I am working on a rather large bookcase for my son. It's frame and panel construction and has a lot of M&T joints. More than I have ever done before. Other than my kitchen cabinets I made ~15 years ago, this is the largest thing size-wise I've ever attempted.
I am in the middle of finishing the panels before glue up and thinking about (worrying about really) when I do glue it all together and the quality of my M&T's. Even though I used my mortising machine, they aren't prefect and there isn't a ton of surface area for the glue and mating surfaces to begin with.
I was reading a FWW article yesterday from 2005 written by Ernie Conover about building a Shaker Rocker. In it, he said he uses 90 minute Epoxy for all the M&T joints.
That got me to wondering if I should consider using some kind of epoxy.
I know zero about epoxy.
I've never used epoxy before with wood.
Should I consider it? What should I know about it?
If assembly time is an issue use liquid hide glue. If joints are in danger of seizing upon assembly use liquid hide glue. If joints are sloppy-loose shim with veneer and then resize with shoulder plane. Don't own one? Buy LV medium as a start.
M&T are strong joints. Glue just keeps then from pulling apart.
Sparse mating surface....be more specific. In one sentence you talk about lots of joints and in another you worry over not enough glue surface. Lots of joints with barely acceptable surface times lots of joints equals robust construction.
I can recall many years ago having to glue up projects with large complex tight fitting joints using Tightbond 1, which was the only one available back then, and having it grab before I could get all the clamps on, having to use the larges hammer hitting a wooden block, then having to move up to 8# sledge. The angst one feels in times like that probably takes a year off one's life.
Like most converts, I started with West System because back then they were the first or only ones to advertise in the mags I read. I liked the pumps for less than 5 pumps, but some of my projects might have required a gallon. As the pumps get old and in cold temperature after getting about 10 of 15 strokes, they would take in a big bubble and I'd need to start over. I then switched to System 3 which I liked better, but then finally settled on FGCI which is about half the price of the other two, and is a 1-1 mix ratio and not all that sensitive to being off a little. I keep clear disposable Solo cups to mix in, and the little 2oz cups for small batches. I can just set two cups beside each other and pour equal amounts in each.
This epoxy is fairly viscous to begin with, but I keep colloidal silica on hand to mix into it for however thick I want it to be. For veneering, I may want it to be about like latex paint. for M&T maybe catsup, or thicker if there is enough space for it to slowly run out of a joint before it kicks.
Epoxy unlike the water born adhesives does not penetrate the cell walls, thus needs a bit of texture or tooth to grab onto, and likes a thicker glueline. It also acts like a lubricant for the joints during assemble, so you may be surprised that it goes together easier than your dry-run. I don't use epoxy for edge gluing, and if I think something is too smooth, scuffing with 80 or coarser grit can quickly add the needed texture.
Since you mention prefinishing, during your dry run fit up, is a good time to use masking tape around the joints to protect against squeezout. Besides adding the silica, I keep the dust from my belt sander bag in jars from various colors to add to the mix. By themselves, they will darken more than the wood species that it's from, but with the silica a little will do the trick.
The one mistake you can make is not mixing enough. I stir vigorously, rotating the cup in my hand and making a conscious effort of scraping the sides and bottom of the container, as well as occasionally scraping the stick off one the lip. Any unmixed resin in the mix will never harden, so mix it good.
A couple of things. Epoxy cures faster the higher the temp, so you can gain LOTS of open time when working large projects if you keep it cool, then warm it up after assembly. And it does need to be warmed up. In cool weather, after measuring into my mixing cup, I move to my sink and mix while the hot water is flowing over and around the outside while I rotate the tilted cup. When it feels warm, I head out to the project and start spreading. The mixing stick becomes a good applicator for M&T joints. Besides the masking tape, alcohol on a rag is a pretty safe clean up unless you used shellac of course. Also, since it cures slowly, you can wait until the cure is about like leather, and trim it off with a knife or chisel.
Another warning though, is don't mix more than you can spread before it starts to generate it's own heat. If you have more than about 1/2 of depth in your container, it can heat up fast and melt the cup you mixed in. This is not an issue after it's been spread out thin. So mixing a second or third batch after spreading the first is less of an issue than having more mixed than you can get spread before it heats up. It may not kick for 6 hours after spreading out, and finding you need to adjust something well after you thought you were done isn't an issue. You can move a part as long as the epoxy is liquid enough and it will still bond the other side as long as they touch.
The link below will get you to the site, but they have a 800 number and are good at answering questions if you tell them what you plan to do.
With any epoxy that you are new to, ALWAYS ask if they have an instruction manual. Get and follow it.
It is the nature of a curing polymer film that it will have huge impact resistance and little shear resistance. For those that don't use epoxy, force stuck joints together with steady clamp pressure. Bashing will likely just demonstrate the impact resistance of the curing glue.
My go-to for all joinery is liquid hide glue. Four reasons: most importantly, it is reversible. This is essential when you need to make repairs. Also, for multiple joints, it has a longer open time. Titebond LHG will clean up in water and does not affect finishes. Fourthly, new hide glue goes directly on old hide glue - that does not work well with yellow or white glues.
I do not use epoxy much, except as a filler of holes when tinted to match the wood. The biggest advantage of epoxy over other glues is that it is gap filling. I would rather have a well-made joint than fill one, but this can be an advantage where the joinery is not easily fine tuned. It is also possible to undo a joint with heat, but I am not sure whether one can epoxy on epoxy.
I am happy to glue up panels with Titebond III. It has a good balance of open time for this task. LHG here takes too long and I suspect (not sure) whether it is as strong as TB III. TBIII is my general purpose glue.
I have used epoxy for M&T joints. If they are looser than ideal, it has excellent gap filing properties. I use saw dust from a sander to thicken it if necessary. The second reason is long open time. I use the 1 hour variety. I try to have my shop at about room temp and have never had trouble with it kicking early. I assemble, use clamps to pull up recalcitrant joints Clean up with Denatured Alcohol, Check carefully and let it cure over night.
On Panels, I pre-finish them, at least stain and one top coat. I sometimes cover the corners with thin plastic like Saran wrap or a plastic bag to prevent gluing the panels into the corners. Once the glue is set you can often pull the plastic out of the joint. If not, cut it away with a sharp knife or razor.
"Fourthly, new hide glue goes directly on old hide glue - that does not work well with yellow or white glues. " Do you have data to support this claim? I have seen unpublished data to the contrary. From the perspective of the polymer chemistry involved I see no reason that new glue should not bond to cured glue. All that has to happen is for the solvents in the new glue to temporarily plasticize the first few molecular layers of the cured glue.
Why do you prefer Titebond III over Titebond original? (I have never used III so I don't know how it differs. just know it costs more)
I use TBIII because it has a longer open time than TB original, and it can be applied at somewhat colder temps, an important consideration for my unheated shop.
Bill, personally, I have not had any success adding new yellow glue to old yellow glue. There is a great deal of similar reports from others. I believe it is accepted that this is so. No, I have not read technical reports.
Why TB III and not II? As I understand III is more strongly water resistant. It also has a longer open time to II.
"I have not had any success adding new yellow glue to old yellow glue"
Was the glue III? If so, that is understandable. If the glue was original,what was the situation where new glue did not bond to cured glue and how did you know the bond was defective? As far as reports from others, while I have seen this claim repeated faithfully by hide glue enthusiasts I have never seen them support the claim with data from a controlled experiment. If you have reference where the claim was supported with strength dataI would like to have a reference to the report.
The common repair situation would be taking apart a M&T joint and regluing. Anyone willing to try, can confirm that a yellow or white glue M&T joint can be disassembled with the same technique as a hide glue joint. Experiments show that this joint can be reassembled with nearly original strength without removing all the old glue.
Possibly the conventional wisdom of yellow glue not adhering to itself comes from years of individuals unsuccessfully re-gluing failed joints. If so, an obvious explanation is that its not the glue-glue bond that is failing, but the loose joint itself.
This is a subject where some actual studies would make for very interesting reading. It sounds like the only work on the subject is unpublished however.
The long chains of molecules of epoxy are too large to penetrate the cell walls of wood like TB and other water born glues, so it is merely a mechanical bond to the surface. The more texture, makes a better bond. Do you have a link to show it's use on polished surfaces?
My local woodworking club received a briefing on adhesives a couple years ago from the director of research at Franklin, the manufacturer of Titebond and their Liquid Hide Glue. When the subject of glue for M&T joints came up he recommended the use of epoxy. I found this unusual because Franklin does not have an epoxy product. His reasoning was based, I believe, on the characteristic of PVA glues, and even Liquid Hide Glue to a lesser extent, to lock up and resist further movement of an M&T joint if there is a time delay from when the tenon is pushed into the mortise and when it is attempted to be fully seated.
I have experienced this issue with the assembly of the undercarriage of a Windsor chair when 10 M&T joints need to be assembled. Because of the need to not fully seat all the stretcher-leg joints before driving the 4 legs into the seat these joints have tendency to lock up requiring the use of persuasion with a hammer. This is not something I look forward to and have gone to epoxy for these joints quite successfully.
Not what you asked for but the principal is the same. The two surfaces were polished steel. Any adhesive that "wets" the smooth surface will work as an adhesive.
I attended a technical presentation on adhesives by someone from VPI. He talked about epoxying aluminum airplane parts together. I specifically asked him about wood and whether the bond was mechanical or adhesion. He said wood is so porous that the two ways of sticking could not be separated.
The glue in Titebond is a suspension of micron sized glue particles. These are a thousand times bigger than an epoxy molecule. Neither penetrates a cell wall. Either can penetrate wood pores size wise. Don't know if the surface properties of epoxy are such that it will wet the wood.....probably would be my guess.
One can epoxy on epoxy. However, as epoxy cures it can produce a surface contaminant called amine blush. The following article explains amine blush and how to deal with it.
A page from West System, Surface preparation *LINK*
Keith Newton
Thoroughly abrade the surface. 80-grit aluminum oxide paper will provide a good texture for the epoxy to “key” into. Be sure the surface to be bonded is solid.
I think all the brands I've used suggest the same.
Not all epoxy exudes this waxy amine blush, which is why I was happy to find FGCI.
Oh, another interesting thing about epoxy is that more epoxy can be added to earlier mix as long as the first isn't cured they are suppose to cross link.
Hey Paul, Good to hear your experience, and interesting that the Franklin rep suggested epoxy as well. I've probably used 100 gallons over the last 30 years, and never had a problem, and can't imagine not having that option.
"oxide paper will provide a good texture for the epoxy to “key” into.
I acknowledge that West should know how adhesion works, but whoever wrote these instructions doesn't seem to. This statement makes no sense. Scratching makes vees. How does a material flowing into a vee "key" into a vee? Scratching is going to increase surface area but even that makes no sense. The projected surface area, which determines tension strength, is the same scratched or not. Scratching will likely improve shear strength on something not porous. Scratching wood is just going to make a weak surface layer of torn wood fibers.
I am unsure if the sculpture I found is made as I describe but if not similar sculptures are. I have watched the process done and discussed the details with the artist. Polished pieces of glass are glued together with an epoxy that is colorless and has the right refractive index so the result does not revel the epoxy layer. This process demonstrates that adhesion, as scientifically defined, has noting to do with "keying". It relies entirely on the liquid (adhesive) wetting the surfaces being bonded. The smoother the better.
The only thing I can think of is that removing a layer that interferes with forming a strong bond may be required of some material. Sanding would accomplish this task.
Is different coefficients of materials can cause expansion and contraction due to various external changes . Wood changes metal responds to temperature changes. Epoxy would more easily shear right off a polished surface.
Most epoxy alone is quite brittle, but the additives allow you to change that, as well as a huge selection of catalyst.
