Diaphragm - Plain Concrete

I don't love it, reduced ductility and all. How confident are you that your concrete deck will have no shrinkage cracks in it when it comes time for it to resist it's seismic flexural demand? In a sense, the beams that form your diaphragm chords kind of will be a form of diaphragm reinforcement.

Samwise - is this the same building you've been posting about with steel deck? If so, then I'd advise you to detail it so your concrete is NOT the diaphragm. I'd use the steel deck as the diaphragm - the concrete on top just stiffens it and drastically increases its capacity (as a diaphragm). SDI diaphragm design manual and the various deck suppliers have special tables for diaphragm strength when their deck is filled with concrete.

If this is a different building, I apologize for the assumption.

Yeah, agree with Kootk. My sense from ACI 318 on unreinforced concrete is: it's ok as long as there is no chance of catastrophic failure from it cracking.

Since you practice in CA I'm assuming the building is located in a high seismic area. ACI 318-14 14.1.4 doesn't allow plain concrete to be used except at foundations meeting specific criteria.

I do not practice in CA. I am in the intermountain region. I am going to still have the angle to behave like a chord member. And if I get any resistance from the diaphragm to act as chord, that just provides me with multiple load paths

Keep in mind you have to have a connection between the diaphragm and the collectors, VLFRS, etc. In steel framing with concrete fill, there is rarely a suitable connection between the the concrete and the steel unless you're doing full composite construction. The bond between composite deck and the concrete isn't enough. You'd need welded studs between beams and your concrete to use the concrete as a diaphragm. Without it, it's just a filler to stiffen the steel deck diaphragm.

Certainly, I agree with phamENG that it's s stiffened steel deck diaphragm unless there is a direct, and convincing, connection between the concrete and the boundary members. Normally that's going to be headed welded studs. Maybe DBA if your slab was a foot thick or it tied into the sides of your beams etc.

For what it's worth (sounds like OP is talking about slab on deck though), we commonly do a similar check when designing PT diaphragms. More often than not, the effective prestress (essentially a proportion of the prestress that accounts for the difference between 1.2DL + 1.6LLf1 vs (1.2 + 0.2Sds)DL + rQE + L + 0.2S) available in seismic load combos (usually around 30-60psi...depending on a lot of stuff..) is enough to overcome the bending tensile stresses, such that the resulting stress is still well below concrete rupture modulus. Also in the intermountain region. SDC = D all day every day.

Effective prestress (or available *extra* prestress in those combos) can be utilized when designing chords and collectors. In the case of collectors, the shear strength of 'plain' concrete (2sqrt(f'c)bd) on it's own is considerable (diaphragm as very deep beam analogy). The amount of drag steel can be reduced dramatically compared to designing drag steel for essentially the entire drag force (i.e., how one would design drag straps/coil straps for a wood diaphragm, for example). Lots of caveats though, of course.

SK Ghosh has 2 good webinars on this methodology.

I feel like there was a thread some time ago discussing something similar about chords and drags - utilizing metal deck as part of the chord 'area'. But in the case of a steel framed structure, I generally just size my beam/column connections to handle the chord forces that may develop, and neglect any contribution from the edge angle. Much easier to do with wideflange columns than with HSS columns. Generally there are lots of jogs and discontinuities in the angle to be useful anyway. And yes, the framing is sized and braced/kicked to handle the axial forces.

Samwise: My mistake, when I checked your profile it showed CA. If you are in SDC D, E, or F my comment stands. Like dold, we also use prestress to reduce the chord / collector demand in PT slabs, but PT slabs are reinforced, not plain as it sounds like you may have been proposing.

Deker , thank you for the response. I don't really like the idea of designing the diaphragm as an unreinforced deep beam. But at the same time, I might use some of the existing mesh and maybe add a rebar.

Samwise - how are you connecting the slab to the collectors?

The deck/ composite slab is connected to the collectors using puddle welds. Given that its a composite beam framing, I also added a note saying that headed studs in construction may be substituted for puddle welds.

Ah, full composite. Got it. If you already mentioned that the beams were also composite, I apologize - I only picked up that the deck was composite. In that case, you can use the concrete as the diaphragm. But the puddle welds do nothing for you. If the concrete is the diaphragm, ALL load has to go into the concrete and then out again through the studs along the beams/collectors.

If there is a way to mix the two, I imagine it would be unnecessarily complex for the vast majority of practical design situations.

phamENG, I am exactly following the load path that you quoted. Diaphragm transfers lateral load per unit length into the collectors/braced members through additional studs that I added in the beams. For the chord forces, I would have liked to simply use the edge angle and be done with it. But given that the edge angle is discontinuous, I was trying to use the concrete diaphragm to take that chord force through a rebar or just the existing mesh I have. I am just not convinced that the diaphragm can be treated as an unreinforced deep beam and just check for cracking.

Correct. If the concrete is going to be a diaphragm, it should be detailed in accordance with ACI 318. I practice on the east coast and I'm not overly familiar with the Intermountain Region - but I take it you are in a relatively high seismic area. That's going to mean plain concrete isn't allowed. You'd have to go with reinforcing and all of the prescriptive ductility requirements. That could very well mean your concrete slab has to get thicker to accommodate everything.

If the loads work, I recommend keeping it in the steel. Forces go into your deck and then get transferred to the collectors (beams, spliced angles, whatever you want to use) through the puddle welds. The concrete just stiffens the diaphragm, which increases its capacity and in most instances allows you to assume a rigid diaphragm for global analysis. If you can't assume (irregularities, etc.), then the calculations usually work out for it to be rigid or very close to the rigid end of 'semi-rigid'.

I am in Seismic Design Category B fortunately and the structural is so symmetric that I have no torsional irregularities.

Oh. Nice. Well that makes things a little easier.

You still have to design it per ACI 318, though...318-14 Chapter 12 if your local code references the 2014 edition. 2019 is probably in the same spot.

I don't think a diaphragm fits into any of the options in Chapter 14 (plain concrete). So you'd be in reinforced concrete. The normal way to handle this is through the steel deck diaphragm. Reinventing this wheel will probably cause headaches in construction.

Samwise Gamgee said:

For the chord forces, I would have liked to simply use the edge angle and be done with it. But given that the edge angle is discontinuous, I was trying to use the concrete diaphragm to take that chord force through a rebar or just the existing mesh I have.

Click to expand...

Your detail indicates that the edge angle is continuous with butt-welded splices. That said, I don't see an issue with using the slab reinforcement as a redundant chord/collector element along with the beam line and edge angle. Doing so makes me *feel* better from a strain compatibility perspective. In other words, there are no breaks in the angle, beam line or slab reinforcement where the axial load has to "jump" from one element to another - just multiple load paths that can strain uniformly.

bones206 said:

Doing so makes me *feel* better from a strain compatibility perspective. In other words, there are no breaks in the angle, beam line or slab reinforcement where the axial load has to "jump" from one element to another - just multiple load paths that can strain uniformly.

Click to expand...

So I am trying to put all of this together. On the roof, yes I will have a continuous angle to behave as a chord member.

However on the slab, I can't have the a continuous angle . So I was going to trim the angle at the column. The angle goes from a HSS 5x5x3/8" to an HSS5x3x3/8".