LVL Reinforcement

[phamENG]

I know this topic has been run ragged, but I have a (probably bad) idea that I want to run by the community that I haven't seen floated in the myriad other threads. In this specific case, I have an existing 4-ply LVL. Building is under construction, and floor finish above has been changed (roughly quadrupling the dead load). The beam is already in place, and there's no way to add reinforcement to the outside face (it would stick out of the building).

My idea is to come underneath with a steel section - a channel, wide flange, something - and fasten it to the underside. This would be really simple if it works (basic shear flow calculation at the joint), but seems like it might not work it practice. Seems like there would be considerable slip at the fasteners, but this would also be a problem in composite flitch beam design.

As far as I'm aware, fastener capacities on the narrow face of an LVL don't change as long as you obey the spacing requirements to prevent splitting.

Any thoughts? (I know, tearing it down and putting it back up would be 'easier' but I'm under pressure to save the day and not "cause" a delay by requiring shoring and tearing out, etc.)

 

[jayrod12]

How much room below can you eat up?

 

[phamENG]

I have a bit of leeway there - six inches or so could be used easily, 8 or maybe 10 if I make a convincing argument (the architect on this one is easy to work with). Sadly no room for a new beam underneath. I have about 2" on the inside face, but would prefer not making it any wider if I can avoid it, or introducing an asymmetrical beam or some torsion inducing headache.

 

[Pmtottawa]

Can you add a steel plate bolted through and design a flitch beam
Only works if lvl is dropped though

P

 

[XR250]

pham,

What size is the current LVL? What size does it need to be if LVL? Is is top flush? How tall are the joists it supports?

 

[phamENG]

14" TJI's, bearing on top of 14" LVLs. I wasn't the original designer - he's out of the picture now and I'm helping to clean up some things during construction. By my calcs, I need (4) 16" LVLs to do it. When asked to try to save material and avoid shoring, I cam up with adding a 16" on the inside, stripping off the outer 14", and adding a 16". That would leave me with a 5-ply, symmetrical (16-(3) 14-16), top loaded beam that barely eeks by. Apparently things have progressed far enough that removing the outer ply is a no-go. That's why I'm trying to just do something on the underside, but I'm not overly confident in avoiding slip in the connections. Maybe the answer is lots of small holes for self drilling screws?

Pmtottawa - thanks, but I'm trying to avoid an asymmetrical built up beam and the inherent detailing challenges. I know it could work, but I don't want to deal with that if possible.

 

[XR250]

If you have 8" to play with underneath, why not install a heavy 8" I-beam non-compositely?

 

[dhengr]

PhamENG:
Do you have access to the two side faces, the bottom and the ends of the 4 by 1.75” x 14” LVL? Consider a couple (one each side, maybe two each side) post tensioning rods, harped at about the third points of the beam. At the ends fab a bearing fixture, an end pl. 7” wide by 13” high (4x1.75” wide ?), two side pls. which accommodate the tensioning nuts on the ends of the rods, no inside weld radius to fit existing beam ends. The end pl. should be fairly thick to distrib. compression to all four of the LVL’s. You jack (pull) the rods and run the nuts down tight. Alternatively, you might use a turnbuckle at mid span. In either case, jack the existing beam up to unload it, camber it, then stress/tighten the rods, then release the jacks. In it’s simplest form, this is a bit more art than science; knowing the thread pitch, thus the turns per inch and shortening of the rods, you know the approx. stress in the rods.

 

[XR250]

Another idea...

Add a 4 ply LVL below ripped down to say 6". Transfer the shear with plywood on each face. (Not sure if the numbers will pan out on the plywood shear area)

 

[phamENG]

dhengr - thanks. I'd thought of a similar idea, but we're too tight to the outside of the building and I can't do much of anything on the outside face. Certainly not enough room to install a post-tensioning fixture, unfortunately. I don't want to do it underneath as the span is long enough to be concerned with relaxation in the rod/cable over time.

XR - The heavy W8 might work. Fasten it to make sure it deflects together, but not worry too much about rolling shear deflection. I'll play with the 6" LVL and plywood idea a little. The plywood might be thin enough to get away with. Thanks.

 

[Celt83]

Sounds like you have a flush header with a bit of room underneath. Would coming in with another LVL beam below and some blocking, almost like a vierendeel truss, be an option. Additional beam will be smaller but sounds like your close with the in place beam so just need a bit of load sharing between the two.

Also have you checked the in place beam using the properties of the LVL that were used, have been seeing "better" stuff getting used in the DC metro area. Any live load reduction?

My Personal Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

Open Source Structural GitHub Group:

https://github.com/open-struct-engineer

 

[phamENG]

Thanks, Celt. I did run it manually using the actual LVL installed. Good call on the live load reduction. It's only 466sf, so I won't get much out of it, but it may help make the numbers check out.

I think your idea is similar to XR's with the plywood skin - essentially using a plywood beam with LVL webs. It might, but I think the steel beam will get more traction - I'll get more bang for inches of depth used.

 

[Celt83]

was thinking something more like this, where your really just relying on the lower beam giving you spring supports for the existing beam and getting some load sharing going on.

My Personal Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

Open Source Structural GitHub Group:

https://github.com/open-struct-engineer

 

[phamENG]

Ah. I see now. Makes sense. Thanks for the idea.