Is there a best joint?

[jtbartlett]

Hello I am brand new to timber framing and need some advice. I am looking for the correct joint for the following application: shed style roof into lower ridge beam. I have Ted Benson's book Building the Timber Frame House but this particular joint is not covered. I have not seen too many shed style timber frames before so I thought someone here might me able to assist. My thoughts are that I shouldn't mortice the rafters into the lower ridge beam compromising the strength of the beam. Since the bearing weight of the rafters is 'into' the beam would a shoulder cut into the beam with the rafter pegged perpendicularly be enough? Thanks.  



 

[Jim_Rogers]

jt:
what is the pitch of the shed roof?
Jim

[Jim_Rogers]

Where the rafter meets the posts you maybe able to use something like this:




Can we see the whole frame?

Jim

[jtbartlett]

Sorry, I should have stated the pitch of the shed roof is 6/12. I'm working from images off the web so I don't have detailed plans. I'm working on a set.

[Jim_Rogers]

ok, I do understand.

first of all there are the basic rules of joints.

For example, if you were to try to cut a mortise in the horizontal beam for a tenon then the mortise would have to go with the grain.

Next, the tenon should be with the grain of the rafter.

this leads to a complex joint and can be very difficult to cut and get right.

let alone how it's going to weaken the horizontal beam.

Here are some examples:



Ok, so let's look at number 1.

You'll see that the tenon follows the rule that the mortise has to be at the cut with the grain of the horizontal beam. But the tenon is cut at an angle and that makes it cross grain and will split and not hold it very well to the timber.

Let's look at number 2

This one has the tenon in the correct position but the mortise isn't right and the shoulders of the tenon are in the wrong position.

Look at number 3.

This one has the shoulders in the correct position, and the tenon is in the right shape, but it has to be made longer in order for there to be enough wood to hold the peg where it goes through the tenon. This cuts a very deep mortise and will greatly weaken the timber. And both the tenon and mortise will be difficult to cut at this angle. (By the way this is probably not the correct angle for 6/12 pitch.)

If it was me doing this, I'd create a housing to hold the timber, and reduce the size of the rafter by say a standard 1/2" on the bottom to create a true shape/sized timber, and just lean it against the horizontal piece and secure it with a couple or three timberlok screws as shown at number 4.

[jtbartlett]

Thank you very much Jim. That's exactly what I needed. Out of curoisity is there an online source for timber joints available or did you post something out of your own collection? Thanks again.

[Jim_Rogers]

I drew that drawing to show you the joints choices.

There are free pdf's of joints you can download at the guild site.

Go to www.tfguild.org and look for the "historic American joints" series written and drawn by Jack Sobon.
I don't have the link right at hand, but you can find it.

If you can't let me know and I'll find it for you.

Jim

[canopy]

I like joinery that does not involve steel so just a question in that regard. What about take joint #1 in the picture and change it to a dovetail tenon that drops down from the top? The dovetail tenon could be made as thick as necessary to compensate for the angle of the wood grain. This in turn might require a wider beam. Just wondering what any of you think of that possibility.

[witterbound]

This is one place in my frame where we didn't take the time to do it the old way.  I have over 50 common rafters coming into ridge beams.  We simply cut shallow pockets, dropped the reduced rafters into the pockets, and used a timber screw.  We used a timber screw on the bottom too, I'll have to admit. 

[Jim_Rogers]

I like joinery that does not involve steel so just a question in that regard. What about take joint #1 in the picture and change it to a dovetail tenon that drops down from the top? The dovetail tenon could be made as thick as necessary to compensate for the angle of the wood grain. This in turn might require a wider beam. Just wondering what any of you think of that possibility.

Personally, I don't like dovetails. The reason is they shrink and this can cause the joint to pull out.

Here is a drawing to show where the dovetail tenon will shrink (the red dimension). To compensate for that some people who use these joints drive in wedges along the side of the dovetail (blue line). However this may not be the best solution.



Also, to cut a dovetail pocket in this beam could greatly reduce it's strength.

[icolquhoun]

i recently ran into the same type of problem with a shed style roof on the back of my cabin...i figured the easiest way to do this without the need to add a lower plate just for the rafters of the shed style roof was shown below using a principal rafter and common purlin style roof just for the rear shed-add-on...this frames isn't finished and is still being designed, hence the braces missing and second floor framing not being complete, just getting ideas out on paper right now

[witterbound]

Ico, it seems in your design it would be helpful to have a girt from the "house" post to the "porch" post, to prevent spreading...but I'm no engineer.

[icolquhoun]

Ico, it seems in your design it would be helpful to have a girt from the "house" post to the "porch" post, to prevent spreading...but I'm no engineer.

since there is no opposing rafter to create a spreading load, shed style roofing systems have no "spreading load" just a dead load (statically) that is straight down.  I can add a tiebeam, but it isn't necessary with the size of joints/timbers and size of that addition (12' wide by 8' deep).  I feel that sheathing (1" T&G) will be more than adequate to deal with any dynamic loads, however, I did consider adding just one tiebeam from the center post out to the rear post.  Hmmmm, that has me thinking......
thanks for the suggestion
 

[canopy]

Personally, I don't like dovetails. The reason is they shrink and this can cause the joint to pull out.

I am just trying to understand the optimal wood joint in a case like this. In this case the comparison is to a tenon that appears to have no pull out resistance at all.

Also, to cut a dovetail pocket in this beam could greatly reduce it's strength.

Which is why I mentioned a bigger beam may be needed if making the dovetail thicker than usual to compensate for the tenon grain angle or is there something else about a dovetail that weakens beams more than a straight mortise and tenon?

[icolquhoun]

Which is why I mentioned a bigger beam may be needed if making the dovetail thicker than usual to compensate for the tenon grain angle or is there something else about a dovetail that weakens beams more than a straight mortise and tenon?

A regular mortise in the case Jim showed above has the material removed in the center'ish of the beam, where there is little stress.  These beams will see compression along their top surface and tension on the bottom, like almost all simply supported beams.  by removing mortise material on top of the beam for the dovetail, you are taking away a considerable amount of material in the fibers of the beam that need to be continuous to resist this compressive stress.  how much is needed there?  well, it depends on the load the beam is going to see.  One could argue that if the dovetail fit perfectly, and a wedge was used to keep it snug, that because it fills the void, that the beam would still have the same amount of compressive strength, which is true if you neglect that the fibers resisting horizontal shear stress haven't been compromised significantly, like if you had, say, 20 rafters going into the beam, each section of the beam between these rafters would be so short that the top of the beam could "pop" off between the dovetailed rafters because the fibers of the wood didn't have enough area to attach them to the wood below.  This is not really an issue in this case however.   

If you have access to Sobon's "Historic American Timber Joinery"  look at page 43 figure 38 (in section five "roof joinery"), which shows a good dovetail shed roof addition joint.  If the beam has access above it to allow such a joint, thats not a bad way to do it. 

[Thehardway]

why not birdmouth the rafter and run it over the top of the beam?  A peg would then be driven in from the top. No wood is removed from the beam and the joint is greatly simplified making it both stronger and quicker/easier to execute.

What is holding the foot of the rafter and controlling thrust at the outer wall? This would also have a great deal to do with my decision on what joint to employ.

[jtbartlett]

The common rafters are the lower roof section of a shed style house. the rafters are 6x8 and the lower ridge beam is 10x12. I worry about removing any material from the rafters since they are much smaller than the beam. Icolquhoun posted above that shed style roofs are static load only since there is no opposing rafter. Is this correct? I think a housing as illustrated by Jim Rogers is probably the solution I will use. I can cut a mortice in the posts for their corresponding rafters but the 16 others will need their own solution. 

[Thehardway]

Icol is correct in theory, however, there are two  real world tendencies other than remaining static:

1. The rafter wants to slide down the incline. This is usually prevented by a joint or birdsmouth at the sill rafter plate which checks the sliding action.
2. If no birdsmouth or joint is used at the sill plate, but is used at the wall intersection, the outside shed wall will be pushed inward by the weight of the roof and a hinge action at the main building wall intersection.


For these reasons a proper joint at the rafter foot is necessary and it is wise to provide a beam to control the inward thrust.

I agree that both Jim and Icol's joint solutions would work but I don't see them as preserving any more wood strength or being easier to cut and fit. with the exception of perhaps Jim's #4 .  # 4 makes me a little nervous due to above issues 1 and 2.

I think #4 would be fine if horizontal beams tied the outside shed wall posts and sills plates to the main structure.

I think the joinery in the below drawing speaks well to your issues.  It removes very little structural material from the rafters or beams, eliminates movement with better load transfer and is very simple to cut and assemble.  The step-lapped joint controls both sliding action and inward thrust.  The upper joint can then be a birdsmouth, Jim's #4, a dovetail, or a soffit tenon.

 

One additional issue that should always be kept in mind is future repair.

Good luck with your project and keep us posted.

{moderator note: I changed a few terms to be plates instead of sills}

[Jim_Rogers]

I can cut a mortise in the posts for their corresponding rafters but the 16 others will need their own solution. 

You'll have to explore the post joinery to make sure your other beams intersecting the posts aren't affected by the rafter tenon mortise.

Three way timber connections are troublesome in timber framing.

If you were to do the joint mentioned above (birdsmouth over) or half dovetail shown below, this would move the beam's connecting point from inline with the rafters. And this would eliminate the 3 way connection.

[Jim_Rogers]

I worry about removing any material from the rafters since they are much smaller than the beam.

Ok, well you haven't included your location in your profile so we don't know where you are. Not knowing we don't have any idea of the snow load you will be seeing on this roof.

And I don't believe you have told us the length of the building or very many details about it. The more we know the more we can help you.

Seeing and understanding the entire design is helpful.

What is the snow load for your area?

[jtbartlett]

This is a picture of what I am planning. This picture is from Northern Lights Timber Framing's website. Clark Bremer is the designer. It's the closest image I could find of what I wanted, a shed style passive solar timber frame. I am stretching out the plan from 24 X 24 to 24 X 40. The bents will remain 24' but the bays will be 14, 12, 14. there will be (2) 12X14 lofts with stairs up the middle.  The roof pitch is 6:12 and I am planning on a steel roof so I'm not terribly concerned about snow weight.  The beams are quite large, 10X10 white pine posts and the ridge beam, front girt and rear girt are 10X12. I do not have detailed plans for this structure so I'm in the process of (learning) drawing them in Google Sketchup. So far so good. The replies here have been very very helpful. I am particularly concerned about: the common rafters connecting to the ridge beams, and the forces acting upon the 10X10 center posts. In my minds eye I am seeing  opposing forces from the shed roofs pushing on the center posts, the upper pushing  toward the front of the house and the lower roof undercutting it, pushing to the rear.  I am very green when it comes to frame designs and whatnot but this project IS going to happen.....hell or high water type of thing. I'm in northern VT. 

[Thehardway]

OK, this is beginning to make a little more sense now.  The rafters are birdsmouthed over the plate and restrained there by the girt between the central post and the shed post. It is likely a shallow housing on the other end as indicated in Jim's drawing #4.   Personally I would run the Rafters up and over the top plate on the second story to protect the clerestory portion from weather a little and provide shade in the summer but still allow winter heat gain.   I would do the same at the eave on the sheds. I would also add a girts  at all posts just below the second story plate,  (think triangulation) this will take care of your opposing thrust issues and also provide more means for securing whatever type of strapping and enclosure you plan to use.

One note, it appears that the plates are uninterrupted...  This is not a true  bent style construction.  to construct in bent style you likely have explained your bents and bays backwards. In this plan, I would say bents would be 40' long with posts spaced 14', 12',& 14' respectively and you will have 2 -12' bays.  Assembly and erection in this direction will be much easier as you are merely standing raising the post and plates to vertical and all are the same height and without interruption.  40' plates may be hard to source and handle. I would likely incorporate a scarf in the center of each plate. This would ease matters.

Does this make sense?

I would be careful of snow loads regardless on weather or not you are using steel roofing, especially in northern VT.  Spacing and spans should be a primary focus.  I am surprised your building inspector is not requiring stamped plans for this project.

[jtbartlett]

Thanks, thehardway, your explanation helps quite a bit. The thread shows 440 reads but I think 420 of them are mine alone. Lots to learn. I do have a question that maybe someone can answer. If I stretch the plans out to 40' I will need to add another set of posts, so 4 posts at 14, 12, 14 spacing. My problem, which you already mentioned, is that a 40' beam will be difficult to locate and handle. I am a little confused how to locate a scarf joint in the 40' span. My solution -which very well may be dead wrong, has been to use the same joinery the lower ridge beam  uses (I am drawing full through mortices with splines, fully housed beams. ) at the upper ridge beam and front eave girt. Is there anyreason why this wouldn't be ok? The load is the same for both roof planes. I understand that the upper ridge beam transfers the weight of the common rafters to the posts and that it would be better on TOP of the posts rather than joined to them, however I cannot figure out how to place a scarf joint in the 40' span. The only thing I have come up with so far is 2 more posts and 2 scarf joints for each beam, but that seems a bit excessive for a small frame. Maybe my explanations aren't so hot, if not let me know and I'll post pictures of what I'm working on.

As far as building inspectors go, the only thing that needs approval are the septic plans and installation and the electrical from utility to the panel. It's a small freedom but a very important one (to me at least). 

[Jim_Rogers]

Locating scarfs can be a tricky situation.

First of all you have to tell us, what is the longest piece of timber you can buy? or make?

Knowing that we can begin to figure out your scarf locations.

[witterbound]

Have you asked Clark if he'll sell you a set of plans?