*
I’m building a hall table for a customer who wants a wide drawer (for a D.C. phone directory). The apron is 7″ wide with mortise and tenon
joints to the legs. The wood is 4/4 air dired (3 yrs in my shop) cherry.
What is the max. width apron that won’t crack due to restricted seasonal
movement?
Bill Lindau
Discussion Forum
Get It All!
UNLIMITED Membership is like taking a master class in woodworking for less than $10 a month.
Start Your Free TrialCategories
Discussion Forum
Digital Plans Library
Member exclusive! – Plans for everyone – from beginners to experts – right at your fingertips.
Highlights
-
Shape Your Skills
when you sign up for our emails
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply. -
Shop Talk Live Podcast -
Our favorite articles and videos -
E-Learning Courses from Fine Woodworking
-
-
Replies
*
I tend to use a mortice and tenon as per the attached sketch for anything over about 130 mm - 150 mm wide rails. True the sketch depicts something more than twice as wide as your 7” rail, but you can scale down accordingly. It’s in metric of course, if you favour imperial measure. If you have a similar rail that emerges from the other side of the leg, and the mortices meet at the bottom, you can mitre the inside face of the ends of the tenon to get the maximum gluing surface. This joint will allow the 7” rail to expand and contract, and prevent it cracking.
< Obsolete Link >
Secondly, say you want a snugly fitting centrally located 5” tall drawer front in your 7” wide rail. Start with a rail at 7-5/16”, rip 1” from both long edges on your table saw, chop the drawer length needed out of the wide central piece on a chop saw, and rejoin the remaining four bits that make up the rail in the same order that they came apart. A skim or two with a jack plane to remove your saw marks prior to gluing it back together should leave an almost invisible join. Don't click on the blue letters below. It's the same image. And my deleted response was a failed attempt to just what I did here. Sliante.
*Thanks again Sgian for your generous contribution of time to helping others. The diagram is very clear and I'm sure to use the concept myself. Keep cool down there. It's only 80 here in Cincinnati.
*Don, Very generous of you, but sometimes I can be bloody minded too as I'm sure you know, and that has something to do with it being hotter than hell and twice as miserable here most of the time. ;-) By the way, I forgot, the joint is called the 'forked mortice and tenon [with extended mortices.]' Did a brain fart too; the 7" rail proposed is a little more than half the width of the rail sketched, not the opposite as I described. Also, just in case it's not clear for some reason, the leg is sketched as if it were rotated through 90 deg so as to look down into the mortices. Sliante.
*Hi Richard, Are the haunches really necessary on this joint? Couldn't you move the upper tenon to the top of the rail and eliminate the center haunch? Of course you would leave an empty space between the tenons. The upper and lower tenon would prevent any cupping. Although I suppose that it is no big deal leaving the center haunch, except that you need a corresponding mortice for it. Could you also just cut one mortice of the proper size to hold both tennons and the center haunch, provided that you did not let the bottom tennon contact the lower edge of the mortice such that the bottom tenon was free to move with expansion and contraction?Bob
*Bobi "Couldn't you move the upper tenon to the top of the rail?"*Yes, but then you aren't making an M&T, but a bridle joint. One of the weaknesses of a bridle joint is that it doesn't very well resist racking leading to the shoulder line opening. This is a table leg, and he probably has two rails coming into one leg at 90 deg. to each other, so I would consider a bridle joint (open at the top) inappropriate here. Too weak. Tenons rarely are placed within the top third of the rails width (in this example.) This is so that plenty of meat is left at the end of mortice towards the top of the leg, so that a plug, approximately the size of the mortice won't be forced to pop out during any racking that does go on. In door making similar calculations are made for the placement of the tenon in the width of the rail for both the top and the bottom rail.i "eliminate the center haunch?"*You could do this, but if we were making a door, this haunch fills the groove worked to carry the panel. Here, it adds a bit of glue strength and frictional resistance to separation. The value of haunches is debatable. They cause more work, but do add a bit of strength. Also a thing to bear in mind regarding cupping of the rail- which is another function of the haunches, is the position of the face of the rail in relationship to the face of the stile, or leg, in this case. If the rail is inset, then the value of the haunch is reduced because cupping won't be very apparent. If the rail and leg faces are flush, then any cupping will immediately show as a step. The haunch will help resist this.i "Could you also just cut one mortice of the proper size to hold both tennons and the center haunch?"*This is an option, and on smaller joints (less than about 160mm wide) this might be okay, but the reason for the fork is so that timber can be left between the mortices. This plug of timber helps to keep the mortices either side stable. There is less chance of the mortice opening up and forming a gap between the cheeks of the tenon, and the cheeks of the mortice. Additionally, in architectural doors the lockset or latch mechanism is often fitted between the forks of the middle rail, so every bit of meat helps.I think I answered all your questions. If not, get back. Sliante.
*Sgian, Thanks for clearing that up for me. I figured that there must be some reason or reasons to do it as you described but I couldn't come up with any. There is so much to learn about this craft and people like you and Kim Carleton Graves take the time to help us along the learning curve. Thanks for taking the time to provide the in-depth answers that you do. Bob
*Thank you Bob. I should get a few pennies for every sale generated from the publisher for every time I recommend this book, but unfortunately I don't; however, it really is a good one and knocks the socks off just about anything that has tried to emulate its depth and breadth of information that's been published in the last 30 years or so. Ancient now, but covers just about all important subjects you care to think of in the field of furniture making. It's not always an easy read requiring a bit of page flipping to refer to sketches and the like, but it's usually my first reference.i Ernest Joyce, The Technique of Furniture making. In the US I'm sure it's The Dictionary of Furniture Making. US ISBN 0 8069 6440 5. In the UK 0 7134 0217. *The online booksellers I know carry a paperback version with an (80's?) update undertaken by Alan Peters, one of the worlds most pre-eminent and most highly regarded furniture makers and designers. My copy is so old he (Joyce) doesn't even mention biscuits and biscuit jointers, but that's really unimportant for a biscuit joint is only a modified form of mortice and tenon anyway. Peters might mention them. I don't know. For Americans, it's also an education in British English, which is handy for reading those quaint, twee British woodworking magazines you all devour every month(?) ;-) Sliante.
*Bill, I'm rather assuming you've not been back to see what responses you got to your question. Was any of the information provided helpful or useful? I'd like to know. I'm just curious. Sliante.
*Sgian, yes, the info was useful. So you haunch the tenons to allow room for tenons that emerge from the other side of the leg.The table has already been made, I only began to wonder about the apron splitting after the fact. My legs were 2 1/4" square so the mortises were about 1 1/8" deep which I thought was plenty. I used two 2" sliding tenons spaced 1" from the top and bottom of the 7" apron with a 1" space in between. So I have effectively restricted movement over 5" of the apron. If I had it to do over (which I may very well) I would use one 5"tenon and glue it only in the middle to allow for seasonal movement. What do you think of that? In your diagram your apron is 12" and you glue the top half of your tenon, aren't you restricting your apron over 6", why don't you glue the middle only.Sorry I was so slow responding. Thanks for your interest.Bill Lindau
*wide sanderswhat is the best double post wide belt or drum sander for under 3k ?
*Essentially the haunch is there to prevent the rail cupping. It does add some strength too. As before, when the visible front faces of a leg/stile and rail are meant to be flush to one another, the haunch helps here by helping to keep the faces aligned. If the rail is set back from the front face of the leg the haunch is of less consequence obviously, but you do provide extra friction and glue surfaces, which is often useful. Where you glue is dependent upon the final use. For example, the sketch I provided came from a sliding table saw off cut extension table with draw leaves. The leaves are attached to bearers that slide in slots cut out of the top edge of the rail. Gluing the top forked tenon causes the bulk of the expansion and contraction to occur towards the bottom edge of the rail, thereby (hopefully) only minimally affecting the height of the slots cut for the leaf bearers.Take another example. An architeectural door. The bottom rail is often 250-300 mm wide (10-12") and traditionally these have a forked M&T. The bottom fork is glued so that the bottom edge of the door remains constant, and desirable for obvious reasons, and the top edge moves up and down on the panel. In the middle rail, which may well be wider, the choice is debatable, but I was taught to generally glue the top fork. This keeps the top edge constant and allows the bottom edge to move, and because the normal line of sight is downwards this movement on the panel is less obvious than seeing the result of movement on the upper panel.In your case, I would glue only the top fork (assuming you decided to go with this rather than a slip tenon. You could of course do two narrow slip tenons, which is essentially the same thing as the traditional M&T.) This should keep the table top at a constant height. It's probably not important if the bottom edge of the rail moves up and down a bit. For instance, one rail might expand more than the others, and if the middle of the tenon is glued this will result in a slight bowing of the top. It's a small point, but worth considering. Sliante
*now I can see your logic, where you glue has to do with which surface you want to remain constant. At what width piece of wood would you no longer need to make allowances for seasonal movement and just go ahead and glue all the tenon? I know the answer to that question has to do with the stability of the wood, so say the piece is dried cherry.Bill
*Good question. In your case, where you have the grain direction of the leg perpendicular to that of the rail, I would say as a rule of thumb anywhere where you don't need or use a forked tenon, glue the whole thing. If you recall my first post I indicated that I start to think of forking the tenon when the rail width reaches about 130-150mm. (5-6")I would say that this dictum is a guide, not God, if you get my drift.There are instances where the whole length of a forked tenon should be glued. The attached pictures illustrates an example, where the long stretcher joins the short stretcher between the legs. Tell us what you see, deduce the makers logic, the technique, and generally talk us through the joint. Sliante.View Image
*Sgian,OK, nobody's taken your challenge, so here's my guess. The joint you picture can have both tenons glued because the grain orientation of both components is horizontal. They'll both expand and contract together in the vertical plane. The grain orientation means that the strong glue joint will be on the horizontal, not vertical surfaces of each tenon.Why two tenons? Is this to avoid making too big a hole in the stretcher, or to increase glue area? Are the wedges mainly for appearance?And another question if I may, what type of joint would you use for a square rail, say 40mm sq. I was think possibly twin tenon - 5mm shoulder, 10mm tenon, 10mm gap, etc? I don't know the rule of thumb for twin tenons.Nice table, by the way.
*AE, It really wasn't a challenge. It was more to do with encouraging the learning process. Bill had a question, and I was just doing a bit to help.You were going in the right direction. There is no significant differential expansion and contraction between the two members so it can all be glued. The layout of the joint is further informed by the generally accepted premise that the strongest glue bond between any two mating surfaces in a joint is between long grain and long grain, and that glue strength between end grain and end grain is minimal, and that glue strength between end grain and long grain is at best poor.By forking the tenon, long grain to long grain gluing surfaces are maximised. There are four tenon cheeks that match with four mortice cheeks that glue well. These faces run parallel to the narrow top and bottom edges of each member. Additionally, you noted and asked about the wedges driven into the end of the tenon. They are definitely structural. The right and left cheeks of the mortices in the short stretcher are tapered wider towards you as you view the joint. There is poor gluing ability here for it is long grain to end grain. To compensate, the wedges force the cheeks of the tenon over to form what is effectively a dovetail which provides mechanical strength. The wedges are driven in to a sawed slot and are glued. (Long to long grain here again.)Your 40 mm square rail might well benefit from being a twin tenon as you describe. The form of the joint does depend a little on where the rail is positioned in a frame (or cabinet perhaps.) If it's at the corner, that's one set of rules. If it's in the middle of a rail or stile, it's an adaptation of that set of rules. It would also depend on how heavy the the piece is that your 40 mm square bit was going in to. If you'd care to get more specific, I'll be glad to try and advise. Generally, furniture makers are looking to maximise long grain to long grain gluing surfaces when forming M&T's. Another example. Dividers between drawers in high quality traditionally made cabinet usually have twin stub tenons into the carcase sides. Dividers typically might be 44 mm wide by 20 thick. The twin tenons are laid out a bit like the joint in the sketch marked 'barefaced forked tenon'. This comes from a chair, so to get the picture you'll need to imagine rotating the side rail and joint through ninety degrees so that the rail has the faces top and bottom as you would have in the cabinet I've just described. You'll also need to mentally shorten the length of the tenon, unless you want it to pierce the cabinet side, which is also a legitimate option in the right place. How come you talk metric anyway? It's like a breath of fresh air to me. Most people in this forum use those old before the wheel imperial inches and feet and fractions and yards and decimal inches and thousandths of an inch, and decimal feet, and other wacky stuff that is mostly just gobbledygook to me! ;-) Sliante.
*Sgian, Thanks for the comments etc. I live in Scotland (although I'm English) and we're all metric over here. Well, not quite. I'm of an age that still thinks in feet and inches for informal measurement. Once it comes down to hard measurements, I'm less error-prone using metric.The question on the 40mm rail was hypothetical at this stage - I'm too far down the line in this project to change from the single tenon I started using. Next time, though. Cutting the twin tenons accurately will set a nice little challenge for my skills as well!Tony
*That explains the metric. We have similar habits when it comes to measurements, so perhaps we are about the same age. Edinburgh is home to me although I currently live in the US. I worked for nearly ten years at the art college there in the furniture department and teaching/instructing students was a big chunk of my job. No doubt you recognised my gaelic handle for what it is. If not, just about any Scot will inform you. Help, if you ever need it, is usually generously given in this forum, but be discerning regarding advice on using table saws. US regulations, working practices, and habits regarding this machine are er,......shall we say just say, different! Removing the riving knife and crown guard from the table saw in the UK is illegal. You seldom see these guards used in the US, even in magazine shots. The 'dado' blade is also an illegal piece of kit in Europe, so you'll need to ignore all advice regarding this tool too. Sliante.
*Sgain,In reference to the book you recommended, I tried looking for it using the ISBN number and found it to be out of print. The ISBN number is for the hardcopy version. They do have a paperback version which is still in print. The ISBN number for that is: 0 8069 7142 8. For $20 it seems ridiculous not to buy it.Steve
*Thanks for the update. Sliante.
This forum post is now archived. Commenting has been disabled