scarf joint or full length beam

[paisan8]

im jumping into timber framing for the first time. its a 22 by 14 ft detached porch with a shed roof. the 22 ft sides have a beam and three posts. i see i can order a 10 x6 beam long enough for the 22ft span. im not sure if i should join 2 shorter beams or just get one 22ft for the whole way. how straight can i expect a douglas fur beam this long to be? how difficult is the scarf joint to get right?

cheers
james

[canopy]

If you have to ask then use a full length beam. If it is not perfectly straight then just lay it out using snap line square rule. Scarf joints are an elaborate, time consuming joint with lots of things that can go wrong. The sky is the limit on how technical they can be. A good scarf joint is quite long, stop-splayed, undersquinted, cogged, wedged and pinned. You also have to place the scarf at the right position along the beam which is never going to be over that center post.

[Brad_bb]

Depends on what you want.  A full length beam would save a lot of time and work of fitting a scarf.  Being a first timber at the scarf, it would be even longer for you.  I'd say it'll save a lot more time doing a fill length beam, and the extra cost for the length shouldn't be too bad.  22 feet is not too bad.  6x though?  I'd say go 8X.  Better to be bigger than to look a the porch and think it looks too skinny for the length.
Scarf joints are better left when you have a little more experience under your belt, unless you really want the challenge.

[Cosmo52]

 

 

 


Scarf joints are not hard to do just takes a lot of time and risk of an expensive piece of wood if you make a mistake (take your time).  The trouble that I had was when the material was removed for the scarf (30" overlap) the tension in the wood was no longer equal on both sides of the heart in the scarf area.  My previously straight 8 by 12 timber has a significant twist near/in the joint.  My building has 3 scarf joints and 2 of the timbers have some twist.  If you can source the timbers in the lengths that you want, you can reduce some risk in making the joint.  The other issue that I had is that as the timber dried some checking occurred along the middle of the timber (heart line) where the cut was stopped.  The engineer for my project specked all thread screws to fortify the joint.  

[Don P]

Engineering a simple or continuous beam is pretty straightforward. I'm curious as to the engineering of scarfs but thus far haven't seen an explanation from a timberframer I would feel comfortable using.

[Brad_bb]

Don, I'm not sure what you mean.   Engineer a continuous beam?  We're talking about milling a continuous beam.  Scarf joints to make longer beams have been around for...hundreds if not thousands of years.  Where you locate a scarf joint is the important part.  It just can't be anywhere.  It can't be in a location where it's going to be under significant bending stress or shear, but in an area of very low bending stress/shear.  Top plates are the most common place for scarfs, also a ridge beam could have a scarf.  Here you see a top plate and the scarf is placed so that one side is supported by the post and the center of the scarf is over the brace.


 

[Don P]

Remember, the point of lowest bending moment is the exact point of highest shear.

The scarfed beam is certainly not a continuous beam, that is a solid one piece beam over 3 or more supports. What do you use for design values?
Which beam equations are most appropriate, I've seen the continuous beam analogy made here and elsewhere many times but that is unconservative to put it mildly.

Edit; Start here for some standard beam equations, the last page has continuous beams, top graphic in each equation is loading, middle graphic is shear, bottom is bending moment, notice the relationship between shear and moment.
http://faculty-legacy.arch.tamu.edu/anichols/index_files/courses/arch331/NS8-2beamdiagrams.pdf

Notice a couple of pages before that the equation for beam overhanging support, that would be one conservative solution.

This gets into why I have no desire and is also why the building code says what it does, make a long beam or multiple simple beams. If you want to do a scarf I'd leave that to the engineers to design or specify.

[Don P]

This is in the interest of understanding what we are doing rather than just assembling parts, which is prone to failure. When I challenge a viewpoint it is usually because I either want to change your understanding of something or adopt it .

Often in the field a contractor will say "we've always done it this way" when I ask why they are doing something questionable. My thoughts go to "a lawyer calls this a class action". In the responses to the op I saw the same way of viewing these joints that I believe to be unconservative. I dislike the "get an engineer " approach if it also carries with it an ignorance of what it is that the engineer is thinking and what we are doing. It leads to my contractor friend above, the repeating of something without understanding it, possibly out of context, certainly without understanding. I've also found even with the engineer doing the heavy thinking the adage "trust but verify" is in ones best interest. That does not mean my understanding is correct, simply that it is considered.

Consider this,
This is a table of cantilever beams of various numbers of spans. The cantilever distance from the support is the bottom section of rows of the table, roughly the moment inflection points. This is I believe a more conservative way of thinking about these joints. Rather than the continuous beam analogy. Study these diagrams and think about the beams more along these lines. One to compare is example 39 in the link above, a 4 span uniformly loaded continuous beam and the forces around the different joint options in the 4 span example below. Notice the higher moments all depending. Typically when scarfs are broken they approach the strength of a solid timber, when well made and when new. That is not the same as equal to a solid timber. Thinking of them as cantilevers with pinned joints is more conservative. If you achieve better joinery, that is gravy. In other words I think the truth probably lies between these two examples.




To compare the two diagrams, this one has the coefficients beneath the table, notice in the 4 span example where the joints are located, they give a couple of layouts.
 the moments for the 4 span cantilever above, .086 between posts(beam overhanging one post), .086 over post 2, .051 Hanging beam), .063, .063, .063 (beam overhanging 2 posts), .096, (end propped cantilever) 

Compare to the continuous beam, .0772(end span), .1071 (over post 2), .0364(center spans), .0714(over center post), .0364(center span), .1071(over post 4), .0772 (end span)

You can see the cantilevered pinned joint example is calling for heavier material and more robust joinery than when it is viewed as a continuous beam in order to stay out of distress.

The options that avoid this and would also meet building code would be to use a full length beam or simple beams that break over the posts.

[paisan8]

thank you for the replys everyone. continuous beam seems like the way to go and will also cut my stress level in half. i just lose out on the aesthetic value of the joint and bragging rights for having done it.

jim

[apratti]

 22' beams milled at 12x8? Where I live, we don't have an abundant supply of trees for this option. So I began practicing scarfs on small 6x6s. I sucked at it. I am not going to be good at them soon. Maybe a solid beam top plate of 8x8?
Mario

[Don P]

If the design requires an 8x12 then dropping to an 8x8 wouldn't be a good thing. If there is a post at midspan then 2 simply supported beams that break over the posts is one solution. This is more typical, it predates the scarf by thousands of years, we are back to stone construction, rather than the 500 or so years that the scarf has been in use. The scarf in modern times is used more often for the reason paisan8 mentioned than it really needs to be.

A scarf develops moment capacity, the ability to resist rotation, or bending, a very good thing. Various forms of that joint have varying levels of resistance to bending. It can also create splitting forces if placed at a point of high moment. Another thing to remember is that at the point where bending moment passes through zero, shear forces are the highest, look at those beam diagrams.

In a roof the loading is generally pretty uniform which helps and is why you see it there. The loads are lower than in a typical floor and are generally more uniform which means the inflection point is more likely in a relatively fixed place. With a floor the loading can move that point of contraflexure which can add greater moment to the joint. One thing to consider, blindly placing that joint over a brace is mentally saying "I don't really trust this joint, here is my backup" The builder did not figure out the inflection point, it is a bit of a cheat. It is generally close though, but again understand what you are doing. If that brace becomes the joint support how is it different than splicing over a post? That brace becomes the point of high moment directly under the joint. In floors and under heavy loads such as heavy timber mill construction the beams usually break over posts or on bolsters over posts. Now, look at pictures of old scarfs. Are those seasoning checks or horizontal shears? I'm simply making points to consider, there are many paths.

[Crusarius]

Use a steel beam then box it in with 1x's and a scarf joint. best of all worlds

[Don P]

There's one in every crowd :D
Yup, many paths.
Last week this came back for a roof support repair, it shows one way of how that detailing might look;

[Crusarius]

There's one in every crowd

Thanks for noticing

[Don P]

Crow doesn't age well, enjoy it while its hot  :D
I was digging in the AITC (Am Inst of Timber Construction) manual last night, another code reference. Check these beam diagrams out;
1st the simple beam;




Next a 2 span continuous beam, notice shear passes thru zero at the point of max moment in bending BUT shear is actually highest over the midpost NOT where moment passes thru zero. At that point shear is no worse or not much worse than the ends.



  

Then a 2 span cantilever, Very similar looking at first blush., notice the hinge connection to the right of the midpost. I haven't delved into it deeply but this doesn't look bad;





This is the connection they have in mind;
Hmm not taking it, continued next post

 

[Don P]

 



There we go. That connection is a hinge, no moment capacity. This is typical glulam heavy timber construction.

At least for the way my brain works I think it would be worth running a scenario or two through all of those equations to see how they compare. Anyway, off to the day but this is more to look at and ponder. To be continued...

[paisan8]

Use a steel beam then box it in with 1x's and a scarf joint. best of all worlds

im just gonna draw a scarf on the continuous beam

[apratti]

That is honestly pretty funny man

[Crusarius]

Use a steel beam then box it in with 1x's and a scarf joint. best of all worlds

im just gonna draw a scarf on the continuous beam

AWESOME!!!

[Don P]

Here we go if anyone wants to play with some comparisons. The equations from the post above put into a calc. Pretty quick and dirty but it gives some ability to plug in and run scenarios to see how the various beams perform. The cantilever is not a true scarf jointed beam, I'd like to find something concrete there, it is assuming a connection like the one posted above. These come from the glulam folks, heavy timber post and beam framing.
http://timbertoolbox.com/Calcs/continuousbeam.htm

[Brad_bb]

@paisan8 still needs to put his location in his profile so it shows up in posts.  That way we quickly know what wood he has access to based on where he is.

[paisan8]

@paisan8 still needs to put his location in his profile so it shows up in posts.  That way we quickly know what wood he has access to based on where he is.

done

[Don P]

Slowly digging through the bookshelf on cantilever beam design. I came across a couple of good worked examples for glulam this morning. It has deflection equations. This is from "Design of Wood Structures, 3rd" Donald Breyer.

From the text, and I've seen some of this before elsewhere, probably in the AITC Timber Construction Manual;

Cantilever beam systems have an internal hinge connection are often used in glulam construction. The reason for this is that a smaller size beam can generally be used with a cantilever system compared with a series of simply supported beams. The cantilever length L_c in the cantilever beam system is an important variable.
...
Cantilever beam systems are not recommended for floors. Proper cambering is difficult and cantilever beam systems in floors may transmit vibrations from one span to another.
...
The case of unbalanced live loads can complicate the design of cantilever beam systems. This is particularly true if deflections are considered. When unbalanced live loads are required, the optimum cantilever span length L_c will be different from those established for the same uniform load on all spans.

In a cantilever system the compression side of the member is not always on the top of the beam. This will require a lateral stability analysis of bending stresses even though the top of the girder may be connected to the horizontal diaphragm.

That last part would apply to a continuous beam as well. When the beam crosses a midspan post and the moment reverses the underside of the beam is the compression face that wants to lay over. The unbraced length there to be checked would be between the inflection point and post. When we get engineering back on continuous LVL beams over multiple posts it usually (always that I recall) calls out lateral beam bracing at each post.

Anyway, just more to think about on the topic.