Wood Material Science & Engineering

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/swoo20

Application of the three-point bending test for

small-sized wood and veneer samples

Carolin Siegel, Beate Buchelt & André Wagenführ

To cite this article: Carolin Siegel, Beate Buchelt & André Wagenführ (2022) Application of the
three-point bending test for small-sized wood and veneer samples, Wood Material Science &
Engineering, 17:3, 157-162, DOI: 10.1080/17480272.2020.1814410

To link to this article: https://doi.org/10.1080/17480272.2020.1814410

Published online: 15 Sep 2020.

Submit your article to this journal

Article views: 239

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=swoo20
WOOD MATERIAL SCIENCE & ENGINEERING
2022, VOL. 17, NO. 3, 157–162
https://doi.org/10.1080/17480272.2020.1814410

ORIGINAL ARTICLE

Application of the three-point bending test for small-sized wood and veneer
samples
Carolin Siegel , Beate Buchelt and André Wagenführ
Professur für Holztechnik und Faserwerkstofftechnik, Institut für Naturstofftechnik, TU Dresden, Dresden, Germany

ABSTRACT ARTICLE HISTORY

The aim of this work is to establish a bending test method as evaluation criteria for mechanical Received 15 May 2020
properties (strength and modulus of elasticity) of veneers. The bending test is one essential and Revised 19 August 2020
easy test method for wood characterization. Actually there are no established methods for Accepted 20 August 2020
characterizing smaller sized wood samples – in particular veneer. This investigation is based on the
KEYWORDS
established test methods for solid wood. The three-point bending test was successfully conducted Veneer; wood; bending test;
with veneer and small-sized samples and evaluated. small-sized wood samples

Introduction section. There are currently no methods for characterizing

smaller sized wood samples – in particular veneer – for
Wood is a proven material for a wide variety of applications.
these new fields of application.
Due to its good mechanical properties and low density,
Accordingly, the existing bending test methods of wood
wood is used in the construction industry, among others.
and wood-based materials are not applicable for veneers.
The fact that wood is a renewable raw material has spurred
The veneer itself has a thickness of 0.6 to 3 mm or more.
its demand. With the socially increasing ecological consider-
Table 1 shows the geometries according to the corresponding
ation of materials, wood becomes the focus for further, pre-
standards. Within DIN EN 310, the flat samples have a
viously unused fields of application. Not only solid wood of
minimum thickness of 7.5 mm, according to DIN EN 52186
larger dimensions but also wood of smaller dimensions –
of 20 mm. The veneer is not available in this thickness corre-
such as veneer – as base material for structural components
sponding to these standards.
is increasingly being investigated (Eckardt and Eichhorn
The aim of this work is to establish a bending test method
2010, Kohl and Böhm 2016, Kohl et al. 2018).
as evaluation criteria for mechanical properties (strength and
For the estimation of the mechanical properties of the
MOE) of veneers. Therefore, the three-point bending test was
wood for established applications respective test methods
conducted and evaluated according to the established test
are known. For construction materials or similar applications
methods. For this purpose, small-sized, thin wood samples
the strength and modulus of elasticity (MOE) are important.
were examined to assess their properties. The problems that
The bending and tensile tests are essential test methods for
should be discussed are:
wood characterization, such as strength. For thin materials,
such as veneers, tests for determining the characteristic
(1) Is it possible to test small samples?
data for model development are known (Dietzel et al. 2016,
(2) Which sample dimensions are suitable?
Zerbst et al. 2020). Dietzel et al. (2016) describe results of
(3) What quality do the determined values have?
tensile tests of veneers with elaborate sample preparation.
Zerbst et al. (2020) have conducted the Nakajima test in
In terms of quality assurance, quick tests on small samples
order to evaluate and model the forming limits of veneer
for other materials are already common. This investigation
samples. Such investigations are not suitable for application-
can provide a basis for transferring such quality assurance
related tests that allow an initial evaluation of small samples
tests to wood.
as suggested in this study.
In contrast to the tensile test, the bending test does not
require any complex preparation of the sample (bonding
Materials and methods
end tabs; tapering the sample). In addition, the bending test
is an application-related test for structural elements. Standard European beech (Fagus silvatica L.) wood and sliced veneers
or conventional bending test methods include DIN 52186 for were used for the investigations. The sliced veneers were
solid wood and DIN EN 310 for wood-based materials. These industrially produced in high quality. Table 2 shows the inves-
test methods were developed for wood or wood-based tigated groups. The reference group corresponds to DIN
materials with dimensions of several millimeters in cross- 52186. All other groups are based on the flatwise test in DIN

CONTACT Carolin Siegel [email protected]

© 2020 Informa UK Limited, trading as Taylor & Francis Group
158 C. SIEGEL ET AL.

Table 1. Sample geometry corresponding to the standards.

Minimal dimensions Distance between the
Standard Sample cross section Thickness (h) Width Length center of supports
mm mm mm mm
DIN 52186 solid wood 20 20 300 15 h

DIN EN 310 wood-based materials 7.5 50 150 20 h

EN 310. Both standards define different distances between Hegewald & Peschke Inspekt 10 testing machine. The testing
the centers of the supports (Table 1). Both distances conditions are shown in Table 3.
between the centers of the supports (15*h; 20*h) were Based on DIN 52186, the bending stress σ was calculated
regarded in the tests. The two control groups are solid as the quotient of the bending moment and the section
wood with correspondingly reduced standard geometries. modulus of the sample (1). In a rectangular sample cross-
The thickness of the veneer samples was based on the section (width b; thickness h) and center force (F) application
available raw material. The 2.4 mm samples represent thick with the distances between the centers of the supports (l),
veneer; conventional veneers are represented by 1.4, 1 and follows (DIN 52186):
0.6 mm thickness.
s = 3∗F∗l/(2∗b∗h2 ) (1)
The relation of distances between the centers of the sup-
ports and thickness (h) deviates from the standards. The dis- The bending strength Rm is the maximum bending stress
tance is set at 15 times the thickness in solid wood (DIN (2). The bending modulus of elasticity is a parameter for the
52186) and 20 times for the flat bending test (DIN EN 310). deformation rigidity of the sample in the elastic range (Δf ),
Based on these different relations and the different thick- where the elastic range is defined between 10% and 40% of
nesses, a variation of 9 series has been tested (Table 2). The the maximum force. It is determined from the elastic deflec-
veneer with a thickness of 0.6 mm would have a distance tion of the sample and calculated as follows (DIN EN 310):
between the centers of supports of 12 mm, based on 20*h,
a very short distance. In order to ensure reproducible hand- E = (l 3 /(4∗b∗h3 ))∗DF/Df (2)
ling, the sample length was set to 20 mm as for the 1 mm Formula (1) and (2) are valid with the boundary condition
thick series. Accordingly, the bearing distance ratio was that the deflection of the straight centerline is small com-
30*h for 0.6 mm veneer samples. With the exception of the pared to the sample cross-section (beam theory).
control group, the width of the samples was kept at 10 mm The results were statistically evaluated using the method of
and the length of the samples corresponded to the relation Analysis of Variance (ANOVA) at the 5% significance level.
of thickness to the distances between the centers of the sup-
ports of the aforementioned standards. Table 2 shows the
sample geometries, the according distances between the Results and discussion
centers of the supports as well as the number of valid
Bending strength
samples that were tested.
The solid wood was cut by a saw. The veneer samples were The determined bending strength is shown in Table 4. The
cut by laser to avoid additional cracks. The samples were con- obtained values are in the range of literature values
ditioned and then tested at 20°C with 65% RH. The tests were between 90 and 125 N/mm² (Ozyhar 2013, Schlotzhauer
conducted parallel to the fiber direction of the wood; thus, the et al. 2017, DIN 68364, Sell 1997, Table 5). Some values are
load was applied in the radial direction (Figure 1). The solid even higher. The comparatively low strength values of solid
wood samples were cut in accordance with the fiber orien- wood can be attributed to the obliquely of the fiber direction
tation of the purchased veneers. The production-related within the samples. Oblique fibers in the wood influence the
surface cracks of the veneers were taken into account. In bending strength. In testing the solid wood samples, obli-
order to reduce the influence of cracks they were positioned quely fibrous broken samples were observed. These strength
facing upwards (Figure 1). All tests were conducted with a values were not so low to be classified as outliers. Though

Table 2. Sample overview.

Group Sample name Material Density Sample geometries Distance between the
Thickness (h) Width Length center of supports Amount of samples
g/cm³ mm mm mm mm n
Reference 20 × 20 × 360 Solid wood 0.77 20 20 360 15 h 16
Control 4 × 10 × 72 Solid wood 0.74 4 10 72 15 h 20
2.4 × 10 × 45 Veneer 0.70 2.4 10 45 15 h 18
1.4 × 10 × 25 Veneer 0.63 1.4 10 25 15 h 20
Control 4 × 10 × 90 Solid wood 0.77 4 10 90 20 h 20
2.4 × 10 × 57 Veneer 0.65 2.4 10 57 20 h 14
1.4 × 10 × 32 Veneer 0.64 1.4 10 32 20 h 20
1 × 10 × 20 Veneer 0.62 1 10 20 20 h 17
0,6 × 10 × 20 Veneer 0.58 0.6 10 20 30 h 19
 WOOD MATERIAL SCIENCE & ENGINEERING 159

only be met approximately. Here, the deflection of the straight

centerline is quite large compared to the sample cross-
section. This fact can explain the comparable lower strength
value of the thin veneer.
A statistical analysis of the determined bending strengths
did not lead to analyzable results. Due to the fact that indust-
rially produced veneers were used, the densities of the
veneers differed. The density has an important influence on
all mechanical properties. Thus the specific bending strength
(ratio of strength to density) was determined in order to
compare the results independently of the sample density.
The results of the specific bending strength are represented
in Figure 2 and show that the specific bending strength is
in a range of 0.15 and 0.2 MPa/kg*m–3. Literature values of
Figure 1. Concept of the 3-point bending test. the specific bending strength result in a range from 0.13–
0.17 MPa/kg*m–3 (Table 5). Compared to the literature
Table 3. Testing conditions in dependence to the sample thickness. values (Table 5) the specific bending strength of the reference
Thickness Velocity Load cell size Roller pins diameter is within this range.
mm mm/min kN mm The influence of sample thickness as well as the
20 13 10 15 influence of the distance between the center of the sup-
4 5 10 6 ports on the specific bending strength are analyzed by
2.4 3 0.5 6
1.4 2 0.5 6 the ANOVA-test.
1 1 0.5 6 First the influence of the sample thickness on the specific
0.6 1 0.5 6 bending strength was verified. Therefore all samples (Table
2) with a distance between the centers of the supports of
15*h were analyzed. The test shows significant differences
Table 4. Bending strength and MOE. between all results with a distance of 15*h. This comparison
Distance between Bending regards different cross-sections, square and rectangular
Sample name the center of supports strength MOE
cross-sections. The ANOVA test of all rectangle cross-section
mm MPa MPa
(without reference) results in no significant differences. That
20 × 20 × 360 15 h 120.6 4.5 11,994 672
4 × 10 × 72 15 h 134.5 8.0 12,175 875 means, there is a significant difference between the average
2.4 × 10 × 45 15 h 130.8 6.6 10,820 524 specific bending strength of the rectangle cross section
1.4 × 10 × 25 15 h 113.6 12.3 8008 886 samples and the specific bending strength of the squared
4 × 10 × 90 20 h 134.4 10.6 14,184 1942
2.4 × 10 × 57 20 h 113.6 3.2 10,069 360 cross section samples (reference). The value of the reference
1.4 × 10 × 32 20 h 116.8 13.9 9933 1302 sample is significantly lower.
1 × 10 × 20 20 h 122.2 13.0 9776 1320 The ANOVA test of the samples with a distance between
0.6 × 10 × 20 30 h 90.6 7.9 6955 820
the centers of the supports of 20*h was carried out with rec-
Standard deviation.
tangular cross-section samples. The ANOVA test shows that
the thickness has no influence on the specific bending
these values have an impact on the average. Nevertheless, strength for thicknesses 4, 2.4 and 1.4 mm, although the
these results are within the range of literature values. specific bending strength of the 1 mm sample thickness
It seems that the strength values tend to decrease with differs significantly from the other samples. The thinner
decreasing thickness. Especially the values of the 0.6 mm samples of 0.6 mm thickness could not be considered
veneer are comparatively low (Table 5). No oblique fiber because the distance between the centers of the supports
breakage was observed when testing the thin samples, was 30*h.
although this observation is difficult to make with thin Thus, it can be concluded that the thickness have no
samples. The values of the 0.6 mm veneer suggest that the influence on the specific bending strength for rectangular
break was always perpendicular to the fiber direction. For cross-section samples for both distances, 15*h and 20*h. Fur-
the thin 0.6 mm samples the boundary condition of the clas- thermore, it can be concluded that these specific bending
sical beam theory for determining the bending strength can strengths are representative results.

Table 5. Mechanical properties literature.

Dimension
Source Thickness Width Length Bending strength MOE Density Specific bending strength Specific MOE
mm mm mm MPa MPa g/cm³ MPa/kg*m-³ MPa/kg*m-³
Ozyhar (2013) 20 20 400 91.8 10.7% 12,400 10.2% 0.649 0.14 19.1
Schlotzhauer et al. (2017) 20 20 380 120.6 8.6 13,708 1223 0.720 0.17 19.0
DIN (68364) 120.0 14,000 0.710 0.17 19.7
Sell (1997) 90 … 125 12,300 … 16,400 0.70 … 0.79 0.13 … 0.15 17.6 … 20.7
160 C. SIEGEL ET AL.

Figure 2. Specific bending strength.

The influence of the distances between the center of the elasticity for the 0.6 × 10 × 20 and 1.4 × 10 × 25 series is com-
supports on the specific bending strength is non-uniform. paratively very low (Figure 3). Also the specific modulus of elas-
The ANOVA test results are as follows: ticity of the reference (15 MPa/kg*m-³) is lower than the literature
values. The values for the 4 × 10 × 90 series are high. The sizeable
. 4 mm thickness: the distance has a significant influence on spreading range also covers almost all of the determined values.
the bending strength The influence of the sample thickness as well as the distance
. 2.4 mm thickness: the distance has a significant influence between the centers of the supports on the specific bending
on the bending strength modulus were analysed using the ANOVA test.
. 1.4 mm thickness: the distance has no significant influence The ANOVA test results of the influence of the distance
on the bending strength between the centers of the supports on the specific
bending modulus show no significant, specific dependencies.
Thus, no uniform conclusion can be drawn. The distance The influence of the sample thickness on the specific
between the centers of the supports of 20*h can be preferred. bending modulus is obvious. For both distances 15*h and
This distance is conform to the DIN EN 310 and the handling 20*h the influence of the thickness is significant, the specific
of larger samples is easier. bending modulus decreases with decreasing thickness. This
result is independent of the cross-section of the sample.
Thus, two conclusions are possible:
Bending modulus
. the specific bending modulus cannot be determined
The determined bending modulus is shown in Table 4. The
obtained values are mostly lower than literature values that representatively,
. the specific bending modulus really decreases with
are in the range between 12300 and 16400 MPa (Sell 1997,
DIN 68364, Schlotzhauer et al. 2017). The tendency – that decreasing thickness.
the modulus of elasticity decreases with decreasing sample
thickness – is more noticeable than for strength (Table 4). It is conceivable that the veneer thickness has an impact on
Concerning the density, this tendency cannot be dis- the specific modulus of elasticity. The size effect may justify
proved (Figure 3), even though the modulus of elasticity this. This means that the smaller the thickness of the wood,
was determined for small deformations and the requirements the smaller the cell compound and the lower the resistance
for beam theory are met. to deflection. Another fact is the damage of the cell com-
Literature values of the specific bending modulus result in a pound caused by veneer production. Although the manufac-
range of 17.6–20.7 MPa/kg*m-³ (Table 5).The specific modulus of turing of sliced veneer causes relatively small cracks in the
 WOOD MATERIAL SCIENCE & ENGINEERING 161

Figure 3. Specific bending modulus.

sheets, the veneer is weakened. It cannot be ruled out that samples. The 20*h ratio is preferred to work with larger
cracks will also occur in the fiber direction, possibly as a samples. In addition to the better handling it is better in
result of the stress. Cracks can be excluded for the 20 and terms of their expected deviation. With these test conditions
4 mm sawn samples. Comparing these series considering the bending strength results in representative values.
the cracks for the veneer samples, the specific bending The quality of the results of the thinner samples (1 and 0.6
modulus will be different for these samples. mm) could not be estimated so that no recommendation can
The described considerations should be also valid for the be given. Although the distance ratio of 30*h is applicable for
results of specific bending strength. But this is not the case. thin veneers it is barely useable for determining the bending
Possibly the moment of inertia and thus the bending strength because the deflection before the break is very high,
stiffness is a critical influencing factor. The geometry depen- so just about in line with the beam theory.
dent variable of moment of inertia differs strongly between As a result of the investigations, the influence of the
the different test series. Possibly the different bending stiff- density should be considered. It is also important to observe
nesses complicate the comparison of the specific bending where the cracks are located in the veneer to achieve compli-
modulus of the samples and with literature values. ant results. The investigation proved that it is possible to
examine small samples concerning bending. This also
enables quality control for wood samples in small geometries.
Summary and conclusion It is also conceivable to determine veneer bend radii with this
The investigation shows that the determination of the test, as an alternative possibility to the test suggested by Sri-
bending strength for small-sized samples is possible. The nivasan et al. (2007).
determined results are in the range of literature values. This Further investigations should consider the influence of the
conclusion cannot be drawn for the determination of the wood structure with decreasing sample thickness on the
bending modulus. The determined values do not correspond bending modulus.
to literature values. The reason is not clear: it is possible that
the geometry conditions and thus the boundary conditions
of the classical beam theory are poorly fulfilled. Possibly the
size effect has an important influence and the bending Disclosure statement
modulus is really different for smaller samples compared to No potential conflict of interest was reported by the author(s).
solid wood.
In the case of determining the bending strength it can be
concluded that the distance between the centers of the sup-
ports of 20*h or 15*h is suitable and delivers good results for ORCID
thicknesses above 1.4 mm and rectangular cross-section Carolin Siegel http://orcid.org/0000-0002-1209-6121
162 C. SIEGEL ET AL.

References Kohl, D., Feldmann, M., Heim, H. P. and Böhm, S. (2018) Characterization of
wood-based multi-material systems under dynamic impact stress.
Dietzel, A., Raßbach, H. and Krichenbauer, R. (2016) Material testing of dec- Wood Material Science & Engineering, 56(5), 1–10.
orative veneers and different approaches for structural-mechanical Ozyhar, T. (2013) Moisture and time-dependent orthotropic mechanical
modelling: Walnut burl wood and multilaminar wood veneer. characterisation of beech wood. PhD thesis, ETH Zurich.
BioRecources, 11(3), 7431–7450. Schlotzhauer, P., Nelis, P. A., Bollmus, S., Gellerich, A., Militz, H. and Seim, W.
DIN 52186 (1978) Testing of wood; bending test. (2017) Effect of size and geometry on strength values and MOE of
DIN 68364 (2003) Properties of wood species-density, modulus of elas- selected hardwood species. Wood Material Science & Engineering, 12
ticity and strength. (3), 149–157. doi:10.1080/17480272.2015.1073175
DIN EN 310 (1993) Wood-based panels; determination of modulus of elas- Sell, J. (1997) Eigenschaften und Kenngrössen von Holz (Zürich:
ticity in bending and of bending strength; German version EN 310. Baufachverlag Lignum).
Eckardt, R. and Eichhorn, S. (2010) Abschlussbericht FNR, Schwingungs- Srinivasan, N., Bhattacharyya, D. and Jayaraman, K. (2007) Thermoforming
und geräusch-dämpfende Leichtbauelemente im Maschinenbau auf of wood veneer composite sheets. Holzforschung, 61, 558–562. doi:10.
Basis von Konstruktionswerkstoffen aus Holz. 1515/HF.2007.105
Kohl, D. and Böhm, S. (2016) Holzformteile als Multi-Materialsysteme für Zerbst, D., Affronti, E., Gereke, T., Buchelt, B., Clauß, S., Merklein, M. and
den Einsatz im Fahrzeug-Rohbau (HAMMER): Schlussbericht. Cherif, C. (2020) Experimental analysis of the forming behaviour of
Verbundprojekt im Rahmenprogramm ‘Werkstoffinnovationen für ash wood veneer with nonwoven backings. European Journal of
Industrie und Gesellschaft – WING’. Wood and Wood Products. doi:10.1007/s00107-020-01494-0