Elsevier

Ultrasonics

Volume 88, August 2018, Pages 43-50
Ultrasonics

The study of wood knots using acoustic nondestructive testing methods

https://doi.org/10.1016/j.ultras.2018.01.004Get rights and content

Highlights

  • The influence of knot on the modulus of wood is studied nondestructively.

  • The sound speed method offers a fast procedure to evaluate the position of knot.

  • The resonance method performs better to quantify the size of the knots.

Abstract

In the process of wood grading with the focus on detecting wood knots, nondestructive testing methods based on sound transmission can assist the traditional characterization methods to achieve a higher efficiency and better results. In this paper, we use two independent methods based on resonance and sound speed measurements to evaluate the elastic modulus of wood beams containing different knots. The results show that the method based on sound speed measurements offers a fast procedure to evaluate whether the knot is in the middle of the cross-section of the beam or not. In this case, both measuring methods are reliable in determining the knot’s characteristics. In the off-center case, the resonance method performs better to quantify the size of the knots.

Introduction

Acoustic nondestructive testing methods applied to wood primarily use the change of sound speed and sound resistance to obtain information about the mechanical characteristics of the wood under consideration and of the location of potential internal defects. Basically one can distinguish three kinds of methods: the ultrasonic pulse method, the acoustic emission technique and the resonance method.

In the early 80′s, Dunlop adopted the ultrasound pulse method to detect defects in wood, and to discriminate the different conditions of wood rot [1]. Shi measured the sound speed of wood along three orthogonal axes and related it – among other parameters – to their moisture content and compression strength [2]. At the end of the 80′s, researchers at EPFL developed an instrument to measure the dynamic modulus of wood based on the longitudinal sound speed [3]. Ross et al. used the difference in time-of-flight measurements to evaluate the presence of rot, cavities and knots in wood [4]. Schafer investigated the energy decay of sound transmission in wood by means of computer simulations in 2000 [5], based on a theory in which the decay of the energy is caused by internal friction and scattering upon transmission of ultrasound through wood.

In the domain of acoustic resonance, Nakayama was the first to introduce and confirm the merit of the resonance method in wood in 1975 [6]. About a decade later, Sobue measured the shear modulus of wood by recording the signal of a complex vibration [7]. Ous reviewed nondestructive evaluation (NDE) methods based on vibrational impulse response of wood and summarized different modes of vibration of logs [8]. Murphy investigated the analytical and numerical solutions for a simply supported vibrating beam, and obtained consistent results for different sample lengths [9]. Kamaguchi et al. related the conditions for transverse vibration to the moisture content of wood [10]. Tonosaki et al. exploited longitudinal and bending vibrations to study the inhomogeneous mechanical characteristics of wood for musical instruments [11].

Concerning NDE on the wood beams, Sandoz measured ultrasonic wave velocity in spruce and fir beams, and correlated the ultrasonic longitudinal wave speed to the modulus of elasticity (MOE) [12]. Diebold examined NDE methods including transverse vibration and ultrasonic inspection on several kinds of wood, and found the ultrasonic inspection had the lowest correlation to the bending strength [13]. Duju applied six methods of measuring MOE to five different timber species, he found the calculated MOEs are correlated to the values of modulus of elasticity (MOR) well [14]. Sandoz and Benoit utilized the ultrasonic wave speed and attenuated energy to characterize the MOE and MOR of wood, Energy damping of the wave was found to be directly dependent upon local singularities such as knots and grain angle [15]. Pires estimated the MOEs of wood using stress wave, transverse vibration, and ultrasound, ultrasound and stress-wave inspection were similarly correlated to those determined from static bending tests [16]. In China, research and use of the resonance method for wood NDE started about several years later. Chen introduced the resonance method by impact excitation in 1991 [17]. Zhong et al. analyzed the static bending modulus of wood using spectral analysis in 1996 [18]. Hu adopted a combination of bending vibration, longitudinal wave propagation as well as resonance to measure the dynamic modulus of chipboards [19]. Later, Cui examined the relation between the dynamic modulus of wood samples and the number of knots contained in these samples by the resonance method [20], while Wang and Yang adopted the ultrasound transmitting method and employed an FFT analysis of the received signals for wood containing holes [21]. The evaluation methods are mostly based on frequency. For transverse vibrations in wood samples of uniform cross-section under free constraint conditions, the resonance frequency spectrum can be related to the modulus and density of the wood material. Internally present knots can significantly affect these two physical properties. Therefore, measuring and comparing the harmonic frequencies of wood beams of the same size by impact excitation can be used to distinguish the presence and characteristics of knots.

The above literature review shows that most of the prior research concerns experiments on wood without defects, in order to identify the wood type and to determine its mechanical parameters. There is little work on assessing wood that contains knots as a kind of defect. Some researchers, e.g. Cui [20], have studied the numbers of knots in wood samples based on the resonance method, however, they neither discussed the size and location of those knots explicitly, nor did they compare this method to a more versatile method based on sound speed. In fact, in this paper, given the fact that the knot’s location affects the accuracy of the sound speed method, we will treat the knot’s position as a primary variable quantity, and use this information to adopt and combine the resonance and sound speed methods to evaluate the modulus of wood. Next, we analyze the obtained results in order to assess the reliability of the knots’ evaluation methods. As part of the conclusion, we formulate advice about some practical measurement procedures at different conditions.

Section snippets

Brief review of the theoretical models for modulus evaluation

In the present study, parallelepiped samples of wood with orthogonal symmetry, as shown in Fig. 1, have been used. The considered beams are long and slender, and have length l, width b (≪l) and thickness h (≪l). The (lengthwise) axial direction corresponds to the axis L in Fig. 1. The samples will be subjected to two types of measurements: longitudinal sound speed measurements along the axial direction, and resonance measurements induced by transverse vibrations perpendicular to the axial

Experimental samples

The experiments were performed on parallelepiped Pinus Sylvestris samples, which is a kind of soft wood. In order to control the variability, we selected two four meter long wood sticks, with rectangular cross-sections, and with the long axis parallel to the axial direction of the tree’s growing orientation, they are both the sapwood cut from the trees. The long sticks were cut into several smaller parts and divided into two groups according to the parental stick origin. As the two sticks were

Results for the first group of samples

For the first group of beams, the comparison of the dynamic moduli inferred from the resonance method (i.e. flexural modulus Ebend, cfr Eq. (4)) and by the sound speed measurements (i.e. compressive modulus E, cfr Eq. (1)) is shown in Fig. 7. The wood samples having number 1–7 are free of knots (and serve as reference samples), sample number 8 and 9 have one knot, and sample number 10–14 contain two knots. The knots of the beams in this group are all in or near the middle of a cross-section

Conclusion

In this paper, we adopted two methods based on sound speed and resonance measurements to infer the dynamic compressive and bending modulus of parallelepiped wood sticks containing knots. A correlation analysis of the two moduli was performed on samples whose knots are located in the middle of the cross-section of the beams or not. When the knots are in the middle of the beam’s cross-section, both methods provide reliable evaluations of the number of knots. When the knots are off-center, it is

Acknowledgements

Project supported by the National Key R & D Program, China (Grant No. 2016YFF0203000), National Natural Science Foundation of China (Nos. 11474160, 11774167), Fundamental Research Funds for the Central Universities (No. 020414380001), Key Laboratory of Underwater Acoustic Environment, Institute of Acoustics, Chinese Academy of Sciences (No. SSHJ-KFKT-1701) and AQSIQ Technology R & D Program, China (Grant No. 2017QK125).

References (27)

  • J.I. Dunlop

    Testing of poles by using acoustic pulse method

    Wood Sci. Technol.

    (1981)
  • B.Z. Shi et al.

    Study of sound speed in wood, I relationship between sound speed and smooth grain compressive strength, moisture content and other physical quantities

    J. Nanjing Instit. Forestry Ind.

    (1983)
  • J.S. Luc et al.

    Detection technology and industrial application of wood acoustics-ultrasonic classification

    Wood Ind.

    (2007)
  • R.J. Ross et al.

    Relationship between log and lumber modulus of elasticity

    Forest Prod. J.

    (1997)
  • M.E. Schafer

    Ultrasound for defect detection and grading in wood and lumber

    Proc. IEEE Ultrasonics Sympos.

    (2000)
  • Y.Y. Nakayama

    Nondestructive testing of wood beams caused by vibration method

    J. Jpn Wood Res. Soc.

    (1975)
  • N. Sobue

    Simultaneous determination of Young's modulus and shear modulus of structural lumber by complex vibrations of bending and twisting

    J. Jpn Wood Res. Soc.

    (1988)
  • D. Ous

    Vibrational and acoustical experiments on logs of spruce

    Wood Sci. Technol.

    (1999)
  • F.M. Joseph

    Transverse vibration of a simply supported beam with symmetric overhang of arbitrary length

    J. Test. Eval.

    (1997)
  • A. Kamaguchi et al.

    Measurement of Heartwood moisture content of Sugi by the lateral impact vibration method Comparison with data obtained from the moisture meter method and measurement of longitudinal distribution

    J. Jpn Wood Res. Soc.

    (2001)
  • T. Mario et al.

    Determination of the vibration of the test method of timber for musical instruments

    J. Jpn Wood Res. Soc.

    (1985)
  • J.L. Sandoz, Nondestructive evaluation of building timber by ultrasound, in: Proceedings, Eighth International...
  • L. Diebold et al.

    Machine grading of structural sawn timber from various softwood and hardwood species

  • Cited by (15)

    • Development of an acoustic method for wood disease assessment

      2021, Computers and Electronics in Agriculture
      Citation Excerpt :

      This would require a sensor that quickly and reliably detects the presence of diseases, regardless of its age and size. With the current wood quality characterization methodologies (Qin et al., 2018), it is possible to obtain the speed of sound, and with it, the elastic modulus of the wood (Olaoye et al., 2019), which is linked to its quality, but not to its degree of dehydration. Although the principles used are similar to the characterization system for boards, the proposed sensor responds to a problem common to commercial solutions: all the trees in a forest are different from each other.

    • Screening for internal defects of timber members based on impact evaluation method

      2020, Construction and Building Materials
      Citation Excerpt :

      Hosainpour et al. developed an automatic separation system for potato harvesters and achieved a detection accuracy of 97% for potatoes and clods [21]. Another method based on sound speed measurements can evaluate whether a knot is in the beam or not; and the resonance method performed better in evaluating the size of the knots [22]. Although numerous researches have proven the feasibility of the IE method, little attention has been paid to a screening method for internal defects of timber members as well as the standards for this method.

    View all citing articles on Scopus
    View full text