Not making much since to me.
Most often this is what's done.
http://www.bing.com/images/search?q...raming+pictures&view=detail&id=3AF4FDE490EFCFD7DFE52E3F9AF7CDDDE31B03E8&first=1

Just how wide is this window?

 

joe: yes, that is a typical, single-beam header. my problem is that i can get 7" psl's real cheap, but a single one is not quite strong enough to carry the load i have. it was therefor suggested that i stack one beam on top of another, to (very approximately) simulate a taller, 9" or 11", single beam. i know a 9" tall beam will carry the load, and can just go buy one and be done w/ it. however, this has become an academic challenge to learn about stacking beams. i don't think two separate beams, stacked atop each other, will even approximate a taller manufactured beam because you likely can not really get the two to "become one" (borrowing from Grasshopper....). so, this is just an academic curiosity that i'd like to learn about. i have an engineering handbook that i am going to look at, but, not being an engineer, it is easy to get misguided by one of those handbooks. btw: the beam will need to be 6' 8" wide; nothing unusual.

 

You keep saying beam. I've never seen a single "beam" used in modern carpentry.
For a 2 X 4 wall it would be 2, 2 X 10's with 1/2 plywood or OSB sandwiched between them.
Whole lot less work then what your suggesting.

 

"beam" = psl, as i mentioned in the second post, or lvl, etc. engineered lumber is scores of gobs stronger than dimensional lumber, size for size. even double 2x12's would not carry my loads. in reality, dimensional lumber headers are fairly weak members, though often adequate, readily available, and relatively inexpensive.

 

really the best way to make the stacked header as one is to glue/screw/nail plywood to both sides of the stack and form a bonding structural connection between the two.if the two headers are 7" each then a 14" piece of plywood would be needed. that in turn leads to which type of plywood would be best?? even without plywood if the two headers sit tight to one another then the top header starting to deflect would have the resistance of the lower header pretty much immediately, the plywood helps to keep any outward kickout force from the separated beam from occuring. the downward force on the header can turn sideways(twist) if there is no resistance to keep it from doing so.

 

really the best way to make the stacked header as one is to glue/screw/nail plywood... •• Yes. Glue, screw, subdue and tatoo; do everything you can. I still wonder if the plywood does much, though, as it is not very wide (2 1/4" max, and then you may as well buy a bigger beam) and still alternate lams are in the wrong orientation.

.... if the two headers sit tight to one another then the top header starting to deflect would have the resistance of the lower header pretty much immediately... •• Correct, but I still think you are getting primarily a point load on the lower beam. That is what it FEELS like to me, anyway. Wish I knew exactly.

Thanks. See after bullets.

 

I have no input for you jklingel but just thought some clarification would get you more answers for your questions.

You have a 6'8" span and a beam comprised of 2 2X12's isn't strong enough therefore the idea for stacked beams, is that correct?

 

I have no input for you jklingel but just thought some clarification would get you more answers for your questions. ••* OK. Send money anyway.

You have a 6'8" span and a beam comprised of 2 2X12's isn't strong enough therefore the idea for stacked beams, is that correct? •• Correct. Neither the the double 2x12's nor a single PSL is quite enough.

Thanks for the reply. See after bullets.

 

My understanding is if the two stacked beams are not fastened/glued or otherwise "stuck" together so that they are working as one, then it is not as simple as 1 + 1 = 2. If you place these two beams side by side, then yes it would be 1 + 1 = 2.

Regarding stacking: You would need to calucate the shear forces inbetween the two beams when placed under load, and compare that to the actual shear resistance of the method you use to fasten the two beams together.

If you do not fasten the beams together, the only shear resistance you are getting is from the friction between the two beams... which will ultimately give you a higher strength beam, but not one that is double in strength.


*Disclaimer: Have an engineering degree but do not practice in structural design. Hopefully someone who does can give you a 100% certain answer.

 

My understanding is if the two stacked beams are not fastened/glued or otherwise "stuck" together so that they are working as one, then it is not as simple as 1 + 1 = 2. •• Agreed. Two stacked are not even close to one manufactured of that height, and my feeling is that the second one is adding about 12% (1/8), at least initially. If 1+1=2, then they'd only manufacture one thickness and tell you to stack what you need. Also, the Karate Kid would never cement all his blocks together and THEN break them; never happen.

Regarding stacking: You would need to calucate the shear forces inbetween the two beams when placed under load, and compare that to the actual shear resistance of the method you use to fasten the two beams together. •• Exactly, which is why I question just nailing and/or screwing. Gluing two stacked, if you could do as good as the manufactured stuff, would likely be pretty good.

If you do not fasten the beams together, the only shear resistance you are getting is from the friction between the two beams... which will ultimately give you a higher strength beam, but not one that is double in strength. •• Again, just a gut feeling, but I'd sure not want to rely on that friction adding much.


** My disclaimer is that I am NOT an engineer.

Roger that. See after bullets. Thanks for the reply. I am going to try to post on an engineering forum; just busy as h3ll building right now..... and pushing snow.

 

That reminds of me of a house I sold years ago. Original load bearing kitchen wall was cut out (balloon frame wall) with second floor joists suspended by 1x6 siding, to put a bathroom addition.

Stacked header, I am guessing, can work, but you are guessing on the load capacity of such a header. Top header will start to bend, taking the bottom header with it, until some balance is reached, or both fail.

Splicing two pieces with plywood will not work. Plywood is weaker than header materials, unless the whole thing is engineered.

 

That reminds of me of a house I sold years ago. Original load bearing kitchen wall was cut out (balloon frame wall) with second floor joists suspended by 1x6 siding, to put a bathroom addition. •• Wonderful engineering on someone's part!

Stacked header, I am guessing, can work, but you are guessing on the load capacity of such a header. Top header will start to bend, taking the bottom header with it, until some balance is reached, or both fail. •• That is what I am feeling, too.

Splicing two pieces with plywood will not work. Plywood is weaker than header materials... •• Typical plywood, yes.

See after bullets. Thanks.

 

Keep in mind that the existing "header/lintel" is already loaded and deflected due to the existing dead loads. Unless the loads are temporarily releived, the straight member under will just pick up the new load and will deflect. The question is whether you are concerned with strength or deflection mainly?

That is why many structures a jacked slightly before a reinforcement is added since the load is eliminated and the existing deflection minimized.

Dick

 

Keep in mind that the existing "header/lintel" is already loaded and deflected due to the existing dead loads. ... •• New construction.

The question is whether you are concerned with strength or deflection mainly? •• Have to consider both; window under header, so deflection will kill it.

Dick

See after bullets. Thanks.

 

The topic of stacking beams has been discussed several times on this forum over the past few years. I wrote a lengthy message discussing the concept of horizontal shear and how that relates to the strength of stacked beams, you may want to do a search if you are really interested in this topic.

The essential fact is that two beams stacked on top of each other WITH NO means of connection are stronger than either beam alone, but not by much, and the exact amount is difficult to calculate. If the beams are properly connected using fasteners, the composite beam acts exactly the same way that a single beam of the same dimensions would have, in other words two beams which are 3.5 inches square (4x4's) joined together vertically have the same strength as a beam 3.5 inches wide and 7 inches deep, provided the fasteners used to connect the beams can carry the horizontal shear developed due to beam loading. Determination of the required fastener schedule is tricky, and beyond a DIY discussion.

 

The topic of stacking beams has been discussed several times on this forum over the past few years. I wrote a lengthy message discussing the concept of horizontal shear .....

Duh. I should have figured that. I will search, and I thank you.

 

From a pure structural standpoint, it does not take much of a plywood layer on each side of separate beams/lintels that are stacked from a flexural and deflection standpoint if you have enough screws/nails to transfer the shear between the units and create a composite beam with as much as 4x the engineering properties of a single beam/lintel. It works well if sources of materials are limited and can be very effective in reinforcing existing structures if loads are relieved before the veneer attachment.

We did similar tricks when reinforcing existing rocket test stands (no time or room for major mods with new more powerful engines) where channels were commonly used as beams and we emptied the fuel tanks to lessen the gravity loads during installation of plates and create the strength that was needed for firing when the upward load were critical after the fuel tanks emptied. - Same principal, different materials, different methods (welding sheets instead or nailing wood) and opposite load directions.

Engineering principals work in almost every structural application.

Dick

 

From a pure structural standpoint, it does not take much of a plywood layer on each side of separate beams/lintels that are stacked from a flexural and deflection standpoint if you have enough screws/nails ....

Like many things in life, that seems almost counter-intuitive. Aren't you still relying on the plywood to convey the loads, and isn't plywood pretty weak, relatively speaking, for this type of loading? Working on those rocket test stands must have been a blast!:wink:

 

If it's for a group of windows depending on the look your going for you can put 2x4's between the windows to help carry the load.

For large open spans replace the ply wood or OSB in the beam with a steel plate. It would be hard to hide the lag bolts for like a window but can be done with for thought.

But when it comes down to it I would talk to an engineer.

 

^ the screw/ nail schedule is an important factor in regards to plywood being used as a connector along with thickness, 5/8" being the minimum and ideal size in some cases.

 

Dan H: I have searched at length, by your name and various permutations of "beam blah blah". I am old and running out of time....:laughing:.... any idea of the title or a key word in the post where you laid out horizontal shear? thanks. i'd love to read it.

 

Found Daniel's discussion of horizontal shear, finally! See http://www.diychatroom.com/f19/strengthening-floor-joist-114616/

Now, Daniel, where do I go to learn the intimacies of ensuring that I have properly addresses horizontal shear when stacking two beams? Engineering school? Surely some wood organization lists general rules of thumb for nail size and pattern, plywood thickness, etc.... any ideas? thanks.

 

Stacking beams is relatively uncommon, and the horizontal shear computations are usually done individually for each installation. The reason they are usually done custom is because the horizontal shear that develops between two components depends on the load and the geometry of each piece. You can find the formula in any standard mechanics of material textbook, however actual application to a real problem is a bit harder than just plugging in numbers.

Once you know the horizontal shear, you select the fastener spacing to provide an adequate safety margin, usually 100 percent, since fasteners are cheap and failure is a bad idea. In other words, you compute the maximum allowable spacing between fasteners, then use half that, so if the allowable spacing were 12 inches OC, you would use fasteners 6 inches OC. The allowable shear per fastener is typically obtained from Wood Products Laboratory or similar white papers on the subject.

As I said, this method of strengthening beams is relatively uncommon, and so far as I know there are no "rules of thumb" similar to the carpenter rules that have been developed to size headers.

 

Once you know the horizontal shear.

In simple terms, how would we calculate the horizontal shear between two 4x2s laid flat, one on top of the other, and uniformly loaded?

 

OK. Understood. I did some Googling last night and it lead me to believe that this was more engineer territory. So, I've learned what I needed to and will just buy bigger beams. Thanks for the info; it is always interesting reading. I did learn that it is permissible to rip PSL's down in width (to 1 3/4"), which is helpful linearly, as you mentioned earlier.

 

The formula is t = VQ/I where t is the shear flow at the neutral axis, V is the maximum vertical shear in lbs, Q is the statical moment of the beam (for rectangular beams Q = Ay' where A is the area of the beam above the neutral axis, and y' is the distance from the neutral axis to the centroid of the area above the neutral axis), and I is the moment of inertia of the section.

So let's take your example. A pair of 2x4's laid flat has dimensions of 3.5 inches wide x 3 inches deep. If the 2x4's are 10 feet long, and are loaded with 20 psf, the total load is 200 lbs, and the maximum vertical shear is 100 lbs, which occurs at either end. For such a beam, Q = 3.9375 in^3. Moment of inertia I = 7.875 in^4. Shear flow at the neutral axis at the ends = 50 lbs/in.

This means that at the ends, the boards need to be nailed together to develop 600 lbs/ft shear capacity (12 inches x 50 lbs/in). A 16d nail has an allowable shear of 96 lbs, so you would need 6.25 nails per foot near the ends. If you use two rows, you would need to space the nails 12/3.125 = 3.84 inches apart, in practice you would use 3.5 inch spacing, two rows of nails. Towards the midspan of the boards you need fewer nails since the vertical shear is lower towards the midpoint (it is zero at the midpoint of the span).

This example shows why it is rare to connect flat boards this way. Far simpler to nail the 2x4's together oriented vertically and let the wood handle the horizontal shear stress. For boards oriented vertically, horizontal shear is irrelevant, and you only need enough nails to hold the two pieces together sufficiently to carry the composite member around.

 

Many thanks, DH. That filled in a little gap for me. (literally!).

 

good work, holzman. i just bid on that last job you we asked to, but i bid half of what you did. "have equations, will engineer." :laughing:

 

on a serious note... if i followed daniel's explanation, i ran the numbers for the original question, stacking a 3.5" W x 7.25" H beam atop another one. i have a tributary width of 23.5' and a roof psf of 65, giving a load of 1527 plf. for a beam of 6.7', V = (1527)(6.7)/2 = 5115 lbs. Q = (3.5)(7.25)(3.63) = 92.1. I =((3.5)(7.25)^3)/12 = 111.1

thus, t = (5115)(92.1)/111.1 = 4240 lbs/inch, which => forget nailing anything together to obstruct such a force. did i run the numbers correctly? thanks.

 

I refers to the moment of inertia of the composite system. In this case, you have two beams, each 7.25 inches deep, so the total depth is 14.5 inches. Therefore I = 889 in^4, not the 111 you computed. The maximum shear flow comes out to be about 529 lbs/in, which is probably too much for nails, but could be handled by bolts.

This member is carrying a lot of load, if you are seriously thinking about installing something like this, you should probably be thinking about a steel beam, not a composite wood beam.

 

Daniel: Thanks for catching my error and your suggestion of using steel. Steel is a last resort because of the thermal conductivity in our 14,000 heating degree days. A 3 1/2" x 9 1/2" Versalam LVL was engineer-approved for a 7' 8" span, using a 60 psf figure. However, that beam barely fails using the city code of 65 psf, so I will add a 2x10 and 1/2" plywood. There will be double 2x6 doug fir trimmers. Thus, the same "beam" on a 6' 8" span should be good. In 39 years, I have never seen a live load exceed 36 psf (by my crude method of cutting a 1' square column of snow, bagging and weighing it), so the 65 psf (50 live) is pretty extreme. Besides, when the snow starts to pile around 3', we generally get out and shovel roofs, just for kicks. Thankfully, that is rare. Again, I thank you for all the education. I've many times wondered if I should have gone into engineering instead of numbers and Greek symbols.