Bouncy First Floor

[jferello]

Hey all,

I bought my house last year, had an inspection done and everything was ok. Over the last year we have noticed that the first floor is very bouncy. When looking down the basement I can see the joists flexing.

Here are some things I have noticed so far:
1) Not one of the joists are 16" on center, they vary from 7" to 17"
2) They are neither 2x8 nor 2x10, they are only 9"....
3) They appear to be Hem-Fir #3 boards

Today I did some googling and found that Hem-Fir 2x10 should not span further than 11-12 feet, mine span 14'

This sounds really unsafe to me... what can I do? This is a 2 story twin with attic and was built in 1978

 

[Smooky]

You could put a heavy timber or beam under the floor joist and support it with a lally column. You might be able to stiffen what you have by scabing larger timbers on the sides of your floor joist. That way you could use 2 X 12 maybe and only lower the ceiling 2-3 inches.

 

[Sponsor]

 

[jferello]

I want to try and stay away from anything that interrupts the basement as we want to finish it at some point. I am not saying I would never do that, but that would be last.

I am thinking solid bridging every 4' along the span across the whole floor.

Can you cut all joists in the middle, insert a new beam and use joist straps to re-connect them all to the new beam? This way the beam would not interfere with the ceiling height. It would rest on the foundation on one side and then I can support it on the other end with a metal post.

 

[Smooky]

Why don't you try scabbing on the side with 2x10's and see if that will stiffen up the floor? That would be the easiest and it may do the trick. You try it on 3-4 joist and see if it makes a difference. I don't think the bridging would help that much and it would be in the way of a real fix. You can and should have bridging between the joist. I would double them first.

 

[Reach4]

You can sister another 2x10 or a 2x8. alongside some of the joists. Screw and glue. Support the middle of the beam from below while you screw and glue. It is the middle part of the beam that does the most flexing.

A 2 x 8 is about 1.5 x 7.5.

 

[Smooky]

Forget glue and screws; just nail the hell out of it.

 

[jferello]

I am thinking about sistering joists, just a pain since the electrical and plumbing is all ran thru the joists rather than under them. The builder ran at least 10+ electrical lines thru 80% of the joists I need to sister

 

[jferello]

Here is what I really want to do, this would cut down the 14' span to about 7'
http://www.familyhandyman.com/walls/how-to-install-a-loadbearing-beam/view-all

Scroll down to:
"Alternative: Hide the load-bearing beam in the ceiling"

 

[Smooky]

Yea that would be a pain. I think you could do your beam idea. It would be a lot of work and might require an engineer. You might be able to relocate the elictrical or redo it after sistering the joist, the plumbing drains are another story. But anyway it is easier to cut out and re do the plumbing than to build around it. It is also stronger when done that way.

 

[Jadnashua]

A 2x10" after finishing (those are the rough dimensions...which is why it's only about 1.5" thick), is typically around 9.25" tall. To get the advantage of any sistering, you need to sister at least the middle 2/3'rds of the joist lengths. While blocking might help with point loads (the subflooring already does that), it does NOT change the overall joist deflection ratings. Blocking can help, especially during construction, by ensuring that the joists can't twist, keeping them at their maximum strength, but it does not strengthen them once the subflooring is on, which does most of the same thing. It can change the resonant frequency of the system, which makes it appear to be stronger. IF the subflooring is glued and proper fasteners are used, that goes a huge length to solidifying the structure. Often, they omitted the adhesive, and may not have used ring-shank nails, or screws to hold it together, and that makes a big difference, too. Coated nails are almost as good, as the heat from driving them helps to anchor them in place by melting the glue coating. Just removing the subflooring and installing new with adhesive and good fasteners will help make things more solid. I ended up doing that on my first floor partly to make it level and flat, and the difference was very noticeable.

Adding a midspan beam, if that beam is in itself strong enough, could help considerably. But, what is the length of that needed beam? TO get it strong enough (assuming support only on the ends), it might need to be something like a glulam or steel - each of which has its own requirements on how to then attach what's left of the existing joists to it. TO be safe, you need the help of a mechanical engineer to run the calculations and give you some reliable options.

 

[Dana]

I am thinking about sistering joists, just a pain since the electrical and plumbing is all ran thru the joists rather than under them. The builder ran at least 10+ electrical lines thru 80% of the joists I need to sister

You can get most of the benefits of a full-depth sistered joist (which is extreme overkill in the first place), by sistering on a pair of 2x3s or 2x4s, one at the top, under the subfloor, the other at the bottom edge. That effectively turns it into an I-beam of a sort. Even if the sistered on lumber has to be interrupted to accommodate some huge hole like drain, it still stiffens it up by quite a bit. If the electrical penetrations are all near the top you can still get real benefit out of a glued & screwed on 2x2, if that's all that fits.

I-joists typically have 3/8-1/2" web with effectively a 2x3 or 2x4 on the top & bottom, eg:

With a milled 2x10 as the "web", the sistered-on lumber makes it substantially stronger.

Inspect the floors above for flatness, and jack up any sagging joists to improve the planar characteristics of the floor, installing the sistered-on element while it's still jacked up. DO use construction adhesive, even if you nail rather than screw it in place.

 

[jferello]

Jad & Dana, very good information, thank you! Yes, if I went the beam route I would definitely consult an engineer. The new beam would need to run about 20' I believe.

I did not think about 2x3 or 2x4 at the tops and bottoms, that could be very feasible. Thanks!

 

[jferello]

If I did the 2x3 routine, would I put them on both sides of each joist or just one side? Also, would the bottom 2x3 go along the bottom of the joist or underneath it?

 

[Dana]

One side is good enough, if it's glued & screwed, but it has to be both top & bottom.

If you put them on both sides you can drop back to 2x2s, everywhere but it's a lot more messing around. Start with whatever fits on one side, see if it's stiff enough for you. It'll likely be satisfactory on just one side when you're only a couple feet beyond the maximum span found in standard engineering tables.

Sistering on full 2 x 10s would work but it is ridiculous- as if you really need 4 x 10 beams 16" o.c. to span 14' !?! If you're sistering on 2x4s at the bottom edge you may even be able to skip the top part, if the subfloor is properly glued to the joists, since the subfloor then behaves like the top flange.

 

[lanachurner]

Do your joists have X bracing between them?
They should have.
You would be surprised how much rigidity 2 rows of X bracing will give you.
The joists in the old part of our house are 2x10 16" OC with about a 15' span. There was no X bracing and there was some bounce to the floor. I installed 3 rows of X bracing and it is much stiffer now.

 

[Jadnashua]

Blocking between the joists helps with point loads, but does nearly nothing for a distributed load when it comes to deflection. With nothing underneath, it is the stiffness of the subflooring and the length of the joist that are the main things to determine the whole floor's load factor. With a point load, sharing the load better with an adjacent joist helps, but the biggest difference is during construction when it keeps the joists vertical until the flooring goes on. Second big thing, it decreases the resonant frequency of the floor, but really does little to help in a tile situation where you have a large distributed load and the maximum deflection is determined by the depth/spacing/length of the joists...adding blocking does not help that. IF both sides of the joists are covered with sheet goods, once installed, the blocking doesn't help much except to help with the resonant frequency. Over time, they shrink, and the resulting gaps don't help much, either.

Picture it this way...take three ropes, each one with a breaking point of say 500#. Put 450# on one, and it will stretch a certain amount. Now, put them close enough together so you can apply that 450#, it will stretch less. But, put 1350# on it, it will stretch as far as the single one did, and 1500# will probably break them. So, it is ultimately the strength of the individual units that determines the system strength when you're talking about a distributed load, which adding tile is, not how they are tied together. Gluing the subflooring to the joists along with ring-shank nails or screws helps immensely. Blocking, not so much. WIth the seasonal dimensional changes, the blocking will not remain perfect, which is what is required for minimizing the resonant frequency and the perceived strength of the flooring assembly.

Blocking also adds to the dead load of the floor system, which, when adding tile, is not a great thing, either.

 

[lanachurner]

I did not say blocking between the joists. I said X bracing.
It ties multiple joists together and makes a big difference in reducing the bounce the OP speaks of.

 

[Jadnashua]

Cross-bracing is just another way to provide blocking, and while it will help distribute the load, it does NOT decrease the deflection of the whole floor assembly. It certainly can help with a point load, but not a distributed load. And, as I said, it reduces the resonant frequency of the floor assembly, but that is a totally different thing than overall strength.

Tile is a distributed load, and is a significant one all by itself. Talk to any structural engineer, and it is the joists - spacing/length/height/species that determine the assembly's deflection, blocking or the specific type you refer to as cross bracing (same thing - harder to get two pieces tight and to work than one!) is not in the equation. Blocking helps keep the joists perpendicular to the load, which maximizes their overall strength, but does not increase it! The subflooring, once installed does the same thing, especially if you also have a panel on the bottom side.

Ever tried to walk across the joists on new construction? That's where blocking can really help to stabilize them. Not so much after the subflooring is installed.

 

[lanachurner]

If you think about it, if blocking or X bracing prevents point deflection it will also help to prevent deflection across the entire floor.
X bracing is superior to blocking for a couple of reasons. One, it transfers the load to the next joist better than blocking - because blocking needs to be accurately cut to be effective. But lumber is an imperfect medium - it has twists and bows in it. Two, it is cheap and easy to install. That's why you see it in all pre truss or I joist construction - going back 150 years.
But I don't care to get into this with you. I asked the OP if his joists had X bracing. It was a legitimate question because it helps to eliminate bounce in a floor.

 

[Jadnashua]

Point load, verses distributed load...two different things. Sharing the load works with a point load, has no significant effect on the overall deflection of the system. Can radically change the resonant frequency and the perceived strength of the system, but not the actual strength. Blocking/bracing the joists, beams, whatever ensures they remain vertical to allow them to support the load along their strongest axis. The subflooring locks the top in place, and if you can do the same on the bottom side, increase the strength far more than blocking would do, creating a box beam. Dealing with tile, the TCNA does not count blocking or bracing at all in a factor for the suitability of the structure for tile for a reason.