i've been presented with plans that call for eight posts in a ring carrying 12 foot beams.
so two beams meet on each post and the connections are at an angle.
could anyone provide some joinery possibilities here?
thanks
-Mark in Ontario.

A very simple technique that works for us is to rip a post lengthwise, in this case with a saw bevel of 22.5 degrees. Then flip over one of the halfs end to end and glue the two ripped edges together (the glue makes it slippery and the "halves" will start to warp immediately so plan on using fastners to hold them together while the glue sets up. This allows the surface of the post presented to the beams to be a simple, traditional plumb, square surface for simple joinery. We use this in conjunction with our splines and it has saved untold hours in detailing, cutting, and head scratching.

Alternatively, if the posts are big enough then the beams could be housed into the sides of the posts, the 22.5 degrees would be cut off the end of the beams and the mortise would slice thru the post at the same angle. Remember, tenon grain should be straight, a smaller extension of the beam so that the wood fibers remain continuous thru the tenon and into the main timber.

clever trick to split and glue like that...
but yes these posts and beams will be at least 8x8 and possibly as big as 10x10
are you suggesting an open mortise and tenon at the top of the post?
also would it not be better to angle the post shoulder rather than the tenon shoulder?

as far as assembly goes something that drops in from above will work best

are there fastener alternatives here?

i know there are places like this being built as the plans are modelled after earthwood.....

THANKS!!!!!
-Mark.

by the way Paul, how big are the posts you split, flip and glue?
-Mark.

Mark—

Polygons are big fun, and I know from experience, jumping into one blind is the kind of thing that can and should make the hair on your arms stand up, so…

No hard advice here, but some simple points to help point you towards the right path

Your five sided posts should be milled at half your angle of interface, in your case 11 ¼,
Less acute angles make for joinery which is simpler to cut, with both less short grain and better edge distances.

The key to simply and traditionally joining a polygon (which oddly requires ignoring there is little such tradition) is to resolve thrust completely before it reaches your plates, this allows you to join them, be it to each other, a tie or the posts themselves as simply as possible, by eliminating tension from the quotient. First ask how to achieve this, then decide how to join it.

In my opinion, the only way to layout anything compound, and most especially anything polygonal is to scribe it. I know if you are not familiar with scribe technique this in no way helps prepare you for your jump…

All the same, I hope this helps

will... could you expand a bit on the five sided post? we have a sawmill and i have some experience in taper sawing and sawing clapboard

as far as thrust goes i'm a novice on circle engineering for sure
i know that this building will have a living roof and it is designed with no roof slope in the timbers, just the overlay materials. there are three rings of posts carrying beams in polygon shapes and the roof members sit on radial lines on top of the beams....the center meeting place of the roof members will be mitered around a continuous roof member spanning the diameter (does this resolve thrust?)
however, i don't feel right about a flat interior ceiling surface
my instinct tells me there should be some kind of rise toward the center
i'm thinking of sawing tapered roof members and keeping the tops flat, let the bottom rise toward the center of the frame.

speaking of layout i'll be starting at least with a snapline and square technique
i will try to lay out the angles on the ends of the posts and work from lines snapped between them
can you point to some reading material on scribing?

do you use slanted bearing shoulders?
i'm thinking of a wider shoulder on the inside of the circle than the outside
my main concern here is the joint at the top of the posts... a 3 way joint between 2 beams and the top the post
are open mortise and tenon joints strong enough here?
is there a way to use a continuous spline through the post here?
how about a bent steel plate?
could this be done as a mitered connection with lag bolts drilled and set from the top of the beams?
the wall system is strawbale and plaster so there will be shear strenghth provided by that system
where do we need braces, if any?
i've already asked the engineer about this but if you could comment?

thanks and more thanks
this forum and the writers here is are a big support!
-M.

Mark we rip 6x10 posts to make our resulting six sided posts, allowing all our plates to connect to the posts in an orthogonal plane. Will's approach makes the most sense if you want to keep your posts as one piece, and are relying on traditional joinery. Also, I doubt there is a better way to fit a joint tightly and cleanly than scribing if your customer is willing to pay for it.

It sounds like you need to employ a designer and engineer for this job, with no central post your rafters will distribute significant horizontal thrust to the outside walls and you will need steel to hold everything together. I have done this with wood in a gazebo since the spans were short enough to overcome thrust, but a building of any significance will exceed the capacity of a wood joint in tension.

There are a number of knowledgeable engineers and designers in the guild resource guide all capapble of advising you in this venture. Will is absolutely correct, be afraid, be very afraid. Paranoia is an engineer's best friend!

as far as fear goes
i've put in a lot of time milking holstien cows and lost most of my fear
those cows can get pretty rowdy, especially the first time you milk them.
the design is being engineered and i have sent several questions through the designer to the said engineers(timber sizing/rafter placement/placement of bracing)

perhaps a bit more info on the design would help
the first ring is a 6 foot radius
contains 8 posts and 8 beams

second ring is at the 16 foot raduis
again 8 posts and 8 beams

the third ring, (which falls outside the wall system and forms the outside wall of a wrap-around porch)is at the 23 foot radius
this third ring contains 16 posts and 16 beams

by the way i saw an interesting solution to the joinery last night.....someone did it with 4 "normal" bents and spaced the bents with 45 degree pieces

-Mark.

I recently worked on a similar building but with an exterior ring and a center post. I was told the living roof support structure needed a minimum slope of 1 on 12 even with the roof membrane. This project was not a residence but rather a classroom so it was required to resist seismic forces, hence knee braces were used. This one used cordwood instead of straw for the infill. There is no code accepted shear values for either straw or cordwood. (The third pig had no problem) Be careful, the living roof system is very heavy.

christopher-
this design does have an exterior ring
at 23 feet radius/46 feet diameter
we are trying to avoid a center post as it would be in the middle of the kitchen

-M.

Mark—

The milling of the posts and the backing angles on your hip rafters only requires a little figuring and the fabrication of a series of kick blocks for the bed rails of your mill.

I can not read enough into your writing to know much about your frame and reasonably comment on the specifics of how it should be configured, but will say this…

Your inner rings provide you your opportunity to resolve your thrust issues inboard of your plates. I don’t know that anyone has quantified the value of SB enclosure as shear walls and don’t buy into the notion that even good ones are reason to eliminate bracing.
In that I find a self aligning frame which is safe to walk while raising and someday safe to strip ( enclosure is in truth a maintenance item and should reasonably be expected to have a shorter service life than the frame it’s self ) I choose to put braces anywhere they do not interfere with millwork or introduce tension I do not wish to impart.

Steel is always a fallback, avoid treating ( and point your engineer towards same ) the main plate line as a tension ring and it can be avoided or used soley as a failsafe.

Again look to the inner rings.

do we need more bracing in the radial plane or the plane which follows the circumference? or, should there be the same number in these two planes?
Also, should all the rafters be on radial lines or is a pricipal/purlin system more efficient?

thanks again,
-Mark.

Will,
I wonder why you suggest not to rely on plate? I would say main plate can and should be treated as tension ring when there is proper tension joinery used.

Michal

Hi Mark,

I am currently working on rebuilding a large post mill which is fitted with a self supporting octagonal roof on a brick roundhouse and it is not permitted for the roof to lean on the centre post !

Octagonal buildings are not that common today and the main ones that can still be easily seen are probably smock windmills. These can have several storeys coming off 8 long continuous slanting posts sometimes arranged in iregular polygon fashion.

The ultimate octagonal building is the lantern at Ely Cathedral see Hewett - English Historic Carpentry page 160 -164. This has 50 foot high posts standing on a 30 foot radius.

You might also want to check out the carpentry on the Globe theatre which is a 20 sided polygon. I seem to recall that there was an article in Joiners Quarterly about this project and I believe also in Timber Framing. Check out the past issues. There is also a chapter on the construction of the Globe in the book Shakespeare's Globe Rebuilt (isbn 0-521-59988-1.

I am about to start work on an energy friendly timber framed house which will have 5 segments of a 16 sided regular polygon arranged in an arc to catch the sun's energy.

Why not post a sketch of your project on the website. Like Will I am struggling a little to visualise the concept.

Regards

Ken Hume
http:\\www.kfhume.freeserve.co.uk

Pardon the lapse, TTRAG and a raising demanded most of my time the past few.

Interestingly, the first polygon I worked on was a ill-fated attempt to replicate a twenty sided Globe with and under the direction of a friend of Ken’s, Paul Russell, the man I had the good fortune to learn scribe from many years ago. Ironically I picked up a long out of print tome on the Theatre in an antiquarian bookshop just yesterday

Mark –

Brace both the radial bents and the inner circumferential ring liberally, the main ring, to the extent that your window and door schedule allows. This is also where you can both negate and redirect thrust through strut/s to the inner ring where it is best resolved

Common purlins are an excellent step towards redirecting and concentrating thrust to the principals/hips.

Michal –

I’m afraid we will have to agree to disagree on the tension ring issue, however I must qualify that with a big ‘ole “in my opinion”

IMO good timberframe design demands the elimination of all possible tension joinery on the drawing board, I choose to do this, in that there are few joints available to us which deal with tension well.

The best case for a polygonal ring, are daps with material projected on beyond the vertices for ample relish. Mark is free to do this even on the main ring, simply by choosing to deal with the work necessary to cut and fit the SB enclosure around the relish, he is freer still to do this on the inner rings.

When we have the freedom to direct thrust away from one joint and in the redirecting now impart it not as thrust, but largely as a dead load to a point in the frame better able to deal with it, why not take advantage of that freedom ?

hi there
i've photograghed three of the design pages with my kodak digital camera and posted them to webpages....

here are the url's

http://www.geocities.com/markola17/photo2.html?1080148218620

http://www.geocities.com/markola17/photo3.html?1080147420170

i appolgise in advance for the quality of the images but these will definitely help those interested to visualize the building.
the design now officially has a center support(will either be masonry or wood) and the inner ring of posts and beams will move to the best placement between the center support and the 16 foot radius ring(8 or 9feet radius?).
again, many thanks to all who are writing, this is an interesting design and hopefully will be worth the effort.
-Mark in Ontario.

blast it
yahoo has cut me off....
i couldn't access the photos because i have exceeded my data transfer..so the above links may not work for you
SORRY!
i'll work on it
-M.

will....
excuse my ignorance but what is a "dap"(s)
-Mark.

another possibility for the joinery that keeps nagging at me is to place the 8 hips or principal rafters directly on top of the posts and move the beams down a bit(6inches minimum), getting away from the lack of relish at the top of the post...but how would it be assembled? you would be fine till you got to beam number 8 and then what? this is not a bent it is a ring so assembly is an issue
perhaps there is a lap joint for the beams?
-M.

Nothing to forgive, the ignorance was mine, in useing a regional term, habit had me forrgetting it was so little used.

Halveings - simply letting one timber into another

Underused, yet highly effective

today i spoke with the man who i apprentised with back in the 80's....
he suggested placing an additional timber across the corner of the beam connection(a relatively dry 6x6 should do[hardwood?])
this timber would dovetail to the top of the beams and receive the bottom of the hip...it would look a lot like a half-dovetailed collar tie
i like it a lot as a solution
for one thing it rounds the shape of the timberframe inside, essentially placing a second octagon shape within the main shape.
structurally the more weight, the tighter the connection -and the hip length is shortened somewhat.
At the post top/beam connection it allows very simple miter with a bit of indexing (t&g between the beams with an index of some kind on the post top should work)there is also the possibility of bringing a tenon from the post, between the beams and into the bottom of the hip(shrinkage in the beams and cross tie becomes a problem here - the peg could hang the hip as the timbers dry)

ok back to the drawing board!!!
-Mark.

Mark –

Your speaking of a dragon beam, an amazing and elegant system which both captures and reverses the thrust which would work to blow a corner joint apart and instead pushes it together and at the same time provides the space which would not otherwise be there, to land a hip on.

However, while a perfect solution for a rectilinear situation, the farther you splay from ninety degrees the increasingly less efficient this system becomes, by forty-five all advantage is lost. Neither still, do you need to create a place to land your hip.

Again, you need to capture your thrust up-roof, and if you have moved to a central post, you have, and the point is mute. Don’t get me wrong I’m not arguing for this as a solution. I think it both avoidable, and structurally and architecturally uninteresting.

Keep looking and keep the core open.

yes there is now officialy a central support
not a post but a masonry chimney
there will be flat steel rings around the circumference where the rafters(hips) bear to take the weight and steel pins will be welded or drilled into the ring for fasteners...just like earthwood.
it may be architecturally uninteresting but central masonry chimneys are very nice in a canadian winter...

as far as the dragon beams go, they do look good on paper, the designer likes them, would we be wasting wood and joinery time to use them? It's hard to see how the beams could escape once you place those ties down on them.

And, as far as daps go, are you thinking of a horizontal lapping of the beams on top of the post? i assume there would be a tenon of some kind rising from the post to index the joint?
the problem with this joinery is that one beam would be cut from the bottom up and this is bad for weight bearing timbers....do you have a way to half lap and get the bottom of both beams on the post top?
questions questions questions
i hope i'm not the only one benefiting from all these questions
a thousand thanks
-Mark.

A post would, yes be boring, your chimney, a far more pleasant story all together.

Hard fasten the hips, ( though it seems to me, this might need be massive or steel reinforced to deal with asymmetrical loads, be they wind or snow ) so they are now tensile members and you have much more freedom to do what you might desire below.

You could perhaps land your hips on dragons in those open areas, on ties in others.

The cant from which you saw the pentagonal posts ( like is often the case ) needs be large for joinery’s sake as much as for load bearing ability. Long story short, wider than the ties/dragon ties, the shoulders are proud of these, and provide full bearing no matter how you join the plates to them.

Seems like a great project you are pulling together…

Be your best and the luck will follow

ok
plan now shows a central masonry support
then rings of posts and beams(girts) at 8 and 16 feet of radius
the porch now has two rings of posts, one buried in the outside of the strawbale wall and a ring at the outside of the porch

The hips (principal rafters) are on top of the posts(through mortise and tenon)
The girts have moved down at least 6" from the hip/post connection (i'm considering timberlinx for the girt connections, they only need a shoulder connection plus a 1 1/8" hole and they can be tightened anytime-the trick will be assembly, the last timberlinx in a ring will have to have a very long hole to the outside face of the girt)
I see common roof purlins with hopefully a dovetailed ring at the 12 foot raduis or so...
I'm thinking 32 braces downstairs and 19 upstairs
The 5 northern posts at the 16 foot radius should be continuous to second floor

We would use dragon beams(or corner ties) on the inner ring if the client's budget allows....

Another word on the posts
i can see taking off the 22.5 degree cuts on the outside face of the posts(similar to the top cuts on a hip) but why remove more wood than that?
why not save the strength and cut sloped shoulders?
this may be something i see because of my layout technique, i use end level lines and quadrant snaplines between the timber ends, all layout as far as shoulders go is done with snaplines...so an angled shoulder is possible(although harder to cut)
i suppose the 11.25 degree post sides would look better?(except where the post meets the hip?)

ok
more later i'm sure
i'll probably exchange the design plans for the framing plans on the webpages when i'm done.
THANKS again to everyone who has helped me grapple with this, especially will truax and mark gillis
-M.

The octagon frame design is now on the web
There are two drawings, a cross-section elevation and a plan of the second floor deck.
The roof plan is quite similar to the second floor plan.
Again I appologise for poor quality on the images, I'm simply taking photos with our digital camera here(the scanner is coming).
Any comments or possible improvements would be welcome.
-M.
http://www.geocities.com/markola17/elevation.html?1082033225710
http://www.geocities.com/markola17/plan.html?1082033939740

hey gang, I wanted to dredge this thread up and post some photos of the octagon frame raising. The frame was in storage here over the fall and winter and last weekend we put it on it's foundation.....
here is a link to photos on a free web server called piczo...watch out for the ads!

http://www.piczo.com/wpltree?g=10263538

Mark

Just a note re: the chimney as masonary support.
You may want to run that by an engineer. It can of course be done but I think the work will most likely be done by a filled cell cinder block core over which you would have a stone facing. If you start with your flue liners (you show 2 flues) then the block and a natural stone facing, the chimney may dimention larger than the smallest dim of 3'
The good news is that the chimney could be engineered to resolve lateral force issues.
Looks like a fun project.

the project has run up a big engineering bill, to be sure.
the chimney is being built just as you suggest, John, block core with some rebar and stone facing(there will be a euro-style heater in the masonry work as well).
The chimney rings are 5' outside diameter for the second floor and 4' outside diameter for the roof ring. The chimney footing is over 8' diameter....
the lateral support is being provided by the strawbale wall system.... our timberframe includes radial bracing and the site-built porch frame has circumfral bracing, but both these bracing systems will be redundant once the strawbale walls are finished.