There are two general categories of fingerjoined lumber in Canada: NLGA SPS-3 “vertical stud use only” and NLGA SPS-1 “structural use”. The latter category is fabricated with phenol-resorcinol-formaldehyde (PRF) adhesive meeting the requirements of CSA O112.7-M1977 or with wood adhesives that meet NLGA “SPS Annex A” requirements. NLGA SPS-1 fingerjoined lumber is considered to be interchangeable with solid-sawn wood of the same species group, grade and size, and accordingly, can be used in either vertical or horizontal applications and under compression, tension or bending loads. NLGA SPS-3 “vertical stud use only” fingerjoined lumber may only be used in “dry service” conditions and vertical applications under axial (or near axial) compression loads, and with bending and tension loads (e.g. wind or earthquake) limited to short duration only. Two types of adhesives are accepted for use in the manufacture of “vertical stud use only” fingerjoined lumber: polyvinyl acetate (PVA) meeting the specifications of CSA O112.8-M1977 Type I and ASTM D 25559 without creep evaluation, and diphenylmethane di-isocyanate based products (also called MDI, isocyanate or polyurethane adhesives).
The use of glulam timber beams and columns has a long history and their fire performance has been well documented. Since NLGA SPS-1 fingerjoined lumber products were traditionally manufactured with the same PRF adhesives used in glulam timbers, there was little reason for building and fire officials to be concerned about the fire resistance of this “glued” product. On the other hand, PVA adhesives are thermoplastic products that creep and lose their bond strength at elevated temperatures. While they are thermosetting polymers, isocyanate adhesives chemically decompose at elevated temperatures. In theory, neither effect should be a problem since studs are subjected to axially-applied compressive structural loads. However, the simple fact that the adhesives used in the manufacture of some fingerjoined lumber products could melt or chemically breakdown at elevated temperatures raised concerns about all fingerjoined lumber in the minds of building and fire officials. In an effort to alleviate some of those concerns, a small exploratory project was carried out to examine the performance of fingerjoined lumber exposed to sub-charring temperatures and tension loads, and the fire resistance of gypsum-board-protected wood-frame walls constructed with fingerjoined studs.
While only a limited number of specimens of each type of fingerjoined lumber were utilised in this study, the moduli of elasticity (MOE) of those studs were far greater than the “modulus of elasticity for design of compression members” assigned by CSA-O86-01 for the respective grades. Depending upon the grades of fingerjoined lumber and adhesives used in their manufacture, the MOE of between 60 and 95% of the pieces were greater than the “specified modulus of elasticity” assigned by CSA-O86-01. Many building officials believe that while the strength properties of fingerjoined studs are more consistent, their average strength properties are less than those of solid-sawn wood studs of identical grade, species and size, and therefore, the fire resistance ratings of walls constructed with fingerjoined studs may not be as great as those for walls constructed with conventional wood studs. While admittedly limited to an evaluation of MOE of a very small “population size” (20-24 specimens of each material), the data generated in this study does not support their contention.
SPF No. 1 and No. 2 lumber lost about one-half of its strength in tension when heated to 200ºC and almost two-thirds when heated to 250ºC. Specimens of SPF No. 2 SPS-1 fingerjoined lumber manufactured with a PRF adhesive lost about one-half of their strength in tension when heated to 150ºC and about 60% when heated to 200ºC. SPF No. 3/Stud grade SPS-3 fingerjoined lumber fabricated with an isocyanate adhesive lost about two-thirds of its strength in tension when heated to 150ºC and about three-quarters when heated to 200ºC. SPF No. 3/Stud grade SPS-3 fingerjoined lumber fabricated with PVA lost more than 90% of its strength in tension when heated to 150ºC and essentially all its strength when heated to 200ºC. Based only on the proportion of “wood” and “glue-bond” failure in each separated fingerjoint, one might reasonably conclude that, when exposed to temperatures up to 250ºC, the adhesives in each fingerjoint in SPF No. 2 SPS-1 fingerjoined lumber bonded with a PRF adhesives has as much strength in tension as the wood fibres in the pieces of lumber that the fingerjoints hold together. Similarly, one could conclude that the adhesive in each fingerjoint in SPF No. 3/Stud grade SPS-3 fingerjoined studs fabricated with an isocyanate adhesive has as much strength in tension, when exposed to temperatures up to 200ºC, as the wood in the pieces that the fingerjoints hold together. Finally, this study indicated that even under ambient conditions, the adhesives in each fingerjoint in SPF No. 3/Stud grade SPS-3 fingerjoined studs fabricated with PVA has less strength in tension than the wood in the pieces that the fingerjoints hold together. These observations do not provide any indication about how fingerjoined lumber products perform as studs in a wall during a fire. Therefore, the wood industry should discourage building officials from specifying tension tests at elevated temperatures for acceptance of fingerjoined studs in fire-resistance-rated walls.
Intermediate-scale fire resistance tests were carried out on gypsum-board-protected wood-frame walls constructed with solid sawn and with SPS-1 and SPS-3 fingerjoined studs. In each test, failure was due to the studs in the walls bowing out of the plane of the wall and away from the fire. The amount of bowing was sufficient that the walls were forced out of the furnace frame at mid-height, and hot fire gases could escape from the furnace between the specimen frame and the outer studs in the walls. However, the bowing was only noticeable during the last three or four minutes of the tests. Most importantly, in no case was there an abrupt collapse of the wall: only a continually increasing amount of bowing outwards by the studs during the final minutes. Fear of the sudden collapse of a wall with no obvious indications of its impending failure is the catastrophic event that had originally raised concerns in the minds of building and fire officials about fingerjoined studs. This phenomenon was not observed during any of the tests on walls constructed with fingerjoined studs. The fire resistances of walls constructed with SPF No. 2 SPS-1 (PRF) and SPF No. 3/Stud grade PS-3 (isocyanate adhesive) fingerjoined lumber were not significantly different from that of a wall constructed with solid-sawn SPF No. 1 and No. 2 lumber.
The primary goal of this research was to determine if there might be any foundation for the concerns of building and fire officials about the fire performance of wood-frame walls constructed with fingerjoined studs. None were identified. However, the performance of walls constructed with SPS-3 (PVA) fingerjoined studs in the intermediate-scale fire-resistance tests requires additional research since all the fingerjoints in those studs separated during the “non-standard” tests used in this study.
The limited resources available for the study (it was only intended to be an exploratory study), resulted in some issues related to the fire performance of fingerjoined lumber not being investigated (hose-stream tests) and a number of others being investigated using non-standard testing procedures (intermediate-scale fire-resistance tests) and limited numbers of specimens (strength in tension at elevated temperatures). Furthermore, whenever fingerjoined lumber is used in fire-rated construction in the USA (e.g. wood-frame apartment buildings and many non-residential buildings), the loadbearing walls of those buildings must have one-hour fire-resistance ratings. Therefore, a committee composed of representatives from major manufacturers of fingerjoined lumber, the committee responsible for writing NLGA standards for fingerjoined lumber, and researchers from Forintek should be convened to review the results of this study and to establish future directions for the wood industry to take when addressing the concerns of building and fire officials about the fire performance of wood-frame walls constructed with fingerjoined studs, and of fingerjoined lumber in general.
Test results for three representative adhesives were obtained for use in the development of a proposed standard for limited moisture exposure (CSA O112.10). The adhesives tested were an emulsion polymer isocyanate (EPI), a polyurethane (PUR) and a melamine-urea formaldehyde with 40% melamine resin content (MUF40). Currently, EPI and PUR are used for I-joists and fingerjoined lumber. MUF40 was included in the study as a non-conforming adhesive. The range of performance of these adhesives, along with that of melamine formaldehyde (MF) and polyvinyl acetate (PVA) evaluated in a previous study, is baseline information used in defining acceptable performance levels for adhesives undergoing block shear tests required in the proposed standard.
Specimens in this study were evaluated under five test conditions: dry, vacuum-pressure wet or re-dried, and three-cycle boil-dry-freeze wet or re-dried. Dry and re-dried test conditions are the proposed test protocols for the draft CSA O112.10 standard.
In terms of shear strength and percentage of wood failure, EPI and MUF40 met the requirements of CSA O112.9 for the dry test condition, and PUR did not.
The following block shear test requirements are recommended for CSA O112.10, based on the 95% lower confidence limit of the EPI test results, and structured to be analogous to the requirements of CSA O112.9:
Median dry shear strength = 10 MPa (1450 psi) (adopted from CSA O112.9);
Vacuum-pressure re-dried median shear strength = 7.4 MPa (1070 psi);
Median percentage wood failure = 85% for all the proposed tests (adopted from CSA O112.9); and
Lower quartile percentage wood failure = 75% for all the proposed tests (adopted from CSA O112.9).
The above requirements will be discussed in the CSA Task Group, which will eventually make recommendations to the CSA Standards Committee.
The premature removal of treated wood from service, due to weathering (checking, distortion and UV degradation), rather than decay has led to increased acceptance of products promoted as low maintenance. Wood-plastic composite decking in the USA is anticipated to post 15% annual growth through 2009 to almost 900 million board feet (Freedonia Group 2003). Plastic and wood-plastic composite lumber are projected to capture 25% of the decking surface board market in the USA. This is a direct threat to the approximately 2 million cubic metres of softwood lumber annually treated with copper amine-based wood preservatives for residential and commercial exterior products. These products have also raised the bar for performance and price, 2 to 3 times that of wood decking. There is a considerable amount of research underway to develop methods to improve the serviceability of decking. These include: material sorting, profiling, water repellants and coatings. Current standard test methods focus entirely on durability and do not take appearance into account. Standard test methods are therefore needed to evaluated processes designed to improve serviceability.
Canadian wood preservations standards do not include any test methods and instead reference the American Wood Protection Association (AWPA) standards. The AWPA has recognized a need to update their standards in the area of serviceability and has set up task forces on test methods and performance criteria. These task forces will be the venue for review, revision and further development of these standard tests. FPInnovations has been conducting tests of deck boards for many years and has included evaluation of checking in some of these tests. FPInnovations has also been conducting tests of coatings and the underlying substrate using methods adapted from those of the USDA Forest Products Laboratory. Elements of these methods have been put together and further developed to create a standard serviceability test using a decking module.
One of the most important characteristics of wood is its tendency to check due to weathering. However, it can take several months or years under natural conditions for significant checking to occur. Hence an accelerated predictive testing method is urgently needed. Dr Phil Evans, formerly of the Australian National University, now with the UBC Centre for Advanced Wood Processing is one of the few researchers that has focussed on the issue of checking. He has developed a prototype accelerated checking machine that shows promise as a screening tool for processes to improve serviceability.
Report #1 presents results on serviceability of a service trial of profiling and coating after one year exposure using the inspection methods in a draft AWPA standard test for serviceability. The draft standard is attached as an appendix to this report.
Report #2 describes the work done to develop an accelerated test method.
Report #3 presents results on an accelerated test of profiling using the inspection methods in a draft AWPA standard accelerated test method. The draft standard for this accelerated test is attached as an appendix to this report.
A literature review was conducted to identify potentially effective nano-based technology for improving wood attributes in order to develop competitive specialty wood products. The review covered both conventional chemical treatment methods and nano-based methods. Traditional chemical treatments have shown to be effective in improving wood hardness, dimensional stability, stiffness, fire resistance, UV resistance, biological resistance and aesthetic appeal. However, nanotechnology offers new opportunities for further improving wood product attributes due to some very unique and desirable properties of chemical materials in the form of particles in nano scale. The advantages of nanotechnology appear to be particularly obvious when applied to create polymer nanocomposites such as wood coatings. Polymer nanocomposites consist of a continuous polymer matrix which contains inorganic particles of a size below approximately 100 nm at least in one dimension. Due to the material nature of solid wood products, creating a continuous polymer matrix with effective inorganic nanoparticles inside wood cells and lumens would be very difficult. The most promising areas of applying nanotechnology to create improvement opportunities would be wood coatings and wood adhesives. It is recommended that research be carried out to explore the potential of nanotechnology in wood coatings and adhesives and their applications in B.C. wood species and wood products.