The use of fluidized beds of hot inert solids for drying wood is a relatively new concept. Recent investigations on fluidized bed drying of thin veneer (1,2) have shown that more rapid drying can be achieved by this method than by conventional means. In the present work, blocks of Western Hemlock wood, 2 in. x 4 in. x 1 ft. containing 70% to 100% moisture (dry-basis) were dried in a fluidized bed of -20 +30 mesh sand at four levels of bed temperature (175, 190, 204, and 217°F) and three air velocities. The drying time required to reach 15% moisture content (M.C.) was 14-15 hrs. for lumber dried at 204°F as against two or more days generally taken in Kiln drying. The quality of the wood dried at bed temperatures of 204°F and below was not adversely affected. Bed temperature had a strong inverse effect on drying time, while the fluidizing air flow rate had little effect. The diffusion equation was employed to describe the movement of moisture during the falling-rate period of drying and the heat conduction equation to describe the unsteady-state movement of heat inside the drying block of wood. Mathematically, drying was treated both as a one and a two-dimensional problem. The resulting equations were solved on a digital computer to predict the average moisture content and the average temperature of the drying block of wood, each as a function of time. The distribution of moisture content within the drying block was also computed. The calculated results showed a good agreement with experimental data. The economics of fluidized bed drying were estimated and compared with the cost of Kiln drying. The results showed that the capital cost of the fluidized bed system is considerably lower while the operating cost is similar to that for kiln drying. / Applied Science, Faculty of / Chemical and Biological Engineering, Department of / Graduate

The purpose of this study was to evaluate the effectiveness of steam conditioning immediately after drying to reduce internal checking resulting from high-temperature drying of Pacific Coast hemlock lumber. Three different levels of conditioning time, 2, 4 and 6 hours, were carried out on 2 inches wide by 4 inches thick by 3 feet long (51 mm by 102 mm by 0.91 m) , and on 4 inches wide by 4 inches thick by 3 feet long (105 mm by 105 mm by 0.91 m), specimens. For comparison purposes, controls of both sizes of specimens were also high-temperature dried without conditioning. Analysis of the results indicated that internal checking was not significantly reduced by steam conditioning and was more likely to develop afterwards during storage at room temperature, and that total degrade observed in the "4x4" specimens was more excessive than that in the "2x4" ones. The defective "4x4" specimens were found over-dried (below the targeted 12% moisture content) with high core-shell moisture content differences. More internal checking was found when the specimens' final moisture content ranged from 7 to 8%. / Forestry, Faculty of / Graduate

Graduation date: 1966

When oven drying Douglas fir to determine moisture content, volatile organics evolving from the wood cause error in the moisture measurement. An apparatus was constructed to measure the carbon contained in the evolved organics and to measure the chemical energy content of the organics which evolved at temperatures up to 103°C. Data were collected and used to estimate errors in moisture content measurement. The data were also used to provide insight into differences that exist in fuelwood and how these differences might affect wood heater test results. For the wood examined, the lower heating value of the evolved organics averaged 283 MJ/kmol carbon. This value indicates that the volatile material had a high oxygen content. Using the average heating value, the mass fraction of oxygen contained in the evolved organics was estimated to be near 0.47. Using the estimated oxygen mass fraction combined with the carbon loss data, errors in moisture content measurement resulting from volatile loss were estimated to range from 0.1% to 2.0% of measured values. The larger of these errors would cause less than a 0.1% error (percent of measured value) in the calculated efficiency of a typical woodstove. The amount of highly volatile organics was found to vary greatly between different pieces of Douglas fir. For the wood tested, carbon loss ranged from 0.02% to 0.35% of the carbon initially contained in the wood. These differences in fuelwood volatility may affect wood heater performance and possibly cause some of the scatter observed in wood heater test data. Conventional proximate analyses were performed on two wood samples, one exhibiting high volatility at 103°C, the other exhibiting low volatility at 103°C. These proximate analyses did not show significant differences between the two samples. Bomb calorimetry was used to measure the heating value of the Douglas fir samples both before and after oven drying. The bomb calorimetry, however, proved to not have the sensitivity to accurately measure the small differences in heating value caused by volatile emissions during oven drying. / Master of Science

The S-D-R (Saw-Dry-Rip) process is a proven method of reducing warp in manufacturing hardwood studs. It has been assumed that its success is partially due to the stress relaxation caused by high-temperature drying. Implementation of this S-D-R process in the Appalachian region has been limited as few mills in this area have this drying capability. The purpose of this study was to investigate the effectiveness of high-temperature and steaming treatments before solar and air drying S-D-R flitches on reducing warp and to determine the cost of application of successful methods. An initial study was conducted to establish the duration of treatment required to relieve growth stresses in short S-D-R flitches. Steaming for 15 hours resulted in a substantial decrease of average stress. Steaming times of 15 and 20 hours and high-temperature durations of 7 and 15 hours were chosen for treating green S-D-R flitches. S-D-R flitches were divided among the four treatments and a control group. One third of the flitches from each group were subsequently dried in a solar kiln; the rest were air dried. All flitches were then ripped into 2 x 4's and bow, crook, and twist was measured for each stud. A comparison by the percentage of studs rejected from No. 2 dimension because of warp showed no significant difference between any treatments (including the control) or between drying methods. Comparisons based on actual amount of warp found small differences between drying methods and between treatments, depending on the type of warp considered. / Master of Science

The Bolivian forest products industry has experienced substantial growth during the last ten years. Exports of value-added products have largely replaced logs and green lumber, and raw materials for wood products have shifted to lesser used species. Important investment has taken place in lumber drying capacity, which unfortunately was not always accompanied by sound drying practices. Several non-governmental organizations, with U.S. financial aid, are currently supporting the industry with technical assistance. This project assists these efforts by assembling much needed information regarding lumber drying, and providing tools for performance measurement of drying practices. A survey was conducted among Bolivian companies to determine lumber drying capacity, technology and practices. Results showed a total drying capacity of 6,104,250 board feet in 167 kilns. Technology and practices used are highly variable. Thirty six percent of kilns are home-made, and 59% are European commercial brands. Upon completion of the survey, a set of analytical tools was developed and tested in six Bolivian firms. These tools were designed to systematically evaluate lumber drying operations and formulate actions for improvement. Equilibrium moisture content (EMC) during storage, manufacturing and shipping was monitored in plants located in three Bolivian cities and inside containerized shipments of wood products. Findings showed differences between EMC and lumber moisture content from -1% to 7%. Differences between EMC inside dry-lumber storage and processing facilities varied between 0% and 3.6% and were greatly influenced by facility configuration. Climate during shipment of wood products largely depended on packaging materials and methods, which attenuate sharp changes in ambient conditions. Monthly values for outdoor EMC for the main cities of Bolivia were calculated based on historic weather data and are reported. / Master of Science

A study is made of the simultaneous heat and mass transfer processes which occur within a finite wood cylinder of circular cross-section when it is convectively dried. Governing transport equations are developed allowing for the functional dependence of both thermal and moisture diffusivities on moisture content and temperature. Derivative type boundary conditions are included in the analysis. The equations are formulated using a cylindrical coordinate system because it is well suited to modelling wood's anisotropy. In the case considered, the axial coordinate direction is aligned with the wood-grain. Due to the coupling and non-linearity present in the transport and boundary equations, an implicit finite-difference solution scheme is formulated. The three-time-level scheme uses an equation splitting technique to simplify its solution on the computer. A mathematical wood-model, as available in the literature, is refined and used to determine moisture and thermal diffusivities, and mass transfer boundary conditions. Results from this wood-model apply to softwoods below the fiber-saturation moisture content. Combined diffusivities of liquid and vapour are calculated for the radial and axial directions and results for the radial direction are compared to those found from diffusion experiments. Desorptional isotherms are used in the mass transfer boundary condition equations to relate surface humidity and moisture content. The wood-model is used in the mass transfer equation to determine wood drying behaviour under isothermal conditions. Initial moisture contents are uniform and equal to the fiber-saturation value. Solutions are presented to give local and average moisture content as a function of time as well as moisture content profiles. The effect of drying temperature was investigated for one case of wood density and shrinkage. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

Experimental drying studies were made on two types of solar lumber kilns, one an external collector type and the other a semi-greenhouse type. Two charges of green sugar maple lumber (5/4 inches  thick) were tested in an external collector solar kiln at the U.S. Forest Products Laboratory, Madison (43°5'N, 89°23'W), Wisconsin, during the summer of 1982. In the first run detailed drying data were obtained and the energy balance was calculated for each day during the entire drying period. Based on these results, the following empirical model for the overall efficiency of the kiln was obtained: EFF = - .0413 + .0102^*(IMC) - .0000562^*(IMC)² where, EFF = overall efficiency of the kiln IMC = average initial moisture content of the lumber in percent Using  this model, the average daily moisture content loss in percent (MCL) can be calculated as follows: MCL = (100*EFF*SI*ACV)/[R*(62.4*V*SG)*{0.53*(212-Ti)+972}] where, EFF = the value obtained from the first equation ACV = area of the collector in ft² SI = average daily solar insolation in Btu/ft² R = ratio of total solar energy incident on the collector cover to total energy available to the system V = green volume of lumber in ft³ SG = green specific gravity of lumber Ti = average initial temperature inside the kiln in °F A comparison of the actual drying time observed in the second run showed good agreement with the predicted drying time obtained from the above equations. A single charge of 9/8 inches green yellow poplar was dried in a semi-greenhouse kiln at Virginia Polytechnic Institute and State University, Blacksburg (35°09'N, 81°30'W), Virginia, during the fall of 1982. Following the same procedure as for the external collector kiln, an empirical model for the overall efficiency (EFF) of the kiln was obtained as a function of initial moisture content (IMC), EFF = - .0767 + .00988*IMC / M.S.

The purpose of this research was to measure and simulate the major perpendicular to gain strain components and associated stresses which develop as the result of mass transfer and restraint of shrinkage in red oak. Particular emphasis was placed on the rheological or creep components of strain. Mass transfer was measured during the first four days of drying under increasingly severe conditions. The resulting moisture gradient profiles were parabolic in shape under nearly all drying conditions. The pattern developed quickly and was modeled with reasonable accuracy using Fick's second law. Three major strain components occur in drying. These are termed elastic strain, visco-elastic creep and "set" or mechano-sorptive creep. The magnitude and variation of each of these components was measured during the first four days of drying under increasingly severe conditions. Using a slicing technique to cut very thin wafers of wood parallel to the surface, the elastic strain was shown to be quite small. The experiments led to the conclusion that the maximum stress develops within a few cell thicknesses of the surface. Much of the experimental work centered on the t of the rheological or creep components of strain, The magnitude of the visco-elastic creep was found to be about the same order of magnitude as the elastic strain and was clearly a function of the applied stress. The major strain component during drying was shown to be mechano-sorptive creep. This type of creep occurs in a number of polymeric materials undergoing diluent movement or temperature change. The maximum magnitude of the mechano-sorptive component was about 30 times larger than the maximum elastic strain. Mechno-sorptive creep was shown to be directly related to moisture loss and, to a lesser extent, a function of applied load. Using relationships derived from the experimental data, a computerized simulation was developed to predict the development of stress and the probability of checking early in drying. The simulation makes use of data on elastic strain, mechano-sorptive creep and the elastic modulus. The concept worked well in this study, but its application is limited by the lack of strain data for the surface layers representing the outer few cell thicknesses where stress is greatest early in drying. As an adjunct, acoustic response tests were performed on green samples stressed perpendicular to grain in flexure under third point loading conditions. The evidence indicates that the onset of surface failure can be predicted prior to the appearance of macroscopic checking. The acoustic response pattern is typical of that which occurs in brittle glassy polymers such as polystyrene. / Ph. D.

Although High temperature drying (HTD) of southern yellow pine (SYP) is extensively used, quality problems still exist. Approximately 25 percent of all green lumber (2 x 4, 2 x 6 and 2 x 8's) are degraded during HTD. Warp degrade, specifically "crook", is the main cause for the loss of quality of SYP dried at temperatures above the boiling point of water. One of the major causes of warp degrade is excessive shrinkage that results from overdrying. In industrial practices, average final moisture contents for HTD of SYP lumber are usually well below the 15% MC maximum permitted by the grading rules. Overdrying the lumber causes not only lumber quality problems, but also constitutes a waste of energy. In order to improve the overall results of HTD of SYP, it is necessary to perform a better control of the process, so unnecessary extended drying can be prevented. In this study, a mathematical model based on the temperature drop across the load (TDAL) concept was proposed. The model relates TDAL to drying rates and according to the results, it appears to be an important alternative for controlling purposes during HTD of SYP. / Ph. D.