What to do about "bouncy" floor?

House Repair Talk

Help Support House Repair Talk:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

AlexEthridge

New Member
Joined
Nov 13, 2011
Messages
4
Reaction score
0
I discovered my "2x8" floor joists are actually 7&1/4 and many actually 7" with NONE at 7&1/2 that one would expect. The span is 13 feet and the floor is "bouncy". My idea to increase rigidity is to add another two and a half inches to the bottom of the existing joists using a 2x3 attached with waterproof glue and screws.

If you think I might benefit from this tactic, please offer advice on glue type and screw spacing. My plan would be to use 2x6's ripped to 2&3/4" with pre-drilled screw holes.

About Me:
I've been a do-it-yourselfer all my life with limited experience removing weight-bearing walls and substituting self-built laminated beams from plywood and lumber. I also owned and operated a small woodworking shop for a few years where I did complete kitchens and baths along with custom, built-in office furniture.
 
Last edited:
If you want it better, but not by a whole lot do it your way. There's two other ways that would work 100%.
#1 Sister 2 X 8's to the sides of the ones you have now, have them run the full lenght as the old ones and use constrution adhesive and 1/2 carrage bolts so it will pull the two together every 2' switching from top to bottom as you go along. I predrill 9/16" holes with a paddle bit so when I'm under there I only need to be drilling one piece not two. Nails will not pull the two halfs together. You need to have them act as one piece.
It takes two people to do this job and a couple of mauls to drive them in place.
#2 Is to install a beam in the the middle of the floor joist made of doubled up 2 X 8's running opposite to the floor joist sitting on piers which are sitting on footings.
For the past 15 years about all I've worked on is 100 plus year old houses and churches and have done both repairs these ways and it's worked perfect, we have even been able to get out 3" sags out of floors using the beam to lift the joist.
 
Alex,

There was a similar bouncy floor thread a few weeks ago, started by papakevin. Look it up by doing a thread search, and read what's been said and done to make things acceptable. Your inquiry didn't say how you plan to orient your ripped 2 x 6s, but you'll have better results installing them "flat" on the bottoms of the joists (making an inverted "T") instead of alongside. Because the finished width of your new "flanges" is less than 78% of full 2 x 4s, you shouldn't expect the total deflection to decrease by the 50% value for full 2 x 4s, as I mentioned with sample calculations in the referenced thread. If I was in your situation, I would use either full 2 x 4s, or even alternate them with full 2 x 6s. The latter would reduce the bounce to imperceptible levels. And they only have to span the middle 2/3 of the joists' full length, where the bending moment and deflection are greatest. Using No. 8 flathead deck screws (I prefer square-drives, no "camming" out like Phillips do), 3" long or better, would work just fine, spaced every 8" or 9".

Or if you enjoy doing things the long, hard and expensive way, listen to what joecaption has to say. Making a floor stout enough to launch Saturn rockets on doesn't do anything for me, but to each his own.
 
Stiffening Engineered Floor Joists | Floors | This Old House
I'm not the only one that feels the way I suggested will work.
I tryed your way once Bridgeman and had to go back for free and do it all over because it was better but not good enough for my fussy customer. Once I installed the sistered floor joist there was no preceptable bounce and the tall high boy cabinet no longer moved when you walked on the floors in the middle and I was able to get the sag out of the floors.
If you look on this and other sites a lot of people have been having problums with the new engineered floor joists bouncing, and there a true I beam in design as your suggesting will work. I think most of it is caused by the builder cheaping out and not building it with one's wide enough or deep enough to save money.
You'll have to trust me on this one, I'm not trying to bust you, I've just been there and done that on some things.
A lot of my jobs came long after the houses were built and I got to see what did and did not workout in the long run.
 
Last edited:
Thanks for the comments.

I forgot to mention this is a full basement house so adding supports in the middle isn't practical.

BridgeMan, I can't grasp how turning the grain flat could possibly be better. That is counter-intuitive to my understanding of wood strengths.

Sandwiching more 2x8's in there isn't going to be worth all the effort as there is wiring, AC duct, plumbing, etc to deal with. I can fit in another 2.5" board on the bottom of what's there on nearly every one of the joists; but, another 2x8 just isn't practical.

What I know about gluing wood comes from my woodworking jobs and from building my own beams into attics in anticipation of removing weight-bearing walls. In those cases, I didn't even have to give much thought to whether it would be strong enough as I had plenty of height to work with so I just made it strong enough to overkill--no call-backs. In this case, I have limited space and many obstructions.

Unless someone knows something about this that shoots my idea down, it looks like my original idea is going to be my most workable.
 
BridgeMan, I can't grasp how turning the grain flat could possibly be better. That is counter-intuitive to my understanding of wood strengths.

Then I'll ask you again to search for papakevin's thread on his bouncy floor joists. I tried to make the explanation as simple as possible in that thread.

Unless someone knows something about this that shoots my idea down, it looks like my original idea is going to be my most workable.

It will be workable, although not very productive. I ran a few numbers for you--adding your ripped 2 x 6s, one to each joist on one side only, will increase the stiffness (and resistance to deflection) by only 12%. So if your deflections are now 1", they will be about 7/8" when you're done. If you can live with that, then by all means go for it.

However, if you want your efforts to be more rewarding and meaningful, glue and screw 2 x 6s, centered, to the bottom of each joist, middle 2/3 of each span. Doing so will increase the section modulus to over twice that of a naked 2 x 8 (30.24 cu. in. divided by 13.14 cu. in. = 2.30), and means they will deflect less than half as much as before. You could use 2 x 4s instead, but the differential would be somewhat less.

And if nothing I'm saying makes any sense to you, then do me one very simple favor (can't help myself for using bridges as an example, as I spent 40+ years working with them to earn a living)--ask yourself when you drive under any steel girder highway bridge and look up at it, why did the designers not orient the bottom flanges parallel to the webs and butting up to them (like you want to orient your ripped 2 x 6s), or even sister in some additional full-depth, parallel plates (like joecaption wants to do), instead of putting the flanges perpendicular to the webs in the form of an inverted "T"? The answer to that question--they simply wanted to utilize the most efficient and economical shape for resisting vertical loads the girders are required to support. If it works for supporting 80,000-lb., 18-wheelers, maybe it will work for your floor joists.

Just something to think about.
 
one to each joist on one side only
That is not exactly what I intend to do. My original idea is to add the 2.875" (5.5 ripped, minus .125 for the saw blade) to the bottom of the current joists, effectively changing the 1.5 x 7 joists to 1.5 x 9.875 and the 1.5 x 7.25 joists to 1.5 x 10.125 (actual).

To be clear, I intend to add the 2.875-inch board to the bottom of the joists, not the side, causing a 2.875-inch loss of ceiling height in the basement.

Several months back, I had a project where I glued two 1 x 12's together, edge to edge to effectively make a 1 x 24 board. It was four feet long. Just recently, as a test, I attempted to separate those boards. I cut the 4' length into two 2' lengths to try two methods of testing the strength of the glue joint. I placed one on a table with blocks on either side and hit it with a heavy hammer. After a few blows, it split, but not on the glue joint. The second method was to put it in a press with precision-placed pressure on the glue joint. Again, it split but not on the glue joint.

I'm not a structural engineer so I admit there is a lot about this that I don't know, such as how to run the calculations that experts do, etc. But, from what I know from a lifetime of gluing wood, adding two inches of width to a two by eight, effectively gives it the strength of a two by ten as a properly-glued joint is always stronger than the wood, itself.

Since the actual measurements of my glued boards will exceed the actual measurements of a store-bought 2 x 10, I have trouble understanding why I am not going to have at least the strength of the 2 x 10.

Please understand I am not attempting to challenge the knowledge of seasoned professionals; I'm just trying to understand why the logic of the above paragraph isn't correct. I'm not trying for zero deflection; I'm just trying to get the deflection I would have if the builder had met span requirements by using a 2 x 10 in the first place.
 
Last edited:
I don't often agree with Bridgeman, but on this one. When we use silent floor system which is OSB with a flat 2x on top and bottom. For short spans they use 2x3s and for long runs they use 2x4s, we never question it because they engineer it and send it out with instructions like trusses.
Like Joe we have allway doubled up the joists because thats what the local engineers call for.
Could it be that all we are doing is doubling the bottom surface and perhaps a 2x4 on the flat would just give you evan more bottom surface under tention.
We have also had good results with 3/4" plywood screwed to the side of joists just in the middle third.
 
Alex,

Trying to make your joists deeper is definitely an option. However, your thinking that it is the best option is flawed, simply because the numbers don't lie. Section modulus (S) of a 2 x 10 is 21.39 cu. in.; section modulus of a 2 x 8 modified by adding a flat 2 x 6 is 30.24 cu. in. Dividing the former by the latter results in 21.39 / 30.24 = 0.71. In other words, your fix will only be 71% as strong (and stiff) as going the flat 2 x 6 route, while costing you an additional 1-1/4" of headroom.

Doesn't make sense in a logical world, but it's what you've chosen to do. Good luck.
 
Ive looked at every concept of removing the bounce in my floor, some possible and some not.
After more researching, ive found something new. IBS bridging. simple enough, but will it work?
I read bridgemans post, and like the idea, only i think if the 2x material was configured(two 2x4's in an L shape) and then applied, it would definitley work.
Hopefully, somebody can push me in the right direction. I really like the IBS idea, which i can manufacture my self.
My span is 14ft. 2x10

P.S. sistering is not an option. to much plumbing, ele.,duct.
thanks Tooch
 
Back
Top