A 32Æ-0ö wide building. The floor

[mreale6828]

Trying to find a soution for excessive floor deflection.

A 32’-0” wide building. The floor structure is 4”x10” Hemlock joists, 4’-0” on center with a 2x6 tongue and v-groove southern yellow pine decking. There is a center girder, effectively reducing the span to 15’-2 ½ ”.

Originally, the floor system was supposed to be designed to support a 50# per square foot load. After I noticed the excessive deflection, almost 1 inch on some joists, we looked back at the design and determined that we had a communication problem and undersized the joists (probably should have been 8x10’s or 6X12’s)

One thought is a steel strap support on the bottom of each joist. By lagging the strap at each end of the joist, then by using a bolt and threaded tube assembly at the mid-point, force the strap away from the bottom placing the joist in compression and the strap in tension. I expect this to eventually persuade the joists back to level.

I would like to read some expert opinions regarding this method. Then if possible, calculate the loads, appropriate size strap, bolts, and fastening so I can get them fabricated.

Of course I’m open to other ideas.

 

[ERV]

From what I read you have an overspan problem. You are carrying about 208 lbs/ft, this is inadequate by at least 100%, Moment of inertia is the biggest problem fully loaded this would give deflections in excess of 1" (L/177).
If at all possible consider breaking the span with another girder. A strap in tension may seem a simple solution, however, the connections may prove to be a major and complex problem.

 

[haynewp]

I find it interesting that the wood joists have not broken with 1" of deflection in a 15' span?

 

[chicopee]

You idea of restoring the chamber appears to have failed when it was noticed that there was excessive deflection of the space frame roof of the Hartford Civic Ceneter. Your idea will probably not worked. I would add support to your decking in the form of columns or girders.

 

[ben1212]

The joists may have experienced some creep already. Deflection from creep is difficult to determine. I believe the NDS code says to multiply your dead load deflection by 2 to account for creep. Please bear with me for I am at home and have no reference material here. I've also read that members that have creeped with spring back at least half the creep deflection when the load is removed.

I'm afraid it may be impossible to remove all the deflection. If possible scab 2x10 or 2x8's to the existing 4x10's then figure what you need in addition. Maybe 2x10 or 2x8 at 16" or 24" o.c. between the existing 4x10's.

The structure you describe sounds like it was meant to be exposed. What I described above is not going to be as nice looking as the beams at 4'-0" o.c. I don't think you are going to get your metal strap/truss idea to work. It's a way out idea and would probably look like hell too.

I suggest reinforcing the joists and adding more and if it is exposed then put some nice bead board on the bottom/ceiling side also.

 

[catanzarope]

Your remedial design proposal is what is called a trussed beam. The tenison you place in the strap is going to induce compression in the joist. I don't recommend doing that as a repair on an excessively sagging member. You would probably fail the joists in compression before much of the deflection came out, plus the end connection is going to difficult.

Boil down the problem by finding out what span your existing joists can safely carry. Then, install supports to cut the span to this length. Obviously, this will clutter the area below this floor but if not a problem, this is your most cost effective way to go.

If it is a problem and you need the space below the floor you will have to scab the joists. I have reversed creep/deflection by jacking the floor system then bolting steel plates/channels to them. 15 foot channels/plates weighing up to 250 pounds each can easily be installed by three men and give you quite a bit of moment capacity. Before installing the steel, I butter the joist with liquid nail, then suck the steel tight (not over tight) to the joist. Let the adhesive cure before jacking down. The floor seldom drops more than a hair.

Another option is to install additional 4 x 10's (or whatever works) in the bays cutting the spacing to 2 feet, 16 inches, etc. However, it is difficult to reverse creep this way and you may not be satisfied with the result.

Sounds like this was done without a structural engineer originally and I would recommend you hire one to design a repair. Let's not spend good money after bad.

 

[DaveDave]

Sounds like a real problem. I think that to call it a deflection problem is an understatement.

I would opt for jacking the existing joists at midspan, inserting intermediate joists (to make the spacing 2'), then releasing the jacks.

I agree...you need a local engineer

 

[jjbaer]

I saw your old post and here's what I think (you verify, I'm only providing my thoughts): a 15 ft long 4' x 10" hemlock beam will deflect 1" at the center with a 7600 lbs uniform load on it. This assumes a modulus of 1,780,000 and that the 4-inch beam dimension rests on the supports with the 10-inch dimension being vertical, thus maximizing the moment of inertia (stiffness) of the beam. Given 4 ft centers, this means the 7600 lbs is spread over 60 sq feet (15 ft long x 4 ft wide). This equates to a loading of 126 lb (7600lb/60 sq ft = 126). This gives a stress of about 2500 lbf/sq inch ( i.e., 2500 psi) which is slightly above the max stress at the proportional limit but nowhere near the breaking point. The perpendicular shear is 95 psi (half the weight divided by the 40 sq in cross section) which is also ok. The compressive (crushing) force at the support ends due to the beam resting on the supports is unknown unless you know how much of the beam is on the support surfaces. If we assume that at a minimum, the beam rests 1.5 inches onto the support at each end, then the crushing pressure is 475 psi (3800 lb/[4 in width x 1.5 in long] = 475) which is within the proportional crushing level and nowhere near the maximum level.

The point here is this: as with most floors, deflections and vibrations are the first thing that owners notice and complain about (and rightfully so) as being unacceptable and this point is usually (but not always) reached long before any other structural limits are reached. You seem to be in the same boat but you need to have your engineer do these calcualtions for you and NOT rely on mine. Adding another beam perpendicular to yours at mid-span would seem to be best, but again, you have to have your engineer tell you this. Adding the mid-span beam would effectively mean that your 15 ft beam now spans 7.5 ft. Deflections (for this same load) go down by a factor of 8 to about 0.125 inches and stresses are cut in half. Good luck, and again, this is NOT legal advice....you need to have your engineer check this out.

 

[Slowzuki]

Another nay on the bottom strap, not gonna happen. You need a structural eng and you may wish to look at slipping some steel in there if preserving the look is important or the span can't be reduced.

Ken

 

[UcfSE]

How about some steel side plates thru-bolted to the joists? Shore the joists to reduce or eliminate the deflection and then bolt a steel plate each side, maybe 1/4-inch or whatever the calculations show. Maybe try some double angles thru-bolted across the bottom of the beam, with legs longs enough in the vertical to give the necessary edge distance. That would create a bottom flange, you could do the same for the top flange. Obviously these won't be cheap but they might preserve the look of 4'-0" o.c. if that is what is required. Maybe try bolting another 4x10 to the side of each joist.

If you are out of your area of expertise then you should hire a local structural engineer.

 

[jtdpsu77]

this is what happens when mechanical engineers design floor systems.

oh snap!