HOW TO GLUE AND REPAIR WOOD WITH EPOXYThe important things you need to know to glue any wood |
||
|
||
Application of the adhesive on the surfaces to be glued For best results any liquid epoxy adhesive should be applied to both surfaces to be glued and allowed to sit long enough for the wood to soak up as much as it wants, so that when the pieces are assembled the wood will not absorb the glue that would otherwise fill the gap between the pieces. That leads to a glue-starved joint. Scarf and butt joints are especially prone to soaking glue out of the joint, as it wicks into the end grain of the wood, which is the open ends of the hollow cellulose tubes of which the wood is made. Edges of plywood are notorious for soaking up liquids. Many glue manufacturers have "solved" this problem by making epoxy glues that are gels or pastes, not liquids. Unless the user smears a thin film of such a glue forcefully against the entire surface to be glued, the paste or gel glue never actually wets the surface, and a low-strength glue joint often results. That alone is not enough to assure a strong glue joint. Many woods have acids [such as oak] or oils [such as teak] in them, and many kinds of epoxy products do not cure in contact with acidic woods, or cannot wet and stick to oily woods. More about this later, but that's not the first and largest reason for failure of wood glue-joints. Improper clamping of wood may be the largest source of glue-joint failure. Poorly-fitting wood elements, clamped to bend them into contact, will have tremendous spring-back forces pulling them apart, as much as a metal C-clamp can develop. Yet, the shear strength of wood is only 200-300 pounds per square inch. The result is that the wood fibers at the glue joint will tear away from their parallel neighbors, often within hours to days after the clamps are removed. Wood splits easily. That, basically, is what is happening here. Curved beams are best made by steam-bending the individual laminations, letting them dry in a fixture that sets the new shape, and then gluing them. Steam-bend wood holds its new shape without stress. Steam-bent ribs for boats have been used for at least hundreds and perhaps thousands of years. If steam bending is not an option, cross-grain fasteners or splines (tenons, biscuits) should be screwed or glued at each end, because the curved structure will want to straighten, and the glue joint will fail by cleavage. Straight, smooth, well-fitting wood elements can yet be made to fail by using excessive clamping force. Squeezing a glue joint down to zero glue-line thickness forces out almost all the glue from between the pieces, and the natural porosity of wood wicks away the remaining microscopic residue of glue. The result is a glue-starved joint. It will often fail when the clamps are removed, or sometimes days or weeks later. If the failed joint shows no divots of wood pulled out of the opposite side, the cause of the failure is almost certainly excessive clamping pressure and not enough time allowed before clamping. The wood element should be clamped gently: Just enough to squeeze out the excess glue and bring the wood pieces into contact at the microscopic high points of the joint. Thick pads of soft rubber under the clamp faces ensure gentle, even clamping forces. The second-major source of glue-joint failure may be Mill Glaze on the cut surface being glued. Wood that has been saw-cut, run-thru-a-planer, routed or otherwise cut or shaped with high-speed cutting tools can potentially have Mill Glaze. This is a thin film of varnish, made from the natural oil in the wood, that is polymerized (made from liquid oil into a solid) by the heat of the cutting-tool. When the tool is new and perfectly sharp, negligible heat is generated in its use, no oil gets "cooked" into varnish and there is no residual varnish-film on the surface of the wood. When the cutting-tool gets slightly dull, some heat is developed; as the tool wears more-and-more, a heavier film of varnish is left on the cut wood surface. A glue joint will easily fail, if Mill-Glazed surfaces are glued together. The reason is that varnish is physically weak, and the best glue in the world will stick to the varnish which isn't-at-all-well stuck-to-the-wood. A chain is only as strong as its weakest link, after all. Mill glaze is hard to detect when the tool is almost-sharp and the varnish film is very thin, but it can destroy glue-joint integrity nonetheless. Eliminating the risk that mill glaze poses to the integrity of your glue joints is very simple: Sand all glue joints before gluing. The sanding does not need to be with something really coarse, such as 100 grit. 220 may very well be adequate; the idea is to ensure a clean wood surface is present before beginning the gluing process. After sanding, clean off the sanding debris with compressed-air or a tack-rag. Don't use paper towels wetted with solvent, as the solvent will be absorbed into the wood and weaken the glue-joint. Woods may have acids or oils; some glues do not like that. Most adhesives, even epoxy adhesives, do not bond hardwoods because the acids, saps and resins in the wood interfere with the bonding chemistry of the adhesive. Our glues are specially formulated (by us - we’re chemists here at Smith & Co.) to overcome this difficulty. We designed a chemical system that would absorb and displace the saps and resins without becoming weakened by the absorbed oils. Our formulations are also compatible with the acids naturally found in many woods, particularly oak. Thus, hardwoods such as maple, acidic woods such as oak, and oily woods such as teak, apetong, araki, pau lope (Ipe), Osage orange, etc., may be glued directly with our epoxy adhesives. Oak & Teak Epoxy Glue is only one of several epoxy adhesives we make. Call the Factory Store at 1-800-234-0330 to discuss your particular requirements. Some woods—particularly ebony—contain a wax rather than oils. Saw cutting or dry sanding can smear this wax over the surface, making gluing difficult, especially on end grain or 45 degree bevels. Wet sanding or light abrasive blasting (such as glass bead or 200 mesh abrasive) can clean such material off the surface to be glued and has been found effective in improving the bond strength of such joints. Side grain bond strength, even with ebony, was found adequate with saw cut or dry sanded surfaces. It is important to remember that wood is a natural product and varies. If you use products that have fairly long thin-film set times, then you will have plenty of time to wipe up drips or shape into the desired form before the epoxy gels. Solvent-cleaning wood before or after gluing Even solvent cleaning hardwoods after gluing (while the glue is still wet) has in some cases, caused glue-line failures. Wiping up drips with paper towels is safe. These comments apply not only to our glues, but to any glue on any wood. Some manufacturers of glues that do not bond well to oily hardwoods, due to their incompatibility with the natural oils of wood, invented the idea of solvent-cleaning the wood so their products would stick to the wood. That is the origin of this idea. It is a bad idea, because solvents are absorbed into the wood, and take a LONG time to diffuse out and evaporate. Those solvents will be absorbed by the glue itself, and when the glue cures, it will be weakened by the high solvent content right where the glue contacts the wood. In effect, much of the glue is bonding to solvents instead of to the wood, and such glue-joints come back apart rather easily. That’s why solvent-cleaning wood or applying solvent-containing products to good-quality wood that has any reasonable amount of strength is a bad idea. Woods that may be directly glued include oak, teak, old-growth redwood [typically 20-50 growth rings per inch] and similar woods. Woods that require consolidation of the fibers before they are glued include “second-growth redwood” a physically weak and very porous wood, with perhaps 4-8 growth rings per inch. In that case Clear Penetrating Epoxy Sealer™ has proven helpful, particularly when glued with the very flexible Tropical Hardwood Epoxy™. Mixing two-component products Hygiene © copyright Steve Smith, 1972 - 2018, All Rights Reserved.
|