Strengthening Floor Joist

[m_struct]

We are looking at some existing 120-year-old floor joist that are “over span” and have a dead load sag of about 2”, which is 2 to 3 times current code deflection limits. For strengthening, we have suggested sistering on LVL to the existing joist. However only 70% of one side of the joist span is accessible and the last 30% is inaccessible, hidden behind historical floor and ceiling line that we are told cannot be removed. Other than have a bit more deflection, do you see any issues with the new joist only 70% of the length?

In terms fixing the two joists together, am I correct in saying that this is NOT a shear flow problem (Q = VQ/I) as the two joists share the same N.A., and thus the ybar = 0. The screw fixing between the two joists along the length of the member should be the linear load that the beam is supporting. At the end of member, the screw fixings should transfer the all the V from the new joist to the existing. Does that sound right?

 

[jayrod12]

Your calculation assumptions are correct.

However, does this really improve things enough? In a best case scenario, you're removing 1/4 of the load from the original joist, I'd bet the moment in the original joist doesn't change drastically. At a rough and dirty number, I get a moment reduction of about 8.7%.

The contribution of the sistering LVL in terms of deflection would be hard to determine without some substantial FEA that takes into account nail slip etc. Although I'll admit I'm surprised, I ran a similar back of napkin calc and got a substantial reduction in the total deflection of the existing joist.

However, other difficulty I see, the existing joists with 2" of deflection will require either the new sistering LVLs to be aligned only at the middle and shimmed up the remaining length, or you remove the sheathing above and align at the bearing points only, essentially rendering the existing joist redundant. How were you planning on dealing with the existing deflected shape?

I'd highly recommend NOT jacking the existing joists the full 2". That is likely permanent set, I always picture wood as a bundle of hollow spaghetti. Over time it will take the deflection as its permanent state and jacking it up could cause damage.

 

[m_struct]

The deflected shape of the existing joist will remain. Good point about not jacking the existing joist. Top of new joist will be level. The flooring would bear on the new joist. For fixing the two joist together, I thought that screws (as opposed to bolts or nails) would minimize the lag or slip deflection between the two members.

 

[SWComposites]

So you are going to disconnect the existing flooring from the deflected joists, then raise the flooring to be level and supported on the new LVL joists? What about the area that the new joists will not cover?

 

[jayrod12]

If you're removing all of the flooring to have the new joists be level, then to me you could extend the new joists all the way to the bearing line at each end. Unless there's something else preventing removing the flooring in the one area.

 

[phamENG]

1st question - how did you determine they are over span? Analysis of historic floors is tricky, and takes a thorough understanding of how our modern wood code works and what constitutes a "failure" and why.

2nd question - how does the 2" 'dead load deflection' align with a creep calculation for the floor?

hidden behind historical floor and ceiling line that we are told cannot be removed.

False. They don't want you to remove it. And for good reason, especially the ceiling. I imagine it's plaster, which is expensive to reproduce. They also want to maintain as much of the historical fabric of the building as possible. A laudible goal. But here's the thing: you're the structural engineer, not them. If that flooring needs to be removed to get new joists in, then it needs to happen. They may have to go with a more qualified bidder rather than a cheaper one to make sure they have the experience and know-how to carefully remove said flooring without damaging it so it can be reinstalled, but it can likely be done (you didn't say what flooring material). The thing is, you have to prove it. You need to prove to them that this floor that has been there for 120 years is no longer safe for occupancy, that the repair options leaving both the floor and the ceiling in place do not do enough to get it there, and that you need to remove one or the other (probably the floor, since you want to level it and it's easier to reinstall most flooring than it is plaster, which you can't).

 

[RattlinBog]

Could use a little clarity on:
Are you intending to install new sisters from below or above (by removing flooring)?
Are there utilities or blocking/bridging in the way?
Are existing joists failing in flexure or shear to trigger this reinforcement? Or just excessive deflection?
What is your goal with the repair--increase flexure or shear strength; mitigate future creep/deflection; reduce existing deflection?

I recently completed similar work in my 100+ year old home. Main issues were from previous damage caused by poor plumbing/hvac work, not so much deflection. There were a few cracked and twisted joists. I sistered full depth 2x8 joists as full length as I could. I designed my connection how you described. Fasteners took the linear load (16" OC with 6" stagger) and ends took full shear. I used Simpson Strong-Tie Strong-Drive SDWS framing screws (0.16x3") and Loctite PL 3x Premium construction adhesive. I laid a bead of adhesive on the top of my new sister joist so it would adhere to the subfloor, too. I wouldn't skip the construction adhesive. I had good luck with installing the screws. Driving nails would be challenging depending on joist spacing.

It's going to be difficult to avoid jacking altogether, especially if installation is from below. When your new LVL (or 2x?) is set up there, you'll have to jack the ends and middle to get it into full contact with subfloor. If your LVL is straight and your floor is crooked, something is going to have to give. In my home, I had a minor amount of sag in the floor and crowning in the new joists to contend with, so I had to jack at various points to get things to play together. I ended up leaving a jack with some positive pressure at midspan supporting both the old and new joists overnight to let the adhesive cure, and then I removed the jack the next day. I was able to eliminate some soft spots and creaks in my floor by doing this. That would have been tough/impossible with 2" of deflection, though.

 

[m_struct]

Installation would be from above, as the flooring has been removed from 70% of the span as part of the renovation. It is bathroom and the arch called for 2" setdown. We said that we would need to see this existing framing commenting. I guess on a positive was the sag already had the set down. For the setdown area, we were debating sistering joist both sides of existing and then plane the top of joist at fit or just not allowing it.

@phamENG - The measured sag is in the ballpark of the calculation creep deflection.

The goal is to reduce the existing deflection the floor. The historic ceiling will still have a similar deflection as it currently does. Good point about the plumbing and blocking. The plumbing will be reinstated and run through middle 1/3 of joist. Joist spacing are 16-18", so about 14" clear once the "sistered" new joist are in place. With historic ceiling, jack is not likely. I also wonder about causing other issues when jacking.