Fire Suppression Systems Update

I had previously posted about this subject here: http://itmanager.blogs.com/notes/2004/03/fire_suppressio.html

Met with a rep of Ansul and a sales tech from the local Ansul authorized dealer in the area last week. They talked to us about the two systems that they install most, Inergen and Saphire.

Saphire was talked about in a comment in by Mark Koskela in the previous post. But here's some more tidbits about it. It's a liquid at room temperature that turns into a gas with very little energy added. It's stored in the same metal canisters as FM-200 or FE-25 but is pressurized with Nitrogen, whereas the other products are turned into a liquid by pressurizing them so they don't need the Nitrogen. It turns into a gas when it hits the nozzles (the energy of hitting the nozzles is enough to turn the liquid into a gas). Pretty neat stuff. The Ansul rep poured a bit of it on our wood table and just rubbed it around with his hand for a few seconds and it evaporated quickly leaving no residue. He poured it on paper and showed how it doesn't react with the paper or ink at all. Poured it directly onto his laptop. Great demo that can only be done with this product as all the other products are gas at room temperature and pressure. Saphire or Novec 1230 works the same way that FM-200 and FE-25 work but has much less environmental impact. It basically has the same global warming potential as CO2 whereas the other two chemicals have global warming potentials over 3500 times that of CO2. Also less toxic. It's used in concentrations 50% below what is known as the no observable adverse effects level (NOAEL). FM-200 is used at concentrations right around its NOAEL and FE-25 is used in concentrations actually above its NOAEL.

The other system, Inergen is a mixture of Argon, CO2 and Nitrogen. It works by pushing the concentration of Oxygen in a room below the point that can sustain a fire. It's more expensive to install an Inergen system because it takes more tanks per room, but unlike the other three agents I've talked about, Inergen can be piped over much longer distances, so you can have a room full of tanks in a central location that can be used to suppress a fire in many different locations in a building. Also, it's much cheaper to refill an Inergen system after a discharge as they basically give the gas away for free and just charge for the labor it takes to refill the tanks. The other agents cost about $30 per pound to refill and considering that a typical room needs more than 200 pounds of the other agents a discharge can be very expensive. According to the rep, an Inergen system costs about 10-15% more to install than the other three.

Some other plusses of Inergen: With the other systems, the chemicals used are heavier than air and tend to disperse out of a room fairly quickly. This means that work has to be done to a room to make sure that it is "air tight" enough to keep the agent in place for a specified amount of time. The gases in Inergen are similar in weight to air and do not tend to disperse from a room quickly, so less work needs to be done to make sure that the room is sealed well. Your local fire marshall might have issues if you use the chemical agents in a room with open plenum false ceiling for instance whereas with Inergen they will not have the same issues. This is because rooms where FM-200, FE-25 or Saphire are installed have to pass a test where they blow air into the room with a fan and measure the air pressure. The pressure has to get to a certain point in order for the room to pass. Rooms where Inergen are installed do not need to even have this test performed.

So basically if you want an agent that is entirely benign to the environment and have a room that fits the profile for an Inergen system, it's the system of choice. Apparently, most of the systems being installed in Europe these days are Inergen systems. Unlike in the states, Halon has been banned for use in any existing system as of last December.

This is unlikely to occur in the states anytime soon, however, as the largest user of Halon in the US is the government.

Anyhow, for companies that are environmentally conscious yet still need proven fire suppression systems, it seems like Saphire and Inergen are the two systems of choice.

As a side note, FM-200 and FE-25 were both initially developed by Dupont. Dupont didn't see a market for FM-200 and sold the licensing rights to Great Lakes Chemical. The Dupont chemical is actually called FE-227ea and FM-200 is a brand name owned by Great Lakes Chemical. Dupont now sells the same chemical under their name of FE-227ea. FE-25 is only made by Dupont and only sold in the US by Fike. Because of its apparent toxicity, it hasn't been adopted widely. Saphire or Novec 1230 was developed by 3M.

BTW the detection systems and delivery hardware are the same for all 3 systems. The difference in the delivery hardware is number of tanks needed and number of nozzles needed in a given room to discharge the gas evenly to all parts of a room in 10 seconds. Inergen uses more nozzles to soften the pressure of discharge at each nozzle to prevent damage to surrounding equipment and assets.

Comments

[Michael]

I'm built many rooms with fire systems. Between TV facilities and IT rooms, there's lots of expensive gear.

The install costs of these gas systems are quite high, not to mention the ongoing maintainance costs. IIRC, Innergen systems need to be serviced every 6 months.

I've come to believe that you are better off with a pre-action water system than anything else. By pre-action I mean that the sprinkler pipes are only filled with air until you get alarm conditions that will flood the pipes. Typically this is two alarms required, heat and smoke. Even after the pipes fill with water, the heat must be enough to pop the sprinkler itself.

You pretty much need a real fire to get water going. If the water does get going, it's only the one sprinkler that got popped. Not your whole room.

The gas systems theoretically are in place to save equipment, but realistically you will have them set to only dump if there is a real fire going on.

The gas systems vendors make a big point about how they leave no trace. It's true, but think about the big picture.

If you really have a fire, just because dumped gas instead of water, you aren't going to be working that room anytime soon. The fire department will probably cut power if there has been a real fire and even pull the occupancy permit.

Do yourself a favor and compare the 5 year costs of a gas system versus a pre-action sprinkler system. I think you might be suprised.

[John]

Inergen - what a great idea. A system malfunction will kill everyone in the server room through oxygen deprivation. Great idea - cheaper than laying those employees off!

Many years ago, I worked for a company with a carbon dioxide fire suppression system in the computer room. Every time I entered, I would disarm the system.

A nuclear plant near here lost several contractors years ago when a CO2 system malfunctioned in a transformer vault.

This is an astoundingly horrible idea. And if any sales critter says that their system can't malfunction, run away!

On the other hand, clean water won't damage your media badly and is a lot more gentle on your employees. What do you value, your employees or some server hardware which can be replaced in a matter of days?

[Alex Scoble]

The system can't catastrophically malfunction like you say since the tanks only hold enough gas to perform the required function.

CO2 systems are quite different however as CO2 in quantities above like 6% of total air is quite toxic to the human body.

Also, when they figure out the amount of Inergen to put in the tanks for a given size room, they normally do it at %35 or so concentration. You can survive at up to %55 concentration, so there's a lot of headroom to work with.

Clean water? Do you not understand how water based systems work? That water sits in those pipes for years and is the nastiest, foulest stuff when it comes out.

Cost to protect a room with one of the clean agent systems is around $30,000 for a good sized room. Cost to protect my server room is around $6,000.

Cost to recover 1 hard drive is about $2000. I have 8 servers with 5 hard drives each. Do the math.

You don't replace a water system with one of the clean agent systems. At least not in California. The clean agent system is used in front of the water based system.

A water based system is only there to protect the building and people. Most companies put hundreds of thousands of dollars into their IT equipment (my company is no exception). It's certainly worth it to spend a small fraction of that to protect the company's most valued asset next to the employees, the data.

[Alex Scoble]

It's not an either/or proposition in the state of California. If you use a clean agent system, it's in addition to the sprinkler system already installed in your building per code.

BTW CO2 is NOT a clean agent. A clean agent has to be environmentally safe, safe for humans and safe for files, equipment and other assets. CO2 is very toxic to humans at levels needed to suppress a fire. Which is why you needed to turn off the system when entering the area to work.

As far as what happens if an Inergen system fails? The fire keeps burning and sets off the sprinkler system. An Inergen system CANNOT dump more gas into a room than it has in the tanks. So the only way for it to fail would be for it not to work at all.

[J Samuel]

Please supply me with more info re the agent. Is it human friendly?

[Samantha]

Just let it burn... and claim it on insurance...

[David]

It's great to see the industry discussing fire protection. If I may add, depending upon the building code in place in your part of the country (most are IBC now), a sprinkler system is still usually required by the local authority having jurisdiction. That doesn't necessarily mean that the pipes above a large computer room are full of water though NOR does it mean that sprinklers are always installed. There are some agencies that will allow just a clean agent system in lieu of sprinklers - it depends on the construction type of the building and other factors but it is possible.

In most 'well engineered' critical assets situations, a fire detection (i.e. smoke) system is connected to the fire alarm panel. Upon activation, the clean agent system goes into a 30 second (or 60 for CO2) pre-discharge alarm. At the same time, another the overhead sprinkler piping may be doing one or two things. If it is a single event pre-action system, water will then flow into the pipes. The clean agent system discharges and hopefully puts the fire out. If not, water is directly above the fire to protect the building as the electronics are likely to the point beyond damage just due to the smoke and decomposition products.

Now if the sprinkler is a double interlock pre-action system, water will not flow through the pipes until both the smoke detection system is in alarm AND enough heat has developed to melt the element in a sprinkler head causing internal air pressure to drop.

With regards to the comment about Inergen killing people during an accidental discharge -- Not True. Take a look at the Ansul web site and you might be able to find the old video clip of people in a room during a discharge. I wouldn't do this personally - but it is a safe agent. Then again, anything except for CF3I is save when compared to CO2.

Now, the trick with 3M's Novec 1230 Fire Protection Fluid is that because it is a liquid at room temperature, it can be delivered with a pump too! No more need for high pressure cylinders. You can get information on Novec 1230 at:
http://cms.3m.com/cms/US/en/2-68/iclcrFR/view.jhtml and learn more about pumps at: http://www.capsfire.com

Now I have done fire testing research with clean agents for over 5 years and working as a firefighter/AHJ in the past, I would argue the the fire department is going to cut the power or no difference in downtime. Most of the time, FF's don't know what the equipment is and are taught to wait until a building rep. is on site before cutting power for these sites. I have seen over a dozen IT/switch rooms back online within hours or a day after a fire with a properly designed clean agent system. The ones with just a sprinkler -- a few weeks time. That's a huge difference. If you want to see the actual comparison, get your hands on the video testing Great Lakes Chemical did for a marketing series. The video speaks volumes. All for now.

[kylina ward]

umm what is this?

[Robert Schmidt]

I need a comparision between Novec 1230 and the other clean agents. We are in the process of upgrading the supression systems in our IT rooms and swicth gear rooms throughout the NWT. Can you help?

[Alex Scoble]

Hi Robert,

What exactly do you need beyond what has already been explained?

[Nick Thompson]

Does anyone have any experience of NAF S-111?

Any info, web address or personal experience - Please.

Thinking of protecting the standby generator.

Regards, NICK

[John Allen]

I googled into your page whilst looking for something else and couldn't help reading the comments. We have a new idea you might be interested in: On the basis that there are areas where installing a room flood system is difficult or not an option, we developed a rack detection & protection system. It contains all the elements of a room fire system in a 19" x 2U high chassis. Options include point smoke or air sampling detectors, FM200, Novec 1230 or Pyrogen/Aero-K fire suppressants, condition monitoring (temp, humidity, vibration sensors) an IP address and Telnet/SNMP.
Our market place is niche but growing, not least because we put detection and suppression where it's needed, in the rack itself. It also means that we won't dump a room full of gas; there is no disruptive installation, or pipework around the room; no need to lock the system off during work hours and if you decide to relocate the room or the rack, the system goes with it.
It's called Redetec and can be found at www.redetec.co.uk
We have a US/Canada agent if anyone's interested

[John Allen]

PS to my message regarding rack fire protection systems:
Sapphire is Ansul's brand name for Novec 1230.

[John Brooks]

I have just found the Blog from a google search. I have been designing, installing and testing Pyrogen Aerosols and other Halon replacement systems for the last ten years. The attributes of the HFC PHFC gasses are will established. One thing to consider is the post combustion by-products. The Hydrofluoric Acid (HF) from HFC227ea, when exposed to a high temperatures (~650F) can do more damage than the fire. Especially in IT applications, the arcing from a power supply will cause enough HF damage to destroy the rack and its contents, plus if you get a breath of it when opening the cabinet it will Kill You.

The condensed aerosols like Pyrogen chemically inhibit combustion, do not replace the oxygen in the space like CO2. It is a new technology under NFPA 2010, and is widely used world-wide and here in the states. It is non toxic, non corrosive, non conductive, and three times more efficient, when compared to Halon 1301. More information available at pyrogen.com.

[Alan Way]

In reply to Michael earlier on, where he says:
"Inergen - what a great idea. A system malfunction will kill everyone in the server room through oxygen deprivation. "

From what I understand of Ingergen, it lowers the O2 percentage in the room to below combustion levels, it doesnt eliminate the oxygen. It also raises the CO2 levels, which stimulates the rate of respiration and increases the body’s oxygen use efficiency.

As a result, the brain continues to receive the same amount of oxygen in an Inergen atmosphere as it would in a normal atmosphere, even if the oxygen concentration falls to below 10%

http://www.wormald.co.nz/products/special_hazards/inergen

[Steven K.]

Query for John Brooks or anyone who may know:

Can you elaborate further about Hydrofluoric Acid production (HF) from FM200 (HFC227ea), when exposed to high temperatures? I have been looking into installing FM200 in my IT Computer Room and specifically in our Silo's. Now I'm having second thoughts. Are there documented cases where this has happened, or was this from laboratory testing? Thanks.

p.s. to Alex Scoble,
Great web site! I've learned more here than weeks of phone conversations with the manufacturers and vendors.

[Hunter Burlew]

HF is real and is real nasty stuff. The reality is, if you have a properly designed FM-200 or Sapphire system the amount of HF produced should be minimal. These systems are designed so the detection (smoke/heat/flame detectors) catch the fire early and put it out. Think of this as well, If you IT room has a fire, the products of combustion from the fire itself are just as bad. Burning insulation of CAT-5 and all the other computer cable is just as bad if not worse.

Your fire protection provider should address if the space is occupied or not. Some agents are rated for occupied spaces some are not. All systems should be designed with pre-discharge alarms as mentiond above. NFPA does require that all systems have a way to manually actuate the system which bypasses the pre-discharge alarm.

[Doug Jones]

What you exhale with every breath is CO2, so the way I look at it, carbon dioxide is, in and of itself, not toxic. It's just that in volumetric concentrations above a certain level (check with NFPA, but best recollection is it's about 18 to 20%), there's insufficient oxygen to sustain human respiration, and it's a bummer, but that's the level needed for effective flame extinguishment. Localized CO2 fire suppression systems are very effective in flame knock-down when applied directly to a flame front, but even with fault-tolerant prewarning systems they should be considered very dangerous in occupied spaces. Halogenated agents, while very effective and having low human toxicity, are considered to be damaging to the ozone layer. The newer agents, I know little about, but I would say they should be subjected to a good deal of scrutiny by the industrial and commercial marketplaces. I may be a bit cynical, but I do believe Dupont, 3M, Forexx, Great Lakes and their vehicles (e.g., Ansul, Kidde, Fenwal) are in it for one thing... the money. I don't believe there's much internal debate in any of those companies about the beneficial impact of their products on the welfare of mankind. It may seem way old school, but water seems to me like the most effective suppression agent in all but a tiny fraction of situations. Just cut the power before using it on electrical equipment and everything is usually pretty cool.

[Alexander Scoble]

Actually, CO2 is toxic above a certain level. If you are exposed to a high CO2 level atmosphere even for a short time, you can die. http://www.emedmag.com/html/pre/tox/0500.asp

As for environmental impact, out of the systems that I have looked at so far, Inergen is by far the most benign to humans and to the environment with almost zero impact.

Water is great, except for the fact that once it's sprayed on your equipment/work environment, it needs to be dried up, carpets need to be replaced, muck needs to be cleaned, etc. A water discharge can put your business infrastructure out of business for weeks whereas a gas system can have you up and running in hours.

If you like where you work and are in charge of replacing the fire suppression system, I'd highly suggest not looking at water...because in the event of a discharge you will either be fired or your company will go out of business due to downtime.

[Marty Gresho]

For most IT equipment, the damage from corrosion of circuit boards due to smoke is more costly and pervasive than water damage. The advantage with a gaseous system (any ofthem) is that they are oftem coupled with a state of the art detection system (like laser based air aspiration type) so they detect and suppress afire very quickly and in its incipient stages. Water based suppression, though code required, requires heat to activate and much more smoke can be created prior to extinguishment. The smoke is hard to get out fast enough before it does its damage and has been the largest driver for gaseous suppression at several of my installations.

Its the data!!

[Scott]

I've read the above discussions with great interest. We manufacture a system that actually protects the individual racks, rather than "total flooding" a room. This offers a couple of advantages. First off, the systems only require enough gas to fill the rack, making the cost of a discharge much, much lower. Also, when this gas is ulitmately bled off into the larger room, the concetrations are low enough to minimize health concerns. Our systems are fully self contained, require no power, and can activate off smoke or flame. In the event of a fire, the system can be set to shut down the power to the rack it protects so that other operations are not affected.

In terms of the NOVEC-FM200-other agent discussion, our systems are available with all of these gases, plus several others that are not ideal for computer applications. Either FM-200 or NOVEC would be a good choice for this type of environment. The concerns about FM-200 byproducts are worthy of discussion, but these byproducts are only created by an active fire. The more quickly the fire is suppressed, the less of these by-products will be produced... lesson here is be sure to cut the power to the fire source. See www.fm-200.com for more detailed information.

Novec is another great product, but it is not inexpensive. The critical concern here is being sure to work with a contractor very familiar with this agent, so that the system will properly deliver the agent.

Alex, in the post above, makes some very good points, ideally you can use a sampling system to detect in the incipient stages, and flood the room, but unfortuantely, these systems are fairly expensive. If your budget allows, this is the way to go.

If not, a rack based solution is a good alternative, as the amount of gas is small (reducing cost and environmental concerns), the areas affected are limited to the equipment in that rack (again, you need to shut down the power to stop fire from restarting - in most cases).

Regardless of the approach you choose, I'd still recommend a gas over water. Good luck!

[sutter kiplangat]

could you please send me Human Toxicity potential
values for agricultural systems

thanks in advance.

sutter.

[Richard Puig]

Your blog seems to have gotten out of hand. There is too much misinformation, incorrect information and slanted information in regards to different folk’s views and their products they seek push.

Just trying to set some of them straight would be a major task. I will limit my comments to just a couple.

HF production in the discharge of HFC (HFC-227ea, FM-200/FE-227) (HFC-125, ECARO-25) is insignificant. They act upon the fire so quickly that the amounts are insignificant. Of more concern are the toxics produced by the fire itself when it burns with plastics and cables in a computer room. Put the fire out with the most effective agents as quickly as possible. NFPA and the EPA would not recognize these agents if they were dangerous to critical equipment.

Aerosols are an EPA approved Halon Replacement but are not approved for occupied spaces by the EPA and they are not included in NFPA 2001.

CO2 should not be considered in occupied spaces.

Inergen does reduce the oxygen in a room and NFPA 2001 standard has stated the human exposure limit should not be more than 5 minutes. This has been its position since its first Edition of NFPA 2001 in 1994. The concern with the reduction of oxygen is hypoxia. Ansul has even printed articles on it.

Inergen might not need a Door Fan Test to determine its room integrity, but it is certainly recommended. No one has mentioned that Inert systems also need venting to prevent over pressurization of the room because you are putting so much Inert agent into the room in a short period of time. Halocarbon Agents do not require venting because there is not concern of over pressurizing a room.

All Halocarbon systems as per NFPA 2001 have a working pressure of 360 psi in their storage cylinders. ECARO-25, FE-277/FE-227 and Novec 1230. All have Nitrogen added to them to pressurize their cylinders and get to a working pressure of 360 psi.

The questions and their answers should be based on what the applicable Standard (NFPA 2001) says in it and what is and is not approved.

[Marco Gauthier]

We just installed a clean agent system in our server room, on top of the pre-action sprinkler system. Inergen was recommended since our mother company is based in Canada, where "green house effect gases" are banned. But since our plant is in Wisconsin where both are allowed, we looked at both Inergen and Ecaro-25. As of December '05, the difference in price was 5:1, the expensive one being Inergen. FE-25 side effect gases being higher than the others, they are merely considered irritants. So is spending the night in a smoky bar or doing repairs on your car inside your garage. In the case of a clean agent discharge, you talk about "maybe" if you get a fire, "maybe" if you don't have time to leave the room. We just tested the system before going "live", believe me, the noise from the bell alone will make you want to get out! As far as the environmental effects, yes the green house phenomenon is bad, but again it's just "maybe"... Coal power plants and heavy diesel truck traffic should be more at the top of your preoccupations.

[Janrdhanan]

Do you have any information about NN 100 SYSTEM?

[Tony Ferro]

We just had a server room remodeled with the addition of an Ansul Inergen system. We were told that when the gas is expended into the room, the running servers would not be affected. Not so much. I had two failed drives in two of my servers.

[Thomas Crowthers, Asst. Chief, Sycamore Township Fire Dept.]

We can not get an inergen to pass the Door Fan test is small area room. Are they doing anything wrong, or can they use other testing systems to test the inergen system for new installing?

Thanks for any information

[Pat Cull]

Does anyone have experience with using these agents on high voltage cabinets and switchgear? (4160 volts and under) A lot of good comments and ideas. The bottom line is the almighty dollar. How much is an owner willing to spend to provide the best possible protection to staff and equipment?

[Dan Ubelhor]

It is great to see so many interested in Clean Agent systems vs. Water. I attended the FSSA (Fire Suppression Systems Association) conference in February of this year. It can be said that Pre-Action Sprinkler systems can be vital in the protection of the building structure. Clean Agent systems are provided to suppress fires in its incipient stages keeping minimal damages limited to the affected piece of equipment. All clean agents require 0 ozone depletion. But, a many have potential Green House effects. FM-200 Green House effect can be up to around 42 years while Sapphire will last only around 5 days. Better yet, Inergen has no Green House effect on the environment. The reason is that the chemicals that make up Inergen (52% Nitrogen, 40% Argon and 8% CO2)is naturally in the environment we breath. While there is potential for HF (Hydrogen Flouride)emittance, the incipiant stage plus a maximum suppression discharge time of 10 seconds required by NFPA 2001 reduces HF levels to extremely low quantity. Inergen is an Inert gas and not a Hyrdo-Flourocarbon. Therefore, it does not require a less than 10 second discharge. Last, what I see most concerning is room integrity. If you as a customer are not willing to properly seal your room, you may as well not have a suppression system. Many jurisdictions require a minimum hold time of ten minutes or depending on how long your local fire department requires to gaining control of your rooms fire condition. Yes, venting is required for Inergen, but did you know it is also important for other gaseous agents such as Fm-200? There have been known cases for rooms protected by FM-200 with heights greater than 12 feet to have experienced entire cinder block walls pushed out and crumble due to overpressurization. To sum up, please consult with experts in the field of Fire Suppression systems to get the facts. Sprinkler systems are vital to the protection of your structure, but there is a place for Clean Agent suppression systems.

[John]

There is a new player in town take a look at www.firepass.com They are a proactive solution to fire fighting in critical facilities. They reduce the oxygen concentration by 6% to stay at 15% oxygen which is equivalent to 8,000 partial pressure O2 or the same as being in a commercial airplane, which is completely safe for humans. My Colleague saw their presentation in Houston and they are presenting at the Uptime Institute National Conference in September and I know they also presnted to the SFPE chapter down in Atlanta and it looks like they will be added to the NFPA 2001 Clean agent standard. From what I was told they are proving OSHA's 19.5% oxygen respiratory standard wrong. NIOSH has backed their claims and they have filed for a permanent variance and currently waiting on the official response and once they get it they will be installing with a lot of top companies. The great thing about their technology is can scale to any size facility from 10,000 sq ft data center or 500,000 records management room or even a full building, and it is very easy to install and very low maintenance required, which I really appreciate. Best of all they can prevent fires from occurring! Their cslogan is "why be reactive when you can be proactive"

[choh choon jin]

May I know the differences between NFPA 2001 and NFPA 2010 standard. If a fire extinguishing system complies with NFPA 2010 standard, can I take it that it can also comply with NFPA 2001 standard. Thank you.

[Erik]

Many of the comments that I have read are accurate in regards to Sapphire and Inergen systems. They are extremely environmentally friendly. Inergen has zero atmospheric lifetime and Sapphire has a lifetime of only a couple of days, where as FM-200 and other "clean agents" have a lifetime of around 33 years! Another thing about Inergen is that while it does lower the level of oxygen in the room it does not lower it below what is safe for humans. Anyone that is going to be in Columbus, Ohio on the 24th of April and would like to attend a Clean Agent Roadshow, Ansul will be putting one on. It will be about an hour and a half presentation with lunch afterwords. Interested email contact information to [email protected]

[RYAN]

Anyone out there intrested in a fike suppression system with a 700lb tank of FM 200 clean agent?

[Radhakrishnan]

Hi Alex,

Your article on FM 200, iergen was exactly i was looking for. Many thanks and keep writing.

Regards

RK

Dubai, UAE.