??s about joist height and brace placement

I’m modifying the “garden toolshed” from Jack Sobon’s first book in a number of ways. One of the things I’m doing is extending the post both “below” and “above” the crossbeam, meaning more headroom on the first floor, and a larger kneewall in the loft up above. I ran into the problem he warns against, and put the brace pocket in the same area as the mortise for the crossbeam. There are two solutions I can think of and I’m definitely leaning towards the first, but would really appreciate any feedback, thoughts, or concerns any of you might have.

Solution #1: Extend kneewall height even further, putting the brace pocket up above the crossbeam mortise. I’ve drawn this out, and that means the kneewall height will be extended a total of 1’8”. This leaves me with 3’8” between the top of the plate and the top of the crossbeam. Jack says not to allow more than 4’ between the top of the plate and the top of the crossbeam because the spreading force of the roof is magnified at the post/crossbeam connection as the distance is increased. So, I’m (barely) within this rule. Do people think 3’8” between the tops of these two members is alright?

Solution #2: Instead of using 3’ braces here, use 4’ or maybe a 30-60 will work. My hesitations are this is my first frame, so I want to keep it simple. It seems like there is lots of information out on how to layout/cut 3’ braces. A 4’ brace wouldn’t be too different, so maybe that could be in the cards. I’ve never even come close to working on a 30-60 brace, know very little about their layout, and am not even sure if they would solve this problem. Also, the other braces in the frame would still be 3’, so this means there would be two different sized braces. I like the idea of that, but in the effort of keeping what feels like a dauntingly unfamiliar project as simple as possible with the goals I have for it, this seems like it could maybe mess me up. I also was going to make a brace layout jig, which then wouldn’t work for a longer brace.

Any ideas? Like I said, I’m leaning towards Solution #1, and have already drawn that out, but want to know if anyone thinks the distance between the plate and the crossbeam is too much, despite being within the 4’ rule. Or, if anyone disagrees with my hesitations about implementing a 4’ of 30-60 brace instead.

That’s it, but it may be relevant to say the posts are 8x8s. Solution #1 would give them a shoulder to shoulder length of 10’11”. On top of them sits the 8x8 plates where the rafters are step-lapped into. The top of the plate with Solution #1 comes in at 11’6”. Difference between top of crossbeam and bottom of plate would be 3’ (a lot?).

Solution #3: make it a shorter brace so the pocket is above the cross beam (tie beam)....

A lot depends on your snow load for your area....where are you located?

Im in Western Massachusetts, which isnt quite northern New England, but years like this one remind us what winters can have in store for us. At Heartwood I was advised to use 50 psf for Massachusetts snow loads.

Dan,

I live in Becket, and I sometimes teach at Heartwood as well. Not sure if we've met. Let me know if you have any questions or need a hand.

At 12' wide it's a pretty small building, so I suspect that you're kneewall height is OK. Especially if Jack OKs it. 30" braces would work fine (42 7/16 length) as Jim mentioned.

Another consideration, that I can't remember whether Jack addresses, is that you'll need some kind of girt or nailer to at the level of the joists to catch the end of your flooring (assuming there is a loft space). That adds even more joinery at the post/TB location. There are few ways to improve that situation.

In discussing this design with Jack last year, I believe he suggested that the snow loads for this building might need to be checked against current codes. I'll ask him tomorrow when he comes to the shop, if I remember.

Cutting a 4' brace is the same as a 3', you just have to calculate the hypotenuse.

I've cut 3-4-5 braces. Layout is with the square. I like to leave the degrees out of the thought process. They are also easy to layout and cut.

Dan,
Can you please cite where Jack says that 4' max is OK between the tie beam and the plate on a kneewall? I'm not saying he's wrong and I don't have a copy of the book with me, but I'm surprised.

A quick calculation shows that an 7x7 with a 3' cantilever above the tie beam to the plate MIGHT be OK if there was no joinery, but after taking out the tie beam housing, mortise and peg hole it wouldn't (assuming an allowable fb of 1000lbs/sq.in, 65lb.sq.ft. combined load, 14' width and 30" rafter spacing). There's about 1500 pounds of roof thrust at the center post top and over 50,000 inch-pounds of moment at the tie beam joint.

In our workshop where we build this frame often, we always limit the kneewall to 2' max.

I'm not an engineer, and I could be missing something. God knows I hesitate to say Jack is mistaken, but he may have assumed something we don't know. But I do know it's dangerous to rely solely on rules of thumb without understanding the underlying principles. These types of calculations we cover in our Timber Frame Design course at Heartwood and I believe can be found in the Guild's two design workbooks. But it's always best to consult an engineer if in doubt, and by doubt I mean not knowing how to figure this out yourself. I realize that's why your asking here.

I remember one time I heard Jack say something different from what he said the year before, and when I asked him he said "Well, last year I was wrong!" And we didn't catch it then, did we? Gotta love the guy.

I talked to Jack today about this briefly, and he did say that moving stuff around presented the problems you ran into. He did not elaborate further. I also asked if he felt the design needed to be reviewed to see if it meets current codes, and he said yes.

Will, page 101,first paragraph, "the distance from the top of the crossbeam to the top of the plate should not be increased beyond 4 feet because of the spreading force". I wonder is this is part of the re-do?

The post in the red book are 8x8,btw.

Not a bad idea to increase the post size, maybe to 9x9. Also, if you had two braces coming together at the same point on a post, one on one side and one on the other, and using 1" reduction/housings on a 7x7 the braces conflict, and on an 8x8 the touch, the 9x9 leaves you with some grace. Just another reason to use a bigger post.

Thanks, Tim,
An 8x8 with a 1" peg, 1.5" thru mortise and 1" housing and a 3' kneewall still doesn't quite make it with an actual f of 1029lbs/sq.in. and an allowable of 1000 (note this is higher than any grade of Eastern White Pine, but hemlock might get you over the hump). One thing that Jack might have considered with the 4' limit was the load range; if there was no snow load then the cantilever can be longer. As with any design, local conditions need to be factored in. There are safety factors built in to these calcs, as well.

Thanks so much guys for your input.

So, here are some options as I see them, either individually or in combination with each other:
1. I can use hemlock and red oak for all my posts. Looking at Will’s quick calculations this could get me over the hump of the high kneewall. This seems like a good idea no matter what, as there are certainly some harvestable oaks and hemlocks that could be used for this project.
2. I can lower the post height and kneewall by 6”, and then use 4’ braces along the long wall. This would leave me with a kneewall height of 3’2” (top of crossbeam to top of plate).
3. This problem is addressed in a couple of books I’m using, and one of the solutions shown throughout is using diagonal struts joining from the top of the tie to the upper ends of the posts. I could do that on the two end bents (wouldn’t really want for the middle bent for loft accessibility). I don’t know how this extra removal of wood in the post would subsequently reduce its strength, but the brace pocket would only be a few inches below where it joins with the plate.

The last thing that’s running thru my head, and I understand this forum won’t help me calculate this, is that the walls/insulation for this building will be straw bales with a earth/lime plaster. In a house this size (small single story), the bales could be structural in themselves. I am going to try and see if they can be relied upon to help prevent this spreading force at the tops of the posts.

timberwrestler, I’m not sure we’ve met, but I’d love to be lent a hand sometime. At least four braces will be cut from curved timber, and that will be a first for me. Also, some timbers may be hand hewn, and I’ve never worked snap-line layout before, and I’m sure I’ll have questions about that.

Thanks for all your input gentlemen; I am invigorated by this process. Makes me want to design my next structure completely (with some more practice and training of course),
Dan

Dan, #3, I don't feel right about. It is tension joinery and If you are looking to gain more room on the second floor the struts will protrude into the space. They may appear like they are doing something but is it worth it? Boost the size of the post, #1.

If the straw bale wall strengths the wall, it most likely won't help with the canted knee wall, in my view.

If you have to use struts, don't use straight mortise and tenons.
You could use a 'suicide' tenon which is a tenon inserted into a mortise that is wider at the bottom and expanded with wedges driven into kerfs in it so that it can never come out again. This is a joint I am familiar with in furniture, but haven't come across it in framing. You have a flared mortise, as described, and a tenon with maybe 2 saw kerfs cut into it. When this joint is put together, you place a wedge in each kerf and then drive the strut home. The wedges expand the tenon inside the mortise, and it can't come back out. This joint resists tension, which is its function in furniture joinery.

Or a lap dovetail could be used to resist tension.

Or, like Tim says, you could just not use struts as they would interfere with interior space.

Another solution is that you could build a truss with a profile kind of like this, which is sort of like an upper cruck, or a German Liegender Stuhl, depending on whether you use straight or curved supports...



This is I imagine a great departure from Jack's design..

But a higher knee wall is possible, it just may require rethinking the roof structure.

note, this drawing is just a very quick sketch, there is nothing technical about it.

I don't imagine that strawbales would do anything for you in resisting thrust. I don't know much about them, but imagine that any structural properties would be along the lines of bearing vertical loads

It is already a small building, therefore the conservation of space would be on my list of things to keep in mind. Now, there is the issue of the short cantilevered joist ending in open space, unused space, so it may not be important.

I must just be a simpleton. Let's see, 7 more pieces, that is 14 more joints, times the 3 bents, that is 42 more joints. Or up size the post?

Not that I am knocking DL's sketch, I think it is a reasonable idea, and on a larger building it would be very reasonable. But the topic at hand is only 14' wide.

Thanks DL. I like the look of that. Something like that certainly could work.

I'm with TIMBEAL on this though, as I'm trying to keep this simple. Sizing up the post, or simply lowering the kneewall are fine options. I wonder with oak and hemlock posts (could all be calculated with hemlocks numbers) how high a kneewall I could get with 8x8s. I'll see if I can run those numbers looking thru these posts and my many books on hand.

timberwrestler, what are the ways you're referring to that could improve that situation?

One way to attach a floor joist at the wall to a post and not create a big joinery problem is to make the floor joist a "slide-in" joist.
That is you raise the frame and afterwards you slide in the joist from the outside before the siding is attached.
This joist can be somewhat smaller then the joist in the middle as it will not hold up as much as the others. It will be somewhat supported by the siding being nailed to it.
To make the joint not show (from the inside), a simple notch can be cut on the side of the post, large enough to hold the load and it could look like this:

Dan,

I've done what Jim mentions, with some structural screws thrown in there. If the enclosure system allows it, I've put strapping on the outside of the post that catches the flooring. With a conventionally framed wall around the frame, I've put blocking at that level. That nailer is not really taking that much weight (depending on the joist locations), it's really just to keep the ends of the flooring from flapping around.

On a separate note, I'm a big fan of Tim's idea of putting the tie beams at different heights, and using soffit tenons on one side.

Brad

Hey Will,

Could you please point me to a source(s) that can 1)explain/walk me thru the type of calculations you did here for my post size in regards to height, load, and removal of wood after joinery, and 2)list the various values for different species of wood. I want to qualify my feeling that Oak and Hemlock will work structurally with a taller kneewall, especially if Oak will be used for the posts in the middle bent. In the meantime I'll see what I can find.

Dan

Dan - Do you have the TFG Joinery and Design books? Great places to start.

http://tfguild.stores.yahoo.net/deswor.html

http://tfguild.stores.yahoo.net/tifrjoanddev.html

Dan,
Not being an enginer, I am reluctant about giving out any definitive engineering advice over the web such as an actual calculation for someone else's project, especially without seeing all of the conditions in the design. I advise you consult an engineer or the Timber Frame Engineering Council. That said, if you want to learn to figure this yourself, you could consult the books that Mike recommends, especially the tension joinery articles in the 1st volume. Design values for the various woods are also there, in the NELMA grading manual and the National Design Specification Supplement published by the AFPA. Calculating the new centroid based on the removed material in the post can get a bit complicated, but I would recommend some of the books in the Parker/Ambrose simplified structural design for builders series. We also cover much of this material, and the calculation I used, in our Timber Frame Design and Joinery Decisions course at Heartwood. I'm reluctant to just throw this calculation out there because there are many underlying assumptions and data that need to be considered, and the calculation needs to be done with a background understanding of basic engineering, such as where to find design values for the wood. In our design course we have a week to build up to this sort of calculation, but it's too much to do here.

Dan,
You are also welcome to come over to Heartwood since you're not that far away - alumni get free consultations! We could look through the library at all the books we've been mentioning and I could show you some of the typical calculations for this issue.

About the end girt to carry the end of the flooring: we have often just put the first and last joist in about 3" from the face of the post (10" or so from the wall surface) and cantilever the flooring boards out to the wall. A trim board is much easier to attach after the fact than putting in a structural member, and the post is not compromised.

Mike and Will, thanks for the info on resources.

Will, wow, what a generous offer. I'd love to take you up on it. Its really just the post size and species in relation to the kneewall and joinery that Im still trying to figure out. There seem to be simpler solutions out there, like sizing up the posts, or most simply bringing down the kneewall, but learning some of those calculations should certainly prove helpful in the future, and hopefully for this projects question as well.

I also plan on cantilevering the floor boards, and with my joist layout it will be 9".

Im close and available, and feel so grateful to have you guys nearby.

Dan