If I’m gluing a joint that involves some end grain, is there something special I’m supposed to do on that end grain part with regard to extra absorption of the glue? I have some vague, vague memory of reading sump’in somewhere, but can’t dredge any details up.
forestgirl — you can take the girl out of the forest, but you can’t take the forest out of the girl 😉
Replies
Sometimes it helps to "size" the end grain with thinned glue. Apply and let dry for a few hours before final glue up.
Jim
Jamie -
You have other mechanisms to hold this joint together rather then just end grain to long grain....right? Endgrain glue joints is the least strong way of joining two pieces together.
Biscuits?? Spline??
[email protected]
Yep, I know that. Here's what the joint looks like -- pardon my scrawl. The material is a little over 2" wide, 5/8" thick. I was planning to put 2 brad nails in each corner. There will be molding applied around the edge that is joined 45 degrees at each corner like a picture frame, pinned and glued to the edge.
The front of the case will be plexiglass, not too heavy I hope. Back is plywood inset into a rabbet, and screwed to the oak case.forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Jamie,
That is a very poor joint. There is nothing but endgrain to long grain contact. The joint may stay together for a while but will eventually fail even without much stress.
End grain joints have no inherent strength without some other mechanical reinforcement. Glue has little or no adhesion to the thin ends of the walls of wood cells and tubes. Only to the side structures (long grain).
The brads will provide almost no strength and molding if mitered at 45° over the joint also will be a very weak reinforcement.
A lap jount allowing long grain to long grain gluing would give you a (much) stronger joint.
R
Well, sounds like I'll have to add some dowels, per Fred. Rich, I'm not following you on the lap joint -- in my very poor drawing, you're looking at the boards on edge. I'm not seeing a way to do a lap joint in this corner, but visualizing 3-D isn't my strong suit.forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Jamie,
Here's an example. #1 has 1 long grain to long grain gluing surface. #2 has two.
R
Better image (I think)
R
Edited 10/23/2002 1:48:49 PM ET by Rich Rose
Hmm...
Ahhhh, I see. #2 is sort-of like a bridle joint, eh? I'd have to consider these joints in terms of aesthetics before deciding to use them in the display case, but they do look like nice strong, and simple, joints. Thanks Rich.
forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Edited 10/23/2002 4:48:34 PM ET by forest_girl
Jamie,
In designing glue joints that don't have any other mechanical holding power (a dovetail joint, or locking miter joint), you must design so that long grain meets long grain. Even locking joints need long grain mating.
Glue cannot adhere to end grain. Long grain surfaces can be oriented parallel to each other or they can cross at any angle up to 90 degrees such as in a cross lap joint or a bridle joint. The important thing is that long grain surfaces mate in the joint.
Cross two boards so that their long grain surfaces meet, glue the mating surfaces, and those boards will probably break if stressed anywhere but the glue joint.
You can't glue two boards end to end, because all that's in such a joint is end grain. But if you create finger joints on the ends of the boards, those fingers expose an enormous amount of long grain surfaces which result in a glue joint stronger than the surrounding wood. Such a joint is essential to the laminated beam industry (glulam).
End grain to long grain is no stronger than endgrain to endgrain. The glue aheres to the long grain surface, but fails on the end grain which is nothing but the thin edges of wood cellular and tubular structures.
You should aways analyze your joints (and those of others!) for long grain to long grain construction. Change any joint that has only endgrain on one of the surfaces. It will fail. It is amazing to see some workers totally unaware of these principles, especially in chair construction and bookcase construction. Dowels and pins seldom keep a poor joint from failing. Some bookcases are built with nothing but dado and rabbet joints which have no long grain to long grain mating at all. They stay together because the back holds them together!
Think of a box joint. It is almost the ultimate in creating mating surfaces of long grain to long grain. The dovetail joint is actually weaker in that the angles in the joint take the mating surfaces away from long grain toward end grain. But the locking nature of the joint makes it very strong in the locking direction.
A dowel in a hole (which many people think of as a strong joint) is really very weak and survives mostly because of other mechanical factors. Its own endgrain has no glue strength. The sides of the dowel are all long grain, but they meet long grain in the surrounding hole only at two narrow tangent lines. All other grain in the hole is either end grain or grain transitioning between endgrain and long grain.
One of the best books to read is Taunton's "FWW on Joinery."
R
OK, I guess all those bookcase joints I see got me off on the wrong track. So...if I want long-grain to long-grain contact, but don't want the joint showing on the sides of the case, my only option is a tongue-in-groove, with the tongue on the top/bottom pieces, and the groove in the side piece (so the only place there is end-grain showing would be on the top and bottom of the case). I could probably convert these pieces to this joint, without losing too much in the dimensions of the case. Unfortunately I don't think I have any scraps to practice on, but could make a bunch.
It's either that, or add splines a la Dennis. Of course, if I can't remedy the finishing problem, it's a moot point :-(forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Jamie,
It's not easy to get good joints without the joint structure showing. I think what you propose as a "tongue in groove" is a dado joint (unless I'm picturing it wrong). And that has no strength either.
My favorite joint for mating the sides to the top and bottom of a bookcase (or similar carcass) is a dovetail joint (or a box joint). But such a joint does not fit with certain styles. A half blind joint can be made. Or a full blind joint can be made consiting of multiple tenons fitting into multiple mortises. Such tenons must be cut to maximize the long grain mating - they are often incorrectly cut so that more end grain mating occurs. (I'll post an illustration, but can't do it for several hours. I'm at work on a machine that has no graphic programs).
James Krenov disagrees with the kind of advice I have given. He finds nothing wrong with small carcass joints being held together with many dowels along the length of the joint. He describes several such cabinets in his books.
Who am I to argue with a god?
R
Edited 10/23/2002 6:00:32 PM ET by Rich Rose
Edited 10/23/2002 6:02:10 PM ET by Rich Rose
A good saying I was tought:
Long grain to long grain is a full joint.
Long grain to end grain is half a joint
End grain to end grain is no joint
Jeff in so cal
Jeff,
Long grain to end grain is no joint. If one side of a joint fails, the joint has failed.
R
Hi Rich
What it means is long to long needs nothing.
Long to end needs some other means to hold it. As in dowels, mortise and tendon. You do get some holding (1/2) with end to long. But it will always fail at the joint with out something else to hold it.
End to end doesn’t hold at all.
Hay my 7th grade Teacher needed something to teache us. Sounded good enough to me to remember it all these years!
Jeff in so cal
Hi Rich -
I've made some mirror frames and such that are similar to the piece Jamie's making and used slip feather splines of contrasting wood to reinforce the corners. I think these, if carefully done, are actually somewhat attractive pluse you can add them after the frame or box is complete. I made a 45 deg. jig support to run the corners through the table saw then made the feather splines by hand to slip fit into the grooves.
Dennis in Bellevue WA
[email protected]
Hi Dennis, thanks for keeping me from feeling like a complete idiot <g>. I got this joint out of a Sam Allen book on joinery, knew it wasn't the strongest, but it was what I felt up to in terms of skill and the time required to make it. Dovetails aren't an option, rabbet and dado didn't go well when I tried it (quite a long time ago, could probably do it now that I understand the process better), and didn't trust my 45-degree table saw set enough to simply miter and reinforce with splines. My saw is now dead-on for the 45's so next one may be done that way.
Just want to get this case up on the wall for now, build a better one next time. So, your splines are just a single spline in each corner? The flat face is parallel to the front and back of the case?
I dunno, might go with dowel pins for this case as I think I'm less likely to mess those up. Have already put a ton of time into this dude, and will be seriously unhappy if I screw it up to the point it has to be rebuilt from scratch. As it is, the finish is giving me fits (maybe you can help, I'll be posting in "Finishing" either a new thread, or in the "Pore filler" thread I have).
See you Friday?forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Hey - I only know what I've read and I haven't read nearly all that's out there so the obvious conclusion is,.... I don't know it all either! (grin)
One of my fav' reference books on joinery is Tage Frid's "Joiner: Tools and Techniques" (the Taunton Press ISBN 0-918804-03-5) Mine isthe fourth printing '82. Tage is a delightful read if only for his subtle humor.
Slip feather or through dowel reinforcement for mitered corner joints are pretty easy after-the-fact type of operations. Once the frame is assembled, square and can be clamped to diagonal braces to keep it from coming apart, make a jig for your table saw that consists of a 90 degree 'cradle', if you will, that is rotate up 45 degrees from the saw table:
/ work piece
--------
________ cardle
-------------- <- saw table
OK, so you can't draw a 90 degree angle with ascii characters but you get the idea. The cradle should be as close to 90 degrees as you can make it. The rotaion off the table should be, ideally, as close to 45 as you can get it as well but this isn't too critical except for the amount of exposed reinforcing feather dimensions; too far off the 45 and it will become noticeable; close and no one will bother to ask (grin).
Just run the frame, all secured from moving around, through the saw blade set to a depth that will *almost* reach the inside of the frame. Then rip some thin pieces of material to fit in this slot. Keep in mind that the direction of the grain of the feather pieces should be at right angles to the grain of the frame to get the most out of them.
When I do this, I take a thin piece of wood and apply the glue to the inside of the slot. Don't know if this is preferable to simply gluing the feather piece but it seems intuitive to me that to get any glue between the feather and the frame pieces at the bottom of the joint you'd have to do this. Perhaps applying a thin film of glue to both the inside of the slot and the feather is even better. Yeah, that's what I did (the memory is the second thing to go; I don't remember the first thing).
Where the frame is deeper than it is wide you might want to set up your cradle jig/fixture such that you could rotate the frame and create two slots offset at 1/4 & 3/4 points of the depth which would be tempered by the door detail as well.
I trimmed the feathers roughly before glueing up but you could probably leave them hanging out since you'll want to be paring them off and sanding flush for a perfectly finished surface anyway.
For a dowel reinforced joint (which could be pretty trick looking - I've not tried it (yet)) you could make another type of jig for the drill press to drill through the corners and stick a dowel all the way through (smooth wall, though) then pare and sand for a little drama to the expression of the reinforcing as well.
I agree with Rich that the dowels don't add as much strength to the mitered joint for the work involved. Some slip feathers, however, do, and they're not hard to do at all.
There's also inside corner blocks, don't forget. You could trick up this approach by running a cove bit along the edge of some wider stock then rip off what would be, in effect, a little cove piece. Whack off pieces to glue in the corners and you've got face-face or long grain/long grain mating surfaces.
For the minimal depth of the cases you're building, all this is probably overkill since the back once glued in place will probably hold everything together. But we like to be sure, don't we! (grin)
Dennis in Bellevue WA
[email protected]
Personally I think, I would have use a dado and the rabbit, instade of the two rabbits.
Maybe two 1/4" dowel pins each corner. If you can’t get biscuits.
Do nothing special. Wood glues need clean non-coated wood cells to develop full strength. Just use a little more glue to account for the somewhat higher absorbsion.
Jamie,
Another thought, although perhpas others would disagree, is a lock miter joint, cut with a router bit of that name in a router table set-up. I occassionally use this joint, even in solid wood, and it is quite strong. It is a pain to set up the router, tweaking the adjustments on scrap is necessary.
>"tweaking the adjustments on scrap is necessary." Har, har, har, you should have seen the pile of scraps I "tweaked" just getting this joint right. Too funny. The locked miter is one I considered for this project, but wanted to go more basic for the time being. I've heard about how hard it is to get those to meet right. forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
This forum post is now archived. Commenting has been disabled