Progress 07/01/12 to 06/30/17
Outputs
Target Audience:This projected started and covered numerical modeling of cutting of wood and wood products. These types of numerical tools are best used for running computer experiments that help one gain insights into cutting mechanics. The target audiences would be companies involved in wood machining, veneer peeling, or design of tools used for wood cutting. One published paper had recommendations for new areas of work for optimizing jack planes used in wood working. The numerical tools provide more detailed information than is possible by trial an error design or by reliance of folklore of cutting tools. In working towards cutting, the project branched out into damage mechanics in anisotropic materials. The target audience for these tools are any researcher or company with concerns about material failure issues. The target audiences are reached by interactions with industry (at Oregon State's Wood Based Composite center), publication of papers, and presentations at conferences. This project work on cutting and damage mechanics reached several industries and was presented at several conferences and in publications. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?One graduate student wasinvolved to developing numerical methods for enhanced stability and noise reduction and modeling of wood using damage mechanics (this student was funded by a different USDA project). How have the results been disseminated to communities of interest?Two publications on cutting and one on XPIC have appeared. The damage mechanics and is relation to fracture modeling was presented at an international conference in Switzerland. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Wood and wood-based composites are very complex materials. Their properties are anisotropic (different in grain direction compared to the transverse direction) and they can become damaged and fail by numerous mechanisms (tensile or shear failure parallel and perpendicular to the grain direction). Understanding this complexity by experiment results is challenging because natural various from specimen to specimen often lead to ambiguous observations. As computers are getting more and more powerful, we now have the opportunity to advance knowledge about many aspects of wood by computer simulations and numerical experiments. One theme of this work was to develop numerical simulations for cutting of wood. A computer package was developed (and available to the public) that can simulate cutting of wood. The modeling accounts for anisotropy of wood properties. Applications of this new tool are to optimize industrial processes that cut wood (e.g., veneer peeling) or to help design new tools for improved cutting of wood (e.g., wood planing). Modern approaches to cutting of wood model it as a fracture process whereby the tool generates a crack that propagates through the wood and is driven by motion of the tool. Working along this lines, this project also explored anisotropic damage mechanics as a tool for modeling wood failure. Some examples considered in wood are failure as a function of grain angle and cracking of layers in wood composites. For example, cross-laminated timber is an emerging mass-timber product finding applications in modest size buildings. Unfortunately, it is very prove to cracking of layers when exposed to changes in temperature and humidity (such as seen by building during seasonal variations). Both theory and numerical modeling has revealed that cracking is caused by use of thick layers in a panels with cross-grained wood bonds. The computer methods could be used to redesign cross-laminated timber to avoid cracking. The first recommendations output from modeling are to use more and thinner layers and to sufficiently dry the timber before fabricating the panels. The anisotropic damage mechanics may provide a tool for predicting cracking and delaminations and help guide new products with much better durability. All computer modeling, especially in complex problems, is prone to noise and numerical stability issues. All simulations in this project used a new computer tool called the material point method (MPM). During the project, we developed a new form of MPM called XPIC for eXtended Particle In Cell method. This method provides significant enhancement of stability in MPM and reduces noise. It was essential for getting the cutting simulations to be stable and is normally required for stable damage mechanics simulations as well. The XPIC methods were recently extended to enhance modeling of temperature conduction and moisture diffusion coupled to mechanical loading of wood products. These combined tools should find applications in predicting long-term durability of wood and wood-based composites.
Publications
- Type:
Websites
Status:
Published
Year Published:
2017
Citation:
J. A. Nairn, "Documentation for use of NairnMPM and NairnFEA Computational Mechanics Software,"
http://osupdocs.forestry.oregonstate.edu/index.php/Main_Page. (Software developed for cutting simulations is available to
the public through this web site. This product indicates the package was updated to latest version during the reporting period)
|
Progress 10/01/15 to 09/30/16
Outputs
Target Audience:The numerical modeling methods developed are targeted at cutting of wood, but the methods and mechanics apply to cutting of all materials. The wood target audiences on specifically wood processing industry and any wood industry involded in cutting. For example, some recent discussions have been held with manufacturers of chain saws. More generally, the numerical methods apply to cutting of materials such as plastics and metals and are most useful to alternative numerical methods have not worked well. I try to reach target audiences with publications, attendence at meetings, and individual discussions. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided? A graduate student has been involved to developing numerical methods for enhanced stability and noise reduction and for better solving the equations for anisotropic failure (although the student was not funded by this project). How have the results been disseminated to communities of interest?A publication on cutting of wood appeared this year in the journal Interface Focus. What do you plan to do during the next reporting period to accomplish the goals?I anticipate use of the material point method (MPM) and new anisotropic failure modeling for more simulaitons of cutting. In particular, the enhanced stability methods (known as XPIC) may allow new simultions that can probe the effect of tool sharpness.
Impacts
What was accomplished under these goals?
Prior work on this project has developed a new numerical methods for simulations of cutting. The new method is based on the particle method called the material point method or MPM. The recent accomplishments have applied this new tool to cutting of wood and focused on the challenges of dynamic fracture modeling and on methods to realistically handle anisotropic mechanical and failure properties of wood. First, we derived a new method to enhance stability and reduce noise in MPM simulations. These simulations will actually apply to all MPM simulations, but cutting simulations are an example of a problem that especially needs stable calculations. Our initial cutting simulations (see published papers) used a very stable MPM update method known as PIC (for Particle In Cell method). Although this greatly improved stability of cutting simulations, it is known to dissipate energy (and we evaluated its damping effects in paper from this project). This year, we derived a new update method termed XPIC (for eXtended PIC). This methodenhances stability and reduces noise but minimizes unwanted dissipation of energy. One of the paper's reviewers described it is perhaps the most significant advance to MPM in the past 15 years. I am now in the process of using XPIC methods for cutting simulations. Other accomplishments are dealing with anisotropic properties of wood. My cutting paper published this year modeled wood using the anisotropic Hill failure criterion. This year we have worked on improved numerical implementation of this approach for greater accuracy and improved efficiency. For an alternate approach, we have worked on using damage mechanics to model initiation and propagation of damage. Unfortunately, most literature models for damage mechanics deal only with isotropic materials and assume the material remains isotropic after damage. Both the assumptions are poor for wood - wood starts out anisotropic and damage induces changes in those anisotropic properties and potential additional aniostropies. In work (funded by different USDA project), we derived anisotropic damage mechanics methods for isotropic materials. Recently, I have extended that work on anisotropic materials. This new damage mechanics will be next used for simulations of cutting in this project.
Publications
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2015
Citation:
J. A. Nairn, "Numerical Simulation of Orthogonal Cutting: Application to Woodworking with a Bench Plane" presented at Cutting Science in Biology and Engineering, October 26-27, 2015, Chicheley Hall, UK.
- Type:
Journal Articles
Status:
Published
Year Published:
2016
Citation:
J. A. Nairn, "Numerical Modeling of Orthogonal Cutting: Application to Woodworking with a Bench Plane," Interface Focus, 6(3), 20150110 (2016)
- Type:
Websites
Status:
Published
Year Published:
2016
Citation:
J. A. Nairn, "Documentation for use of NairnMPM and NairnFEA Computational Mechanics Software," http://osupdocs.forestry.oregonstate.edu/index.php/Main_Page. (Software developed for cutting simulations is available to the public through this web site).
|
Progress 10/01/14 to 09/30/15
Outputs
Target Audience:This project is on experiments and modeling of the cutting process in wood. The numerical modeling is aimed at developing new tools for optimizing cutting methods in a variety of wood processes. The two target audiences are the numerical modeling community (for all the new modeling methods being developed) and the wood products industry (for applications). Some work this year has also fucused on tools, and therefore would have target audience of companies involved in wood machining or an manufacturing of wood tools (planes, chisels, saws, chain saws, etc.) Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?A graduate student has been involved to developing numerical methods for better solving the equations for anisotropic plasticity theory (although the student was not funded by this project). How have the results been disseminated to communities of interest?The most recent wood cutting experiments was presented at the "Cutting Science in Biology and Engineering" conference hosted by the Royal Society in the UK. That paper was expanded to a referred publication to follow the conference for a special issue of "Interface Focus." The numerical methods used for must of the wood cutting modeling, was the focus of a special topics course are Oregon State University. One student in that class did his class problem on cutting of wood for application in understanding experiments with chain saws. What do you plan to do during the next reporting period to accomplish the goals?I anticipate improved efficient in modeling wood failure and those methods will be turned to more cutting problems and to veneer peeling. This latter goal, especially, needs efficient failure modeling.
Impacts
What was accomplished under these goals?
The numerical modeling of wood was improved to account for failure. Two approaches are being used. The first was to use anisotropic plasticity theory (known as the Hill criterion). The failure is modeled as plastic yielding, but can represent different strengths of wood in different loading directions. For numerical modeling, it is important to recognize some significant difference between methods used for plasticity of isotropic materials and extra methods needed when dealing with anisotropic materials. I have implemented the proper methods now for anisotropic wood. Although they work, they are inefficient. Some new efforts are looking at an alternative numerical solution to partition elastic and plastic deformation on each time step (which requires numerical solution with 7 unknowns). The initial results are encouraging and hopefully will produce more efficient code. An alternative approach to modeling wood behavior is to use "softening" materials with different softening laws depending on the damage direction in the material. The softening approach should be more efficient and many be more realistic description of wood. I have worked out a new theory for softening materials (needed because much of the literature is vague and most of it reverts to isotropic damage when actually implemented). The new methods are partially implemented (for anisotropic failure in isotropic materials). The next stage will implement them for wood (for anisotropic failure in anisotropic materials). The first failure modeling (the plasticity method) was used for many simulations for planing of Douglas fir with a wood working bench plane. I looked at effects of rake angle, chip breaker location, mouth opening in the sole of the plane, friction on tool surfaces, and sharpness of the tool. All of this issues can now be studied by numerical modeling. The first results were present an conference on cutting science and will appear in a special issue of Interface Focus (in 2016).
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2015
Citation:
J.A. Nairn, "Numerical Simulation of Orthogonal Cutting using the Material Point Method," Eng. Fract. Mech., 149, 262-275 (2015)
- Type:
Journal Articles
Status:
Accepted
Year Published:
2016
Citation:
J.A. Nairn, "Numerical Modeling of Orthogonal Cutting: Application to Woodworking with a Bench Plane," Interface Focus, in press (2016)
|
Progress 10/01/13 to 09/30/14
Outputs
Target Audience: This project is on experiments and modeling of the cutting process in wood. The numerical modeling is aimed at developing new tools for optimizing cutting methods in a variety of wood processes. The target audiences are the numerical modeling community (for all the new modeling methods being developed), the wood research community (for modeling methods specific to wood), and the wood products industry (for applications). Three examples industries are veneer peeling, chain sawing, and wood chipping. We have initiated discussions with Blunt Industries in Portland, Oregon, who are a major supplied of chain saw blades. We have started simulations on wood chipping with Forrest Concepts in Seattle, Oregon. Changes/Problems: The current numerical simulations are working well. The new damping method has solved almost all issues. The remaining issues are simulations of very thin cuts and simulations of cutting with blunt cutting tools. These problems do not require a change in the project, but will require a breakthrough to solve certain types of cutting geometries. A series of undergraduate students had problems using our experimental cutting apparatus. This apparatus was developed by a former graduate student, but not left in a state for easy use by other students. Just recently we seemed to have solved these problems and experiments are underway. What opportunities for training and professional development has the project provided? Several under graduate students have worked with our cutting apparatus and worked towards running new cutting experiments. The experimental methods have proved challenging. Although all students received significant lab training, several students were unable to complete experiments before their training period ended. The current student, however, has succeeded in getting everything running and is collecting results for cutting of various wood specials in various directions with respect to wood grain. We will undertake simulations of these experiments when they are completed. The NairnMPM wiki page (http://osupdocs.forestry.oregonstate.edu/index.php/Main_Page) is available for anyone interested in self-guided training on numerical modeling tools capable of simulating cutting. We have ported most feature used for current cutting simulations from OSParticulas (our in-house code) to the available, open-source NairnMPM package. How have the results been disseminated to communities of interest? The recent modeling results were presented at the European Structural Integrity Society (ESIS) conference in Switzerland in 2014. We followed up this conference with a publication that has been submitted to Engineering Fracture Mechanics. We have started discussions with a local manufacture of chain saw blades on whether or not numerical simulations can help them develop better blades. We have started simulations with Forest Concepts of a wood chipping process. What do you plan to do during the next reporting period to accomplish the goals? The next period will include continued development of the modeling methods and application to more problems in wood cutting. In addition, once the undergraduate student completes wood cutting experiments, we will interpret the results with cutting simulations.
Impacts
What was accomplished under these goals?
The numerical modeling of cutting is particularly complex because it combines multiple physical processes (fracture, yielding, friction, large deformations, dynamics, and more). In previous reports, we mentioned this work has made significant advances by implementing large-deformation, elastic plastic material models (for the material being cut) and developing a new scheme for generating output of cutting forces. These new feature greatly improved the simulations, but the overall simulation still struggled with instabilities and noise due to dynamic effects. In this reporting period, we made a major breakthrough by developing a new of damping for MPM simulations that appears to work especially well for all problems involving crack propagation (such as cutting). This new scheme was motived by recent use of MPM at Disney Animation Studios to animate snow mechanics in the movie "Frozen". They suggested a modified extrapolation method when updating particle velocity based on combined extrapolation of both velocity and acceleration. Their use, however, was incomplete. This project developed their method and demonstrated that in corresponds to an artificial (and useful) damping mechanism. Furthermore, it is not enough to change just the update to particle velocity - the new updating method needs to modify the update of particle position as well. Our new damping scheme is documented in a publication where we showed in results in cutting simulations that are always stable and nearly free of noise. We used the scheme to studying cutting mechanics and to compare to analytical models in the literature. We then moved beyond the capabilities of analytical models to consider cutting problems with wood and the role friction and cutting geometry (e.g., when planeing wood). Out now-accurate, cutting simulations have modeled the metal tool as a rigid material. To move on to important issues such as tool sharpness and tool wear, we needed to replace the rigid material with real materials, but that is a challenge for all numerical methods with explicit cracks. To solve this problem, we developed a new MPM scheme whereby any material can be contained within a crack. This option allows simulations such as cutting with a metal tool, driving nails or screws into wood, and much more. It extends MPM to an entirely new field of numerical simulations that is difficult to approach by other numerical methods. We have developed 3D simulations of a wood chipping process where wood veneers are fed into a row of cutters that look much like a paper shredding machine. The work has just started in collaboration with Forest Concepts in Seattle, Washington.
Publications
- Type:
Journal Articles
Status:
Submitted
Year Published:
2015
Citation:
J.A. Nairn, "Numerical Simulation of Orthogonal Cutting using the Material Point Method," Eng. Fract. Mech., submitted (2015)
|
Progress 01/01/13 to 09/30/13
Outputs
Target Audience: This project is on experiments and modeling of the cutting process in wood. The numerical modeling is aimed at developing new tools for optimizing cutting methods in a variety of wood processes. The two target audiences are the numerical modeling community (for all the new modeling methods being developed) and the wood products industry (for applications). Two example industries are veneer peeling and chain sawing. We have initiated discussions with Blunt Industries in Portland, Oregon, who are a major supplied of chain saw blades. Changes/Problems: The numerical simulations of cutting typcially look at a cutting geometry and run simulations as a function of depth of cut. We are getting good results for many depths of cuts (and fortunately for most depths of cut of interest to the wood products field), but more detailed validation of the model would be better if the simulations could be continued to thinner depths of cuts. Currently the simulations either fail or get too long when the depth of cut gets small. We are working on alternative modeling methods to handle a wider range of cut depths. This problem does not require a change in the project, but will require a breakthrough to solve. A recent power failure caused a malfunction in our testing equipment thaat ended up destroying a key triaxial load cell. We are in the process of replacing the load cell and will use it in experiments during the next reporting period. What opportunities for training and professional development has the project provided? An undergraduate student was hired to run cutting experiments. His work will be in the next reporting period. The new software wiki (http://osupdocs.forestry.oregonstate.edu/index.php/Main_Page) is available for anyone interested in self-guided train on numerical modeling tools capable of simulating cutting. How have the results been disseminated to communities of interest? The recent modeling results will be present at ESIS conference in Switzerland in 2014. We have started discussion with a local manufacture of chain saw blades on whether or not numerical simulations can help them develop better blades. What do you plan to do during the next reporting period to accomplish the goals? The next period will include continued development of the modeling methods. In addition, an undergraduate student will collect experimental result on cutting of wood to compare to modeling predictions.
Impacts
What was accomplished under these goals?
The numerical modeling of cutting is particularly complex because it combines multiple physical processes (fracture, yielding, friction, large deformations, dynamics, and more). In the reporting period we made some significant advances by implementing large-deformation, elastic plastic material models (for the material being cut) and developed a new scheme for generating output of cutting forces (which was a drawback in previous cutting models). The large deformation model has greatly improved accuracy as the chip curls during the cutting process. The output forces are now giving quantitative results that we are comparing to analytical models. In my opinion, the computer models are now much more reliable and realistic then prior mathematical models. The new modeling methods were submitted (and accepted) for presentation of a talk on cutting mechanics for a conference in Switzerland in 2014 (ESIS conference of deformation and fracture of materials).
Publications
|
Progress 07/01/12 to 12/31/12
Outputs
OUTPUTS: The goal of this project is to develop numerical modeling methods for orthogonal cutting of wood with applications to veneer peeling and general machining methods for wood and wood products. The problem is a "grand challenge" comprised of multiple physical phenomenon that all need to be included. Our approach is based on recent advances in cutting theory that incorporates fracture analysis into the cut plane process. Older theories assumed it was all determined by plasticity, but older theories had several inconsistencies compared to experiments, especially for wood, which behaves very different from metals plasticity used in prior cutting models. The key elements of the model are material elastic and plastic properties, contact between the knife and wood, friction at the contact point, and fracture properties in the curt direction. We have successfully incorporated all these effects in a material point method (MPM) simulation using custom MPM computational mechanics code. We have also completed numerous cutting experiments on wood and wood plastic composites that will be available for interpretation by modeling methods. Our experiments used an instrumented knife and automated data acquisition system that records normal and tangential force during cutting. When cutting wood transverse to the grain direction, which mimics veneer peeling cutting, we could observe transient peaks in force during the process. These peaks could by directly attributed to formation of transverse cracks or checks in the layer being cut from the wood. In other words, this lab method reproduces lathe checking in an instrument that can monitor the entire process. We plan to include these experimental observations into modeling capabilities to then developing numerical simulation of lathe checking. PARTICIPANTS: John A. Nairn (PD) runs the project leads the project, hires and trains and students in the project, develoops the numerical modeling software, and prepares publications and presentation. Kalin Semrick recently finished is MS in Wood Science & Engineering working or othogonal cutting of wood plastic composites. During the past year he rans new experiments on transverse cutting of wood. His thesis experiments and his wood experiments will be important validation experiments for the numerical modeling work. TARGET AUDIENCES: Wood products and wood-manufacturing industries, other industrials the make use of cutting, academic researchers involved in material point method or other computation mechanics method. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
The outcome of this project will be computer simulation tools for modeling for orthogonal cutting of wood and wood products including veneer peeling. The code development is being done in a public domain computational mechanics software written by PI of this project called nairn-fea-mpm. The latest cutting simulation methods are now in this project. The project is available at http://code.google.com/p/nairn-mpm-fea/. At the beginning of this project, a former student who worked on cutting did new experiments on wood and on wood plastic composites. His work is written in a thesis and a draft paper is in preparation. The experiments that reproduce lathe checking are documented and ready for analysis. A major outcome this year was incorporation of a large-deformation, hyperelastic-plastic material model and this material is leading to improved simulations. It is particularly good at tracking large strain in the chip as in curls and rotates after being cut from the main material. The current state of the simulation tools will be presented at an MPM workshop in Salt Lake City, UT in March 2013.
Publications
- No publications reported this period
|
|