Storm Shelter Host Building Design 2

[SamB35]

We are designing a building in a tornado region of the country with a storm shelter as part of the building. The main building is a multi-unit residential facility. The storm shelter is attached to the host building. I have a couple questions I hope someone can help me out with. We are using ICC-500, FEMA 361 and 2012 IBC.

1) The risk category for the shelter would be IV as and emergency shelter and the host building as a risk category of II. Per 2012 IBC section 1604.5.1 when you have two or more categories in one structure, then you use the higher risk category for both, making the host building risk category IV for wind and seismic. Can someone confirm that this is the correct procedure?

2) Per ICC-500 section 308.1, host building framing is to be designed for wind forces equal to or greater than the design wind forces for the storm shelter when it is connected to the storm shelter. So my interpretation is that the diaphragm of the host building adjacent to the shared shear wall should be designed for the full tornado wind loads (per ICC-500) and the shared shear wall should be designed to take the tornado wind loads from both the shelter and the adjacent host building diaphragm, but the the rest of the host building can be designed for the standard wind loading for this area.

Thanks in advance for any insight.

Sam

 

[SlideRuleEra]

The answer to your first question is: The entire structure does NOT have to be Category IV.

The question appears to have come up before. See ICC's interpretation and explanation here:
Storm Shelter & Host Building

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[JAE]

ICC 500 - 14 has revised 308.1 (from ICC 500-08)...which was very confusing.

The new section in the 2014 version reads:
[blue]304.8 Shielding of Storm Shelters by host and adjacent buildings
Storm shelters enclosed in, partially enclosed in or adjacent to host buildings or adjacent to other buildings not designed for the load requirements of Chapter 3 shall be designed considering the host buidling and adjacent buildings to be destroyed and the shelter to be fully exposed.[/blue]

[blue]304.9 Storm Shelters connected to host buildings
Where an element or component of the host building is connected to a storm shelter, the storm shelter shall be designed to resist the maximum force that could be transmitted to the shelter equal to the ultimate failure strength of the connection or element being connected, whichever is lower, concurrent with the other wind loads on the storm shelter required by Chapter 3.[/blue]


So with the above, your diaphragm issue would be that you need to design the shelter for the tornado loads assuming all other surrounding buildings are "gone".
And
...you need to design the storm shelter for added loads at each point where the host building is connected to the shelter. These loads would be based on the ultimate failure strength of the connection or element being connected, whichever is lower.

Under ICC 500-08 your importance factor for tornado shelters is listed as 1.0. Thus, the risk category does not affect how you determine the tornado wind loads.
For your main question - does the host building then require a Risk Category IV just because it has a storm shelter in it....

I've never heard of requiring a host building to be kicked up to a Risk Category IV for that reason. Seems counter-intuitive.
The commentary in IBC 2012, section 1604.5 does not mention tornadoes but does mention hurricanes when speaking of Risk Category IV.

Also - in the commentary of 1604.5.1, it states that you can "separate" the two occupancies and thus have two different risk categories.
It states: "Alternatively, the engineer can structurally separate portions of the structure containing distinct occupancy categories and design each portion accordingly based on its risk category.
For the quoted ICC 500 sections above, it seems to me that designing the storm shelter as its own tornado-proof entity, and assuming the host building is destroyed, is a sense of a structural separation in that the shelter doesn't depend on the host building for strength and thus doesn't require the host building to be kicked up a notch.

You also don't typically have egress paths going from the host building through the shelter area. But you might need egress from the shelter through the host building....if so, they you need to kick up the host building or provide a separate, distinct egress from the shelter independent from the host building.











Check out Eng-Tips Forum's Policies here:
faq731-376

 

[JAE]

Seems that what SRE posted (I didn't see it before I posted mine) was a nice response and helps clarify the issue.

I don't totally seem to agree with the issue of having an exitway/egress from the shelter and through the host building though. That has always concerned me that shelter occupants leaving would have to wade through collapsed steel and even open their shelter exit door into debris blocking the exit.

Check out Eng-Tips Forum's Policies here:
faq731-376