Using Dowels instead of Mortises and Tenons in Timber Frame Construction

Hi folks,

I am designing an octagonal cabin about 12' in diameter from corner to corner. I have a bunch of red oak and would like to build a timber frame, but I am intimidated by the process of making mortises and tenons - I don't have much experience with chisels and making 24 mortises (using lap joints) and 16 tenons seems beyond my capabilities.

Instead a friend suggested I use dowels - essentially drill holes into the post ends, the lap joints of the beams, and the floor, and use these to hold everything together. However, he wasn't really sure of what sizes everything should be. He thought maybe 1" dowels would be enough. I plan to use 5x5 posts and either 6x6 or 4x6 beams. There will of course be braces as well (4x4s), and the roof will be evenly octagonal with 8 3x5 rafters. I am getting the lumber rough cut so these will be true sizes instead of nominal ones. It will be green as well

Does anyone have any experience with building in this way and/or know of a guide to do it? I am most interested to know how wide the dowels should be and how deep they should be inserted. Thank you!

I know pegged construction  was used in timber framing in the past - square-ish riven pegs pounded into round holes (the term "drawbore" will help you when googling this technique). You still may have to do mortise and tenon with this joint (or at least half-laps). Dowels could work for this application but they need to fit just right - loose pegs won't do.

Red oak is easy to work, so it could be a good learning experience. You may not be able to do it quickly, but you should be able to do it accurately if you are careful. Large holes for dowels may invite splitting, depending on the forces and wood. If you must go the dowel route, I think I'd opt for something like steel lag-bolts, or at least rebar.

I wonder if instead of inserting one large dowel I could use two smaller pegs. I would think that would decrease the chance splitting and prevent rotation?

My barn is pinned together, being built before nails were cheap so each join is either a half lap with 2 pins or for the cross beams it is a mortise and tenon which is then pinned. I don't think you can use the pin with just a butt join.

First of all I don't know anything so don't ask me any questions. However I do read a lot. I recall seeing the peg technique and one aspect that was stressed was that it's important which way the wood shrinks as it dries. Different types of wood shrink differently.
So the type of wood of each makes a difference.
I'm pretty sure they used a different type of wood for the peg. And couldn't pegs being small be pre-shrunk ie dried.
If you do it right the peg won't shrink as much as the piece with the hole or at least not in the same direction. Also I think I've seen in books a dowel holding two pieces together and then another peg through the dowel itself. Surely this is described fully in an old book somewhere, probably in the uk. Maybe amazon.uk has a book on it.
Apologies for my non-technical language and also that I can't refer you to a source.

Good luck!

The peg in a mortise and tenon joint is to keep it from pulling apart.  Dowels can't do that.   There are ways to avoid the joinery with metal brackets and bolts or lags, but they get expensive fast.

> [may be too hard...]

Well, it's a tall undertaking! <g>

How much time can you put in and how soon must you be finished?

Do you have an excellent, "might be doable", or "nothing, and no clear way to get one" work space?

What tools do you have skills in? What tools might you acquire and learn?

Will you be working alone most of the time?

Those are the first and very critical questions that need good answers before plans and decisions might be made. IMHO.

And I'll play the devils advocate, just a little.

You have a building type you like and you have some resources, but, from your question, zero experience and very significant deficiency in needed skill.
MaayBe... It would be worth talking with people that have done and do this type of building a lot and have examples of their work standing, in use and with happy clientele.. Or people ditto who have years of timber framing behind them. It sounds like you have building plans of some sort, so you have a good starting place for some consultation. People who can provide strategically targeted input to fill in some of your knowledge and insight holes.

Consider: That is a LOT of weight over your head. It's a BIG investment on  your part, no matter who/how it happens. In such a situation, where you will be putting in at least the equivalent of $20k (and probably much more)  of your time, heart, hope and money paying $1000 (or whatever) up front for somebody that _does_ have  years of this behind them in a good way might make sense. Starting right. Hedging your bet.

Finding the "right person"  will likely be challenging. After  you get some feedback online and hone your questions and learn some options, you hopefully will start to get a feel for the scope of what you're talking about and that will help finding and recognizing competent and honest people in the alt-construction field and in timber frame. This is not a pay$ get-goodies. Finding a good person takes real effort. Then there will be thinking carefully about what you expect that person to deliver to you. When you do any transaction it helps a LOT to know what you want to receive at your end. You're just human, your expert is just human - it helps to be clear what will make this a satisfactory, even valuable deal, for you.   Eg. do you want full plans? I'm guessing not, but a few detail might be helpful. Do you want lists (eg. a Bill of Materials) so you can plan your resource? Do you want a written evaluation of different construction methods (including your joints)? Believe me, it help a LOT to put stuff on paper. If you "consult" try, nay, refuse to accept any completing w/out a written summary of everything you _think_ was covered during interviews, walk through, etc.

Look at the possible value  of getting help "over the counter", decide what you think is worth it. As you can see, I'm a "standard procedure" person. I believe those hoops can help a lot in avoiding the simple problems, allowing us to get on to the real troubles. <g>

Regards,
Rufus

Nope, don’t do it.  When properly constructed, you will have shouldered mortises on your posts, thus carrying the full strength of the beam onto the post.  If you don’t shoulder your mortises, you effectively reduce the load bearing capability of your beam.  Say you start with a 6x6, then cut off an inch and a half on each side to make a tenon but don’t put a shoulder on it, you have effectively turned that 6x6 into a 3x6 for load bearing capacity and created a fail point on the end of each beam.  If you use pegs instead of joinery, the problem gets even worse.  You use a 1”peg, expect 1” load bearing capacity.  Do a little research on making mortises and tenons, shouldered and housed tenons, get a skilsaw, tape measure, hand saw, framing square, 2” hand auger bit, 1” bit, hammer, sledge, some ratchet straps and a 1” chisel and you can do everything you need. A combination square and a 4” angle grinder with sanding discs are handy tools too.  I use either a draw knife or the grinder to make pegs, but you could just buy some hardwood dowel I suppose, make them too long to start with and sharpen one end just a tad bit to make it start in the hole easier.   Blocklayer.com has a gazebo calculator that will be very handy in figuring out your angles for your octagonal building, including rafter layout.  Timberframing isn’t nearly as hard as it seems and once you cut the first couple of notches, you’ll forget all about the worry of it.  I’m hosting a teaser class on timberframing without specialty tools in August.  We will be using round log, so as to show people this method of construction without the need for a sawmill.  The above tool list is about all you need to enter in to the world of timberframe construction.  Add a chainsaw and chalk line for round wood construction.  It isn’t until you desire to go into larger production that you will want or need to spend the money on the specialty tools like 16” skilsaws and chain mortises.

Jesse, is this what you're describing?  Blue is the peg/dowel?

Mike, that is basically what I am suggesting, with a similar method to connect the post to a sill on the floor.

T, thanks for the advice, after doing some research it does seem easier to me than I thought. I have found many examples of plans out there. However, this octagonal structure seems hard, and I have found no plans that could help me make my own. Do you know of a place where I might be able to find something like that? Even a place with info on the joints I would need. I was playing around in Sketchup and came up with the following (attached), but I don't know if this is a good joint or not, for example.

I see what you are going for in the sketch up you have here, but one thing I notice right off is I don’t like the tenon placement on the beam located on the right.  It could work if you move it over or change the angle of the joint so that the full width of both beams rests on the top of the post.  From what I see in the drawing, the beam on the left is resting fully on the post, but the beam on the right is mostly just the tenon over the beam.  I’ll try to get you a sketch of some alternative solutions soon.  
With your bracing, it is perfectly acceptable to have them attached to the face of the post like is shown.  This is definitely easier than trying to do a full mortise.  They are a little better if you make a dove tail on the end of it, but a straight lap like shown is fine.  Make sure you add the shoulder, which I believe you show in your sketchup.  I’m not sure what you have going on with your foundation beams, but it looks like you may be trying to use straight beams, with a short piece attached to make the bend?. For this just lap the foundation beams right under the post, dove tail them is better and cut a square mortise pocket for the post to set in.  This is assuming you have something under the beam that is bigger than the notch. If sitting right on the ground, ignore that, if up on blocks or pads, make sure each section of beam is well supported underneath. Hope I’m helping.

This is very helpful T, thank you. It is so hard to find anything about octagonal timber frame structures. Any sketches would be excellent - I try my best with Sketchup but I realize there are simply geometries I don't have in my head yet. I did work up something new for the top joinery based on your suggestions - is this the sort of thing you were talking about? Are the two joins not too close to each other?

You are right about my foundation beams, I was trying to make a short piece for the bend. I guess lapping is ok on the sill, which makes sense now that I think about. I haven't worked shoulders into my designs yet, but yes I plan to put them on the braces. Now that I think of it I was planning to flatten off the bottom of the brace tenon. I included a closeup of a brace join (without the shoulder or the flattening) since it was a little messy in my first diagram.

For your small simple one-story structure, I think lap joints for the top beams similar to Mike Haasl's sketch would be fine. I would hold the lap cuts back so the upper beam has 1" of full bearing on the post. No graphics available right now, but something like this:

.............|
.............V (pin)
__________________________
..................|
.........._____|
..........|
______|__________________
.....|.............|
.....|.............|
.....|.............|

(ignore the dots)

I recommend Tedd Bensons book “Building the Timber Frame House” for starters.  It has some decent graphics of the joints involved and layouts.  I’m sure there are other useful books, but that’s the one that got me started.  I think Glen and Mike Haasls recommendation of the half lap joint would be sufficient, a tongue and fork would be better, but like Glen said, make it so that the post holds at least 1 full inch of each beam.  I wouldn’t rely solely on a peg though, I’d use a tenon on the top of the post that goes through both of the beams, similar to what would be used on a post to sill connection.   For hardwoods, a 1.5” thick tenon is usually sufficient. 2” for soft woods.  For simple but effective joints, look at the tongue and fork, shouldered mortise and tenon, lapped dovetail knee brace and dovetail collar tie.  Your knee braces to the post connection will likely be your most challenging notch because it’s not coming into the post square but at a 67.5 degree angle.  Use a king post for your rafters.  Take a short 6x6 and knock off the corners to make it an octagon, cut mortises on all 8 sides, top and bottom.  Your rafters get tenons, as well as the collar ties.  Add a little something artistically to the end of it like some sort of pendant. If your sawyer hasn’t cut your logs yet, see if he can cut you pentagon posts with angles 135, 90, 112.5,112.5,90 in that order.  Then your outside corners of your posts will match the octagon shape, the knee braces will come in square to the posts, and the collar ties and rafters will still come into a square face as well.

It seems to me that placing the beams on top of the posts makes uncomfortable joints. So, I'm trying to figure it out with the beams hanging off of the sides (such as in the diagrams). This brings up a problem of tension though, as if this were a square structure I could use a wedged half-dovetail joint - is there something similar for a 135 degree angle instead of 90? The diagrams show two different possible joins.

Go with the second image and shave off the extra wood from the beam that protrudes past the post on the outside.  Use two 1” pegs per tenon and make your posts 6x6 at least or you won’t have much tenon to peg in to.  Other than that, looks good. Using a pentagonal post with the outside having a 135degree angle would allow you to have a similar joint but with tenons coming straight off the beam, not at an angle.  You could do the dovetail but I don’t think it would typically be used in this manner.  It should just be a straight tenon and pegged, so long as your tenons are big enough. With a 6x6 post, using a shoulder like you should, you are only going to have 2” of tenon to peg per side unless you overlap the tenons in the joint.  I’ve never seen it done that way but I don’t see why it wouldn’t work.  Cut the mortise 3” wide, make the tenons 1.5” thick and have them pass by each other in the mortise, then run 2 pegs thru the whole thing?  I hope it’s clear what I’m talking about in my super sketchy 2D sketch pad drawing.  

I were building something like this, I would definitely take the time to mill the posts octagonal in order to keep your joints at right angles as much as possible.. Getting odd angles right on end cuts, much less mortises and tenons, is at least a magnitude (ie. 10x, for those that haven't suffered through science classes lately) up in the difficult scale.


FWIW
Rufus

Thanks T Hayden, I'm going to try this on a sort of model first, so I will give that a try.

I see what you mean Rufus, but unfortunately I have no way of getting my logs milled other than square. I would think laying the cuts out with pencil ahead of time should be enough? I use Sketch-up to help me get the angles right.

I'm no timber guy, but depending on the size and number of your posts, and on the trueness of the face cuts, it seems  this should be doable "by hand" with a circular saw. The large ones with the 8-1/4" blade would let you tackle posts an inch larger. You would _most_ definitely want a good new ripping blade and you want to wax (or other lube) it regularly. A ripping blade makes a _huge_ difference in the ease and quality of a rip cut.

Mark the octagonal off on each end. If the resulting octagonal is less the 6" (or maybe seven if your saw spins a large blade) wide on each face, set the foot of the saw to 45degrees and tack a straight edge where it needs to go the whole length of the post. Then cut 1/2 of one octagonal face. Before moving the post, position the straight edge and run the saw the other side and to 1/2 of the face of the next octagonal. Rotate the post, set your straight edge and make the next cut back to one of the first cuts you made and there's one octagonal face. Reposition  the straight edge and run the saw down the other side for 1/2 of the next face. Etc.

The saw needs to be a good one with a strong stiff foot you can adjust accurately. It is _well_ worth an extra $150, or whatever, for a saw that won't wimp out on you and twist where it should stay straight. I don't know the good ones - the Home Depot Bosch I got 10 years ago is NOT what you want. The foot and how strong it is against twisting when it's set to full bevel is what really matters. That and a good blade.  When thinking about saws, consider which hand you normally will use to push the saw. The consider which side of that hand you want the blade to be on when you're using the saw. I'm right handed, but I plain cannot recall which side of the saw the blade is on. It's workable. But it does make a difference, so if buying  a saw, take a stab at figuring out what you want and try for that. This also affect which way the foot pivots. Read up or get some opinions, etc. Buy a $10 saw at a garage sale and try different cuts; get another $5 saw with the blade different and try _that_ and see what you feel.

Correct layout. You need to position the straight edge correctly for each cut. 8 cuts make a square post octagonal. With a blade with a depth of cut equal or greater than the octagonal face, you'd only have to do 4 cuts. After initial layout, 5 minutes to secure the straight edge, 2  minutes for the long rip, 3 minutes to recover composure and stop swearing, 10 min total for second rip w/out repositioning, then reposition post 15 minutes = 35min. x4 = 130 minutes per post. This _should_ speed up considerably, maybe down to 60+ minutes per post.  (edited: bad math fix, hopefully)

So it kinda depends on:
- the truness of the posts you get
- your layout skills
- the quality of the saw you can use
- the solidity and ease w/which you can secure the post for ripping; and the speed and ease w/which you can reposition it.
- you energy

Not sure this helps you directly, but maybe gives some ideas.

Ah. A _real_ 12" table saw with long tables in/out will do this job, also.  But that kind of saw needs a true 2+ horsepower and most people don't know where/how to find them at reasonable prices; it's a tool that requires some commitment. And I'm actually not sure, even if you had the perfect set up, it would be as practical as the circ saw with a straight edge.

Cheers,
Rufus

So much here. Loads of good advice. Well worth paying attention to. Dowels are not joints ;) It will be easier to learn to do mortise and tenon joints than to make dowels that aren't holding mortise and tenon joints together work to hold things ;)