Jurnal

Full Paper
Teknologi
ASSESSMENT IN BENDING AND SHEAR Article history
Received
STRENGTH OF GLUED LAMINATED TIMBER 15 June 2015
Received in revised form
USING SELECTED MALAYSIAN TROPICAL 16 September 2015
HARDWOOD AS ALTERNATIVE TO TIMBER Accepted
17 December 2015
RAILWAY SLEEPERS
*Corresponding author
Norshariza Mohamad Bhkaria*, Zakiah Ahmadb, Afidah Abu [email protected]
Bakarc, Paridah Md. Tahird

aFaculty of Civil Engineering, Universiti Teknologi MARA,

Pahang, Malaysia
bInstitute Infrastructure Engineering and Sustainable
Management, Universiti Teknologi MARA. Malaysia
cFaculty of Civil Engineering, Universiti Teknologi MARA

Malaysia
dInstitute of Tropical Forestry and Forest Product (INTROP),

Putra Infoport, Universiti Putra Malaysia, Selangor, Malaysia

Graphical abstract Abstract

This paper presents a pilot study on bending and shear strength of glued laminated (glulam)
Manufacturing
timber using selected tropical timber namely, Kekatong (Cynometra spp.) and Melagangai
of glulam
(Potoxylon melagangai) as an alternative for timber railway sleepers. Selected timbers were
beams
manufactured in accordance with MS758:2001 and the bending test was conducted
according to ASTM D198:2013. The shear test for glue line integrity was performed to observe
the bond performance in glulam accordance to MS758:2001.The results showed both species
can be used as structural members since the bending strength obtained from the laboratory
Bending Test work is greater than the allowable bending strength. In terms of the percentage of wood
failure, the bonding characteristics of both glulam satisfied the bonding requirement
stipulated in the standard and have the potential to be used as glulam timber railway
sleepers.

Shear Test Keywords: Bending strength, glued laminated timber, tropical hardwood, shear strength,
timber railway sleepers

© 2016 Penerbit UTM Press. All rights reserved

1.0 INTRODUCTION lower percentage installed in railway tracks

nowadays is instigated from the limited resources of
Hardwood solid timber sleepers are widely used in the required grade hardwood timber (from SG1 and
railway industry as reported by [1]. There are more SG2). Moreover, depleting supply in large diameter
than 2.5 billion timber sleepers are installed timber logs limited the number of sleepers production
throughout the world. In Malaysia, the number of and cause a lot of wastages. Consequently, the
solid timber sleepers was recorded by local railway price for solid timber sleepers would increase. Timber
authority (Keretapi Tanah Melayu Berhad) as 26.4% sleepers are generally exposed to deterioration due
from the total track length of 1724.5 kilometres. This to deleterious environmental effects [2], [3]

78:5–5 (2016) 111–117 | www.jurnalteknologi.utm.my | eISSN 2180–3722 |

112 Norshariza et al. / Jurnal Teknologi (Sciences & Engineering) 78:5–5 (2016) 111–117

particularly for non-durable timber and not properly manufactured at a glulam factory in Johor Bahru.
treated which limited its structural strength. The timbers were visually graded in accordance with
Amongst the above shortcomings, timber sleepers [18] by certified grader. Only timbers in HS grade
are still in demand since, in certain location of railway were chosen and the production of glulam was done
track, the PSC sleepers are inappropriate to be used in accordance with [19]. Two numbers of each
such as at station, yard and industrial lines, bridges Kekatong and Melagangai glulam beams with size of
and the soil area contributing pumping to 100mm x 200mm x 4000mm length were prepared.
locomotives [4]. Therefore, there is a need to find an The production of glulam beams were taken from
normal and controlled manufacturing processes and
alternative to solid timbers sleepers and this has
the grain for all timber pieces are parallel to the
spurred the use of engineered timber products such
longitudinal axis of the sleeper. Phenol-resorcinol
as glued laminated timber or glulam.
formaldehyde (PRF) adhesive was used as
Glulam is a structural timber product recommended in [19]. All the test specimens were in
manufactured by gluing together individual pieces of dried condition.
dimensioned and strength graded timber under
controlled manufacturing conditions [5], [6]. Glulam
can be designed and engineered to meet the needs
of the construction industries. Presently, glulam has 3.0 EXPERIMENTAL PROGRAMME
been broadly used in building structures such as in
beams and roof trusses, and infrastructures for 3.1 Bending Test
example in bridge girder. Nevertheless, the used of
glulam as railway sleepers is still limited and focusing The bending test is conducted under third-point
to softwood [7]–[9] and low grade hardwood [10] loading method as shown in Figure 1. The test
which required chemical treatment. Other study was apparatus including roller support, reaction bearing
conducted for untreated hardwood sleepers from plates, load bearing blocks were set up according to
Brazilian timber (Eucalyptus Citriodora) presented the ASTM D198:2013 [20]. These supports allowed the
good performance in bending. However, the specimen to deflect without significant resistance
bondability of this timber is not discussed [11]. and prevented buckling. Transducers were attached
Thus, this study proposed natural durable timber in vertical positions to monitor the deflection data
from the Malaysian tropical hardwood species during the test.
namely Kekatong (Cynometra spp.) and Melagangai The bending strength, 𝑓𝑚,𝑎,// , modulus of elasticity,
(Potoxylon melagangai) from strength group SG2 and permissible stress, 𝑓𝑚,𝑎𝑑𝑚,// were calculated using
and SG3, respectively. These timber species are listed the Equation (1), Equation (2) and Equation (3)
in the technical document of KTMB Permanent Way respectively which 𝑃𝑚𝑎𝑥 is ultimate total load; 𝐿 span
Manual [12]. of beam; 𝑏 is width of beam; is depth of beam;𝑃 is
The bending and shear strength of glulam increment of applied load below proportional limit;
Kekatong and Melagangai are more depend on the is deflection; 𝑘1 duration of load factor; 𝑘6 is depth
manufacturing process of lamellas by removing or factor;𝑘11 is size factor; and 𝑘20 factor for service
dispersing the knots within layers particularly placing class. In this study, service class 3 was considered
it at the neutral plane[13] and the integrity of their since the glulam will be used as railway sleepers
glue lines. With appropriate lamella configuration, which fully exposed to the external environment.
the glulam timbers are more homogeneous and However, grade stresses can be obtained from the
stronger with reliable properties. As in [14], the preliminary study conducted before and not
strength of glulam beam has been proven to described in this study.
improve into two grades higher from the original
strength grade of medium hardwood timber. 𝑓𝑚,𝑎,// =
𝑃𝑚𝑎𝑥 𝐿
(1)
𝑏ℎ2 ℎℎℎ
However, the naturally durable heavy hardwood
timber, the strength of glulam timber is expected at 23𝑃𝐿3
least at par with the value of solid timber sleeper [15]. 𝐴𝑝𝑝𝑎𝑟𝑒𝑛𝑡 𝑚𝑜𝑑𝑢𝑙𝑢𝑠 𝑜𝑓 𝑒𝑙𝑎𝑠𝑡𝑖𝑐𝑖𝑡𝑦, 𝐸𝑓 = (2)
108𝑏ℎ3 ∆
Therefore, the study is performed to assess the
bending strength of glulam beams and the shear 𝑃𝑒𝑟𝑚𝑖𝑠𝑠𝑖𝑏𝑙𝑒 𝑠𝑡𝑟𝑒𝑠𝑠
strength of glue lines for Kekatong and Melagangai. = ((𝐺𝑟𝑎𝑑𝑒 𝑠𝑡𝑟𝑒𝑠𝑠𝑒𝑠 × 𝑘20 ) × 𝑘1 × 𝑘6 × 𝑘11 ) (3)
The strength and integrity of these glue lines are
dominant for the reliability of whole pieces of
sleepers.

2.0 MATERIAL
Kekatong and Melagangai species were selected as
glulam beam in this study. These timbers are naturally
durable [16], untreated and suitable for outdoor
environment conditions [17]. The glulam beams were
113 Norshariza et al. / Jurnal Teknologi (Sciences & Engineering) 78:5–5 (2016) 111–117

3.2.1 Testing Method

Prior to testing, the specimens were keeping in

laboratory environment for climatization in order to
have equilibrium moisture content in a range of 10%-
14%. The specimens are placed into shear test
apparatus with the glue line oriented parallel to the
loading direction. The quasi-static tests were
performed under displacement control at a constant
rate of 0.3mm/sec.
The sheared area and the maximum loads were
Figure 1 Bending test setup recorded and the shear strength(𝑓𝑣 ) was computed
for each specimen using Equation (4) where 𝐴 is the
shearing area; 𝐹𝑢 is maximum load and; 𝑘 = 0.78 +
3.2 Shear Test 0.0044 is modification factor that modifies the shear
strength for test specimens when the sheared area is
Shear test of glue lines is required for quality control less than 50mm.
measurement in glulam manufacturing. The strength
and integrity of glue lines in glulam lamellas can be 𝑓𝑣 = 𝑘
𝐹𝑢
(4)
𝐴
derived by the test on block shear samples
according to [19]. The requirement for glue line
Besides the shear strength, the wood failure
integrity is based on testing of glue lines in a full cross-
percentage (WFP) has to be determined to identify
sectional specimen which cut from the
the quality of a bond which usually measured by
manufactured glulam member as representative of
visual examination. However, the WFP over a cross
the production. The acceptance criteria for shear
sectional specimen in average or any individual
strength of each glue line shall be at least 6.0N/mm2.
value shall exceed the minimum wood failure
For lighter density timbers, shear strength of 4.0N/mm2
percentage as stated in MS758:2001.
shall be regarded as acceptable if the wood failure
percentage is 100%. The quality of the glue line can
vary significantly between and within members.
Therefore, multiple samples must be taken into 4.0 RESULTS AND ANALYSIS
account in order to get reliable global estimations of
member's properties for the effect of irregularities 4.1 The Relationship of Load and Deflection
[21].
Figure 3 shows relationship between load and
3.2.1 Samples Preparation deflection for Kekatong and Melagangai glulam
beams. In general, both glulam timber species show
Test samples were cut in the form of block with similar behaviour in performance until the test sample
dimensions of 50mm x 50mm x 50mm for Kekatong reached the first fibre crack (𝑃𝑝 ) then continued and
glulam beam. For Melagangai glulam timber beam, attained the maximum load (𝑃𝑚𝑎𝑥 ). From the graph,
test specimens were cut into a shape form and Kekatong showed the steepest slope in elastic region
dimensions as in Figure 2 since the lamella thickness with higher stiffness compared to Melagangai but
was less than 25mm thick. The loaded surfaces were then failed abruptly. Table 1 show the average value
cut parallel to each other and perpendicular to the of 𝑃𝑝 for glulam Kekatong beam is 33.18% higher than
grain direction. Test specimens were taken from the that of glulam Melagangai beam. For the average
full cross-sectional of glulam timber beam with at ultimate load, the glulam Kekatong beam is 46.66%
least three (3) glue lines were tested in each of the higher than that of glulam Melagangai beam with
lower, middle and upper part of the glulam timber small difference in average midspan deflection of
beam cross-sections. 1%. In order to reach the average ultimate load,
glulam Kekatong beam achieved 34% higher after
the first fibre crack occurred while for glulam
Melagangai beam is 17% after the first fibre crack.

Figure 2 Glue line shear test specimen; (a) Dimensions; (b)

Loaded surface at transverse surface
114 Norshariza et al. / Jurnal Teknologi (Sciences & Engineering) 78:5–5 (2016) 111–117

Figure 3 The relationship of load and deflection graph

Table 1 The average bending strength for Kekatong and Melagangai glulam beams

4.2 Bending Strength Properties crack before it reached the total failure but this
scenario is not applicable for the glulam Melagangai
Table 1 presents an average basic properties on beam.
strength (MOR) and stiffness (MOE) for the Kekatong The average MOE value for glulam Kekatong
and Melagangai glulam beam calculated using timber is higher compared to glulam Melagangai
Equation (1) and Equation (2). MOR for glulam timber with the percentage of difference of 48.67%.
Kekatong beam showed the higher average value According to the graph in Figure 3, proves that the
compared to glulam Melagangai beam which the steepest slope for glulam Kekatong timber indicates
percentage of difference is 46.66%. During the the higher stiffness of the beam before reaching the
bending test, glulam Kekatong beam shows the first fibre crack.
higher fibre stress at proportional limit with the
average value of 50.15N/mm2 while glulam 4.3 Permissible Bending Strength of Glulam Beams
Melagangai beam reached the average fibre stress
at the proportional limit of 33.51N/mm2. The results Table 2 shows the bending strength and the
indicate that the glulam Kekatong beam attained permissible stress of glulam beams for Kekatong and
the average maximum stress about 34% after the first Melagangai timber. The bending strength values are
fibre cracked and glulam Melagangai beam obtained from Table 1whereas the permissible stress is
reached the maximum stress about 17% after the first calculated using Equation (3).
crack. From the observation, the glulam Kekatong
permits the beam to sustain longer time after the first
115 Norshariza et al. / Jurnal Teknologi (Sciences & Engineering) 78:5–5 (2016) 111–117

Table 2 Bending strength and permissible stress

The bending strength and the permissible stress the wood failure percentage is at least in 74%. Table
values for both timbers are compared and the results 3 summarised the global shear strength and wood
indicated that the value of bending strength is higher failure percentage for Kekatong and Melagangai
than the permissible bending stress design. It can be glulam test specimens. Both timbers of Kekatong and
concluded that the glulam beam for both species Melagangai show good performance in shear
are suitable to be used as structural member. strength with the average values of 12.25 N/mm2 and
10.67 N/mm2 respectively. The average values of
4.4 Glue Line Shear Strength wood for both timbers show the decreases as timber
density increases while the average values of shear
For Kekatong and Melagangai timber, the minimum strength increases.
requirement for shear strength is set as 6N/mm2 while

Table 3 Shear strength and wood failure percentage

Figure 4 shows the results for each individual test

specimen for Kekatong and Melagangai glulam
timber considering their simultaneous fulfillment of the
acceptance criteria requirements established in MS
758:2001. Considering the criteria of shear strength
and wood failure percentage, generally Kekatong
and Melagangai test specimens comply with the
requirements.
In this study, two types of failure modes are
identified which are cohesive failure and adhesive
failure. Cohesive failure was found as dominant
failure surface for both timbers of Kekatong and
Melagangai block samples. However, a few block
samples in Kekatong are failed in adhesive mode
generally for wood failure percentage below than
75% as shows in Figure 5(a). From the observation in Figure 4 Shear strength and percentage of wood failure in
this test, 36% of test blocks were failed in adhesion test specimens of kekatong and melagangai glulam timber,
failure mode for Kekatong timber while in the solid line represents the requirements outlined in
MS758:2001
Melagangai this type of failure is not found. Figure
5(b) shows cohesive failure mode in Kekatong and
Figures 6 (a) and (b) in Melagangai block samples.
116 Norshariza et al. / Jurnal Teknologi (Sciences & Engineering) 78:5–5 (2016) 111–117

The dark area on the surface as shown in those Acknowledgement

figures is the PRF adhesive. For the failure surface
occurred in timber there is less appearance of PRF The provision of glulam and solid timber railway
indicating that there is a good adhesion and wetting sleepers by Woodsfield Timber Industry (M) Sdn. Bhd.
between the adhesive and timber. and supported documents by Keretapi Tanah
Melayu Berhad (KTMB) are gratefully acknowledged.
The funding of this project is from Ministry of Science,
Technology and Innovation (MOSTI) under research
grant 03-01-01-SF0724, the Malaysian Timber Industry
Board and Universiti Teknologi MARA. The authors
wish to thank the laboratory technicians of Civil
Engineering Faculty for their assistance.

References
Figure 5 Surface Failures for Kekatong Test Blocks; (a)
Adhesive Failure at 50% of Wood Failure; (b) Cohesive [1] Ets Rothlisberger, S. A. 2008. History and Development of
the Wooden Sleeper. [Online]. Available: www.corbat-
Failure at 100% of Wood Failure
holding.ch.
[2] Sadeghi, J. 2012. New Advances in Analysis and Design of
Railway Track System. In X. Perpinya, ed. Reliability and
Safety in Railway. InTech. Available at:
www.intechopen.com. 76-100.
[3] Hamzah, S. H. and Din, K. 2008. Appraisal of used Wooden
Railway Sleeper. J. Eng. Sci. Technol. 3(3): 224-233.
[4] Mohamed, A. R. 2012. Engineer of Permanent Way, KTMB
in Personal Communication.
[5] The Malaysian Timber Industry Board. 2011. Glulam:
Freedom of Shape and Design.
[6] Anon. 2008. Handbook 1 - Timber Structures. Educational
Materials for Designing and Testing of Timber Structures
Figure 6 Surface Failures for Melagangai Test Blocks; (a) (TEMTIS).
Cohesive Failure at 100% of Wood Failure; (b) Cohesive [7] Antonio Silva, Martins, A. C., Feio, A. O. and Machado, J.
Failure at 75% of Wood Failure S. 2014. Feasibility of Creosote Treatment for Glued-
laminated Pine-timber Railway Sleepers. J. Mater. Civ.
Eng. 1-11.
5.0 CONCLUSIONS [8] Felipe Hideyoshi Icimoto, Fabiane Salles Ferro, Carlito Calil
Nato, and Carlito Calil Jr. 2013. Glulam Ties Manufactured
with Oocarpa Pinus and Polyurethane Adhesive.
The bending and shear strength of glulam using International Conference on Wood Adhesives 2013. 600-
selected Malaysian tropical hardwood in different 607.
strength group were investigated and concluded as [9] Gallery, D., Gauntt, J. C. and Webb, D. A. 1999.
following: Progressively Engineered Hybrid Wood Crossties for the
Next Century. The Roadmaster’s Conference Proceedings
of American Railway Engineering & Maintenance-of-Way
1. Glulam Kekatong and Melagangai have the Association.
higher value of maximum bending capacity [10] Gong, M., Delahunty, S., Chui, Y. H. and Li, L. 2013. Use of
compared to the permissible bending Low Grade Hardwoods for Fabricating Laminated Railway
strength, therefore these glulam are suitable Ties. Constr. Build. Mater. 41: 73-78.
[11] Carrasco, E. V. M., Passos, L. B. and Mantilla, J. N. R. 2012.
to be used as structural member and can be Structural Behavior Evaluation of Brazilian Glulam Wood
recommended as glulam timber railway Sleepers when submitted to Static Load. Constr. Build.
sleepers. Mater. 26(1): 334-343.
2. For good bonding, the bond need to have [12] KTMB. 2005. Technical Specifications. Malaysia: KTMB.
good shear strength and the failure should [13] Thelandersson, S. and Larsen, H. J. 2003. Timber
Engineering. England: John Wiley & Sons Ltd.
occur in timber rather than in the glue line. [14] Wan Mohamad, W. H., Razlan, M. A. and Ahmad, Z. 2011.
Both Kekatong and Melagangai glulam has Bending Strength Properties of Glued Laminated Timber
satisfactory bonding performance. from Selected Malaysian Hardwood Timber. Int. J. Civ.
3. Data from the above experimental work can Environ. Eng. 11(4): 7-12.
be used as pilot data for selecting the [15] Norshariza, M. B., Ahmad, Z. Bakar, A. A. and Tahir, P. 2013.
Ultimate Strength of Kekatong Glued Laminated Timber
suitable material and adhesive type for Railway Sleepers. The International Civil and Infrastructure
application of glulam timber railway Engineering Conference 2013. 97-105.
sleepers. The reliability of glulam timber as [16] MS 544: Part 2. 2001. Code of Practice for Structural use of
railway sleepers will investigate in the next Timber - Permissible Stress Design of Solid Timber. Malaysia:
testing phase under the sleeper Department of Standards Malaysia.
[17] MTIB. 2010. 100 Malaysian Timbers. 2010 Ed. Kuala Lumpur:
performance tests. Malaysian Timber Industry Board (MTIB).
117 Norshariza et al. / Jurnal Teknologi (Sciences & Engineering) 78:5–5 (2016) 111–117

[18] MS 1714. 2003. Specification for Visual Strength Grading of [20] ASTM D198. 2013. Standard Test Methods of Static Tests of
Tropical Hardwood Timber. Malaysia: Department of Lumber in Structural Sizes. United States: ASTM
Standards Malaysia. International
[19] MS 758. 2001. Glued Laminated Timber - Performance [21] Tannert, T., Müller, A. and Vallée, T. 2010. Assessment of
Requirements and Minimum Production Requirements the Shear Strength of Glued-laminated Timber in Existing
(First revision). Malaysia: Department of Standards Structures. The Final Conference of COST Action E53.
Malaysia.