As of 2010, almost 7.6 million hectares of land was under rubber (Hevea
brasiliensis) cultivation in the South East Asian region and it is envisaged
that the cultivation acreage will increase substantially in years to come as
demand for latex or natural rubber continue to increase in worldwide (Ratnasingam
and Scholz, 2009). Although cultivated primarily for latex or natural rubber
production, the rubber trees also produce a large amount of woody biomass, of
up to 180 m3 per hectare upon harvesting for replanting activities
(Hong and Sim, 1994). However, only about 50 m3
from this amount of the woody biomass is converted into sawn lumber, commercially
known as rubberwood (Hevea brasiliensis). Despite its inherent harvesting
and production shortcomings, rubberwood has emerged as the most important wood
raw material for the wood products manufacturing sector. It is used extensively
for particleboard, medium density fibreboard, parquet flooring, joinery and
furniture productions throughout the South East Asian region, particularly in
Malaysia, Thailand and Indonesia (Ratnasingam and Scholz,
2009). The fact that the material is derived from man-made forest plantations
makes rubberwood an environment-friendly material, which further boosts its
status as a premier wood raw material. Despite, its success in the wood raw
materials market, rubberwood is plagued with low processing yield, both in the
sawmilling and other machining operations due to the presence of tension wood
and other processing defects (Ratnasingam and Scholz, 2007,
2009; Ratnasingam and Ma, 2010).
Tension wood is the abnormal wood found commonly in leaning hardwood trees and
its branches (Archer, 1986; Lim and
Nadzri, 1995) and its presence generally impairs the overall wood yield
and quality (Wardrop and Dadswell, 1955; Ratnasingam
et al., 2010). Although the formation and properties of tension wood
has been well documented (Lim and Nadzri, 1995; Plomion
et al., 2001), information on minimizing its effect on wood yield
and quality is limited (Ratnasingam and Scholz, 2009).
The use of mobile sawmills in rubberwood processing is predominant throughout
South East Asia, as it is suitable for on-site conversion of rubber trees in
plantations that are dispersed and remotely located. Consequently, the resultant
processing yield is low as the technology is a better choice for re-sawing operations,
rather than primary sawing. Inevitably, wood yield and quality is impaired to
an extent that affects the competitiveness of the material as a primary wood
resource for the value-added wood products industry (Ratnasingam
and Scholz, 2007, 2009).
One possible technique available to improve wood yield and quality is the Saw-Dry-Rip
(SDR) process which has been used for the sawing of hardwoods in other parts
of the world (Maeglin and Boone, 1983). The SDR process
is a simple process based on well-known manufacturing techniques combined, in
a different order, to account for wood structure and stresses. In this process,
the logs are first live sawn into flitches and subsequently dried, before being
ripped to dimension. The difference in the processing order is that, the conventional
sawing cuts the lumber to dimension, with an allowance for shrinkage and planning,
before drying. Research has shown that the SDR process show significant improvements
in yield when sawing hardwoods (Maeglin and Boone, 1983).
Generally, the green logs are live sawn into thick flitches in the SDR process
which allows the inherent stresses to be balanced and restrained. Additionally,
the drying stresses induced during the drying process offset, to some degree,
the growth stresses in the flitches. After drying, the flitches are straight-line
ripped to desired dimensions which results in higher sawing yield. In essence,
the SDR process may be useful for converting rubberwood, which has a significant
proportion of tension wood and growth stress, which reduces its yield during
sawing and drying (Ratnasingam et al., 2010).
Therefore, a study was undertaken to compare the conventional and SDR sawing processes for rubberwood in terms of product yield and quality. Further, the study also evaluated the economics of the two sawing processes, to provide the necessary industrial guidelines for improvements in the sawing of rubberwood.
MATERIALS AND METHODS
As most of the rubber plantations were located in the southern states of Peninsular
Malaysia, the study conducted in three parts, was carried out in the state of
Johor between the months of January to October of 2010.
Part I evaluated the processing yield and quality during the sawing of rubberwood
in one of the largest sawmills, with a sawing capacity of 50 m3 of
saw logs per day. The saw logs, aged more than 25 years, were in the 28-32 cm
diameter class, with an average length of 1.9 m. The sawing process was carried
out using mobile sawmills, with vertical saw blades of 100 mm in width to reflect
current industrial practices. The saw blade characteristics used in this study,
are as shown in Table 1. Sawn rubberwood of 25, 38 and 50
mm in thickness and 50 and 75 mm in width were produced to reflect the most
common cutting sizes within the industry (Ratnasingam and
|| Saw blade characteristics
|| Kiln drying schedule for rubberwood
|Values in parentheses reflected the dry-bulb and wet-bulb
temperatures used in high temperature drying
The study was conducted over a period of a week, which ensured sufficient sawn
material to fill-up the kiln-drying chamber on-site.
The sawn materials were then dried in two steam-heated kilns, using two different drying schedules (Table 2). The air speed in the drying chamber was maintained at 3.0 m sec-1 while a restraining pressure of 100 kg m-2 was applied on each bundle as the weight down to minimize warping during both drying schedule. The equal amounts of the different sawn materials were subjected to the two different drying regimes. Kiln samples were used to determine the moisture content of the materials. An equalizing treatment was applied to ensure uniformity in the final moisture content of 10±2%, before discharge.
Part II of the study focused on the use of the SDR process in the conversion of rubber logs. The sawing was carried out to produce flitches of 55 mm in thickness. The sawn materials were then dried in two steam-heated kilns, using the two different drying schedules (Table 2). The drying charge were of similar quantity and quality and at the end of the drying cycle, an equalizing treatment was applied to ensure uniformity in the final moisture content of 10±2%, before discharge. The dried sawn materials were sent to the sawmill for straight-line ripping to the final desired dimensions.
The processing yield measurements, under the two sawing processes were compiled
and evaluated on a volumetric basis. In terms of quality, the degree of warp
(specifically crook, bow and twist) was measured by placing the piece on a flat
surface to determine the maximum deviation from the straight line for each type
||Comparison of yield and quality of the two different sawing
|Values in parentheses indicate Standard deviation
This deviation was measured using a calibrated depth gauge recording to the
nearest 1 mm. The maximum acceptable warp limits for the material was set as
per recommended in the Grading guideline for rubberwood issued by the Malaysian
Timber Industry Board (MTIB). The results obtained were statistically analysed
using the Analysis of Variance (ANOVA) technique, which ascertained the statistical
differences due to the different experimental parameters. The degree of confidence
of p = 0.05 reflected the reliability of the results obtained, in comparison
to other studies.
Part III of the study involved a survey of current industrial yield at 125 rubberwood saw mills, located mostly in the states of Johor, Melaka and Negeri Sembilan in Peninsular Malaysia. The selected mills were based on the listing provided by the Malaysian Wood Industries Association (MWIA).
RESULTS AND DISCUSSION
Part I: Comparative sawing yield: The sawing yield from the two different processes was derived by calculating the nominal volumes of sawn timber and multiplying it by the number of sawn timber produced. Percent yield is the total volume recovered as a percentage of the total log scale for each sawing process. Although, differences in log diameters were unavoidable, all efforts were taken to ensure equal distribution between the two sawing processes.
The percent volume recovery for the two sawing methods showed significant difference
(p = 0.05), since the SDR process resulted in higher nominal yield as it allowed
greater versatility during the ripping operation for different dimensions. A
similar finding was also reported by Layton et al.
(1986) who found that the SDR process allows greater recovery due to the
possibility of a better cutting plan for increased recovery and the avoidance
of shrinkage allowances provided in the conventional sawing process (Table
Based on the rejects for warp only, the SDR process had a significantly lower
percentage of rejects (p = 0.05) compared to the conventionally sawn material
(Table 3). The study showed that 81% of the SDR produced material
was a grade material (no defective pieces beyond the acceptable limit). While
only 9% were B grade (less than 15 defective pieces per 100 pieces) and the
balance were grade C material (more than 15 but less than 30 defective pieces
per 100 pieces) (Table 3). Hence, it is apparent that the
SDR process outperforms the conventional sawing process, in terms of both volume
recovery and product quality. This is particularly true when the sawn materials
were subjected to high temperature drying regimes. Studies by Mårtensson
and Svensson (1997) and Lazarescu and Avramidis (2008)
have shown that drying wood at high temperatures tends to counter the inherent
growth stresses in the wood. On the other hand, the study by Sik
et al. (2009) and Ratnasingam and Scholz (2009)
have shown that drying rubberwood at high temperatures will counter balance
the inherent growth stresses in the wood but also minimize the imbalance due
to the tension wood by plasticising the lignin in wood. In essence, it is apparent
that high temperature drying of rubberwood is advantageous to increase both
wood yield and quality.
Part II: Treatment effects on warp development: The effects of the SDR
process on the development of warp and other degrades were also studied. The
SDR process has been reported to have a strong influence on warp (Layton
et al., 1986) but the interaction of the various factors on the type
and degree of warp in sawn Rubberwood remain inconclusive. Table
4 shows the comparison of the effects of sawing and drying treatments on
the type of warp in sawn material from this study. As shown, the effectiveness
of the SDR process in controlling crook was substantial (p = 0.05), with more
than 50% reduction in crook, both using the conventional and high-temperature
drying techniques. Similarly, the defect, twist was reduced by more than 50%
(p = 0.05) through the SDR process, under both the conventional and high-temperature
drying techniques. In terms of the defect bow, almost 45% of the conventionally
sawn material exceeded the allowable bow limits, while only 8% of the SDR sawn
material manifested the defect. Previous reports by Sik
et al. (2009) and Ratnasingam and Scholz (2009)
have shown that by turning around the logs during the sawing process as practiced
in the SDR process, significantly reduced the defects, twist and crook, in the
sawn wood material.
|| Effects of treatments on drying defects
|Statistical significance was tested at p = 0.05
|| Comparative cost of the two sawing processes
|Based on the average log cost of US 25 m-3, labour
cost in sawn timber production of US 35 m-3, saw-doctoring cost
of US 10 m-3, energy cost of US 15 m-3, drying cost
of US 48 m-3 and high temperature drying cost of US 115 m-3
Turning around the logs minimized the effect of grain deviation in the wood
due to the radial-tangential surfaces counterbalance (Huber
et al., 1984; Simpson, 1991). Further, the
high temperature drying also reduces the amount of warp in the sawn material,
as high temperature drying under the weight-strain plasticizes the sawn material
to reduce warp significantly (Diawanich et al., 2010).
Therefore, the use of the SDR process significantly reduces warp which remains
the most common defect in sawn rubberwood, as reported by Ratnasingam
et al. (2010).
Part III: Cost implications: Table 5 shows the economic
implications of the two different sawing processes in rubberwood conversion.
It is apparent that, although the SDR process results in a lower throughput
level compared to the conventional sawing process at the saw mill, it leads
to a significantly higher quality sawn rubberwood (p = 0.05) which fetches a
premium price in the market place. Further, the application of the high temperature
drying process, not only reduces the drying time but also lead to a better quality
dried stock, with lesser defects. The study also proves that high temperature
drying of rubberwood is both practical and economically viable and is parallel
with the earlier report by Sik et al. (2009).
Nevertheless, the maximum temperature used in this study proves to be optimum,
as the unappealing darkened colour of the wood surface is minimized, although
the effect of higher temperatures on the colour of the wood surface is worthy
of further studies. The colour of the rubberwood surface is a primary criterion
in using the materials for the manufacture of value-added wood products (Ratnasingam
and Scholz, 2009).
Industrial implications: A survey of 125 rubberwood saw mills in Peninsular Malaysia revealed that the average green yield was 27%, with a further reduction in yield by almost 7% due to drying losses. Table 5 shows the comparison in process economics between the two sawing techniques used in this study. It is obvious that although the SDR process requires a much larger capital outlay, its lower operational cost and higher economic returns should prompt more rubberwood saw mill operators to seriously consider the SDR process as the conversion technique of choice for rubberwood. This is particularly true against the background of increasing production cost and the demand for better use of the raw materials available. In this context, it must be emphasized that the SDR process should be the conversion process of choice, especially when dealing with logs of more than 28 cm in diamet.
The results of this study shows that the Saw Dry and Rip (SDR) process is a more viable option compared to the conventional sawing process for rubberwood. The higher volume recovery as well as the significant reduction in the defect, warp, implies that the SDR process provides the economic benefits for saw mills operators to adopt this technique. The conventional sawing process, although is practical and widely used, is less economical due to its inherently higher wastage and low quality product. In fact, the SDR process offers a more productive solution to the conversion of rubberwood throughout South East Asia.