Construction and Building Materials 101 (2015) 622–627

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Mechanical properties of laminated strand lumber and hybrid cross-

laminated timber
Zhiqiang Wang a, Meng Gong b,⇑, Ying-Hei Chui b
a
College of Material Science and Engineering, Nanjing Forestry University, 159 Longpan Road, Nanjing, Jiangsu 210037, China
b
Wood Science and Technology Centre, University of New Brunswick, 1350 Regent Street, Fredericton, New Brunswick E3C 2G6, Canada

h i g h l i g h t s

Hybrid CLT was fabricated using lumber and/or LSL.

The bending properties of generic CLT were improved by using LSL as core or outer layers.

Planar shear failure in lumber core layer was the key failure mode of CLT with a lumber cross layer.

Tension failure in bottom layer was the key failure mode of CLT with a LSL core layer.

a r t i c l e i n f o a b s t r a c t

Article history: Hybrid cross laminated timber (HCLT) was fabricated using lumber and/or laminated strand lumber (LSL),
Received 15 March 2015 the mechanical performances of which were evaluated. To reach this goal, the mechanical properties of
Received in revised form 1 September 2015 LSL and the bending properties of CLT and HCLT were measured in this study. The properties of LSL mea-
Accepted 13 October 2015
sured included the tension strength (only in the major direction), shear strength, shear modulus, and
modulus of elasticity (MOE) and modulus of rupture (MOR). The failure mode of each kind of specimens
was visually examined and recorded. Four types of CLT panels, one generic CLT (used as control) and three
Keywords:
types HCLT were fabricated. The properties measured included the bending properties (in the major
Hybrid cross laminated timber
Bending properties
direction) and planar shear properties (in both major and minor directions). It was found that the
Planar shear properties HCLT had better bending and planar shear properties than that of generic CLT. The MOE and MOR of
Failure modes HCLT having LSL as the outer layers were 19% and 36% higher than those of generic one, respectively.
Laminated strand lumber The MOE and MOR of HCLT having LSL as core layer (replacing the cross lumber layer) were 13% and
24% higher than that of generic CLT, respectively. The failure modes of four types of CLT observed
included the planar shear failure of cross lumber layer, tension failure of bottom LSL, and tension failure
of bottom lumber, especially tension failure of lumber originated at a knot(s).
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Cross-laminated timber (CLT) is defined as a prefabricated solid
engineered wood product made of at least three orthogonally
bonded layers of solid sawn lumber or structural composite lumber
(SCL) using adhesive, nails or wooden dowels. CLT is an innovative
wood product that was introduced in the early 1990s in Austria
and Germany [1], which has been recently gaining increased pop-
ularity in residential and non-residential construction in form of
roof, floor, or wall applications in Europe and North America. Due
to its cross lamination, CLT shows good dimensional stability
⇑ Corresponding author.
E-mail addresses: wangzhiqiang@njfu.edu.cn (Z. Wang), meng.gong@unb.ca
(M. Gong), yhc@unb.ca (Y.-H. Chui).
compared to solid wood, and high in-plane stiffness and strength
when used as wall element. Another advantage of CLT is that it
can be prefabricated, reducing construction time and wastes. Com-
bined with other engineered wood products, such as I-beams, lam-
inated veneer lumber (LVL) and structural plywood, CLT
demonstrates great potential of serving as crucial elements in the
construction of buildings made entirely from timber. The multi-
storey buildings made of CLT exhibit their excellent resistance to
seismic tests [2]. Up to date, the tallest timber apartment building
in the world is the Forté apartment building in Melbourne, Aus-
tralia, which is 32.17 meters high. By using CLT, Forté reduces
CO2 equivalent emissions by more than 1400 tonnes when com-
pared to concrete and steel [3].
However, generic CLT is known to be prone to the so-called pla-
nar (rolling) shear failure and excessive deflection when subjected

http://dx.doi.org/10.1016/j.conbuildmat.2015.10.035
0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.
 Z. Wang et al. / Construction and Building Materials 101 (2015) 622–627 623

to out-of-plane loading. This is particularly critical where a lumber
layer doesn’t have edge gluing. To address this issue, many studies
have been conducted, including the measurement and prediction
of the planar shear properties of CLT, evaluation of the effect of pla-
nar shear on CLT, and improvement of the planar shear properties
of CLT [4–8]. Fellmoser and Blaß [4] investigated the planar shear
modulus of solid Norway spruce (Picea abies) board using a bend-
ing vibration test and the effect of span-to-depth ratio on planar
shear properties of a 3-layer CLT panel via shear analogy method.
They found that the planar shear modulus of 3-layer panels ranged
from 40 MPa to 80 MPa, and the planar shear strength was strongly
influenced by the span-to-depth ratio being below 20. Zhou et al.
[5] carried out the bending tests and two-plate shear tests for
examining the influence of growth ring orientation and laminate
thickness of cross layer on its planar shear properties of CLT spec-
imens, indicating that it should be feasible to adopt the two-plate
shear test for determining the planar shear modulus and strength
of cross layer in CLT. Chen [6] investigated the structural perfor-
mance of a box based CLT system in floor applications. She devel-
oped comprehensive three dimensional finite element models and
tested the bending properties of CLT board. She found that the
numerical analysis well agreed with the experimental data in
terms of vertical deflection and bending stiffness. Li [7] evaluated
the duration-of-load and size effects on the planar shear strength
of CLT, suggesting that the planar shear duration-of-load strength
adjustment factor for CLT was more severe than the general
duration-of load adjustment factor for lumber. Such a difference
should be considered in the introduction of CLT into the building
codes for engineered wood design. Wang et al. [8] studied the fea-
sibility of using fast-growing poplar as cross layer to fabricate
cross-laminated timber (CLT). Their study showed the mechanical
properties of CLT panels containing poplar were similar to those
made of non-poplar wood, such as Douglas fir.
These performance issues of planar shear can be addressed by
designing new type of CLT or directly replacing one or more of
the layers in a laminated strand lumber (LSL) panel(s). Researchers
at the Wood Science and Technology Centre (WSTC), the University
of New Brunswick, Fredericton, New Brunswick, Canada, have been
developing hybrid CLT by using laminated strand lumber (LSL). LSL
is a composite structural lumber consisting of oriented wood
strands that are glued and compressed to form panels up to
90 mm in thickness. LSL shows greater mechanical properties than
solid lumber of the same species as well as less variability [9]. LSL,
a revolutionary product, is used for a broad range of applications
including rim board, millwork and window, door and garage door
headers, as well as for many industrial uses. New uses for this pro-
duct are still evolving, including the use of LSL for vertical mem-
bers in commercial applications where the framing member
heights are long, and the wind loads are substantial. This study
was mainly aimed at evaluating the bending properties of HCLT
made of LSL and/or lumber. To reach this goal, the selected
mechanical properties of LSL were measured as well.
LSL was aspen poplar (Populus tremuloides). The MC of LSL at test
was about 7%. One-component polyurethane adhesive (Purbond
HB E452) was provided by Henkel Corporation.
2.2. Preparation of LSL specimens
The mechanical properties of LSL were directly measured in this
study. The test program is given in Table 1. These properties
involved the modulus of elasticity (MOE) and modulus of rupture
(MOR) under static bending, shear modulus, shear strength and
tension strength. The specimens of each type had their face grain
direction parallel and perpendicular to the major direction of LSL,
except the tension strength test due to the limitation of the panel
width, which was only tested in the major direction. The groups A,
C, F, I, and K had their face grain direction parallel to specimen
length direction, and the groups B, G, J, and L had their face grain
direction perpendicular to specimen length direction. 2 or 3 spec-
imens of each type were cut from one LSL panel and four LSL panels
were used, generating a total of 8 or 12 specimens for each group.
All test specimens were stored in a conditioning chamber of a tem-
perature of 20 °C and a relative humidity of 65% for more than
30 days before testing.
2.3. Fabrication of HCLT panels and preparation of bending and shear
HCLT specimens
Three types of HCLT combining lumber with LSL were prepared
in this study. The layups of 3-ply CLT and HCLT panels are given in
Table 2, which were fabricated in Alberta Innovates Technology
Future, Edmonton, Alberta, Canada. The HCLT had similar structure
with CLT, i.e. all panels had three glued layers of boards placed in
orthogonally alternating orientation, except the type LTL consist-
ing of lumber and LSL placed in the parallel to major/longitudinal
direction Table 2. Due to the dimensional limitation of LSL panel,
the centre layer of TTT contained three LSL pieces leading to two
gaps that were 0.6 m from two edges in the panel length direction.
After preparing each layer, the surface of each layer was first air-
blown to remove dust and debris and then was sprayed with
95 g of water for wetting surface. The reason of water spray was
that LSL panel had relatively lower MC that the required value (lar-
ger than 8%) for curing the adhesive [1]. It should be noted that the
lumber layers were not applied with edge gluing and moisting (due
to its MC being higher than 8%). The adhesive was only applied at a
rate of 220 g/m2 on one surface of each layer right after two min-
utes of water spray. The total assembly time varied from 15 to
35 min which was well controlled under the required maximum
assembly time (i.e. 45 min) of the adhesive. A 1.5 m by 2.7 m steam
press was used to apply the pressure. A 0.13-mm-thick Teflon
sheet was used to separate the panel and pressing plate. The one
or two panels were pressed without applying side pressure under

Table 1
Types of LSL specimens.
2. Methods
Specimen Testing Length Width Face grain No. of
type type (mm) (mm) direction* specimens
2.1. Materials
A Bending 848 89 // 12
B Bending 848 89 \ 12
2-by-4 Lodgepole pine (Pinus contorta) lumber used in this
C Tension 1829 140 // 8
study was provided by Weyerhaeuser, which had a grade of #2 F Bending 350 89 // 8
and better J-Grade with an average moisture content (MC) of 12% G Bending 350 89 \ 8
at test. The nominal dimensions of lumber were 38 mm in thick- I Bending 300 89 // 8
ness, 89 mm in width, and 2.44 m in length. The LSL panels of a J Bending 300 89 \ 8
K Bending 406 89 // 8
grade of 1.5E were provided by Weyerhaeuser as well, the dimen- L Bending 406 89 \ 8
sions of which were 2.44 m in length, 1220 mm in width and
*
38 mm in thickness. The wood species of strands used for making // and \ represent the parallel and perpendicular to the strength direction of LSL.
624 Z. Wang et al. / Construction and Building Materials 101 (2015) 622–627

Table 2
Panel configurations of 3-ply CLT and HCLT.

Panel type Panel layup (top–center–bottom) Layer orientation

(top–center–bottom)
LLL Lumber–Lumber–Lumber //-\-//
LTL Lumber–LSL–Lumber //-//-//
TLT LSL–Lumber–LSL //-\-//
TTT LSL–LSL–LSL //-\-//

the ambient temperature 16 °C for 2 h at a pressure of 1.3 MPa. The

selected manufacturing parameters of CLT panel, such as adhesive
coverage, assembly time, pressing pressure, pressing time and
pressing temperature, were based on the findings from WSTC.
Fig. 2. Failure of a typical LSL specimen under tension.
Three panels of each type were fabricated, giving a total of 12 3-
layer panels that had the dimensions of 114 mm in thickness,
1220 mm in width, and 2440 mm in length.
After fabricating the panels, the specimens of proper dimen-
sions were cut and shipped to WSTC. Four bending specimens
and four shear specimens were cut from each panel in the
parallel-to-the major direction, producing a total 12 specimens of
each type of HCLT panel.

2.4. Properties of materials and panels

Group C was used to determine the tension strength of LSL in its

major direction by the Metriguard Testing Machine (Model 401)
following ASTM D3500 [10]. Groups A, B, F, G, I, J, K and L were
used to determine the shear modulus (G) of LSL in both panel
directions under the variable span bending test with center-point
bending referring to ASTM D198 [11]. Groups A and B were used
to determine MOE and MOR of LSL in both directions under the
third-point bending method. Groups I and J were used to test the
shear strength of LSL under the center-point loading referring to
ASTM D198 [11].
Four span-depth ratios (L/h) were used in the variable span bend-
ing tests with center-point loading, 18.42 (groups A and B), 7.89
(groups K and L), 6.57 (groups F and G), and 5.26 (groups I and J).
A specimen was loaded using a universal testing machine at a
cross-head movement of 3 mm/min. The mid-span deflection (D)
was measured using a 25 mm linear variable differential trans-
former (LVDT) until the load reached a preset load value that was
lower than the elastic proportional limit that was estimated in a
separate study, 3 kN for group A, 0.5 kN for group B, 10 kN for
groups K and F, and 2 kN for groups L, G, and J. Em and Gapp can be
therefore calculated using Eq. (1), which is illustrated in Fig. 1 [11].
 2
1 1 1 h
¼ þ
Eapp Em KGapp L
Fig. 3. Failure of two LSL specimens in the major and minor directions under
bending.
where Eapp is the apparent modulus of elasticity that incorporates
the influence of shear deformation, Em is the true modulus of elas-
ticity, Gapp is shear modulus, h is the depth of cross section, L is the
span (i.e. the distance between two supports of a bending setup), K
is the shear coefficient. K is a reciprocal of so-called shape factor,
dependent on the cross-sectional dimensions of the beam. K is
assigned to be 0.84 for a rectangular section [11].
The bending properties of CLT/HCLT panel specimens were
tested under the third-point loading method as per ANSI/APA
ð1Þ PRG 320 [12]. The dimensions of a bending specimen were
2438 mm in length, 197 mm in width and 114 mm in thickness.
The bending span was 2286 mm and the span/depth ratio was
20. The load speed used was 10 kN/min. The mechanical properties
measured included MOR and MOE in the major direction. The shear
specimens were tested as well but will be reported in another
paper.

3. Results and discussion

3.1. Mechanical properties and failure modes of LSL

3.1.1. Tension and bending properties

Fig. 1. Determination of planar shear modulus (G) by means of the variable span
bending test.
The mean tension strength in the major direction of LSL speci-
mens (group C) was 45.11 MPa with coefficient of variation
(COV) of 4.08%. The simple tension failure near the middle of a
specimen length was observed during testing, Fig. 2. Under the
third-point bending tests, the failure modes of the LSL specimens
were found to be tension failure at the bottom near the loading
 Z. Wang et al. / Construction and Building Materials 101 (2015) 622–627 625

Fig. 4. Fitting result of GXY (GXZ) and GYZ of LSL specimens.

Fig. 5. Failure modes of LSL specimens under short span bending.

point showing a brittle fashion in both major and minor directions. Table 3
The load–deflection curves of two typical LSL specimens tested in MOE of variable span-depth ratio LSL specimens.
the major and minor directions are plotted in Fig. 3. The mean
Strength direction Specimen type Span–depth ratio Eapp (MPa)
MOE values of LSL specimens (groups A and B) tested at a span-
to-depth ratio of 18.42 were 11563 MPa with COV 3.92% and Major A 18.42 11,089
K 7.89 5,620
1301 MPa with COV 3.81%, respectively. The mean MOR of LSL F 6.57 5,823
specimens (groups A and B) were 61.77 MPa with COV 10.4% and I 5.26 4,473
8.86 MPa with COV 16.43%, respectively. Minor B 18.42 1,350
L 7.89 1,079
G 6.57 929
3.1.2. Shear properties and Em J 5.26 878

The MOE values, i.e. Eapp, of LSL specimens tested under the
variable span bending tests are listed in Table 3. From Table 3, it
can be found that the, Eapp decreased with decreasing the span-
to-depth ratio. This is attributed to the contribution of shear com- Table 4
Bending properties of CLT and HCLT.
ponent on deformation. Fig. 4 illustrates the calculation of Em and
Gapp using Eq. (1), which are 10,789 and 324 MPa in the major Panel type MOE (MPa) COV (%) MOR (MPa) COV (%)
direction, and 1373 and 96 MPa in the minor direction. It can be LLL 9,727 7.37 35.37 9.85
found that the Em of LSL in the major direction is similar to that LTL 10,975 6.19 43.95 9.94
of lumber, and Gapp of LSL is much higher than that of lumber that TLT 11,221 3.64 44.18 12.30
is reported to be around 50 MPa [1]. TTT 11,609 8.95 48.19 8.93

The mean shear strength values of LSL (groups I and J), which
were tested at the span-to-depth ratio of 5.26 according to ASTM
D198 [11], were 6.58 MPa with COV 6.12% and 1.43 MPa with 3.2. Bending properties of CLT and hybrid CLT
COV 15.05%, in the major and minor directions, respectively. It
was, in general, found during tests that the LSL specimens failed The bending properties of CLT and HCLT specimens are listed in
in shear in the major direction and in simple tension in the bottom Table 4. The MOE and MOR values of HCLT are increased after using
in the minor direction, respectively, shown in Fig. 5(a) and (b). LSL in HCLT. Group LLL (i.e. generic CLT) had the lowest MOE and
626 Z. Wang et al. / Construction and Building Materials 101 (2015) 622–627
Fig. 6. Failure modes of CLT and HCLT specimens tested in bending.
MOR values, which were about 19% and 36% lower than group TTT,
and 15% and 25% lower than that of group TLT, respectively. Such
an increase could be attributed to the use of LSL as the outer layer,
which had uniform mechanical properties without defects such as
knots. In addition, compared with the bending properties of group
LTL panels, which had LSL as the core layer and lumber as outer
layers, group LLL (i.e. generic CLT) had lower MOE and MOR values,
which was about 13% and 24% lower than the HCLT (type LTL).
Such an improvement could be attributed to the higher GLT (GLR)
of parallel LSL layer compared to the lower planar shear modulus
(GRT) of cross lumber layer. The planar shear modulus is depended
on many factors, such as species, cross-layer density, and sawing
pattern configurations (i.e. annual rings orientation). [1]. In the
CLT Handbook [1], the GRT of spruce-pine-fir (SPF), Douglas fir-
larch (D. Fir-L) and hemlock-fir (Hem-Fir) lumber, and MSR and
visually graded boards are assumed as 50 MPa, however, the GLT
(GLR) of LSL was 324 MPa measured in this study. This suggests that
use of LSL can improve the planar shear properties of CLT.
The failure modes differed among four groups. Groups LLL and
TLT failed in form of the planar (rolling) shear failure of the cross
layer lumber within the shear span range, Fig. 6(a) and (b). Due
to the wood anatomical features in the radial-tangential (RT) plane
in wood in comparison to the longitudinal-radial (LR) or
longitudinal-tangential (LT) plane, the planar shear modulus is
very low, leading to planar shear failure in the cross layer of lum-
ber in CLT in the shear span range under bending. Groups LTL and
TTT failed in simple tension in bottom layer of lumber or LSL, Fig. 6
(c) and (d). As for group LTL, failure happened in tension in the bot-
tom layer lumber, especially around a knot(s), Fig. 6(c).
4. Conclusions
The mechanical properties of LSL and bending properties of
generic CLT/HCLT were measured. The following conclusions were
drawn:
(1) The average tension strength of LSL in the major direction
was 45.11 MPa. The average shear modulus GLR (GLT) and
GRT of LSL were 324 MPa and 96 MPa, respectively. The aver-
age shear strength values of LSL in the major and minor
directions were 6.58 MPa and 1.43 MPa, respectively. The
average bending MOE and MOR of LSL in the major direction
were 11563 MPa and 61.7 MPa, respectively. The average
bending MOE and MOR of LSL in the minor direction were
1301 MPa and 8.86 MPa, respectively.
(2) The bending properties of generic CLT had been improved by
using LSL as core or outer layers. For example, the average
MOE and MOR of group TTT were 19% and 36% higher than
those of generic CLT (group LLL), respectively. The average
MOE and MOR of group LTL were 13% and 24% higher than
those of generic CLT (group LLL), respectively.
(3) Two typical failure modes were discovered in the CLT/HCLT,
i.e. planar shear failure in lumber core layer and simple ten-
sion failure in bottom of LSL/lumber.
Acknowledgements
The authors wish to acknowledge Jiangsu Government Scholar-
ship for Overseas Studies, the National Natural Science Foundation
of China (Grant No. 31570559), the Priority Academic Program
Development of Jiangsu Higher Education Institutions (PAPD),
and the Jiangsu Science and Technology Project (Grant No.
BY2014006-05). The authors would also like to thank Weyer-
haeuser and Henkel for testing materials and technical advices.
References
[1] E. Karacabeyli, B. Douglas, Cross-Laminated Timber Handbook, U.S ed.,
FPInnovations, Quebec, Canada, 2013.
[2] J.W.G.V.D. Kuilen, A. Ceccotti, Z. Xia, M. He, Very tall wooden buildings with
cross laminated timber, Procedia Eng. 14 (2011) 1621–1628.
[3] L. Lease [Internet], Melbourne, Australia: Forté, Available from:
http://makeitwood.org/made-from-wood/forte.cfm. Cited 2014 Nov 26.
[4] P. Fellmoser, H.J. Blaß, Influence of rolling shear modulus on strength and
stiffness of structural bonded timber elements, in: H.J. Larson, J. Munch-
Andersen (Eds), Paper 37-6-5, Proc CIB-W18 Meeting, 31 August to 3
September, 2004.
[5] Q.Y. Zhou, M. Gong, Y.H. Chui, M. Mohammad, Measurement of rolling shear
modulus and strength of cross-laminated timber using bending and two-plate
shear tests, Wood Fiber Sci. 46 (2) (2014) 259–269.
 Z. Wang et al. / Construction and Building Materials 101 (2015) 622–627
[6] Y. Chen, Structural performance of box based cross laminated timber system
used in floor applications (dissertation), The University of British Columbia,
Vancouver, Canada, 2011.
[7] Y. Li, Duration-of-load and size effects on the rolling shear strength of cross
laminated timber (dissertation), The University of British Columbia,
Vancouver, Canada, 2015.
[8] Z.Q. Wang, H.M. Fu, Y.H. Chui, M. Gong, Feasibility of using poplar as cross
layer to fabricate cross-laminated timber, in: Proceedings of the 13th World
Conference on Timber Engineering, August 10–14; Quebec, Canada, 2014,
paper 520.
627
[9] H.G.L. Prion, H. Li, W. Boehner, D.M. Moses, Composite behavior of laminated
strand lumber, Wood Sci. Technol. 37 (2003) 59–77.
[10] ASTM, Standard Test Methods for Structural Panel in Tension, Designation: D
3500-90 (Reapproved 2003), American Society for Testing and Materials
(ASTM), West Conshohocken, PA, USA, 2003.
[11] ASTM, Standard test methods of static tests of lumber in structural sizes,
Designation: D198-09, American Society for Testing and Materials (ASTM),
West Conshohocken, PA, USA, 2009.
[12] APA, Standard for performance-rated cross-laminated timber. ANSI/APA PRG
320–2012. Tacoma, WA, USA, APA – The Engineered Wood Association, 2012.