I was reading in an old joinery book by Tage Frid (pub. in 1979, I believe) where Prof. Frid wrote that finger (or box) joints are stronger than through dovetails because the fingers offer much more glue surface. I’m curious as to what others feel about this, given the evidently strong attraction of dovetailing.
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
GJ,
Although I roughly sketched it out, for the same number of pins and tails, I think the through dovetail has the greater gluing area due to the greater angle. The half-blind dovetail will have an even greater gluing surface area since there are four surfaces for adherence as compared to 3 for through joints. From a strength perspective, dovetails offer a superior construction technique since the angles provide a greater "lock" on the joint thus helping to prevent the joint from loosening or breaking free.
On the other hand, finger joints are limitless in size of their application. Try cutting half-blind dovetails in a box with 3/16" sides. They also offer better proportions on small pieces such as jewelry boxes.
At least that is my take on it. For me, the bottom line is that I think dovetails just look classy.
Doug
If the finger joint has the same number of fingers as the dovetail joint has dovetails, then the two have nearly equal glue surface. However, I think Tage was presuming that the box joint would be cut with lots more fingers than the dovetail joint would have dovetails. This is typically how box joints are built.
Loved the man, but I disagree. An unglued, but well-cut dovetail will hold together. Not so with a finger joint. I don't believe the incrementally greater gluing surface of the finger joint would make it surpass a glued dovetail joint.
Can't agree with you more. The beauty of a dovetail is the mechanics behind the design.
J.P.
If Roy Underhill made a box using fingerjoints, he would probably need a quart of glue to keep it together, yet somehow his dovetail boxes can stay upright with no glue at all. I'm going with the dovetail.
My two cents is that since he includes info on how to make a jig for the table saw for making finger joints that with smaller more numerous pins/fingers, there would be more glue area. Whether that makes it a better joint than a dove tail, I don't know.
Datachanel
Doing things the hard way
I like Doug's answer (post #2).
Dovetails have the obvious mechanical advantage, but I prefer finger joints on thinner stock.
Common sense would seem to indicate that dovetails are stronger, but as a practical matter, if your finger joints are machined with a nice, close fit, I doubt if they are much weaker than a dovetail.
On the other hand, if you're making a box subject to a lot of abuse and/or racking (say, a tool carrier), then dovetails would be my choice.
gj13,
It would seem to me Tage's comments would be accurate if you have more fingers than dovetails and you buy into the glue joint is stronger than the wood itself. They join together short pieces with a kinda finger joint and glue and I'm sure that is as strong as long piece of wood without a joint.
That said, however, personally, designing a piece where you have to rely on the glue or any fastener for strength is, in my opinion, to be avoided.
Finger joints are not box joints .
All I ever learned of life was while at my mother's knee. (And other low joints).
Steinmetz
How do they differ?
UNK,
Finger Joints look like long thin saw teeth (Used to join two in-line strips to obtain longer lengths. Also, to glue harder wearing wooden soles on planes (Germany)
Or, in the case of Home Depot, to make crappy molding from short 'Cut offs' To sell to suckers.(Paint grade) And to charge more for the good stuff (Stain grade)
Router bits are available to cut finger joints. Stein.
Edited 6/15/2004 2:38 am ET by steinmetz
Edited 6/15/2004 2:39 am ET by steinmetz
Edited 6/15/2004 2:43 am ET by steinmetz
UD
Ditto Stein.. I built a work-bench top with short scraps and joined the pieces with finger joints produced by machine. The finger joints are repetiously side by side and provide a large amount of glue surface. If I had used DT's cut by hand and got as glue surface as the finger joints, I would probably still be cutting them.
For joints that show and are spaced random, I prefer DT cut by hand. If it is merely a matter of joining for utilitarian purposes, I would use the finger joints for time saved and to get the maximum amount of glue area.
BTW, the work-bench top moved 1/8" over the course of 4 seasons here in GA. Corrected in 45 minutes with a fore, #4 and LA smoother plane. The finger joints have their place and in other situations are out of place.
sarge..jtProud member of the : "I Rocked With ToolDoc Club" .... :>)
I approached the question as which one makes the best corner joint. I still think the dovetail beats the finger joint for that application.
I'd assume he was referring to box joints with lots of fingers 1/8 - 1/4" wide for a corner joints. The tapered finger joints used to make molding out of kindling don't work for corners. There also used for making "woody" parts. The comparison is therefore apples to oranges. One is a machine joint the other can be done by hand or by machine. This is further complicated by hardwood vs. sheetgoods and then adhesives. For utility work 1/4" - 3/8" box joints in 1/2" BB Ply with TB glue and pegs (terriaki skewers actually) through the end fingers will make a very attractive, durable box, very quickly, whether it be for tools or a kitchen drawer. For show work in hardwood, hand cut dovetails are tough to beat and sure look good. I like to use the joint that's most appropriate to the work and material. I think a well machined "machine" joint looks and works better than a machine joint trying to imitate a hand cut joint. That's why I don't own a DT jig.John O'Connell - JKO Handcrafted Woodworking
The more things change ...
We trained hard, but it seemed that every time we were beginning to form up into teams, we would be reorganized. I was to learn later in life that we tend to meet any new situation by reorganizing; and a wonderful method it can be for creating the illusion of progress while producing confusion, inefficiency, and demoralization.
Petronious Arbiter, 210 BC
Let me preface what I say here with a little disclaimer. About thirty years ago, I took some courses in strength of materials and in wood engineering. I haven't done a thing with it since, and what I say below is dredged from memory. If I make mistakes, particularly in terminology, I apologize.
An engineer would want you to be a little more specific on what you mean by strength. We can talk about the load the joint will bear at complete disintegration (ultimate strength), or we can talk about the load at which the first signs of cracking occur (yield strength). I assume we are talking about yield strength.
I am not aware of a published study comparing these two joints, but this isn't my field. I do know that Fine Woodworking published an article in issue #148 by Bruce Gray that compared the strength of M&T, floating tenon, biscuit, and various dovetail joints for corners – no finger joints.
Bruce built joints and broke them in a machine that measured load vs. deformation. He then analyzed the broken joints and compared their modes of failure. He built and destroyed four types of dovetail joints (wide pins, narrow pins, low angle, and high angle). He made and broke six samples of each.
In his discussion, he noted that all the dovetail joints failed in the same way. First, the tails split and slipped, followed by failure at or near the glue lines, and finally the pins may have been broken as the joint completely failed. Interestingly, he found that the initial failure occurred the same in tight and in loosely constructed dovetail joints.
It struck me that the initial failure would have occurred with or without the glue. It was a physical breaking of the wood, along the grain in the short grain created when the tails were formed. The glue failed after the wood.
If a finger joint were formed in straight grain wood, and parallel to the grain direction, there would be no short grain to fail. In that case, failure would be expected to occur in the glue. Probably the finger joint, even with the same finger spacing as the dovetail spacing, would be stronger as long as the glue remains at full strength.
I have always heard that modern glues, particularly like PVA which has some give, are far superior to traditional glues (mainly hide glue). As hide glue ages, I understand it tends to crystalize and become brittle. Glue joints break as the wood moves in response to changes in humidity. Therefore, hide glue joints tend to loose strength over time. I don't know how PVA joints behave over time.
I don't have data to support it, but my opinion is that the finger joint will have greater strength, even with the same glue area if we define failure as initial cracking. Again, that is as long as the glue retains its strength. The structural advantage of dovetail joints is that they retain some strength even without the glue. Dovetails were used traditionally because the glues were not reliable.
So far as joint selection goes, I think the aesthetics and ease of construction probably rule. In most applications, either joint will be many times as strong as is needed.
Thanks for the detailed response. I'm sure you've illuminated this subject.
I agree with....and would add you have to look at what kind of forces are acting on the joint: tension, compression, racking, shear, etc. Each joint will perform better in some, worse in others, than any other joint. A dovetail for example is very good resisting tension forces....the shape of the dovetail contributes to that, whereas with a box joint (as noted above, box joints and fingerjoints are not the same thing), it's all about glue area. In compression, I don't see how one could be stronger than the other.
BTW, I agree with the earlier posts saying that with many smaller box joints 'fingers', lots of glue surface, and modern glues, a box joint is incredibly strong....remember in a well cut joint with properly mating surfaces, most of the wood should fail before the actual glue joint.
It's all about weighing your otions in the case of a specific engineering challenge.....like which joint to use to make a piece strong enough, in the face of the forces that will be expected to work on it.cabinetmaker/college woodworking instructor. Cape Breton, N.S
I completely agree. The forces that are to be resisted should be a principal consideration. In the case of drawers or boxes, though, those forces are relatively small compared to the strength of the joints.
Several contributors have stated that there is a distinction between box and finger joints. As I understand the terms, what are being called "finger joints" are cut with a router or shaper bit and designed to join short pieces of wood end to end. The joint elements usually have tapered sides. What are being called "box joints" are used to join two pieces in a corner. The elements of "box joints" have parallel sides. Perhaps you can explain the distinction between the two joints.
I do not see universal agreement in the literature on using different terms. I think it is relatively recent, probably arising from shaper and router bit marketing. In his 1979 book, Tage Frid Teaches Woodworking: Book 1 - Joinery, Frid indicates that the terms are interchangeable, but he seems to prefer "finger joint" to refer to the corner joint. More recently, in The Complete Illustrated Guide to Joinery (2002), Gary Rogowski also refers to the corner joints as "finger joints". Lon Schleining calls the same joint a "box joint" in his article in FWW #148 (2001).
In my opinion, especially in industrial woodworking, 'fingerjoint' means one thing....the end to end joint with tapering fingers. There are different types (structural/non-structural, European pattern/N.American pattern etc.), but it is a very important industrial process, and personally I reserve the term fingerjoint for that. The router and shaper cutters that do this joint, especially the router ones, aren't even in the same ball park as the joint that is made on industrial machinery.......the fingerjointing lines today are incredibly fast and precise, and the joint is very, very strong. The router bits do an okay job, we do a test to show how strong it is.....glue up a bunch of short blocks, and try to break it.I'm seeing more and more production furniture companies using a finger joint in exposed surfaces now, as a decorative element.....NOW, nobody start howling about 'traditional quality' and all that, I'm not making a value judgement here, just pointing out a trend that relates to fingerjointing. Millworkers making doors and especially windows and related, have been using fingerjoints a lot, for a long time, and there I will make a value judgement.....it does improve quality. If you want to see what an industrial fingerjointing head looks like, try http://www.wkwinc.com the price list on there is for EACH cutter....there should be some pictures where you can see how many teeth are in a cutterhead. The tolerances are pretty amazing.
The interlocking corner joint, personally, I call a box joint.....but, in small shops and a lot of books aimed at the small shop or hobbyist woodworker, I agree, it's often called a fingerjoint. I just did a quick check; all my English books call it that. My newer cabinetmaking text calls it a 'box joint, sometimes called a finger-lap joint'. If someone used the term finger joint to me, in the context of a corner joint, I'd know what they meant. But, like I say, I prefer to reserve the term for the end-to-end joint.cabinetmaker/college woodworking instructor. Cape Breton, N.S
I tee-totally agree. A true finger-joint is an elongated triangle and used to end join. If you attempted the joint on a 90* degree corner, it wouldn't mesh with full contact. It was not designed for corners.
A box-joint is a series of rectangular cuts similar to mortice and tenon, but open on one end. It is considered a locking joint, but I prefer to call it a floating joint.
The dove-tail is a true inter-locking joint. If you could get as many thin dove-tails in the same amount of space as finger and box joints, I personally believe it would be stronger bond than the finger or box. Consequently, if you put 4 dove-tails on a corner of a box compared to 4 box joints; the dove-tail rules as it provides approximately the same amount of glue space and also truly inter-locks from a structural point.
The finger joint is ruled out for corners. Anyone that doesn't believe that should cut some and try to corner join them. They mesh wonderfully end to end, but take a completely different confirguration in a 90 degree corner and won't mesh at all.
In the auto world you have "rag joints or coupling joints", same thing. You have "center links and drag bars", same thing. But in WW world, even though the term is used out of context the finger joint and the box joint are two completely different animals regardless of how respected a source used it that way. That's my story and I'm sticking with it.
Off to Wally World to loosen up what used to be floating joints but want to become inter-locking joints if I get lazy and allow them too. ha.. ha...
Regards from the deep south ...
sarge..jt Proud member of the : "I Rocked With ToolDoc Club" .... :>)
Regards to you, Sarge, from the Far East (Coast)cabinetmaker/college woodworking instructor. Cape Breton, N.S
>> ... two completely different animals regardless of how respected a source used it that way.
Well if respected sources don't define correct usage, who does? And what are we going to do when you're not around anymore?
I'll bet a dollar that "finger joint" was in use long before the tapered, end to end finger joint was even invented, and that it was understood to mean a corner joint, what you are insisting can only be called a box joint.
When the end to end finger joint was invented, it was obvious that it would also be called a finger joint, but this created no conflict with the already existing usage of finger joint because it's almost always clear from the context which one is intended.
I don't have a problem with box joint and finger joint both being used to denote the corner joint. It makes a useful distinction between the case where the fingers are as wide or wider than the thickness of the stock, which I would call a box joint, and the case where the fingers are narrower than the thickness of the stock, which I would call a finger joint.
Neither joint is that old.....my 'box joint' is in Ellis'es 'Modern Practical Joinery' (1902..... mine is a reprint of the 1908 edition) as a 'corner locking joint.....lately introduced from America'. (Goes on to say " It is used instead of dovetailing, as it is much stronger than the latter, though not so neat in appearance").
The finger joint (then called a lock-joint, as in lock-joint grooving head), was patented in 1929.
Not making a point, just adding info as it turns up.
cabinetmaker/college woodworking instructor. Cape Breton, N.S
Edited 6/18/2004 2:56 pm ET by Adrian
I bought the Frid Joinery book back in 80'. Pg. 90 he explains how to cut a finger joint or box joint as he stated. He uses the term as one. I respect Frid immensely, but with the info Adrian introduced could he have possibly made a mistake?
I also respect Rogowski also, but he wasn't around when the finger joint got started. Neither were you or I. I believe the two terms were merged as the finger is an un-common joint to see outside of industrial.
Do you suppose that even respected sources as these could have made a mistake in the usage of the terms and there is a remote possibility that they are truly two different joints? In my opinion that did happen. It's just my opinion and I cannot support it with hard-core evidence as I wasn't there when the "same" but "different" joints evolved.
I saw my first box joint around the 50's and it was called a box joint. I saw my first finger joint around the early 70's and it was called a finger joint. Whatever these two different joints are, they are indeed that. Different. I can physically see the vast difference.
What you referred to about usage actually happened with the bridle joint. I bet you have seen lot of open mortices, but you may never have seen a bridle joint. The open mortice is a bridle joint as it was originally called. They look and are cut exactly the same because they are. But what I know as a box joint for 40+ years is not the same as what I know as a finger joint for about 33 years.
Again, this is my opinion an I feel I'm en-titled to it. You have yours and are en-titled to it. As far as "what we (or you) are going to do after I'm not around anymore", I could really care less. But what you're not going to do while I am around is deny my right to have my own strong opinion that there is a distinct difference between a box and finger joint. Proud member of the : "I Rocked With ToolDoc Club" .... :>)
>> ... my right to have my own strong opinion that there is a distinct difference ...
I cherish your right to hold that opinion and to share it with us. My only dispute is with the notion that it is incorrect to call a box joint a finger joint because the term is already being used for something else. There are many, many words in English that mean two - or more - different things, and somehow we manage to struggle along.
The only standard for correctness is popular usage, and if the population that deals with joinery decides to use finger joint to refer to two different joints, then they are both finger joints. End of discussion. What do you gain by insisting that only one of the two joints is _really_ a finger joint? It won't help you understand what others are saying, or help you make yourself understood. If the function of language is communication, your position that finger joint can only mean one thing strikes me as a dead loss.
I should add that my question, what will we do after you're gone, was meant to tease you for making yourself the arbiter of the correct usage of finger joint when you disagree with a respected source. I certainly did not mean that I wish to hasten the day or that I bear you any ill will.
I have given this some thought after Adrian's post as to when the joints were created and what they were originally called. My conclussion may be closer to the truth than either of our views as we currently see them.
My theory this morning (and it changed over-nite after carefel thought) is that the inventor of the box-joint intended for it to be called a box joint. The inventor of the lock joint intended for it to be called a lock joint. Somewhere down the road the word finger joint sprouted with one of the two and it caught on at some point with the other.
Finger joint seems to be slang for both joints. But if you refer to a finger joint, are you referring to a rectangular cut open ended joint or a triangular cut open ended joint? Both have different characteristics and aren't inter-changable.
If you inter-lock your fingers on your hand and turn it over to view the back side, you note that the fingers are spread at an angle. The web at the base is not a point and as a triangle and it's not square as a rectangle. Still both resemble to some degree what is actually a box joint and a lock joint. Somebody, somewhere hung the term finger joint on one or the other. At some point the other joint evolved with the same slang and it caught on also.
My conclusion is we have box joints and lock joints and the true finger joint with rounded tips has not been invented. Will all the books that have been printed recently pull them back to add a foot-note * finger-joint triangle cut or * finger-joint rectangle cut. Nah!! Could the guru's be wrong (with no intention to decieve but hanging on to the way they were oriented to use the term in their early WW beginnings)? Yep, I think that could happen and has!
Sam Maloof told me once that "nobody knows all there is to know about WW, nobody". I believe he was correct and that even guru's can miss something in the translaion. How many guru's you know that have spent intense periods of study on the history of all faucets of the craft backed by bare facts? Could anything be mis-interupted as it was handed down from craftman to craftsman? I think so as guru's put their pants on the same as everyone else.
My conclusion it that from this point forward I will call a box joint a box joint and a lock joint a lock joint. That will save time asking the question rectangular or triangular that could be better spent cutting the grass or making sure the barbeque on the grill doesn't burn. There is no confusion using the terms in their original form as they were intended.
That leaves the door open for some enterprising young man or woman to invent and patent the true finger joint. In the meantime I will continue to cut box and lock joints and leave finger joints on the capable hands of those that are experienced at box and lock joints.
Final Answer...
Back to the yard and tending the grill, keeping a close eye on the ice chest to make sure the beer doesn't run low. A few finger joints around those brown bottles would be a welcome addition to my already expert ability at emptying the bottles. ha.. ha... ha.. ha..ha...
Ya'll have a good day...Proud member of the : "I Rocked With ToolDoc Club" .... :>)
Sarge,
I thought about it too last night..and came to the opposite conclusion. Listen to my logic....back when this discipline (woodworking, cabinet making, jointing, whatever) was being develop and taught to the apprentice...very few people could read words...and using a word or a name for a thing or place that people could easily relate too was the norm. Therefore, dovetail joint looks like a dovetail or bridle joint looks like... (we'll leave that one alone)...anyhow, except for dado (which is probably french) most names made sense.
In that context then, it would seem to make sense that finger joint preceeded box joint. To me, for clear understanding, a finger joint, box joint or dovetail is different and should be differentiated. If we were going to modernize the terminology (as if someone is asking us too...lol) we should probably call all of these finger joints with a positive, negative or zero slope to distinguish them.
I suspect the box joint came into being with the machine age...as a distinguishing characteristic to sell there better boxes at lower than hand made prices. All of this is guess on my part.
I agree they probably got modified by someone because of the look they present. Even though one is an end-to-end and the other is a corner joint. It's likely as one is used commonly in shop situations and the other in industrial that the person or persons responsible for the modification never dreamed that it could create confusion in idenity.
The good news is it gave us something to argue about other than Bush-Kerry, conservative-liberal, left-right tilt or which TS blade can do it all with one tooth count. The bad news is someone on an entry level that was reading this post would probably be confused as to what the h*ll we're talking about to start with. Most have probably never seen a (what I know as a true finger, which turns out is not a true finger if you go back to roots) as they have never seen moulding, etc. made in a mill using this common machined joint.
Frid's section of his joinery book shows how to make the "finger-joint" or box joint (and if you turn to pg. 90 that's exactly how it was wrote). I almost ask why he called a box joint a finger joint when I saw him in seminar in the latter 70's. But I had heard it called that regularly by then and did not feel it relevant since I knew what both joints being called fingers were and there different uses.
I suppose it's ir-relevant now as I doubt the term will be dropped and the original coinage given the two joints be again adapted. I can see no great pilgrimage on the horizon to correct the usage.
My final solution is to call a box-joint a box-joint. I will call a lock-joint a lock joint. Others can chose their own poison. Perhaps they should be numbered like planes. Hey BG, give me a couple of # 43's. Over easy and a side of hash-browns smothered in glue.
Or perhaps I got better things to do on fathers day.
ha.. ha...
sarge..jtProud member of the : "I Rocked With ToolDoc Club" .... :>)
>> ... and the true finger joint with rounded tips has not been invented.
I've seen them. Unfortunately I can't remember where, but I have seen them. Fat stubby fingers with rounded ends.
I can't say that I have, even though I see no benefit in rounding the tips really. I have cut a ton of lock joints as I get a lot of donated scraps and turn short pieces into longer pieces for donated furniture for shelters, etc.
When I say I have cut a ton of lock joints (what until yesterday I called finger joints.. ha.. ha...), let's say that I have cut a ton on a machine. I have never cut a lock joint by hand. These things can be tiny to large and are spaced side by side repetiously. They would be a "pain" to cut by hand. The machine does it accurately and without hesitation in seconds. The machine cost a ton an that's why you won't see them in other than an industrial situation that can warrant the cost.
As for a real need for a "finger-joint", I can see none. Rounding the tip would be the same truck with a different grille. Besides, the box-joint and lock-joint seem to already have all the ground covered in that area. ha.. ha...
sarge..jtProud member of the : "I Rocked With ToolDoc Club" .... :>)
I'm going to stick with box joint, and finger joint....introducing, or re-introducing the term lock joint would just add confusion I think, and fingerjointing is very well established. Plus there are other lock joints that could get mixed up in that....drawer-lock joints, and corner-lock mitre joints/lock-mitres etc. I've never had occcasion to use a box joint anyway, apart from trying it out when I was starting out, and very rarely run across them.....there are probably some big box makers somewhere, but I haven't crossed paths with them.
I do however reserve the right to berate anyone I catch in the act of using terminology counter to the rules I have just laid down! You've all been warned!
Now, on to the next terminology debate.....rebate (correct), vs. the TOTALLY unacceptable term rabbet....my dukes are up....lets get started....cabinetmaker/college woodworking instructor. Cape Breton, N.S
Talk about beating a dead horse......
"Talk about beating a dead horse......"
Since everyone seems to be insisting on the proper terminology in this thread I beleive the original saying was "Talk about assaulting an expired equine." Please use the proper phrase in the future:)Tom
Douglasville, GA
I believe "flaggelating an expired equine" would be more descriptive and therefore preferable, but I'm going to let this one pass. Just this once.
cabinetmaker/college woodworking instructor. Cape Breton, N.S
I'm pretty sure we're done. You can move back over to Cafe now for your next dose of web knowledge.
Talk about beating a dead horse......
ha.. ha... ha..ha..ha..
Proud member of the : "I Rocked With ToolDoc Club" .... :>)
Frid is right.
Since far more finger joints running continuously fit in a given area than dovetails could fit in that area, the finger joints offer more glue surface and are stronger.
This forum post is now archived. Commenting has been disabled