Splitting a building into two risk categories 6

[ggcdn]

Does anyone know of any examples or references that describe buildings with dual risk categories?

We are doing a performance-based seismic design in California, and would like to propose splitting the project into two risk categories because of the different usage and structural configuration (typical residential tower above, larger office podium below).

I know of one case where this has been done for multiple towers landing on a common podium structure. Any other references or past precedent would help our case.

Thanks,

 

[Agent666]

So which one are you trying to argue is higher/lower risk?

Way I see it, if tower is on podium the entire thing should have one risk category, all designed by the same loads.

Think of it this way if podium is designed for higher load, then tower is closer to collapse limit state under that load (compared to podium). Surely part of life safety for someone in the podium is the assurance that the tower isn't going to come down on top of them...

Only way I'd see this being acceptable is when structures are separate, seismic separation between them.

 

[ggcdn]

The residential tower qualifies for risk category II, but the podium pushes the structure into category III.

I understand your viewpoint, and have been struggling with this argument myself. Hence why I'm looking for examples.

Through the PBD procedure, we are already arriving at a design that has a far lower probability of collapse than a code-based design. I suppose one path forward would be to quantify that probability, and show it lands within risk category III. But that is easier said than done.

Regarding your hypothetical, consider these scenarios:

- Risk Cat II tower beside a theatre (or other venue which would be Cat III).

- Risk Cat II tower with a podium that would qualify for Risk Cat II, except it extends horizontally a block away and includes a theatre

- Multiple Risk Cat II towers on an interconnected podium, that, if it were divided up to its tributary towers with separations, could also qualify for Risk Cat II.

 

[Agent666]

Clause 1.5.2 of ASCE 7 basically says if there is something category III in the building, even part of it, then the entire thing is to be classified as category III. (I'm not from the states but I assume this is what you are working to)

What are the exact wording that classifies the podium as III, maybe there is some leeway there?

 

[ggcdn]

Thanks Agent, I've reviewed the relevant ASCE 7 clauses and commentary, it's pretty clear that, taken at face value, it would require the entire structure with a single category unless structurally separated. And yet I know of two examples where this hasn't been followed. The thing about PBD is we get to essentially write our own code (design criteria) as long as we can convince the peer review panel. So I am trying to determine the reasoning why it was acceptable for these past examples.

The most recent LATBSDC guidelines describe how the split risk category should be handled, analytically. But it seems to be targeting the situation of multiple towers on a shared podium.

 

[KootK]

And yet I know of two examples where this hasn't been followed.

Were both of those examples different superstructures over a common podium?

I dealt with something related for wind on a NY tower. 40 stories of residential condo over a couple of stories of senior care space. It seemed punitive that the whole building should be designed for higher importance wind just for the sake of those couple of stories at the bottom. We concluded that was, in fact, rational however for reasons similar to Agent's thinking. The intent is a higher margin of safety for the folks in the precious occupancy. And it's tough to argue that the seniors would have had that margin of safety if only their floors had been designed for the higher wind, which is something that was briefly considered.

I suspect that you'll struggle to sell different risk categories stacked vertically, particularly with the higher risk category being at the bottom. Where the spacial separation between risk categories is horizontal, that makes more sense to me. Especially if the following are true which, often, I'm sure is the case:

1) The two occupancies have independent lateral systems above the podium and;

2) The two occupancies have independent egress through the podium.

 

[jayrod12]

I don't practice south of the border nor for seismic, so forgive my ignorance, but why would designing the tower above the higher priority have any effect on the lower portions, provided the lower portions are designed for the high level loads.

To be clear, my approach would be, design and detail the upper portions with the low risk occupancy for the appropriate loads, design the higher occupancy level at the lower floors as if the entire building above is at the higher loads. I'm not sure why that exposes the higher risk occupancy to a problem.

 

[KootK]

I'm not sure why that exposes the higher risk occupancy to a problem.

For my wind example, the higher importance factor is basically saying that the higher risk occupancy should be able to survive a more demanding wind event. And it's not as though that hurricane is just going to blow against the two stories full of old folks. If that hurricane blowing against the forty stories above the old folks would push the building over, then I don't feel that you could plausibly claim that the code intended hazard risk reduction had been achieved at the lower stories.

I know, it makes one's head hurt a bit. And it produces results that feel a bit unreasonable at times.

 

[KootK]

Actually, I think that I see what you're saying a little more clearly now having noodled on it a bit. For my example, one would design the lower floors for the hurricane assuming that the demand in the lower floors originated from the hurricane wind applied to all the floors. Then the upper floors could just be designed for the wind event associated with the lower risk occupancy. I'd think that would be kosher so long as a hurricane event would not cause damage to the upper floors that would, itself, compromise the lower floors.

With respect to seismic, if the more severe event would trap egress elevators up high, or cause the upper part of the building to pancake onto the lower, then obviously that would be a problem.

 

[jayrod12]

Your interpretation is correct of my intended scenario.

I can also understand the second scenario

 

[ggcdn]

Koot, one example was two towers on a shared above-grade podium, the other example is that building I posted above, which is a single tower with 5 story podium. In my project, there is a separate bank of elevators that exclusively service the office.

I was thinking about this scenario similar to jayrod's interpretation. The entire building is designed for Risk III demands from foundation to top of podium (including our code-based design and MCE collapse-prevention assessment). In fact, we could do the entire code-based design for Risk Cat III, it doesn't really matter. My main objective for this is that I don't want to arbitrarily increase the demands coming from the nonlinear collapse-prevention analysis by Ie=1.25 for the upper tower portion of the building (or reduce the deformation capacities of nonlinear elements by 1/Ie).

I think this makes sense, intuitively. It's of little consequence to the office below that some coupling beams or slab outriggers in the tower went to 6% rotations instead of 5%, or that the tower drifted 3% instead of 2.4%. We aren't designing a post disaster facility - there is no expectation of damage-free performance.

 

[Agent666]

One way to think about it as well is for the higher importance structures, all the codes are saying is that the actual design load is closer to the MCE (maximum credible earthquake) level event, it's not raising the bar for the MCE level event as well in proportion as I understand it.

My understanding of the 2.5% typical drift limit is that it's intended to ensure in the typical swing of things that a structure still has sufficient robustness to get to the MCE with drift that might well exceed 2.5% but with a low probability of collapse occurring.

Higher importance structures simply have a higher degree of robustness, especially at MCE level events.

It's of little consequence to the office below that some coupling beams or slab outriggers in the tower went to 6% rotations instead of 5%, or that the tower drifted 3% instead of 2.4%.

Possibly this is the way to prove it, provided that drift isn't initiating some sort of collapse limit state that might impact on the tower, and hence podium below. The code you posted the picture from clearly states you can separate it out.

I've seen one structure in this part of the world that has a new dual rail tunnel going diagonally through part of the basement with the 39 storey podium and tower straddling it with a few mega columns between the tunnels, this forced the entire structure to be designed for 1.8 times the typical important level seismic load because of the higher importance level tunnel. I believe it was an active decision taken by the designer that obviously kept the tunnel stakeholders happy. There wasn't any attempt to weasel their way around it that I know of even though tunnel and building are seismically separate as far as I understand.

 

[KootK]

Having spent some more time noodling on this, I have these additional thoughts:

1) Here, I'll elaborate a bit more on the wind example that I mentioned previously. It'll become clear why further on. In a previous design iteration, the building was a fair bit taller and had a true outrigger system comprising story high CIP walls that reached out from the central core to grab outrigger columns. As one would expect, those outrigger floors were placed well above the lower, risk category III occupancy levels. This was really where our concern originated with respect to the appropriate risk category business. With this setup, a cat III event acting over the entire building would compromise the outriggers located on the cat II levels.

And, once the outriggers were compromised in this way, it would be difficult to argue that the lateral system for all levels of the building would not be compromised. Composite action between the core and the outriggers would be lost and you'd have a very different set of forces in the lower level walls and, in particular, in the foundations. So, in this instance, it makes sense to me that the lateral system of the entire building ought to be designed to cat III unless it can somehow be demonstrated that, during a cat III event, the building would still be viable without the outriggers providing the connection required for composite behavior among the various elements of the intended lateral system.

I think this makes sense, intuitively. It's of little consequence to the office below that some coupling beams or slab outriggers in the tower went to 6% rotations instead of 5%, or that the tower drifted 3% instead of 2.4%. We aren't designing a post disaster facility - there is no expectation of damage-free performance.

2) I disagree and see this as being quite analogous to #1. All of the elements that you mentioned are the very ones that will be producing a degree of composite flexural behavior in your lateral system. And, as with my wind outrigger example, it cannot be said cavalierly that the assumed composite behavior of the lateral system in the Cat III levels will not depend on the integrity of the structural elements up in the cat II levels that help to provide that composite behavior.

If a coupling beams's curvature level would render it no longer fit for purpose during a cat III event, then it seems rational to me that that beam's contribution to composite action of the entire lateral system ought to be discounted as well. Of course, if you've got 200 coupling beams in the cat II tower and only four of them would see excessive damage, then a sound argument surely could be made for allowing that to happen.

Surely part of life safety for someone in the podium is the assurance that the tower isn't going to come down on top of them...

Regarding your hypothetical, consider these scenarios...

3) I agree with Agent's reasoning on this and, frankly, I believe that it probably should apply to an independent, cat II tower constructed next to a shorter, cat III building. That said, I also find SOA's hypotheticals to be salient examples which make it abundantly clear that this is not what we are doing regardless of what we ought to be doing. This may ultimately just be a mater of practicality. The logistics of making one, cat II building owner design his building to a higher standard such that it not collapse on a current, or even future, neighboring cat III building would surely get a bit nuts. Lastly, the LATBSDC tower/podium example makes it clear that we're also not to be worried about cat II towers collapsing upon cat III podiums during cat III events.

4) I read into the LATBSDC docs a bit last night and, as a result, need to change my answer on whether or not you'll be able to sell a dual risk category tower to your peer review panel. Based on what I've seen in the LATBSDC, docs, I think that you've got a legitimate shot. I can only think of one thing that makes the LATBSDC examples a bit different from yours and it is, admittedly, a reach. In general, a podium will possess lateral resisting systems separate, and in addition to the lateral systems coming down from the tower(s). As such, tower collapse excepted, a loss of lateral system composite behavior in level is probably not as much of a concern as it might be when the podium level lateral systems are really just vertical extensions of the tower lateral systems. I didn't see anything in the LATBSDC docs that specifically mentioned this aspect of things however.

So yeah, I now think that you've got a pretty good shot at being able to push this through. At the least, hopefully the extra vetting here will help you to cut your teeth one the problem and be better prepared to answer any objections that may be voiced.

 

[KootK]

I'm going to be sneaky and summon Deker and Sandman21 over to this thread. They're both LA guys and would be perfect for this.

 

[sandman21]

The large towers I know had the same risk category, III in the couple cases I know about. But the lower portion was designed as ELF and tower as PBD, creating two systems. The 2017 LATBSDC section 3.6.3.2.3 seems to cover your condition, "Where multiple towers on a common podium or base create a situation in which the number of occupants at or below the podium or ground level may exceed 5,000 persons, then the value of Ie = 1.25 shall be used in conjunction with application of Equations 5a, 5b, 6a, and 6b for all force-controlled actions including those of the podium diaphragm and below, including the foundations placed under the Risk Category III portion of the structure and all force-controlled elements of the tower passing through the Risk Category III portion of the project. Acceptance criterion for all deformation-controlled elements of the tower passing through the Risk Category III portion of the project as well as all deformation-controlled actions of the first level of each tower immediately above the common podium shall be considered as 1/Ie times the acceptance limit for a similarly situated the Risk Category II building.". I am surprised by the fact that your office building and tower would have more than 5000 load. If you are in a jurisdiction outside LADBS, they will likely follow LATBSDC for how to handle the conditions.

 

[wannabeSE]

In simple terms, I wouldn’t buy into a risk category II on a risk category III. If the risk category II fails, it falls on top of the risk category III. I can’t imagine designing a podium for loads from a collapsing tower. I can’t think of a compelling argument to justify a II on top of a III.

 

[Deker]

As long as the tower itself qualifies for Risk Category II, it would be common to design the podium and only the deformation-controlled elements of the tower adjacent to the podium for the enhanced acceptance criteria per the LATBSDC procedure referenced by sandman21. Some additional commentary is provided in the PEER/TBI Guidelines which I have copied below (emphasis added). As noted, some engineers will try to relax the above requirements even further by designing only the deformation-controlled elements of the tower adjacent to the podium for the enhanced acceptance criteria.

"The Authority Having Jurisdiction should make the final assignment of Risk Category based on the requirements specified in the building code. These Guidelines provide the recommended technical approach to demonstrate that a tall building is capable of providing performance equivalent to that anticipated of buildings conforming to the prescriptive requirements for Risk Categories II, III, or IV.

Risk Category assignments are defined through the building code consensus process at the national level. The current assignments, located in IBC Table 1604.5, are a culmination of decades of discussion and balloting. Some local jurisdictions have legislative authority to modify these assignments.

Under IBC Table 1604.5, Risk Category II is assigned when the building is not assigned to Risk Category I, III, or IV. Most tall buildings fit into this category. Risk Category III is assigned to “Buildings and other structures that represent a substantial hazard to human life in the event of a failure…” Included in this category are buildings “with an occupant load greater than 5000.” The IBC approach to calculating the number of occupants is based on egress considerations, which are intended to be conservative and typically result in occupant loads in excess of that normally present. Despite this conservative occupancy calculation, most residential towers will not exceed the 5000-person threshold. However, office buildings with a building area of approximately 750,000 sq. ft. or more will exceed this threshold.

In some cases, tall buildings may share a common podium with other buildings, resulting in a combined occupancy for the podium and supported buildings that exceeds the 5000 person threshold. In such cases, the Risk Category assignment should consider whether the multiple towers rely on overlapping areas for egress.

As noted, Risk Category assignments are the responsibility of the Authority Having Jurisdiction. For multiple towers on a common ground-level podium, some jurisdictions have adopted the approach of assigning the Risk Category for the podium based on the combined occupancy of the supported towers. For the case where Risk Category III is assigned, some jurisdictions only require that the enhanced acceptance criteria be applied through the plastic hinge region of towers that utilize concrete shear walls as their seismic force-resisting system. This approach is based on the judgment that the most significant damage is likely to occur in the plastic hinge zone, and if these zones are protected, the building complex should perform to the expectations for a Risk Category III structure at, and adjacent to, the podium level. Where seismic joints are provided at the podium level, the Risk Category should be assigned based solely on the occupancy within the bounds of the expansion joints because each building responds independently to the earthquake shaking."

 

[ggcdn]

Thanks everyone for your comments. This gave us lots to think about and preempt the responses we think we will receive. I prepared a small exhibit for the city showing the occupancy calcs from the Architect, risk category division, how we will modify our design criteria, and our rational for doing this.

Now we get to wait and see what they think.

 

[beaklio]

Informative thread. For those who have building with multiple categories, I think, it would be better to reassess the structure and make necessary changes.

 

[ggcdn]

Update: Unsurprisingly, the city has said they want the code-based design to be completed solely as Risk Category III, and the dwgs to state one risk category. But how that affects the MCE analysis performance criteria is up the the peer reviewer panel.

So now we move on to that discussion.