Non-destructive Evaluation Applied In Wood Research

Non-destructive evaluation (NDE) is becoming increasingly important as a means of maximising and optimising the value (economic, engineering, utilitarian, etc.) of every tree that comes from the forest. Much of the work has been completed in conjunction with the U.S. Department of Agriculture, Forest Service, Forest Products Laboratory as well as the Agricultural Research Service with the overall intent of improving lumber and wood products standards and valuation. Thus, the future impact of NDE needs to be strengthened. By R Dan Seale, Rubin Shmulsky and Frederico Jose Nistal Franca, Forest and Wildlife Research Center, Mississippi State University

In an ongoing effort to enhance forest sustainability and utilisation, non-destructive evaluation (NDE) research is permeating into an increasing number of applications. At this stage, this type of research is often more applied than basic. 

That is, it seeks to solve current and pressing real-world problems at hand. A combination of federal-level (mainly through the US Department of Agriculture and Department of Interior), state-level (through the state of Mississippi), and commercial/industrial interest and funding is supporting NDE research. 

This manuscript provides a review of the research stemming from these partnerships. The purpose of the manuscript is to pull together and summarise the findings of this targeted NDE work to help scientists and practitioners increase, improve, and adopt these types of technologies.

In short, NDE seeks to predict how stiff or strong an individual member will be. While the ultimate strength of solid sawn lumber is more variable than that of composites, the prediction of each mechanical property can be improved with the application of NDE. 

As basic forms of NDE are well-described elsewhere, they will be mentioned and discussed herein with the assumption that the reader has some level of familiarity with their basic underpinnings.

The research shown below highlights a broad swath of NDE development and application. Most relate to new construction. Some relates to existing products in service or in-situ. Both have implications for improving forest sustainability. 

The longer the time any given product or structure stays in service, the fewer hectares of timber need to be cut in a given year. As the accuracy of predicted strength increases, more homes can be built and the basic needs of an increasing population can be met from the same timberland area. That type of philosophy provides great hope toward meeting the home and shelter needs of an ever-increasing global population.

The research highlighted herein has occurred as part of ongoing partnerships between federal, state, and private interests. This three-legged stool approach provides great stability and a high degree of accountability. Industrial and commercial stakeholder input has been gleaned throughout. 

Their input and guidance has helped steer the research toward the most pressing needs. The federal and state input and interest has assured that the work maintains broad interest and influence across the state, region, and national levels.


Fundamental of NDE

Fundamentally, the interrelationships among specific gravity, stiffness, and strength are largely at the heart of NDE. As a means of monitoring mill production, quality, and timber resources over time routine stiffness evaluation often provides the best “reasonable” indicator.

Work with NDE tools at Mississippi State University accelerated in 2004 with the acquisition of a Metriguard E-Computer and a Fiber Gen HM200. The impetus for this change arose for the need to grade a pilot-scale, high-strength and stiffness-engineered wood product that had no surface defects. 

There, NDE was used to evaluate both the finished product and the raw materials. While encouraging, these results were not sufficient to adequately grade the engineered product. Ultimately, an X-ray technology was used in conjunction with the NDE to determine on-grade product. 

The X-ray could spot low-density zones which impacted modulus of rupture (MOR) and the NDE tools accurately predicted static bending stiffness. Basically, the X-ray technology worked as an analog to knot allowances in solid sawn machine stress rated (MSR) and machine evaluated lumber (MEL). That work resulted in a report issued by APA The Engineered Wood Association in 2007.

Idealising the MOE and MOR relationships among perfectly homogeneous materials is routine. Applying these relationships to actual small clear specimens adds a level of variability. Then, applying these to full-size pieces of lumber adds an increasing amount of variability. 

To glean a better understanding of how NDE can be best applied to grading or performance assessment of graded lumber at a mill level, it seems appropriate to investigate NDE at the mill level prior to lumber grading. To this end, researchers have sampled structural lumber from varying production facilities. 

At a fundamental level, Anderson et al. found that MOE and/or MOR may change at a given mill over time. This finding can perhaps be somewhat explained by raw material resource changes throughout the year. At wetter times of the year, logs are taken from higher and drier ground. During the drier summertime, loggers can pull logs from what might otherwise be wetter bottomland.


Dimension Of The Lumber

In work by Dahlen et al. both southern pine and Douglas fir were sampled. In that case, the MOE and MOR were correlated in 2 × 4 lumber from six pine mills (from the states of Alabama, Arkansas, Georgia, Mississippi, and Texas) along with six Douglas fir mills (from Washington, Idaho, Oregon, and Canada) that were sampled. 

Neither of these was considered a production weighted sample however the geographical representation was widely reaching. All specimens were testing in bending per ASTM D198. In sum, 744 pine specimens were considered and the MOE to MOR r2 value (adjusted to 15 percent MC) was 0.52. Similarly, 733 Douglas fir specimens were considered and the MOE to MOR r2 value (adjusted to 15 percent MC) was 0.66.

Additional research on Douglas-fir and southern pine 2 × 4 s by Dahlen et al. showed great variability among mills with respect to MOR variation and MOE vs. MOR correlations. In each case, variations were statistically significant at the α = 0.05 level. These findings highlight the conservatism in developing global design values for an individual species. It also provides an impetus for implementing NDE, such as MSR or MEL, as a means of capturing the otherwise lost utility value of stiffer and stronger material at sawmills that convert high-quality timber resources.

Generally, it has been observed that mills with a wider range of raw materials, such as those mills that run both small logs and relatively large logs, see better MOR to MOE correlations. This finding seems to be because they produce a lumber with a wider range of density and a wider range of MOE. 

Often, this manifests itself as 2 × 4 and 2 × 6-inch lumber being manufactured from both the juvenile-wood centre of small diameter trees (relatively weaker properties) and from the outside (jacket boards) of larger logs (relatively higher properties). 

This factor typically leads to a wider range of MOR values and thus their respective r2 values increase. These findings suggest that if a given mill wishes to investigate NDE as a means of capturing utility value, which mill should to first evaluate their particular resource and if implemented, that mill will need to dial-in the performance of their equipment and routinely calibrate it. With respect to strength distribution, the wide variation in properties between juvenile wood vs. jacket board lumber often makes the 2 × 4-inch and occasionally the 2 × 6-inch sizes appear bimodal.


Roundwood Dowels

Climate change appears to be causing greater variation in extreme weather events. In the U.S., the Gulf South region is heavily timbered. It is also highly susceptible to tropical storms and hurricanes stemming from or passing through the relatively warm Gulf of Mexico. 

As a result, wind or storm damaged timber is not uncommon. Fully damaged, broken, and twisted tree stems are rarely salvaged into usable logs. The costs and risks associated with getting them to the mill in a timely manner are high. 

That said, high wind events often only partially damage wood or forest tracts. Sometimes, high wind events partially wind throw trees. The result (post-storm) may be fully upright or partially leaning trees, forest stands, or vast forests. While visually “normal” there may be inherent damage (in the form of ring shake or mild to moderate timber break). 

In some cases, these may only become visible after sawing, after drying and planing, during peeling wherein the core separates from the log, worse yet after being put into service which invariably causes expensive claims and seemingly unnecessary consternation. 

Following hurricane Katrina (late summer 2005) a storm which damaged thousands of hectares of timberland research by Slay et al. investigated the ability to use acoustic velocity to assess non-visible damage in small round stems that had been turned down to 4-inch diameter dowels. 

Dowels were selected for their uniform section and low potential for cross grain from end to end. There, the acoustic velocity in green wood dowels was measured and then the dowels were stressed in bending. This process was repeated at increasing stress increments. 

Of particular interest was the ability to use NDE to detect if each dowel had been stressed beyond the proportional limit and was thereby permanently weakened. The rationale was that if NDE could be employed in this manner, then a given processing facility could measure incoming raw materials and either deduct value as appropriate or merchandise damaged logs more appropriately.

In related work, Shmulsky and Snow investigated the interrelationships of MOE, acoustic velocity, rings per inch (a surrogate for density) and MOR on 5-inch diameter pine dowels. 

In this work as well, dowels were chosen as their uniform section simplifies analysis while also maintaining a high degree of straight grain throughout the length. There, the combination of acoustic velocity plus rings per inch were strong predictors of MOE (r2 values of 0.72 (green wood) and 0.76 (dry wood)). 

The best predictions for MOR used either acoustic velocity plus MOE (r2 values of 0.45 (green wood) and 0.51 (dry wood)) or acoustic velocity, rings per inch, and MOE (r2 values of 0.50 (green wood) and 0.53 (dry wood)). Given the added complexity of using three predictors versus two, the combination of acoustic velocity and MOE seemed like the most favorable choice of predictors.


Utility Poles And Crossarms Of Wood

Wood utility poles remain the lowest cost solution for distributing electric power and utilities throughout the U.S. While other materials are used extensively, particularly in specialty applications, wood poles with their low cost, wide availability, and 30–50+ year life, remain the material of choice. 

Around two million new poles (either as new construction or line-rebuilding) are put into service each year. If one estimates 125 pole-class stems per acre then one can quickly surmise that approximately 16,000 hectares of land are required to grow these poles. 

At a 40-year rotation, one can project that 640,000 hectares of land (about the size of the state of Delaware) are continually associated with growing pole class stems. Thus, anything one can do to extend their service life relieves the pressure on this land area. To assess novel technology intended for in situ assessment of wood utility poles, a study of 50 specimens was developed by Seale et al.. 

There, during routine 8–10-year infrastructure inspection, approximately 200 poles were selected for removal and replacement. Among these, 50 poles were identified for further study. These poles were tested via NDE in the field with the novel technology, removed from service, brought to Mississippi State University, tested via NDE again, and then tested to failure per ASTM 1036. 

Of these 50 specimens, 17 were reinstalled in the ground, the ground was compacted, and they were tested in an upright orientation. A total 33 of the specimens were tested horizontally in a dedicated utility pole testing fixture. 

Among all 50 poles, the r2 value for the actual breaking strength vs. the predicted value was 0.56. This value is similar to that commonly observed with dimension lumber during manufacturing. For the 17 specimens that tested in the upright orientation (installed in the ground), the r2 value between actual vs. predicted breaking force was 0.73. This finding suggested that the NDE technology showed great promise in potentially evaluating in situ wood utility poles during their requisite routine inspections.

Wood utility crossarms are produced to a national standard, ANSI O5.3. Based on a series of visual standards, these specialised industrial products either make the grade or are culled. There is a need for higher capacity arms in certain high load situations such as end-of-lines, generally as distributions circuit capacity is increased and as the electrical grid is hardened to improve resilience. 

Work by Catchot et al. evaluated both Douglas fir and southern pine cross arms. There, manufactured cross arms of these two species were measured via varying NDE technology and then destructively evaluated. Both acoustic velocity and longitudinal vibration most accurately predicted MOR and MOE. Results also indicated that these technologies could be used to identify candidate stock for a premium type grade that would have superior mechanical properties as compared to the general on-grade population.


Smartphone Application For NDE

To push NDE to the consumer level, researchers have developed a smartphone application that calculates lumber MOE. This work describes the development and accuracy of a program that uses either the smartphone microphone or its accelerometer to calculate the MOE of solid materials. 

While not robust and fast enough for the production setting, it is useful for builders and building contractors to assess their lumber particularly when trying to choose pieces for beams and headers. 

Furthermore, technology such as this can be helpful for assessing material performance over time, such as in the case of scaffold planking. Related work by Han et al. investigated market acceptance and interest in this type of smartphone application.


Research Conclusion On Different Kinds Of Woods

For investigation on hardwood lumber, it is widely used for flooring, stair systems, rail and guard systems, and others. In some of these cases it necessarily provides structural capacity. The building code(s) in the US require specified strength and stiffness performance levels for structures and their various sub systems, such as stairs and guards. 

To meet these requirements, building products must have publically available bending strength and stiffness values. In the case of the grades, sizes, and species most often used in stair and guard systems, these mechanical properties are not readily available. 

As part of a study to investigate potential changes among red oak, white oak, hard maple, and yellow poplar lumber in these applications, NDE-related findings are reported in Turkot et al. This work is critical toward maintaining, and potentially growing, the markets for U.S. hardwoods that are to be used in load bearing applications.

For investigation on the engineered wood, Yang et al. describe the production and mechanical properties of a novel type of engineered lumber that incorporated machine stress rated lumber stock at the extreme edges of structural beams. 

There, the machine stress-rated lumber, when applied to the extreme edges of beams, greatly improved the design bending strength of lower quality (number 3 grade) lumber. 

Further work related to NDE of cross laminated timber is ongoing at Mississippi State University. This work is geared to in plant or in-field assessment of bondline quality. 

This type of information is critically important for the quality control and quality assurance related to mass timber which ultimately minimises its risk for failure and maximises its uniformity.


Significance Of NDE

NDE has gained wider and wider commercial acceptance during the past three decades and research at Mississippi State University is pushing the technology toward increased adoption and application. There are a wide variety of NDE technologies and a great many ways in which they can be applied to real-world production or in-situ products and structures as a means of improving product valuation and structural assessment. It is anticipated that the coming decades will see:

•Greater use of NDE in mass timber/cross laminated timber production;

•Increased use by saw mills and other structural lumber producers;

•Improved means of identifying strength and stiffness reducing characteristics;

•Potential adoption of automated visual grading systems as candidates for producing machine stress rated lumber and machine evaluated lumber;

•Novel engineered composites that incorporate NDE in their manufacturing quality control and assurance;

•Additional field-based devices that allow contractors, engineers, builders, and others perform some level of NDE on wood members and structures either at the time of (or post) construction; and

•Development of wave analysis techniques to improve trim saw solutions.

Rate this item
(0 votes)
  • Last modified on Thursday, 28 January 2021 09:29
  • font size

FDMAsia About Us

For almost 30 years, FDM Asia has been the publication of choice for woodworking professionals in Asia Pacific, providing the latest news and expert insights of a diverse range of topics including process technologies, furniture production, panels manufacturing, raw materials handling and sustainability issues.

We have one of the most comprehensive woodworking database and e-media programme to assist you in your e-marketing and give you an exclusive opportunity to connect to our 25,152 online subscribers.