Sobon shed and snow load - for dummies please

[Jim_Rogers]

Many years ago, I was designing a shed for a customer. I told him the size his plate needed to be in order to support his load.
He was confused as his shed would be about the same size and loads that Jack's shed would be and my plate was larger.
He asked why?
I ran the numbers of the plate in Jack's book per my spreadsheet calculator and it failed.
So, I called up Jack, as I do know him personally, after taking his timber framing workshop.
And I asked him how that plate size is correct.
He told me that the distance or span that I was using for my calculations were too long.
I said that my distances of the span on the plate was from post to post, face to face to make it the "unsupported span". He said that distance is not right because it's not taking into consideration of the brace coming from the post to the plate on each end of the gable bents.
He told me that an engineer told him that the span could be reduced by the distance of the brace on one end of the plate. This means part of the load on the plate is being picked up by the brace and transferring it to the post.
When I used this shorter distance (one less brace distance) the plate calculations passed.
You may need to adjust your unsupported span of the plate.
Just something to consider.
Jim Rogers

[Don P]

Not in my world, and for dang sure not with green timber. That brace is loose and unsupporting when it dries. The correct span per the NDS is the clear span from post to post plus half the required bearing length at each end.

[Jim_Rogers]

That brace is loose and unsupporting when it dries. The correct span per the NDS is the clear span from post to post plus half the required bearing length at each end.

If it is truly "un-supporting" when it dries, what good is it?
Why bother to even put it in?
How do you figure it is loose?
If the joints are cut correctly then the brace tenon peg, if draw bored, should keep the brace tight to the housings on both ends.
Jim Rogers

[Don P]

In green timber that subsequently dries they are effectively worthless. Or really, they are only as good as the peg, there is no bearing. The post shrinks in width, the angle on the shrinking brace becomes acute so that its first point of contact in a racking situation it a small pointed toe which then needs to compress wood until it becomes wide enough to bear the load. Draw boring only means it has pulled that loose brace into contact but think about what has happened it is not in full bearing, it is not bottomed. At best the peg is what would take load, the bearing surfaces have left the area. The sheathing does the real bracing. On open structures there really isn't much there.

Try an experiment. Build a Y braced T out of green timbers, a post with a section of beam, make braces tight and well fitted but don't peg them. Peg the post and beam. Check it a few years later. In oak you can slide the braces around about 1/2". To make it structural you would need to drive wedges into the brace mortises in either the post or beam or both.

Trusses, same thing can come into play. On the kingpost trusses in my shop I made one with a discontinuous bottom chord, kind of a proof of concept exercise. I passed a spline through the bottom of the kingpost and attached the 2 halves of the bottom chord to it from each side. I was proving to myself that the bottom chord was totally incapable as a beam supporting the kingpost, it is just a tension member. That worked fine. Till the kingpost shrank in width. I should have anticipated that and started with a little camber to the bottom chord. The bottom chord is now slightly sagged. No one has noticed, but its there. At some point I'll lift it and insert shims alongside the joggle at the head of the kingpost.

Anyway, in common usage the span is the clear span, do not use braces for support. I gave the technical span above which adds a couple of inches most of the time. Sobon and his engineer are overstressing the members.

[swmn]

I talked to my local saw mill about plates.  They are willing to sell me a 20 foot 12x12 (#2 SPF) for $360, and they have a couple dozen in stock for me to choose my two plates from.  So $720., plus I'll need at least one helper for several hours to pick through the stack.  Now I can call an engineer and ask how much he would charge me to figure out how much money he could save me.

Would I probably register that against the outside surface of the shed and leave the extra width of the plate inside the building?  Leave the braces also registered on the outside surface of the building so as to not weaken the posts?

I used 97" for my span.  When I put the 12x12 into the beam calculator as 12 wide and 10 deep after notching for rafters it passes with plenty to spare, so that's good.  I asked them about a 12x10, the guy said he would just sell a 12x12 for the $360 and then charge me shop time to cut two inches off the width.  With no rafter cutouts a 10x10 passes.  With rafter cutouts a 10x10 doesn't pass without an engineer to do his thing, 20 foot 10x10s are $250.  

So if an engineer can get me into 10x10 plates ($720 for 12x12 - 500 for 10x10) for less than $220 for the stamp I will come out ahead, and I probably need to add a few inches to the span.

Appreciate the brace and shrinkage discussion.  Looking at the workbench I built with drawbored MT joints in January 2019 there's got to be some internal stresses in there.  As the legs have dried down from 20% MC the tenons from both the long and short stretchers that were flush are now sticking out of the perpendicular leg surfaces.  

I also did a lapped dovetail diagonal brace on one of the top to leg corners of the same bench, the brace shrank more than the leg and top as everything dried.  I had to hammer that thing in, but it would fall out now that the diagonal brace shrank more than the opening I had cut for it, just the pegs holding it in.  Because of that I suspect I will cut the through dovetails on the crossbeams first, leave them a little fat and let them dry during the rest of frame cutting and then trim to fit when all the other joints are cut and I am assembling bents.

EDIT: I though my brace stock was fully cured, +/- 11%MC, and I knew the legs and top were up around 20% MC, therefore the cutout for the dovetails should have shrunk more than the dovetails on the ends brace.  Clearly the brace stock was not fully cured as it shrank.

[Chilterns]

This is getting out of hand.

I have watched so many Building Alaska episodes that I know for sure that Alaska is stuffed full of trees and so obtaining an 8 x 8 wall plate should not pose a problem. If this really is such a big deal then chop a tree down and hew it out to the required size. A hewing axe will not cost you $720.

You seem to want to put too much play on what calculations and your engineer might conclude about the adequacy of 8 x 8 plates. The Sobon shed is tried and tested with possibly hundreds of these now standing across North America and elsewhere in the world.

Stick to Jack's recipe.

[Jim_Rogers]

If you are using the standard Sobon shed layout with three bents then the span of the plate is not 97"
the center post for the center bent reduces the span to a lot less
you need to re figure your spans lengths

Jim Rogers

[swmn]

What I did was stretch the frame to 18 feet so I could definitely fit my 16 foot boat (outboard skiff with the motor not mounted) up in the loft area over the winter.  And my ground snow load is 60 psf.   Both of those factors point to a larger plate - the length of the shed and the snowload.  The boat is ~100# and will be resting on crossbeams, not hanging from the roof.

With 7.5 x 7.5 actual posts on nine foot centers the clear span I come up with for three bents is 96.75.  Adding one inch to each end of the span to cover the required 6.4 sqin of bearing area I ran it, again, with a 99 inch span, and no change in the outcome for plate size. 

Since I apparently haven't screwed up anything too terribly I will price up timber for each bent and see how long a shed I could build with 4 bents and a smaller plate.

Thanks for your help.

[Jim_Rogers]

In green timber that subsequently dries they are effectively worthless. Or really, they are only as good as the peg, there is no bearing. The post shrinks in width, the angle on the shrinking brace becomes acute so that its first point of contact in a racking situation it a small pointed toe which then needs to compress wood until it becomes wide enough to bear the load. Draw boring only means it has pulled that loose brace into contact but think about what has happened it is not in full bearing, it is not bottomed. At best the peg is what would take load, the bearing surfaces have left the area. The sheathing does the real bracing. On open structures there really isn't much there.

Try an experiment. Build a Y braced T out of green timbers, a post with a section of beam, make braces tight and well fitted but don't peg them. Peg the post and beam. Check it a few years later. In oak you can slide the braces around about 1/2". To make it structural you would need to drive wedges into the brace mortises in either the post or beam or both.

I have to say DonP, that I totally disagree that braces will loosen up and be worthless in a dry frame.
I have conducted the experiment in real life.
Here is that story:
https://youtu.be/-nE1eVfird0
Jim Rogers

[Don P]

Not really, I've done the experiment I described, but that is a fine looking frame.

Before we wander too far off OP, what I'm really disagreeing with is the assertion above that a knee brace can be used to reduce the clear span.

I'll just quote from folks smarter than me. From the Guild's timberframe standard;

3.5 Seasoning Effects

For connections fabricated from unseasoned lumber, consideration shall be given to the potential effects of distortion due to shrinkage or warping prior to and after assembly of the structure. In addition, the potential effects of seasoning on the assembled connection shall be
considered. In particular the possible loss of bearing contact between two members and the change in orientation of bearing surfaces shall be considered.

From Dick Schmidt's paper on the behavior of traditional timberframe structures subjected to lateral load;

Using an allowable deflection of height/400, the allowable drift due to wind load on an eight-foot high frame would be approximately 0.25 inches...

The stiffness of all frames was lower than this value. This
indicates that traditional timber frames with knee braces as the primary lateral load support do not have adequate stiffness to resist typical wind loads. However, at displacements far beyond serviceable limits, all frames continued to carry increasing load. Therefore, these frames have sufficient strength to resist lateral load...




All frames exhibited free-displacement at low load levels...




The joint stiffness is initially relatively low at small displacement but increases as displacement increases. A step increase in compression side stiffness is assumed to be due to the additional joint stiffness realized as the joint surfaces bear against one another.

By the time the frames had the lateral resistance needed provided by kneebraces alone the frames had drifted from 1-3".

I think we can agree that load goes to stiffness. If the kneebrace is not the stiff element the load is carried by the post. Or, as I've said before, if you are depending on a kneebrace to support a beam, by the time that kneebrace "hooks up" the beam is already in trouble.

I would not suggest using the kneebrace as a beam supporting element, the span is from post to post. If the kneebrace is the primary lateral load restraint in an open structure it is probably best to make it just as long as it can be and fit the frame up as dry as possible to minimize seasoning effects.
 
I don't think I'm really all that interested in youtube fame so I'll let you all carry these discussions on.

[Don P]

 Jim, I recognize I have no control over actions other than my own. What made me feel sideways there was that you took this to another platform where I was muted while you and your client were busy banging on a pinned frame in the wrong plane and jumping to conclusions. That was more of a rant for your TV audience than a discussion. Pop the pins and insert a pry bar if you want to see if its really tight. Nuff said, moving on.

Let me show the other end of the shrinkage spectrum so the concept clicks.









My partner and I sawed the timbers, members of the Guild, who you know, notched and erected the frame. It was tight, angles correct, but it was green. It is not yet fully dry, this would be maybe 2-1/2 years old. I can go get similar pics from a much older Hundegger frame. That CNC machine advertises I think 5 thousandths accuracy, but the end result was the same. The run here is in the 60' neighborhood, but that doesn't really matter, what is going on and what we are talking about here is in the 6' brace area. As you can see the brace angles became more acute. With the post shrinking in width the braces also become too short and lift off the bearing. If the wood shrinks this is what happens. The degree to which it happens depends on the amount of shrinkage. That is moisture content and species dependent. It does not matter if you are draw boring or believe in magic, the sides of the triangle have changed dimension. This is just basic geometry. For a 45° brace, the brace, the hypotenuse, becomes too short by 1.414 times the gaps at the bearings caused by that shrinkage. And it has a pointed toe bearing. That is not a smart beam support. It is not the stiff point. As I mentioned in a pm to you last week, if the timbers are close to emc this becomes less of a concern. We can do whatever we want but the building always knows the real load path.

Since the pegs are under tremendous tension the brace will ring if struck. Notice what is probably a tension perp failure at the one post peg. All of that is what led me to do the pegless test I described above.

Where is the real lateral coming from on those side sheds? The pegs and braces certainly contribute but really, the roof plane, screwed tight through the flats of the steel roofing for tight contact, collecting and delivering lateral load back to the log core of the building is the stiff element. I saw comment on another thread last night asking why screwing moved to the flats on roofing. That happened when post frame builders began using the diaphragm action of the steel deck in their engineering.

[swmn]

snip...

Let me show ... [a] concept.



That is moisture content and species dependent. ... the sides of the triangle have changed dimension. 

Notice what is probably a tension perp failure at the one post peg. 

Don, I heavily snipped your post but you have demonstrated a thing I was trying to figure out about braces and girts.

I can see drawboring the plate down on to the post tops, so as the plate seasons and shrinks "most" of the plate shrinkage will be measurable as the rafters getting closer to earth, and a very minimal portion of plate seasoning able to be measured as increased vertical clearance between the floor and the bottom surface of the plate.

But what about those darn posts?  They are going to shrink in cross section, so the girts and braces, which aren't going to change in length much at all, are going to need shoulders cut in the posts so there is a path other than the peg to get loads to the foundation after everything is dry.

I am going to keep reading about this, but you have pictured a thing that I thought could happen.  I am just going to have to make some scale model joints, maybe 2x2 post and beam with 1x2 braces, and cut the joinery open (lapped instead of mortised) so I can see inside the joint, and drink probably several beers.

Your picture did confirm my suspicion though, so I can go on to trying to figure out the next step.

[Don P]

Darn computer ate another tome, I'm becoming a one tree a day logger :D. A couple of thoughts.

Check Sobon's "Historic American Timber Joinery" start on pg 31 for braces;
https://www.ncptt.nps.gov/wp-content/uploads/2004-08.pdf
Look at the pics of packing pieces. If they and the brace are tapered they can be snugged up as the frame seasons. I'd make them out of hardwood.

Bolsters on top of a post can reduce the clear span of a plate but it does add another shrinking issue.