tube collectors between joist seats 3

[smvk3]

I am designing an occupancy category IV structure that is in a seismic design category of D. The structure is steel bar joists on CMU bearing-shear walls. Where the joists bear on top of the CMU walls, to transfer the diaphragm load to the top of the wall, I am debating whether to use tube collectors between the joists seats per the attached or utilize the joist seats themselves as a rollover load. The question I have is would I need to amplify the load on either the joist seat or the tube collector by the overstrength factor to satisfy the intent of section 12.10.2.1 of ASCE 7-05? Without the overstrength factor applied, the rollover load at each joist seat would be between 2,000 and 4,000 lbs.

 

[KootK]

If you'll notice, I'm not as hung up on the EXACT force to be used as I am about making an effort to ensure that collectors and shear transfer elements are proportionally stronger than the SRFS.

Acknowledged. That's precisely what I was alluding to with this statement:

But, yeah, maintaining a consistent design philosophy is important, even when great precision is not available to be had.

Glad to have you on board!

It may be short lived. I'm now thinking that:

1) Connection ductility is largely irrelevant and;

2) Every collector connection, whether ductile or not, should be designed for over strength (or whatever is meant to capture a degree of capacity design).

I'll post a sketch when I've got it worked out in my head to my satisfaction...

 

[Deker]

2) Every collector connection, whether ductile or not, should be designed for over strength (or whatever is meant to capture a degree of capacity design).

Certainly if it's a discrete collector connection in the traditional sense, which is spelled out clearly in the code. And ideally it would be applied to all distributed shear transfer and collector elements, although that bit is open to interpretation, i.e. where does the diaphragm end and the collector begin? For those who wish to consider these distributed elements as part of the diaphragm, at a minimum the failure mode should be considered when choosing whether or not to amplify the loading.

Looking forward to your sketch...

 

[sticksandtriangles]

In reality, you almost always have distributed chords which could amplify the peak shear connection demand by 50%. This, in reference again to the potential unzipping of brittle things.

I am not following this train of thought, care to explain? I've heard of distributed chords before, but I am not seeing how this could increase peak shear.

S&T -

http://www.re-tug.com

 

[bones206]

Just stepping back a bit and looking at this structure holistically, the out-of-plane anchorage for top of CMU wall will be critical. The code explicitly amplifies these anchorage loads based on observations of poor performance in earthquakes, especially with flexible diaphragms. These OOP anchorage forces may govern the connection design and supersede the in-plane shear forces. Seems like the more important consideration from a collapse prevention perspective.

 

[KootK]

@S&T: you know how the peak shear stress in a rectangular cross section is [1.5 V / A] rather than just [V / A]? Similar thing here. Let me know if that doesn't do the trick.

 

[KootK]

where does the diaphragm end and the collector begin?

Yeah, that. Maybe I can work with that instead of a sketch for now. Humor me.

In your opinion, what is it about a diaphragm that justifies it being designed without over strength? If the answer is "ductility", then what, specifically, is it about ductility that justifies not amplifying the diaphragm load?

 

[bones206]

Excerpt from FEMA P-1026 - Seismic Design of Rigid Wall Flexible Diaphragm Buildings: An Alternate
Procedure

Observation of earthquake damage to the in-plane rigid shear walls or the main flexible roof diaphragm has been rare, except for collateral damage from the out-of-plane wall anchorage issues. The perimeter shear walls often consist of largely solid wall portions with relatively few penetrations, resulting in inplane lateral strength significantly in excess of that required for seismic forces. This inherent overstrength of the shear walls transfers the inelastic building behavior into the diaphragm. It is important to consider that the out-of-plane detachment of the heavy walls from the diaphragm in the past may have protected the diaphragm from experiencing in-plane seismic forces which could have led to global failure. The overstrength of the walls, combined with the code required higher wall anchorage and collector force levels, could potentially make diaphragm yielding, foundation rocking or sliding, and global response more critical for RWFD buildings in future earthquakes.

 

[bones206]

Some further relevant discussion from the FEMA Design Guide:

The main point here is that RWFD-type structures' yield mechanism in the diaphragm. So the boundaries need to be super strong to permit a more distributed development of inelastic behavior deeper into the interior regions of the diaphragm. In my mind, it's clear that the boundary connections should be designed with a substantial overstrength factor. Especially given the fact that metal deck yielding is not a well-defined mechanism and thus actual boundary forces in a seismic event could significantly exceed the design forces we come up with using R-factors.

4.2.1 Encouraging Distributed Inelastic Behavior
Analytical studies have demonstrated that the performance of a diaphragm during strong earthquake shaking can be improved if yielding is spread over a large portion of a diaphragm’s span instead of focused at the boundary. The spread of diaphragm yielding is improved if the location of initial yielding is shifted away from the boundaries of the diaphragm. Surprisingly, this can be achieved either by intentionally weakening a portion the diaphragm’s interior areas below current building code levels, or by increasing the strength-to-demand ratio of the diaphragm near its boundaries.

Distributing the inelastic behavior deeper into the diaphragm also requires that the diaphragm connectors that yield first exhibit sufficient positive post-yield stiffness behavior. In other words, once the critical connectors begin to yield, they need to resist increasing load rather than maintaining the initial yield load so connectors elsewhere in the diaphragm also reach their yield load. The connectors that yield must also have sufficient post-yield deformability so the diaphragm as a whole has adequate post-yield deformability to provide the required collapse resistance. The overall diaphragm deformability increases if connector yielding spreads over a large portion of the diaphragm. Also, the spread of yielding reduces the deformation demand on individual connectors. The hysteretic responses of the nail connectors in the wood structural panel diaphragms have sufficient positive post-yield stiffness to effectively move the location of first yield away from the diaphragm boundaries and spread the yielding over a large portion of the diaphragm’s span. In the case of steel deck diaphragms, insufficient data is available to make that determination. Thus, the recommendations for spreading diaphragm yielding are limited to wood structural panel diaphragms at this time. An alternative for steel deck diaphragms could involve simply using a smaller Rdiaph until sufficient data is obtained to justify an approach that distributes inelastic behavior. An appropriate Rdiaph for steel deck diaphragms where the spreading of inelastic behavior is beyond the scope of this report.

 

[KootK]

I'd not meant to target the RWFD typology in this discussion but I think that I can use it to make the point that I've been thinking of from another angle.

For energy dissipation, all buildings designed conventionally for seismic require some degree of inelastic, inter-story drift between the ground and the significant mass sources. In a RWFD building, that happens within the diaphragm. On a 30' tall big box building, that drift my be on the order of 6" at the middle of the diaphragm. And it's that capacity for significant ductile displacement that allows us to cap the forces developed within the structure at roughly the VLFRS mechanism formation values.

Conversely, there aren't many ductile, diaphragm boundary connections schemes that could tolerate even 1" of ductile displacement without being torn apart. As such, it seems to me that ductility cannot be assumed to cap the forces developed within the structure near the VLFRS mechanism formation values.

I'm not saying that ductility in the boundary connections is bad. Ductile stiff is almost always an improvement over brittle stuff. I simply question whether the ductility available in these kinds of connections provides enough inelastic drift capacity that one can consider it to effectively aid in the capping of seismic force development. If not, then I feel that boundary connections ought to receive overstrenght treatment even if they are nominally ductile.

 

[bones206]

I'd not meant to target the RWFD typology in this discussion but I think that I can use it to make the point that I've been thinking of from another angle.

Understood. I was making a conscious effort to steer the discussion towards this specific typology, since I think the OP's detail used mostly within that classification of structures. Given that RWFD buildings response to earthquakes is fundamentally different, I thought it important to put the discussion of boundary/collector overstrength into that context.

I agree with everything you stated in your last post Koot. Given the nature of the connection details we have for metal deck to CMU or tilt up walls, there's not much ductility to be had. That's why I say overstrength is the only real way to go in this situation. Not only for in-plane shear, but for the out-of-plane anchorage as well.

 

[Deker]

In your opinion, what is it about a diaphragm that justifies it being designed without over strength? If the answer is "ductility", then what, specifically, is it about ductility that justifies not amplifying the diaphragm load?

I don't think it is justified in the spirit of traditional capacity design where we are expected to limit all yielding to the vertical elements. But I think it's implied by the design requirements of ASCE 7 that some diaphragm inelasticity is expected and is perhaps even beneficial in terms of total energy dissipation and force reduction by way of period elongation. However, it does present issues with inconsistency between the R values used for vertical elements and the available ductility in the diaphragm. There's a lot of work that's been done recently to address these issues, with FEMA P-1026 that bones206 referenced being one resource. There's also the alternate diaphragm design procedure that was introduced in ASCE 7-16, the 2015 NEHRP Recommended Provisions Resource Papers, and a ton of research that's been taking place as part of the Steel Diaphragm Innovation Initiative.

As for the logic behind designing a distributed shear transfer element without overstrength, I think it goes something like this: If some diaphragm inelasticity is expected to result from normal use of the ASCE 7 provisions, that inelasticity is likely to occur at high shear regions near vertical elements / collectors. If there are ductile distributed shear transfer elements at the same location and directly downstream from the point of peak shear in the diaphragm, it doesn't matter if yielding at that location occurs in the diaphragm or in the shear transfer elements. Yielding in either will limit the force that can be delivered to the other, and the global behavior of the diaphragm would be the same in either case.

I don't necessarily agree with that design strategy, but I can't say it's unreasonable compared to standard practice.

 

[bones206]

If there are ductile distributed shear transfer elements at the same location and directly downstream from the point of peak shear in the diaphragm, it doesn't matter if yielding at that location occurs in the diaphragm or in the shear transfer elements.

I agree with this as a hypothetical concept, but in practice the available ductility of these shear transfer elements is very limited. Maybe someone will invent some innovative rigid shear wall to metal deck seismic connectors that absorb tons of inelastic energy, but I think the current reality is that ductility at the boundaries is inadequate for significant earthquakes. Therefore you gotta go overstrength.

 

[Gopher13]

1) If I can make a go of it with joist rollover, I need to do that to stay competitive.

2) If I can make a go of it with a tube lug every other joist space, I need to do that to stay competitive.

I have been out the real world for a number of years now (government employee), but is omitting the tube or going with half the number going to make much difference in the overall cost of a building? Just curious.

 

[jayrod12]

I have been out the real world for a number of years now (government employee), but is omitting the tube or going with half the number going to make much difference in the overall cost of a building? Just curious.

I wouldn't say it makes a substantial difference. However to the more frugal owners/contractors, every dollar matters. So if engineer A calls off the welded collector all the time everywhere, and engineer B only calls it off when absolutely necessary, those people are going with engineer B 100% of the time. And most extremely profitable clients are the frugal ones who pay attention to the small things.

My grandpa used to say "Worry about the nickels and dimes, the dollars will take care of themselves." Sometimes I look at my bank account statements and wish I listened to him even more.

 

[Gopher13]

Thanks jayrod12!!! Makes sense.

 

[KootK]

@Deker: thank you for your detailed explanation. For now, I think that I'll just have to accept that as the limit to my understanding when it comes to reconciling ASCE design philosophy with general, capacity design philosophy.

I have been out the real world for a number of years now (government employee), but is omitting the tube or going with half the number going to make much difference in the overall cost of a building?

My answer matches Jayrod's:

1) Let's say we're talking $20K worth of cost on a big building to install those lugs on top of masonry wall embeds.

2) That $20K is probably nothing in terms of the overall cost of the development. Like, < 0.5% of the cost kind of "nothing".

3) That same $20K might buy four used cars for the kids of four of the contracting firm's principals.

So it's that constant juxtaposition between what is a small number for the owner and what is a large number for the contractor. I can tell you that, unless the owner has also been the contractor, I've never had an owner call me up to bitch at me because my deck angle is bolted to a block wall at 24" oc rather than 32" oc.

 

[jayrod12]

unless the owner has also been the contractor

Which is one reason I hate doing Design-Build projects. So much of that bitching it makes the work not worth the hassle.

 

[bones206]

Would be nice if you could make the tubes continuous, with shop welded studs. Then offset them enough to clear the joist seats and run by. Then the masons just have to set them on top of the wall and grout. Way easier and cheaper than having a field welder up there measuring and welding a thousand little welds for a week.

 

[Gopher13]

kids

Soft spot. ya got me. I will rely on joist rollover from now on.

 

[jayrod12]

Wouldn't offsetting it to miss the joist seat just put a whole bunch of additional eccentricity in the wall to change that design? Most joists need a minimum of 2 1/2" of bearing, if centered on the wall to minimize the eccentricity to the minimum value then that really only leaves you with 2 1/2" from face of CMU to edge of joist seat. Not really a whole ton of room. I guess you could put a 4x2HSS LSV there, but your studs you were talking about would never make it into the cores.

It's one of those damned if you do, damned if you don't type scenarios. Welding costs more, or your wall does. There's no free lunch.