firewall design

huangyhg

firewall design
what kind of lateral loads (wind, seismic) does a rated wall (now cantilever)see after the structure around it has burned down?  what if the wall is 25' tall?
tia
i got into this not too long ago with another engineer. i don't think you should have to design the firewall for full exterior wind, but there is nothing that i could find anywhere that supports that (no pun intended).
sorry never done a fire wall in the us.
what is the code intention of this wall:separation(insulation) or structural integrity.
if it is the latter then wind may be important, but the former has already failed to serve its purpose if there is no structure on either side.
my understanding is that the wall is to resist fire only - like 2 or 3 hours.
it may or may not be a structural wall.  i always have assumed that the surroudning structure would keep it in place.
the fire walls i've done are solely for fire separation between a new and existing structure (i'm in the u.s).  depending on the rating requirements, no structural element can attach to it (i.e. it has to be free standing), and i have designed them to take the wind load as if it were fully exposed.  since it can't attach to the structure, it's a cantilevered wall.  the reinforcing gets pretty intense if it's 25' as the op suggested.
cantilevered wall design is not the only way to design fire walls but is generally the most practical for short, one story buildings.
the ibc code does not spell out that an interior fire wall has to be designed for full wind. it, however, does need to be designed for seismic whether it is interior or exterior.
nor does fm, nfpa, ncma, but the canadian code suggests 10 psf but i believe that is an ultimate load. some us engineers use only 5 psf (interior pressure)or seismic load whichever is higher.
i believe that it should be higher than 5 psf to accomodate thermal shock (heated wall hit with cold water from a fire hose)and incidental forces from those caused by flashing pulling away, etc. perhaps the canadian code has it about right but i do not know what research that was based upon.
here鈥檚 what ibc says about partition walls鈥?
1607.13 interior walls and partitions.
interior walls and partitions that exceed 6 feet (1829 mm) in height, including their finish materials, shall have adequate strength to resist the loads to which they are subjected but not less than a horizontal load of 5 psf (0.240 kn/m2).
i think the code requires that it not be attached to the structure on either side.  one thing we do at my office is build two walls so that one will fall down but the other remains standing (i don't think this was the intent of the code).  otherwise we make it a cantilevered wall designed for 5 psf or the seismic load whichever is greater.  the 5 psf also would be about a 35-45mph wind, which i think is appropriate for a temporary situation.
the intent of the code regarding a firewall is that it is not self supporting but rather supported by either of the two structures on either side of the firewall.  either building can collapse and the wall still stands.  it is used to create, literally, 2 buildings in the code (for various code reasons that the architect is trying to solve).  one reason is to reduce the size of a building to avoid sprinklers, or to keep a lower type of construction.  if the wall is supported on both sides by the 2 structures, you should be okay.  i am not sure what loads to apply but you would not be wrong (perhaps conservative)to design the wall as an exterior wall of both buildings.
don phillips
double firewalls (as lowlax mentioned) are normally an acceptable type of firewall design.
when a structure collapses, the remaining firewall must resist forces produced by:
thermal shock
flashing pulling away
pressures created by the collapsing structure
it would be great if there was some research on the magnitude of forces that these actions produce.
i've always designed cantilevered fire walls with the specified wnd load for exterior walls;  it seems reasonable that a fire and subsequent collapse may occur during a severe storm. the reinforcing requirements are increased, and the footing tends to get relatively wide. for new construction, i believe this is the best solution for single story structures. for additions, the footing design can be challenging.
the nfpa handbook says, "fire separation between buildings can also be provided by two blank exterior walls of adequate fire resistance and located very close together. each wall is tied to the frame in its building; collapse of the frame pulls the wall in, leaving the other wall to resist the fire."  i've been involved with this solution on most multi-story projects. the catch is if there is access throughthe fire wall, a rated door is needed in each wall, which can lead to architectural function and detailing challenges.
to an extent, i agree with donphillips' assessment with respect to tieing the wall to the structure on each side with the caveat that we must be able to assure the collapsing bulding will not pull the wall down with it. factory mutual's "property loss prevention data sheet 1-22 says that if the wall is tied to the frame on either side, the "pull of the collapsing frame on the fire side of the wall can be resisted by the strength of the frame on the non-fire side".  they go on to present an equation for calculating the "horizontal pull", but i've never felt comfortable enough with this approach to use it.