Side-by-side 'sistering' of joists?

Couldn't find any posts dealing with sistering of joists that could increase maximum spans..... Since I have a situation where shallower ceiling joists would be anvantageous (they would be placed floating between the above floor's floor joists), I was wondering if 2 steel studs would give better strength if they were boxed or attached back-to-back. Yes, we are dealing with a cellar studio project where any preservation of ceiling height is crucial.... Any ideas? anybody?

My gut tells me that you'd be better off not sistering the joists, but rather, just reducing the distance between them (i.e., 12" o.c.). Have you seen American Wood Council's Maximum Span Calculator for Joists & Rafters? --Keith :mrgreen:

sharward wrote:

My gut tells me that you'd be better off not sistering the joists, but rather, just reducing the distance between them (i.e., 12" o.c.). Have you seen American Wood Council's Maximum Span Calculator for Joists & Rafters? --Keith :mrgreen:

In effect, wouldn't it seem like somewhat the same thing? (though sistering seems like it might provide better 'stiffening') When sistering, the 2 joists are glue and/or screwed together... that would definitely make the resulting individual joist stronger, and would use the same material as if the spacing were halved (not just like going from 16 OC to 12 OC) Btw, I have seen that span calculator and I was a bit confused; I was surprised to find that oak or maple seemed to give a shorter span than some weaker softwoods. Counterintuitive. (I'll try it again) Plus I downloaded a table from a different reputable source that seems to conflict with values from that calculator. I have also downloaded Steel Stud Manufacturers Association technical product manual (50 plus pages), and they talked about stiffening joists and studs by sistering, either back to back or boxed. Why do you think sistering is any less than decreasing joist spacing? If I sister joists at 16 OC, I would get 2 joists per 16 inches (might that be like one every 8 inches?). K see: http://www.thisoldhouse.com/toh/print/0 ... 11,00.html

My roof and ceilings are held up with engineered trusses 24" o.c. If sistering them and doing them 48" o.c. was stronger, that's what they would have done. In your case, sistering on 16s is probably inferior to not sistering on 8s. I'm not a structural engineer though... :roll: I think sistering is usually done when there is some obstacle in the way. For example, if I am strengthening a floor assembly, but there are ducts running parallel to the joists, I might run additional joists between the existing joists (splitting the distance), but where there are ducts that prohibit that, I might sister the existing joists on either side of the duct. Did I mention I'm not a structural engineer? ;-) --Keith :mrgreen:

sharward wrote:

My roof and ceilings are held up with engineered trusses 24" o.c. If sistering them and doing them 48" o.c. was stronger, that's what they would have done.... In your case, sistering on 16s is probably inferior to not sistering on 8s....

The issue of going from 24 OC, to 48 OC could start to involve other issues; joists don't just support material in the direction of the span, but also in the direction of their spacing: 1.) Since the joist spacing is getting so great, the span of OTHER members (i.e., the facings/sidings) across that spacing increases to the point where they need to be able to safely make that increased span (4 feet as opposed to 2 feet); increasing joist spacing at this scale might actually increase costs of adjoining materials. 2.) The number of joists decreases, and so if any one fails for whatever reason, you might have a 'span' between them of 8 feet; in other words, the failure of one covers alot of area. 3.) Where the joist sits on a sill or crossbeam, there would be a much more concentrated crushing/compression force across the grain (i.e., the force is concentrated at a fewer number of points, instead of divided up into twice as many); this too becomes a big factor for those members, since we are not talking about a difference between 16" and 12", or even 8". So I CAN see that a closer spacing would be better at distributing the load to other members, or allowing floorings, facings, or ceilings to more easily distribute their loads to the joists, but at the scale we are talking (8", 12", 16") this factor seems not nearly as important (as it is at 2' to 4'). Also, I'm not talking about increasing joist spacing from 16", I'm talking about increasing the strength of each joist; it's not as if the joist spacing WAS 12" or 8" and I was increasing that spacing. We know that 12, 16, 24 OC are acceptable joist spacings in of themselves (if we just for a moment disregard weight load, joist depth, or joist material) since they ARE used in certain situations; 48 joist OC seems unheard of (at least in house construction(?)). In any case, if I have room for 8" OC I will consider it. Then the question would be: what effect would that have on sound transmission? thanx, K

Luftweg wrote:

In any case, if I have room for 8" OC I will consider it. Then the question would be: what effect would that have on sound transmission?

I believe that isolation decreases as the joist/stud spacing decreases. If this is not the case, what the effect may be is that with closer spaced joists/studs, you are changing the resonant frequency of the wall (i.e., smaller spacing = higher frequency). Did I read the first post here wrong or did you want to use steel JOISTS? Is that possible? I don't recall seeing them in anything but top and bottom plates and studs. len

Yes, there are steel joists. Yes, there are benefits isolationwise to have greater spacing at certain frequencies. Floppier can be better. But stability, strength, safety, and code compliance should trump isolation. Bottom line, if you could do it two ways, both of which are totally safe, but one offers greater isolation results, then obviously you'd choose the one that is superior isolationwise. :) --Keith :mrgreen:

len-morgan wrote:

Did I read the first post here wrong or did you want to use steel JOISTS? Is that possible? I don't recall seeing them in anything but top and bottom plates and studs. len

I believe steel load-bearing studs and steel joists are possibly the very same members; i.e., they have the same shape -- and thus code designation. On the span tables that I have found (search: Steel Stud Manufacturers Association, for a 50+ page technical product listing), there ARE steel ceiling joists (studs) with a 3.5 inch depth (and I think, 1.68 inch thickness, and appropriate gauge steel) that can support the span I'm in need of -- even at 16 OC (and most definitely at 12 OC) without even back-to-back or boxing of 2 pieces. K

bump for Knightfly.... thanx, K

Luftweg Your biggest challenge w/ sistering the joists is that you also need to double up the joists at the bearing points (the ends, where they rest on the foundation) for them to get the increased capacity. It's tough to do, esp. for an entire room, because they need to span bearing point to bearing point continuously, to get that extra capacity. Make sure you do this right, with the approval of the building dept. or at least an experienced framing carpenter, to make sure you get the bearing for the joists correct. You don't want to end up with all that add'l. drywall on your head. Good luck Frank

Luftweg, The answer to the question is this: If I place 2 - 4" x 1 5/8" joists together (side by each) I have essentially created a 4" x 3 1/4" joist (which is not the same as if I stack them one above the other and create an 8 x 1 5/8" member) Short - stocky square members do not necessarily gain you a lot - you would get a lot more out of them constructed the other way........ but it would help to understand exactly what it is you're building - so a complete description of your conditions would be expected. Rod. .

rod gervais wrote:

If I place 2 - 4" x 1 5/8" joists together (side by each) I have essentially created a 4" x 3 1/4" joist (which is not the same as if I stack them one above the other and create an 8 x 1 5/8" member) Short - stocky square members do not necessarily gain you a lot - you would get a lot more out of them constructed the other way........

Ahh yes, there are diminishing returns on simply thickening the joists. Yet, essentially this is what is almost happening when 'on-center' distances decrease. Not sure if sistering may or may not provide a better option; on the one hand, if the distances are far apart already, it seems better support could be given to the structures that need to 'span' the on-center distance itself (not the span of the joist) if the OC distance decreases; on the other hand, if the sistering of the joists approaches a 'lamination' I could see a more stiffening effect than decreasing OC distances might give with the same members. Other factors I was wondering about could be the material: hardwoods vs softwoods; laminated woods (sorta mentioned, if gluing 2 joists); steel (heavier gauges, shape of cross-section). I mean, is it wrong to think white oak or rock maple ISN'T alot stronger than standard softwood building materials? And what about laminated 4 inch strips of a couple baltic birch sheets plywood (would figure that could hold up half the house! lol. kinda sorta 'half'engineered joists)?

rod gervais wrote:

but it would help to understand exactly what it is you're building - so a complete description of your conditions would be expected. Rod.

Rod, please see my project thread: http://www.johnlsayers.com/phpBB2/viewtopic.php?t=5980 I'm putting ceiling joists between the above floor's 2x8 joists (to de-couple); don't have tons of room to run 2x6 and thought a shallower, but stronger option was possible. thanx, K

Luftweg wrote:

[Ahh yes, there are diminishing returns on simply thickening the joists. Yet, essentially this is what is almost happening when 'on-center' distances decrease. Not true - The closer together you put the joists - the more the diminishing law of returns comes into play. As an example - 2x4 DF Larch - Select Structural - L/360 - 40# LL 10# DL - max span at 24" oc would be 5'9" - yet you only gain 1' 6" of span if you go to 12" centers - for a total of 7'3" - this because although you are cutting the load in half - you are doubling the dead load of the joists itself - it ends up getting worse the closer you bring the joists together.

Other factors I was wondering about could be the material: hardwoods vs softwoods; laminated woods (sorta mentioned, if gluing 2 joists); steel (heavier gauges, shape of cross-section). I mean, is it wrong to think white oak or rock maple ISN'T alot stronger than standard softwood building materials?

It would be wrong to think that hardwoods are much stronger than softwoods........ very wrong in fact. Look - same requirements for all materials - 2x6 - 40/10 (LL/DL) - L/360 - 24" oc:

Species............. Modulus of Elacticity (E) - Fiber Bending (Fb) - Span Doug Fir.................1,900,000psi....................2242.50psi..........9'-0" Hem Fir..................1,600,000psi....................2093.00psi..........8'-6" SPF........................1,500,000psi....................1868.75psi..........8'-4" Red MAple..............1,700,000psi....................1934.50psi..........8'-8" Red Oak.................1,400,000psi....................1791.25psi..........8'-2" White Oak...............1,100,000psi....................1794.00psi..........7'-6" data from Washington State University

And what about laminated 4 inch strips of a couple baltic birch sheets plywood (would figure that could hold up half the house! lol. kinda sorta 'half'engineered joists)?

I don't see a building official letting you get awaqy with some make believe laminated construction without some engineering and tests to support it. Sincerely, Rod

Aha, so the dead weight -- actual weight of the wood itself -- limits the increase in span.... In the case of say changing from a 2x4 joist to a 2x8 joist at the same OC separation, wouldn't there also be a doubling of the dead weight? But wouldn't doing that greatly increase the maximum span? Just as a random comparison on the 'calculator': A hem-fir 2x4 (stud grade, 16 OC, at L360, 5 dead, 10 live) gives 8' 11"; A hem-fir 2x8 (stud grade, 16 OC, at L360, 5 dead, 10 live) gives 17' 5". (The weight of the wood of the joists would have roughly doubled.) I'm supposing what I'm doing wrong here is not including the weight of the joist itself? (if that's true, that's a shame because wouldn't it be better if the calculated span already accounted for the weight of the joist's wood? After all, you can't really construct anything without the weight of what your constructing it with -- and the average weight of a particular member should be known?) (On the hardwood issue:) I guess I'm confused. Hardwoods are usually 'harder' than alot of softwoods, but I guess that does not mean they are stronger?.... hardness is not strength (I thought if they were harder and fibrous that they 'might' also be stronger.... guess I was wrong about that one). thanx, K

Luftweg wrote:

(On the hardwood issue:) I guess I'm confused. Hardwoods are usually 'harder' than alot of softwoods, but I guess that does not mean they are stronger?.... hardness is not strength (I thought if they were harder and fibrous that they 'might' also be stronger.... guess I was wrong about that one).

The problem is you're mixing up the concepts........ Why (for instance) should harder mean stronger? let's take 2 different material - one very hard - one much softer - both exactly the same thickness and length - and see which one will carry more...... a 4' length of 1/4" annealed glass - and a 4' length of 1/4" 3 strand nylon rope. support the glass rod at each end and place a 50# load on it and the glass will snap - support the rope at each end and it will carry 1650 pounds. The glass is much harder - the rope is much stronger. Hardwoods tend to be brittle (glass like) and do not fare well under bending stresses - softwoods tend to handle those stresses well - so softwood make better structural members (in general) than hardwoods. Rod

[quote="rod gervais] The problem is you're mixing up the concepts........ Why (for instance) should harder mean stronger? .... Hardwoods tend to be brittle (glass like) and do not fare well under bending stresses - softwoods tend to handle those stresses well - so softwood make better structural members (in general) than hardwoods. Rod[/quote] So it's like I said in the previous post(?), strength is not hardness; of course I do/did know that, but it's not like the differences between hardwoods and softwoods is like that easy to 'see' as it is between diamonds and steel .... Actually (according to the 'calculator'), 'beech-birch-hickory' IS stronger than 'hemlock-fir' for an example 2x6 with similar specs ( 10' 7" versus 9' 11" (L360 10 live, 20 dead)) -- similar situation for northern red oak (other hardwoods 'fair' less). I'm trying to find a calculator that 'figures IN' for the weight of the joists themselves..... I'm starting to really like the idea of steel joists for the ceiling, as it seems (since they are pretty light), figuring-in for their weight is not AS important..... I really thank you for all the help with these concepts... K

Luftweg wrote:

Actually (according to the 'calculator'), 'beech-birch-hickory' IS stronger than 'hemlock-fir' for an example 2x6 with similar specs ( 10' 7" versus 9' 11" (L360 10 live, 20 dead)) -- similar situation for northern red oak (other hardwoods 'fair' less). The hickory looks right (although I would hate to think of the cost) but the northern red oak doesn't- what span tableare you using?

I'm trying to find a calculator that 'figures IN' for the weight of the joists themselves.....

The calculators give you a span with particular live and dead loads - those loads do take into account the weight of the jist themselves. So I don't understand why you keep refering to that.........

I'm starting to really like the idea of steel joists for the ceiling, as it seems (since they are really light), figuring for their weight is not very important.....

Knowing their weight is exactly as important as it is with wood....... For example - a 20 ga 6" member weighs .92 plf - whereas a 2x6 piece of Doug Fir weighs about 2 plf.......... So keep it important in your mind either way.............. Sincerely, Rod

rod gervais wrote:

.... what span tableare you using?

http://www.awc.org/calculators/span/cal ... ontal+Span Northern red oak gives a different, greater span than does 'regular' red oak.....

I'm trying to find a calculator that 'figures IN' for the weight of the joists themselves.....

rod gervais wrote:

The calculators give you a span with particular live and dead loads - those loads do take into account the weight of the jist themselves. So I don't understand why you keep refering to that.........

The only reason I was trying to find out for sure is because some people (here?) have told me that the span tables DON'T take into account the weight of the joist in question. I thought that that would be a silly table/calculator, since -- if they know all the other parameters of the wood -- they could easily figure-in for it's weight (and thus avoid having to calculate it in afterwards). Are you saying then that you DO have to add the weight of the joists in to the dead weight, OR is that weight alread considered, and all you have to use in the dead weight rating is the insulation, sheet-rock, etc. weights?....

I'm starting to really like the idea of steel joists for the ceiling, as it seems (since they are really light), figuring for their weight is not very important.....

rod gervais wrote:

Knowing their weight is exactly as important as it is with wood....... For example - a 20 ga 6" member weighs .92 plf - whereas a 2x6 piece of Doug Fir weighs about 2 plf.......... So keep it important in your mind either way..............

Hmmm, then that indicates that the dead weight must be figured for the weight of the joists; that is, one MUST include the weight of the joists as part of the dead weight. Your statement, above, about the calculators is a little confusing for me. ... Don't worry, I find out which way it is..... I guess I should have written about the steel members that it's 'less' important to consider their weight, and not implied that it was 'unimportant'..... thanx, K

Luftweg wrote:

The only reason I was trying to find out for sure is because some people (here?) have told me that the span tables DON'T take into account the weight of the joist in question. I thought that that would be a silly table/calculator, since -- if they know all the other parameters of the wood -- they could easily figure-in for it's weight (and thus avoid having to calculate it in afterwards). Are you saying then that you DO have to add the weight of the joists in to the dead weight, OR is that weight alread considered, and all you have to use in the dead weight rating is the insulation, sheet-rock, etc. weights?....

What I am saying is that the dead load calculations (for example 10#) would be the TOTAL WEIGHT the joist can carry - including the joist itself. So - if the dead load is 10 psf - and the span is 2' o.c. - and the joist weighs 2plf then the total load that you can add to the joist (per linear foot) would be 8 pounds per bay - that includes the decking - any additional isolation above or below the decking - plus the ceiling. This also includes any piping or ductwork supported by the joists. Now - that is the case if you're using a span table.......... an engineer (however) can examine your actual construction - and determine if you can safely exceed that restriction. In the case of an independantly framed ceiling - you can use the dead load along with the live load (seeing as you can't impose any real live load) in your calculations. I hope that makes it easier to understand.

I guess I should have written about the steel members that it's 'less' important to consider their weight, and not implied that it was 'unimportant'.....

Nope - it is not less important - it is exactly the same level of importance as it is with a wood joist. Sincerely, Rod

Okay... One of the specific issues I have that prods me to employ joist sistering is this: It involves adding between-joist sheetrock to the underside of a floor from the basement, for the outer-leaf of the basement studio ceiling..... I'm worrying about the added weight of 2 layers of 5/8" sheetrock to the subfloor. The house is circa 1959-60 construction. The above floor joists have a span of 11' 10" from sill to center beam. Floor joists are 2x8's, 16 OC, #2 hem-fir (northern hem-fir? I'm north of Boston), with mid-span cross-bracing. The above room is a bedroom, so it's 30 live load I guess. There is no rot, mold, water, or insect damage to the joists. The underlayment is dimensional 1 by 8's. Over that is red oak flooring, and over that is wall to wall carpet. There is no current sag or bounce problem with this floor. I have full access to the floor from below. There are only the center beam, and the main heating duct next to it, running below the joists (there ARE 4 in-between joist ducts, but these don't appear to be much in the way (I can sister from the adjacent OC spacing)). There are no electrical wiring or plumbing issues that hinder access to the existing joists. I cannot add a beam below the midspan of the joists, as I don't have the ceiling height for that (thus nor can I sister deeper joists than 2x8's); even with the plan now, the ceiling will be just a hair over 7' (that's the 'code' by the way). I'm planning on sistering additional 2x8 hem-fir joists, for most of, if not the entire length of the span AND over the sill and beam on each end. I realize that I will likely have to notch the ends to have them fit (I'm not planning on jacking, as there is no real current sag, there will be a re-loading of the joists after the sheetrock is added, and I just plain want to avoid all that work). They would be screwed/glued. Also planning on blocking at mid-span, possibly even quarter-span too. I Don't want to use cross-bracing -- unless it would be way better that way; I will, however, need to notch the blocking a bit to allow room for isolated inner-leaf steel ceiling joists. From checking span tables, it appears that I already have sufficient specs to add 2 layers of sheetrock, if I consider a dead load of 20. However, I don't want a sag or bounce to develop after adding them on, and would like a safety margin of sorts. **** Is sistering the best option when considering all the factors of cost, work, reasonable increase in strength/stiffness? How is this best done (with respect to glue/screw, methods, etc.)? I have looked a little bit into some other ideas (flitching, adding a steel strap, etc.); am I overlooking a better option? Finally, I do realize that the strength in joists increases by the depth much more so than by the thickness (or OC spacing), but again, I don't have that option on this project.... Oh, and also realize that the additional weight of the sister 2x8's will somewhat mitigate the increase in weight capacity, but I still think it would be an acceptable improvement (from what I have read so far).