Static and Fatigue Strength of Oil Palm Wood Used in Furniture

MATERIALS AND METHODS

Commercial grade lumber of the five most common furniture wood materials (namely, Rubberwood (Hevea brasiliensis), Nyatoh (Palaquium gutta), Light Red Meranti (Shorea platyclados), Ramin (Gonystylus bancanus) and Sepetir (Sindora coriacea)) and OPW of the size 30x60x000 mm³ were obtained from a local materials supplier. All the samples were kept in a conditioning room at 20°C and 65% relative humidity, for about two months to ensure a final moisture content of 12±2%. All static-bending tests were carried out on an INSTRON universal testing machine, while the edge-wise fatigue tests were conducted with a specially designed air-cylinder loading system. All the experimentations were carried out at the Materials Testing Laboratory of the Science Company Inc., in Singapore, between June and October of 2009. The controlled environment in the testing facility ensured that the properties of the materials were kept constant during the test duration, which is important as strength properties of materials is influenced by humidity and temperature (Ratnasingam et al., 1997).

One hundred pieces of each of the experimental material were first ripped full length to a final dimension of 25x50 mm and then cross cut at mid length to form two matched pieces. One of the matched pieces was tested in bending in order to determine its density, moisture content and ultimate bending strength, while the remaining matched piece was used in the fatigue test. All the parts were coded accordingly to designate the test pieces. The dimensions of the specimens were in accordance to the dimensional requirements of components used in the manufacture of educational furniture frames in the country (Ratnasingam and McNulty, 2009). The static bending tests were conducted in accordance with the ISO 3133 (1975) for experimental samples as described in Bao and Eckelman (1996) and Eckelman et al. (2001) and the value of Modulus of Rupture (MOR) and Modulus of Elasticity (MOE) was established through calculation from the load-deflection curve after each test.

All the specimens were subjected to six different stress levels (i.e., 30, 40, 50, 60, 70 and 80%), expressed as a percentage of their respective average MOR. These load levels were chosen based on the previous study by Bao and Eckelman (1996) and after discussions with educational furniture manufacturing experts. Loads were applied to the specimens by means of air cylinders. A support span of 900 mm was used in keeping with the overall configuration of the test equipment. During the fatigue test, the air cylinder system exerted and released a non-reversal load at a rate of 20 cycles per minute. Mechanical counters recorded the number of cycles completed. Limit switches were used to stop the test when the specimen failed, or when 1 million cycles were completed.

RESULTS

Table 1 shows the bending test results for all the specimens used in the study. Generally, the wood materials exhibited the higher bending strength (MOR) and bending stiffness (MOE) values compared to the OPW, as expected.

Table 1:	Average values of physical and mechanical properties of specimens
Valus are represent as Mean±SD

Table 2:	Average values of fatigue life at different stress levels for different specimens
MOR expressed in N mm-2. Results were significantly different at p<0.05. RW: Rubberwood, LRM: Light red meranti, NY: Nyatoh, SPT: Sepetir, RM: Ramin, OPW: Oil palm wood

The MOR value of the OPW averaged 31 N mm^-2, while its MOE averaged 4324 N mm^-2. On the other hand, the average MOR for the other wood materials was 78.4 N mm^-2, with an average MOE of 12584 N mm^-2. The performance of the wood materials was however in the flowing order: Ramin>Sepetir>Nyatoh>Light Red Meranti>Rubberwood. These values were found to be statistically significant at p<0.05, indicating that the static strength properties between the different materials are different.

Table 2 shows that fatigue life of the OPW specimens for each of the different loading levels. Results indicate that at 30% of the ultimate strength, all the specimens could be expected to have a fatigue life above 1 million cycles. However, when the stress level was increased to 50% of the MOR, fatigue life decreased to an average of 203,000 cycles for OPW, while the other wood materials showed a fatigue life of more than 500,000 cycles. At higher stress levels, the fatigue life of all specimens was significantly reduced further. Similar to the static strength, the wood materials outperformed the OPW in terms of fatigue resistance. Statistically the results were found to be significant at p<0.05, indicating markedly different fatigue strength among the materials used in this study. Nevertheless, the results suggest that the order of fatigue strength among the wood materials were quite similar to the trend shown in the static strength.

DISCUSSION

If the loading to which a piece of furniture is subjected were static or almost static without any kind of cyclic loading, then the allowable design stress for the components could be determined from a consideration of the ultimate bending strength of the material and the desired factor of safety (Eckelman, 1988). If however the components are subjected to repeated live-loading environment, as it often the case for educational furniture, it is more reasonable to use fatigue strength rather than static strength to determine working stresses (Ratnasingam et al., 1997). Thus, when repeated loading cannot be ignored, allowable design stress should be taken as the percentage of ultimate strength that will ensure that the expected fatigue life of the components will exceed the expected number of service-life load cycles.

Test results indicate that fatigue life regularly decreased as the level of stress increased, similar to the report by Bao and Eckelman (1996). The study also reveals that solid wood specimens out-performed the OPW specimens at all levels of stress, clearly reflecting the superior fatigue resistance of solid wood materials (Eckelman, 1974; Thompson et al., 2002, 2005; Sugimoto and Sasaki, 2006; Sugimoto et al., 2006). At a stress level of 40% of its MOR, the OPW had a fatigue life of almost 500,000 cycles, while the other solid wood specimens all had 1 million or more cycles in fatigue life. Therefore, OPW could be used in furniture, where the repeated live-loading does not exceed 40% of the material’s ultimate strength. The results of this study suggest that OPW has fatigue strength similar to that of wood-based panels, as previously reported by Bao and Eckelman (1996) and Sugimoto and Sasaki (2006).

It is difficult to estimate the number of load cycles to which a piece of furniture will be subjected to in service. The number of cycles that a piece of furniture must withstand to pass accepted performance tests is well defined, however. For instance, the number of cycles stipulated in the British Standard BS 4875 for chairs and settees is 200,000 cycles (BSI, 1972). Hence, the allowable design stresses for the OPW furniture components might be set at 40% of the MOR of the material. The values obtained could then be further reduced, if desired, to take into account an appropriate safety factor and other adjusting factors required in the educational furniture frame design procedures (Ramirez-Coretti et al., 2009).

Design implications: In most furniture designs, the components are often oversized to accommodate the aesthetic needs of the joint construction. Such oversized components, while not contributing to the overall strength of the furniture, also results in wasteful application of the material (Ratnasingam et al., 1997; Huber and Eckelman, 1999). On this premise, it is apparent that OPW sized according to their fatigue strength can be effectively used as furniture components in the manufacture of low-cost education furniture construction, especially for school chairs and desks. Nevertheless, the fact that oil palm has a markedly different anatomical structure, which impairs its static and fatigue strength in comparison to solid wood materials (Ratnasingam and McNulty, 2009), suggested that furniture constructed from oil palm wood should include a higher factor of safety to accommodate the repeated live-loadings experienced in service. The vascular bundles, which constitute the building blocks in oil palm, embedded in a matrix of parenchyma cells, which in turn renders it with a lower ability to absorb repetitive loadings (Sreekala et al., 1997; Karenlamp et al., 2003). Nevertheless, with properly sized members and well-constructed joints, OPW could be successfully used in the manufacture of educational furniture that should be able to surpass the general acceptance standards of educational furniture in many countries within the South East Asian region. It must however be recognized that from an economical perspective, as long as the cost of OPW remain competitive to support the need for larger sized components with a higher margin for safety, its application in the furniture industry may still be a viable option.

CONCLUSION

Oil palm wood was subjected to edge-wise static load and constant amplitude cyclic load in an effort to investigate the fatigue properties of this material, in comparison to the other common furniture wood materials. Results of cyclic loading analyses suggest that fatigue life regularly decreases as the level of stress increases. When considering cyclic load effects, the allowable design stress for oil palm wood should be no more than 40 percent of its MOR, respectively. This number would provide furniture manufacturers a quick reference to estimate the appropriate sizes of oil palm wood furniture components that would satisfy the different GSA performance testing acceptance levels of educational furniture.

ACKNOWLEDGMENTS

The authors are very grateful to Dr. Stephen Avery of the Science Company Inc., in Singapore for his assistance during the experimentation stage of the study.