SciELO - Scientific Electronic Library Online

vol.29 issue1CFD modeling of basic convection cases in enclosed environments: Needs of CFD beginners to acquire skills and confidence on CFD modelingStructural damage detection: comparison between GA and PSO techniques author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google


Revista ingeniería de construcción

On-line version ISSN 0718-5073

Rev. ing. constr. vol.29 no.1 Santiago  2014 


Shear strength parallel of the wood fiber stress by punching – proposed test

Mário Massayuki *, Adalberto Matoski¹*, Cláudio Magajewski*, Juliana Machado*

* Universidad Tecnológica Federal do Paraná (UTFPR). BRASIL

Dirección de Correspondencia

This research aims to propose a method for testing the shear strength of wood parallel to fiber and compare with other proposed methods. The method proposed by NBR 7190/1997 Brazilian Standard, that is similar to ASTM D 143-09/2010, the uses a rectangular prismatic specimen with an indentation to prevent its rotation, which it is caused by the asymmetry in the application of force that promotes the shear. In order to eliminate this asymmetry, the proposed procedure applies shear stress by punching a simple prismatic test specimen without cuts, with the distribution of shear section symmetrically in relation to a point. There upon, it's developed a male and female tool adapted to a universal test press for force application. Once the consistency of the method is verified, it's been carried out a series of tests with three different wood species (Pinus sp, Erisma uncinatum and Mezilaurus itauba), distinguished by their mechanical strength, inspecting the method by Santos Neto (1999) under the NBR 7190/1997 Brazilian Standard, and comparing it with the proposed method - punching. The outcomes showed that the testing with NBR 7190/1997 Brazilian Standard methods and the proposed method displayed similar and close results between them.

Keywords: Wood, parallel shear force, punching

1. Introduction

Wooden structures are widely used in civil engineering mainly by the great supply of wood available, by the simplicity of the tools and processes for its purpose.

The good performance of wood as a building material requires knowledge of its physical and mechanical properties. There are many available types in the market with very different properties allowing several scopes. The physical and mechanical properties for the application in civil engineering are normalized. NBR 7190/1997 Brazilian Standard, that is similar to ASTM D 143-09/2010, establishes the procedures for obtaining the wood mechanical strength for different types of stresses.

The shear force parallel to fiber test deserves special attention in this study because of the importance of its results for sizing of structural elements where the use of inexact values can cause property damage and other irreversible consequences. EVANS (2000).

In this test, which is similar to a guillotine, compressive forces act on a standard test specimen in order to cause shear force in a studied section. Researchers have observed that during this test there are distortions that have been studied and analyzed with proposals of new methodologies in order to lessen possible deviations.

Thus the main objective of this research is to propose a test method for determining the shear strength parallel to fiber by applied-stress punching of the test specimen, and compare the results of the tests with the ASTM-D143-09/2010 that the (NBR7190/1997) is similar. The second objective is also to compare with other proposed method.

This work is based on tests of punching that are already used and standardized for metallic materials as cited in study of Acharya and Ray (2013).

2. Literature review

The interest in determining the mechanical wood properties is justified by the fact that in most situations of use, the stresses that the wood is controlled by must be known, thus enabling the correct sizing of the various elements which make up the structural set as a whole.

Authors with LIY (1983) studied new methods for the test for the wood shear strength. These authors saws that the ASTM D143-09 method not achieved the pure shear.

In other article Gupta and Sinha (2011) conclude that de ASTM D143-09 shear test specimen produces stress concentrations that which creates peak shear stresses.

European authors as Yoshihara (2012) also studied the shear timber comparing this time with the ISO 15310 standard.

The authors cited above show concern for the shear test, and they seek the real values. Similar works is based on tests punching already commonly used for metallic materials.

The property of wood (shear strength parallel to the fibers) is of great importance in pegged wood, rollers, sheaves, and shears. The shear is the separation of the fibers by a parallel stress applied to them. The test consists of applying and measuring the stress of the wood fibers, to cause the separation between them.

The shear strength of wood is directly proportional to its density, but it depends mainly on the direction in which stress is applied in relation to the wood growth rings (usually the break is determined in the tangential or radial plane). In the tangential plane, there is a great influence of the initial difference between the early and later wood. In the radial plane there is a great influence by the wood rays and wood with wide rays has low resistance to shear compared with narrower growth rings. In the tangential plane there is a strong influence of the tree growth, or the difference in resistance between the early and late woods.

The shear test of wood is problematic due to the stress concentrations, overlapping of tensions, especially as the bending, compressive stress and rupture developments that cover-up the shear force phenomenon, making the results questionable. We can take Santos Neto (1999) method, which proposes the modification of the test specimen. (Figure 1) In this proposal, the test specimen of small thickness, 20mm, runs three cracks, aiming to obtain two planes that make the shear stresses to appear easily.

To calculate the shear strength, the load applied is considered as well as the area where the test specimen breaks.


Figure 1. Specimen according Santos Neto Method. Santos Neto (1999)

Coker e Colemann apud Mendes (1984) noted that all the specimens broke in the reduced area and only by shear force parallel to the fibers. He also noticed that the break in the weakened region of test specimen presented, it is not by pure shear, but shows a combination of shear and tensile strengths.

In the model proposed by NBR 7190/1997 Brazilian Standard, the stress is applied to the test specimen through a metal plate called knife with defined thickness, which when in contact with the test specimen, transfers all the compression stress to it. The test specimen suggested by the standard is rectangular prism, in a cubic shape (Figure 2). Note that the specimens are similar to those proposed by the ASTM – D143-09 (2010).


Figure 2. Specimen by model proposed by Brazilian Standard and Similar to specimen proposed by -D143-09/2010

The compression force applied to the top of the test specimen is transferred to the support plate that sits on the lower top of the test specimen. The face of the support plate coincides with the plane formed by the inner side of the knife. This model provides a shear plane in the test specimen, equipment used in Figure 3, creating a shear between the top plate and the supporting plate and a sheared area between them, in the test specimen.


Figure 3. Equipment used for testing the shear strength of wood in Brazilian Standard

It is observed that the contact of the knife with the test specimen requires a contact area with dimensions that avoid rupture due to compressive crushing of the timber, prior shear breaking. The center of applicable force to the shear plane of the test specimen is equivalent to half the thickness of the knife. Similarly, the same occurs in the lower supporting plate resulting in an eccentricity between the forces applied to the top edge of the test specimen and at the top bottom of equal value to the thickness of the knife. Since we have a force of action and reaction, parallel to each other and of opposite sign, separated by an arm, which in this case is the eccentricity, we have a moment created by the binary forces. This moment is the cause of the rotation that occurs in the shear test. This is a significant flaw which solution can be the insertion in the test specimen of traction and compression fields simultaneously to the shear force, making the outcomes questionable.

Analyzing the processes used in the metalworking industry to make holes in metal parts, it was observed that punching is the most widely used and economical method, used in mechanical presses and automatic machines. The process consists of a set of metal pieces called male and female, usually with circular shape for cylindrical holes, which may have variants. This set is attached to a machine that induces force, usually a hydraulic press.

The set is assembled in a way so the action of the male is inserted on the female, which can be penetrated in a few millimeters or piercing it through. The male part is attached to the machine that performs the movement upward and downward of the tool and also transmits the force generated by the machine. The female part is attached to the base of the machine. The set must be assembled so that the movement of the male part is accurate over a vertical axis passing through the longitudinal center of the cylinder of the male part. Any deviations due to imperfections in the movement can lead to a clash between male and female damaging the tool.

This process solves the major problems described above, such as the ratio between the shear and compression stresses. The fact comes purely from the process' geometry where the compression will act in a circular area equal to the cross section of the male part, and this has a relation with the square of the section radius. The shear area of the test specimen has a linear relation with the radius of the cross section of the male part. This characteristic allows the rupture of the specimen is only by shear.

Another important aspect is the stability of the test, as this will be multi-symmetric in relation to the longitudinal axis of the male part to which we shall call the test director axis. With this feature we assure that the test specimen will not rotate.

The three species adopted in this work, are the most sold in Brazil where the common name is Itauba, Cedar and Pinus. The scientific name of Itaúba is Mezilaurus itauba and it is very common at state of Pará and Mato Grosso in Brazil. The wood is heavy and hard, density about 0,8 g / cm³ with high compressive strength and low shrinkage, high natural resistance to decay and insect attack (ROBERT et al., 2012).

Pinus sp has distinct varieties: Pinus elliottii var. etc.. Part of the group of species with pine range in Canada and the United States. However there are many areas planted with this species in South America, particularly in southern Brazil. GARCIA et al. (2012).

The Cedar (Erisma uncinatum) as the International Tropical Timber Organization (ITTO) reports that the species is an important source of timber. The timber is reported to be exported regularly, and is imported from Brazil into China. The timber is reported to have good resistance to attack by decay-causing fungi and dry-wood insects. Its resistance to attack by termites is rated as poor.

3. Methodology

To start the process, it was used a test specimen with an area equal to the shear test specimen of ASTM D 143-09/2010, it was established the male part punch diameter of 40 mm and a gap between male and female of 1 mm resulting 20 mm thick for the test specimen (Figure 4).


Figure 4. Proposed test specimen

The compressed area for the proposed test are 1256.61 mm² (40 x 40 x 3.14159 / 4) and the sheared area are 2513.21 mm² (3.14159 x 40 x 20). While the sheared area of the standard's test specimen are 2500 mm2 (50 x 50) (see Figure 3).
The suggested initial parameters produced perfect tests and direct measurements for comparison with the results generated by the ASTM standard method and (the same) Brazilian Standard method, which are sufficient reasons for not promoting new attempts and avoid new costs. The adaptation of the punctures to the mechanical platform was made on the same machine shop and later the connection of the set was made in the universal mechanical press testing.

The punching method provides advantages over other methods:
Effortless preparation of the test specimen for examination;

The arrangement for the test shear parallel to the wood fibers is simpler;
The cost of the tool is more affordable (around 25% compared to the standardized tool);
Easy reproduction of the method.

Three types of species woods were chosen for this study: Pinus (Pinus sp), Cedar (Erisma uncinatum) and Itaúba (Mezilaurus itauba). These types were adopted, following the same Brazilian standard (NBR 7190/1997), as specified in Table 1. The wood classification of Pinus is exotic conifer. For Cedar and Itaúba, the wood classification are native dicots.

Justifies the choice of these species because they are the ones that have increased trade in Brazil, in the last 5 years. The pine is an exotic species and is widely grown in southern Brazil. The Cedar (Erisma uncinatum) and Itauba (Mezilaurus itaúba) are native species and beyond their internal use are also exported.


Table 1. List of woods used (adapted from NBR 7190/1997) Brazilian Standard

3.1 Characterization of wood
Once defined the types of wood, commercial wood boards were used with (75 x 100) mm section, with a 2500 mm length.

The wood boards were selected based on their availability, aiming to meet the Brazilian Standards guidance for obtaining samples for testing. To determine the shear strength, the measures of the sides of the test specimens were made with minimum accuracy of 0.1mm. The test was performed with steady increasing loading, with a rate of 2.5 MPa / min.

The basic density of wood is the density obtained by the conventional relationship between the dry mass of the specimen by the volume of the specimen saturated. The bulk density is obtained by taking the specimen with 12% moisture content. Their values were determined according to NBR 7190/1997.

3.2 Proposed method
The motif of the choice of the punch method is because the proposed method seeks to eliminate the undesirable effects and determine the real value of shear strength, following the same line of reasoning Liy (1983) and Gupta and Sinha (2011). The proposed method is an adaptation of the method used by Acharya and Ray (2013) for metallic materials.
The cross section of the punch is circular with a diameter of 40mm, so the test specimen should cover the hole created by the test area and leave an extra sufficient area to withstand the compression test, taking into account the evident principle of execution.
To define the size of the test specimen, they were prepared with the three wood types, varying sizes and thicknesses, as shown in Table 2.


Table 2. Types, size and thickness of the test specimens

After the test specimens, it was concluded that the best dimension of the test specimen is (75 x 75) mm (length x width) with a thickness of 20 mm. The modification of the thickness of 20 mm corresponds to a shear area of 2513 mm² which is equivalent to the area of NBR 7190/1997 Brazilian Standard method, which allows the direct comparison of the results.

The finishing of the surface of the test specimen where the load will be applied can be finished with fine sandpaper.

The unfinished body makes it easy for the homogeneous application of compression load, at the same time the irregularity allows the highest point squeeze and uniformly distributing the compressive stress.

Figure 5 shows the arrangement of the proposed method.


 Figure 5. Arrangement of Test for Punching

After the shear tests, the moisture content of specimens was determined and the results of shear were corrected for moisture. The goal is to correct the shear strength with various moisture levels, to the standard moisture of 12%. The expression (1) is defined by the NBR 7190/1997 Brazilian Standard.



= shear strength to the standard moisture of 12%;
= shear strength to moisture in the test specimen;
U = moisture content (%) of the test specimen.

3.3 Equipment
The application of forces in the test specimens completed through the Universal Testing Machine, in order to determine the mechanical properties of the shear. The equipment's brand is EMIC, model DL-10000, with a maximum capacity of 100kN, equipped with 100 kN load cell, controlled by specific software that ensures the constant speed / load and the load data acquisition as well as the deformations of the tests.


4. Results and discussion

Table 3 presents the results for the Pinus sp. The average specific gravity result is 363.50 kg/m3 and the apparent density is 434.52 kg/m³.


Table 3. Results for Tests of basic and apparent density of the Pinus sp

The results for tests of specific gravity and apparent density to the Cedar (Erisma uncinatum) wood are presented in Table 4.


 Table 4. Results obtained for tests of basic and apparent density of Cedar wood

Table 5 present the results for Itaúba, (Mezilaurus itauba ) the average specific gravity and apparent density are the highest compared between the three types of wood, which is an expected result.

The results found in literature for the basic density is around 800kg/m3 and 960kg/m3 for apparent density. We obtained average values of 705.73kg/m3 for specific gravity and 795.79kg/m³ for apparent density on the tests performed on Itaúba (Mezilaurus itauba) wood. There are several factors that can influence these results, but the main ones are the tree age and the sample position.

It is observed from Table 5 that the Mezilaurus itauba has the highest specific gravity between the species analyzed, thus being a harder wood which may explain the high results for the shear strength obtained for both standardized tests and also the proposed test.


Table 5. Results obtained for tests of basic and apparent density of Itaúba

4.1 Results of shear force test for Pinus sp

Table 6 presents the results of shear force tests, in the three methods, (method proposed by Santos Neto (1999), method according NBR 7190/1997 Brazilian Standard and the proposed method) for the Pinus sp wood.


Table 6. Results of shear force tests parallel to the fibers of the Pinus sp wood, obtained from three methods

The results of the tests proposed by Santos Neto (1999) obtained very high values so far from the values found in the tests made according to NBR 7190/1997 Brazilian Standard and tests the proposed method (see value Table 6).

From the results obtained in Table 6, it was calculated the average, standard deviation, variance and coefficient of variation for the method of NBR 7190/1997 Brazilian Standard and for the proposed method (Table 7).

Table 7. Statistical comparison of the test results of the NBR 7190/1997 Brazilian Standard and proposed method of punching of the Pinus sp wood

The range considered in the method of NBR 7190/1997 Brazilian Standard is between the values of 3.52 and 10.16 MPa.

Analyzing the results of tests of Pinus sp wood by the "t" test of student, it is observed that p-value in the Student's t test was 0.28047, then do not reject the hypothesis that the means are equal it can conclude that there is no significant difference between the tests with a confidence level of 95%.

4.2 Results of shear force test for Cedar (Erisma uncinatum) wood
Table 8 presents the results of shear force tests for the cedar (Erisma uncinatum) wood.

Table 8. The results of shear force tests parallel to the Erisma uncinnatum fibers, obtained from the three methods

After obtained these results, it was calculated the average, standard deviation, variance and coefficient of variation for the method of NBR 7190/1997 Brazilian Standard and proposed method (punching), displayed in Table 9.

Table 9. Comparison of results between ASTM (similar to NBR) and the proposed method

The considered range in the method of NBR 7190/1997 Brazilian Standard is between the values of 2.76 and 14.44 MPa for Cedar (Erisma uncinatum ) wood.

After the statistical analysis of the results for the cedar wood, (Erisma uncinatum) The p value found in the Student's t test was 0.11023, then does not reject the hypothesis that the averages are equal. It can be concluded that there is a significant difference between the results of tests with a confidence level of 95%. This aspect shows that the test of the Brazilian Standard may be influenced by the asymmetry test specimen. Table 9 presents the results of shear force tests for the (Erisma uncinatum Wood).

4.3 Results of shear force test for Itaúba wood.(Mezilaurus itauba)
Table 10 presents the results of shear force tests for the Itaúba (Mezilaurus itauba ) wood.

Table 10. Results of shear tests parallel to the Itaúba wood, obtained from the three methods

From the results obtained in Table 10, it was calculated the average, standard deviation, variance and coefficient of variation for the method of NBR 7190/1997 Brazilian Standard and proposed method – punching (Table 11).

Table 11. Statistical comparison of test results of the NBR 7190/1997 Brazilian Standard and proposed method - punching of the Itaúba wood

As noted in Tables 6 and 8 same results happens here. The of shear tests are close when comparing the proposed method and the methods of D 143-09 and NBR 7190/1997, but they are far from the values obtained by Santos Neto (1999).

The range considered in the method of NBR 7190/1997 Brazilian Standard is between the values of 9.09 and 11.31 MPa (without sample No. 13), and for the proposed method - punching the range of values is between 9.79 and 13.47 MPa.

However for Itaúba (Mezilaurus itauba) kind of Amazonian origin p value found in this test was almost nil, p = 0.0007 showing a significant difference between the means, adopting the confidence level of 95%. This is the only species that showed a statistically significant difference to the results. This shows that there is indeed an influence of the asymmetry observed in the specimens of the standard.

On the other aspect, the statistical analysis for all the results, regardless of the species, showed no significant difference between the tests with a confidence level of 95%. This result can be explained by the number of tests carried out (45 repetitions in this case) and the higher variation of them.

It can be concluded that the test of proposed method - punching is viable and can replace the test method proposed by NBR 7190/1997 Brazilian Standard. It can be noticed also that the equipment used for the test is cheaper and the time for the preparation of the test specimen is shorter.

The Box plot shown in Figure 6 compares the results of the shear tests for all three species of wood. The first aspect that we observe are the outliers that are larger for tests of the standard. In other words, this method is subject to an apparently larger number of failures. The second aspect of this chart is presented that Mezilaurus itauba for being hard wood obtained the highest shear values for both tests, correctly showing the influence of the species in the results. Finally, the graph shows that there is a similar behavior for both trials, despite the different wood species adopted. This shows that the effect of bending forces due to asymmetry of the specimen obtained according to ASTM D143-09/2010 test or NBR 7190/1997, is small and can be tolerated.

Figure 6. Box Plot Chart for shear strength for three specimens

5. Conclusion

The first result shows that the of Santos Neto, method achieved very high results when compared to the results obtained by the proposed method and obtained by the ASTM D143-09/2010 method that is the same of the NBR 7190/1997 method.

Comparing the test results obtained according to the method of ASTM D143-09/2010 is similar to that of NBR 7190/1997, Brazilian Standard method to the results obtained by the proposed method we found that were very close.

The similarity of the results and the closed values between the proposed test and ASTM (and NBR) tests allow us to admit the asymmetry has low influence.

Statistical analysis showed a significant difference only for the Itauba species and for other species there was no significant difference at the level of 95%.

It was also observed that the difference was statistically significant for the species of hardwood. Thus it is suggested that further work to a greater amount of species will be tested.

And finally, the results show that there is little influence of the asymmetry of forces applied tests found in the standards.


6. References


Acharya S., Ray K. K. (2013), Assessment of tensile properties of spot welds using shear punch test. Materials Science & Engineering A – v.565 March P. 405 – 413 access in 01/12/2013.         [ Links ]

ASTM D 143-09 (2010), Standard test methods for small clear specimens of timber. American Society for Testing and Materials, West Conshohocken. USA        [ Links ]

Associação Brasileira De Normas Técnicas (1997), Projeto de Estruturas de Madeira. NBR 7190. Rio de Janeiro.         [ Links ]

Evans J.L.W., Senft J. F., Green D. W. (2000), Juvenile wood effect in red alder: analysis of physical and mechanical data to delineate juvenile and mature wood zones. Forest Products Journal, v.50, n.7/8, p.75-87.         [ Links ]

Garcia F. M., Manfio D. R., Sansigolo C. A., Magalhaes P. A. D. (2013), Rendimento no desdobro de toras de Itaúba (Mezilaurus itauba) e Tauari (Couratari guianensis) segundo a classificação da qualidade da tora. Floresta e Ambiente, 2012 out./dez.; 19(4): 468-474 access in         [ Links ]

Gupta R., Sinha A. (2011), Effect of grain angle on shear strength of Douglas-fir wood. Holzforschung, v. 66, p. 655-658 – Berlin.         [ Links ]

Liu J. L. (2013), New Shear Strength Test for Solid Wood. Wood and fiber Science – v 16, nº 4 october 1984. Access in         [ Links ]

Mendes A. P. (1984), Resistência da Madeira ao Cisalhamento. 1984, 157p. Dissertação (mestrado), Escola de Engenharia de São Carlos, Universidade de São Paulo, São Carlos, 1984.         [ Links ]

Robert R. C. G., Santos A. S., Duarte Dos Santos L., Fantini A. C. (2012), Caracterização do Abastecimento de Madeira Serrada Comercializada no Município de Florianópolis – SC. Revista Floresta - Curitiba, PR, v. 42, n. 1, p. 85-94, jan./mar.2012.         [ Links ]

Santos Neto A. B. S. (1999), Cisalhamento em Corpos de Prova de Madeira. 1999. 146p. Dissertação – Mestrado em Engenharia /civil. Universidade Federal de Santa Catarina, Florianópolis, 1999.         [ Links ]

Yoshihara Hiroshi (2013), Shear modulus and shear strength evaluation of solid wood by a modified ISO 15310 square-plate twist method. Drvna Induystrija, v. 63 nº 1 p. 51-57- march 2012 - access in in 01/12/2013.         [ Links ]



Fecha de Recepción: 01/10/2013 Fecha de Aceptación: 09/11/2013