Structural Framing Question

[Bull]

This is something I have wondered for a while.

Let's say you have some dimensional lumber across a span. It's obviously supported on each end, and unsupported across the length of the span. If you want to sister another piece of dimensional to it to increase strength, do you have to use a single piece, or can you use smaller pieces butted together? Also, do you need to sister the new wood along the entire length of the original beam, or can you just beef up the middle, where the most strain is? Finally (for now) does the new beam have to be the same size as the original? In other words, is it bad practice to sister a 2x4 onto a 2x6 in an effort to strengthen it?

Looking forward to the answers...I am not in possession of a good knowledge of either physics or geometry.

 

[Motofixxer]

Well I would say speaking just from a common sense standpoint. Any joint is a weak point. So if you have multiple joints then it clearly will not be as strong as no joints. Also the larger the lumber the stronger it is. So if it were me I would put the largest piece reasonably possible. But it depends on the application and the need. If it were a real load bearing nothing really beats steel or an LVL\Microlam. But anything extra added is going to help.

 

[Daedalus]

I assume you want to beef up the stiffness of the joist to reduce bending, rather than shear. The short answer is yes, you can use shorter pieces of reinforcement to provide localized stiffening. If you were to stiffen only a portion of the joist near the center with a theoretical "infinitely stiff" sister piece, you could remove all bending along that length. However, there would be no benefit to bending in the rest of the joist not reinforced.

Using a smaller-cross-section piece of lumber to reinforce a joist is poor practice because of the math behind it. Stiffness in a rectangular beam is derived from 1/12 * b * h^3, where b is the base of the rectangular cross-section and h is the height. Note the exponential relationship. For a 2x4, I = 5.36/in^2. For a 2x6, I = 20.8/in^2.

So using a 2x4 to stiffen a 2x6 only increases the stiffness by 26% along its span, even though it's 64% the volume (and weight and cost).

 

[Torque1st]

Bull, just sister the 2x6 with another 2x6 for the entire length. Do not use short sections.

 

[Bull]

Ok, there is a lot of science to this that I am just not going to understand, like that formula up there

I have most often wondered this with respect to ceiling rafters and floor joists, of course.

In the current case, I have gutted a bathroom that has a sloped ceiling...it's an old Craftsman style house. The rafters in this section are, as I recall, 2x5s. Undersized, IMO. I want to bolster them to help eliminate any future sagging, and for plain old peace of mind. I did this in another room that I renovated in the spring. In used full-length beams, which was a PITA to get them in to the house and upstairs, working solo. If I can even just use two pieces and **** them together, it would make the process easier on me.

Also, when you are sistering, do you need to make sure that the ends of the new piece continue to the framing/post that supports the original piece? In other words, if you have a beam spanning two vertical posts, does your sister need to begin above the first post and terminate above the second, or can it terminate several inches short on both ends? How much strength is lost this way? I ask because these rafters terminate in angles. Having to duplicate the angles on the sister takes some work for me. If I am not gaining much by doing it, then I wont. If it's crucial/good practice, I will do it.

Thanks for the help so far, guys.

 

[Torque1st]

Is there some sagging problem? If so some jacking to straighten the rafter may be needed to start. If you have a real structural problem you may need a good carpenter to check it out.

A 2x6 is only 1-1/2 x 5-1/2".

Use a full length piece if possible, no short pieces.

It is best to have the sister beam supported on both ends but I understand your problem. I would use as long a piece as you can get in there. Use some liquid nails construction adhesive along the rafter like:
|\|\|\|\|\|\ -You get the idea. Put some big C-clamps on it every foot or so then nail it real good driving many of the nails in at a slight angle.

 

[Bull]

What you describe with the glue/clamps/angled nails/long pieces is what I did the last time. I guess I wont cut any corners this time, either. I think that or one or two of the rafters, I even made a plywood sandwich (rafter, 1/2ply, rafter), but that is probably overkill for this.

The house is 75 years old...there is some slight sag in the roofline in general. But, I am not going to worry about jacking up this little section...the sag is not severe. I just figure that my hope is to have these rooms ripped apart once in my lifetime, so while everything is exposed I'd like to strengthen/repair as needed. Also, these rafters are smaller than a modern 2x6 as I recall...I will measure them and be specific about their dimensions. I thought that 75 years ago, a 2x6 was really 2x6, too...?

 

[Torque1st]

I thought that 75 years ago, a 2x6 was really 2x6, too...?

Somewhere along the line they started reducing the dimensions but I am not sure 75 years old is old enough to get back to before that time.

Good luck with the project. It sounds like you are on track. I have got to hit the sack.

 

[Kevin54]

There is a difference between strength and straightness. Look at a lot of door jambs and trim. It is made from shorter pieces of lumber that has been finger jointed. This will keep the length straighter than one long continuous piece. Reason being is that short pieces have more strength and less flex than one long piece. Look at laminated beams. They are various short pieces laminated together to make one long beam.

Bull.....in your case, I would dampen the wood down. Either by hand with a sponge, towel or something. Then I would take the bow out by jacking it up slightly. A little bit over a day or two, then sister a 2x6 or 2x8 to it. Or for strength, you could glue and screw some 3/4 OSB to each side. OSB is supposed to be structurally stronger than dimensional lumber as it has the strands running in all different direction vs. a 2x that the grain runs in one continuous direction lengthways.

 

[galwaytt]

...there is another way, which is something I only came across a couple of years ago, as it has appeared to have fallen out of practice, here: flitch plates.

Basically, using a flat steel plate, bolted through a lintel etc, to increase it's capacity, but without introducing huge volume - e.g. where space is tight, or causing you to have lintels wider than the supporting walls (where located axially)....or without introducing the need for requiring more cripples at the end of the beam.

For what you're doing - the plate could be very light. Indeed, you could use sheet material, esp if you had a shape folded into it - e.g. a hem.

If someone on here can help, great, otherwise, Google's your friend.


http://www.strucalc.com/engineering-resources/flitch-beam-bolting/


Indeed, I see a comment in THIS that might be useful.

 

[OccupantRJ]

3/4" X 6" pieces of plywood cut with the grain running the long direction, glued, clamped, and screwed along the length of the rafter as far as possible or practical. Strong, light, and can be bowed to help get into place easier. Jacking optional, depends on the sag. Get the plywood with the most lamination layers.

 

[rieferman]

a bevel gauge might help you with the angles at the very ends. If you have it ripped apart, you might as well take the sister to the edge I think.

As for butting smaller pieces together... Each place where they **** together, if you again cover that joint (so, sister on top of the sister) you will reduce the weakness of the joint. For your application (one where it doesn't sound like a jacuzzi will be immediately on top of the area you're strengthening) this may be more than enough.

I like to through bolt in a few places when sistering in addition to using ample amounts of ringed nails driven in at an angle. I use 1/2 threaded rod, lock washer and flat washer.. This way I can make it perfect length and sawzall off the ends. Plus it's much cheaper than buying carriage bolts of similar length.

And I'd jack it up while you're doing it. If the sag is slight I wouldn't expect that you'd need to do the little by little approach. You could just take care of it all at once. Of course, listen to the wood and use common sense. Stop if you hear alarming amounts of popping etc. A post jack from HD or Lowes (or similar) will set you back about $40 and will be useful for countless things over the years. Honestly, it's not as scary as is sounds.

Of course, this is all from a pencil pusher not a pro etc. So take the advice for what you paid for it

 

[69lm69gp]

Since it was mentioned twice, I though I would pass on some information I got from my contractor while I was having my new garage build (build thread to follow).

LVL/Microlam's are not as strong as solid wood of the same dimension. LVL's are used where getting a single length of solid wood would be impossible or cost prohibitive.

 

[CraigFL]

In general, the maximum bending moment and deflection is at the center of the beam. This means that if you strengthen the beam in the middle 50%, it would probably be significant in your case.

If you think about a joint in the stiffener, the cross section of the beam becomes the strength of the original beam without the stiffener no matter how "tight" you make the **** joint because the stress cannot "jump" across this joint. This results in a stress concentration there that may or may not cause problems so it should be avoided. If the change in section occurs at the quarter points-- a strengthener beam 1/2 the length of the original beam--the bending(moment) is usually smaller enough that it is not so significant.

It is also interesting to note that beams can be strengthened much more efficiently by adding material to the bottom of the beam. Sistering a 2X6 to another 2X6 doubles the stiffness while adding a 2X4 to the bottom of the 2X6 can more than triple it. The compromise is that the height of the ceiling gets lowered.

 

[Stuart in MN]

Another method I've read about to stiffen joists is to fasten a piece of metal strapping along the bottom edge. I found a discussion about it here, along with some other interesting information: http://www.eng-tips.com/viewthread.cfm?qid=158165&page=1

 

[bimmer1980]

CraigFL is hitting on a critical point. The "bending moment" is the critical item here. If you look at the engineering equations and the bending moment diagram, the bending moment is at it's highest at the middle of the beam. So yes, if you sistered a 2x6 in the middle of the beam, it would help out on the deflection. (i.e beam is 10 foot total length and a 6' board is sistered in the middle--2' on either end would not have any additional boards) It will not add any shear capacity to the whole beam because it is not supported at the ends. So it all you are trying to do is remove the sag, then a shorter piece that is sistered in the middle will help. (please note--this is not perfect, but if you struggling getting the long length up there, this will help.)

Make sure that there are no joints in the middle of the beam.

The steel plate idea is also good. The key to this is to make sure that it is securely bolted to the beam. It has to be constrained so that it cannot buckle.

Finally, there is no substitution for the depth of the beam. You are better off going from a 2x6 to a 2x8 or 2x10 than trying to double or triple a 2x6.

 

[StingRay]

When looking at the idea of joint in the sistered portion of the beam think of it as the loads are really tension and compression. Most of the strength of any horizontal beam is in the very top and bottom (flanges). If loaded from the top the material along the top of the beam is in compression or is being pushed together. In bending a avery tight joint here might (don't take this literally as there other considerations) be acceptable just as joints in a brick arch are. But along the bottom the stress is tension and could be likened to to just pulling on a rope. Two pieces of rope end to end aren't going to do much good unless you tie a knot in them.
If you look at any truss or any beam like an I beam you see the flanges along the top and bottom which are taking most of the tension and compression and a web in between that is for the most part just seperating the top and bottom. The further apart the top and bottom are the better the beam will be in bending up to the point that the tension and compression members are no longer strong enough or the beam fails in another mode. Look also at a wooden I joist and you will see just a wood I beam. The wood along the top in compression and on the bottom in tension. A taller Ijoist being able to span a larger distance having better stiffness for comfort factor in a home. A standard 2 x 10, 2 x 12 etc is no different. The wood at the top and bottom is really the bending strength and the wood in between is seperating. Cut even a small notch along the bottom and you have cut the rope. Your 2 x 10 is now a 2 x 8 and it has a stress riser to boot. Notch the top and you have a gap that compression is trying to close and you have a stress riser again. Not good. Drill a hole in the middle where forces are minimal and there is no problem. The closer to the middle or neautral axis the better. If you look at any stucture that is timber framed you will often note that there is a crack running the length of the timber. Above the crack the wood is pushing together and below it's pulling apart. Right at the crack is the neutral axis where there are no compression or tension forces. There are alot of other considerations in a beam besides bending. Examples are shear and lateral deflection so just taller is not going to be useful to the layman. Don't take this kind of information here and use it out of context.

I know it's not the exact topic but Look at a roof truss and it's a little more complicated. You still have tension and compression members. Any cuts, holes or weaknesses introduced are not a good thing either.

 

[Jack90210]

Since it was mentioned twice, I though I would pass on some information I got from my contractor while I was having my new garage build (build thread to follow).

LVL/Microlam's are not as strong as solid wood of the same dimension. LVL's are used where getting a single length of solid wood would be impossible or cost prohibitive.

I don't believe that this information is correct, and that your contractor needs to look more carefully at the specs for the materials. Look at this pdf table from the UMass construction website. Note that for every application you need significantly more thickness for Douglas Fir lumber than you do for the LVL product -- and the LVL product allows you to span longer distances. Below, from the website:

Sawn Lumber has it limitations. Its bending strength is often only 1/2 that of engineered wood products. As a result, it doesn’t clear-span long distances, comes in sizes only up to 2×12, and select structural grades are not always available. Select structural grades are special-ordered in many locations. Also, not every species is readily available. For example, Douglas-fir is difficult to buy in some eastern markets. But overall, for short spans, sawn-lumber is tough to beat.

 

[viper86]

I don't believe that this information is correct, and that your contractor needs to look more carefully at the specs for the materials. Look at this pdf table from the UMass construction website. Note that for every application you need significantly more thickness for Douglas Fir lumber than you do for the LVL product -- and the LVL product allows you to span longer distances. Below, from the website:

I will second this - unless you're fortunate enough to use the highest quality southern pine or perhaps a machine-stress-rated lumber product. We'll typically spec spruce-pine-fir #2 for our wood-framed projects. The allowable bending stress of this material is pretty much 1/3 that of an LVL product. The LVL product is also a stiffer material than the SPF.

 

[neverenoughtools]

I'm not a carpenter, But I'd advise you to use the same 2x as you are sistering to and use the same length of wood as the one you are beefing up. Use construction adhesive and screw the heck out of it. Small pieces butted together will not do a thing......avoid them.!!!!!

 

[tcianci]

It really sounds like you have "small potatoes" project going on here. Wood is a somewhat plastic material and it will sag over time, it's not that the structure wasn't properly engineered in most cases, it's just that it is made out of wood. The simplest way to strengthen the frame that you are referring to is to sister the original rafters with 2x material for as much of the length as is reasonably possible, don't get all jammed in if you cant fit the new pieces to the ridge and down to the top plate of the wall.
When you install the new pieces, note the "crown" of the new lumber and install it against the old piece crown up and start your fastening at one end of the new piece and "train" it into place with another piece of 2x lumber as a jack against the floor. As you push the new material into place and stitch it off you will be putting a force on the roof structure that will tend to counter the existing sag. Don't be surprised if it talks back at you, you can't expect a 75 year old building to just up and run. Be sure that you bend over any nails that are protruding from the side of the old rafter or the roof deck so that you won't be fighting them as you install the new stuff. Declare victory and go home proud.