Return to index: [Subject] [Thread] [Date] [Author]

RE: Firewall

[Subject Prev][Subject Next][Thread Prev][Thread Next]
I believe the definition given below for a firewall applies only to a MFL
(maximum foreseeable loss) wall as defined by FM (Factory Mutual).  I
believe a "firewall" falls somewhere between a fire separation wall and an
MFL wall.

There are "fusible" fire link structural connectors available. Composite
nylon/ceramic all-thread bolts, nuts and washers with fairly high allowable
shear and tension values are readily available through out the U.S.  This
material melts at a temperature below that of steel so that the connection
disappears before the steel begins to pull away from the supporting wall.  I
don't know how the combustible temperature of fire-treated wood compares to
the Nyltron bolts.

Also, in my opinion the exposed wall after a fire would only have to be
designed for the same temporary loads required by code for a
"under-construction" situation (which I believe is only 15 PSF wind load per
OSHA) rather than a full seismic or wind load event.

-----Original Message-----
Sent: Thursday, February 22, 2001 12:49 PM
Subject: FW: Firewall:(was)CMU Wall Supported by Plyw

-----Original Message-----
From: Roger Turk [mailto:73527.1356(--nospam--at)]
Sent: Thursday, February 22, 2001 12:32 PM
To: seaint(--nospam--at)
Subject: Firewall:(was)CMU Wall Supported by Plyw

Scott Maxwell wrote:

. > Now about firewalls...


. > The first thing is to realize that there is a difference between a
. > firewall and a fire separation wall.  A fire separation wall is a wall
. > between two spaces that is meant to stop or slow the spread of fire.  A
. > fire wall is essentially a fire separation wall that MUST remain
. > AFTER one side of the building has collapsed (or is one of the two
. > buildings has collapsed).

. > If you must have a firewall, then that wall between the two buildings
. > be designed in a fashion that it would remain standing if either
. > collapses.  Thus, if the new building falls, then the fire wall MUST
. > remain standing or if the existing building falls, then the fire wall
. > remain standing.

. > This can typically be done one of several ways.  You can:

. > 1) [snip]

. > 2) Do a single wall that is tied to each structure that has "break away"
. > connections so that one side can fall without causing the other to
. > collapse or bringing down the wall.  With a masonry wall with wood
. > this has traditionally been done with a "fire cut".

. > 3) [snip]

It seems to me that a "fire wall" would be virtually impossible in seismic
country as the code requires that the diaphragm be tied to masonry/concrete
walls.  "Fire cuts" would be ineffective as the joists are tied to the wall
with "tension ties" or other anchors.  A "fire cut" would only be effective
for wood joists resting on or in walls without any other type of connection
between the joist and the wall.  For a fire cut to be effective with tension
ties, the tension tie would have to be a fusable link and even then it is
questionable whether that would work, e.g., a fire destroys the wood joist
mid-span, but does not reach releasing temperature at the fusable link at

A. Roger Turk, P.E.(Structural)
Tucson, Arizona