Built-up Beams (2x lumber)

[North Sask]

There are some stale posts on this general topic but I thought it might be worthwhile to start a fresh discussion. I am planning to construct built-up wood beams for both my floor beams and my ridge beam. Due to access issues (boat access only), large engineered beams are not a consideration. There will be four supports for the 24' floor beams. The ridge beam (also 24') will have three supports, one at each end and one in the middle.

I have the beam sizing sorted out but the splice locations seem to require further thought. Note that I want to build a continuous beam and not multiple simply supported beams, due to the numerous structural advantages. I have come across some conflicting information in the various sources I have reviewed. Some say that the splice must be located over a support. For the floor beams, supported at four locations, this will not be a problem. Combinations of 8'/16' and 16'/8' boards (if you get my drift) will work fine. The ridge beam is where this issue arises. If the splice is to be located over a support, two simply supported beams would be required. Some sources seem to recommend avoiding splices between the supports. Other sources indicated that splices between the supports are fine as long as they are within 6" of the quarter point of the clear span and in not more than half of the members. I should mention that the beams will be 3 ply, but the issue would be the same if they were 4 or 5 ply.

Any thoughts from the CP village? I'm not familiar with the IRC, so I don't know what it recommends. I have been relying on the CMHC information and I have found their info on framing quite useful. It is based on the National Building Code of Canada, if that is where you hang your hat.

Thanks in advance.

[Don_P]

 The IRC requires splices over posts, I believe Canada does allow splices near the inflection point ~ 1/4 span where the bending moment theoretically passes thru zero but you'd have to confirm that.

[Erin]

My main beam is a 3ply 2x12.  It runs the 40' length of the house and supports the 24' floor joists.

I have three posts, 10' apart. 
My beam boards are differing lengths.  The outer pair are 10', 20', 10' and the inner is 20' and 20'.  Consequently, all of my joints are supported over posts, but they overlap each other. 


_____ __________ _____
__________ __________
_____ __________ _____

         ||       ||        ||

My roof, OTOH, is supported by trusses.  No ridge beam needed.



Where your roofline is only 24' long, could you just get 24' stock?   Or, go with trusses? (BTW, I have a full loft.  My "trusses" are more like timber-frame bents)

[North Sask]

Don_P, if I was building to IRC I would have to either use two 12' simply supported beams or one unspliced 24' beam? The first option is a little undesirable for structural reasons and the second is undesirable for logistical reasons. Ah yes, the third option is to have an engineer sign off.

I have a bridge background so I can understand why the Cdn Code would allow 1/4 point splices. But I am used to bolted plate splices in a steel girder, and wood is a different animal. I did some further reading on a building code forum and the unsupported splice in wood seems pretty undesirable due to the nature of moment transfer in wood. I think it would take a lot of nails to transfer the moment across an unsupported splice.

It seems like anyone using a built up beam supported at three points would need to use either two simply supported beams or long members. This shouldn't be an issue with good site access. I might have to come to terms with transporting 24' pieces of lumber to build a continuous beam. I am going to see what the ol' internet has to offer in terms of splice connections for wood beams...

[Squirl]

By recollection, all building codes in the U.S. require splices over a support.  To do otherwise requires loading calculations.  Normally these are done by a qualified engineer familiar with the conditions and design.

How are you going to carry the load from the ridge beam support?  By your description a quarter of the weight of your entire roof in snow country is going to fall in the middle span of a floor beam.  This should be carried directly to a large footing to carry the weight.

http://publicecodes.cyberregs.com/st/ny/st/b400v10/st_ny_st_b400v10_5_par018.htm

Also, I believe you stated you are going to create the ridge beam of 3 - 2x12s.  My snow load is 50 p.s.f. in central  NY.  I am only guessing here, but I assume northern Canada has a higher snow load, so I used the 70 p.s.f. chart.  You didn't mention the buildings width.  Depending on the width, 3 - 2x12's may be just a little under recommend sizing charts that I use.

Depending upon the buildings width, even if you could get a solid beam to run the 24 length unsupported in your snow region, it would be massive.  Giving advice on how to home engineer that much weight with DIY mechanical connections is more than I feel comfortable recommending.  Don_P had some good beam sizing calculators.  Maybe you can find some timber at the site that could be used.  By a few rough calculations (very rough) it may have to be as large as 16x16, depending upon the conditions.

[North Sask]

Erin, I will investigate the 24' long lumber but that is not my first choice due to my access limitations (by boat). For the same reason, trusses are definitely not an option.

Am I understanding your floor beam situation correctly? Do you have a 5 foot cantilever on each end? I would be interested to know more about that situation.

[North Sask]

Hi Squirl, the plan is to support the centre of the ridge beam with a built up wood column, then floor beam, then 6x6 post, then a footing that will bear on crystalline bedrock. I mentioned earlier that the floor beams will have four supports (i.e., there would be no floor beam support at the mid span), but I plan to reconfigure the centre floor beam to ensure there is a direct support for the ridge beam load.

A lot of our neighbours to the south have the misconception that Canada is entombed by snow in the winter. This is only true for very limited portions of the country. Many parts of Canada receive very little precipitation. Saskatchewan is stuck in the middle of the continent, thousands of miles from any of the liquid source material. I am using a snow load of 1.25 kPa (~26 psf). For anyone interested, snow loads in Canada can be found here.

The proposed dimensions for the cabin are 16'x24'. I think a 3 ply 2x12 beam, supported at each end and the middle, will be more than enough.  Actually, if you use the Canadian beam tables and do some healthy amounts of extrapolation, I think you could get by with a 3 ply 2x10 beam. Of course the extrapolation should only be done by a qualified engineer (I don't consider myself a qualified structural engineer when it comes to residential wood construction, I only know enough to be dangerous).

[Erin]

Erin, I will investigate the 24' long lumber but that is not my first choice due to my access limitations (by boat). For the same reason, trusses are definitely not an option.

Am I understanding your floor beam situation correctly? Do you have a 5 foot cantilever on each end? I would be interested to know more about that situation.

Sorry!   No, that's where my basement walls are.  No cantilevers.
And our trusses were site-built.  We used standard dimensional lumber to create what is basically a timber-frame bent.

[North Sask]

Sorry!   No, that's where my basement walls are.  No cantilevers.
And our trusses were site-built.  We used standard dimensional lumber to create what is basically a timber-frame bent.

My bad, I assumed that you were using post and beam (if I am doing it, everyone must be, right?). 

Site built trusses would be a good option but I am stubborn in my desire for an open cathedral ceiling, in spite of all the associated challenges. Maybe I could consider site built scissor trusses but that sounds like a lot of work.

[Erin]

I AM doing post and beam...for the house itself. 
The foundation, OTOH, is a block basement. 

[Don_P]

Timberframers like to splice plates with impressively designed scarf joints off the post, and they have tested a number of designs to failure. It makes me nervous. I can wrap my head around the concepts but still have no desire to actually do that, hence my comment about "theoretical inflection point" I don't believe it performs in real life the way it does on paper. My guess is something between an overhanging beam supporting a point load on the end and their ideal continuous beam. Glulam constructed roofs of 3 or more spans use the overhanging beam approach fairly often and then hang a beam saddle on the cantilevered ends of the main beams to support a lighter center beam. The overhanging beams benefit from the  overhang lifting the outer spans a bit and the mid span is reduced so can be a lighter beam. These are treated as pinned connections though. If you can live with simply supported beams that would be my choice. If you want to weave something across the joint to help hold the world together I've sandwiched heavy flashing in between the plies of build up girders in the past. That puts your snow load at about the same as mine in the Virginia Blue Ridge, interesting.

[North Sask]

Erin, I really like the looks of what you are building! I assume you must have a thread in the owner-builder projects forum. Please pass along the title so I can take a closer look.

Don_P, thanks for all the info. I will have to give some further thought to the final ridge beam system. The snow load might be similar in our respective areas but I bet you are not fortunate enough to wake up to -40 degrees in January! 

[Erin]

Actually, I don't!  Just haven't gotten around to making one...