UBC Undergraduate Research

Compression Kiln Research & Development FOR: Forest Enterprises Australia Ltd. Fairclough, Duncan T. Apr 14, 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52966-FaircloughDuncan_WOOD_493_Graduating_Essay_2008.pdf [ 4.26MB ]
Metadata
JSON: 52966-1.0103120.json
JSON-LD: 52966-1.0103120-ld.json
RDF/XML (Pretty): 52966-1.0103120-rdf.xml
RDF/JSON: 52966-1.0103120-rdf.json
Turtle: 52966-1.0103120-turtle.txt
N-Triples: 52966-1.0103120-rdf-ntriples.txt
Original Record: 52966-1.0103120-source.json
Full Text
52966-1.0103120-fulltext.txt
Citation
52966-1.0103120.ris

Full Text

Compression Kiln Research & Development FOR: Forest Enterprises Australia Ltd. Duncan T. Fairclough  WOOD 493 A Report Submitted in Partial Fulfillment of the Requirements for the Degree of Bachelor of Science in Wood Products Processing In The Faculty of Forestry  April 14, 2009  Abstract With growing competition in the timber processing industry in Australia, forestry companies are forced to look for new and innovative technologies to remain competitive. Forest Enterprises Australia (FEA) is dedicated to sourcing new equipment and technologies that can improve their various product lines whether it is in their plantations or their timber processing facilities. FEA’s timber division is specifically committed to always improving their product line with their in-house research and development department. Attempting to utilize the TeknoComp compression kiln is just another example of FEA’s commitment to finding innovative ways to produce timber. The focus of this study was to determine the feasibility of using the compression kiln. Prior to this study there had been some drying tests done on E. nitens in the compression kiln. The bulk of these tests were done at lower temperature ranging from 60°C to 100°C. The overall results from the testing were not very promising as much of the timber was not dry. When the timber was dry it usually contained severe collapse and checking both on the surface and internally. During these tests there was no way of tracking the drying conditions inside the kiln due to the software not allowing for data to be downloaded. For this study the bulk of the E. nitens timber was dried at higher temperatures than that of the previous trials, temperatures ranged from 70°C to 140°C. For the study the software was changed so the data for the various kiln parameters could be downloaded. This allowed for kiln parameters such as dry bulb, wet bulb, dew point, and relative humidity to be tracked. From this it was determined that the kiln was capable of recreating similar drying conditions for charges with the same schedule. During the study it was possible to obtain acceptable values for moisture content in the timber however, the timber was still prone to many defects mostly collapse and checking. It was often found that even timber that had excellent surface quality still contained internal checking when the timber was cross sectioned. Overall the results for drying E. nitens from green were not promising. Three mainland species were also selected for drying in the compression kiln E. dunnii, E. saligna, and E. sieberi. The species were selected based on the hopes that their natural properties would achieve a better result. Drying schedules used for these species were very similar to those of E. nitens. When compared to the E. nitens boards there was some improvement in quality however E. dunnii and E. saligna were both prone to surface and internal checking as well as collapse. E. sieberi did show some very promising results even though it was only done in small quantities. It appeared that the species was much less prone to surface and internal checking as very little of these defects appeared in any of the test samples. Collapse in E. sieberi was almost nonexistent. The final species to be tested was pine. The results from this trial were promising as the final dried quality of the timber was quite good. Drying temperatures ranged from 100°C to 140°C with drying times of 15 to 24 hours. The main problem came from the cost of drying the pine in the compression kiln versus the value of the timber as the pine that was Compression Kiln R&D  Duncan Fairclough ii  dried was a low valued product (100x38mm heartwood). Unfortunately due to time restrictions it was not possible to test larger section pine that could be sold as a high value product. This would be a good starting point if further trials of drying pine in the compression kiln were to be done. Throughout the study there were various difficulties running the kiln mainly due to equipment failure. Building the timber stacks proved to be quite difficult due to the Joulin stacking machine often dropping the timber. Also the hydraulic systems inside the kiln used to compress the stacks and drive the conveyors were in desperate need of maintenance causing difficulties during the study. There was also the issue of the kiln constantly tripping breakers causing the kiln to shutdown in the middle of a drying charge. If the kiln were to be put back into production thorough maintenance would need to be completed.  KEYWORDS Kiln Drying, Australia, Tasmania, Radiata Pine, Eucalyptus, Nitens, Dunnii, Saligna, Sieberi, Compression Kiln  Compression Kiln R&D  Duncan Fairclough iii  Acknowledgements I would like to thank the staff at Forest Enterprises Australia for their support during the course of the project. Without them this project would not have been possible.  A special thank you to Dr Trevor Innes for taking the time to answer my many questions and for providing me with information and advice when it was needed.  Compression Kiln R&D  Duncan Fairclough iv  Table of Contents ABSTRACT ...................................................................................................................... II ACKNOWLEDGEMENTS ........................................................................................... IV TABLE OF CONTENTS ................................................................................................ V 1.0  INTRODUCTION................................................................................................. 1  1.1 BACKGROUND ON WOOD BIOLOGY AND DRYING ...................................................... 1 1.2 BACKGROUND ON THE TEKNOCOMP COMPRESSION KILN ......................................... 8 1.3 COMPRESSION KILN ................................................................................................... 9 1.4 PROJECT OVERVIEW................................................................................................. 14 2.0  EUCALYPTUS NITENS TRIAL ...................................................................... 16  2.1 EUCALYPTUS NITENS ............................................................................................... 16 2.2 EUCALYPTUS NITENS BATCH SUMMARIES ............................................................... 18 2.2.1 Batch #1 Summary (E. nitens) ......................................................................... 18 2.2.2 Batch #3 Summary (E. nitens) ......................................................................... 19 2.2.3 Batch #6 Summary (E. nitens) ......................................................................... 21 2.2.4 Batch #7 Summary (E. nitens) ......................................................................... 23 2.2.5 Batch #8 Summary (E. nitens) ......................................................................... 27 2.2.6 Batch #9 Summary (E. nitens) ......................................................................... 30 2.2.7 Batch #10 Summary (E. nitens) ....................................................................... 33 2.2.8 Batch #11 Summary (E. nitens) ....................................................................... 36 2.2.9 Batch #12 Summary (E. nitens) ....................................................................... 39 2.2.10 Batch #15 Summary (E. nitens) ..................................................................... 42 2.2.11 Batch #18 Summary (E. nitens) ..................................................................... 45 2.3 EUCALYPTUS NITENS HARDNESS TESTING .............................................................. 49 2.4 TEMPERATURE DIFFERENCE VS. MC ........................................................................ 52 2.5 EUCALYPTUS NITENS TRIAL SUMMARY ................................................................... 54 2.5.1 Current E. nitens Trial Summary ..................................................................... 54 2.5.2 Previous E. nitens Trial Summary ................................................................... 56 3.0  E. DUNNII, E. SALIGNA, & E. SIEBERI TRIAL .......................................... 58  3.1 E. DUNNII, E. SALIGNA, & E. SIEBERI ......................................................................... 58 3.2 E. DUNNII, E. SALIGNA, & E. SIEBERI BATCH SUMMARIES ..................................... 60 3.2.1 Batch #7 Summary (E. dunnii & E. Saligna) ................................................... 60 3.2.2 Batch #8 Summary (E. dunnii) ......................................................................... 65 3.2.3 Batch #9 Summary (E. saligna) ....................................................................... 68 3.2.4 Batch #11 Summary (E. dunnii) ....................................................................... 71 3.2.5 Batch #18 Summary (E. dunnii) ....................................................................... 74 3.2.6 Batch #20 Summary (E. sieberi) ...................................................................... 78 3.2.7 Batch #21 Summary (E. dunnii) ....................................................................... 81 3.3 E. DUNNII, E. SALIGNA, & E. SIEBERI TRIAL SUMMARY ............................................... 82  Compression Kiln R&D  Duncan Fairclough v  4.0  PINE TRIAL ....................................................................................................... 85  4.1 RADIATA PINE .......................................................................................................... 85 4.2 RADIATA PINE BATCH SUMMARIES ......................................................................... 87 4.2.1Batch #1 Summary (Pine) ................................................................................. 87 4.2.2 Batch #2 Summary (Pine) ................................................................................ 88 4.2.3Batch #4 Summary (Pine) ................................................................................. 89 4.2.4 Batch #6 Summary (Pine) ................................................................................ 90 4.2.5 Batch #7 Summary (Pine) ................................................................................ 92 4.2.6 Batch #8 Summary (Pine) ................................................................................ 93 4.2.7 Batch #13 Summary (Pine) .............................................................................. 94 4.2.8 Batch #15 Summary (Pine) .............................................................................. 96 4.2.9 Batch #20 Summary (Pine) .............................................................................. 98 4.2.10 Batch #23 Summary (Pine) .......................................................................... 100 4.3 PINE STANDARD RACK & REVERSE TWIST TESTING .............................................. 101 4.3.1 Batch #25 Summary (Pine Standard Rack) ................................................... 101 4.3.2 Batch #26 Summary (Pine Wedge Rack) ....................................................... 105 4.4 PINE HARDNESS TESTING ....................................................................................... 107 4.5 PINE BENDING TESTS ............................................................................................. 110 4.6 PINE TRIAL SUMMARY ........................................................................................... 112 5.0  CONCLUSIONS, DIFFICULTIES, & RECOMMENDATIONS ............... 114  5.1JOULIN .................................................................................................................... 114 5.2 COMPRESSION KILN ............................................................................................... 117 5.3 GENERAL ............................................................................................................... 119 REFERENCES .............................................................................................................. 120 APPENDIX .................................................................................................................... 121  Compression Kiln R&D  Duncan Fairclough vi  1.0 Introduction 1.1 Background on Wood Biology and Drying The bulk of plantation hardwood in Australia has been planted with pulping in mind, however some hardwood sawmills are attempting to utilize these trees as sawn timber products. There are many difficulties in producing high quality sawn timber products from fast grown plantation eucalypt species, one of the major problem faced by hardwood sawmills is seasoning the timber with minimal degrade. For many years sawmills have recognized the problem of internal and surface checking in timber as well as collapse however there has been very little successful research done on avoiding these defects during the drying process.  To fully understand the problems that occur during the timber drying process there needs to be an understanding of wood at the microscopic level. Trees can be classed into two main categories hardwoods and softwoods. Softwoods are gymnosperms and hardwoods are angiosperms.  On a microscopic level hardwoods and softwoods have distinct cell structures, softwoods are primarily made up of one cell type (tracheids) while hardwoods contain many more cell types. Tracheids are the major cell type in softwoods (90%) and run vertically through the tree stem. The two main functions of theses cells in the movement of sap in the tree as well as providing mechanical support of the tree stem. The tracheids walls form the bulk of the solid wood structure in softwoods.  The middle lamella is a  cementing structure found between the various cell units. The second type of cell found in softwoods are wood rays (WR) which run perpendicular (radial direction) to the tracheids. The main function of these cells is to store and horizontally transport food substances in t the tree. Simple pits (SP) and border pits (BP) are holes in the various cells and are used to transfer sap between the rays and tracheids (USDA).  Compression Kiln R&D  Duncan Fairclough 1  Figure 1.1: Softwood cell structure (USDA). Hardwoods tend to have a more complex cell structure than softwoods. There are two major types of vertical cells in hardwoods the larger ones are called vessels and they are used to moves sap vertically through the tree. On the end grain the vessels appear as small holes or pores (P). The vessels are set one on top of the other and continue as open tubes for relatively long distances. The smaller vertical cells are called fibres (F) and they provide the mechanical support for the tree stem. As with softwoods there are wood rays (WR) present that run perpendicular (radial direction) to the vessels and fibres and are used to store and transport food in the horizontal direction. Again the middle lamella layer is used as a cement to hold the cells together. When hardwoods are pulped the middle lamella is removed using chemicals which allow the fibres to be separated (USDA).  Compression Kiln R&D  Duncan Fairclough 2  Figure 1.2: Hardwood cell structure (USDA). All wood cells are made up of three main chemical components, cellulose, hemicellulose and lignin. Cellulose can be considered the framework that provided reinforcement and tensile strength, lignin the encrusting substance which provides compressive strength and hemicellulose as the matrix which provides a link between the cellulose and lignin.  Wood cell walls are made up of two different cell walls; the primary cell wall and the secondary cell wall. The primary cell wall is much thinner than the secondary cell wall which consists of three layers; the S1, S2, and S3 layers. Within each of these layers microfibrils are oriented at various angles. Microfibrils are cellulose molecules that are bundled together into fibrils. There is a great deal of debate on the actual size of microfibrils however a likely approximation is 100Ǻ to 300Ǻ wide (Kollmann & Cote, 1968) and a cross section can contain up to 2,000 individual cellulose molecules. In the S1 layer the microfibrils are between 50° to 70°, for the S2 layer angles range from 10° to 30° and in the S3 layer they are 60° to 90°. The thickness of each layer also differs, the S1 layer is very thin at approximately 0.1μm to 0.3μm, the S2 layer is the thickest at 1μm to 5μm, and the S3 layer is again thin with a thickness of 0.1μm (Walker, 1993/Wardrop, 1964). Compression Kiln R&D  Duncan Fairclough 3  Figure 1.3: Shows cell wall structure and how microfibrils are oriented in S1, S2, and S3 layers of secondary cell wall. Wood is a hygroscopic material which means that it will gain or lose moisture depending on the temperature and humidity of the atmosphere. In living trees water is present in both the lumen and the cell wall. As wood is dried the water in the lumen known as “free water” is lost. The point when all the water has been removed from the lumen but the cell walls remain filled with water is called the fibre saturation point (FSP). If the timber continues to dry the water in the cell walls known as “bound water” is removed. The bound water has the greatest affect on the normal shrinkage and strength properties of the timber.  If timber is left out in stable conditions it will gain or lose moisture until it is in equilibrium with the surrounding air humidity. At this point the timber is said to be at its equilibrium moisture content (EMC). This EMC is rarely ever achieved due to the constant change in ambient conditions from daily changing temperature, weather conditions, and relative humidity. There are many other factors that can affect the EMC such as the wood species, heartwood/sapwood, and the compositions of the cells inside the wood. In timber drying there are two types of shrinkage that occurs: normal and collapse shrinkage. When the moisture content in timber that is being dried falls below the FSP, Compression Kiln R&D  Duncan Fairclough 4  bound water starts to be removed from the cell wall, at this point normal shrinkage occurs. This normal shrinkage is further complicated due to the fact that wood is an anisotropic material that have different shrinkage properties in all three planes (longitudinal, radial, and tangential).  A general approximation of the amount of  shrinkage in each plane in relation to the other is as follows: tangential shrinkage is 1.5 to 2.5 times greater than radial shrinkage for most species; longitudinal shrinkage can be ignored in most cases due to it being so negligible (Koehler 1931; US Forest Service Forest Products Laboratory 1999). As wood is not a consistent material throughout it is quite obvious that timber will not dry uniformly. The outer edges of the timber will reach the fibre saturation point soon after timber drying commences while the core of the board remains above the FSP. This causes a moisture gradient in the board which can lead to drying stresses. As the outer edge of the board will have no water in cells they will want to shrink, however as the cells located in the core of the board still contain water theses cells will not shrink. This causes two different stresses to occur in the board, on the outer edge there will be tension stress and at the core there will be compression stress. As drying continues the core of the board will begin to reach FSP, at this point stress reversal will occur as the outer edge of the board tries to limit the amount of shrinkage that occurs in the core of the board.  Understanding the various stresses that occur during drying and how drying temperatures and moisture contents affect the drying stresses is important when trying to determine the optimal drying schedule for boards. These stresses are directly related to surface and internal checking found in many species of wood. Much of the stress that occurs during drying is related to factors such as the grain orientation and board dimensions of the boards.  As stated earlier longitudinal shrinkage in normal wood is very minimal, in timber dried from green to oven dry moisture content shrinkage is usually between 0.1% and 0.3% (Hann 1969; Skaar 1988). The exception to the rule is typically found in reaction wood (compression wood for softwood and tension wood for hardwoods) and juvenile wood Compression Kiln R&D  Duncan Fairclough 5  where excessive longitudinal shrinkage can occur.  Commonly 1-2% longitudinal  shrinkage can be found in compression wood, but it can go as high as 6-7% (Haygreen & Bowyer 1982). This can prove to be problematic when larger boards are dried due to the higher possibility of juvenile or reaction wood being mixed in with the normal wood causing uneven shrinkage throughout the board. A special case of shrinkage called collapse can be found to some degree in a variety of species, but is particularly prevalent in Eucalypt that have low to medium density (Rozsa & mills 1997). Collapse differs from normal shrinkage in that it occurs above the FSP during the early stages of the drying process. This means that the shrinkage occurs when water is still present in the cell lumen, unlike normal shrinkage which occurs when water only remains in the cell walls as bound water. The most commonly accepted theory behind the cause of collapse is the hydrostatic tension theory that was proposed by Tiemann in 1915. This theory was further developed by Kauman (1964). Kauman used two main equations to analyze the theory the first equation; Laplace’s equation involves relating the total liquid tension to the radii of the curved water surface. The second equation, Kelvin’s equation relates the total liquid tension to the relative vapour pressure above the meniscus. From these two equations Kauman developed an equation for liquid tension collapse. For this theory there are two assumptions, first sap completely fills the lumen and does not contain bubbles with a radius greater than that of the meniscus (this would relieve tension by expanding) and second the sap has sufficient cohesive strength to transmit the tension without cavitation.  One of the main problems related to collapse is internal checking. It often occurs in species where the difference between the earlywood and latewood density is quite high. These internal checks are caused by the difference in collapse potential of the lower density earlywood and the higher density latewood. Generally the internal checks will remain in one growth ring, however if there is severe stress caused by drying the checks will cross over multiple rings.  This severe checking is often referred to as  “honeycombing”. In most cases collapse has little effect on the strength and stiffness  Compression Kiln R&D  Duncan Fairclough 6  properties of a board, however when there is extreme checking or honeycombing there can be a loss of structural integrity of the board, particularly a decrease in shear strength.  Despite all the problems caused by collapse it is possible to recover collapsed timber with the use of steam. This reconditioning to recover collapse was first discovered by James and George Grant (father and son) in 1917.  Steam reconditioning timber involves  placing the timber in an insulated chamber and filling the chamber with atmospheric pressure steam for a certain period of time based on board dimensions and species. Steam reconditioning has been proven to reverse the physical appearance of collapse in timber however; once collapse and checking has occurred in timber the physical damage it causes cannot be reversed. With this in mind most research on collapse has been done to determine how to prevent collapse rather than reversing it.  There is limited knowledge regarding at what moisture content the timber should be steam reconditioned, in Australia most hardwood sawmills follow the guideline that boards should have a moisture content of 15-20% before they are reconditioned (CSIRO, 1942). This target is used to make the assumption that the core moisture content of the board is less than 25% or below FSP. If the board’s moisture content is above FSP there is the possibility of further collapse to occur in the board after the steam reconditioning process.  It can be difficult to recondition larger dimension timber as the moisture  gradient within a single board may be quite high meaning the some parts of the board may not be below FSP which would allow collapse to occur. Recent studies have shown that a moisture content of 15% is the minimum cut-off point for reasonable collapse recovery. A moisture content of 20% prior to steam reconditioning showed positive results for collapse recovery, even higher moisture contents of up to 25% showed the best results. However, with these higher moisture contents the core to surface moisture gradient of individual boards had to be minimized to obtain the best results (P.A. Blakemore 2008). Greenhill came up with the theory that in the cell wall there are two structural layers that affect reconditioning, one layer being elastic and the other inelastic. He believed that when collapse occurs there is a larger amount of energy that is stored in the elastic layer. Compression Kiln R&D  Duncan Fairclough 7  This energy was very unstable and required the inelastic layer to restrain it in the collapsed state. With this in mind, when steam reconditioning is applied the inelastic layer becomes heated and less force is required to deform it. Essentially as heat is applied to the timber the energy and force in the elastic layer is able to overcome the force of the inelastic layer and the cell is able to expand back to its normal state. However, the cell will not be completely restored to its pre-collapsed state as the application if heat will reduce the elastic properties of the cell which in turn lowers the amount of stored potential energy in the elastic layer of the cell. This two layer theory has not been completely proven. However, it has been proposed that the S3 layer of the cell may in fact be the elastic layer due to the high microfibril angle (high microfibril angle relates to high stiffness), and the lignin hemicellulose matrix near the primary cell wall may be the inelastic layer (Kaum 1964). 1.2 Background on the TeknoComp Compression Kiln The compression kiln that FEA purchased was manufactured by TeknoComp which is a Finish company that manufactures various compression based timber drying kilns. TeknoComp was founded in 2003 and is still currently privately owned. The company works closely with the Mari Group. Both companies share the same principal owner. TeknoComp utilizes two brand names, CompKiln and Wood3. The compression kiln used by FEA falls under the CompKiln brand name.  TeknoComp is essentially trying to overcome the drying problems and issues that conventional drying methods are prone to. The main advantage of the compression kiln stated by TeknoComp is the ability to decrease drying times of timber while not sacrificing quality. TeknoComp claims that boards will be straight after drying due to the pressure being applied during the drying process along with an increase in the bending strength of boards dried in the compression kiln. Hardness values are also said to be improved. Other advantages of the compression kiln listed by TeknoComp are as follows: -  Increase the value of low valued wood species.  -  Allowing use of plantation woods (such as eucalyptus in) for sawn timber.  Compression Kiln R&D  Duncan Fairclough 8  -  Increase the use of wood species that are notoriously difficult to dry.  -  Improving the efficiency of raw material used by drying sections of wood that are notoriously difficult to dry (pine heartwood for example).  -  Increase inventory turnover through reduced drying times.  -  Energy savings through reduced drying times.  -  Highly automated process that reduces inconsistencies in the final product.  As compression kilns are a relatively new technology there is not a great deal of background information on them.  Through the compression kiln research and  development project FEA hopes to determine how many of the claims listed above are indeed true. 1.3 Compression Kiln The compression kiln comprises of three main chambers. The first chamber is the preheat chamber this is followed by the drying chamber and finally the cool down chamber. After a kiln charge is made it will move through the three chambers before the drying cycle is completed. During the drying cycle the chambers are separated by aluminum doors. The kiln charges are moved through the different chambers by a series of chain conveyors.  Inside the pre-drying chamber there are no heat exchangers present. The pre-drying chamber draws heat and steam from the drying chamber through the use of the kilns extraction fan. The extraction fan is located at the entrance to the pre-drying chamber and draws heat and steam from the drying chamber through a channel located on the door connecting pre-drying chamber to the drying chamber. The steam and heat is pulled through the pre-drying chamber and is used to preheat and steam the timber in the charge and then is extracted through the vent located behind the extraction fan. It is hoped that the pre-steaming will even out the moisture content of the charge prior to it entering the drying chamber. There are ten hydraulic rams located in the pre-drying chamber. These are used to compress the timber in the kiln charge during the pre-drying phase of the drying cycle. The kiln charge is pressed against the I-beams located on the ceiling of the Compression Kiln R&D  Duncan Fairclough 9  pre-drying chamber for the duration of the pre-drying period. The hydraulic rams create 150BAR of pressure, this translates to a pressure of roughly 95MPa being applied to the stack.  The drying chamber consists of a bank of heat exchangers with four fans above them on one of the walls. On the opposing wall there is another bank of heat exchangers with a set of four fans located below them. Both sets of fans only turn in one direction and therefore the air flows only in one direction. The only measuring devices present inside the kiln are a set of four dry bulbs and a set of four wet bulbs. Initially the software for the compression kiln did not allow for the data to be downloaded for further analysis; however the software has been updated to allow this. Like the pre-drying chamber the drying chamber has ten hydraulic rams that are used to press the kiln charge against the Ibeams mounted on the ceiling of the compression kiln during the entire drying cycle. The hydraulic rams create 150BAR of pressure which translates to a pressure of roughly 95MPa being applied to the stack.  The cool down chamber has no major equipment inside of it. As the name implies the kiln charges are left in the chamber to cool down prior to being un-stacked. The kiln is capable of temperatures up to around 200°C however it is unlikely that these temperatures would be used. The hydraulic rams are also capable of achieve over 200 BAR or pressure this relates to a pressure being applied to the timber charge of roughly 125MPa.  Compression Kiln R&D  Duncan Fairclough 10  Figure 1.4: Overhead cross section of compression kiln.  Compression Kiln R&D  Duncan Fairclough 11  Figure 1.5: Cross section drawing of compression kiln drying chamber.  Compression Kiln R&D  Duncan Fairclough 12  The kiln charges are made up of rows of timber that are separated by aluminium sheets (not solid sheets as there are gaps). Aluminium was selected for its favorable heat transfer properties.  Figure 1.6: End of compression kiln charge with 100x38mm pine heartwood.  Figure 1.7: Aluminium sheets.  Figure 1.8: Aluminium sheets.  Compression Kiln R&D  Duncan Fairclough 13  Figure 1.9: Side view of compression kiln charge with 100x38mm pine heartwood. 1.4 Project Overview The main focus of the compression kiln project is to determine the feasibility of using the compression kiln to dry the two main species produced by FEA, Eucalyptus nitens and Radiata Pine. The project will also look at two Australian mainland species for the drying trial, Eucalyptus dunnii, Eucalyptus saligna, and Eucalyptus sieberi. The Eucalyptus saligna and Eucalyptus sieberi will only be done in small amounts. All of the mainland species were chosen as they have properties that could make them favorable to compression drying. A sidebar to the project is also to look at further understanding the physics of drying using the compression kiln as there has been very little research done in this area.  The main focus of the project will be on gathering data and developing drying schedules for various wood species. The pine trials were mainly focused on final moisture contents, defects that occurred during drying, and visual grading of the dried timber. During the second half of the project (when the various hardwood species were being dried), the kiln software was changed making it possible to track various kiln parameters such as wet bulb and dry bulb temperatures as well as the temperature drop across the charge. From Compression Kiln R&D  Duncan Fairclough 14  this data it was possible to determine the relative humidity inside the kiln during the drying process. With this information it was possible to compare various charges and determine how consistent the kiln conditions were from charge to charge.  Compression Kiln R&D  Duncan Fairclough 15  2.0 Eucalyptus Nitens Trial 2.1 Eucalyptus Nitens Currently E. nitens needs to be air dried for around four months before it is dried in the kiln over 24 to 48 hours. Eucalyptus nitens (E. nitens) was the main driver for FEA’s decision to purchase the compression kiln. The idea being that the kiln could be used to dry green timber to moisture content of 10% to 12% in approximately 24 hours. This would allow FEA to cut holding/inventory costs even if cut volumes increase. There was also hope that the compression kiln would decrease drying induced degrade. The timber used for the study will be from plantation forest located in Tasmania. As this is the main hardwood produced by FEA it would be very beneficial to be able to dry E. nitens to a relatively good quality. Unfortunately this is very difficult as E. nitens is prone to many drying defects.  The bulk of the material that has been used in the compression kiln prior to this project has been E. nitens. The results that FEA has achieved with this species have not been very promising. The major defects seen in E. nitens are surface checking, internal checking, and collapse. Collapse appears as surface rippling along the surface of E. nitens boards. It is one of the most common defects in hardwood that is dried quickly. There is limited information on what causes collapse however it appears to be occur during the drying process when the fibre cells in the wood physically collapse. This can also cause internal and surface checking which appears as cracks on the surface of the board. Internal checking is difficult to see as a cross section of the board is needed. In the cross section internal checking appears as small holes or voids in the wood.  Compression Kiln R&D  Duncan Fairclough 16  Figure 2.1: Cross section of E. nitens board containing internal checking. However the results have not all been bad, there have been some positive findings. It seems that E. nitens boards dried at a lower temperature have achieved decent quality. The only issue with these low temperature kiln charges is the time required to dry a charge is quite high. The fact that the compression kiln can only dry a small volume of timber in one charge it requires short drying times to produce large volumes of timber.  Compression Kiln R&D  Duncan Fairclough 17  2.2 Eucalyptus Nitens Batch Summaries 2.2.1 Batch #1 Summary (E. nitens) Batch #1 was run over a period of approximately one day (20.0 hours) at a temperature of 110°C. There were 147 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.52m3). The temperature was chosen to set a starting point for future trials to be based off. Table 2.1: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 30.0 31.6 30.1 30.6 30.3  Table 2.2: Stack recovery based on surface quality. Good Average 20% 10% Stack Recovery  Standard Deviation (%) 5.5 4.8 6.0 5.5 5.4  Poor 70%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been 100% poor due to none of the boards being dried to acceptable moisture content. The main defects seen other than the boards not being dry were surface checking and collapse. Both defects were seen in all of the rows to varying degrees. Collapse was seen on almost every board while surface checking appeared on approximately half of the boards.  Drying time and or temperature will need to be increase to achieve more acceptable average moisture content in the charge.  Compression Kiln R&D  Duncan Fairclough 18  2.2.2 Batch #3 Summary (E. nitens) Batch #3 was run over a period of approximately a day and a half (34.0 hours) at a temperature of 70°C. There were 147 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.52m3).  The temperature of 70°C was reached three hours into the drying process and remained relatively stable during the cycle.  The temperature was chosen to try and reduce the  harshness of the drying conditions in hopes that it may improve overall board quality. Kiln Conditions 100.0 Wetbulb Depression  90.0  Wetbulb Temperature Drybulb Temperature  80.0  Temp (°C) / Humidity (%)  Relative Humidity Dewpoint  70.0 60.0 50.0  40.0 30.0 20.0 10.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200 1260  Time (min)  Figure 2.2: Drying conditions for batch. Table 2.3: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 35 35 35 35 35  Table 2.4: Stack recovery based on surface quality. Good Average 60% 30% Stack Recovery  Standard Deviation (%) n/a n/a n/a n/a n/a  Poor 10%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been worse as none of the boards were dried. The main defects seen Compression Kiln R&D  Duncan Fairclough 19  in most boards were collapse and surface checking. The majority of these defects were not severe and there was minimal down grading. The minimal amount of defects was probably due to the timber not being dried completely, had the timber remained in the same conditions for longer to bring down the moisture content more defects may have appeared. The temperature will need to be increased and or the drying time will need to be longer to achieve better average moisture content.  At the end of the drying cycle the average temperature drop across the charge was 0.7°C. Temperature Differences 9.0  8.0 7.0  Temperature (°C)  6.0 Upper Temperature Difference  5.0  Lower Temperature Difference  4.0 3.0  Combined Average  2.0 1.0 0.0 -1.0  -2.0 Time (min)  Figure 2.3: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 20  2.2.3 Batch #6 Summary (E. nitens) Batch #6 was run over a period of approximately one day (20.3 hours) at a temperature of 130°C. There were 144 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.43m3).  The temperature of 130°C was reached during the final hours of the drying cycle. The temperature was chosen to try and reduce the amount of time required to dry the boards and see the effects of high temperature on drying quality. Kiln Conditions 140.0 Wetbulb Depression Wetbulb Temperature  120.0  Drybulb Temperature  Relative Humidity  Temp (°C) / Humidity (%)  100.0  Dewpoint  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200  Time (min)  Figure 2.4: Drying conditions for batch. Table 2.5: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 15.2 15.1 13.2 14.5 14.9  Table 2.6: Stack recovery based on surface quality. Good Average 60% 10% Stack Recovery  Standard Deviation (%) 6.1 6.1 5.0 5.8 5.0  Poor 30%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 21  Figure 2.5 (left): E. nitens sample #8 Figure 2.6 (right): Sample #5 had a which had a grade of good based on grade of poor due to severe collapse and surface quality. checking. Moisture content was not taken into account. If moisture content were included the recovery would have been worse. The main defect seen in most boards was collapse. There was also surface checking present in a few boards. Also there was severe internal checking, even in boards graded good. The average moisture content was too high. The temperature will need to be increased or drying time will need to be increased.  Compression Kiln R&D  Duncan Fairclough 22  At the end of the drying cycle the average temperature drop across the charge was 2.8°C. Temperature Differences 10.0  9.0 8.0  7.0 Upper Temperature Difference  Temperature (°C)  6.0 5.0  Lower Temperature Difference  4.0  Combined Average  3.0 2.0 1.0  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200  -1.0  -2.0 Time (min)  Figure 2.7: Temperature difference across the load. 2.2.4 Batch #7 Summary (E. nitens) Batch #7 was run over a period of approximately one day (20.7 hours) at a temperature of 130°C. There were 104 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (3.20m3). The boards were placed in the top 8 rows and the bottom 5 rows with E. dunnii and E. saligna boards placed in the middle of the charge.  The temperature of 130°C was reached during the final hours of the drying cycle. The temperature was chosen to try and reduce the amount of time required to dry the boards and see the effects of high temperature on drying quality.  Compression Kiln R&D  Duncan Fairclough 23  Kiln Conditions 140.0 Wetbulb Depression  120.0  Wetbulb Temperature  Drybulb Temperature Relative Humidity  Temp (°C) / Humidity (%)  100.0  Dewpoint  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200  Time (min)  Figure 2.8: Drying conditions for batch. Table 2.7: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 17.2 19.3 16.3 20.3 17.6  Table 2.8: Stack recovery based on surface quality. Good Average 60% 0% Stack Recovery  Standard Deviation (%) 5.7 6.7 5.1 6.2 6.0  Poor 40%  *(Based on surface quality)  Figure 2.9: E. nitens sample #7  Figure 2.10: E. nitens sample #3  Compression Kiln R&D  Duncan Fairclough 24  Figure 2.11: E. nitens sample #7 which Figure 2.12: Sample #3 had a grade of had a grade of good based on surface poor due to severe collapse. quality. Moisture content was not taken into account. If moisture content were included the recovery would have been worse. The main defect seen in most boards was collapse. There was also surface checking present in a few boards. Also there was severe internal checking, even in boards graded good. The average moisture content was too high. The temperature will need to be increased or drying time will need to be increased.  Compression Kiln R&D  Duncan Fairclough 25  At the end of the drying cycle the average temperature drop across the charge was 2.8°C. Temperature Differences 6.0  5.0  Temperature (°C)  4.0  Upper Temperature Difference  3.0  Lower Temperature Difference  2.0  Combined Average  1.0  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200  -1.0 Time (min)  Figure 2.13: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 26  2.2.5 Batch #8 Summary (E. nitens) Batch #8 was run over a period of approximately one day (19.3 hours) at a temperature of 130°C. There were 128 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (3.94m3). The boards were placed in the top 9 rows and the bottom 7 rows with E. dunnii boards placed in the middle of the charge.  The temperature of 130°C was reached during the final hours of the drying cycle. The temperature was chosen to try and reduce the amount of time required to dry the boards. Kiln Conditions 140.0 Wetbulb Depression  120.0  Wetbulb Temperature  Drybulb Temperature Relative Humidity  Temp (°C) / Humidity (%)  100.0  Dewpoint  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140  Time (min)  Figure 2.14: Drying conditions for batch. Table 2.9: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 19.4 20.4 21.1 20.3 20.7  Table 2.10: Stack recovery based on surface quality. Good Average 10% 45% Stack Recovery  Standard Deviation (%) 5.2 5.6 7.6 6.2 5.7  Poor 45%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 27  Figure 2.15: E. nitens sample #6  Figure 2.16: E. nitens sample #4  Figure 2.17: E. nitens sample #6 which Figure 2.18: Sample #4 had a grade of had a grade of good based on surface poor due to severe surface checking. quality. Moisture content was not taken into account. If moisture content were included the recovery would have been worse. The main defect seen in most boards was surface checking. There was also collapse present in many boards. Generally boards with lower moisture content appeared to have fewer surface defects. Also there was severe internal checking, even in boards graded good. The average moisture content was too high. The temperature will need to be increased or drying time will need to be increased.  Compression Kiln R&D  Duncan Fairclough 28  At the end of the drying cycle the average temperature drop across the charge was 3.2°C. Temperature Differences 7.0  6.0  5.0 Upper Temperature Difference  Temperature (°C)  4.0  Lower Temperature Difference  3.0  2.0  Combined Average  1.0  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140  -1.0  -2.0 Time (min)  Figure 2.19: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 29  2.2.6 Batch #9 Summary (E. nitens) Batch #9 was run over a period of approximately one day (20.7 hours) at a temperature of 140°C. There were 144 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.43m3). The boards were placed in the top 18 rows of the charge with the bottom rows being filled with E. saligna boards.  The temperature of 140°C was never reached. The temperature was chosen as there had been promising results from previous kiln charges dried at this temperature. It is unknown why the final drying temperature of 140°C was never reached. Kiln Conditions 160.0 Wetbulb Depression  140.0  Wetbulb Temperature Drybulb Temperature  Relative Humidity  120.0 Temp (°C) / Humidity (%)  Dewpoint  100.0  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200  Time (min)  Figure 2.20: Drying conditions for batch. Table 2.11: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 16.0 16.4 18.6 17.0 17.1  Compression Kiln R&D  Standard Deviation (%) 5.3 4.9 6.4 5.7 4.2  Duncan Fairclough 30  Table 2.12: Stack recovery based on surface quality. Good Average 25% 10% Stack Recovery  Poor 65%  *(Based on surface quality)  Figure 2.21: E. nitens sample # 6.  Figure 2.22: E. nitens sample #7  Figure 2.23: E. nitens sample #6 which Figure 2.24: Sample #7 had a grade of had a grade of good. poor due to severe surface checking. Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. The main defect seen in most boards was surface collapse and checking, which was seen in most of the boards to varying degrees. Surface checking was the most severe defect in the stack. Generally boards with lower moisture content appeared to have fewer surface defects. Also there was severe internal checking. The average moisture content was acceptable. However, it still could be slightly lower. It would be worth adding an extra one or two hours to the drying schedule to achieve this.  Compression Kiln R&D  Duncan Fairclough 31  At the end of the drying cycle the average temperature drop across the charge was 2.4°C. Temperature Differences 5.0  4.0  Temperature (°C)  3.0  Upper Temperature Difference  2.0  Lower Temperature Difference  1.0  Combined Average  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200  -1.0  -2.0 Time (min)  Figure 2.25: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 32  2.2.7 Batch #10 Summary (E. nitens) Batch #10 was run over a period of approximately one day (21.2hours) at a temperature of 140°C. There were 144 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.43m3).  The temperature of 140°C was never reached. The temperature was chosen as there had been promising results from previous kiln charges dried at this temperature. It is unknown why the final drying temperature of 140°C was never reached. Kiln Conditions 160.0 Wetbulb Depression  140.0  Wetbulb Temperature Drybulb Temperature  Relative Humidity  120.0 Temp (°C) / Humidity (%)  Dewpoint  100.0  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200 1260  Time (min)  Figure 2.26: Drying conditions for batch. Table 2.13: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 12.8 18.5 15.1 15.5 15.6  Table 2.14: Stack recovery based on surface quality. Good Average 10% 20% Stack Recovery  Standard Deviation (%) 4.7 6.6 6.6 6.4 5.7  Poor 70%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 33  Figure 2.27: E. nitens sample #7.  Figure 2.28: E. nitens sample #4.  Figure 2.29 (left): E. nitens sample #7 Figure 2.30 (right): Sample #4 had a which had a grade of good. grade of poor due to severe collapse. Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. The main defect seen in most boards was surface collapse, which was seen in most of the boards. Generally boards with lower moisture content appeared to have fewer surface defects. Also there was severe internal checking. The average moisture content was acceptable. However, it still could be slightly lower. It may be worth adding an extra one or two hours to the drying schedule to achieve this.  Compression Kiln R&D  Duncan Fairclough 34  At the end of the drying cycle the average temperature drop across the charge was 2.3°C. Temperature Differences 5.0  4.0  Temperature (°C)  3.0  Upper Temperature Difference  2.0  Lower Temperature Difference  1.0  Combined Average  0.0  -1.0  -2.0 Time (min)  Figure 2.31: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 35  2.2.8 Batch #11 Summary (E. nitens) Batch #11 was run over a period of approximately two days (49.4 hours) at a temperature of 90°C. There were 128 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (3.94m3). The boards were placed in the top 16 rows of the kiln charge.  The lower temperature was chosen to create a less harsh drying environment for the timber by decreasing wet bulb depression and maintaining a higher value for relative humidity. Drying time was increased over previous low temperature charges as previous charges had high average moisture contents. Kiln Conditions 100.0 Wetbulb Depression  90.0  Wetbulb Temperature Drybulb Temperature  80.0  Temp (°C) / Humidity (%)  Relative Humidity Dewpoint  70.0 60.0 50.0 40.0 30.0 20.0 10.0  96 0 10 80 12 00 13 20 14 40 15 60 16 80 18 00 19 20 20 40 21 60 22 80 24 00 25 20 26 40 27 60 28 80  84 0  72 0  60 0  48 0  36 0  24 0  0  12 0  0.0  Time (min)  Figure 2.32: Drying conditions for batch. Table 2.15: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 21.4 20.4 20.3 20.7 20.5  Table 2.16: Stack recovery based on surface quality. Good Average 25% 25% Stack Recovery  Standard Deviation (%) 7.5 6.4 7.2 7.0 4.9  Poor 50%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 36  The main defect seen in most boards was surface checking and collapse. When boards were cross sectioned there was minimal internal checking in the boards.  Figure 2.33: E. nitens sample #4 which Figure 2.34: Sample #3 had a grade of had a grade of good. Poor. Moisture content was not taken into account. If moisture content were included the recovery would have been somewhat worse. Generally there was an improvement over previous low temperature charges. The amount of defects such as checking and collapse was similar to that of previous low temperature charges. Also there was minimal internal checking. The moisture content was closer to acceptable levels but was still slightly high. It seems that a longer drying time would be required to bring average MC into the 15% range.  Compression Kiln R&D  Duncan Fairclough 37  At the end of the drying cycle the average temperature drop across the charge was 0.6°C. Temperature Differences 5.0  Temperature (°C)  4.0  3.0  Upper Temperature Difference  2.0  Lower Temperature Difference  1.0  Combined Average  96 0 10 80 12 00 13 20 14 40 15 60 16 80 18 00 19 20 20 40 21 60 22 80 24 00 25 20 26 40 27 60 28 80  84 0  72 0  60 0  48 0  36 0  24 0  0 12 0  0.0  -1.0  -2.0 Time (min)  Figure 2.35: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 38  2.2.9 Batch #12 Summary (E. nitens) Batch #12 was run over a period of approximately one day (21.1hours) at a temperature of 140°C. There were 156 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.80m3).  The temperature was left at 140°C and the vacuum fan setting was changed so that it would turn on at a temperature of 120°C. This was done to see if the relative humidity could be raised during the early part of the drying cycle. For an unknown reason the final drying temperature of 140°C was never reached. Kiln Conditions 140.0 Wetbulb Depression  120.0  Wetbulb Temperature  Drybulb Temperature Relative Humidity  Temp (°C) / Humidity (%)  100.0  Dewpoint  80.0  60.0  40.0  20.0  0.0 0  120  240  360  480  600  720  840  960  1080  1200  Time (min)  Figure 2.36: Drying conditions for batch. Table 2.17: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 28.3 27.4 26.0 20.7 27.4  Table 2.18: Stack recovery based on surface quality. Good Average 20% 30% Stack Recovery  Standard Deviation (%) 7.4 7.5 7.6 7.5 4.6  Poor 50%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 39  Figure 2.37: E. nitens sample #6 which Figure 2.38: Sample #3 had a grade of had a grade of good. poor due to severe collapse and surface checks. Moisture content was not taken into account. If moisture content were included the recovery would have been somewhat worse. There did not seem to be a much difference in the relative humidity when compared to previous charges. The main defect seen in most boards was surface collapse Also there was minimal internal checking.  The  moisture content was too high, most likely due to the final drying temperature of 140°C no being reached.  Compression Kiln R&D  Duncan Fairclough 40  At the end of the drying cycle the average temperature drop across the charge was 1.1°C. Temperature Differences 4.0  3.0  Upper Temperature Difference  Temperature (°C)  2.0  Lower Temperature Difference  1.0  Combined Average  0.0  -1.0  -2.0  Time (min)  Figure 2.38: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 41  2.2.10 Batch #15 Summary (E. nitens) Batch #15 was run over a period of approximately one day (22.7hours) at a temperature of 140°C. There were 124 boards of E. nitens with dimensions 150mm x 38mm x 5.4m boards (4.45m3). The boards were located in rows 1 to 12 and 17 to 21.  The temperature was left at 140°C and the drying time was increased to try and bring the average moisture content of the stack down. Kiln Conditions 160.0 Wetbulb Depression  140.0  Wetbulb Temperature Drybulb Temperature Relative Humidity  Temp (°C) / Humidity (%)  120.0  Dewpoint  100.0  80.0  60.0  40.0  20.0  0.0  Time (min)  Figure 2.39: Drying conditions for batch. Table 2.19: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 7.5 8.5 10.5 8.8 9.2  Table 2.20: Stack recovery based on surface quality. Good Average 0% 45% Stack Recovery  Standard Deviation (%) 2.7 3.5 4.3 3.7 3.5  Poor 55%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 42  Figure 2.40: E. nitens sample #8 which had a grade of good based on surface quality. When it was cross sectioned it could be seen there was severe internal checking.  Figure 2.41: Sample #6 had a grade of poor due to severe collapse. There was also severe internal checking.  Moisture content was not taken into account. If moisture content were included the recovery would have been almost identical. The main defects seen in most boards were surface checking and collapse. The average moisture content of the stack was at an acceptable level which was most likely due to the extended drying time. The overall stack quality was not very good as most boards contained at least some collapse and/or surface checks. This was probably due to the extremely dry conditions (humidity was quite low) in the drying chamber.  Compression Kiln R&D  Duncan Fairclough 43  At the end of the drying cycle the average temperature drop across the charge was 0.6°C. Temperature Differences 13.0 12.0 11.0  10.0 9.0 Upper Temperature Difference  Temperature (°C)  8.0 7.0 6.0  Lower Temperature Difference  5.0 4.0  Combined Average  3.0 2.0 1.0  0.0 -1.0 -2.0 -3.0 Time (min)  Figure 2.42: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 44  2.2.11 Batch #18 Summary (E. nitens) Batch #18 was run as a two part charge over the period of approximately two days (Part#1: 21.0hosurs, Part#2: 15.8 hours).The first part ran at a temperature of 70°C and the second part ran at a temperature of 100°C. There were 104 boards of E. nitens with dimensions 175mm x 38mm x 5.4m boards (3.73m3).  The two stage drying process was chosen to try and reduce the harshness of the drying conditions to minimize drying defects. The first part was run without the extraction fan running to increase humidity. For the second part the fan was run but only during the final drying stages.  Kiln Conditions Part 1 100.0 Wetbulb Depression  90.0  Wetbulb Tem perature Drybulb Tem perature  80.0  Temp (°C) / Humidity (%)  Relative Hum idity Dew point  70.0  60.0 50.0 40.0 30.0 20.0  10.0 0.0  Time (min)  Figure 2.43: Drying conditions for part one of batch.  Compression Kiln R&D  Duncan Fairclough 45  Kiln Conditions Part 2 120.0 Wetbulb Depression  Wetbulb Tem perature  100.0  Drybulb Tem perature  Temp (°C) / Humidity (%)  Relative Hum idity Dew point  80.0  60.0  40.0  20.0  0.0  Time (min)  Figure 2.44: Drying conditions for part 2 of batch. Table 2.21: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 24.8 28.1 21.7 24.9 23.4  Table 2.22: Stack recovery based on surface quality. Good Average 60% 10% Stack Recovery  Standard Deviation (%) 10.5 9.6 6.7 9.4 11.1  Poor 30%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 46  Figure 2.45 (left): E. nitens sample #6 Figure 2.46 (right): Sample #4 had a which had a grade of good. grade of poor due to severe surface checks. Moisture content was not taken into account. If moisture content were included the recovery would have been somewhat worse. About 25% of the boards actually had a moisture content that was within the acceptable range. Boards located in the top quarter seemed to have to best quality as well as better moisture content. In the first part of drying (no exhaust fan running) the humidity was significantly higher than previous charges however during the second part the humidity was comparable to other charges. The main defect seen in most boards was surface collapse and checking. Also there was minimal internal checking. The moisture content was too high, most likely due to the boards being dried at lower temperatures for too short a time.  Compression Kiln R&D  Duncan Fairclough 47  At the end of the drying cycle the average temperature drop across the charge was 0.8°C. Temperature Differences Part 1 12.0  11.0 10.0  9.0 8.0 Upper Tem perature Difference  Temperature (°C)  7.0 6.0  Low er Temperature Difference  5.0  4.0 Com bined Average  3.0  2.0 1.0 0.0 -1.0 -2.0  -3.0 Time (min)  Figure 2.47: Temperature difference across the load for part one. Temperature Differences Part 2 11.0  10.0 9.0  8.0 7.0 Upper Tem perature Difference  Temperature (°C)  6.0 5.0  Low er Temperature Difference  4.0  3.0 Com bined Average  2.0  1.0 0.0 -1.0 -2.0 -3.0  -4.0 Time (min)  Figure 2.48: Temperature difference across the load for part two.  Compression Kiln R&D  Duncan Fairclough 48  2.3 Eucalyptus Nitens Hardness Testing The hardness of the E. nitens timber dried in the compression kiln was tested against a control batch that had been put through a standard kiln to determine if TeknoComp’s claims of increased hardness were true. To test this theory a sample of 30 pieces of timber dried in the compression kiln and 30 pieces of conventionally dried timber were tested using the Janka hardness testing method. The compression kiln sample pieces were 175x38mm rough sawn E. nitens heartwood samples cut to a length of approximately 600mm. As these boards were rough sawn they first needed to be planed to provide a flat surface to perform the hardness tests on. The control sample boards were 90x35mm cut to a length of 300mm. The samples were tested twice on each face. Hardness values were determined at exactly a depth of 5.64mm by determination of slope in the range 5.59-5.66mm and calculating the intercept at 5.64mm.  Moisture a content of each board was also determined using a resistance moisture meter (Delmhorst J2000), setup to measure pine and correcting for temperature. Moisture content readings were taken from the centre of the sample board entering from Face 1.  Compression Kiln R&D  Duncan Fairclough 49  Table 2.23: Hardness values for E. nitens.  Compression Kiln R&D  Duncan Fairclough 50  Hardness Values Comparison (Both Faces) 5.00  4.50  4.00  Hardness Values (KN)  3.50  Compression Kiln  3.00  Control 2.50  2.00  1.50  1.00  0.50  0.00  Figure 2.49: Comparison of hardness values for E. nitens trial. From the above data it can be seen that there is little difference between the hardness of the control and compression kiln boards. The combined average for both face of the compression kiln boards and the control boards were exactly the same at 3.00kN. The E. nitens boards appear to be slightly harder than the pine boards that were tested. Despite the small sample size it would be safe to assume that the hardness value of timber dried in the compression kiln is not affected one way or another.  Compression Kiln R&D  Duncan Fairclough 51  2.4 Temperature Difference vs. MC With the current compression kiln software there is no way of tracking the moisture content of the kiln charge during the drying process however, it is possible to monitor the temperature drop across the load (difference in temperature between the hot side and cold side of the stack). It was thought that it may be possible to determine a correlation between the moisture content and the temperature drop across the load. If this were possible there would be no need to implement a moisture content measuring system in the kiln as the stacks moisture content could be tracked using the temperature difference. To test this theory the value for the temperature difference at the end of the kiln drying process was recorded and compared against the average moisture content of the boards in the stack (moisture content of the boards was measured after the stack had been allowed to cool down, usually 24 hours after drying cycle finished).  Moisture Content vs Temperature Difference 3.5  Temperature Di fference ( C)  3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0  5.0  10.0  15.0  20.0  25.0  30.0  35.0  40.0  Moisture Content (% )  Figure 2.50: Moisture content vs. temperature difference.  Compression Kiln R&D  Duncan Fairclough 52  Table 2.24: Moisture content vs. temperature difference.  As can be seen from the above data there is no correlation between the moisture content of the stack and temperature difference. Despite the small sample size it would be safe to assume that there would be no correlation found even with more testing due to the data being so sporadic.  Compression Kiln R&D  Duncan Fairclough 53  2.5 Eucalyptus Nitens Trial Summary 2.5.1 Current E. nitens Trial Summary Over the course of the study approximately 82m3 of E. nitens was dried in the compression kiln.  The schedules used ranged from low temperature slow drying  times(90°C for 48 hours) to high temperature fast drying times (140°C for 22 hours). There was a great deal of variation in the quality of timber that was produced however overall results were not very good. Much of the timber contained collapse and checking on the surface of the board as well as internal checking when the boards were cross sectioned. The best quality timber when looking at both surface and internal quality came from lower temperature charges that were dried for longer periods of time. This is most likely due to the fact that these lower temperature schedules create milder drying conditions (smaller wet bulb depression and higher relative humidity).  Compression Kiln R&D  Duncan Fairclough 54  Table 2.25: Summary of drying schedules Batch Temperature Drying Time Notes and Change from Previous # (°C) (Hours) Charge Overall quality was poor. MC was too 110 20.0 1 high. Lower temperature/longer drying time. 70 34 3 Quality was worse due to high MC. Board quality was better. Moisture 130 20.3 6 content was within an acceptable range. Final drying temperature was not reached. 130 20.3 7 Most boards were not dry. Drying time was decreased. Final drying 130 19.3 temperature was not reached. MC was too 8 high. Overall quality was poor. Drying temperature and time were 140 20.3 increased. Final drying temperature was 9 not reached. MC was acceptable. Drying time was increased. Quality was 140 21.1 10 slightly worse. More checking/collapse. Drying temperature decreased and time 90 49 increased. MC was too high. Quality was 11 poor. Drying temperature was increased and time was decreased. Extraction fans were 140 21.1 turned on at 120°C to increase relative 12 humidity. Only reached 120°C. MC was too high. Drying time was increased. MC was a bit 140 22.7 15 low. Overall quality was poor. Two part schedule, no extraction fan in part 1. Lower temperature longer drying 70 then 100 36.8 18 time. MC was too high. Surface quality of boards was acceptable. As stated earlier in the section there were no real advantages to timber dried in the compression kiln over timber dried in conventional kilns when looking at bending strengths and surface hardness. Also there was no correlation to be made between the temperature drop across the load and the moisture content of boards within the stack. Overall the statement that the compression kiln would allow for E. nitens timber to be dried from green with little or no defects is farfetched. To truly accomplish this some form of pre-conditioning prior to drying and re-conditioning after drying would need to be implemented. Compression Kiln R&D  Duncan Fairclough 55  As stated earlier in the report kiln conditions for the E. nitens trial were able to be recorded and it was possible to determine if the compression kiln could create repeatable drying conditions from charge to charge. It was found that with that the kiln was capable of creating repeatable drying conditions for charges with the same or similar drying schedules. 2.5.2 Previous E. nitens Trial Summary Prior to this study taking place there had been ongoing research and development of kiln schedules for E. nitens. During the study approximately 60 charges were run with varying drying times and temperatures. Throughout the study there were some schedules that showed promising result however the overall outcome of these trials was not good.  For the study a variety of temperatures and drying times were used. Generally the drying temperatures used in this study were lower than the ones used in the most recent study. One of the main defects noted in the trial was consistently low moisture content; this was most likely due to the lower drying temperatures. The lower temperatures were selected to create drying conditions that would prevent collapse and checking in the timber, unfortunately this did not seem to work as collapse and checking are present throughout most kiln charges.  During the study the kiln software was only capable of setting a final drying time with the change over point for the first five drying steps being set by the temperature difference across the load. At the time of the study it was still believed that there was a correlation between the moisture content of the stack and the temperature drop across the load. As stated in the section before this theory was proven incorrect the current study as there was no relationship found between moisture content and temperature difference. The lack of ability to truly control the drying time would have made it difficult to create repeatable schedules for comparison. Overall the results for this study were less than promising. The majority of the charges had a stack recovery rating of poor to average. In most cases the timber in the stacks had Compression Kiln R&D  Duncan Fairclough 56  a moisture content that was too high and when the moisture content was within an acceptable range the timber was prone to severe surface and internal checking as well as collapse.  Compression Kiln R&D  Duncan Fairclough 57  3.0 E. dunnii, E. saligna, & E. sieberi Trial 3.1 E. dunnii, E. saligna, & E. sieberi As this was a joint project between FEA and the Australian federal government there were some requirements that the government had laid out. The government wanted some of Australia’s mainland species tested in the trials, to satisfy this requirement four species were originally selected however due to supply issues only three species were selected for the tests with Eucalyptus dunnii (E. dunnii) being the first species, Eucalyptus saligna (E. saligna) being the second species, and the last species being Eucalyptus sieberi (E. sieberi). E. dunnii (Dunn’s white gum) is found in north eastern New South Wales and south eastern Queensland. The species grows best in areas with relatively high rainfall. E. dunnii’s wood is quite hard with a density of around 800kg/m3. Currently the main uses for E. dunnii have been pulp and paper; however there has been a shift towards light construction use.  Early drying results from drying E. dunnii have shown some promise for the species being grown as a saw log. The main drying defects that occur in E. dunnii are cupping, spring, end split, and collapse. Currently it can take up to 6 months to completely dry E. dunnii boards after they are cut. The general process for drying is as follows; first the boards are racked and then air dried for 3 to 6 months depending on the climate they are dried in, this brings the moisture content to 20 to 25%.  After this the packs are  reconditioned by steaming over the course of a few hours usually at a temperature of around 100°C to recover collapse. The boards are then put into the final drying stage to bring the moisture content down to around 12%.  E. saligna (Sydney blue gum) occurs naturally in open coastal forests. The tree is suited to a broad range of soils but grows best in sandy gravel that drains well. Due to its shallow roots the tree does not do well in dry sites. Within 20 to 25 years sawlogs with a  Compression Kiln R&D  Duncan Fairclough 58  diameter of 45 to 50cm can be produced provided the trees are planted in areas with sufficient rainfall (550 to 650mm annually). E. saligna’s heartwood is quite hard however it is only moderately durable.  At a  moisture content of 12% density of mature wood is around 850kg/m3, with rapidly grown plantation timber the density is significantly lower at approximately 500kg/m 3. Due to the heartwood having a desirable colour (dark pink/reddish brown) it is commonly used in appearance applications such as flooring, paneling, furniture, and general construction. It can be used for pulping, however this is not preferred due to its low pulp yields and low tensile strength. The timber is prone to collapse when it is dried.  E. Sieberi (Silvertop Ash) is found in New South Wales, Victoria, and Tasmania. It generally grows in mountainous regions. It is slow in drying and is quite prone to surface checking in the tangential plane. Currently the main use for the timber is in general construction. The wood is quite hard with a dry density of 850 kg/m3. It is difficult to distinguish the between the heartwood and the sapwood as they are both very similar in colour.  All of these species are not currently processed by FEA however if positive results are seen FEA may consider them as a production species. Unfortunately it was only possible to obtain a very minimal amount of E. sieberi for the trial making it nearly impossible to make any finite conclusions on the feasibility of the species.  Compression Kiln R&D  Duncan Fairclough 59  3.2 E. Dunnii, E. Saligna, & E. Sieberi Batch Summaries 3.2.1 Batch #7 Summary (E. dunnii & E. Saligna) Batch #7 was run over a period of approximately one day (20.7 hours) at a temperature of 130°C. There were 70 boards of E. dunnii with dimensions 100mm x 38mm x 4.8m boards (1.28m3). The boards were placed in rows 9 to 13 with E. nitens boards above and E. saligna and E. nitens boards below.  There were 42 boards of E. saligna with dimensions 100mm x 38mm x 4.8m boards (0.77m3). The boards were placed in rows 14 to 16 with E. dunnii and E. nitens boards place above and E. nitens boards placed below.  The temperature of 130°C was reached during the final hours of the drying cycle. The temperature was chosen to try and reduce the amount of time required to dry the boards and see the effects of high temperature on drying quality. Kiln Conditions 140.0 Wetbulb Depression  120.0  Wetbulb Temperature  Drybulb Temperature Relative Humidity  Temp (°C) / Humidity (%)  100.0  Dewpoint  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200  Time (min)  Figure 3.1: Drying conditions for batch.  Compression Kiln R&D  Duncan Fairclough 60  Table 3.1: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards (E. dunnii) Sample Boards (E. saligna)  Average MC (%) 17.2 19.3 16.3 17.6 20.9 16.1  Table 3.2: Stack recovery based on surface quality. Good Average 0% 20% Stack Recovery  Standard Deviation (%) 5.7 6.7 5.1 6.0 4.7 5.0  Poor 80%  *(Based on surface quality)  Figure 3.2: E. dunnii sample #5.  Figure 3.3: E. dunnii sample #3.  Figure 3.4: E. dunnii sample #5 which Figure 3.5: Sample #3 had a grade of had a grade of average due to some poor due to severe surface checking and surface checking. cupping.  Compression Kiln R&D  Duncan Fairclough 61  For the E. dunnii boards, moisture content was not taken into account. If moisture content were included the recovery would have been worse. The main defect seen in most boards was severe surface checking. There was also cupping present in many boards. Also there was severe internal checking in all sample boards. The average moisture content was too high. The temperature will need to be increased or drying time will need to be increased.  Figure 3.6: E. saligna sample #6.  Figure 3.7: E. saligna sample #7  Figure 3.8 (left): E. saligna sample #6 Figure 3.9 (right): Sample #7 had a which had a grade of average due to grade of poor due to severe surface some cupping and surface checking. checking. Moisture content was not taken into account again for the E. saligna boards. If moisture content were included the recovery would have been worse. Although the average MC was around 15% there was great variation in board MC throughout the stack. The main defect seen in most boards was surface checking. There was also cupping present in many boards.  Also there was severe internal checking. The average moisture content  Compression Kiln R&D  Duncan Fairclough 62  was too high. The temperature will need to be increased or drying time will need to be increased.  Compression Kiln R&D  Duncan Fairclough 63  At the end of the drying cycle the average temperature drop across the charge was 2.8°C. Temperature Differences 6.0  5.0  Temperature (°C)  4.0  Upper Temperature Difference  3.0  Lower Temperature Difference  2.0  Combined Average  1.0  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200  -1.0 Time (min)  Figure 3.10: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 64  3.2.2 Batch #8 Summary (E. dunnii) Batch #8 was run over a period of approximately one day (19.3 hours) at a temperature of 130°C. There were 70 boards of E. dunnii with dimensions 100mm x 38mm x 4.8m boards (1.28m3). The boards were placed in rows 10 to 14 with E. nitens boards placed above and below.  The temperature of 130°C was reached during the final hours of the drying cycle. The temperature was chosen to try and reduce the amount of time required to dry the boards. Kiln Conditions 140.0 Wetbulb Depression  120.0  Wetbulb Temperature  Drybulb Temperature Relative Humidity  Temp (°C) / Humidity (%)  100.0  Dewpoint  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140  Time (min)  Figure 3.11: Drying conditions for the batch. Table 3.3: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 19.4 20.4 21.1 20.3 21.3  Table 3.4: Stack recovery based on surface quality. Good Average 20% 60% Stack Recovery  Standard Deviation (%) 5.2 5.6 7.6 6.2 1.7  Poor 20%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 65  Figure 3.12: E. dunnii sample #2.  Figure 3.13: E. dunnii sample #1.  Figure 3.14: E. dunnii sample #2 which Figure 3.15: Sample #1 had a grade of had a grade of good based on surface poor due to severe surface checking and quality. cupping. Moisture content was not taken into account. If moisture content were included the recovery would have been worse. The main defects seen in most boards were surface checking and cupping. There was also collapse present in a few boards. Generally boards with lower moisture content appeared to have fewer surface defects. Also there was severe internal checking in most boards. The average moisture content was too high. The temperature will need to be increased or drying time will need to be increased.  Compression Kiln R&D  Duncan Fairclough 66  At the end of the drying cycle the average temperature drop across the charge was 3.2°C. Temperature Differences 7.0  6.0  5.0 Upper Temperature Difference  Temperature (°C)  4.0  Lower Temperature Difference  3.0  2.0  Combined Average  1.0  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140  -1.0  -2.0 Time (min)  Figure 3.16: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 67  3.2.3 Batch #9 Summary (E. saligna) Batch #9 was run over a period of approximately one day (20.7 hours) at a temperature of 140°C. There were 42 boards of E. saligna with dimensions 100mm x 38mm x 4.8m boards (0.77m3). The boards were placed on the bottom 3 rows of the kiln charge with E. nitens boards placed above.  The temperature of 140°C was never reached. The temperature was chosen as there had been promising results from previous kiln charges dried at this temperature. It is unknown why the final drying temperature of 140°C was never reached. Kiln Conditions 160.0 Wetbulb Depression  140.0  Wetbulb Temperature Drybulb Temperature  Relative Humidity  120.0 Temp (°C) / Humidity (%)  Dewpoint  100.0  80.0  60.0  40.0  20.0  0.0 0  60  120  180  240  300  360  420  480  540  600  660  720  780  840  900  960 1020 1080 1140 1200  Time (min)  Figure 3.17: Drying conditions for the batch. Table 3.5: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 16.0 16.4 18.6 17.0 18.1  Compression Kiln R&D  Standard Deviation (%) 5.3 4.9 6.4 5.7 6.9  Duncan Fairclough 68  Table 3.6: Stack recovery based on surface quality. Good Average 33% 33% Stack Recovery  Poor 33%  *(Based on surface quality)  Figure 3.18: E. saligna sample #3.  Figure 3.19: E. saligna sample #2.  Figure 3.20: E. saligna sample #3 which Figure 3.21: Sample #2 had a grade of had a grade of good. When the sample poor due to severe surface checks. was cross sectioned it contained multiple internal checks. Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. The main defect seen in most boards was surface checking and spring, which was seen to some degree in most of the boards. Generally boards with lower moisture content appeared to have fewer surface defects. Also there was severe internal checking in most boards.  The average moisture content was  acceptable. However, it still could be slightly lower due to a few boards having a much  Compression Kiln R&D  Duncan Fairclough 69  higher than average moisture content. It may be worth adding an extra one or two hours to the drying schedule to achieve this.  At the end of the drying cycle the average temperature drop across the charge was 2.4°C. Temperature Differences 5.0  4.0  Temperature (°C)  3.0  Upper Temperature Difference  2.0  Lower Temperature Difference  1.0  Combined Average  0.0 0  60  120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200  -1.0  -2.0 Time (min)  Figure 3.22: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 70  3.2.4 Batch #11 Summary (E. dunnii) Batch #11 was run over a period of approximately two days (49.4 hours) at a temperature of 90°C. There were 70 boards of E. dunnii with dimensions 100mm x 38mm x 4.8m (1.28m3). The boards were placed in the bottom 5 rows of the kiln charge.  The lower temperature was chosen to create a less harsh drying environment for the timber by decreasing wet bulb depression and maintaining a higher value for relative humidity. Drying time was increased over previous low temperature charges as previous charges had high average moisture contents. Kiln Conditions 100.0 Wetbulb Depression  90.0  Wetbulb Temperature Drybulb Temperature  80.0  Temp (°C) / Humidity (%)  Relative Humidity Dewpoint  70.0 60.0 50.0 40.0 30.0 20.0 10.0  96 0 10 80 12 00 13 20 14 40 15 60 16 80 18 00 19 20 20 40 21 60 22 80 24 00 25 20 26 40 27 60 28 80  84 0  72 0  60 0  48 0  36 0  24 0  0  12 0  0.0  Time (min)  Figure 3.23: Drying conditions for the batch. Table 3.7: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 21.4 20.4 20.3 20.7 20.1  Table 3.8: Stack recovery based on surface quality. Good Average 20% 40% Stack Recovery  Standard Deviation (%) 7.5 6.4 7.2 7.0 6.1  Poor 40%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 71  The main defect seen in most boards was surface checking and collapse. When boards were cross sectioned there was minimal internal checking in the boards.  Figure 3.24: Sample #5D was given a Figure 3.25: Sample #2D was given a grade of good based on surface quality. grade of poor based on severe surface When it was cross sectioned there was checks. severe internal checking present. Moisture content was not taken into account for grading purposes. If moisture content were included recovery would have been worse.  Generally there was minimal  improvement over previous low temperature charges. Defects such as checking, collapse, and cupping were seen in large quantities as previous charges. The moisture content was closer to acceptable levels but was still slightly high. It seems that a longer drying time would be required to bring average MC into the 10 to 15% range.  Compression Kiln R&D  Duncan Fairclough 72  At the end of the drying cycle the average temperature drop across the charge was 0.6°C. Temperature Differences 5.0  Temperature (°C)  4.0  3.0  Upper Temperature Difference  2.0  Lower Temperature Difference  1.0  Combined Average  96 0 10 80 12 00 13 20 14 40 15 60 16 80 18 00 19 20 20 40 21 60 22 80 24 00 25 20 26 40 27 60 28 80  84 0  72 0  60 0  48 0  36 0  24 0  0 12 0  0.0  -1.0  -2.0 Time (min)  Figure 3.25: Temperature difference across load.  Compression Kiln R&D  Duncan Fairclough 73  3.2.5 Batch #18 Summary (E. dunnii) Batch #18 was run as a two part charge over the period of approximately two days (Part#1: 21.0hosurs, Part#2: 15.8 hours).The first part ran at a temperature of 70°C and the second part ran at a temperature of 100°C. There were 70 boards of E. dunnii with dimensions 100mm x 38mm x 4.8m boards (1.28m3). They were located in the bottom 5 rows of the charge.  The two stage drying process was chosen to try and reduce the harshness of the drying conditions to minimize drying defects. The first part was run without the extraction fan running to increase humidity. For the second part the fan was run but only during the final drying stages. Kiln Conditions Part 1 100.0 Wetbulb Depression  90.0  Wetbulb Tem perature Drybulb Tem perature  80.0  Temp (°C) / Humidity (%)  Relative Hum idity Dew point  70.0  60.0 50.0 40.0 30.0 20.0  10.0 0.0  Time (min)  Figure 3.26: Drying conditions for part one of the batch.  Compression Kiln R&D  Duncan Fairclough 74  Kiln Conditions Part 2 120.0 Wetbulb Depression  Wetbulb Tem perature  100.0  Drybulb Tem perature  Temp (°C) / Humidity (%)  Relative Hum idity Dew point  80.0  60.0  40.0  20.0  0.0  Time (min)  Figure 3.27: Drying conditions for part two of the batch. Table 3.9: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 24.8 28.1 21.7 24.9 22.0  Table 3.10: Stack recovery based on surface quality. Good Average 20% 40% Stack Recovery  Standard Deviation (%) 10.5 9.6 6.7 9.4 6.0  Poor 40%  *(Based on surface quality)  Figure 3.28: E. dunnii sample #1D (left) Figure 3.29: Sample #2D had a grade of which had a grade of average due minor poor due to severe surface checks. collapse and spring.  Compression Kiln R&D  Duncan Fairclough 75  Moisture content was not taken into account. If moisture content were included the recovery would have been somewhat worse. About 25% of the boards actually had a moisture content that was within the acceptable range. Boards located in the top quarter seemed to have to best quality as well as better moisture content. In the first part of drying (no exhaust fan running) the humidity was significantly higher than previous charges however during the second part the humidity was comparable to other charges. The main defect seen in most boards was surface collapse and checking. Spring was also present in a couple boards in most rows. The moisture content was too high, most likely due to the boards being dried at lower temperatures for too short a time.  At the end of the drying cycle the average temperature drop across the charge was 0.8°C. Temperature Differences Part 1 12.0  11.0 10.0  9.0 8.0 Upper Tem perature Difference  Temperature (°C)  7.0 6.0  Low er Temperature Difference  5.0  4.0 Com bined Average  3.0  2.0 1.0 0.0 -1.0 -2.0  -3.0 Time (min)  Figure 3.30: Temperature difference across the load for part one.  Compression Kiln R&D  Duncan Fairclough 76  Temperature Differences Part 2 11.0  10.0 9.0  8.0 7.0 Upper Tem perature Difference  Temperature (°C)  6.0 5.0  Low er Temperature Difference  4.0  3.0 Com bined Average  2.0  1.0 0.0 -1.0 -2.0 -3.0  -4.0 Time (min)  Figure 3.31: Temperature difference across the load for part two.  Compression Kiln R&D  Duncan Fairclough 77  3.2.6 Batch #20 Summary (E. sieberi) Batch #20 was run over the period of approximately one day (28.0 hours). The final drying temperature was set to 105°C.  There were 70 boards of E. sieberi with  dimensions 100mm x 38mm x 5.4m boards (1.44m3). They were located in the bottom 5 rows of the charge.  The lower drying temperature was chosen to try and reduce the harshness of the drying conditions to minimize drying defects. Kiln Conditions 120.0 Wetbulb Depression  Wetbulb Tem perature  100.0  Drybulb Tem perature  Temp (°C) / Humidity (%)  Relative Hum idity Dew point  80.0  60.0  40.0  20.0  0.0  Time (min)  Figure 3.32: Drying conditions for the batch. Table 3.11: Moisture contents. Average MC (%) 15.1 Top 1/3 of Stack 14.5 Middle 1/3 of Stack 14.4 Bottom 1/3 of Stack 14.7 Total Stack 15.7 Sample Boards Table 3.12: Stack recovery based on surface quality. Good Average 65% 35% Stack Recovery  Standard Deviation (%) 3.0 3.9 3.0 3.3 3.5  Poor 0%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 78  Figure 3.33: E. sieberi sample #1 which Figure 3.34: Sample #3 had a grade of had a grade of good. poor due to severe spring caused by the board being quarter sawn. Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. Almost all of the boards had a moisture content that was within the acceptable range. The main defect seen in most boards was spring. This was mostly due to the boards being quarter sawn. There was almost no collapse in present in the stack however; some boards did contain some surface checking. The overall quality of the E. sieberi boards was quite good with most boards showing very little in the way of internal defects. Even though the sample size was very small E. sieberi has shown the most promise out of the mainland species.  Compression Kiln R&D  Duncan Fairclough 79  At the end of the drying cycle the average temperature drop across the charge was 1.0°C. Temperature Differences 10.0  9.0 8.0  Temperature (°C)  7.0 Upper Tem perature Difference  6.0 5.0  Low er Temperature Difference  4.0 Com bined Average  3.0 2.0 1.0 0.0  -1.0 Time (min)  Figure 3.35: Temperature difference across the load.  Compression Kiln R&D  Duncan Fairclough 80  3.2.7 Batch #21 Summary (E. dunnii) Batch #21 was run over a period of approximately one day (28.0 hours) at a temperature of 105°C. There were 120 boards of E. dunnii with dimensions 75mm x 25mm x 4.8m boards (1.08m3).  As the boards were thinner than most boards the lower temperature was chosen to try and reduce the harshness of the drying conditions in hopes that it may improve overall board quality. Table 3.13: Moisture contents Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 11.1 12.1 11.4 11.5 9.8  Standard Deviation (%) 2.9 3.3 3.3 3.2 2.6  Table 3.14: Stack recovery based on surface quality. Good Average 60% 30% Stack Recovery  Poor 10%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. The main defects seen in most boards were similar to that of the larger dimension boards with collapse and surface checking being the most common. When surface checking appeared it was often quite severe and would split the entire board. Overall there was no significant difference in the timber quality of the smaller dimension timber and the larger dimension timber.  For an unknown reason the kiln was unable to trend the data from the charge so it is not possible to determine the values for relative humidity, dew point, etc.  Compression Kiln R&D  Duncan Fairclough 81  3.3 E. dunnii, E. saligna, & E. sieberi Trial Summary Over the course of the study approximately 11m3 of E. dunnii was dried. Out of all three mainland species that were tested it showed the least amount of promise. A variety of schedules were tried; ranging from high temperature with fast drying times (140°C for 22 hours) to low temperature with slow drying times (90°C for 48 hours). The schedules that produced the best results were the lower temperature schedule which was expected as the drying conditions would be very mild. However, despite these mild conditions the overall quality of the timber was quite poor based on surface and internal grading. Severe surface and internal checking was often present in many of the boards. Even when surface quality was quite good there would often be severe internal checking causing the timber to be worthless. It is also worth noting that after drying much of the timber would begin to cup making it very difficult for the Joulin stacking machine to pick up the boards. In many cases the boards would have to be manually removed from the stack as the Joulin was just unable to lift the boards.  Table 3.15: Drying schedules summary. Batch Temperature Drying Time Notes and Change from Previous # (°C) (Hours) Charge Drying temperature was not reached. MC 130 20.7 was too high. Surface quality of boards 7 was poor. Shorter drying time. Quality was worse 130 19.3 8 due to boards not being dry. Lower temperature longer drying time. 90 49.4 Board quality was similar to batch #8. 11 MC was too high. Two part schedule. For part 1 extraction fan were not run to increase relative 70 then 100 36.8 humidity. Lower temperature longer 18 drying time. Same quality as batch #11. MC was too high. MC was good. Board surface quality was 105 28.0 good. Thinner boards were used (25mm 21 thick) As there was a limited supply of E. saligna logs available only 5m3 was dried. The drying results for the species were better than that of E. dunnii however, the overall Compression Kiln R&D  Duncan Fairclough 82  results were still not very positive. Again a variety of schedules were used to dry the timber ranging from high to low temperature with fast to slow drying times. Similar to E. dunnii the best results came from timber that was dried at lower temperatures. The main defects seen in the species were surface and internal checking. Though not as frequent as the checking found in the E. dunnii there were still enough to cause significant down grade. As with E. dunnii many of the boards were cupped after drying and the Joulin was unable to lift these boards.  Table 3.16: Drying schedules summary. Batch Temperature Drying Time Notes and Change from Previous # (°C) (Hours) Charge Drying temperature was not reached. MC 130 20.7 was too high. Surface quality of boards 7 was average. Higher temperature same drying time. 140 20.7 Surface quality was similar. MC was 9 closer to acceptable range. Same temperature increased drying time. 140 22.7 Boards were over dried. Overall surface 15 quality was average to poor. E. sieberi was the final species to be tested. There was an extremely limited supply of the species to the point that only one charge containing the species was run. The volume of timber dried was 1.5m3. It is unfortunate that there was such a limited supply of the species as the dried quality showed great promise. The species was run at a temperature of 105°C for 28 hours. The main defect seen was spring however; this can mainly be put down to the boards being quarter sawn rather than a drying defect. Unlike E. dunnii and E. saligna E. sieberi had very little surface and internal checking. When internal checks were found they were often very small in size. Also the surface quality of the boards was quite good with almost no collapse or cupping present. This allowed for the Joulin to actually be able to lift the boards off of the stack.  For the mainland species study it would be safe to assume that with more testing E. sieberi would prove to be the best species out of the three for drying in the compression  Compression Kiln R&D  Duncan Fairclough 83  kiln. This would be followed by E. saligna and E. dunnii, the latter two species being quite poor in quality when compared to E. sieberi.  Compression Kiln R&D  Duncan Fairclough 84  4.0 Pine Trial 4.1 Radiata Pine Radiata Pine (Pinus Radiata) was the first species used for drying trials in the compression kiln; specifically boards made from the heartwood of the log were chosen for this drying trial. The trials have been done on pine boards with dimensions of 100 x 38mm as they were readily available and earlier trials on the species had shown some promise.  When radiata pine is grown in a plantation situation the tree will usually grow tall and straight. The exception to this is on trees located at the edge of the plantation where there is access to light and space in these areas trees will often grow large branches. There has been a great deal of work done on the genetics of the tree and silvicltural practices to improve overall characteristics of the tree such as growth rates and form of plantation grown trees. Trees used for solid wood products are generally harvested at an age of 25 to 30 years. The tree is susceptible to a small number of diseases and pests which can attack pine plantations.  Most if these can be controlled through proper forest  management plans.  Radiata pine wood is pale yellow or brown in colour and is generally straight grained with prominent growth rings. The green density is around 1000 kg/m3 and the dry (12% moisture content) density is 300 to 400 kg/m3. If the harvesting and processing of the logs is not done quickly enough the wood is prone to a fungus that leaves a blue stain in the wood. This blue stain does not damage the wood durability and strength however it does limit the woods use in products used for appearance.  In general radiata pine is relatively easy to dry, however the heartwood is prone to twisting when dried. The wood can be dried quite rapidly at high temperatures with some commercial operations drying at 180°C (some operations dry at 200°C).  At these  temperatures the drying time is extremely short usually around 3 to 5 hours. In these rapid drying schedules the wood is prone to twisting; to moderate this wood should be Compression Kiln R&D  Duncan Fairclough 85  weighted down during the drying process and pre-steamed for several hours prior to drying. The pre-steaming phase is generally used for evening out the moisture content among the boards in the drying charge however recent studies have shown that there is little benefit to doing this and pre-steaming is becoming less common. After the timber has been dried it is often steamed again this is done to even out the moisture gradient of the board (board surfaces will be very dry compared to the core) and the stress gradients.  Figure 4.1: 100x38mm pine heartwood boards containing twist and spring. The smaller pine heartwood timber used for the preliminary drying trials is generally sold as a low value product. This makes it difficult to justify using the compression kiln to dry the material, as the total cost of the kiln including overhead costs and labor costs is quite high. However using the data gained from this smaller timber it could be possible to develop a drying schedule for larger section pine timber that could be sold as a high value product to furniture manufacturers and other secondary manufacturing companies.  Compression Kiln R&D  Duncan Fairclough 86  4.2 Radiata Pine Batch Summaries The following pages contain batch summaries of various trials. 4.2.1Batch #1 Summary (Pine) Batch #1 was run over a period of approximately one day (20.0 hours) at a temperature of 115°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3).  The temperature was chosen to set a starting point for future trials to be based off. Table 4.1: Moisture contents. Sample Boards  Average MC (%) <5  Table 4.2: Stack recovery based on surface quality. Good Average 0% 0% Stack Recovery  Standard Deviation (%) n/a  Poor 100%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been the same due to the timber being over dried. The main defects seen in the stack were bow twist and spring. Twist was the most common defect seen with almost 100% of the boards containing it. This was followed very closely by bow and spring with the amount of 80% and 70% respectively.  There were some hydraulic issues which caused the charge to remain in the drying chamber after the drying cycle had finished. This means that although the schedule was only 20 hours long the charge remained in the drying chamber for 47 hours during which time the temperature in the kiln was still above 60°C. The stack did not have a proper preheat cycle before drying.  Drying time will need to be decreased to achieve more acceptable average moisture content in the charge. This will also hopefully minimize the frequency and severity of the defects listed above.  Compression Kiln R&D  Duncan Fairclough 87  4.2.2 Batch #2 Summary (Pine) Batch #2 was run over a period of approximately one day (20.0 hours) at a temperature of 115°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3). Table 4.3: Moisture contents. Sample Boards  Average MC (%) 7.3  Table 4.4: Stack recovery based on surface quality. Good Average 0% 65% Stack Recovery  Standard Deviation (%) 1.7  Poor 35%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been the same due to much of the timber being over dried. The main defects seen in the stack were bow, twist, and spring. Twist was the most common defect seen with 50% of the boards containing it. This was followed very closely by bow and spring with the amounts of 40% and 30% respectively.  There were some hydraulic issues which caused the charge to remain in the drying chamber after the drying cycle had finished. This means that although the schedule was only 20 hours long the charge remained in the drying chamber for 27 hours during which time the temperature in the kiln was still above quite high. There was an improvement in the quality of the timber over the previous charge (batch #1). Even though the schedules were identical batch #1 remained in the drying chamber for a longer period of time. This caused over drying of the timber which most likely resulted in a higher defect rate.  Drying time will need to be decreased to achieve more acceptable average moisture content in the charge. This will also hopefully minimize the frequency and severity of the defects listed above.  Compression Kiln R&D  Duncan Fairclough 88  4.2.3Batch #4 Summary (Pine) Batch #4 was run over a period of approximately one day (20.0 hours) at a temperature of 105°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3).  The temperature was chosen to see if a lower temperature would decrease the amount of defects seen by reducing the harshness of the drying conditions. Table 4.5: Moisture contents. Sample Boards  Average MC (%) 11.5  Standard Deviation (%) 3.9  Table 4.6: Stack recovery based on surface quality. Good Average 40% 40% Stack Recovery  Poor 20%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been the same due. The main defects seen in the stack were bow and twist. The lower temperature seemed to greatly decrease the amount of both defects seen. Also when the defects were present they were not as severe as in previous charges. In most rows there were 1 or 2 boards that contained major twist.  With the lower temperature some of the boards were not dried (moisture content >25%). This was expected as the drying time was not changed from the time used on the higher temperature charges. It is also worth noting that there were significant differences in the board quality seen in various locations in the stack. The centre 1/3 of the stack contained the best quality boards. Boards located on the top and bottom of the stack had similar quality. Twist and bow was more prevalent in boards located in these areas of the stack.  Compression Kiln R&D  Duncan Fairclough 89  4.2.4 Batch #6 Summary (Pine) Batch #6 was run over a period of approximately one day (21.0 hours) at a temperature of 105°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3).  The temperature was chosen to see if a lower temperature would decrease the amount of defects seen by reducing the harshness of the drying conditions. The drying time was increased by an hour over previous charges to try and ensure that all boards were dried to acceptable moisture content. Table 4.7: Moisture contents. Sample Boards  Average MC (%) 11.9  Table 4.8: Stack recovery based on surface quality. Good Average 50% 40% Stack Recovery  Standard Deviation (%) 3.4  Poor 10%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been the same due. Overall the moisture content of most boards was at an acceptable level. The main defect seen in the stack was twist. Bow and spring were seen in a few boards however not nearly as much as twist. The lower temperature seemed to greatly decrease the amount of both defects seen.  In previous charges with lower temperature some of the boards were not dried (moisture content >25%) this charge used a slightly longer drying time to ensure all boards were dried adequately. This seemed to work as most of the boards were dried to an acceptable level. There were some cases where boards were either too dry or not dry enough, however in reality this is to be expected as it would be nearly impossible to have all the boards dried within the proper moisture content range.  There was an improvement in the stack recovery over previous charges as can be seen in the above table. It is also worth noting that there were significant differences in the board Compression Kiln R&D  Duncan Fairclough 90  quality seen in various locations in the stack. The centre 1/3 of the stack contained the best quality boards. Boards located on the top and bottom of the stack had similar quality. Twist and bow was more prevalent in boards located in these areas of the stack.  Compression Kiln R&D  Duncan Fairclough 91  4.2.5 Batch #7 Summary (Pine) Batch #7 was run over a period of approximately one day (19.0 hours) at a temperature of 105°C. There were 238 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (4.88m3).  The temperature was chosen to see if a lower temperature would decrease the amount of defects seen by reducing the harshness of the drying conditions. The drying time was decreased by two hours over previous charges. Table 4.9: Moisture contents. Sample Boards  Average MC (%) 8.9  Table 4.10: Stack recovery based on surface quality. Good Average 25% 50% Stack Recovery  Standard Deviation (%) 2.0  Poor 25%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been somewhat worse due to over drying. This over drying was most likely due to the fact that the charge had two pre-drying stages due to the stack tipping in the pre-dry chamber requiring it to be restacked and sent through the drying process again. The main defects seen in the stack were twist and bow. Spring was seen in a few boards however not nearly as much as twist and bow. It will be difficult to draw any concrete conclusions from this report due to the charge tipping affect the overall drying schedule.  Compression Kiln R&D  Duncan Fairclough 92  4.2.6 Batch #8 Summary (Pine) Batch #8 was run over a period of approximately one day (19.0 hours) at a temperature of 105°C. There were 238 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (4.88m3).  The temperature was chosen to see if a lower temperature would decrease the amount of defects seen by reducing the harshness of the drying conditions. The drying time was decreased by two hours over previous charges. The main test for this charge was to determine what effect not having a pre-dry sequence would have on the deviation of the moisture content of the charge. Table 4.11: Moisture contents Sample Boards  Average MC (%) 11.1  Table 4.12: Stack recovery based on surface quality. Good Average 40% 40% Stack Recovery  Standard Deviation (%) 4.7  Poor 10%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. The overall stack quality was very similar to other charges dried under the same schedule with the main defects seen in the stack being twist and bow. Spring was seen in a few boards however not nearly as much as twist and bow.  The main test for the charge was to see if the standard deviation of the moisture content would be higher due to the charge not having a pre-dry sequence. The purpose of the pre-drying chamber is to reduce the moisture gradient of the board in the stack. From this charge it can be seen that the pre-drying cycle does not have any significant effect on the moisture content deviation.  Compression Kiln R&D  Duncan Fairclough 93  4.2.7 Batch #13 Summary (Pine) Batch #13 was run over a period of approximately one day (18.0 hours) at a temperature of 110°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3).  The temperature was chosen to see if drying time could be reduced while maintain decent quality of the boards by increasing the temperature. Table 4.13: Moisture contents. Sample Boards  Average MC (%) 11.2  Table 4.14: Stack recovery based on surface quality. Good Average 60% 30% Stack Recovery  Standard Deviation (%) 2.8  Poor 10%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been very similar. The overall stack quality was good with the main defects being twist and spring. There was a drop in quality from rows 12 through to 17. Quality picked up again for the bottom 4 rows of the charge.  From the results of this charge it seems that it is possible to maintain a decent level of quality in the boards while drying at higher temperatures if the drying time is decreased.  Compression Kiln R&D  Duncan Fairclough 94  Figure 4.2: Sample #3 came from row Figure 4.3: Sample #7 came from row #5; both the row and board were given a #13; both the row and board were given quality rating of good. a quality rating of poor. Both the sample and the row contained twist which resulted in their quality rating.  Compression Kiln R&D  Duncan Fairclough 95  4.2.8 Batch #15 Summary (Pine) Batch #15 was run over a period of approximately one day (17.5 hours) at a temperature of 115°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3).  The temperature was chosen to see if drying time could be reduced by increasing the temperature while still maintaining decent quality of the boards. Table 4.15: Moisture contents. Sample Boards  Average MC (%) 8.1  Table 4.16: Stack recovery based on surface quality. Good Average 10% 60% Stack Recovery  Standard Deviation (%) 2.8  Poor 30%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been worse due to many of the boards being over dried. The main defects seen in the charge were twist, bow, and spring. All of the defects were seen in similar amounts throughout the charge. Boards located on the outside edges of the stack were most susceptible to the defects. Boards located in the center of the row appeared to be of slightly higher quality than those on the outer edges.  Although the drying time was only set to 17.5 hours the charge remained in the drying chamber for approximately 37 hours due to hydraulic issues. This may have been the cause for the timber being over dried which may have caused the overall stack quality to be poor.  Compression Kiln R&D  Duncan Fairclough 96  Figure 4.4: Sample #3 came from row #5; the board was given a quality rating of good while the row was given a rating of average.  Figure 4.5: Sample #1 came from row #1; both the row and board were given a quality rating of poor. Both the sample and the row contained twist which resulted in their quality rating.  Compression Kiln R&D  Duncan Fairclough 97  4.2.9 Batch #20 Summary (Pine) Batch #20 was run over a period of 16.5 hours at a temperature of 120°C. There were 294 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (6.03m3).  The temperature was chosen to see if drying time could be reduced by increasing the temperature while still maintaining decent quality of the boards. Table 4.17: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 6.8 6.7 6.1 6.5 6.8  Standard Deviation (%) 1.4 2.4 1.5 1.8 1.8  Table 4.18: Stack recovery based on surface quality. Good Average 10% 30% Stack Recovery  Poor 60%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been worse due to many of the boards being over dried. The main defects seen in the charge were twist, spring. Bow was seen in some areas of the stack but was nearly as common as the other two defects. Quality seemed to drop off in the bottom 2/3 of the stack with the best quality boards located near the top of the stack.  The schedule did show promise as there were some boards that were of good quality. These boards tended to be the ones with moisture contents within the acceptable range. If drying time were reduced it may be possible to increase the average moisture content to an acceptable level and possibly reduce the amount of defects seen.  Compression Kiln R&D  Duncan Fairclough 98  Figure 4.6: Sample #4 came from row Figure 4.7: Sample #5came from row #9; #7; both the row and the board were both the row and board were given a given a quality rating of good. quality rating of poor. Both the sample and the row contained twist which resulted in their quality rating.  Compression Kiln R&D  Duncan Fairclough 99  4.2.10 Batch #23 Summary (Pine) Batch #23 was run over a period of 14 hours at a temperature of 130°C. There were 292 boards of radiata pine with dimensions 100mm x 38mm x 5.4m boards (5.99m3).  The temperature was chosen to see if drying time could be reduced by increasing the temperature while still maintaining decent quality of the boards. Table 4.19: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 7.4 7.3 7.6 7.4 7.1  Standard Deviation (%) 1.6 2.4 3.0 2.4 2.3  Table 4.20: Stack recovery based on surface quality. Good Average 40% 20% Stack Recovery  Poor 40%  *(Based on surface quality)  Moisture content was not taken into account. If moisture content were included the recovery would have been worse due to many of the boards being over dried. The main defects seen in the charge were twist and bow. Spring was seen in some areas of the stack but was nearly as common as the other two defects. The best quality boards seemed to come from the middle of the stack. Also boards located on the outer edges of the charge were more susceptible to defects and being over dried. Board quality was divided into two distinct groups either being good or poor with very few in the middle range of quality.  The schedule did show promise as there were some boards that were of good quality. These boards tended to be the ones with moisture contents within the acceptable range. If drying time were reduced it may be possible to increase the average moisture content to an acceptable level and possibly reduce the amount of defects seen.  Compression Kiln R&D  Duncan Fairclough 100  4.3 Pine Standard Rack & Reverse Twist Testing 4.3.1 Batch #25 Summary (Pine Standard Rack) Batch #25 was run over a period of just under one day (18.8 hours). The final drying temperature was set to 115°C. There were 392 boards of pine with dimensions 100mm x 38mm x 6.0m boards (8.94m3).  The temperature of 115°C was chosen as previous charges run at this temperature had positive results in dry timber quality. Kiln Conditions 140.0 Wetbulb Depression  120.0  Wetbulb Tem perature Drybulb Tem perature  Relative Hum idity  Temp (°C) / Humidity (%)  100.0  Dew point  80.0  60.0  40.0  20.0  0.0 0  120  240  360  480  600  720  840  960  1080  Time (min)  Figure 4.8: Drying conditions for batch. Table 4.21: Moisture contents. Top 1/3 of Stack Middle 1/3 of Stack Bottom 1/3 of Stack Total Stack Sample Boards  Average MC (%) 8.2 8.6 8.7 8.5 8.6  Table 4.22: Stack recovery based on surface quality. Good Average 30% 40% Stack Recovery  Standard Deviation (%) 2.3 2.3 2.1 2.3 1.0  Poor 30%  *(Based on surface quality)  Compression Kiln R&D  Duncan Fairclough 101  Figure 4.9: Sample #3 which had a grade Figure 4.10: Sample #4 had a grade of of good. poor due to severe spring. Moisture content was not taken into account. If moisture content were included the recovery would have been slightly worse. As the rack had been sitting outside for two months prior to entering the kiln some of the timber had begun to dry out, causing some of the boards to have moisture contents that were too low. As the drying schedule was based on timber that was very green when entering the kiln the final drying time was too long. The main defects seen in most boards were twist and spring. This was mostly due to the boards having been already quite dry prior to entering the kiln.  Compression Kiln R&D  Duncan Fairclough 102  At the end of the drying cycle the average temperature drop across the charge was 1.2°C. Temperature Differences 10.0 9.0 8.0  7.0 6.0 5.0  Temperature (°C)  4.0 3.0  Upper Tem perature Difference  2.0 1.0  Low er Temperature Difference  0.0 -1.0 -2.0  Com bined Average  -3.0 -4.0  -5.0 -6.0 -7.0 -8.0 -9.0 -10.0  Time (min)  Figure 4.110: Temperature difference across the load.  Figure 4.1: Crushed rack sticks.  Figure 4.13: Rack stick marks on pine boards.  As the compression kiln would apply more weight to the stack than a standard kiln there was some concern that the rack sticks my cause indentation in the boards. The rack sticks did cause some indentation in the boards but in most cases could be removed in the Compression Kiln R&D  Duncan Fairclough 103  drymill (figure 2.3.5). Often the rack sticks would be severely crushed, in these cases more significant marks would be left in the timber (figure 2.3.4).  Compression Kiln R&D  Duncan Fairclough 104  4.3.2 Batch #26 Summary (Pine Wedge Rack) As twist was found to be one of the major defects that occurs during the drying of pine there has been some research done on utilizing reverse twist by using wedges. Basically the theory behind reverse twist is timber generally only twist in one direct (clockwise when viewed from the end). By using wedges to create a positive twist (anti-clockwise direction) the net result is zero twist. Early results have shown some positive results and it was determined to try and duplicate these results using the compression kiln.  Figure 4.14: Pine rack being loaded onto wedges.  Figure 4.15: Pine rack being loaded onto wedges.  Unfortunately as the racks from the greenmill had been sitting outside for two months they had become quite dry. This prevented the timber from flexing when it was placed on the wedges. Due to this the rack was not very stable and during the pre-drying phase many of the wedges fell over. This caused the whole rack to shift and made it impossible to move it through the kiln. At this point the study was stopped as time was a limiting factor. For the study to be run again it would be crucial to have timber that was very green which would allow the timber to flex on the wedges.  Compression Kiln R&D  Duncan Fairclough 105  Figure 4.16 Pine rack with wedges that have fallen over during pre-dry stage due to uneven lifting of the stack.  Compression Kiln R&D  Duncan Fairclough 106  4.4 Pine Hardness Testing TeknoComp has claimed that one of the advantages to the compression kiln is that timber dried in the kiln will have an increased hardness value over conventionally dried timber. To test this theory a sample of 30 pieces of timber dried in the compression kiln and 30 pieces of conventionally dried timber were tested using the Janka hardness test. The sample pieces were 90x35 pine heartwood samples cut to a length of approximately 300mm. The compression kiln boards were spray painted pink and labelled 1 to 30 and the control boards were painted yellow and again labelled 1 to 30. The side with the number on it was designated face 1 and the opposite side was designated face 2. As the probe diameter is only 10mm all hardness tests were done in the earlywood of each sample, this was done to maintain some form of control during the experiment. The samples were tested once on each face. Hardness values were determined at exactly a depth of 5.64mm by determination of slope in the range 5.59-5.66mm and calculating the intercept at 5.64mm.  Moisture a content of each board was also determined using a resistance moisture meter (Delmhorst J2000), setup to measure pine and correcting for temperature and density. Moisture content readings were taken from the centre of the sample board entering from Face 1.  Compression Kiln R&D  Duncan Fairclough 107  Table 4.23: Pine hardness values.  Compression Kiln R&D  Duncan Fairclough 108  Hardness Values Comparison of Average (Both Faces) 4.00  3.50 Compression Kiln  3.00 Hardness Value (kN)  Control 2.50  2.00  1.50  1.00  0.50  0.00  Figure 4.17: Hardness values comparison. From the above data it can be seen that there is little difference between the hardness of the control and compression kiln boards. The combined average for both faces of the compression kiln boards was 2.34kN which is slightly less than the combined average for both faces of the control boards as 2.35kN. Despite the small sample size it would be safe to assume that the hardness value of timber dried in the compression kiln is not affected one way or another.  Compression Kiln R&D  Duncan Fairclough 109  4.5 Pine Bending Tests To test the claim that the compression kiln increased bending strength of timber 30 samples of timber dried in the compression kiln were sent to the drymill to be moulded down to 90x35mm. From this same moulder run 30 sample boards that were dried in conventional kilns were also selected to be the control batch. The bending test was carried out on the four point bending rig in the lab. Table 4.24: Compression bending test values.  kin  boards Table 4.25: Control boards bending test values.  Compression Kiln R&D  Duncan Fairclough 110  Bending Test Comparison 70.00  60.00  50.00  Compression Kiln Boards Control Boards  MOR (Mpa)  40.00  30.00  20.00  10.00  0.00  Figure 4.18: Bending test comparison. From the above data it can be seen that the compression kiln boards had a slightly higher average value for MOR (Modulus of Rupture) but a slightly lower average for MOE (Modulus of Elasticity). Overall there was very little difference between the compression kiln boards and the control boards. Due to the small sample size it is difficult to make any concrete conclusions, to determine if TeknoComp’s claim of increased bending strength is true or not further testing would need to be done.  Compression Kiln R&D  Duncan Fairclough 111  4.6 Pine Trial Summary Although only 38mm thick material has been tested the results from these early tests of pine heartwood have shown some promise. It is impossible to eliminate all drying defects however they have been minimized through the development of the drying schedules. For this trial approximately 144m3 of pine was dried.  It seems that pine heartwood dries quite well at a higher temperature which allows overall drying times to be decreased as the water is removed from the wood more rapidly. Initially lower temperatures were used with longer drying times which produced good quality dry timber. This allowed a baseline to be established to refer back to as changes were made to the drying schedule.  From this point on drying times have been  progressively decreased and drying temperatures have been slowly ramped up as needed to achieve the desired moisture content.  Currently the maximum temperature used in the compression kiln has been 115°C with a drying time of approximately 17 hours. The quality of timber being produced by this schedule has been acceptable. The next set of trials will be aimed at decreasing the drying time further by drastically increasing temperature up to 150°C. It is hoped that these higher temperatures will not affect quality and will allow for a significant increase in the total output volume of the kiln.  As stated earlier in the report in order for the compression to be feasible drying times must be kept to a minimum to maximize the amount of volume it produces. One of the major problems has been achieving a consistent board quality across the entire kiln charge. There were problems with the boards located on the side of the kiln charge next to the heat exchangers were often over dried. These boards frequently contained defects such as twisting and bowing. It is hoped that this will be overcome through higher drying temperatures and shortened drying times. The theory behind this is at higher temperatures water moves faster within the board increasing air humidity, this  Compression Kiln R&D  Duncan Fairclough 112  should lower the moisture content gradient within each board which will lower the moisture content gradient across the stack.  Generally the 100x38mm pine heartwood trials have shown promise in the quality of timber produced. The next phase will be to test larger thicknesses to see if similar results are achievable. The thicker pine material if dried to a decent quality could be dried in lower volumes (suiting the compression kiln) and sold as a high value product. One foreseeable problem with drying this larger dimension timber would be building the kiln stacks as the Joulin would likely struggle with the greater weight of the timber. Unfortunately it was not possible to test the larger sections due to time restrictions.  Table4.26: Summary of drying schedules. Batch Temperature Drying Time Notes and Change from Previous # (°C) (Hours) Charge Boards were over dried. Overall quality 115 20.0 1 was poor. MC was slightly higher. Still overall 115 20.0 2 quality was poor. Board quality was better. Moisture 105 20.0 4 content was within an acceptable range. Drying time was increased. Quality was 105 21.0 6 slightly better. MC was good. Drying time was decreased. Overall 105 19.0 quality was average to poor. MC was too 7 low. No change in schedule. Board quality was 105 19.0 8 average too good. MC was acceptable. Temperature was increased drying time 110 18.0 was decreased. Board quality was good. 13 MC was good. Temperature was increased drying time 115 17.5 was decreased. Board quality was poor to 15 average. MC was too low. Temperature was increased drying time 120 16.5 was decreased. Boards were over dried. 20 Overall quality was poor. Temperature was increased drying time was decreased. MC was just below 130 14.0 23 acceptable levels. Board quality was average. Compression Kiln R&D  Duncan Fairclough 113  5.0 Conclusions, Difficulties, & Recommendations 5.1Joulin The Joulin is a key component to the compression kiln as it is the machine that is used to build and take down stacks that are dried in the compression kiln. It consists of nine suction pads that are used to pick up the timber and aluminum sheets that are placed inbetween the timber. It is essentially the first and last step to the compression kiln drying process.  For the first part of the study that only involved pine timber the Joulin stacking machine worked relatively well however, it still did drop some of the timber and aluminium sheets from time to time. It would pick up green pine timber with little difficulty as generally the timber was relatively flat allowing the vacuum pads to get good suction however, if there was standing water on the timber from being left in the rain the Joulin would be unable to pick up the timber. The Joulin would struggle a great deal with dried pine that contained twist due to the vacuum pads not having a flat surface to create suction. In many cases the Joulin would be unable to pick up entire rows. In these cases it was mandatory to manually take down a stack.  For the second part of the study mainly hardwood timber was used. It was at this point that the Joulin really began to struggle. Again it would be able to pick up the green timber as long as there was no standing water on the boards, but it would struggle to pick up the dried timber. This was especially true with the mainland species which were cut to smaller dimensions and more prone to cupping.  Compression Kiln R&D  Duncan Fairclough 114  Figure 5.1: Aluminium sheet that has Figure 5.2: Aluminium sheet that has been dropped by the Joulin. been dropped by the Joulin. It was also noticed that stacks that contained only E. nitens would have a bow in them after drying leaving the two ends of the stack lower than the middle. This caused a problem as the Joulin’s outside edge vacuum pads were not able to grip the aluminium sheets which resulted in many sheets being dropped. This caused further problems as after a sheet is dropped it loses a great deal of rigidity. The aluminum sheets are made up of many smaller aluminum bars that are held together with clips (the bars are hollow which allow for airflow during the drying process). When a sheet is dropped the clips get bent and do not hold the individual bars together which causes a loss in the overall sheets rigidity. When a sheet loses its rigidity is picked up by the Joulin it begins to droop between the suction pads, when this happens the Joulin loses suction as the sheet begins to pull away from the vacuum pads, this in turn causes to the sheet to fall. After a sheet has been dropped three or four times it generally becomes useless as it is bent too much to be reused.  To allow the Joulin to function properly the design of the aluminum sheets would need to be changed. Ideally the sheet would be made up of one piece however this would most likely not be possible. An alternative to this would be making the sheets out of three or four larger pieces to allow for a more rigid structure by eliminating the vast majority of the clips. This could be achieved by welding many of the existing small bars (Figure 5.4) together into larger pieces then three or four of these larger pieces could be joined  Compression Kiln R&D  Duncan Fairclough 115  together using clips to create a full sheet. As time ran out on the project it was not possible to test this theory.  Figure 5.3: Aluminium sheet.  Figure 5.4: Top view of aluminium sheet showing edge of sheet with clips.  However, the best case scenario for loading and unloading charges from the kiln would be to remove the Joulin and utilize standard kiln racks (designed for the compression kiln) and use a forklift to load and unload them. The ideal racks would be 6m long approximately 1.7m tall and have a width between 1.4m and 1.5m. It is very important to ensure that the width of the stack is inside this range, as there is very little clearance in the kiln.  Compression Kiln R&D  Duncan Fairclough 116  5.2 Compression Kiln The first problem that was been encountered during this project is the re-commissioning of the compression kiln.  At the beginning of the year FEA Timber switched all  production to their new site. As the new mill was still in the process of troubleshooting not all of the equipment was brought to the new site. Machines deemed to be not vital to production such as the compression kiln were left at the old sawmill’s site. This meant that the compression kiln had not been in use for a period of time during which little to no maintenance was done on the kiln to keep it ready for operation. The first couple of months were mainly spent doing maintenance to the kiln and re-commissioning it. Also new software was needed to make the kiln more user-friendly and allow for certain kiln parameters to be tracked. The software had not been completely installed before the trials started and it needed to be tested ensure it was working correctly prior to the project starting. This new software did improve the user friendliness of the kiln however it still did contain some bugs. The main issue was every so often it would trip a variety of errors for no apparent reason which would shutdown the kiln. This proved to be quite a problem as it would slow production.  The hydraulic system that operated the rams inside the kiln was not functioning at 100% for most of the study and caused some major problems. During the drying process the rams in the drying chamber would often lose pressure very quickly causing the hydraulic pump’s electric motor to run continually. This in turn would cause a circuit breaker to trip and would stop the kiln from running. The most likely cause of this loss of hydraulic pressure is a worn seal in one or more of the rams.  During the study there were multiple cases of kiln charges coming off the chain conveyors inside the kiln. This was generally caused by the hydraulic rams not raising and lowering the stack evenly. The problem was never solved as it occur sporadically however the most likely cause of the problem was the severe corrosion that was present on most of the hydraulic rams. This corrosion was causing individual rams to stick while a stack was being raised or lowered, when this happened the stack would tip and slide to one side causing it to come off the chain conveyors. Compression Kiln R&D  Duncan Fairclough 117  If the kiln is to be put back into production it will need a significant overhaul. All of the conveyor chains and bearings would need to be greased and oiled. Also a larger drive motor for the chain conveyor in drying chamber would be needed, as sometimes the motor would not be able to move a charge out of the drying chamber. Some form of moisture content measuring system would need to be implemented to give the operator some idea as to what the moisture content of the charge is during the drying process, as currently there is no was of measuring the moisture content during drying. Although the software has been upgraded there is still room for improvement.  Features such as  calculating relative humidity and tracking it during the drying process would be useful features as currently the data has to be downloaded to a computer then humidity has to be calculated from the data.  There is a great deal of corrosion in many of the internal areas of the kiln; this is mainly due to the wrong type of materials being used (mild steel rather than stainless steel). The door between the pre-drying and drying chamber was one location where this was quite evident. On many occasions this door would not be able to open and close fully due to large buildups of rust on the door’s runners. Also the chain conveyor in the drying chamber was prone to rust and constantly needed to be oiled to allow it to run.  Before the kiln is to be run for any significant amount of time it would need a thorough going over by maintenance ensuring all chains are well oiled and bearing are greased. Also it would be worth upgrading the drying chamber’s conveyor motor to something with more power. The entire hydraulic system should be checked to ensure there are no external leaks in lines or leaks in the seals. All hydraulic pressure sensors should also be checked and calibrated. The hydraulic pump should also be checked to ensure it is still in good working order. Many of the smaller electric motors such as the one that drives the hydraulic pump would also need to be upgraded as in many cases they were under powered. Finally, once this is all completed it may be worth upgrading the software even further to make the kiln easier to use, as well as implanting some form of moisture content measuring system s to track the moisture content of the charge during drying. Compression Kiln R&D  Duncan Fairclough 118  5.3 General One major issue with drying timber in the compression kiln is the extremely high running costs. The energy costs for the kiln are quite high and the amount of volume that it is capable of producing per charge is quite low. The claim was made that the kiln would be capable of drying timber that was well above fibre saturation point, this would cut inventory costs by eliminating the need to have timber racks left outside in the yard to be air dried. This was shown to not be true as even though the kiln was capable of drying the timber to acceptable moisture content the quality of the timber was quite poor.  Supplying timber for the kiln has also been an issue for the project. As all of the Eucalyptus dunnii and Eucalyptus saligna has come from plantations located in New South Wales it has taken a great deal of time to source decent quality logs due to the limited amount of mature plantation resources (majority of plantations are only 10 years old or younger). The logs then needed to be cut by a third party sawmill in New South Wales as FEA Timber’s current sawmill is not set up to handle small production runs of test logs. Also there was a very limited supply of E. sieberi logs available. Due to this it would be very difficult to make any concrete conclusions from the study.  If the  compression kiln were to be run in the future and further mainland species were to be tested E. sieberi would be a good starting point.  Compression Kiln R&D  Duncan Fairclough 119  References Blakemore, P. A. 2008. Optimization of steam reconditioning for regrowth-ash and plantation-growth eucalypt species. PhD Thesis, School of Chemical & Biomolecular Engineering, The University of Sydney. House, S. 2006. Species Profile for Dunn’s White Gum. Queensland Government. Huth, J. 2006. Species Profile for Dunn’s White Gum. Queensland Government.  Innes, T. C. 2003. Australian Hardwood Drying Best Practice. Manual parte 1 & 2.  Kerruish, C. M. 1991. The Young Eucalypt Report. Lawson, S. 2006. Species Profile for Dunn’s White Gum. Queensland Government. Lee, D. 2006. Species Profile for Dunn’s White Gum. Queensland Government.  McGavin, R. 2003. Australian Hardwood Drying Best Practice. Manual parte 1 & 2.  Nolan, G. 2003. Australian Hardwood Drying Best Practice. Manual parte 1 & 2.  Rawlins, W. H. M. 1991. The Young Eucalypt Report.  Redman, A. 2003. Australian Hardwood Drying Best Practice. Manual parte 1 & 2.  Rob, C. Sydney blue gum for sawlogs in the 450-650mm rainfall zone.  Siau, J. F. 1984. Transport Processes in Wood.  Skaar, C. 1988. Wood Water Relations.  Compression Kiln R&D  Duncan Fairclough 120  Appendix See attached CD containing individual batch sheets and pictures for all kiln charges.  Compression Kiln R&D  Duncan Fairclough 121  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52966.1-0103120/manifest

Comment

Related Items