Printer Friendly





Electromagnetic radiation, such as radio-frequency heating has been shown to rapidly heat wood. Previous research at The University of British Columbia focused on the drying of sawnwood using a radio-frequency/vacuum (RF/V) dryer with flat parallel plates. The study described in this paper examined the potential of RF/V drying to rapidiy dry roundwood. The investigation was carried out in a laboratory RF/V dryer at a stabilized frequency of 13.65 MHz. During drying, the change in the temperature with time, within each roundwood section, was continuously monitored. Examination of all the experimental runs confirmed that the RF/V unit with a flat electrode design, could dry 165- to 240-mm diameter and 2.0- to 2.4-m-long pole sections, from their initial sapwood moisture content (MC) of over 80 percent, to a final MC of less than 25 percent. This was accomplished with limited loss in quality, in less than 16 hours. Furthermore, RF/V drying produced poles with a uniform final MC.

Currently, a significant number of utility poles are still commonly air-seasoned in North America prior to preservative treatment (7). However, air-seasoning is time consuming and requires pole producers to retain a large inventory, leading to high overhead costs. In addition, when large numbers of poles are needed quickly, such as during the replacement of poles damaged during the ice storm in eastern Canada and the United States in January 1998, air-drying is too slow. A further disadvantage for thick sapwood species is the difficulty in preventing loss in quality due to colonization by decay fungi. This can result in premature pole failure in service (6). With the increased production of poles treated with chromated copper arsenate, kiln-drying of utility poles prior to treatment is becoming more common.

The drying of timber using dielectric heating at radio frequencies (RFs) has been successfully demonstrated, to be independent of wood thickness. The results also showed that it minimized checking and produced a uniform final moisture content (MC) in the timber (1,3-5).

This study was designed to determine whether radio-frequency/vacuum (RF/ V) heating could efficienfly dry roundwood. The hypothesis was tested using a laboratory RF/V kiln, which has flat horizontal parallel electrodes and is located at the Wood Drying Laboratory at The University of British Columbia (UBC). During the heating and drying process, the temperatures at different depths from the wood surface of each pole section were recorded and efficiency of the process was evaluated from the final MC profiles and the total drying times.


Western red cedar (Thuja plicata), Douglas-fir (Pseudotsuga menziesii), and red pine (Pinus resinosa) roundwood sections, recovered from freshly processed poles, were individually dried in a series of experimental runs. Most of the green pole sections were approximately 165 to 240mm in diameter and 2 to 2.4 m in length. The red cedar and Douglas-fir pole sections had an average sapwood depth of 10 to 25 mm and 33 to 47 mm, respectively, and were sourced in British Columbia. The red pine pole sections contained 45 to 80 mm sapwood, and were sourced in Ontario. All of the experimental pole sections were shipped directly to the Wood Drying Laboratory at UBC. The size of the roundwood sections was dictated by the limitations of the RF heating unit. Since the RF/V drying takes place mostly from the ends of the timber, end-coating was not applied so that the benefits of the process would be realized. Previous experiments have shown that doubling the length of the timber from 1 to 2 m caused a 5 percent increase in the drying time [8]. However, it is anticipated that some increase in the drying time would be required when comparing the drying of 2.4-m-long pole sections to that of full-length poles.

There were 32 experimental runs carried out with the 3 wood species in a 0.25-[m.sup.3] capacity laboratory RF/V dryer. The detailed description of the RF/V dryer can be found in a previous publication [4]. The drying experiments were done using an RF oscillator that generated the electromagnetic field. The oscillator operated at a stabilized frequency of 13.56 MHz, and a maximum output of 10 kW at a maximum electrode voltage of 5 kV.

The temperatures inside the pole sections and the drying chamber were monitored by fiber optic temperature sensors. These sensors were placed close to the midpoint of each pole section and separated by about 5 cm from each other (Fig. 1). A plastic cap, in which a small hole had been drilled to allow insertion of each temperature probe, was used to fasten it inside the wood. Two probes were used in runs 8 to 10. They were attached through 10-mm-diameter holes, located 30 mm away from the surface (outer shell), and at the center of the pole cross section (core). The drying experiments 1, 2, and 27 to 32 used three probes. They were also placed in holes 10 mm in diameter with one in the outer shell, one at the midpoint of radius of the cross section, and the last at the center of the cross section. In order to monitor the temperature of the environment inside the chamber during drying, 4 temperature probes were used in the remaining 14 runs. In this case, one of the probes was left outside of the pole section, while the other three probes were placed as just described.

Each experimental run was conducted using one roundwood section, and all of the drying experiments started from ambient temperature (approximately 20[degrees]C). In order to increase the rate of initial heating and reduce the total drying time, the kiln was evacuated when all of the temperature sensors inside the pole exceeded 60[degrees], or when one of the sensors exceeded 95[degrees]C. The drying was then carried out at a pressure range of 2.66 to 3.60 kPa. A computer connected to a data-acquisition system recorded the temperatures every 30 minutes during the first 12 runs, and every 4 minutes for the remaining 25 runs.

Immediately at the end of each drying run, the pole section was removed from the kiln and the temperature sensors were carefully disconnected. Samples were recovered for determination of heartwood and sapwood MC. They were measured on randomly selected cores removed from each section using the ovendry method.


The results of the temperature variation with time were grouped according to the section's initial MC. Since the purpose of this project was to examine the possibility of fast and uniform drying of poles using the RE/V process, analysis of the relationship between the energy consumption, the electrode voltage, and their impact on the drying speed was not considered. During this discussion, representative plots of temperature changes with time for the series of drying runs with high initial MC will be presented.


Of the 32 drying runs, the first 2 series involved 14 experiments using pole sections with a sapwood MC ranging from 30 to 86 percent. Most red cedar sections had a sapwood MC between 30 and 40 percent, and only 1 or 2 hours were required to reduce the sapwood MC to less than 25 percent, which would allow the poles to be preservative treated. When the MC was between 50 and almost 90 percent, 6 to 9 hours were required to reduce the sapwood MC to about 10 to 24 percent.

Eighteen experiments were conducted using Douglas-fir and red pine pole sections, in which the initial sapwood MC was over 100 percent. The initial heartwood MC for Douglas-fir was close to the fiber saturation point (30%). The drying time ranged from 10 to 16 hours. Figure 2 shows the temperature changes with time in runs 15 and 32, which are representative for drying of Douglas-fir and red pine, respectively. In run 15, it can be seen that the ambient temperature in the kiln remained relatively constant at approximately 25[degrees]C. The temperature in the core of the section before introducing the vacuum was the lowest, at around 60[degrees]C. It is also interesting to note that before introducing the vacuum, there was a steep temperature rise, especially in the probe at the outer shell where the MC is at its highest level. After the vacuum was applied, the temperature inside the pole section decreased rapidly from 1200 to about 60[degrees]C in the outer shell, presumably due to evaporative cooling from th e surface of the section. Much smaller reductions (from 80[degrees] to around 65[degrees]C) were noted at the midpoint of the radius. At the same time, there was no significant decrease in the temperature of the inner wood. Following the initial response to the rapid removal of moisture from the kiln by the vacuum, the temperature at the three locations rapidly stabilized, and then remained relatively constant until the end of the drying.

The red pine sections contained over 80 percent of sapwood. The inner initial sapwood MC was higher than that of the outer shell. This would result from the initial drying of the freshly cut pole material. It is clearly shown in the temperature graph of run 32, that the rise of the temperature at the center was more rapid than that of the outer shell (Fig. 2). The core temperature was reduced rapidly by almost 35[degrees]C when the kiln was evacuated, while the temperatures in the midpoint of the radius and the outer shell were reduced 20[degrees] and 10[degrees]C respectively. Within about 30 minutes of introducing the vacuum, the temperature in the three locations inside the pole increased to their maximum, and then gradually decreased due to the loss of moisture from the wood.

After 10 to 16 hours of drying, all pole sections had a final MC less than the 25 percent target suggested for poles prior to preservative treatment. Almost one-third of the pole sections had an average final MC of less than 18 percent. The average difference in final sapwood and heartwood MC of the Douglas-fir sections was calculated at approximately 5.8 percent. The difference in final inner sapwood and outer sapwood MC of red pine was about 10.8 percent. This relatively large MC difference following drying was due to the much higher initial MC in the inner sapwood of red pine, which is different from the red cedar and Douglas-fir pole sections. All of the Douglas-fir specimens had a heartwood final MC below or equal to that of the sapwood. On the other hand, all of the red pine pole sections had final inner sapwood MCs higher than that of the outer sapwood. Even so, the inner sapwood MCs were about 25 to 29 percent.

Figure 3 shows the differences of initial and final MC between heartwood and sapwood and between inner sapwood and outer sapwood. The excellent potential of RF/V drying to produce roundwood with a uniform MC in both heartwood and sapwood (or in the case of red pine between the inner and outer sapwood) is demonstrated by these results. The initial variation is relatively high, with some of the sapwood values approaching 140 percent. After drying, the differences are all less than 10 percent (except for red pine at 10.8%). This uniformly low MC is clearly helpful in reducing deep check formation during drying, providing a high quality utility pole in terms of enhanced resistance to in-service decay.


Due to equipment limitations, the energy consumption during RF/V drying could not be monitored. Based on previous research of RF/V drying of thick timbers conducted using the same kiln, the small variation in the power could be neglected, if the control "paddle" was kept at the same level throughout the whole initial heating process. The initial heating rate at the vicinity of each probe was calculated as the temperature change with time, from the beginning of the heating to the first temperature maximum. Using this basis, and neglecting the small variations due to wood density, the initial heating rate at the pole section surface was plotted against the initial MC of the outer sapwood (Fig. 4). It is clear from the plot that the higher the initial MC in the outermost (sapwood) zone, the greater the initial heating rate. This is consistent with a proposal made in a previous publication [2], which stated that wood with a high MC will heat up faster than dry wood, at a fixed frequency. A plot of the loss of moi sture in the sapwood, determined for all drying runs, shows that there is a linear relationship between the drying time and MC decrease (Fig. 5) over the range of MCs evaluated.

The heartwood initial MCs in the red cedar and Douglas-fir were much less variable. Indeed, they could be considered as a group with a range of less than 10 percent. Comparing the two, the more dense Douglas-fir had the lower heating rate. The red pine with an intermediate density had an intermediate heating range, although the inner sapwood MC was much higher than the red cedar and Douglas-fir heartwood. The linear regression of the initial heating rate versus the wood density for the heartwood and inner sapwood, the zone further from the surface, gave an [r.sup.2] of 0.999 (Fig. 6) suggesting that the two are closely correlated.


It may be concluded from this study that a laboratory RF/V dryer with a flat electrode design could effectively dry western red cedar, Douglas-fir, and red pine pole sections. It took less than 16 hours at internal temperatures over 65[degrees]C to dry 165- to 240-mm-diameter and 2.0- to 2.4-m-long pole sections, from their initial sapwood MC of over 80 percent to a final MC of less than 25 percent. The heartwood intial MCs were lower, but they too could be reduced within the same time frame to less than 25 percent MC. This drying rate compares favorably to the 1 year required by the current air-drying strategy or the several days required by conventional kiln-drying methods. The dried pole sections were uniform in final MC. The loss in moisture from the wood surface was linear with drying time. However, further from the surface, the initial heating rate was strongly correlated with wood density.

The authors are, respectively, Former Graduate Student and Professors, Dept. of Wood Sci., Forest Sciences Centre, 4041 - 2424 Main Mall, The Univ. of British Columbia, Vancouver, BC, Canada, V6T 1Z4. The authors are pleased to acknowledge the financial and material support provided by NSERC, Manitoba Hydro, Timber Specialties Ltd., and Stella Jones. This paper was received for publication in June 2000. Reprint No. 9139.

(*.) Forest Products Society Member.

(c)Forest Products Society 2001.

Forest Prod. J. 51(7/8):56-60.


1. Avramidis, S. and J. Dubois. 1992. The study of dielectric properties of spruce, hemlock, w.r.cedar and Douglas-fir at varying MC, temperature, grain orientation and radio frequency. Rept. No. 93 (SA-3), Sci. Council of British Columbia, Vancouver, BC, Canada. 50 pp.

2. _____ and F. Liu. 1994. Drying characteristics of thick lumber in a laboratory radio-frequency/vacuum dryer. Drying Technology 12(8):1963-1981.

3. _____ and R.L. Zwick. 1992. Exploratory radio-frequency/vacuum drying of three B.C. coastal softwoods. Forest Prod. J. 48(7/8): 17-24.

4. _____, F. Liu, and BJ. Neilson. 1994. Radio-frequency/vacuum drying of softwoods: Drying of thick western redcedar with constant electrode voltage. Forest Prod. J. 44(1):41-47.

5. _____, R.L. Zwick, and BJ. Neilson. 1996. Commercial-scale RF/V drying of softwood lumber. Part 1. Basic kiln design consideration. Forest Prod. J. 46(5):44-51.

6. Highley, T.L., C.A. Clausen, S.C. Croan, F. Green Ill, B.L. Illman, and J.A. Michales. 1994. Borate diffusion from fused borate rods in Douglas-fir transmission poles. Res. Pap. No. FPL-RP-529. USDA Forest Serv., Forest Prod. Lab., Madison, WI. 20pp.

7. Newbill, M.A. and JJ. Morrell. 1991. Effect of elevated temperatures on survival of basidiomycetes that colonize untreated Douglas-fir poles. Forest Prod. J. 41(6): 3 1-33.

8. Zhang, L., S. Avramidis, and S.G. Hatzikiriakos. 1997. Moisture flow characteristics during radio frequency vacuum drying of thick lumber. Wood Sci. and Technology 31:265-277.

[Graph omitted]

[Graph omitted]
COPYRIGHT 2001 Forest Products Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Forest Products Journal
Article Type:Statistical Data Included
Geographic Code:1USA
Date:Jul 1, 2001

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |