A friend wants an arbor over her patio for vines to climb on. We envision an open structure with four posts supporting two main crossbeams. These crossbeams would support "rafters" on two foot centers. One of the rafters will carry a porch swing, but that rafter is near the posts, not near the center of the span. No other significant loads other than the weight of the vines that will eventually cover the rafters. Main crossbeams would be 20 feet long, with 18 feet of clear span between the posts. Rafters will be 12 feet long, with 8 feet of clear span and 2 feet of overhang on each end.
My plan is to buy 3/4 plywood intended for sub-floor and cut it into strips 7 7/8" wide. These strips will glue and screw together with Titebond II, and the screws will serve as clamps until the glue dries. I'll stagger the joints by using 2, 4, 6 and 8 foot strips.
I plan to protect the lamintate beam from the weather by enclosing it in a 3-sided box of pressure-treated lumber.
Here's the question. How thick does the laminated beam need to be?
DonkeyHody "Every man is my superior in that I can learn from him" - Thomas Carlyle
Sun, Jul 1, 2007, 5:33pm (EDT-3) snipped-for-privacy@bellsouth.net (DonkeyHody) did query: I just know somebody here has experience . . . A friend wants an arbor over her patio One of the rafters will carry a porch swing, My plan is to buy 3/4 plywood intended for sub-floor and cut it into strips 7 7/8" wide. These strips will glue and screw together with Titebond II, and the screws will serve as clamps until the glue dries. I plan to protect the lamintate beam from the weather by enclosing it in a 3-sided box of pressure-treated lumber. Here's the question. =A0 How thick does the laminated beam need to be?
Yep, got plenty of experience. Whatcha want to know about?
But if you put the swing there, and later find out that's a bad place, it's gonna be a bitch moving it. I'd rather have a stand-alone swng, just in case.
Why screws? You figuring on taking it apart later? I'd nail it.
Well I knew a guy that did something like this. He took two 2X8's & routed out the center of them & glued in a plywood lamination like you speak of. Then glued up the whole thing so it looked like one beam. I can't remember how thick he made them, but I think it was three pieces of plywood. Since your going to do this anyway, why don't you glue up three and see how it feels? Then you'll know if it feels right. Also I would use Titebond III, it's a great glue for outdoors.
The screws are to draw the layers of plywood together, each acting as a small clamp. To lift it, I thought I'd stand one of the rafter boards on end and screw it alongside the post. Put an eye-bolt at the top and use a block and tackle.
DonkeyHody "In theory, theory and practice are the same, but in practice, they are not."
You really want to check out the cost of an engineered beam like Anthony's GlueLam ... they may be cheaper in the long run than your materials and labor.
These are similar to the ones I use:
formatting link
I just had an extra 18' one hauled off the other day (Habitat for Humanity))
My gut agrees. 1/2 the ply is going in the wrong direction to offer any strength. Of course the middle doesn't do much anyway; it's the top an bottom of the beam that are exposed to all of the force. 18' is a pretty big span.
I too think you should check with your local lumber yard about pre-made beams.
If I'm understanding correctly, your patio roof will be 18 feet long by
16 feet wide (not counting overhang on the ends), with two beams and four posts?
Typically, you would calculate the loads using 40 lbs per square foot, but since this is an open structure, I'd probably just use 10 lbs per foot for the dead weight of the materials (Keep in mind this doesn't account for any snow loads, weight of the vines, etc.).
So you would be looking at roughly 2880 pounds on the roof structure (18x16 x 10psf).
Half of that weight (1440 pounds) is carried by each beam.
According to the charts I have, you would need a MINIMUM 4x10 beam to support that weight over an 18 foot span. If it were me, I'd go with a
4x12 beam. 18 feet is a long distance to free span.
If possible, I would span the beams across the 16 foot distance, rather than the 18 foot distance. This would let you use 4x8 beams, though I'd still opt for the next size up, 4x10 beams.
Of course, adding additional beams and posts would make a huge difference in reducing the spans and the corresponding beam sizes.
Sounds expensive and a lot of work. In addition, for outdoor projects you really want as few joints and seams as possible. I'd go with solid beams, or a couple of 2x10's nailed together as a bare minimum. Plywood isn't really a good material for outdoor use.
20 foot long 2x12's would be fairly easy to set in place by yourself, though it would help if you could get another person to help balance things while you nail things together.
I'd also recommend metal post caps or hurricane straps to hold everything together. You don't want the whole structure coming down in a strong wind or earthquake. And, you'll need to sink the posts deeply in the ground and/or install some kind of diagonal lateral bracing to keep the whole structure from toppling over.
You might want to pick up a book on outdoor roof structures. I have one by Sunset books called "Patio Roofs and Gazebo's" that is pretty good (ISBN 0-376-01440-7).
Thanks to all for the advice. I'm re-thinking the plywood idea. The reason I considered laminates in the first place is that 20 foot lumber in any size bigger than 2X6 is pretty scarce around here. I'll price a factory glue-lam.
DonkeyHody "Every man is my superior in that I can learn from him." - Thomas Carlyle
SFWIW, right after WWII, materials were still difficult to get and expensive.
Some how, my dad managed to collect some kind of pallet that used 4x4 runners, about 12 ft long, that were probably "Ironwood" or something similar that required drilling pilot holes for every nail.
He used those runners for posts, then ran #9 galvanized baling wire overhead to provide a place for the vines.
That puppy lasted at least 20 years.
Might be an idea or two in there that you might find useful.
As an Architect, maybe I can add a little insight...
I would question the use of 'multiple' plywood laminations if for no other reason, the material strengths are indeterminate (the two main things needed for a rational structural design is 1) knowing accurately the loads one is to carry and 2) knowing accurately the ability of the material to carry those loads). While the wood generally used for plywood would have known properties, remember that plywood is generally used in an application (as in decking) that loads it in the plane of the laminations and in your case, you are loading it perpendicular to that. Couple that with the gaps that sometimes occurs in interior laminations (which is more critical in your application) and much uncertainty enters the equation - exactly what you don't want if a rational structural design is the goal.
Don't get me wrong, we're not talking some major critical issue here (it's not holding up a second floor of a house, etc) - but, it seems reasonable to not get into areas that a trained structural engineer would avoid, no? There are better ways, anyway.
One basic thing to remember in structural members is the *depth* of a member -- as this adds rigidity far more than width (i.e. a 2x12 is far better than a 3x8 or 4x6 (all with the same cross-sectional area)). It's a 'moment of inertia' (Ix) thing (you can google around
-- or just remember that the 'Ix' goes up with the square of the DEPTH and you can see the effect). That is - provided the width doesn't get too slender such that the beam can flex in the horizontal axis (as it will) - but that is what the rafters above will provide -- 'lateral support'.
Remember that a beam under load has tension stress in the bottom and compressive stress at the top and you can see that the wood fibers need to be oriented horizontally (which every other lamination in plywood would defeat that purpose - since the fibers would be oriented incorrectly) and you can see how solid, sawn wood would serve your purposes better.
I believe Lew mentioned a few 2x12s and I would agree that something like this would be a smarter solution. Also, the American Plywood Association
formatting link
has some pretty good tech reports on built up beams -- and don't get me (us) wrong - one *can* build up beams with plywood, but it is generally done with a plywood panel sandwiched between sawn lumber (or, two panels in layers 2 and 4 of a
5 member sandwich).
Then again, we also (now) have LVL (Laminated Veneer Lumber), which is just a big hunk of plywood (but with the laminations oriented correctly and with documented strength values) or LSL (Laminated Strand Lumber (similar). These are (generally) about twice as strong, size for size, as solid sawn lumber but are, of course, more expensive. Being an exterior application, you *should* use pressure- treated or naturally decay resistant wood. I would not use a LVL or LSL without capping the top surface with a conventional 1x pressure treated piece - just to keep water from working it's way between the laminations (or, cap it with some copper sheet - would make a nice design element -- just paint the covered surfaces with some oil-based paint and maybe use a few copper washers to space the sheet slightly away from the wood surface for ventilation (letting it dry between rains) -- but, then, this begins to complicate things more than it should be. Simpler is to simply use multiple 2x solid sawn members (and, maybe, that exterior plywood inner lamination) and cap with a
1x6 pressure-treated 'rain cap'.
Loading? There wouldn't be much - fundamentally the dead load of the materials and vines, etc. It is possible that wind-loading could be your greatest load - especially if covered completely with vines. I would double the rafters where it carries the swing and run your bolts from the top of the rafters (vs eye-bolts screwed into the bottom of the rafter) -- there are 'edge distance' issues with bolted connections and driving a lag-type eye-bolt into a 1.5" member (nominal 2x width) maybe cutting things a little close. Better to use a conventional threaded (i.e. machine threads) eye-bolt hanging from the top (between the two 2x rafters) - which would load the assembly in compression and not tension (i.e. pullout) - which the conventional lag-screw type eye-bolt will do. Since running a bolt right thru the interface between the 2 2x's might could cause delamination over time, smarter to run the bolt thru a smallish 'cap piece' on top of the
2-2x's (and, with a little overhang, this could be a decorative piece, as well) and this would not cause delamination.
That's probably enough to get some thinking going...
John, You've given me plenty to think about. While I can't argue with your structural analysis concerning the orientation of the wood fibers in plywood, my experience has been that plywood had less deflection when loaded "edgewise" than solid wood of equal dimensions. I deduced (perhaps incorrectly) that the advantages any grain defect being so thin and localized offset the obvious disadvantage of half the layers being oriented in the wrong direction. Or maybe I was simply wrong about it being stronger in the first place. Good food for thought in any case.
I failed to mention one of the reasons I was considering building my own beams. For esthetic reasons, we wanted the elevation of the rafters to equal the facia board on the house, even though they won't be attached to the facia. The house is slab-on-grade, with long eaves yeilding a fairly low roofline. While I recognize the strength advantage of a "tall" beam, headroom under it becomes an issue pretty fast. So . . . I was looking to substitute some width (got plenty of room for that) for some height. In other words, I wanted to make the beam wider instead of taller, while realizing the diminishing returns involved in doing so. I understand the tradeoffs, but I can't calculate them. All the tables I can find assume loads far higher than these beams will ever see. Maybe it's time to abandon wood and go to a steel beam. We could wrap it in pressure treated wood to hide it.
Thanks again for taking the time for such a detailed and thought- provoking answer.
DonkeyHody "In theory, theory and practice are the same, but in practice, they are not."
Pretty straight forward, you can't get there from here.
The key issue is "Moment of Inertia".
The formula to compute "Moment of Inertia" for a rectangle:
I=(b)*(h^3)/12 where b = base dimension & h = height dimension.
As you can see, "Moment of Inertia" is only changed linearly with a change in width while the "Moment of Inertia" is changed by the cube of the height dimension.
"Moment of Inertia" explains why wide flange and I-beams have the shapes they do.
Greatest strength for the least amount of material.
I guess it's a little late in the conversation to admit that I have an engineering degree. I understand the concepts well enough, but I don't have the resources to calculate the amount of deflection I can expect. Note that my load is lighter than usual and my amount of allowable deflection is greater than usual. In this case, if the deflection isn't noticable to a casual observer, it's acceptable, since there's no drywall to crack or doors to bind etc. etc. I was hoping somebody would chime in and say, "Oh, I've done that lots of times. Make your beam 6 inches wide and you'll be fine."
I've about come to the conclusion that, while I CAN get there from here, the young couple's budget won't support it. So we'll probably just add a center post, which will simplify life considerably, even though it's not an elegant solution.
DonkeyHody "In theory, theory and practice are the same, but in practice, they are not."
One idea no one has mentioned is to laminate the beam out of 2x4's laid flat. Stack them as high as needed so you end up with a beam 3-1/2" wide by how ever many 2x4's tall. Probably not cost effective, but possible. When you get it done you could plan it flat and smooth, as long as you have the tools and support to do it! Greg
"Don" (I'm just just this moniker for convenience)...
Well, your further elaboration of the conditions sheds more light on possible solutions (as does your engineering degree).
Height limitations always rear their ugly head in everyday design - and, when that is all you have to work with, well, that is all you have to work with. Shallower beams can be used - it just takes more material (as you obviously know) to accomplish the same result - i.e. less efficient, but then, as noted, that's life. (Lew is correct and I mistated about the Ix varying as the cube of the depth).
You might consider a smallish (i.e. low slope) truss construction - even a simple 'king-post' truss (i.e. 1 vertical strut connecting top & bottom chord at mid-span) would probably work in this case. This would provide considerable rigidity over a simple beam and even a very low slope (say, 2 feet of height at mid-span) should work and would lessen material needs - as well as getting the end "under" the fascia (as I understand the situation). That may not fit with your aesthetic desires, tho -- just have to look the situation over.
Just more thoughts.
My gut tells me this would be overkill. Just remember with this new ACQ pressure-treatment the corrosion issues with steel (over the older (and now banned for residential) CCA treatment). These concerns (with corrosion) are real.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.