerical model.

figure 58: Numerical modelling approach. (Pang & Rosowsky, 2010)

Pang and Rosowsky applied the principle of sub-modelling to get correct stiffness values as input for the spring
elements. Therefore, more detailed modelling was done on shear-wall level:

figure 59: Detailed FEM modelling of a timber-frame shear-wall. (Pang & Rosowsky, 2010)

The horizontal floor diaphragm in figure 58 was modelled using the so-called two-node-beam element. Therefore,
first an analysis was performed using a detailed FEM approach. The information on stiffness of the diaphragms
could thereafter be implemented in the two-node-beam element approach. See figure 60, and figure 61.

page 42 / 49

figure 60: Deformed shape of floor diaphragm, calculated using detailed FEM approach.
(Pang & Rosowsky, 2010)

figure 61: Two-node-beam element modelling approach, beam stiffness derived from more
detailed FEM modelling. (Pang & Rosowsky, 2010)

Using different methods of modelling for different levels of scale is a usual approach. Dujic and Zarnic followed
this modelling procedure as well. In a detailed FEM model, the fasteners and hold-down anchorage were
modelled using non-linear springs. The framing elements were represented by the normal dimensions, and were
connected with plastic hinges. The sheeting and glulam beams were modelled using linear elastic panel elements.

figure 62: From a complex and detailed FEM model (L) to the equivalent strut model (R). (Dujic
& Zarnic, 2002)

Using this type of modelling also perforated shear-walls can be analysed. The results of the detailed modelling can
thereafter be used in the less complex models using equivalent bracing.

page 43 / 49

figure 63: Details of the FEM model. (Dujic & Zarnic, 2002)

The simplified bracing approach can be used to construct a larger 3D model of the whole building.

figure 64: 3D equivalent strut model. (Dujic & Zarnic, 2002)

How the modelling aspect of the masters thesis will exactly be dealt with will be determined in a later phase of
the project. The information stated in this paragraph only presented the findings from the literature research.

page 44 / 49
4.2 Eurocode 5 methods for strength verification
In Eurocode 5 (NEN-EN 1995) can be chosen from two analytical methods to calculate shear-wall diaphragms. The
methods are denoted as Method A and Method B. In this paragraph, Method A will be highlighted because this is
the recommended method. The Eurocode approach can be used to derive the design value for the strength of a
wall loaded in plane (
). The method is based on the design strength of a single fastener (
) . This design
strength can be calculated using the Johansens equations which are given in the code. The strength of a single
fastener can be used to calculate the design strength of a single wall panel (
), taking into account the width

(
) of the panel, and fastener spacing () and panel ratio. (For a drawing of a standard shear-wall panel, see
figure 8 on page 10.)

(12)

(13)

In
, the height-to-width ratio is included. The fastener strength may be increased with a factor 1,2 because
of system-action. A reduction of the weakest side has to be calculated with when different sheeting is applied on
both sides of the wall panel, this reduction is 50% or 75% (2.1.6). Perforated panels with windows or door
openings do not contribute to the strength of the shear-wall assembly (2.1.7). Additionally is stated that the
external tensile and compression forces on the edges of the panel need to be taken up by the supporting
structure. Sufficient hold-down capacity has to be provided using tie-down anchorage or vertical load. Vertical
load however is not considered in the Eurocode method. No rules or guidelines are given on stiffness aspects of
wall and floor diaphragms. In practice, the limit state approach will be limited to strength considerations only.
This is one of the reasons to perform this masters study (See 1.1).
4.3 NEN-EN 594 Racking test method
The racking strength and stiffness of timber-frame wall panels can also be determined with experimental testing.
Different monotonic and cyclic testing standards are available around the world. These tests can be different not
only because of the boundary conditions (fixing the panel to the test rig), but also the procedure of applying the
load is often different. Although the tests may be performed according to different standards, for purpose of
calculating stiffness values from the test-data (3.1 & 3.2) use will be made of NEN-EN 594.

Experimental tests of timber-frame shear-wall panels are often assigned to research institutes by manufacturers
of sheeting materials such as: Oriented Strand Board, Plywood, Particleboard, Medium Density Fibreboard,
Gypsum Paper Board (GKB) or Gypsum Fibre Board. The manufacturers use the tests to gain European Technical
Approval (ETA) to sell their products for structural purposes on the European market. Strength values taken up in
the product information are often based on these ETAs. Although the tests are done to acquire characteristic
strength values in most cases, the load-displacement graphs can also be used to derive stiffness values. This
information however, is almost never given in the product information. From literature and the archive of
Boorsma Consultants, load-displacement graphs were found for a number of shear-wall configurations. Using
NEN-EN 594, stiffness values can be determined from this test-data.

According to NEN-EN 594 the racking stiffness of the panel can be calculated from the following equation:

(14)

Where: F2 is the racking load of 0,2 Fmax
F4 is the racking load of 0,4 Fmax
v2 and v4 is the deformation on these load levels

page 45 / 49
Fmax is reached when either the panel collapses, or the panel attains a deformation v of 100 mm, whichever
occurs first.

figure 65: Test procedure. (NEN-EN 594)

page 46 / 49
5 References
Andreasson, S., 2000. ISSN 0349-4969 Three-dimensional interaction in stabilisation of multi-storey timber frame
building systems. Licentiate Thesis. Lund (Sweden): Lund University division of structural engineering. TVBK-1017.
Arkana, 2011. Arkana houtskeletbouw - Lage-energie-woningen. http://www.bouwsite.be (07-09-2011).
Banga, J. & Graaf, P.d., 2002. Handboek Houtskeletbouw. 2nd ed. Rotterdam: SBR Stichting bouwresearch.
Bott, J.W., 2005. Horizontal stiffness of wood diaphragms. Virginia Polytechnic Institute and State University,
Masters thesis.
Bouw- en woningtoezicht Rotterdam, 1989. Richtlijnen stabiliteit houtskeletbouw d.m.v. gipskartonplaten.
Rotterdam.
Ceccotti, A. & Karacabeyli, E., 2002. Validation of seismic design parameters for wood-frame shearwall systems.
Canadian journal of civil engineering, 29, pp.484-98.
Christovasilis, I.P., Filiatrault, A. & Wanitkorkul, A., 2007. Seismic testing of a full scale two-story light-frame wood
building. Buffalo: University at Buffalo.
Conte, A., Piazza, M., Sartori, T. & Tomasi, R., 2011. Influence of sheathing to framing connections on mechanical
properties of wood framed shear walls. ANIDIS Bari 2011.
Dolan, J.D. & Heine, C.P., 1997. Monotonic tests of wood-frame shear walls with various openings and base
restraint configurations. Virginia Polytechnic Institute and State University, Department of Wood Science and
Forest Products.
Dolan, J.D. & Johnson, A.C., 1996. Monotonic tests of long shear walls with openings. Virginia Polytechnic Institute
and State University - Department of Wood Science and Forests Products.
Doudak, G. & Smith, I., 2009. Capacities of OSB-sheated light-frame shear-wall panels with or without
perforations. Journal of structural engineering ASCE, 135(3), pp.326-29.
Douglas, B.K. & Sugiyama, H., 2001. Perforated shearwall design approach. American Forest & Paper Association.
Dujic, B., Aicher, S. & Zarnic, R., 2006. Testing of wooden wall panels applying realistic boundary conditions.
Portland (Oregon): World conferenc on timber engineering (WCTE 2006).
Dujic, B., Klobcar, S. & Zarnic, R., 2006. Influence of openings on shear capacity of masive cross-laminated wooden
walls. International workshop on "Earthquake engineering on timber structures", Coimbra(Portugal), pp.105-18.
Dujic, B. & Zarnic, R., 2002. Influence of vertical load on lateral resistane of timber-frames walls. International
council for research and innovation in building and construction - Working commission W18 Timber Structures,
pp.CIB-W18/35-15-4.
Dujic, B. & Zarnic, R., 2003. Comparison of hysteresis responses of different sheating to framing joints., 2003.
Falk, R.H. & Itani, R.Y., 1989. Finite element modeling of wood diaphragms. Journal of Structural Engineering,
115(3), pp.543-59.
Foschi, R.O., 1974. Load-slip characteristics of nails. Wood Science, 7(1), pp.69-76.
page 47 / 49
Girhammar, U.A., Bovim, N.I. & Kllsner, B., 2004. Characteristics of sheathing-to-timber joints in wood shear
walls. World conference on timber engineering WCTE 2004 , 8, pp.165-70.
He, M., Lam, F. & Foschi, R.O., 2001. Modeling three-dimensional timber light-frame buildings. Journal of
structural engineering, 127(8), pp.901-13.
Jorissen, A., 1998. Double shear timber connections with dowel type fasteners. PhD-thesis Delft University of
Technology.
Kllsner, B., 1984. Panels as wind-bracing elements in timber-framed walls - Wood Technology Report Nr. 56.
Stockholm: TrteknikCentrum.
Kllsner, B. & Girhammar, U.A., 2009. Plastic models for analysis of fully anchored light-frame timber shear walls.
Engineering structures, 31(03), pp.2171-81.
Kllsner, B. & Lam, F., 1995. Diphragms and shear walls. In H.J. Bla, R. Grlacher & G. Steck, eds. STEP 3:
Holzbauwerke nach Eurocode 5: grundlagen, entwicklungen, ergnzungen. Dsseldorf: Fachverlag Holz. p.15/13.
Karacabeyli, E. & Ceccotti, A., 1996. Quasi-static reversed-cyclic testing of nailed joints. Bordeaux, 1996.
Kasal, B., Collins, M.S., Paevere, P. & Foliente, G.C., 2004. Design models of light frame wood buildings under
laterl loads. Journal of Structural Engineering ASCE, 130(8), pp.1263-71.
Kessel, M.H. & Kammer, T.z., 2005. Rumliches Tragverhalten mehrgeschossiger Gebude in Holztafelbauart.
Bauingenieur, (may).
Kessel, M.H. & Sandauw-Wietfeldt, M., 2004. Wood panels with thin low buckling resistance sheeting. World
conference on timber engineering WCTE 2004, 8, pp.643-47.
Komatsu, K., Hwang, K.H. & Itou, Y., 1999. Static cyclic lateral loading tests on nailed plywood shear walls. Graz,
1999.
Kuilen, J.W.G.v.d., 2008. Creep of timber joints. HERON, 53(8), pp.133-56.
Leichti, R., Anderson, E., Sutt, E. & Rosowsky, D., 2006. Sheathing nail bending-yield strength - Role in shearwall
performance. World Conference on Timber Engineering WCTE.
LHT Labor fr Holztechnik - Fachbereich Bauingenieurswesen Hildesheim, 29-08-2002. Bericht 01/P1002. 18180th
ed. Archive Boorsma projectnr. 05654 (Knauf): Industriegruppe Gipskartonplatten IGG im Bundesverband der
Gips- und Gipsbauplattenindustrie Darmstadt.
McCutcheon, W.J., 1985. Racking deformations in wood shear walls. Journal of structural engineering ASCE,
111(2), pp.257-69.
NAHB Research Center Inc., 2005. Full-scale tensile and shear wall performance testing of light-frame wall
assemblies sheathed with Windstorm OSB panels. Test report. Upper Marlboro: NAHB Research Center Inc.
Norbord Windstorm OSB.
NAHB Research Center, 2003. Light-frame wood shear walls with various overturning restraints. 2003rd ed. Upper
Marlboro: NAHB Research Center.
page 48 / 49
Nakajima, S., 2001. The effect of the moisture content on the performance of the shear walls. CIB-W18
International council for research and innovation in building and construction - working commission w18 - timber
structures, CIB-W18/34-15-2.
Nakajima, S. & Okabe, M., 2004. Effects of dry and humid cyclic climate on the performance of nail joints and
shear walls. WCTE 2004 World conference on timber engineering Lahti (Finland), pp.117-22.
Ni, C., Shim, K. & Karacabeyli, E., 2010. Performance of braced walls under various boundary conditions. Trentino
(Italy): World conference on timber engineering (WCTE 2010).
Ni, C., Shim, K. & Karacabeyli, E., 2010. Performance of braced walls under various boundary conditions. World
Conference on Timber Engineering WCTE 2010.
Pang, W.C. & Rosowsky, D.V., 2010. Beam-spring model for timber diaphragm and shear walls. Structures and
Buildings - Proceedings of the Institution of Civil Engineers, (163 (SB4)), pp.227-44.
Richard, N., Daudeville, L., Prion, H. & Lam, F., 2002. Timber shear walls with large openings - experimental and
numerical prediction of the structural behaviour. Canadian journal of civil engineering, 29, pp.713-24.
Rotho Blaas GmbH Rothofixing, 07-04-2010. Prfbericht Nr. 106115 Versuche zur Untersuchung des trag- und
verformungsverhaltens von verbindungen mit GH hold-down-verbindern Rothofixing WHT 80/620. KIT Karlsruher
Institut fr Technologie: Rotho Blaas GmbH.
Rothoblaas GmbH, n.d. Information on Rothoblaas WHT angle bracket hold-down device. Generously provided by
Ing. Fabio Verber from Rothoblaas GmbH.
Salenikovich, A.J., 2000. The racking performance of light-frame shear walls. Dissertation. Blacksburg: Faculty of
the Virginia Polytechnic Institute and State University.
Salenikovich, A.J. & Dolan, J.D., 2003. The racking performance of shear walls with various aspect ratios - part 1 -
monotonic tests of fully anchored walls. Forest products journal, 53(10), pp.65-73.
Sinha, A. & Gupta, R., 2009. Strain distribution in OSB and GWB in wood-frame shear walls. Journal of structural
engineering ASCE, 135(6), pp.666-75.
Songlai, C., Chengmou, F. & Jinglong, P., 2010. Experimental study on full-scale light frame wood house under
lateral load. Journal of structural engineering ASCE, 136(7), pp.805-12.
Vessby, J., Kllsner, B. & Olsson, A., 2008. Influence on initial gap between timber members on stiffness and
capacity of shear walls. Miyazaki: World conference on timber engineering (WCTE 2008).
VHT Versuchsanstalt fur holz- und trockenbau Darmstad, 10-06-2002. Prfung P-187/4-02/He. 150th ed. Archive
Boorsma Consultants projectnr. 05654: Firma Rigips GmbH.
VHT Versuchsanstalt fur holz- und trockenbau Darmstad, 29-05-2002. Prfbericht P-187/3-02/He. 150th ed.
Archive Boorsma Consultants projectnr. 05654: Firma Rigips GmbH Dsseldorf.
VHT Versuchsanstalt fur holz- und trockenbau Darmstadt, 04-10-2001. Prfbericht P-187/2-01/He. 150th ed.
page 49 / 49
VHT Versuchsanstalt fur holz- und trockenbau Darmstadt, 16-02-2000. Prfbericht P-186/1-00/He. 125th ed.
Yasumura, M., 1991. Seismic behavior of wood-framed shear walls CIB-W18/24-15-3. Oxford: CIB working
commission W18 - timber structures.
Yasumura, M. & Karacabeyli, E., 2007. International standard development for lateral load test method for shear
walls CIB-W18/40-15-5. Bled: CIB working commission w18 - timber structures.
Yasumura, M. & Kawai, N., 1997. Evaluation of wood framed shear walls subjected to lateral load. Vancouver,
1997.

Literatureandknowledgereview

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

Literatureandknowledgereview

Uploaded by

Copyright:

Available Formats

literature review for MSc thesis:

length of full height wall segments

is ratio of openings in wall element

is ratio of full wall segments

). A number of fasteners are used. When calculating the combined

). The method is based on the design strength of a single fastener (

), taking into account the width

You might also like