Vehicle barrier load

[WAstruc10]

Section 1607.7.3 of the 2003 IBC requires a 6000# horizontal load (due to vehicle impact) applied on a 1.0 square foot area. I have a parking garage situation with CMU infill wall along part of the edge, which must be designed for this load. My question is, how wide of a "design strip" would seem appropriate for analyzing as a vertical "beam" - obviously 1 foot is way too conservative considering the effect of the horizontal bars, but going much past 3 feet sounds uncomfortable...and even a 3-ft width seems to require solid grout for shear and a whole lot of rebar...

Any advice?

 

[youngstructural]

You've just encountered the very good reason you don't see many concrete masonry unit parking garages. I know the Canadian code has a similar clause (sorry, can't remember it too well at the moment, and am currently working in New Zealand so it's not exactly in my back pocket). You might want to try looking into dip top blocks (the kind which can have actual steel bar horizontal reinforcing rather than steel ladders for your horizontal bars. If you did that you'd be able to use a two way slab model (either by effective strip methods or yield line theory) to carry the load two ways into your structural frame. This type of block is pretty rare in Canada (although is available), but is very common here in New Zealand. See this link for a picture of men working with them (sorry, couldn't find a good drawing on short notice:

http://www.firth.co.nz/ProductItem.aspx?pitemid=15

I would imagine they are available in the US.

In a similar situation (not my design, but a project I was involved in) the engineer of record specified a steel guarding rail with the horizontal at ~600 (2ft), and argued that with it in place, the code clause requiring the horizontal thrusting load for the wall no longer applied (the cars would, at least in theory, never impact the wall).

Hope that helps at least a little. Tough situation... Good luck, and let us know how it all turns out!

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[WAstruc10]

Thanks youngstructural...a 2-way slab analysis (besides being time-consuming) will not help much because of the length of my wall - definitely 1-way action, so if there was a way to justify spreading the load wider than 3-ft without a FE analysis it would be helpful.

But more importantly, who do I send my resume to in order to get a structural engineering job in New Zealand?

 

[Taro]

With appropriate load-distributing reinforcement, I would use ACI 530 Section 2.3.3.3 as a guide. This gives the effective width for flexure in running bond walls as the least of:
a. bar spacing
b. 6x wall thickness
c. 72"

I would also think that you would want to solid grout the wall no matter how the numbers calc out. A car hitting the hollow part of a partially grouted wall could send a lot of small missiles flying.

 

[miecz]

Taro-

ACI 2.3.3.3 is the "Effective compressive width per bar". I don't think it applies here, as that would mean that the design strip width would increase with wider bar spacings.

WAstruc10-

AASHTO (17th Ed.) Article 2.7.1.3.6 calls for a 10 kip transverse force on concrete barrier walls to be spread over a longitudinal length of 5 feet.

 

[Taro]

The ACI provision not strictly applicable, that's why I said I would use it as a guide. The same engineering principles are at work, though. Of course, you will need to apply sound engineering judgment.

A concentrated force on the wall will produce high localized stresses near the point of application. These stresses will diminish with distance. But to simplify the problem and produce a workable solution, you can assume uniform stresses over an effective width. Whether the concentrated load is an internal force due to the reinforcement tension or an externally-applied point load, the concept is the same. For a typical 8" thick wall, you could use 6x the thickness and come reasonably close to the 5 foot width you cited.

 

[MichSt]

WAstruc10,

The 5’ distribution width from AASHTO that jmiec cited is standard in the design of highway bridges. I have used this extensively in the design of bridge railings and decks for vehicle impact loads.

You could also assume a 1:1 distribution from the application of the point load to the base of wall/barrier where your design moment occurs, if this gives a more conservative distribution width. This is a common engineering assumption for distributing concentrated loads in this type of situation.

 

[Octapus]

Hey there!
I did a barrier wall design for a parking garage in Florida. Basically the barrier wall is cantilevered off the spandrel beam or slab. In my case, I did not have spandrel beams so the moment from the cantilever barrier wall has to be carried by the edge slab. It was a post-tensioned slab and did not have the capacity to do so even at 9" thick. I ended up pouring a concrete beam between supporting columns to take the vehicular impact load. I found the 8" cmu wall reinforced with #7 @ center of cell @ every cell was not sufficient to take the load even distributing over 3 cells. 12" block will work but the placement of reinforcement is toward the edge and hard to enforced during construction. Be certain the supporting structure of the barrier of wall is suffieient to carry the induced moment from the impact.