Windsor Chair Resources

Windsor chairmaking tips-tools


 || Tip Index ||

How to make the perfect tapered tenons.

One of the issues that always come up when talking with other chairmakers is how to get the perfect tapered tenon that is tight. I have played around with this for years, through trial and error. I have now found a way which makes it easy to do and is alway accurate. I use the Fred Emhof reamer. But when I am turning the leg and making the tenon (I rive green maple and then let it dry before turning the leg but I have turned the leg green before) I make sure the tapered tenon is facing the tailstock and is very close to the end of lathe. I then purchased a Ashem Crafts 5/8" rounder but I realized it has a natural taper on it. It turns out that this angle and length matches Fred's reamer perfectly. All I do is turn the tenon about 1" in diameter, start a natural taper with my lathe tool, remove it from the lathe and let the rounder do the rest. Everytime it makes a perfect tenon that matches my reamed leg hole. I turn the tenon about half way through the leg so I can make any adjustments needed. Once the taper is done, I then turn the rest of the leg. Perfect every time. Just thought I would share this tip. DG 3/14/03


QUESTION:  Searching the site I came across the spindle and socket illustration in the STRUCTURE section. It features James Rendi's description of the round bottom socket and rounded spindle tenon as the basis for a strong spindle to seat platform joint. I would like to get some feedback from others on this position. I advocate the joint is strengthened by driving a faceted hardwood dowel into a softwood mortise(round hole). The end grain of a rounded spindle seems a bit "far fetched" to me as the source of enduring strength. Any comments?? DO 11/20/02

REPLY: I agree with DO, the roundness of the socket and end of the spindle may present less end grain in the glue joint, but I doubt if it adds any appreciable strength. The facets of an octagonal cross section hardwood spindle compressing the softer seat wood would provide much more holding power.

In Kassay's book, the first chair with measured drawings is a comb back side chair. The spindles are through tenoned and wedged on the bottom of that chair. Now *there* is a strong joint. There is no reason you can't do it for every chair. Another thing I have done is to pin the spindle/seat joint with a treenail similar to the way the top of the spindles are pinned to the crest rail.

Has anyone had experience with the spindle/seat joint in a chair they have made coming loose? If so, what would they have done differently?  GB 11/20/02

REPLY:  I agree with the faceted tenons. My old wood shop teacher always said gluing with out clamping is like not gluing at all. I think to get the pressure needed for a good glue joint, the tenon would need to match the hole perfectly. The chances of perfect fit between a bored hole and a hand carved tenon are slim, where as a faceted tenon will bite into the seat. BG2  11/20/02

REPLY:   I think there is a bit of reinventing the wheel going on. The faceted hardwood spindle into the softwood seat is probably one of the strongest joints on the chair. The rounded bottom was only the result of spoon bits and getting the spindle all the way in. I've never seen one of these joints fail. The only spindle joints I've ever seen fail were the short ones next to a failed armpost joint and always both were a straight tenon joint in a hardwood seat. You could through tenon and wedge them but why? They don't fail anyway. Pinning will only help if the glue fails and then it's harder to remove and reglue it. JT 11/20/02

REPLY:  I agree with JT. Also, I think that one of the reasons these joints don't fail is the mechanical force leveraged to the sides of the tenon in the seat mortise by imperfectly "straight" spindles stretched, bent, and fixed in the bow, arm, or comb. And how about gravity? I've never seen one float out of the hole in my shop. BG 11/21/02

Tapered vs. Round Tenon on Bowback

QUESTION: When attaching a bow back to the seat of a bow back sidechair, which is the perfered method to use. In John Kassay's excellent book it shows to use a parallel round tennon, while in Drew Langsner's book, also an excellent reference, he shows using a tapered round tenon.

REPLY:  I use a 5/8"Veritas tenon cutter on the ends of the bow. Straight tenon,no taper. I use Ash. Tenon protrudes through the bottom of the seat, with a scribed shoulder at the entry point. This joint is then glued and wedged like any other through joint on a Windsor. I learned this method from Michael Dunbar, and carefully done really well for me. Also....I bead the bow before bending. MWT.  3/16/02

REPLY: I use the round tenon with scribed shoulders on my chairs. I also wedge it on the underside of the seat though that may be overkill. You can use guide block or other aids to help you drill a more accurate hole the first time. The only advantage of the tapered socket in this use is that you can adjust it. Since the pressure is back on a bow rather than down on the legs I don't think the strength advantages of the taper are useful or needed in the bow. JT 3/16/02

Tenon Joint on Philadelphia Windsor

I am just about to make my first Philadelphia chair, and have been wondering about what sort of tenon to use (tapered or straight) to fit the arm post to the seat. Looking at drawing in Kassay's book it shows tenons with very extreme angles (such as 1 11/16 in. reducing to ¾ in. over a 1 ¾ in. seat), but as it is hard to believe that such extreme angles would grip I wondered if anyone had direct experience of this and could let me know. JM 11/1/99

Reply: I've made a number of large Phil. fanbacks and the armposts are usually perpendicular to the seat so with a wedged straight tenon and a large flange they've proven to be a very secure joint. The armpost on the chair you are interested in is tilted back 4 degrees but I'd still be tempted to use a straight tenon, in turning I'd make a slightly thicker shoulder and scribe it to fit. I can't say if this is the case here, but unless a chair is taken apart figuring out some joinery is just a guess. JT 11/3/99


QUESTION:  In "Make a Windsor Chair with Michael Dunbar", Mike refers to egg shaped tenons. Are egg shaped tenons necessary for strong joints? Do very many chairmakers use them? Thanks, CRR 1/13/03

REPLY: Lately, in his classes at the Windsor Institute (based on my attendance at the sack back course in Oct. 2001), Mike does a little recanting on this topic.

For the undercarriage, he now teaches that the mechanical means of assuring tight joints is to make the stretchers 1/4" longer than the actual distance required. This is done by measuring the distance between the legs after you've fitted them, precisely contolling the hole depths, and cutting the stretcher lengths accordingly.

For the spindles, the mechanical means that Mike teaches to keep the joints tight is to make them oversized, and faceted, so that they actually bite into (compress) the softer seat when you fit them.

The egg shape theme may have been derived from the examination of joints of old chairs taken apart for repairs. The holes had round bottoms from the use of the chairmaker's drill bits, and the tenons were formed to match. Differential shrinkage would transform round cross sections to egg shape.

I believe there *is* some mechanical holding power (other than the glue) of the socket and tenon joint. I am basing this on my ladderback chair experience. In making these chairs, I kiln dry the rungs to below "equilibrium" moisture content), and insert them into posts that (ideally) are riven, worked into shape, drilled and assembled the same day they are split from the log. The rungs take on moisture and swell, the posts' holes actually will shrink in the tangential direction. If the hole/rung fit is tight on assembly, it gets tighter when the wood reaches an equilibrium moisture content. Over the years, the seasonal shrinking/swelling cycles will compress fibers and loosen the grip somewhat. GB 1/15/03

REPLY: The egg shaped tenon was an attempt to get some benefit out of wet/dry joinery idea. It didn't work too well, the downside of protruding tenons in the seat and some joints that still came loose proved it wasn't a good idea for windsors. The egg shaped tenons were also a way to allow you to trim the tenons to fit. So if you aren'[t using a tenon cutter they may help in that regard. It's hard to recommend the Dunbar book anymore. There is some good info, but many of the methods haven't any validity anymore. JT 1/15/03

REPLY: Dunbar's book says to use egg shaped tenons in two locations, the leg stretchers and on the bottom of the four short spindles just behind the arm posts. I don't see any reason to use an egg shaped tenon on the short spindles as he uses faceted tenons on the rest of the spindles. The seat should be dry, thus it won't be shrinking around the tenon. I think the faceted tenon biting into the seat will make for a stronger joint than the one he describes.

If your stretcher mortises are bored wet and your tenons dry, the wood will shrink around the shaped tenon making for a strong joint. However, I question whether this is really necessary since the legs should be in constant tension, thus holding the joint together. I think Mike probably rounds the end of his tenons to match the round bottom of his spoon bit holes. Instead of fooling around trying to make an egg shaped tenon, I just use narrow gouge on the stretcher tenons as a last step when I size it on the lathe and cut off the end with a saw, I then let the shoulder of the stretcher tenon take the presure pushing the legs out rather than worring about hole depth and tenon length. BG2 1/15/03

REPLY: The choice of a leg tennon is dictated by the chair you are reproducing, if you do reproductions. Early Philadelphia Comb-backs used mainly a cylindrical tennon. Many of the New York C-armchairs used a tapered tennon, Walter MacBride's chairs is one example. Rhode Island/ Conn. border area used stub-tennons, that is a tennon that does not pierce the seat blank. The smaller diameter of a tapered-tennon showing on top of the seat is aesthetically more pleasing than a 1" diameter (as might be used with a cylindrical tennon). The larger diameter of the same tapered-tennon underneath the seat offers more strength due to its size. The tapered-tennon tightens in its respective mortise each time a sitter sits. I would guess most chairmakers do various types of chair legs styles and tennon types. PW 1/16/03


QUESTION: All that being said about tenon joints, how many of you are still using the wet/dry tenon joint on the stretcher/leg joint. There is obviously a timing issue on assembly of these joints. Is anyone assembling dry/dry with glue and relying only on the tension created with 1/4 inch longer tenon for strength and longivity? BG2 1/16/03

REPLY:  For Windsor chairs I use dry wood for both the legs and stretchers, relying on an extra 1/4" stretcher length to keep the joints tight. (BTW, this puts the stretchers in *compression*, not tension.) If you use green wood for the legs, you'll end up with your tapered leg/seat joints protruding a considerable amount once they dry and shrink. GB 1/16/03

REPLY:  I leg up under compression. This other method is a waste of time, (in my opinion), and I believe will disappear completely in time. Mike Dunbar credits himself as starting this practice, the selective drying of tenons. And he has recanted this some time ago.It's common knowlege that early chairmakers were using dry parts while assembling their chairs, which has been mentioned here many times. MWT 1/16/03

REPLY: As far as I know the wet dry method for Windsor chairs was started by Mike Dunbar. He has since recanted and says he will probably spend time in chairmakers purgatory for it. It works in post and rung chairs because the wet piece is all wet end to end and the same with the dry piece. In a windsor one end is wet and one is dry, eventually the moisture in the piece equalizes. Which means the dry end will gain some moisture and expand. Eventually though it all drys out and shrinks. This means protruding tenons and loosening joints. In a post and rung chair when all pieces gain or loose moisture equally there always is an advantage or at least no disadvantage. If you are making chairs using this method you will make better, longer lasting chairs by using completely dry components. You can find many old chair labels where completely dry components was listed as a feature. This was with good reason. Using a small foam kiln as previously mentioned works great, you get the ease of turning green wood and 36-48 hours later, put them back on the lathe touch up and sand and you have nice dry pieces. JT 1/17/03

REPLY: While JT's comments seem to make sense, I use the wet/dry method all the time and haven't had a leg loosen up on any of the several hundred chairs I have out there. I use hard maple legs and soft hardwood seats. Maybe the reason is that I don't use the hard maple green right off the lathe. I put the legs in grocery sacks to dry as long as I can - from 5 days to a month if I have the luxury (without the sack to slow down moisture loss, the hard maple legs will dry too quickly and develop drying cracks), and finally I use the hot sand method - 130 degrees for a couple of hours before insertion. I've tried the kiln method - it dried the hard maple too quickly and caused drying cracks. (It's worth having the kiln though - I use if for spindles, bows, combs ...) I do think that whatever method you use, unless the part is all dry, there will be some degree of moisture equalization. Even using the kiln method on other woods I would think that this may happen because the thinner tenon end should dry more completely than the bulkier inner portions along the length of the leg. BG 1/20/03

REPLY to BG:, Can't argue with what works. My guess would be you are getting the legs drier than you think. Actual legs loosening though is less common than the tenons protruding above the seat and that can take a few years. I typically use my kiln with the door open a inch or two for the first 18-24 hours and I seldom have problems with legs cracking. Which may be a result of too much heat too fast. Often when they do crack it's something not 100% with the wood itself. JT 1/20/03

FOLLOW-UP: GB, I turn my legs green then wrap the middle portion in tin foil so the tapered tenon will dry. I've even put them in the oven at about 175 degrees to dry the taper. When you pull the foil off after it comes out of the oven, the middle is still wet. Thanks for your input BG2 1/16/03

FOLLOW-UP:  I've heard of another method used to attempt to achieve what you are describing with tinfoil. A coffee can of sand on the woodstove with the tapered ends of the legs stuck into it. I'm not sure about the success or failure of this method, any one out there tried this? GB 1/16/03

REPLY: Stretcher to leg techniques/leg to seat techniques. Over the years I have moved from the green legs to completely dry legs and stretchers. I don't operate on the 1/4" rule as carefully as some others(not being schooled) but I do keep my stretchers under compression. My reason for avoiding the uniform length increase: I stretch the legs for and aft a bit more on seats with the grain running front to back(my continuous arm). On my sacks, I tend to back off the "feel of the stretch" to avoid a split running with the grain. As I pull my legs to my "low-tech" feel, I put a stick from the cove of the front leg to the cove of the back leg(baluster) and hold the compression. I now use the bevel gage on the stick and the leg to pick up the "compression angle" on the front and back legs. After I bore the holes, I reset the legs, stretch again and use a small telescope tool(5/8" pvc tube with a dowel inside)to measure the exact stretcher length from the bottom of the leg holes. Its really close to what others are doing I just get a warm fuzzy when I feel the amount of pressure I'm using. Truthfully, I think the tenon to leg joint is the achilles heel of the windsor- not the leg to seat joint. I've agonized with going back to hide glue in the joints to allow for some future repair. The wood glue I'm using just can not be repaired without removing the old before the new. Oh decisions .... This has been wonderful to revisit the subject and gather ideas. Thanks.DO 1/21/03

© Copyright Windsor Chair Resources, All Rights Reserved