Drawboring Mortise and Tenon Joints

[Wayne Cannon]

In reading Christopher Schwartz's new workbench book, he is constantly arguing the merits of drawboring for mortise and tenon joints.

A recent woodworking magazine joint test, however, highlighted for me the fact that drawbored mortise and tenon joints all failed at the bore/pin and all failed before an equivalent non-drawbored joint. This also correlates with two old pieces of furniture with failed joints that I repaired -- either the mortise split at the hole, or the portion of the tenon behind the bore separated and released the joint. Granted, in these "old" cases, the glue failed, in which case a non-drawbored joint would also come apart.

Granted, drawboring has the immediate benefit of self-clamping and of drawing the joint tight initially. Does anyone else wonder if the cost in weakening the joint might not outweigh the short-term initial benefits of drawboring, or have any counter arguments?

[Dan Barr]

why perforate something that needs to be contiguous?

just asking for a failure.

v/r

dan

[Mike Cutler]

Wayne

I think we have to look at the purpose and history of drawboring.

It's really a construction technique,and can be seen in effect in barns and houses all throughout New England. Some of these structures and joints are 100's of years old and are still intact.

Properly executed in furniture, it shouldn't "create" a weak area. I would also argue that the test was exerting force in a manner inconsistent with normal use to the point of failure. A misconception of a drawborne joint is that the pin is pulling the two pieces together. This is not true. The holes are offset a slight amount initially, true, and the pin is there to mitigate the effects of expansion and contraction over time, and add mechanical integerity to the joint. If the pin has to be pounded in, or the two pieces clamped tightly to insert it, in furniture (building joints are a different animal alltogether). The joint has not been prepared correctly.
it is still a very viable, and necessary joinery technique.

[Randy Klein]

Does anyone else wonder if the cost in weakening the joint might not outweigh the short-term initial benefits of drawboring, or have any counter arguments?

I would argue that it's the long term benefits that drawboring is important. It offers a mechanical lock to hold the joint together when (not if) the glue fails. Granted, this may be 75-100+ yrs down the line, so you must consider the planned life of your piece. IMO, for heirloom furniture, yes; for kitchen cabinets, no.

[George Bregar]

First, at one time thi joint was used without glue, I've read more than one piece where an antique furniture restorer has drilled out the pin and the pieces slid apart. So this idea of it weakening the joint, while maybe true, is probably just a matter of inconsequential degrees.

As mentioned, it's been used in TF for centuries, and you will often see old barns where the frames have survived intact when all else has failed.

Not sure if it really a necessary joint with today's modern glues. In old pieces, they used hide glues, which were a sure failure over time.

I use pinned tenons as a decorative touch, and to pull joints and hold them tight initially. I think they will not effect the pieces longevity over the long term.

[Chris Friesen]

A misconception of a drawborne joint is that the pin is pulling the two pieces together. This is not true. The holes are offset a slight amount initially, true, and the pin is there to mitigate the effects of expansion and contraction over time, and add mechanical integerity to the joint. If the pin has to be pounded in, or the two pieces clamped tightly to insert it, in furniture (building joints are a different animal alltogether). The joint has not been prepared correctly.

After reading multiple articles on drawboring, I'd have to disagree with you. The very nature of drawboring is that the pin is under tension and is pulling the joint tight. This is why it's important to use straight-grained wood for the pin, otherwise it can break. That said, the _amount_ of offset must be scaled based on the material being worked and the size of the joint.

Also, drawboring can be used where you don't have any long enough clamps, or where you want to be able to disassemble the joint later on (of course you don't use glue in this case).

[Wilbur Pan]

My two cents on drawbored mortise and tenon joints:

Penny #1: A drawbored mortise and tenon joint is self clamping, which can be a huge advantage sometimes when doing a glue up.

Penny #2: Chopsticks make great drawbore pins.

Having said that, I don't buy the "drawbores make the joint weaker" theory, but I also keep my drawboring holes on the smaller side.

[Bob Lang]

I used to be in the "don't need it, liable to break something" camp, but Chris Schwarz's research and testing on this convinced me otherwise. As with any technique, it needs to be done properly, and if the pin is oversized for the mortise/tenon joint, or in the wrong location it can break the tenon, as has been seen in "other" magazines tests and examples. When done the right way, the pin acts as a spring, holding the joint tight if the surrounding wood shrinks.

Bob Lang

[Mike Cutler]

Chris

I agree with you.

This sentence in your post expresses it better than I did.

"That said, the _amount_ of offset must be scaled based on the material being worked and the size of the joint."

We have one point though that we don't completely agree on, but I believe it's just a use of terms, or definitions that seperates us.
The pin is under tension. No doubts there,or there would be no reason for the "draw" part of draw bore. If however the joint cannot be put together and fitted properly without the pin being pounded in. Something may need to be re-evaluated.

The breadboard I made in this attached link was done with a drawbored ,M&T breadboard end.

http://www.sawmillcreek.org/showthread.php?t=61157

The center 3"-4" of the breadboard end(the area of the through mortise) is the only place there is adhesive, that pin in the center is drawbored. The rest of the breadboard/ breadboard end is free to move. The two pins toward the ends of the breadboard end are also drawbored, and elongated, but there is no adhesive in that part of the joint. It is the pin that provides all of the integerity here.
The mortised end of breadboard fits flush against the shoulder of the breadboard, without the pins inserted. With an approximate 1/32-1/16 offset on the drawbored holes there is pressure on the pin, but the pin was not required to draw the joint into position.
The pin is not pulling, per se, the joint tight. It is maintaining the joint tight.

[Brad Olson]

Granted, drawboring has the immediate benefit of self-clamping and of drawing the joint tight initially. Does anyone else wonder if the cost in weakening the joint might not outweigh the short-term initial benefits of drawboring, or have any counter arguments?

This is why I do them.

However, in something like a workbench you need to think about the type of failure you are likely to see.

In a workbench you are not likely to see complete joint failure, rather, I argue you would be more likely to see glue failure due to the racking forces over time. Thus the joint would "break", but would start to loosen or separate. A drawbore, would resist racking failure of the glue and also keep the joint together without glue.

Same rationale goes for regular M&T vs. other variants. I always think about what the piece will be used for and the likely mode of failure and design accordingly

[Chris Friesen]

With an approximate 1/32-1/16 offset on the drawbored holes there is pressure on the pin, but the pin was not required to draw the joint into position.
The pin is not pulling, per se, the joint tight. It is maintaining the joint tight.

Oh, sure. It should be a tight joint before the pin is inserted. The insertion of the pin will put some tension on the tenon to lock the shoulders tight.

I see it almost like prestressed concrete. The tension on the tenon will help ensure that even under moderate pressure the shoulder of the joint never opens up. The downside is that under extreme pressure the pin may totally blow out the mortise piece or shear out the tenon, which are the failure modes in the magazine tests.

[Alan Schwabacher]

Schwarz advocates making it so strong there is no trace of movement. From that point of view, failure of the joint happens when it gets a tiny bit of movement, and that's when you would repair it. Your grandchildren probably don't need to worry about that.

If you get an earthquake, tornado, or explosion sufficient to cause the kind of failure to a workbench leg that will pull out the side of the pin, you will have other things to worry about, like where your house went.

[Dan Barr]

Brad, if i have a failing joint i dont think i want a through bored M&T holding it together solely by a pin. I would rather repair it on the spot. if i wait til the TBM&T fails thereafter, i probably wont have a chance of repairing it because of a split tenon or possibly a split apart mortise.

a well fitted M&T will hold itself to some extent. its not like the mortise is going to completely slide off of the tenon. i might see an exposed shoulder, and then i would know to fix it. if i have a TBM&T, i wont necessarily see an exposed shoulder and i wont know to fix it until too late.

I still see no purpose in a TBM&T other than appearance, maybe. maybe im not enlightened to something here. shouldnt a M&T be good enough on its own? What specific application is a TBM&T going to give me an advantage?

if i have a racking force and a TBM&T, wont that pin force the joint to split apart in a more destructive manner? Yes, i will be realizing the full strength of the M&T, but it will be in a destructive manner and threfore never repairable.

the shear strength of the pin is far greater than the shear strength of the cross-grain mortise member. The pin is endgrain and will compress existing material. the mortise member will not compress, it will split. therefore, im going to have split apart cross-grain before i'll ever shear the pin's end-grain. see what im saying? the sides of the mortise member will split or the tenon will split long before the pin will ever shear.

i would rather have repairability.

am i wrong here?

please be objective,

cheers,

dan

[Randy Klein]

am i wrong here?

I think so. :D

The forces that these magazines use to test strength are beyond any normal force a piece will ever see. And these mag tests have shown that when the normal M&T joint fails, they don't fail at the glue line, but break along the grain somewhere in the mortise, making repair about as difficult.

The repairs for normal M&T that are easy to do are when the seasonal cross-grain movements (after many years) have weakened the glue bond to the point that the tenon can slide out. But a drawbored joint will still hold in this case and will not need repair.

[Brad Olson]

A drawbored M&T doesn't really require glue. In fact i've seen some "green woodworking" pieces that were entirely constructed without glue.

It is the same situation as dovetails. They are both joints that do not derive their strength from the glue. Further, these joints don't require glue to remain intact.

Thus if the glue fails within the joint, the joint stays together. While in theory this shouldn't happen, it is probably one of the more common type of failures, especially in joints subjected to racking forces over time.

If you are putting enough force on a any type of M&T joint that causes and instantaneous catastrophic failure (like the tests in magazines), either your design is incorrect, or you are doing something very wrong with what you are making.

[Walter Phelps]

I happened on this discussion and it interested me enough to register and post an opinion. When I first started woodworking, I had read all about draw boring and why and how it was done. I've since come to believe that it is a technique that may be useful in hardwood or on a large scale such as timber framing, but in the window and door joinery work that I did, it was impractical.

The supposed method is to cut all of your joints, bore the peg holes in the mortised pieces, then do a trial assembly to mark the hole position on the tenons. The work is then disassembled, and the tenon peg holes are bored slightly off center. When the work is reassembled. and pegged, the off center holes cause the work to draw tightly together.

In practice, it is far too time consuming to assemble a door or window twice. If the joints fit well, it is a difficult and (often damaging) task to knock them apart again. Furthermore, boring a hole exactly where you want it in softwood is nearly impossible as it is very difficult to get any style bit to follow a punched center mark.

When antique mortise and tenon joints are disassembled it may appear that they were draw-bored, whereas in fact, the swelling of the mortised piece may have simply caused that appearance.

I can say with a fair amount of certainty that draw boring window and door joinery is too impractical to have been a normal procedure.

[Kirk Poore]

I've tried drawboring, and in most circumstances I don't believe it is superior to clamping tight then boring the hole straight through. Drawboring a clamped M&T joint isn't going to make it any tighter. The "prestressing" rationale might add a little strength, but if the joint is down to the pin strength alone (as opposed to pin + side friction, both assuming the glue no longer functions), I can't see the extra strength being enough most of the time.

I think drawboring does have a place where the joint can't be clamped tight, and you have to rely on it being tapped together. This would include irregularly shaped pieces and pieces too large for clamps, as in timber framing.

Kirk

[Wayne Cannon]

That has NOT been my experience. I just finished repairing an almost-antique dresser a chair, and a modern cheap wooden glider, all of which used pinned mortise and tenon joints. The glue failed, and the mortise sides split, the portion of the tenon beyond the hold pulled out, or both. A 2" dresser leg (with a 1/4" tenon on a side rails) split due to racking forces after the glue failed.

If, as in the case of the barns, mentioned earlier, and Chris Schwarz' bench, the mortise walls and tenons are so thick that this type of failure is very unlikely. However, in normal furniture, the failures in the recent magazine article test are exactly the kinds of failures I have seen.

Unglued joints (as in Chris Schwartz' leg-to-top joints) and no-clamp-required joints make sense.