Alaskan remote 16x28 1.5 story in a Winter Wonderland

Started by ChugiakTinkerer, January 20, 2016, 01:21:07 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


Looking good.
Just a quick look, the tie is unnecessary, but I like it.
The kneebraces under the floor are also unnecessary to my way of thinking IF the posts are sized correctly. I'm viewing them as (look at awc DA-6) a beam overhanging a post with a point load on the end. The beam is your post, the ground is the point load, the support in that scenario is the floor. There is also an axial load on our "beam". I think you are following my thoughts there. The kneebraces are then redundant, just fine but not needed, the braced walls connected to an unbroken post have answered that need.

Check your floor structure dimensions, kneebraces do not decrease span.

Umm, I never cleaned this calc up and put it in the beam index, point load overhanging a post, looks like I just scanned the DA6 pic onto the calc for reference;
Bear in mind this isn't taking the combined load into account, just the bending.
(ahhh, I see what I was doing, that calc was set up to check the studs on a story and a half design without ridgebeam or ties... fail) aaanyway, you can back out inputs and insert yours.

The interior wall needs a double joist under, I'm sure it would help as a braced wall (an undesigned wall) as opposed to a shear wall, a designed and properly secured wall.

Are you sawing timbers on site or bringing them in?
edit to add, I've been playing with sawing posts with crotches included in the upper part of the post. Essentially the kneebrace, a branch, is part of the post. In that case I believe the "brace" does reduce the span, it is part of the post. The 8" thick post I got yesterday is 24" wide at the top. With one of those on each end of a beam I do believe it has decreased the clear span by 4' where a conventional kneebrace will have lost its ability to rigidly support the beam as the post width shrinks, in that case the true bearing, the rigid support, moves back out to being over the post during seasoning.
Can you do metal fabrication?


The calc at I can use to look at how the unsupported posts respond to a transverse wind force, or equivalent force induced by an earthquake.  I can back out those numbers for the beam and plug in ones for an 8x8 white spruce.  I'm thinking the numbers for the load I can get by looking at the total wind force on the structure and treat the ground resistance as the point load on the end of the beam.  I'm thinking I should look at uniform distribution of the wind load on all the posts, and also look at double or triple that load on a single post and see how it holds up.  If it passes then I'll be happy with no additional knee braces at ground level.

Timbers will be hauled in, I'm thinking.  I don't have enough big trees on the property to pull this off.  I can look around and see if there are areas on the state land where I can fell some decent trees.  A permit for firewood and/or house logs can be had for free.  Actually now that I think about it there is a stand on my trail in that has some nice trees.  The problem is I don't have a mill.  Also the local guy I'm considering can supply my needs for under $2k.  If he has the spruce on hand.  His other species for timbers is poplar, probably balsam poplar.  There's not much info on mechanical properties, but what I can find suggest it's not nearly as suitable for timbers, plus a little more prone to rot.  So if I can't find a supplier of white spruce timbers I may be felling and milling my own.  For just doing timbers a chainsaw mill would work good enough, I reckon.
My cabin build thread: Alaskan remote 16x28 1.5 story


I wouldn't use bam for timbers or lumber but it would be fine for sheathing or boards. The price for timbers seems quite reasonable if the haul isn't too tough... all stuff you'll need to weigh.

You're following dead on with my thoughts on the engineering, which only means 2 novices agree  :D.


So I've drawn up a couple cost estimates on connectors for my wannabe post-frame constructed cabin and got a bit of a shock at what the price is even for the Simpson connectors.  Bottom line is if I use the fancy connectors for every joint then I'll be spending more in steel (or aluminum) than I will in timber.  That just seems wrong.

I'm not a fan of the big black metal plates, so I'm considering connectors that are discreet or invisible.  My connector options are:

1) T-Rex connectors from Connecticut Post and Beam

2) Timberlinx connectors

3) Simpson Concealed Joist Tie.  Simpson also has some other connectors such as the HUCQ that was pointed out earlier.  Might be an option where the joint isn't visible.

4) or something from the school of joinery,as shown in Jim Rogers' picture here:

I find myself drawn to the simplicity and efficacy of option #4 as shown.

For the joining of the 8x8 timbers I will probably go with a Timberlinx connector.  I especially like the 17.5" one meant for connecting three timbers together.  For the smaller members, such as the rim board or the loft joists, I'm wondering if I can't get way with doing #4 and cutting out a shallow pocket then securing it in place with a couple of Timberlok or similar structural screws.  The shoulder will be holding up the joists and the screws are good for several thousand pounds in tension and shear.  I just need to be sure and provide the minimum bearing length to allow the joist to have its full carrying capacity.

That brings me to another problem.  I really like the exposed beams and t&g decking on the underside of the loft.  Here's a sketch of what I am talking about.

El problemo is that an 8x8 can't span 15' and support the floor too.  Turns out an 8x10 has trouble and can only do it with some select and #1 of a few species.  I could just use sawn or BCI joists and be done with it, but I'd lose the rustic charm of the exposed joists and 2x6 tongue and groove, not to mention a little extra head room.  So I've looked at glulams and a 6-3/4 by 9 inch glulam can easily manage the load of the loft floor.  Would I be compromising the glulam overly much if I were to carve out a pocket 1/2" deep or so and drop in the 4x6 nominal joist?  Or do I need to consider going up to the next size of glulam to ensure I've got enough meat on the beam after carving out the pockets?

My cabin build thread: Alaskan remote 16x28 1.5 story


That felt undersized. The stud wall is only about 10' from the rear frame... post underfloor and turn the summer beam 90 degrees with joists then also turned 90?

I've used timberlinx, they work fine but are a bit fussy and not cheap either. Whenever I can simply set a joist or rafter on top of a beam and block between, I do if possible.

Recessing into a glulam is getting into deep water. I've done it very lightly at a ridge/rafter connection that then had thru bolts taking the load. Minimum bearing for a seat is 1.5" unless engineered. Modelling how a beam responds to notching is still not entirely clear. Jim Rogers and I were at a conference where Dick Schmidt showed full scale breaking of notched glulams. Video of the tech diving for cover as a large beam broke and bounced around the lab was impressive. A variation of the soffited tennon in #2 performed best. The tennon is in the neutral axis of the girder and diminishes back to leave a full dimension top edge (the extreme fiber in compression). The beam in #4 acts pretty much like a beam of full depth but only the thickness of the remaining wood between notches. The lower remaining wood under the joist does not seem to form a T beam as one would expect, or as many have modeled. That part is trickier than it first seems. The folks at FF have much more experience than me.

Free downloads on this page from the Glulam folks;
I do have an older copy of the manual, 3rd, (they are up to 6th now) if there is anything inside you need.
You are not out of (relatively) concealed connection options. I'll try to sketch another that I've seen.


You can fabricate connectors as well. The correct code path is to have an engineer design it, a certified welder fabricates and you install to design. The awc connections calc has in it most of the tabulated steel plate design numbers from the NDS. Bolt size, spacing and count through whatever guage plate into whatever specific gravity wood gives design capacity.
Anyway this is one very quick sketch of a way to mostly conceal a connector while providing good bearing and some tension/uplift restraint

For a floor joist/girder, I'd be tempted to put a top flange that bears on the girder rather than relying on bolts and possible tension perp in the girder. Over the summer beam they could be welded to a plate that saddles two of these connectors back to back over the girder. (one possible derivation for summer beam, "sumpter" the most powerful ox on the farm)
Other shear developing connectors for heavy timber are split rings and shear plates, happily I've never had to go there. Those are concealed within a bolted connection... I guess really the drilling precision is probably no tougher than timberlinx.


Thanks for the ideas.  My welder is only a 110 AC mig welder, and it's only good up to about 1/8" steel.  So I'd need to go to someone else for fabrication, at which point I'd probably just buy a connector like the Simpson CJT and call it good.  I didn't take connector cost into consideration so one option is to drop a joist completely and spread them out a little more.  4x6 nominal might be adequate, or I can upsize to 4x6 rough sawn, or even bigger if needed.  The joist isn't the real problem here though, it's the summer beam.  A post running through the stud wall is definitely an option.  Turning the beam 90 degrees as well as the joists and decking would make that work.  I'll sketch that up as an option.  I'll run these past the boss and see what direction she wants to go.  And then I'll have to figure out how to join the rotated beam in at each end.  Despite my desire to avoid challenging mortise and tenon joints this one looks unavoidable.  Might just end up with "Screw it, throw some BCIs up there".

Edit: your comments on notching the glulam are well heeded.  We won't go there.
My cabin build thread: Alaskan remote 16x28 1.5 story


Here's another thought, all of this is crudely sketched in. It looks like uniform load on the summer beam at 30/10psf is about 4200 lbs. A triple 9.25" lvl works. With a 1-3/8" thick wood ledger well attached to that triple lvl on each side under nominal 4x6 joists (840lbs uniform) and the same wood blocking between joists, a bottom wooden cap to cover the edges and it should work.

We're below the prescriptive 1.5" min bearing, so check bearing capacity. The joist area bearing on the ledger is 3.5" wide x 1.375" deep=6.125 sq inches of bearing seat. 420 lbs at each end of a joist/ 6.125=68.57 psi in side grain compression on joist and ledger, from NDS 425 psi is allowable...check.


I checked in with SWMBO and she's on board with just putting the joists on top of the summer beam.  When I told her it's a lot cheaper and easier I got the thumbs up.  She's half a foot shorter than me and won't miss the headroom in the loft.   :P

Putting a post under the brace wall takes enough load off the other loft beam that I can go with an 8x10.   Current design looks like this:

Note that the brace wall will go all the way up to the decking.  The post in that wall will probably be some built up 2x6.  A 4x4 would be strong enough.

I've been doing some analyses on the 8x8 posts.  I've decided I need to get up close and personal with each post and beam.  To start with I've numbered my posts in a clockwise fashion:

Posts 4 and 8 have the most load at 14,213 lbs, nearly double that of posts 1 and 11 at 7,470 lbs.  Doing some simple column load calculations from the Toolbox, I can stack all the load at the top of a 16' unsupported post and a 7x7 made of SPF #3 will hold it.  I figure, and this is where I get dangerous, that the post is braced and the load isn't all at the top.  So I ran some calcs on the three stages of unsupported beam: a 5' column bearing the roof load, a 9' column bearing the roof and loft, and a 3' column bearing the roof, loft, floor, and walls.  For each of those columns a 5"x5" is sufficient.  So I figure I can use 8x8 columns and notch as needed to support my girts without compromising the load-bearing core.

I also re-ran the top plates using 8x10 SPF #3.  With that grade of wood I could conceivably get by with a 7x10 to carry the trusses.  I went with 100 psf load calcs, which was 70 for snow, 20 for dead load, and 10 for wind.  So I'm comfortable that an 8x10 will be sufficient, provided I can get timbers that are at least as strong as indicated for Spruce/Pine/Fir #3.

I'm pretty comfortable with the major frame members as specified.  I still need to do load calcs on the girts and size them appropriately.  I'm thinking that 3x10 rough-sawn is good for  girts at floor and loft level.  I'm also thinking I can get away with a full housing 1.5" deep for a quick and dirty joint with the posts.  Three Timberloks and Bob's your uncle!

My cabin build thread: Alaskan remote 16x28 1.5 story


I like it, agree with the column bracing analysis. You can drop the girders some to gain loft room if desired, just keep them above 7'. we were deep into it on one job when I realized the injunear had dropped a beam in the entry below 7'... I think he could tell I was ready to reach out and touch someone  :D. The different post loads is part of the difficulty with spot footings, differential settlement. Keep that in mind as you prep the foundation.


Well, I'm going to have to upsize my girts.  A 3x10 of SPF #3 just barely passes carrying the floor load.  Adding in the dead load of the 2x6 wall framing and sheathing and it fails.  A 3x11.25 of Doug fir #2 passes, as does a rough 3x12 of SPF #3.  I would rather go deeper than wider, from a simple frame construction standpoint.  Wider than 3" girts and I don't think I can safely notch the posts.  A wider girt is feasible but then I need to look at mortise and tenon joinery. 

Because I'm way out of the prescriptive realm I'm doing all my calcs with conservative numbers.  Load for floor is 40 live and 20 dead.  Load for walls is 10 lbs per s.f. of wall.  I'll confirm this at some point down the road, but I believe those dead load numbers are at least 50% over what actual framed load would be.

I hope quality spruce logs yielding 12" cants are available.  My local guy is getting a load of spruce in August.  I should high-grade all the good stuff.  At least my girts aren't more than 6'-6" long, so if big enough logs are in short supply I can buy 7' rather than 14' boards.
My cabin build thread: Alaskan remote 16x28 1.5 story


That gives room for a redundant ledger under the joists, not a bad thing. Going deeper rather than wider builds strength and stiffness much faster.

Grading, generally... you know generally, most people will balk at a true #2 IF they are paying attention. Most people do not pay attention. I spotted a rafter in a thread a few days ago that was below a #3 and have sure done it myself. If you can tail the sawmill you can begin looking at it as it opens up. If you stock just enough then you are stuck with those decisions, I like having a good bit of extra so I'm not pinched. Things twist or check or I failed to really look a timber over. I like to block between timbers as well as things like light mounts between joists, etc can use up low grade or extras. Point there is, if you keep your head in the game #2 numbers are fine. If you don't keep your head in the game #3 numbers are not going to save you.

Slope of grain is very often the control. The knot often isn't the real problem, it is the grain running out of the timber short and fast in the area surrounding the knot. You want straps of good straight and sound wood running along especially the tension faces. Funny thing about that knot though, because of the cross grain swirl surrounding it, that face draws and shortens during seasoning so that when you crown the timber that knot wants to be on the bottom according to crown and on the top according to strength. That timber just became something shorter.

For awhile I tried to hide or cut out those knots, aligning them into mortises or notches. In reality that simply further compromises the already short grain surrounding the knot and concentrates stress around my sharp corners (a human failing, nature doesn't do sharp corners, she flows stress out) Anyway, morning coffee rambling, off to help a plumber make something go downhill through all my blocking  d*


Just an update of where I am now.

Here's the post frame structure with all the members sized appropriately.

The 2x6 framing will go flush with the outside, leaving some exposed timbers inside.

And here it is all dolled up with windows, doors, and siding (sorta).

Exact size and placement of the windows are still a big question mark, but it will be something like this.  The door wall which is where the 12 o'clock post is will be facing southeast.  Lots of windows for sunshine and view.  Actually getting to this stage, I'm realizing that the loft is rather dark and enclosed.  Am contemplating a dormer looking out on the lake.  I know they are challenging enough with rafters, what might I need to keep in mind with scissor trusses?  I guess I need to start looking at every dormer framing job and get a sense for how it's done.  Or not.  Keeping it simple has it's up side.
My cabin build thread: Alaskan remote 16x28 1.5 story


I take it a ladder will be used to access the loft? I really like your model. Look forward to your build.


It's a tight fit, but I'm planning on the Jefferson stair.  There's a link to it here.

I'm starting to feel marginally capable in Sketchup now.  The plugins that are available make it an amazing platform.  A simple stick frame structure can be sketched out in literally just a few minutes.  It's a great way to play around with design ideas.  The two I use are Medeeek Truss Plugin and HouseBuilder.

My cabin build thread: Alaskan remote 16x28 1.5 story


The way we typically frame dormers with trusses is to slide several trusses together at each end of the dormer area to form a girder truss (these are designed) and run beams between those girder trusses to carry the dormer. With scissors a gable dormer will want to plane in with the interior ceiling plane. I'm seeing an upper and lower beam there to carry the inner and outer roof planes. That looks nice and adds a window but little room. A shed dormer would probably run from the outside peak at the top end down to the level of the interior peak at its lower end. This would give more useable room but gives some hard interior lines to work around visually. I like that option more if there is a dividing wall between loft room and the greatroom, getting into personal taste. Or the roof could be structural ridge the whole way and rethink that part.


Quote from: ChugiakTinkerer on July 28, 2016, 04:36:10 PM
I'm starting to feel marginally capable in Sketchup now.

It's not an easy app to learn, and I'd have to start from scratch again if I ever need it again. But it is pretty cool. It really helped my project with window placement and interior "feel". It's also fun to show off the fly-through feature. It was extremely gratifying to have the result end up so close to the drawing.

Looking forward to see your build come together, especially in such an awesome location. Best wishes!


There are also light tubes for additional light.

Are you planning any bear determent?


Quote from: ChugiakTinkerer on July 28, 2016, 04:36:10 PM
It's a tight fit, but I'm planning on the Jefferson stair.  There's a link to it here.

I'm starting to feel marginally capable in Sketchup now.  The plugins that are available make it an amazing platform.  A simple stick frame structure can be sketched out in literally just a few minutes.  It's a great way to play around with design ideas.  The two I use are Medeeek Truss Plugin and HouseBuilder.

I'm also doing a "Jefferson" stair with a landing and turn in our cabin.  I'd never seen one before and it was just what I needed for the limited space I had available.


A structural ridgebeam in lvl with a midpost could be as small as 3 1/2x 11 7/8" (I think I'd bump it to a triple or a double 14"). An 8x12 in spruce if it is very nice. I think that might make for more useable space and easier dormers if you go that route, another option to think about anyway.


Thanks all for the feedback and encouragement.  I've been consulting with my client, and I'm doing a lot of modifications here and there to get a floor plan she's happy with.  Still trying to keep posts out of the open floor, so am favoring trusses at the moment.  There are two interior posts now, both located in interior walls.  We'll see what the implications are of some of our design choices.  I'm liking this 18x30 layout with a 6' knee wall.  Dropping roof pitch to 8:12 to keep the overall height the same.  Will post more once I know this will pass my necessary load calculations.

@ kenhill,
I haven't made bear deterrent or mitigation a design concern.  Exterior doors will open out, so that the door jamb will provide reinforcement on any bear pushing on the door.  I'll probably have heavy plywood covers for the windows.  The area is not heavily infested with bears, to my understanding, so things like nail boards may not be necessary.  Hopefully we can find the right degree of deterrence as we go along.
My cabin build thread: Alaskan remote 16x28 1.5 story


Here's where I should probably trust the numbers, but I want to sketch out what I think is a viable joint where the tie beam and girts all meet on the post.  In addition to holding the frame together these are supporting the loft floor.  Size of beam is 8x10 and girts are 3x12, rough sawn.  I've notched a full housing for the girts and beam 1-1/2 inches deep.  That leaves a core in the post that is 5" x 6-1/2".  According to my load calcs a post 4.9" x 4.9" will carry this load.  That is a cross-section of 24.01 square inches, whereas my notched post still has 32.5 square inches, an upgrade of 35%.  Is that cutting too close to the bone?  Calcs are assuming HemFir #2.  Anyone have any thoughts if this is an acceptable joint design, or should I use hangers for the girts?

Edit: Forgot about the part of the post between the beam and the girt.  On each side that is an area of 3.5" x 1.5", for a total of 10.5 square inches.  So my notched post has an area of 43.0 square inches, which is 79% more than the minimum.
My cabin build thread: Alaskan remote 16x28 1.5 story


I'd be tempted to frame the walls under the girt area, install a rim joist resting on the wall, unnotched at the post, then frame above the floor.


Hmm, a hybrid platform framing / post and beam construction.  That would put more load on the girt below, but I don't doubt that it would hold it fine.  Hang on a sec, let's be certain.  Go to the Toolbox and select a uniformly loaded beam, with double the load for the floor girt.  And...

Nope, misses by || that much.

I think where I've got the girts and beams coming in at the same point I will opt to use a hanger.  Something like the HUC212-2, which has an allowable load of ~3,000 lbs.  Two of those should be able to hold up a girt loaded with 3200 lbs.  I could frame in the 2x6 wall underneath beforehand to give it a little extra oomph, although I think that might be overdoing it.  These hangers will be covered by the framed wall so I don't have to worry about it looking too industrial.

So my new and improved joint will look like this:

The Simpson bracket does great against uplift and down load.  I have my doubts about pullout resistance since it has no double-shear nailing flange.  So a Timberlok screw or two should help in that regard.

Edit: Actually, the girt loads for the loft aren't as big as the loads for the floor, because there are two posts supporting the tie beams.  If I were to have a frame and rim joist supporting the loft, all supported by a floor girt, the max design load would be ~4850 lbs.  So the floor girt could hold that.  But to give me some peace of mind I'll proceed with hanging those girts with the StrongTie hangers.
My cabin build thread: Alaskan remote 16x28 1.5 story


Additional load is half a 2nd floor bay to the next joist and the upper wall load. Visualize it with tall studs and a nailer at 8' across the studs.
The timberlocks will provide a little lateral in the hanger (diameter, depth, density and toenail factor) but the sheathing will do the job if it spans across the post. Another way to provide that tie is a strap from girt to girt over the post and hidden by the sheathing. A roll of Simpson strapping is very handy to have around, or repurposed heavy lumber banding in a pinch.