The following paper
appeared in the 1999 National Conference on Cultural Property Protection
Proceedings:Cultural Property Protection from the Ground Up:
Fire Protection 201
By: Deborah Freeland
Deborah L. Freeland joined Arthur J. Gallagher & Co.,
the world's fourth largest insurance broker, in 1997 as an Area Vice President.
She worked as Engineering Manager for thirteen years with Johnson & Higgins
as Regional Engineering Manager. She worked for five years as an HPR Consultant
with Kemper Groups, a long-term member of the National Fire Protection
Association's Committee on the Protection of Cultural Resources. Kemper Groups
writes the National Fire Codes for museums, libraries, places of worship, and
historic buildings.
A fire protection consultant
for almost twenty years, Freeland has extensive experience in highly protected
risk consulting, and developing special expertise in the protection of cultural
institutions. She has worked with museums and libraries of all sizes in
developing fire protection alternatives and programs to reduce their potential
risk.
Debbie holds a Bachelor of Science degree in chemistry, with an emphasis in
biochemistry, from the University of Washington. She was named a Fellow in the
Institute for the Advancement of Engineering, and is a past president of the
Southern California Chapter of the Society of Fire Protection Engineers. She
also holds the Associate in Risk Management and Certified Property Casualty
Underwriters designations. Debbie has been a member of the National Fire
Protection Association Committee on Protection of Cultural Resources since
1985.
ABSTRACT
Fires are the single greatest
cause of property loss in cultural facilities. In the United States, we
currently average just over one hundred museum fires per year, and over two
hundred library fires, with an annual property loss exceeding $2 million and $6
million, respectively. The need to provide better protection for our cultural
properties is evident. However, the practical issues of how to accomplish this
can be daunting while balancing budget constraints, different levels of risk,
and a flood of information on new products and technologies. By using a
risk-based approach to fire protection that weighs the various alternatives and
levels of protection against the cost and potential risk, we can achieve this
goal.
DISCUSSION
BALANCING THE RISK
Traditional approaches to fire
protection have been "performance-based," i.e. applying a universal
fire code or other protection requirements without regard to the level of risk
or value of the property. With ever-tightening financial constraints imposed on
most cultural institutions, this approach is often ineffective at best. It does
not encourage (and often does not allow) the limited amount of funds available
to be applied to those projects that will make the greatest impact on the total
potential for loss. Instead, by using a risk-based approach, one can identify
the more critical elements, and direct fire protection efforts and funds where
they will be most beneficial.
The methodology in developing
a risk-based approach is quite simple, with only three steps; applying these
steps is more difficult. The three steps are:
1. Identify the protection
goal (acceptable level of risk) for each area.
2. Assess the current level of risk for each area.
3. In areas where the risk is unacceptable based on the protection goal,
develop and implement measures to reduce the level of risk.
IDENTIFYING PROTECTION
GOALS
Perhaps the most difficult
step is identifying the desired protection goals. This process involves
determining the relative value of different areas and defining the acceptable
level of risk for each area. For example, the essential assets of a museum's
collection would require a higher degree of protection than lesser pieces.
Similarly, libraries would consider their rare book storage area more critical
than the newer books, which the facility can replace easily. The process of
determining relative value and acceptable levels of risk should incorporate
input from directors and management, curatorial staff, and protection
consultants.
Determining the relative value of different pieces, collections, or areas is
not simply a matter of looking at replacement costs. In addition to the dollar
value, it is also important to consider a number of other factors. For example,
the cultural significance of a piece or collection is important; the relative
importance is a function of its interest to the general public, special
visitors, or scholars/historians.
Additionally, the value to a
particular institution may depend on whether the primary purpose of the
institution is to draw the general public or to support the research of
scholars. Similarly, one must consider the political value of a collection; for
example, collections or pieces belonging to major benefactors.
Another type of value has to
do with the effect a loss of a particular piece might have on future exhibits.
For example, if an important piece or collection on a loan from another
institution is lost or damaged, it would probably inhibit one's ability to borrow
such items in the future. The assigned relative value should also incorporate
the importance of a piece to the actual stated purpose of the facility.
Different institutions focus on different media, periods, subjects, or
audiences; ancillary collections that do not support the primary focus may have
less significance to a particular facility.
Once one identifies the
relative value, one can determine the acceptable level of risk for each area.
As one renowned fire protection engineer noted: "You tell me what you want
left after the fire, and I'll design a system to save it." Of course,
one's first instinct is to reply that no loss can be tolerated. But in the real
world, especially in institutions with serious budgetary constraints,
protection of all areas to this level may not be possible. Thus, it becomes
necessary to return to the concept of relative value and expend the greatest
amount of energy and funding to protect those areas housing collections
considered to be the most valuable.
ASSESSING THE RISK
Assessing the potential risk
involves evaluating the susceptibility of a given area or collection to damage
from fire and the consequences of fire (heat, smoke and other by-products,
water, and other extinguishing agents, etc.). One must consider both the
potential frequency and severity of a fire in the area under consideration.
When examining the probability
of a fire occurring, we evaluate nearby hazards, such as cooking, flammable
liquids, combustibles that might ignite, electricals, heating equipment, as
well as the level of access that staff and the public have to the area, etc. To
get a better idea of the items that are generally associated with the
origination of fires in cultural properties, it is useful to look at historic
data on causes of fires in cultural properties.
The National Fire Protection
Association's most recent study on the causes of fires indicates that among
those that occurred in museums, approximately 23 percent were caused by
electrical systems. The rest were caused by arson (13 percent), heating
equipment (9 percent), cooking equipment (9 percent), by open flame/torch (9
percent), and by other equipment (12 percent). For libraries, the figures read:
39 percent arson, 19 percent electrical, 5 percent heating equipment, 5 percent
torch/open flame, and 7 percent other equipment.
Smoking, playing children,
natural causes, and exposure caused a few fires, but nowhere near the number
associated with the major causes listed above. This tells us that, although we
want to continue to be diligent in addressing those hazards associated with
fewer fires, current levels of protection in general may be adequate. Instead
of spending additional resources in those areas, we need to focus our efforts
on increasing protection for those hazards that cause the greatest number of
fires, such as arson and electrical and heating systems.
One aside: the NFPA does not
have a separate category for fires that occur during renovation, as this is an
indirect, rather than a direct cause. However, the increase we are seeing in
this area is alarming. Almost all of the fires caused by open flame/torch
occurred during renovation work, as were many of the electrical and arson
fires. It is therefore important to be particularly diligent in our fire
protection efforts during renovations.
The main features affecting
the potential severity of a fire, once it has been started, include the type of
construction (combustible, non-combustible, fire resistant), the number and
location of fire divisions, such as fire walls, the degree of combustible
loading in the area, and detection and protection systems. An analysis of these
features allows us to estimate the probable extent of fire spread in different
scenarios. Combining this information and the extent to which a particular area
is exposed to the causes of fire noted above allows us to determine the
relative susceptibility of different areas to fire.
DEVELOPING ADDITIONAL
PROTECTION MEASURES
The last step in the process
is to identify those areas where the risk or susceptibility to fire is
currently unacceptable, based on the relative value and protection goals
determined in the first step. Based on the information developed, one can
prioritize and target fire protection efforts in certain areas to reduce their
risk. These efforts might include providing additional automatic fire
protection systems, reducing hazards in those areas, providing different
detection systems that might respond faster, installing partitions to segregate
critical areas, relocating critical collections, modifying security and other
loss control programs, etc.
Several matrices and flow
charts have been developed that outline this process, including one found in
the appendix of NFPA 909, the National Fire Protection Association Standard for
the Protection of Cultural Resources.
NEW TECHNOLOGY: THE GOOD,
THE BAD, AND THE UGLY
With the flood of new products
and rapid changes in technology in the fire protection field, this discussion
would not be complete without a review of some of the emerging products that
are becoming available. However, we should preface this section with a word of
warning. As Steve Bush, retired head of safety and fire protection for the
National Library of Congress, cautions:
If I could give one piece of
advice to people responsible for the protection of cultural properties: Beware
of New Technology. Some cultural property professionals will grasp at any fire
protection device that will protect their collections from water damage without
considering that it may not perform as promised or may even introduce exposure
to new hazards or risks. Would you want your facility to be the guinea pig for
debugging a new product?
Before adopting a new
invention, find out if an adequate record of success has been established in
other people's facilities (not just testing laboratories) before adopting it in
your own. Moreover, find out from consumers as well as from sales
representatives. Constructive cynicism is a healthy virtue.
With that said, let's review
some of the pros and cons of some of the new technology that is becoming
available on the marketplace.
Plastic Sprinkler Piping
(The Good)
Plastic piping has been in use
long enough that it hardly qualifies as new; it has been well tested and is
approved for use primarily in light-hazard occupancies. Especially good for
retrofits, it is flexible, does not require welding (so it is less hazardous to
install), cutting oils or heavy equipment for installation (so there is less
exposure to damage for items in nearby areas). This saves installation labor
costs because it is easier to install. However, the piping must be protected
from sunlight due to deterioration issues and must be installed above a
ceiling. The largest installation of this type to date was in the historic
Supreme Court of Scotland, involving some 41,000 linear feet.
On/Off Sprinkler Control
Valves (The Bad)
At first glance, the idea of a
sprinkler system that turns itself off once the fire is extinguished is a
terrific idea, but the Firecycle system currently being marketed is not
recommended. The heat detectors used to determine when the fire is out are
subject to cooling by the sprinklers, resulting in the possibility of the
system shutting itself off before the fire has actually been extinguished, then
turning back on as the fire builds up again. This scenario would put great
stress on the sprinkler system components, increasing the risk of failure, and
resulting in increased damage from smoke, fire, and water.
Omega Heads (The Ugly)
Recently recalled, these heads
deserve mention because they are an example of new technology that was adopted
quickly for use in cultural and light hazard occupancies (even installed in the
White House) and later proved to have serious problems. After several years it
was determined that the silicone o-rings incorporated in this otherwise good
design deteriorate over time. Currently we are seeing about a 35 percent
failure rate among these heads.
Smoke Detection
Tremendous improvements have
been made in the area of smoke detection. Standard units have been made more
sensitive, yet capable of screening out background dust and other false alarms.
New air-sampling units have been developed that use tiny tubes which are easy
to hide, and yet are extra sensitive and provide early warning. And
multi-sensor units with a combination of photoelectric, ionization, thermal, or
gas detection are coming onto the market, which allow better fire detection and
screening of false signals. However, air-sampling units involving a cloud
chamber require the monthly addition of de-ionized water, an undesirable
feature from a maintenance standpoint, and the combination detectors tend to
draw more current, making it harder to provide battery back-up power.
Inert Gas Extinguishing
Agents
A potential replacement for
Halon that should no longer be installed due to their ozone depletion
potential, Inert gas systems such as Inergen and Argonite are becoming quite
popular for total flooding systems, especially in Europe where they were
developed. Environmentally inert and fairly inexpensive, the systems present a
lower cost alternative compared to some of the other systems, and no clean up
is required after accidental discharge. And the technology is older and more tested
than other replacements described below.
On the other hand, since the
inert gasses cannot be stored as a liquid at ambient temperatures, and require
many more gas canisters and more storage space; venting must be incorporated
into the room design and the small risk of oxygen deprivation to personnel
makes evacuation recommended upon agent discharge. These agents cannot be used
with existing piping for Halon.
Halocarbon Extinguishing
Agents
In the United States, the most
popular replacement for Halon for total flooding systems are the Halocarbon
extinguishing agents, primarily FE-13 and FM 200. Again, these are not
generally compatible with existing piping for Halon. The volume and weight of
the agent required is significantly lower than that for the inert gasses, and
there is no clean up required after accidental discharge. On the other hand,
this product in an actual fire situation does break down into corrosive gasses,
especially hydrofluoric acid; the quantity generated depends on the design and specific
conditions at the time of the fire. This is of significant concern to those
trying to protect valuable collections. Additionally, a gas exhaust system
would be required and may soon face restrictions because of its global warming
potential and atmospheric lifetime (about 1/3 that of Halon). There is also a
risk of cardiotoxicity making evacuation required upon agent discharge.
These agents also cannot be
used with existing piping for Halon.
Water Mist Systems
Perhaps the most promising of
the total flooding agents under development are the water mist systems. These
create a fine fog that fills the room and extinguishes primarily by cooling
action. The mist droplets are tiny, so there is no wetting of surfaces and,
like a gas, the mist can be drawn into areas such as underneath equipment that
standard sprinkler droplets cannot reach. However, these systems are still
primarily experimental, equipment for these installations has not yet been
listed or approved, realistic test scenarios need to be established and tried
to address issues such as obstructions, and no system is yet nearing
development for protection of occupancies such as cultural properties or
computer rooms. Nonetheless, it is something to watch for in the future.
Arc Fault Circuit Interrupters
It is estimated that the new
arc fault circuit interrupters could eliminate a large percentage of the
electrical fires that are the number one cause of loss in museums and number
two for libraries. These are designed to sense a change in electrical demand
and act as a circuit breaker (only faster), shutting off the electricity in the
event of problems like an intermittent short circuit, frayed wiring, or poor
connections. Current pricing is $40-50 each, but the price is expected to come
down dramatically over time.
Compact Modular Storage
Finally, new technology in
storage has resulted in a new and difficult challenge for protection engineers.
This densely packed shelving, when filled with combustible storage such as
books or papers, presents a unique hazard and results in fires that are very
difficult to extinguish. Tests conducted by the Canadian National Archives
resulted in the following scenario. Once a fire has started in the storage, the
solid shelves radiate heat downward and also trap the smoke and heat so that
operation of the ceiling smoke detectors and sprinkler systems are greatly
delayed. Meanwhile, as the fire smolders and the oxygen supply becomes limited,
flammable gasses build up under the shelves. As the fire grows, eventually the
sprinkler system may be activated, but the shelves shield the burning material
from the water. When the fire department responds and opens the shelving better
to access the fire, the flammable gasses that have built up under the shelves
flash over, due to the provision of additional oxygen. In a risk assessment
process, without special protection, one should anticipate a total loss of the
contents of the module and the associated fire area.
Appropriate protection for
this type of storage has yet to be defined to the extent that would allow it to
be incorporated into existing codes. However, the following items have been
demonstrated to be effective in reducing the extent of loss: provision of a
very high density sprinkler system, installation of early warning smoke
detection, placing one inch spacers between modules, providing metal vertical
barriers for each shelf, and use of quick response sprinklers with wet pipe
sprinkler systems only.
CONCLUSION
By implementing a loss control
program that incorporates a risk-based approach to fire protection, along with
careful application of new and existing technology, we will be much more
effective in controlling the number one threat to our cultural properties.
Moreover, fire protection is preservation.