Level 1 Manual

Level 1 Training for Fire Enclosure Integrity Design, Testing, and Witnessing
For complying with • NFPA 2001 - Appendix C • NFPA 12A - Appendix B • ISO 14520 – Annex E

Revision History
Date
May 2, 2003 Aug 14, 2003 Sept 03, 2003 Oct, 2003

Person
TL TL TL TL

Rev
0.00 0.01

Notes
Initial Copy Major additions from Old Level 1 manual More Major Additions Added Copyright Added design section Added Retention Time Section Changed E2001 to CA2001 in disclaimer Added Appendix C – Enclosure Design Specification Added witness chapter Finished witness chapter Added Enclosure Verification Form Added Glossary or Terms Reviewed entire document Modified NFPA excerpts

0.03

Nov 2003 Nov 2003 Nov 2003 Nov 19 2003 Sept 20, 2004

TL TL TL CG JG

0.04 0.05 0.06 0.10

Copyright © 2004 Retrotec Energy Innovations Ltd All rights reserved. This document contains materials protected under International Copyright Laws. All rights reserved. No part of this document may be copied or reproduced in any form or by any means without the prior written consent of Retrotec Energy Innovations Ltd. Retrotec makes no warranties with respect to this documentation and disclaims any implied warranties of merchantability, quality, or fitness for any particular purpose. The information in this document is subject to change without notice. Retrotec reserves the right to make revisions to this publication without obligation to notify any person or entity of any such changes. CleanAgent 2001 and CA2001 are Trademarks of Retrotec Energy Innovations Ltd. Other trademarks or brand names mentioned herein are trademarks or registered trademarks of their respective owners.

9/22/2004 9:59 AM M:\Projects\Training\Fire\Level 1\Manual\Level 1 Manual (rev 0.10).doc

Level 1 Training Manual

Table of Contents
1
1.1 1.2 1.3

WHAT IS ENCLOSURE INTEGRITY TESTING ............................................................ 6
Why Clean Agent Systems? .....................................................................................................................................6 The Discharge Test ..................................................................................................................................................6 The Door-Fan Test ...................................................................................................................................................7 What is a Door-Fan test?..........................................................................................................................................8 Static Pressures.........................................................................................................................................................8 Total Room (Whole Room) Leakage .......................................................................................................................8 Measuring Lower Leaks...........................................................................................................................................9 Predicting the Retention Time..................................................................................................................................9 Other Door-Fan Applications...................................................................................................................................9 Pressure Relief Vents ...............................................................................................................................................9 Smoke and Contaminant Movement ......................................................................................................................10

1.4

2
2.1 2.2

AGENT LOSS MECHANISMS .......................................................................... 11
Pressures Across Holes: How Agent is Lost ..........................................................................................................11 Pressures That Cause Agent Loss ..........................................................................................................................13 Dynamic Discharge Pressure .................................................................................................................................13 Gravity....................................................................................................................................................................13 Convection .............................................................................................................................................................14 Forced Air Circulation ...........................................................................................................................................15 Static Pressures.......................................................................................................................................................15 Wind Pressures.......................................................................................................................................................15 Enclosure Behavior during Retention ....................................................................................................................16 The Descending Interface.......................................................................................................................................16 Continual Mixing ...................................................................................................................................................18

2.3

3
3.1 3.2 3.3 3.4 3.5

CALCULATING RETENTION TIME..................................................................... 19
What is Retention Time?........................................................................................................................................19 Retention Time in the Descending Interface Case .................................................................................................20 Retention Time in the Continual Mixing Case.......................................................................................................20 Retention Time in the Extended Discharge Case ...................................................................................................20 Retention Time Dependencies................................................................................................................................21 Leakage Area (ELA) ..............................................................................................................................................21 Agent Initial Concentration and Final Concentration ............................................................................................21 Equipment Height and Enclosure Height...............................................................................................................22 Measuring Maximum Agent Height?.....................................................................................................................22 Measuring Minimum Protected Height? ................................................................................................................22

3.6 3.7

4
4.1 4.2 4.3 4.4 4.5 4.6

RECURRING DESIGN PROBLEMS...................................................................... 24
Cascading Pressures ...............................................................................................................................................24 Common Sub-floor.................................................................................................................................................25 Pressurized Sub-floor .............................................................................................................................................26 Common Above-ceiling Spaces: Discharge will pull smoke in .............................................................................26 Common Above-ceiling Spaces: HVAC Leakage .................................................................................................27 Suspended Ceilings Too Low ................................................................................................................................27

5
5.1 5.2 5.3 5.4 5.5 5.6

GOOD ENCLOSURE DESIGN PRACTICES .............................................................. 28
Superior Protection for Less Money! Interested? .................................................................................................28 Run walls slab to slab.............................................................................................................................................28 Eliminate T-bar suspended ceilings .......................................................................................................................29 Maximize the room height and volume..................................................................................................................29 Select an appropriate retention time .......................................................................................................................30 Fit automatic door closers ......................................................................................................................................30

Level 1 Training Manual 6 THE DOOR-FAN TEST ............................................................................... 31
6.1 6.2 How a Single Door-fan “Sees” a Room .................................................................................................................31 The Lower Leaks Test............................................................................................................................................32

7
7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12 7.13 7.14 7.15 7.16 7.17 7.18 7.19 7.20

WITNESSING AN ENCLOSURE INTEGRITY TEST ....................................................... 33
Technician Training ...............................................................................................................................................33 Software Conformance...........................................................................................................................................33 Room Pressure Gauge Calibration Certificate .......................................................................................................33 System calibration ..................................................................................................................................................34 Field Calibration check procedure .........................................................................................................................34 Return Path.............................................................................................................................................................34 Room and Equipment Set-up .................................................................................................................................35 Static Pressure Check .............................................................................................................................................35 Gauge Set-up..........................................................................................................................................................35 Flow and Room Pressures Entered Correctly.........................................................................................................36 Range Selection......................................................................................................................................................36 Testing in Both Directions .....................................................................................................................................36 Determining the Leakage Split – The BCLA Test .................................................................................................36 Technical Judgment................................................................................................................................................37 Yearly Retests ........................................................................................................................................................37 Commonly Needed Inert Gas Clarifications ..........................................................................................................37 Enclosure Integrity Test Verification Form ...........................................................................................................38 Standards and How They Apply ............................................................................................................................43 Range List for Door Fans .......................................................................................................................................44 Flow Range Pictures for 2000 Series Door-fans ....................................................................................................44 Flow Range Pictures for 900 series Door-fans.......................................................................................................47 Small Room Retention Times ................................................................................................................................48 Selecting an Appropriate Retention Time ..............................................................................................................48 Recommended Times for Small Rooms.................................................................................................................48

8
8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8

APPENDIX A – AGENT COMPARISON ................................................................. 50
Standards ................................................................................................................................................................50 Agents ....................................................................................................................................................................51 Specific volume and density constants for agents..................................................................................................52 Concentration Ranges ... from an enclosure leakage perspective ..........................................................................53 Comparing Retention Times ... descending interface case.....................................................................................55 Comparing retention times ... continual mixing case .............................................................................................56 Conclusions ............................................................................................................................................................57 Agent Comments....................................................................................................................................................58

9
9.1 9.2 9.3 9.4 9.5

APPENDIX B – NFPA STANDARD EXCERPTS .......................................................... 59
NFPA 2001 Standard (Year 2004 Edition) ............................................................................................................59 NFPA 2001 Standard (Year 2000 Edition) ............................................................................................................62 NFPA 2001 Standard (Year 1996 Edition) ............................................................................................................65 NFPA 12A Halon...................................................................................................................................................68 NFPA 12 for CO2 ...................................................................................................................................................69

10
10.1

APPENDIX C – SAMPLE ENCLOSURE INTEGRITY TEST SPECIFICATION .................................. 70
General Enclosure Design Guidelines....................................................................................................................70 Slab To Slab Walls or Solid Ceiling ......................................................................................................................71 Avoidance of Attached Volumes ...........................................................................................................................72 Penetration Planning ..............................................................................................................................................72 Document Passageways .........................................................................................................................................73 HVAC Dampers .....................................................................................................................................................73 "Un-closeable" Openings .......................................................................................................................................73 Location of Dedicated HVAC units .......................................................................................................................73

Level 1 Training Manual
10.2 10.3 10.4 10.5 10.6 10.7 Minimum Protected Height ....................................................................................................................................73 Summary ................................................................................................................................................................75 Enclosure Integrity Specifications..........................................................................................................................75 Enclosure Integrity Performance Specification......................................................................................................76 Enclosure Integrity Prescriptive Specifications......................................................................................................76 Clean Agent System Specifications .......................................................................................................................78 HVAC Specifications.............................................................................................................................................79 Ductwork................................................................................................................................................................79 Approval/Acceptance of Clean Agent System .......................................................................................................80 Approval/Acceptance of Enclosure Integrity .........................................................................................................80 Warranty.................................................................................................................................................................82

11 12 13
13.1 13.2 13.3 13.4 13.5 13.6 13.7 13.8 13.9 13.10 13.11 13.12

APPENDIX D – ENCLOSURE INTEGRITY VERIFICATION FORM .......................................... 84 APPENDIX E – GLOSSARY OF TERMS ................................................................. 88 APPENDIX E – CA2001 DEMO EXAMPLE ............................................................ 95
Home tab ................................................................................................................................................................95 Building/Room tab .................................................................................................................................................96 Agent/Test tab ........................................................................................................................................................96 Total Leaks tab .......................................................................................................................................................97 Retention tab ..........................................................................................................................................................97 Total Leaks tab .......................................................................................................................................................98 Field Cal tab ...........................................................................................................................................................99 Wind Losses tab ...................................................................................................................................................100 Venting tab ...........................................................................................................................................................101 Saved Tests tab.....................................................................................................................................................102 Calibrations and Reports ......................................................................................................................................104 Field Calibration Report.......................................................................................................................................109

Level 1 Training Manual

1

What is Enclosure Integrity Testing

1.1 Why Clean Agent Systems? Water-based suppression systems provide only a minimum amount of fire protection. For critical company information systems, data centers, paper archives, museums and other enclosures whose contents are susceptible to water damage, sprinkler systems will indeed protect the building from fire damage, but at the expense of the contents being protected! It was due to this necessity to provide sufficient fire protection for the building and also to mitigate or eliminate damage to the contents that non-water-based suppression systems were introduced. Halon was one for the first of these fire suppressants. Halon (and other gaseous systems) presented a new problem to the designer however; to extinguish a fire and to keep it suppressed, the gas needed to be present in the enclosure for many minutes. Enclosures now had to be “tight” enough to retain the Halon in sufficient concentration and for sufficient time to ensure that re-ignition did not occur. 1.2 The Discharge Test Prior to 1988, the capability of an enclosure to retain its fire suppressant was assessed by a Discharge Test. Sensors that detected fire suppressant concentration were installed at various points of interest around the room and then the fire suppressant system was discharged. During the discharge, these sensors where monitored, usually with strip-chart recorders. A room would pass or fail the test by examining the agent concentration at the top of the equipment over time. The room passed the test if sufficient agent concentration was present after the required hold time at the top of the equipment. In the event that the room failed, usually the only recourse was for a sealing job to be undertaken and then, for the discharge test to be repeated.

FM200 Discharge Courtesy of Great Lakes Chemical

Aside from the cost of repeated discharge tests, Halon, the predominant agent at the time, was also known to be an ozone depleter. In 1989 the EPA mandated the industry to eliminate all future Halon discharges for the purpose of enclosure integrity verification. Even today, the discharge test is of limited use due to: 1) High Cost: Costs of labor and product to repeatedly recharge system is high. 2) Disruption: Discharge test is very disruptive to occupied enclosures. 3) Failure Identification: In the event of failure, the discharge test offers no opportunity to identify leak locations. 4) Repeat Testing: Although the Standards encourage annual retest, the above factors virtually preclude any retesting

Section 01 – Introduction to Door-Fan Testing Page 6 of 109

Level 1 Training Manual 1.3 The Door-Fan Test Towards the end of the discharge era, several progressive installers found a unique way to ensure that they would always pass the discharge test. They used a fan mounted in a doorway to create pressure which in turn allowed them to locate hidden leaks using chemical smoke. When the leaks were sealed, the room would always pass the discharge test. It worked so well that the discharge test has now been replaced by the Door-Fan Test. NFPA 2001, NFPA 12A, and ISO 14520 now all require an enclosure integrity test as part of the acceptance procedure for all clean agent systems, including all halocarbon and inert gas agents. This comprehensive test and calculation procedure predicts how long the agent will stay in the room if it were ever discharged. The Enclosure Integrity Test’s primary goal is to predict the enclosure’s retention time in the event that the Clean Agent Fire Suppression System is discharged. The discharge test typically only verified agent distribution in one location, usually the most favorable. This often led to assuming that other approval steps for the enclosure could be overlooked. To make matters worse, the discharge test was never repeated. The room leakage would increase steadily, compromising the system from day one. The Enclosure Integrity Test’s simplicity and accuracy, encourages trouble-shooting of problem rooms and retesting either periodically or after enclosure modification. In the past, enclosures were often designed merely to pass the discharge test. This often left rooms with fire barriers on only 5 sides and with the top completely open. Often only ceiling tiles stood between the protected enclosure and an adjacent unprotected area! Smoke or fire could readily enter from above. The Enclosure Integrity Test is also the best way to ensure that the enclosure is protected from smoke events occurring OUTSIDE of the protected room! Now, the EPA, Industrial Risk Insurers, Factory Mutual, other insurers, Fire suppression equipment manufacturers, and the FSSA all encourage Door-Fan tests on every installation. Both NFPA (Sec 4-4) and ISO (Annex E) require Door-Fan tests to be repeated every 12 months, or whenever new holes are made in the enclosure.

Section 01 – Introduction to Door-Fan Testing Page 7 of 109

Level 1 Training Manual What is a Door-Fan test? The Door-Fan Test measures the size of holes in the enclosure using a door-fan to pressurize the room to the same pressure it would experience during discharge. Knowing the pressure inside the enclosure, the pressures across each wall, and the flow through the door-fan to maintain the enclosure pressure, a computer calculates the retention time.

Door-fan pressurizes room

Air flows out through leaks in floors, walls and ceiling

Static Pressures When conducting a door-fan test it is very important to be aware of, and to measure any static pressures in an enclosure. There are two static pressures to be aware of; the static pressure at the time of the door-fan test and the static pressure that will exist during discharge. These static pressures can be created due to damper or duct leakage, wind loading on the enclosure and many other reasons. This static pressure will act to push or pull the agent out faster than normal, reducing the retention time and must be taken care of in any calculation of retention time. State-of-the-art software, such as Retrotec’s CA2001 Windows Software, takes this calculation into account. Refer to C-2.5.2.3 & C-1.3.21 in NFPA2001 for additional information on this concept. Total Room (Whole Room) Leakage Using a single door fan to pressurize the room will measure the total leakage area of the entire room; floor, walls, and ceiling. This result is called the Whole Room or Total Room Leakage. Because this measurement includes the leakage in upper area of the room, where the agent would not normally leak out, it often results in unrealistically large leakage and unrealistically short retention times. A room which passes a Whole room test would most certainly pass a discharge test. A room which fails the Whole Room Test however, might very-well pass a discharge test if the majority of leaks were at ceiling level. When conducting a Total Leaks Test, a single door-fan is temporarily installed in a doorway leading from the protected room to a large open area or outdoors. The fan speed is adjusted to obtain a pressure difference between the test room and the volume surrounding the room. This pressure (usually 10 to 15 Pa) is similar to the steady state pressure (column pressure) exerted by the agent at floor level immediately after discharge. The computer converts flow and pressure readings into an Equivalent Leakage Area (EqLA), the total area of all the cracks, gaps, and holes in the room.

Section 01 – Introduction to Door-Fan Testing Page 8 of 109

Level 1 Training Manual The measurement is done by first blowing air out of the room (depressurization) and then into the room (pressurization). The two readings are averaged to reduce errors due to static pressure, HVAC operation, wind and faulty gauge zeroing. Measuring Lower Leaks A room that fails the Total Leak Test can easily pass a discharge test if all of its leaks are located at the ceiling level. In this extreme case, the heavier-than-air agent settles to the floor and, with nowhere to leak to, remains there indefinitely. Since the leakage area of the above-ceiling space is generally far greater than the belowceiling leakage area (BCLA), measurement of the BCLA can dramatically increase the calculated retention time. The BCLA can be measured separately using a flex-duct or plastic on the ceiling to neutralize any leaks in the above-ceiling space. These techniques eliminate the upper leaks for the purpose of measuring the more important lower leaks. Both leakage measurements are then used to make a more accurate prediction of retention time. Predicting the Retention Time After discharge, the heavier-than-air agent creates a small positive pressure within the enclosure. Flow develops whenever holes have pressure across them. The greater the pressure and the larger the hole, the greater the amount of agent lost. As the agent leaks out the bottom, a small negative pressure develops at the top. This pulls air in through the higher level leaks. Each agent creates a slightly different pressure as indicated by the densities as shown in NFPA 2001. The door-fan test measures the size of the holes within the enclosure. The quantity of agent and height of the room determine the pressure difference across holes in the enclosure. Knowing the size of the holes, the pressure difference, and the minimum equipment height or agent concentration, Retrotec’s CA2001 software predicts how many minutes will pass until the equipment is no longer protected. This time, from discharge until the equipment is no longer protected is called the Retention Time. 1.4 Other Door-Fan Applications In addition to the enclosure integrity test, the door-fan equipment can be used for a number of other applications. Pressure Relief Vents Of interest to the clean agent installer is the testing of rooms for adequate pressure relief. In all cases, if a room is too tight, potentially damaging pressure can develop after agent discharge. Using Retrotec’s CA2001 software, the door-fan test equipment can be used to predict the maximum expected pressure in the room during discharge and calculate the amount of venting required. Section 01 – Introduction to Door-Fan Testing Page 9 of 109

Level 1 Training Manual

If pressure-relief vents are installed in the enclosure, the door-fan test equipment can be used to pressurize the enclosure and test correct design and functioning of the vents. Retrotec’s experience having tested 100’s of enclosures and supported 1000’s of testers world-wide is that pressure relief vents rarely open at the prescribed pressures and if they do, rarely open fully as required. The door-fan equipment can also be used to test for blockages in venting capacity not apparent to visual inspection. Crushed duct-work, debris in duct-work, blocked weather covers and malfunctioning dampers are all easily tested using the door-fan. Smoke and Contaminant Movement An emerging industry is the evaluation and prediction smoke and other contaminant movement through multi-floor building. Conventional predictions have been based on guessed or “typical” leakage areas in floors, walls and ceiling. The door-fan test however can be used to isolate and measure leakage areas of individual walls, ceilings, shafts and floors. New techniques have been developed using three specially designed door fans that allow the tester to measure the leakage of each floor slab separately. Floor to elevator shaft and floor to stairwell leakage can also be measured. Individual shafts can be measured in their entirety. The preliminary results of this testing indicates that most buildings leak ten to 100 times more than they should in order to be safe from smoke moving through the building. Check the Retrotec website at www.retrotec.com or e mail [email protected] for the latest paper on this topic.

Section 01 – Introduction to Door-Fan Testing Page 10 of 109

Level 1 Training Manual

2 Agent Loss Mechanisms
2.1 Pressures Across Holes: How Agent is Lost Consider the case of a bucket; regardless how large an opening at the top, fluid will not leak out.

The same is true for an enclosure full of agent. Shortly after the discharge, once the agent settles out, a room with large openings at the ceiling level will contain the agent indefinitely. With a hole at the top and a hole at the bottom, water will leak out at a rate dictated by the ratio of upper and lower leak areas. If the upper leak is reduced enough, water leakage out the bottom will be reduced. Generally, even if the upper leaks are entirely sealed, water will still slowly leak out as air bubbles into the bucket.

The same is true in the enclosure full of agent. Sealing the upper leaks will eventually slow the flow of agent out of the room, but will not stop it. Agent leakage is primarily governed by the lower leaks. Section 02 – Agent Loss Mechanisms Page 11 of 109

Level 1 Training Manual Given an unlimited capacity for replacement air to enter the bucket, leakage is directly related to the total area of the lower leaks. Double the size of the holes and the leakage doubles. Triple the size of the holes and the leakage triples.

2x holes = 2x leakage

3x holes = 3x leakage

Depending on the density/concentration of the fluid in the bucket; for the same sized hole, as the concentration increases, so does the flow out of the bucket increase. It is not directly proportional however, doubling the leakage area only adds 50-60% leakage.

0% Agent

5% Agent

10% Agent

This is of significant importance in enclosures that fail. A very common misconception is to simply add more agent to increase the retention time. This is, in fact, wrong. Adding more agent increases the concentration and will actually reduce the retention time! Section 02 – Agent Loss Mechanisms Page 12 of 109

Level 1 Training Manual 2.2 Pressures That Cause Agent Loss There are a number of influences that will act on the agent/air mixture as the agent enters the enclosure and during the time it remains within the enclosure. These influences will play a key role in how agent is lost from the enclosure. Dynamic Discharge Pressure In the seconds during and immediately following discharge the agent mixes violently with the air in the enclosure. Swirls and eddies due to various combinations of warming, cooling and expansion lead to a homogeneous concentration of agent throughout every corner of the flooded enclosure. Pressures created in the first few seconds of discharge (dynamic discharge pressures) are ignored in the retention time prediction model. While these pressures can be very large, they often swing wildly between positive and negative values, making them difficult to predict. In addition, their duration is very short making their contribution small when compared to steady state losses. Non-the-less, a small amount of initial agent loss is assumed to occur and the equations used to calculate the required concentration per NFPA have a loss factor built in. What happens next, during the remaining minutes of the retention time however is virtually impossible to model. It is useful however to understand the mechanisms at work within the enclosure and how they affect the distribution of the agent. Gravity Gravity acts on the heavier-than-air clean agent/air mixture within the enclosure and, in the absence of any other influence, will cause the agent/air mixture to leak out the bottom of the enclosure. Air is then pulled into through the leaks in the top of the enclosure. A short period of time after the discharge a well-defined agent-air interface may form, much like the interface between water and air in a bucket, or the way fresh water flows on top of salt water. Agents that are significantly denser than air (such as Novec which is 10x the density of air, and most of the other halocarbons) will be more influenced by gravity than agents that have densities similar to air (such as Argon and the other inert gases). The more dense agents tend to merely run out faster than the less dense. For example, 40% CO2 runs out about twice as fast as 40% Argon. N2, which is actually less dense than air, should theoretically rise to the ceiling, though cooling due to expansion upon discharge should increase its density beyond that of air.

Section 02 – Agent Loss Mechanisms Page 13 of 109

Level 1 Training Manual

2. To be replaced be air

1. Agent/Air Mixture Leaks Out

If there were no leaks in the floor, a pressure would be created at the floor due to the column of heavier-than-air agent pressing upon the floor. This pressure is referred to as the column pressure. If floor leaks exist, as they always do, the column pressure is dropped partially across the floor and partially across the ceiling depending on the ratio of leaks, much like how voltages are dropped across two resistors. It the holes are the same size, then about half the Column Pressure is dropped across the ceiling and half across the floor. The absolute sum being the column pressure. A positive value will be felt at the lowest portions of the enclosure, near zero in the middle of the column and a negative pressure at the top of the air-to agent interface. Convection “Hot air rises” and this is no exception in an enclosure. Even though computer and other electronic equipment may be shut down at the time of discharge, 100’s or 1000’s of watts of heat are still contained in their chassis. This heat will cause localized “plumes” to form around the equipment, drawing in gases from the floor level, transporting them up the equipment towers, to be released above the equipment and cycled back to the floor.

Section 02 – Agent Loss Mechanisms Page 14 of 109

Level 1 Training Manual Even in the absence of hot equipment, upward convective currents will form along locally warm walls and downward currents will form along locally cooler walls. Lighter gases will be more significantly affected by convection than heavier gases. Forced Air Circulation Self-contained air conditioning units, circulating fans, HVAC systems, and even equipment cooling-fans all act to mechanically circulate air around the room. If these fans are running during discharge and during the retention Liebert time, they will act to continually churn the agent and air within the enclosure. Static Pressures Leaky ductwork, HVAC systems with a fresh air supply, and stack effects are just a few of the things that can cause a higher pressure to exist in the enclosure relative to neighboring rooms or outside. This higher pressure, not due to the column pressure of the agent itself is called a Static Pressure. Due to conditions beyond the control of the tester, a static pressure may be present during the Door-Fan Test. By sealing up the blower with a cover or a piece of cardboard, this static pressure can be measured and by opening up a small hole in the blower (for example, by removing a single plug in the low-flow plate) and using the smoke puffer, the direction of the static pressure can be observed. A static pressure that is present during the Door-Fan Test, is called the Static Pressure During Test and must be measured in both magnitude and direction for the test to be accurate. A static pressure within the enclosure during the retention time will act to significantly increase the flow of agent out of the room. In extreme cases, with very leaky HVAC systems or extreme wind conditions, the agent is literally “blown” out of the room. This Static Pressure During Retention must also be assessed and entered into the retention time calculation. Wind Pressures Wind blowing onto the wall of an enclosure can cause huge pressure fluctuations within the enclosure will literally blow the agent out of the enclosure. Add to this the vacuum effect in Section 02 – Agent Loss Mechanisms Page 15 of 109

Level 1 Training Manual the lea of the building and we get a recipe for rapid failure of enclosures that are exposed to frequent windy conditions. Neither NFPA nor ISO standards address this issue. 2.3 Enclosure Behavior during Retention After the initial effects of the discharge have concluded, the enclosure will revert to some form of stead-state behavior. This behavior will be a result of the combined influences of gravity, convection, and forced air circulation as discussed above. The combined behavior in any enclosure will be different based on which of the above influences are present, which are predominant, and which agent is used. The resultant steady-state behavior will lead to one of two modes of retention within the enclosure, Descending Interface and Continual Mixing. The Descending Interface In enclosures with little heat-generating equipment and no forced-air circulation (such as museums and paper archival rooms) the over-riding influence will be gravity. The heavierthan-air agent-air mixture will settle to the floor of the room and a layer of any displaced air will migrate to the ceiling. Due to the column pressure of the mixture, a pressure difference will develop between the area within the enclosure and the rooms outside the enclosure. This pressure difference will cause agent to flow outwards through any holes in the enclosure. The greater the pressure difference, the faster the flow of agent. Agent will flow out of holes at floor level faster than similar sized holes mid-wall. As agent flows out of the holes near the bottom of the enclosure, air will rush into the enclosure at the ceiling level to replace it. As this happens, the agent-air interface will slow drop. This is called a Descending Interface. Equipment will be protected so long as the descending interface is above it. Once the descending interface touches, or drops below the equipment, it is no longer considered to be protected. This protected equipment height should be specified by the enclosure designer, physically measured, or can be taken to be 75% of the room height, as agreed upon by the AHJ. The following three graphs depict the results from an actual Halon discharge with a descending interface. During the test, three probes measuring Halon concentration were set up, each at a different height. Time is given on the horizontal (X) axis and Halon concentration is given on the vertical (Y) axis. In all three cases, the concentration rises rapidly to the approximately 6.5% design concentration.

Section 02 – Agent Loss Mechanisms Page 16 of 109

Level 1 Training Manual At the 173” (14.5’) level, concentration is held for about a minute, and then slowly drops as agent drains from the room. By 4 minutes, the concentration reaches 1% and equipment at that level would no longer be protected.
%Halon. 7 6 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 10 Time (minutes) 11 12 13 14 15 16 17

Probe at 173" from floor slab

At the 146” (12’) level, concentration is held for about 4 minutes, and then rapidly drops over the next 2 minutes to 1%, at which time equipment at this height is no longer protected.
Probe at 146" from floor slab
%Halon 7 6 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Time (minutes)

At the 86” (7’) level, the initial concentration is held for almost 12 minutes and then slow drops to 1% concentration at 17 minutes.
Probe at 86" from floor slab
%Halon 8 7 6 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Time (minutes)

Section 02 – Agent Loss Mechanisms Page 17 of 109

Level 1 Training Manual Continual Mixing In enclosures containing hot and massive equipment, or circulating fans, chillers and HVAC systems that remain running during the retention time, the agent will be continually circulated. As a result, there will be a uniform concentration of agent throughout the enclosure at any point in time. As the mixture leaks out of the enclosure, concentration at the floor will decay at the same rate as the concentration near the ceiling. This is called Continual Mixing. The agent concentration throughout the room will begin at the initial design concentration and over time, decrease. Equipment in the room will be protected so long as the agent concentration is greater than some minimum concentration. This minimum concentration should be stipulated by the enclosure designer or with the agreement of the AHJ or witness, may be based upon the manufacturers’ suggested minimum concentration to prevent reignition. The minimum concentration is often confused with initial concentration, which it is not. For example, the typical starting concentration for FM-200 is 7.5% but according to some manufacturers, approximate minimum concentration to prevent re-ignition is about 5.6%.

Section 02 – Agent Loss Mechanisms Page 18 of 109

Level 1 Training Manual

3 Calculating Retention Time
3.1 What is Retention Time? Retention time is the amount of time it takes until the first piece of protected equipment is no longer protected. Retention time begins the moment that agent begins to be released into the enclosure. The actual discharge itself takes only seconds to complete. This discharge time is considered to be part of the retention time. The retention time ends when the first piece of equipment is no longer protected. This can happen in one of two ways, depending on the mode of retention in the enclosure; either the descending interface reaches the highest piece of equipment; or the agent concentration drops below the specified minimum concentration to prevent re-ignition. Contrary to popular belief, there is no specification in the NFPA standards for a 10 minute retention time. What NFPA 2001 does say is: A.5.6 In establishing hold time, designers and authorities having jurisdiction should consider the following or other unique factors that can influence the performance of the suppression system: 1) Response time of trained personnel 2) Sources of persistent ignition 3) Excessive enclosure leakage 4) System enclosure venting requirements 5) Inertion and re-flash hazards 6) Wind down of rotating equipment The hold time for the duration of protection should be sufficient to control the initial event and allow for support should resurgence occur once the agent has dissipated. Under the NFPA standard, it is up to the designer or AHJ to specify the retention time, and they should as a minimum, take into consideration the time it will take for trained personnel to respond to the situation. A remote cell-site may warrant a hold time of significantly more than the commonly used 10 minutes, while the server room in a fire hall might get away with less than 10 minutes. ISO 14520 Section 4 of Annex E does specify a minimum 10 minute retention time and refers to the guidelines set out in section 7.8.2 to determine potentially longer retention times.

Section 03 – Calculating Retention Time Page 19 of 109

Level 1 Training Manual 3.2 Retention Time in the Descending Interface Case In the absence of any circulation, a defined interface will form between the agent-air mixture and the air as the agent-air mixture flows out of the lower leaks and the external air rushes in through the upper leaks. As the agent leaks out of the enclosure, this interface will slowly descend in height. Initially all equipment in the enclosure will be bathed in agent, but over time, as the agent escapes, the higher equipment will slowly begin to be exposed to the air as the interface drops. 3.3 Retention Time in the Continual Mixing Case In enclosures with circulating fans, self-contained cooling or heating systems, racks of computer equipment with cooling fans, or large heat generating equipment, there will be significant circulation and mixing of the agent and the air. In these cases, no descending interface will form. Instead, the concentration of the agent will remain the same throughout the enclosure and will slowly drop, throughout the enclosure, as the agent leaks out. The concentration will begin at the design concentration immediately after discharge. As the agent leaks out of the enclosure, the concentration will slowly fall. The equipment throughout the enclosure will lose protection when the concentration falls to the minimum concentration. This minimum concentration is usually taken to be the manufacturer’s minimum concentration to prevent re-ignition but must ultimately be specified by the designer, engineer, or AHJ. 3.4 Retention Time in the Extended Discharge Case After the initial discharge, some systems may continue their discharge for many minutes to maintain the agent concentration. This is especially true in enclosures that are extremely leaky. In these cases, the retention time begins at the start of the initial discharge, continues through the extended discharge and then continues through the normal retention time after the extended discharge finishes. The extended discharge is an excellent way to obtain an indefinite retention time without conducting significant room sealing. The flow from the extended discharge need only be enough to account for the leakage from the enclosure, which is usually relatively low. Software such as Retrotec’s CA2001 will calculate the required extended discharge flow rate based on the door-fan test.

Section 03 – Calculating Retention Time Page 20 of 109

Level 1 Training Manual 3.5 Retention Time Dependencies There are a number of factor and enclosure characteristics that affect the retention time. Knowledge of these dependencies up front, before the enclosure is designed can save time, money and aggravation in the long run. Leakage Area (ELA) It may seem obvious, but for the record, leakage area affects retention time. The larger the leakage area is - the shorter the retention time will be. Reducing the leakage area will increase the retention time If only a whole-room test is performed, the worst case 50% upper leaks – 50% lower leaks leakage split must be assumed. If leakage split between upper and lower leaks can be determined, the retention time may be able to be lengthened to compensate for reduced lower leakage. Agent Initial Concentration and Final Concentration When there is no mixing, a descending interface will form. Below the descending interface, the agent concentration will remain constant at the original, design concentration. It is important to understand that, without mixing, increasing the initial agent concentration will increase the column pressure, increase the speed at which the agent escapes from the enclosure and hence DECREASE retention time. A common misconception with a failed room is that adding more agent will fix the problem. With a descending interface, the opposite will occur. 1. With a descending interface, adding more agent will decrease the retention time. When there is mixing, the agent concentration remains homogeneous throughout the enclosure and slowly decreases, throughout the enclosure, as agent leaks out. The concentration begins at the initial design concentration and immediately begins to decrease. The retention time ends when the concentration reaches some specified minimum concentration. This minimum concentration must be lower than the initial concentration. A common misconception is that the enclosure fails when the concentration drops below the initial concentration. This happens almost instantly as agent begins to leak from the room. 1. Retention time is dependent on the difference between the starting concentration and the final concentration. The larger the difference, the longer the retention time. 2. With mixing, adding more agent to a failed enclosure will usually increase the retention time.

Section 03 – Calculating Retention Time Page 21 of 109

Level 1 Training Manual Equipment Height and Enclosure Height When there is no mixing, a descending interface will form. Immediately after discharge, the interface will be located at the highest flooded point in the room which is usually at the upper slab, or at the suspended ceiling if present. This is called the Maximum Agent Height. As agent begins to leak out of the room, the descending interface begins to drop in height, to be replaced with fresh air from above. When the descending interface reaches the highest piece of protected equipment, protection is assumed to be lost. When there is no mixing, retention time is the time from discharge until the descending interface reaches the protected equipment. When there is no mixing, retention time is dependent on the difference between the maximum agent height (the height of total flooding) and the minimum protected height (equipment height). 1. With a descending interface, increasing the ceiling height (increasing the maximum agent height) will increase the retention time. 2. With a descending interface, decreasing the height of the protected equipment will increase the retention time. When there is mixing, retention time does not depend on equipment or enclosure height. 3.6 Measuring Maximum Agent Height? The Maximum Agent Height is measured from the lowest point in the room, to the highest intentionally flooded spot in the room. When the enclosure has a suspended floor, the Minimum Protected Height is measured from the lower slab. Do not measure above suspended ceilings if there is NO agent discharged there. Do NOT measure to the bottom of trenches if there is very little leakage there.
Minimum Protected (Equipment) Height Maximum Protected (Agent) Height

Minimum Protected (Equipment) Height

Maximum Protected (Agent) Height

3.7 Measuring Minimum Protected Height? The Minimum Protected (Equipment) Height is measured from the lowest point in the room to the highest piece of equipment being protected. In cases where there is no equipment in the room, 75% of the overall room height is chosen. When the enclosure has a suspended floor, the Minimum Protected Height is measured from the lower slab. Section 03 – Calculating Retention Time Page 22 of 109

Level 1 Training Manual

Do NOT measure to the bottom of trenches if there is very little leakage there.

Section 03 – Calculating Retention Time Page 23 of 109

Level 1 Training Manual

4 Recurring Design Problems
At Retrotec we’ve tested hundreds of enclosures and provided technical support and consulting on thousands more. Here are some of the most common design problems. Problems we see again and again and again. Most of these problems have simple fixes that are inexpensive and simple to implement during the design phase of the enclosure and are typically expensive and difficult to implant upon discovery during the door-fan test. 4.1 Cascading Pressures
Pressure +25 Pa Flow Pressure +10 Pa Flow Pressure 0 Pa

Protection lost in seconds! Cascading room pressures will create a horizontal flow through the enclosures which will add to normal agent losses. In the above example, the center room is protected with clean agent. After the discharge, the agent will effectively “pile up” on the low pressure side of the room, and be blown out of the room far quicker than predicted. In one case, with a leaking cable tray, protection would have been lost in seconds! This room pressurization system must be shut down at discharge to prevent these uncertainties, or if not possible, the clean agent enclosure must be specifically engineered to resist these pressures.

Section 04 – Recurring Design Problems Page 24 of 109

Level 1 Training Manual 4.2 Common Sub-floor

Sub-floors always seem to present many opportunities for mistakes. One of the most significant problems is when the protected zone is connected to an adjacent room by common sub-floor. This could be due to some (poor) design, or because sections of the partition wall have been mistaken uninstalled (a common occurrence under doorways). Agent is discharged into the protected room and sub-floor but will quickly flow up through the sub-floor into the next room as if it were water finding its level. The Solution to this problem is that either both rooms need to be discharged at the same item or that the walls must be continued to the lower slab.

All too often we observe partition walls that, for some reason, do not continue to the lower slab in the subfloor

This wall rests on a false floor. We are looking from the protected zone into the adjacent unprotected space. Agent will flow under the wall causing the protected enclosure to lose concentration immediately. Flooding the entire sub-floor is often used in an attempt to address the problem. This doesn’t work however, because the agent just forces its way up through the floor from the underside, unintentionally flooding the adjacent room.

Section 04 – Recurring Design Problems Page 25 of 109

Level 1 Training Manual 4.3 Pressurized Sub-floor Wall cavities that connect the sub-floor space to the above-ceiling space will cause agent to be forced above the false ceiling if the sub-floor is pressurized during the retention period.

Pressurized sub-floor pushes agent into the above-ceiling space 4.4 Common Above-ceiling Spaces: Discharge will pull smoke in Common above-ceiling space

smoke

smoke

Smoke event in neighbouring room

Protected Zone

Walls that only extend up to the T-bar suspended-ceiling rely on flimsy tiles and clips to hold in the gas at discharge and to protect the enclosure from fire and smoke events outside the protected zone. The result is inadequate protection. Agent can be lost at discharge when the tiles get blown away, reducing agent concentration. Worse however is that smoke from an external event (smoldering trash bin or neighbouring event) can cause the clean agent system to discharge. As the agent naturally leaks out, smoke will be “pistoned” into the enclosure, causing unnecessary damage to the equipment and unnecessary costs for clean-up, service, and recharge of the system. Section 04 – Recurring Design Problems Page 26 of 109

Level 1 Training Manual 4.5 Common Above-ceiling Spaces: HVAC Leakage HVAC pressures arising from either leaky supply or leaky returns will act to push or pull agent out of the enclosure faster than expected.
HVAC

Leaky Supply Dampers

Passive Return Leaks

Increased Pressure Forces Agent out Faster

4.6 Suspended Ceilings Too Low
Ceiling void

Better design using slightly more agent to act as reserve over the equipment. No connection to rest of building. Fire barrier on all sides.

Equipment

Equipment Worst possible design. Leaky T-bar, suspended ceiling connects enclosure to events in other parts of the building. No reserve over the equipment. Short retention time.

Note: Containment on all 6 sides is required by NFPA 75.

Section 04 – Recurring Design Problems Page 27 of 109

Level 1 Training Manual

5 Good Enclosure Design Practices
5.1 Superior Protection for Less Money! Interested? Consider the case of a 500 ft2 room where $10,000 and two weeks were spent to seal it tight enough to pass the enclosure integrity test. If a few $100 more had been spent on more agent, and the ceiling raised several feet, this could have been avoided. Often the general contractor finds himself rebuilding a room that was not designed to be tight enough to hold agent. There are 4 guidelines that will: • Ease passing the door fan test • Dramatically improve fire and smoke protection • Solve 90 % of the design problems that have to be solved at the last minute just prior to occupancy 5.2 Run walls slab to slab Include construction details that would allow for sealing of the wall to the upper slab. extending walls to the upper slab and sealing them airtight is often the only defense from fire and smoke entering the enclosure from the outside. This sealing is the MOST important thing that can be done to improve protection in the enclosure. Refer to C-1.2.1 (b) in NFPA2001. This is the easiest way to get slab to slab walls sealed. This spray on flexible rubber is available from 3M and Grace and it has a fire rating! Better yet, it is flexible and will not crack and fall out as many other treatments will. Loaded floors can move ½ an inch, but this sealant will remain flexible over that range. Just stuff Rockwool backing in any size gap and start building up layers of this rubber sealant. And we don’t even get any commission for saying this.

Section 05 – Good Enclosure Design Practices Page 28 of 109

Level 1 Training Manual 5.3 Eliminate T-bar suspended ceilings Eliminate T-bar suspended ceilings in enclosures where the walls do not go slab to slab. Use a solid sheetrock ceiling with access hatches and walkways above it. a. Clipping tiles is often used but is ineffective. Clips are lost almost immediately. As soon as the tiles are opened the clips go flying and are never replaced. This is an example of a practice that was commonly used to keep tiles in place during a discharge but has no practical application for long-term protection. This recommendation is in spite of its mention in NFPA2001 section C-1.2.3.10 b. Discharge agent above the false ceiling. Often for a few added pounds of agent, an immense improvement in protection can be gained. For starters, ceiling tiles will usually go flying during a discharge, causing agent to get lost above the suspended ceiling. This agent will mix with the above-ceiling air to provide a concentration that is lower than the initial concentration. Some of this agent may come down to replace losses below, but at a decreased concentration. Discharging agent above the false ceiling solves the displaced ceiling-tile problem and in most cases will triple the retention time. Good value. c. Use Fire Rated ceiling tiles. This would be an option for existing installations. The positive view of T-bar ceilings T-bar suspended ceilings are low cost, do conceal ductwork and wiring, and tiles can be removed to gain access. On the negative side T-bar suspended ceilings usually begin to look tattered even before the construction job is complete. The tiles often go flying when the system is discharged. Fire and smoke events occurring outside the protected zone are far more likely to cause damage in the enclosure than events that occur within it. The sheetrock ceiling option The sheetrock ceiling provides a complete enclosure to protect the contents of the room from externally generated smoke damage. This increased protection or compartmentalization shows up when the leakage of the room is measured using the door fan. Instead, install a sheetrock ceiling with access hatches; cover it with plywood, then sheetrock above so it can be walked on while servicing the equipment above. 5.4 Maximize the room height and volume Place the ceiling as high as possible. More clean agent = more protection. In small rooms, run pipe and supply nozzles to fill the above-suspended-ceiling space. Section 05 – Good Enclosure Design Practices Page 29 of 109

Level 1 Training Manual

The greater volume of clean agent in the enclosure, the greater the protection. Merely putting a higher concentration in the room will only ensure the agent will run out of the room faster (this is not true in the less common case where the agent is continually mixed after discharge). 5.5 Select an appropriate retention time NFPA 2001 states “... the design concentration ... shall be maintained for a sufficient period of time to allow effective emergency action by trained personnel.” The following guidelines are suggested for small enclosures. For example, a remote site where re-ignition was possible, and where it would take 30 minutes for trained personnel to arrive, should be specified as 30 minutes. On the other hand, a small room with little or no potential for a deep-seated fire and where personnel would respond within 5 minutes would need a retention time of 5 minutes. NFPA 2001 does not recommend any specific time. The AHJ must ultimately decide what time is appropriate. ISO however, does specify 10 minutes. All rooms typically have at least one door that will generally leak about 5 to 20 in2. A 350 ft3 room with a 10 minute hold time requires a leakage of 7 in2 or less to pass. Since that is not really practical, reducing the specified hold time or adding an extended discharge is the only option. For room volumes of: Minimum achievable leakage area is: Suggesting a retention times for inerts: And suggesting retention times for halocarbons: 2,500 62 10 8 1,250 42 10 6 625 350 ft3 32 8 4 23 6 3 in2 minutes minutes

Alternatively, make the room bigger or discharge agent above false ceilings. 5.6 Fit automatic door closers Doors often get wedged or propped open when the room is in use. This practice must be discouraged because the clean agent system will not work properly with perimeter doors open. A better solution is automatic door release mechanisms that will close the doors whenever the first alarm sounds. Choose a door opener that will close the door when it is de-energized so on power failure the doors close.

Section 05 – Good Enclosure Design Practices Page 30 of 109

Level 1 Training Manual

6 The Door-fan Test
6.1 How a Single Door-fan “Sees” a Room

3 units

1.5 units

1 unit

1 unit

3 units

1.5 units

1 unit

Each of the above four enclosures has 3 units of leakage area. The first one has its three units located at the ceiling level, the second at the floor level, the third equally split between the ceiling and floor and the fourth equally split between the floor, ceiling and mid-way up the wall. Each enclosure, however, has a significantly different retention time as indicated below.

3 units
Infinite retention time Longest retention time

1.5 units
Shortest retention time

1 unit
Middle retention time

1 unit

3 units

1.5 units

1 unit

A single door-fan will measure the size of the leakage area in each room at 3 units. Both the NFPA and ISO standards make the assumption that a 50-50 split of leakage between ceiling and floor exists (the third room in the example). This assumption leads to the most conservative calculation of retention time. This calculation is called the Whole Room Test or the Total Leaks Test and is always the first step in the enclosure integrity procedure. Retrotec’s CA2001 software automatically performs this calculation and generates the retention time based on the 50-50 leakage distribution.

Section 06 – The Door Fan Test Page 31 of 109

Level 1 Training Manual 6.2 The Lower Leaks Test As stated above, the whole room test will produce the most conservative calculation of retention time. If the total leakage of the enclosure is small enough compared to the volume of the enclosure, it is possible that the enclosure will pass with sufficient retention time. The vast majority of rooms however have significantly more total leakage than the retention time calculation permits. Rooms with suspended ceilings are the most problematic, as the areas above the suspended ceiling are rarely sealed sufficiently to pass the Whole Room Enclosure Test. In this case, the standards dictate that the leakage distribution must be measured. This is done by measuring the lower leaks. To accurately measure the Lower Leaks (those leaks below the ceiling), one of two test techniques may be employed – The Plastic-on-the-Ceiling-Test or The Flex Duct Test. Both of these tests attempt to isolate the Above-Ceiling Leaks from the Below-Ceiling Leaks. In doing so, a more accurate calculation of retention time can be made. If only considering agent retention rooms with excessive above-ceiling leaks will hold agent just fine. However, this is a poor design practice as they will not prevent smoke ingress from adjacent rooms, nor will they prevent external events from triggering the suppression system.

Section 06 – The Door Fan Test Page 32 of 109

Level 1 Training Manual

7 Witnessing an Enclosure Integrity Test
The witness plays a key role in ensuring that the test is completed correctly. Many witnesses are unaware however, of exactly what they need to look for and what factors affect the test results. What follow is a short explanation of each of the points that the witness should be aware of during the test. The 4-page Enclosure Integrity verification Form that follows is a comprehensive check-list showing each and every point that must be inspected to complete an approval according to NFPA 2001. 7.1 Technician Training 6.6.1 All persons who could be expected to inspect, test, maintain, or operate fire extinguishing systems shall be thoroughly trained and kept thoroughly trained in the functions they are expected to perform. NFPA 6.6.1 requires that testers be appropriately trained. The testing technician should be certified to at least Level 2 for conducting a single door-fan test and at least Level 3 when conducting a dual door-fan test. Testers who have completed any part of Retrotec’s training program are listed on our website at www.retrotec.com, with complete details of their level of certification. Technician certification is stored within the CA2001 software, is displayed prominently in the software and on reports, and is easily viewable by the authority upon request. 7.2 Software Conformance Analysis tools in use by the technician to produce ELA and retention time calculations must conform to the standard being tested to (either NFPA or ISO). Retrotec’s HA6 DOS software and CA2001 Windows software conform to the NFPA 12A and 2001 standards exactly. CA2001 conforms to the ISO 14520 standard exactly. 7.3 Room Pressure Gauge Calibration Certificate The NFPA standard requires that the pressure gauge used to measure the room pressure be calibrated annually. Experience shows however that unless the gauge has been damaged or moves unevenly it is generally within 10%.

Section 07 – Witnessing a Test Page 33 of 109

Level 1 Training Manual Calibration certificates for each piece of equipment are stored within CA2001. The witness should ensure that the gauge being used by the technician has a current calibration certificate within CA2001 and that the serial number on the gauge matches the serial number in the software. The ISO standard recommends calibration but does not suggest the interval. It does require +/- 1% accuracy. Retrotec recommends annual calibration of all pressure gauges. 7.4 System calibration C.2.2.1.6 Door fan systems should be checked for calibration every 5 years under controlled conditions, and a certificate should be available for inspection at all integrity tests. The calibration should be performed according to manufacture’s specifications. The certificate should include the following: 1) Description of calibration facility and responsible technician. 2) Date of calibration and serial number of door fan. 3) Room pressure gauge error estimates at 8, 10, 12, 15, 20 and 40 Pa measured by both ascending and descending pressures (minimum). 4) Fan calibration at a minimum of 3 leakage areas (approximate): 0.5 m², 0.25 m², and 0.05 m² measured at a pressure of 10 Pa. 7.5 Field Calibration check procedure A field calibration check can be requested by the witness to see if the equipment and operator can actually measure a hole of a known size. This test takes very little time to perform and is the perfect way to gain confidence in the tester, test equipment, and test technique. It is preferable to inform the operator beforehand of the expectation to perform a field calibration check so the operator can bring the requisite equipment. 7.6 Return Path There must be a complete and unobstructed flow path from every leak in the enclosure back to the Door-fan otherwise some leaks may not be measured. This may entail opening stairwell or elevator doors to floors above and below, neighboring room doors, and perhaps windows and doors leading outside (if the enclosure under test borders an external wall). The witness should ensure that the operator has examined and accounted for the return paths from all leaks.

Section 07 – Witnessing a Test Page 34 of 109

Level 1 Training Manual 7.7 Room and Equipment Set-up All doors within the zone must be open. The HVAC system and all dampers must be set as they would be at or during a discharge. A doorway must be selected that opens into the largest and most open space. Applying tape to seal up the Door-fan panels is OK as long as the doorway is tighter than the panels. The volume of the room should be calculated on-site using a tape measure or by counting tiles. Taking volume measurements from blue-prints or site-staff as gospel, without confirmation, should not be permitted. Maximum agent height and minimum protected height must be physically (and correctly) measured on-site. Similarly, the operator must confirm the weight or volume of agent through actual inspection of the bottles, not by consulting design documents or talking to site staff. 7.8 Static Pressure Check Before beginning the test, the operator should accurately determine the static pressure. With the doorway completely sealed and the door-fan completely sealed, the magnitude of the static pressure must be measured. By opening a small hole (perhaps a single low-flow plug) and using smoke, direction of the static pressure should be determined. Relying on the positive/negative sign as read from a digital gauge is very error-prone and should be discouraged. The origin of pressures above 3 or 4 Pa should be determined and rectified if possible. 7.9 Gauge Set-up In the case of analog gauges, gauge leveling and zeroing must be completed before any tubes are hooked up. Most analog gauges are meant to be mounted vertically (usually on doorframe or in a pre-designed rack or console). Once mounted, the gauges should be carefully zeroed using small screw-driver and tapped gently to ensure stability. Once zeroed, analog gauges should be quite stable over the course of the test and need only be spot-checked occasionally or tapped prior to each measurement. In the case of digital gauges, zeroing is usually done through a pneumatic switch, either manually or automatically. Many digital gauges are susceptible to position and once zeroed, should not be moved. Digital gauges also tend to drift slightly over time as they warm up. Digital gauges should be zeroed prior to each test measurement. The witness should ensure that gauges are zeroed correctly. Section 07 – Witnessing a Test Page 35 of 109

Level 1 Training Manual

7.10 Flow and Room Pressures Entered Correctly In addition to ensuring that the gauges are correctly reading the pressures, the witness should also confirm that the operator actually records the pressure correctly and enters it correctly into the software. 7.11 Range Selection The “Ranges” on a Retrotec door-fan refers to the selection of flow-restrictor-plates or orifices that change the airflow through the blower. Testing on the correct Range is extremely critical for good results for two reasons. Often the same room-pressure can be achieved with a number different Ranges (just like a manual transmission car can achieve the same speed in a number of different gears). For the best results, the door-fan should be running above ½-speed and the flow-pressure through the door-fan must be greater than the room-pressure and should be at least double the room pressure. If these conditions aren’t met, the operator should change to the next smaller flow range. If the operator does not input the correct flow-range that the test was conducted on into the software, the results can be significantly affected. Retrotec’s CA2001 Windows software displays a picture of the selected Range, which must be acknowledged, before the results of each test are entered. On older versions of software, the ranges can only be identified by consulting a pictorial table. The witness must confirm that the range on the printout or shown in software during the test matches the range that is actually used by the operator. 7.12 Testing in Both Directions Rooms must be tested positively and negatively to eliminate bias due to duct leaks and other static pressures. Testing in both directions is not done due to positive pressures after discharge. 7.13 Determining the Leakage Split – The BCLA Test To measure leaks in the lower part of the room is not the same as taping up leaks to pass a discharge test. The leakage of the whole room, including the ceiling is measured in the first test, the Total Leaks Test. The lower leaks can be measured separately to get a more accurate prediction of retention time. The Total leaks test MUST be performed first and then, and only then, can the BCLA test be performed.

Section 07 – Witnessing a Test Page 36 of 109

Level 1 Training Manual During a ceiling-neutralization/flex-duct test, the witness should ensure that the operator has indeed neutralized smoke flow across the ceiling. Smoke must be used in this test as equalizing pressures using a gauge is not adequate. 7.14 Technical Judgment As a last resort to other BCLA techniques, Retrotec software has a spreadsheet method of determining leak location. For enclosures with extremely large overhead leaks they may be passed using “technical judgment” as per the following section: Section C-1.2.2 (e) of NFPA 2001 App C version 2000 states: “Enclosures with large overhead leaks but no significant leaks in the floor slab and walls will yield unrealistically short retention time predictions. Experience has shown that enclosures of this type can be capable of retaining clean agent for prolonged periods. However, in such cases the AHJ might waive the quantitative results in favor of a detailed witnessed leak inspection of all floors and wall with a door fan and smoke pencil.” It must be understood that, although this test will show the enclosure’s ability to hold agent in a discharge test, a large measure of the passive fire protection has been eliminated due to the absence of an upper sealed smoke and fire barrier. 7.15 Yearly Retests All enclosures must be re-tested yearly if any doubt exists as to whether the room has had any more holes put in it in the last year. 7.16 Commonly Needed Inert Gas Clarifications Inert gases are all heavier than air (with the exception of Nitrogen) and will, according to NFPA 2001, run out of room leaks. They are not as heavy as halocarbons and usually only run out at half the rate. They do run out though. Inert agent enclosures have a very narrow window where they must be tight enough to hold the agent but loose enough to vent peak discharge pressures. All inert clean agents need relief vent areas according to their manufacturers. Inert agents still must pass the door-fan tests. If the enclosure has a vent, its area can be checked with the door fan and some vents can be tested for their ability to open under pressure.

Section 07 – Witnessing a Test Page 37 of 109

Level 1 Training Manual 7.17 Enclosure Integrity Test Verification Form Because there are so many aspects for the witness to keep track of during the enclosure integrity test, Retrotec has developed an easy-to-follow check-off sheet. You can also find this form in the appendices at the end of this document. Please feel free to copy this form and customize it for your own organization as you require.

Section 07 – Witnessing a Test Page 38 of 109

Level 1 Training Manual

Enclosure Integrity verification form
Building Room Test # Testing technician Witness Date and time of test
Check off __ yes __ no __ yes __ no __ yes __ no __ yes __ no __ yes __ no __ yes __ no Screen tab Name on Computer Screen “View” button will display the current certificate “View” button will display the current certificate “View” button will display the current certificate Elevation What to look for

Home

Is the One Year Calibration Certificate up to date?

Home

Is the Five Year Calibration Certificate up to date? Does the technician have the correct level of training? See Level 1-A, page 15 Is it correct within 1000 ft.? See Level 1-A, page 12 This is used to re-calculate the design concentration. It must be re-measured, was it? See Level 1-A, page 12 Was the temperature expected during a discharge within 10F or 5 C? It may differ from the temperature at the time of test. See Level 1-A, page 12 Was it re-measured from floor slab to highest combustible? Enter the maximum agent height from lower slab to highest point that is flooded with agent. See Level 1-A, page 13 Do you agree with the time shown? See Level 1-A, page 13

Home

Building/Room

Building/Room

Net protected room volume

Building/Room

Room operating temperature

__ yes __ no __ yes __ no

Building/Room

Maximum agent height

Building/Room

Minimum agent retention

Section 07 – Witnessing a Test Page 39 of 109

Level 1 Training Manual
Check off __ yes __ no __ yes __ no __ yes __ no __ yes __none __ yes __ no __ n/a __ yes __ no __ yes __ no __ yes __ no __ yes __ no Total Leaks Total Leaks Total Leaks Enter untested values Name on Computer Screen Agent weight

Screen tab

What to look for Were you able to confirm the agent weight or volume? See Level 1-A, page 14

Agent/Test

Agent/Test

Agent volume

Only used for INERGEN in North America

Agent/Test

Initial Concentration

Does the concentration meet the specification? See Level 1-A, page 14 Remove all temporary tape or get sufficient assurance it will be replaced with a permanent seal. If untested values were entered, do you agree with their validity? See Level 1-A, page 16. It would be unusual to have untested values.

Smoke

Did you see the smoke movement test at the doorway? See Level 1-A, page 17

Total Leaks

Test both directions:

Was the enclosure tested in both directions?

Total Leaks

Static pressure

Did you observe the static pressure measurement at the time of the door fan test?

Temperature during Was the temperature within 100F or 50C of that test(0F) recorded? The NFPA Procedure requires a measurement if the difference is greater than 18 Temperature during 0F or10 0C test(0C)

__ yes __ no Total Leaks Range for room pressures:

Was the room pressure reading within the range specified?

Section 07 – Witnessing a Test Page 40 of 109

Level 1 Training Manual
Check off __ yes __ no Total Leaks Blower range Name on Computer Screen

Screen tab

What to look for Was the range used verified? A picture of the range will pop up on CA2001 when the room pressure is entered or see the range section a few pages ahead. Were the correct room pressures recorded with the door fan running? Were the pressures checked across each wall or was there sufficient return path from the door fan to the enclosure leaks to ensure the pressure was the same across all enclosure boundaries. Was the flow pressure accurately recorded while the door-fan was running without pop ups in CA2001 warning of the pressure being too low? Applies only to multipoint tests and here the value must be 6% or less for the test to be good Must be between 0.45 and 0.9 to be acceptable applies only to ISO tests Must be 99% or higher on multipoint tests applies only to ISO tests Must be 0.07 or less - applies only to ISO tests Were the Total Leaks measured first and was the smoke was neutralized at ceiling openings before readings were taken? The resultant leakage area must be entered in the Lower Leak tab, NOT the Total Leak tab (common mistake). Were the Total Leaks measured first? The resultant leakage area must be entered in the Lower Leak tab, NOT the Total Leak tab (common mistake). Are you comfortable with the method of estimation?

__ yes __ no

Total Leaks

Room pressure

__ yes __ no ISO only ISO only ISO only ISO only

Total Leaks

Flow pressure:

Total Leaks Total Leaks Total Leaks Total Leaks

Error %: Slope n: Correlation: Standard Error:

__ yes __ no

Lower Leaks

Flex-duct test

__ yes __ no __ yes __ no __ yes __ no

Lower Leaks

Plastic on ceiling test

Lower Leaks

Estimated

Retention

Mixing during retention

Would this occur? See Level 1-A, page 24

Section 07 – Witnessing a Test Page 41 of 109

Level 1 Training Manual
Check off __ yes __ no __ yes __ no __ yes __ no Name on Computer Screen No mixing during retention

Screen tab

What to look for Would conditions be calm enough after discharge for no mixing of agent and incoming air? Is there an extended discharge. If so, what is the quantity and duration? Have the enclosure set up as it would be just prior to agent discharge. Do you think the smoke direction recorded is the same as what it would be during the retention period? Was the height above the lower slab to the equipment being protected, properly measured? Only applies if there is no mixing but it is useful to have it recorded either way. Do you agree that the value recorded will prevent re-ignition at the end of the retention period? Only applies if there is mixing but it is useful to have it recorded either way. This is the retention time given all the variables input into CA2001 so far. If this is greater or equal to the “Minimum agent retention” specified on the Building/Room tab, the room PASSES.

Retention

Retention

Extended discharge

Retention

Smoke

__ yes __ no __ yes __ no __ not sure __ pass __ fail

Retention

Minimum protected height

Retention

Minimum concentration

Retention

Time,t

Note any other concerns you had about the test ->

Section 07 – Witnessing a Test Page 42 of 109

Level 1 Training Manual 7.18 Standards and How They Apply Standards NFPA 2001, Appendix C ISO 14520, Annex E standard NFPA 12A, Appendix B NFPA 12, not included Agents All Agents except halon and CO2 All Agents Halon only CO2 only Latest edition 2000 2000 DRAFT 1997 2000

Section 07 – Witnessing a Test Page 43 of 109

Level 1 Training Manual 7.19 Range List for Door Fans Flow Range Pictures for 2000 Series Door-fans Always start in the Flow Away position. Adjust speed till room pressure is reached. Fan must be running at least at half speed and flow pressure must be greater than room pressure. If not, insert the next lower ranges until the motor is running at least half speed and flow pressure is greater than room pressure.

Range 22 - Flow Away

Range A - Flow Away

Range B - Flow Away

Range C8 - Flow Away

Section 07 – Witnessing a Test Page 44 of 109

Level 1 Training Manual

Range C4 - Flow Away

Range C2 - Flow Away

Range C1 - Flow Away

Range 22 - Flow Towards

Range A - Flow Towards. The inlet is now on the other side. Look at “Flow Away” picture to see what the inlet must look like.

Range B - Flow Towards. The inlet is now on the other side. Look at “Flow Away” picture to see what the inlet must look like.

Section 07 – Witnessing a Test Page 45 of 109

Level 1 Training Manual

Range C8 - Flow Towards. The inlet is now on the other side. Look at “Flow Away” picture to see what the inlet must look like.

Range C4 - Flow Towards. The inlet is now on the other side. Look at “Flow Away” picture to see what the inlet must look like for C2 and C1 ranges below this one.

Section 07 – Witnessing a Test Page 46 of 109

Level 1 Training Manual

Flow Range Pictures for 900 series Door-fans Flow Ranges must appear exactly as shown

Range 18F for flow Away from the operator

Range 18R for flow Towards the operator

Range 9 for slightly tighter rooms

Range 5

Range 3

Range 1.4

Range 1.3

Range 1.2

Range 1.1

Range 0.1

Section 07 – Witnessing a Test Page 47 of 109

Level 1 Training Manual 7.20 Small Room Retention Times For the purposes of this discussion, small rooms will be defined as 2,500 ft3 or less. Some compromises can be made for small rooms for two reasons:

• •

They are not likely to have a large fire that would threaten the rest of the building Once trained personnel arrive and open the door, the enclosure integrity (and some of the agent) is lost anyway so shorter retention times can be considered

Selecting an Appropriate Retention Time NFPA 2001 states “... the design concentration ... shall be maintained for a sufficient period of time to allow effective emergency action by trained personnel”. The following guidelines are suggested for small enclosures. At a remote site, for example, where re-ignition was possible and where it would take 30 minutes for a responsible party to arrive should be specified as 30 minutes. On the other hand, a small room with little or no potential for a deep-seated fire and where personnel would respond within 5 minutes would need a retention time of 5 minutes. NFPA 2001 does not recommend any specific time. The AHJ must ultimately decide what time is appropriate. ISO does specify 10 minutes. Each room must have at least one door, and that door will leak about 5 to 20 in2. A 350 ft3 room with a 10-minute hold time requires a leakage of 7 sq.in. or less to pass. Since that is not really practical, reducing the specified hold time or an extended discharge is the only option. Recommended Times for Small Rooms For room volumes of: Minimum achievable leakage area is: Suggested retention times for inerts: Suggested retention times for halocarbons: 2,500 1,250 625 350 cu.ft. 62 10 8 42 10 6 32 8 4 23 sq.in. 6 3 minutes minutes

Section 07 – Witnessing a Test Page 48 of 109

Level 1 Training Manual

Appendices

Page 49 of 109

Level 1 Training Manual

8 Appendix A – Agent Comparison

8.1 Standards
Standards NFPA 2001, Appendix C ISO 14520, Annex E Standard NFPA 12A, Appendix B NFPA 12, not included Agents All Agents except Halon and CO2 All Agents Halon only CO2 only Latest Edition 2000 2000 DRAFT 1997 2000

Appendix A – Clean Agent Comparison Page 50 of 109

Level 1 Training Manual

8.2 Agents Agent’s Trade Name
Air

Manufacturer with registered trademark 
n/a Air

NFPA name

Chemical name
n/a

Density kg/m3
1.205

INERT Agents
Argon Argonite  CO2 Inergen  Nitrogen Minimax GmbH Ginge Kerr Chemetron, others Ansul Cerbex AG IG-01 IG-55 CO2 IG-541 IG-100 Ar N2 50%, Ar 50% CO2 N2 52%, Ar 40%, CO2 8% N2 1.700 1.410 1.832 1.430 1.165

HALOCARBONS
CEA-308  CEA-410  FE-227  FE-241  FE-25  FE-36  FIC-1311 FM-200  Halon SIII  Novec 1230  3M 3M DuPont DuPont DuPont DuPont Great Lakes Chemical Recycled only NAF 3M FC-218 FC-3-1-10 HFC-227ea HCFC-124 HFC-125 HFC-236fa FIC-1311 HFC-227ea Halon1301 HCFC BlendA None yet CHClF2 82% ? C3F8 C4F10 CF3CHFCF3 CHClFCF3 CHF2CF3 CF3CH2CF3 CF3I CF3CHFCF3 7.905 9.850 7.260 5.830 5.060 6.545 8.051 7.260 6.283 3.840 12.937

Appendix A – Clean Agent Comparison Page 51 of 109

Level 1 Training Manual

8.3 Specific volume and density constants for agents GAS
K1 Air Argon Argonite CEA-308 CEA-410 CO2 FE-13 FE-227 FE-241 FE-25 FE-36 FIC-1311

METRIC
K2 kg/cu.m. K1

IMPERIAL
K2 lb./cu.ft.

1.202 0.56850 0.65980 0.11712 0.09410 0.49747 0.31640 0.12690 0.15750 0.18250 0.14130 0.11380 0.12690 0.14871 0.65799 0.24130 0.07100 0.00208 0.00242 0.00467 0.00030 0.00203 0.00120 0.00050 0.00060 0.00070 0.00060 0.00050 0.00050 0.00057 0.00239 0.00088 0.0003 9.850 1.832 2.915 7.260 5.830 5.060 6.545 8.051 7.260 6.283 1.430 3.840 12.9366 1.4090 7.4472 4.7250 1.8850 2.3395 2.7200 2.0980 1.6830 1.8850 2.2062 9.8579 3.6120 1.0627 0.00310 0.01806 0.01070 0.00460 0.00580 0.00060 0.00500 0.00440 0.00460 0.00505 0.02143 0.00790 0.00267 1.700 1.410 8.5140 9.8809 0.01850 0.02150

0.0750 0.1061 0.0880

0.6149 0.1144 0.1820 0.4532 0.3640 0.3159 0.4086 0.5026 0.4532 0.3922 0.0893 0.2397

fm-200
Halon

Inergen 
SIII Novec 1230

• • • •

ss= k1 + k2 * T ss= specific volume in cu.ft. / lb. or cu. m. / kg. T = temperature in F or C Densities are shown in lb./cu.ft. or kg/cu.m. for the agent in the gaseous state at 70F or 21C Higher densities and concentrations will result in greater column pressures. See the table (2 pages forward) labeled “comparing retention times…descending interface case.” The higher the column pressure, the shorter the retention time

Appendix A – Clean Agent Comparison Page 52 of 109

Level 1 Training Manual

8.4 Concentration Ranges ... from an enclosure leakage perspective
Warning! Not for use in designing systems. The purpose of this is to review the range of concentrations used in calculating retention times. Small changes in concentration do not affect the retention time very much unless the agent is continually mixed throughout the retention period. Typical design % for occupied spaces Europe 5 % N America 6 % Maximum design % for occupied spaces 7.5 % not suitable 7% 7.44 % 8.6 % 5 -12 % 16- 20 % 9% 9% 13 % 24 % * 24 % * 28 - 65 % 5.5 % 5.5 % 8.6 % 5% 15 % 9% 9% 10 % 40 % 50 % 7.5 % 37.5 % 43 % ** 43 % 34 – 50 % ? 28.6 % 43 % 43 % 10.5 % 10.5 % > 10 % > 40 % > 50 % 10 % 52 % 52 % Design range for unoccupied spaces Minimum % at the end of the NOAEL LOAEL retention period 2.5 % 5% 7.5 %

Agent

Halon CO2 FM-200 for Fike & Kidde-Fenwal FM-200 for Chemetron SIII CEA-410 FE-13 FE-25 Inergen  Argonite FIC-1311 FE-241 FE-36 Argon CEA-308 Novec 1230

1% 10 % 43 % 8.8 % 5% 4.6 % ? 43 % 30 %

2.5 % 15 % 52 % > 30 %

Appendix A – Clean Agent Comparison Page 53 of 109

Level 1 Training Manual
Typical design % for occupied spaces Most commonly specified. Usually for the typical room where a descending interface is assumed for purpose of calculating the retention period per NFPA 12A & 2001 Appendix B. Higher concentrations yield somewhat shorter retention times due to increased pressure.

Maximum design % Design concentration specified where continual mixing will occur during the retention for occupied period such that the concentration degrades at all elevations at the same time. Higher spaces concentrations yield somewhat longer retention times due to increased pressure. Design range for Typical range for agent use in unoccupied areas. Presumably, any concentration could be unoccupied spaces used here. Minimum % at end Assuming continual mixing during the retention period, the concentration will drop of the retention gradually over the entire period. The higher the initial % and the lower the final %, the period longer the retention time (often 10 minutes). NOAEL LOAEL From 2001. No Observable Adverse Effect Level. From 2001. Lowest Observable Adverse Effect Level.

Appendix A – Clean Agent Comparison Page 54 of 109

Level 1 Training Manual

8.5 Comparing Retention Times ... descending interface case
Example: 1,000 cu.ft. room at 70 F, 10 ft. high, min. protected height of 7.5 ft., ELA (leakage area) of whole room measured with door fan of 22 sq.in. Warning! Agent concentrations shown are for demonstration of comparative retention times in the same enclosure and may not reflect their ability to put out a fire or keep it out. Consult a qualified fire protection engineer for appropriate design concentrations.

Agent Argon Argonite CEA-308 CEA-410 CO2 FE-13 FE-227 FE-241 FE-25 FE-36 FIC-1311 fm-200 halon Inergen  SIII

Quantity of agent 470 cu.ft. 470 cu.ft.

% concentration 37. 5% 37.5%

Column Pressure Pc 5.6 Pa 2.3 Pa

NFPA calculated retention time 14 minutes 21 minutes

61 lb. 52 lb. 40 lb. 37 lb.

9% 37.5 % 18 % 7.55 %

23.1 Pa 7.1 Pa 9.2 Pa 13.7 Pa

8 minutes 14 minutes 11 minutes 10 minutes

37 lb. 25 lb. 470 cu.ft. 24 lb.

7.55 % 6% 37.5 % 9%

13.7 Pa 9.2 Pa 2.6 Pa 7.2 Pa

10 minutes 11 minutes 20 minutes 13 minutes

Column Pressure Pc NFPA calculated retention time

This is created by the weight of agent pressing on the floor. As soon as the agent begins to leak out, this pressure is reduced by the pressure drop across the upper leaks and the loss of agent. In this example, this is the time for a descending interface between the agent below and the air above to drop to 7.5 ft. above the slab.

Appendix A – Clean Agent Comparison Page 55 of 109

Level 1 Training Manual

8.6 Comparing retention times ... continual mixing case
Same example: 1,000 cu.ft. room at 70 F, 10 ft. high, ELA (leakage area) of whole room measured with door fan of 22 sq.in. - but min. concentration at 10 minutes shown in table. Warning! Agent concentrations shown are for demonstration of comparative retention times in the same enclosure and may not reflect their ability to put out a fire or keep it out. Consult a qualified fire protection engineer for appropriate design concentrations.

Agent Argon Argonite CEA-308 CEA-410 CO2 FE-13 FE-227 FE-241 FE-25 FE-36 FIC-1311 fm-200 halon INERGEN  SIII * ** ***

Quantity of agent 559 cu.ft. 559 cu.ft.

Initial % 43% ** 43% **

Final % 31.3% 28.6%

Column Pressure 6.4 Pa 2.7 Pa

NFPA retention time 14 minutes 27.2 minutes***

No mixing case results 14 minutes 21 minutes

196 lb. 559 cu.ft. 57.5 lb. 45 lb.

24% * 42.8%** 24% * 9%

5% 31.3% 15% 5.5%

62.1 Pa 8.1 Pa 12.3 Pa 16.4 Pa

23 minutes 13 minutes 15 minutes 14 minutes

8 minutes 14 minutes 11 minutes 10 minutes

45 lb. 31.5 lb. 559 cu.ft. 26.6 lb.

9% 7.5% 43% ** 10%

5.5% 2.5% 28.6% 8.6%

16.4 Pa 11.4 Pa 2.9 Pa 7.9 Pa

14 minutes 32 minutes 26.1 minutes*** 7 minutes

10 minutes 11 minutes 20 minutes 13 minutes

Maximum design is 24 % for occupied spaces to prevent O2 from falling below 16 % per NFPA. Maximum design is 43 % ** for occupied spaces for inert agents to give 12 % O2 per NFPA. Note recalculated time. Initial % Final % Highest initial concentration was used to give longest retention time Lowest concentration used in example to show the longest retention time at end of the retention period for enclosure example. Pc is created by the weight of agent pressing on the floor. As soon as the agent begins to leak out, this pressure is reduced by the pressure drop across the upper leaks and the loss of agent. Time for the concentration to drop from the initial concentration to the final %.

Column Pressure Pc NFPA calculated retention time

Appendix A – Clean Agent Comparison Page 56 of 109

Level 1 Training Manual 8.7 Conclusions The inert agents have longer retention times because their density is close to that of air. This would allow their enclosures to be about twice as leaky to maintain the same concentration. Caution is advised in making them too leaky because then smoke damage from outside the enclosure may become a problem. Some agents have the same retention time while others are higher and lower. The agent’s performance must be checked during the design phase to ensure the enclosure’s dimensions maximize retention time. For example: For descending interface cases, rooms must have the highest possible ceilings and the equipment in the room must be at the lowest possible level. For continual mixing cases, rooms must have the greatest volume and maximum initial versus final concentration to increase retention times. Ideally, the enclosure can be optimized for both situations because it may not be known whether mixing will occur during the retention time or not.

Appendix A – Clean Agent Comparison Page 57 of 109

Level 1 Training Manual

8.8 Agent Comments
Argon Argonite CEA-308 CEA-410 CO2 FE-13 FE-241 FE-25 FE-36 FIC-1311 FM-200, FE-227 Halon Inert clean agent - see INERGEN  - same comments apply Inert clean agent - 50% Argon and 50% Nitrogen - see INERGEN  - same comments apply Not currently in 2001 but slated for next issue - in draft of ISO standard Large range between design and NOAEL - would allow for high design concentrations where continual mixing was desired to provide protection at high levels in an enclosure For unoccupied areas only UL listed to - 40 F and ceilings up to 25 feet - Kidde literature states NOAEL of 30 % - large range between design and NOAEL would allow for high design concentrations where continual mixing was desired For unoccupied areas, cabinets & compartments - useful down to –32 F For unoccupied areas For portable extinguishers Listed in 2001 but no information available Popular replacement for Halon in many halocarbon clean agent applications - puts out fires by removing heat at the molecular level so combustion cannot continue The old standard used Halon as a familiar reference for comparison to other agents Inert clean agent Initial oxygen content 13.1 % for 37.5 % INERGEN and 12 % for 43 % INERGEN Unoccupied areas can have greater concentrations than 52 % This can rise up to 15 % O2 (28.6 % INERGEN) before it loses its fire suppression abilities Inergen has longer retention times due to its density being only slightly heavier than air - even so, the higher concentrations still cause considerable loss from leaks NFPA 2001 requires a door fan test on every installation - in addition, relief venting must be added no matter how leaky the enclosure is, according to Ansul Widely used in Australia

• • • INERGEN  •

SIII

Appendix A – Clean Agent Comparison Page 58 of 109

Level 1 Training Manual

9 Appendix B – NFPA Standard Excerpts
9.1 NFPA 2001 Standard (Year 2004 Edition)
Call NFPA @ 1-800-344-3555 for a complete copy

Standard 5.3 Enclosure 5.3.1 In the design of a total flooding system, the characteristics of the protected enclosure shall be considered. 5.3.2 The area of unclosable openings in the protected enclosure shall be kept to a minimum. 5.3.3 The authority having jurisdiction shall be permitted to require pressurization/ depressurization of the protected enclosure or other tests to assure performance meeting the requirements of this standard (see Appendix C).

Comments

5.3.4 To prevent loss of agent through openings to adjacent hazards or work areas, openings shall be permanently sealed or Extended discharge equipped with automatic closures. Where reasonable recommended where confinement of agent is not practicable, protection shall be openings can’t be sealed. expanded to include the adjacent connected hazards or work areas or additional agent shall be introduced in the protected enclosure using an extended discharge configuration. 5.3.5 Other than the ventilation systems identified in 5.3.5.1 and 5.3.5.3., forced air ventilating systems shall be shut down or closed automatically where their continued operation would adversely affect the performance of the fire extinguishing system or result in propagation of the fire. 5.3.5.1 Completely self contained recirculating ventilation shall not be required to shut down. 5.3.5.2 The volume of the ventilation system and associated ductwork shall be considered as part of the total hazard volume when determining the quantity of agent.

Appendix B – NFPA Standards Excepts Page 59 of 109

Level 1 Training Manual 5.3.6 The protected enclosure shall have the structural strength and integrity to contain the agent discharge. If the developed pressures present a threat to the structural strength of the enclosure, venting shall be provided to prevent excessive pressures. Designers shall consult system manufacturer’s recommended procedures relative to enclosure venting. 5.6* Duration of Protection. It is important that the agent design concentration not only shall be achieved, but also shall be maintained for a sufficient period of time to allow effective emergency action by trained personnel. This is equally important in all classes of fires since a persistent ignition source (e.g. an arc, heat source, oxyacetylene torch, or “deep-seated” fire) can lead to resurgence of the initial event once the clean agent has dissipated. Exception: Ventilation systems necessary to ensure safety are not required to be shut down upon activation of the fire suppression system. An extended agent discharge shall be provided to maintain the design concentration for the required duration of protection. 6.1.1 Inspection & Tests At least annually, all systems shall be thoroughly inspected and tested for proper operation by competent personnel. Discharge tests are not required. 6.4 Enclosure Inspection Other than as identified in 6.4.1, the enclosure protected by the clean agent shall be thoroughly inspected at least every 12 months to determine if penetrations or other changes have occurred that could adversely affect agent leakage or change volume of hazard, or both. Where the inspection indicates conditions that could result in the inability to maintain the clean agent concentration, the conditions shall be corrected. If the uncertainty still exists, the enclosures shall be retested for integrity in accordance with 6.7.2.3.

Venting shall be provided if there is any risk of overpressure during the discharge.

This time is usually 10 minutes but this should be considered more carefully. Take into account response time for fire dept. or other personnel; the mass of the fuel; the extent of “deepseated” potential.

There will almost always be cause for a re-test since enclosures always become leakier with time. It is far easier to set up the door fan and retest than to do a detailed inspection to find holes in the enclosures.

6.5.3 Any penetrations made through the enclosure protected by the Door fan test will show clean agent shall be sealed immediately. The method of enclosure is sealed. sealing shall restore the original fire resistance rating of the enclosure Appendix B – NFPA Standards Excepts Page 60 of 109

Level 1 Training Manual 6.6.1 All persons who might be expected to inspect, test, maintain, or operate fire extinguishing systems shall be thoroughly trained and kept thoroughly trained in the functions they are expected to perform. 6.7.2.2.10* If a discharge test is to be conducted, containers for the agent to be used shall be weighed before and after discharge. Fill weight of container shall be verified by weighing or other approved methods. For inert gas clean agents, container pressure shall be recorded before and after discharge. 6.7.2.3* Review Enclosure Integrity. All total flooding systems shall have the enclosure examined and tested to locate and than effectively seal any significant air leaks that could result in a failure of the enclosure to hold the specified agent concentration level for the specified holding period. The currently preferred method is using a blower door fan unit and smoke pencil. If quantitative results are recorded, these could be useful for comparison at future tests (For guidance, see Appendix C). All of Appendix C This clumsily worded section is usually taken to mean that all clean agent systems must have a door fan test. The quantitative results are used when enclosure is re-tested.

AHJ may ask to see certificate of course completion on door for testing.

Appendix B – NFPA Standards Excepts Page 61 of 109

Level 1 Training Manual

9.2 NFPA 2001 Standard (Year 2000 Edition) Standard 3-3 Enclosure 3-3.1 In the design of a total flooding system, the characteristics of the protected enclosure shall be considered. 3-3.2 The area of unclosable openings in the protected enclosure shall be kept to a minimum. 3-3.3 The authority having jurisdiction shall be permitted to require pressurization/ depressurization of the protected enclosure or other tests to assure performance meeting the requirements of this standard (see Appendix C).

Comments

3-3.4 To prevent loss of agent through openings to adjacent hazards or work areas, openings shall be permanently sealed or Extended discharge equipped with automatic closures. Where reasonable recommended where confinement of agent is not practicable, protection shall be openings can’t be sealed. expanded to include the adjacent connected hazards or work areas or additional agent shall be introduced in the protected enclosure using an extended discharge configuration. 3-3.5 Forced-air ventilating systems shall be shut down or closed automatically where their continued operation would adversely affect the performance of the fire extinguishment agent system or result in propagation of the fire. Completely self-containment recirculating ventilation systems are not required to shut down. The volume of the system and associated ductwork shall be considered as part of the total hazard volume when determining quantity of agent. 3-3.6 Venting shall be provided if there is any risk of overpressure during the discharge.

Appendix B – NFPA Standards Excepts Page 62 of 109

Level 1 Training Manual 3-6* Duration of Protection. It is important that the agent design concentration not only shall be achieved, but also shall be maintained for a sufficient period of time to allow effective emergency action by trained personnel. This is equally important in all classes of fires since a persistent ignition source (e.g. an arc, heat source, oxyacetylene torch, or “deep-seated” fire) can lead to resurgence of the initial event once the clean agent has dissipated. Exception: Ventilation systems necessary to ensure safety are not required to be shut down upon activation of the fire suppression system. An extended agent discharge shall be provided to maintain the design concentration for the required duration of protection. 4-1.1 At least annually, all systems shall be thoroughly inspected and tested for proper operation by competent personnel. Discharge tests are not required. 4-4 Enclosure Inspection. At least every 12 months, the enclosure protected by the clean agent shall be thoroughly inspected to determine if penetrations or other changes have occurred that could adversely affect agent leakage or indicates conditions that could result in inability to maintain the clean agent concentration, they shall be corrected. If uncertainty still exists, the enclosures shall be tested for integrity in accordance with 4-7.2.3. There will almost always be cause for a re-test since enclosures always become leakier with time. It is far easier to set up the door fan and retest than to do a detailed inspection to find holes in the enclosures.

This time is usually 10 minutes but this should be considered more carefully. Take into account response time for fire dept. or other personnel; the mass of the fuel; the extent of “deepseated” potential.

4-5.3 Any penetrations made through the enclosure protected by the Door fan test will show clean agent shall be sealed immediately. The method of enclosure is sealed. sealing shall restore the original fire resistance rating of the enclosure 4-6.1 All persons who might be expected to inspect, test, maintain, or operate fire extinguishing systems shall be thoroughly trained and kept thoroughly trained in the functions they are expected to perform. AHJ may ask to see certificate of course completion on door for testing.

Appendix B – NFPA Standards Excepts Page 63 of 109

Level 1 Training Manual 4-7.2.2.10* If a discharge test is to be conducted, containers for the agent to be used shall be weighed before and after discharge. Fill weight of container shall be verified by weighing or other approved methods. For inert gas clean agents, container pressure shall be recorded before and after discharge. 4-7.2.3* Review Enclosure Integrity. All total flooding systems shall have the enclosure examined and tested to locate and than effectively seal any significant air leaks that could result in a failure of the enclosure to hold the specified agent concentration level for the specified holding period. The currently preferred method is using a blower door fan unit and smoke pencil. If quantitative results are recorded, these could be useful for comparison at future tests (For guidance, see Appendix B (they meant C)). All of Appendix C This clumsily worded section is usually taken to mean that all clean agent systems must have a door fan test. The quantitative results are used when enclosure is re-tested.

Appendix B – NFPA Standards Excepts Page 64 of 109

Level 1 Training Manual

9.3 NFPA 2001 Standard (Year 1996 Edition) Standard 1-4.2.7 Where clean agent systems are used, a fixed enclosure shall be provided about the hazard that is adequate to enable the specified concentration to be achieved and maintained for the specified period of time.

Comments Large holes and/or duct leaks can cause “lumps” of air to get pulled in upon discharge. Local reduction in agent concentration may compromise protection.

1-5.1.2.1 Unnecessary exposure to all halocarbon clean agents and their decomposition products shall be avoided. Halocarbon agents for whom the design concentration is equal to or less than the NOAEL shall be permitted for use in normally occupied areas. Halocarbon agents for which the design concentration is greater than the NOAEL shall not be permitted for use in normally occupied areas. 1-5.1.2.2 To maintain oxygen concentrations above 16 percent (sea level equivalent), the point at which onset of impaired personnel function occurs, no halocarbon fire extinguishing agents of concentration greater than 24 percent addressed in this standard shall be used in a normally occupied area. 3-3.2 The area of uncloseable openings shall be kept to a minimum. The authority having jurisdiction can require pressurization/depressurization or other tests to assure proper performance as defined by this standard. 3-3.3 To prevent loss of agent through openings to adjacent hazards or work areas, openings shall be permanently sealed or equipped with automatic closures. Where reasonable confinement of agent is not practicable, protection shall be extended to include the adjacent connected hazards or work areas. INERGEN  may be the exception, where oxygen may fall to 12% but the increased CO2 will increase the breathing rate to compensate.

Appendix B – NFPA Standards Excepts Page 65 of 109

Level 1 Training Manual 3-3.4 Forced-air ventilating systems shall be shut down or closed If not shut down, additional automatically where their continued operation would adversely agent must be provided to affect the performance of the fire extinguishment agent system compensate for losses. or result in propagation of the fire. Completely selfcontainment recirculating ventilation systems are not required to shut down. The volume of the system and associated ductwork shall be considered as part of the total hazard volume when determining agent quantities. 3-7 Duration of Protection. It is important that the agent design concentration not only shall be achieved, but also shall be maintained for a sufficient period of time to allow effective emergency action by trained personnel. This is equally important in all classes of fires since a persistent ignition source (e.g. an arc, heat source, oxyacetylene torch, or “deepseated” fire) can lead to resurgence of the initial event once the clean agent has dissipated. 4-1.1 At least annually, all systems shall be thoroughly inspected and tested for proper operation by competent personnel. Discharge tests are not required. 4-4 Enclosure Inspection. At least every 12 months, the enclosure protected by the clean agent shall be thoroughly inspected to determine if penetrations or other changes have occurred that could adversely affect agent leakage or indicates conditions that could result in inability to maintain the clean agent concentration, they shall be corrected. If uncertainty still exists, the enclosures shall be tested for integrity in accordance with 4-7.2.3. There will almost always be cause for a re-test since enclosures always become leakier with time. It is far easier to setup the door fan and retest than to do a detailed inspection to find holes in the enclosures. This time is usually 10 minutes but this should be considered more carefully. Take into account response time for fire dept. or other personal; the mass of the fuel; the extent of “deep- seated” potential.

4-5.3 Any penetrations made through the enclosure protected by the Door fan test will show clean agent shall be sealed immediately. The method of enclosure is sealed. sealing shall restore the original fire resistance rating of the enclosure. 4-6.1 All persons who might be expected to inspect, test, maintain, or operate fire extinguishing systems shall be thoroughly trained and kept thoroughly trained in the functions they are expected to perform. Appendix B – NFPA Standards Excepts Page 66 of 109 AHJ may ask to see certificate of course completion on door for testing.

Level 1 Training Manual 4-7.2.2.10 A discharge test is generally not recommended; however, if a discharge test is to be conducted, containers for the agent to be used shall be weighed before and after discharge. Fill weight of container shall be verified by weighing or other approved methods. For inert gas clean agents, container pressure shall be recorded before and after discharge. 4-7.2.3 Review Enclosure Integrity. All total flooding systems shall have the enclosure examined and tested to locate and than effectively seal any significant air leaks that could result in a failure of the enclosure to hold the specified agent concentration level for the specified holding period. The currently preferred method is using a blower door fan unit and smoke pencil. If quantitative results are recorded, these could be useful for comparison at future tests. A-1-6 Do not perform unnecessary discharge testing A-4-7.2.3 If the authority having jurisdiction wants to quantify the enclosure’s leakage and predicted retention time, Appendix B of NFPA 12A, Standard on Halon 1301 Fire Extinguishing Systems, may be used. Adjustment to the existing formulas must be made to account for differences in gas density between Halon 1301 and the proposed alternate extinguishing agent. Specifically, Equation 8 in paragraph B-2.7.1.4 of NFPA 12A must be modified by substituting the alternate agent’s gas density in (Kg/m3) for the existing value of 6.283, which is the value for Halon 1301. See Appendix B of this Standard. All of Appendix B B-1.1.4 This procedure should not be considered to be an exact model of a discharge test. The complexity of this procedure should not obscure the fact that most failures to hold concentration are due to the leaks in the lower surfaces of the enclosure, but the door fan does not differentiate between upper and lower leaks. The door fan provides a worst-case leakage estimate that is very useful for enclosures with complex hidden leaks, but it will generally require more sealing than is necessary to pass a discharge test. The formula adjustment is made in the Retrotec software using values supplied by NFPA This clumsily worded section is usually taken to mean that all clean agent systems must have a door fan test. The quantitative results are used when the enclosure is re-tested.

Appendix B – NFPA Standards Excepts Page 67 of 109

Level 1 Training Manual 9.4 NFPA 12A Halon 3-6 Altitude Adjustment. The design quantity of Halon 1301 shall be adjusted to compensate for altitudes of more than 3000 ft. (1000 m) above or below sea level and pressures that vary by 10 percent above or below standard sea level pressure (29.92 in. Hg at 70 degrees F). The Halon 1301 quantity shall be corrected by multiplying the quantity determined in 3-5.1 and 3-5.2 by the ratio of average ambient enclosure pressure to standard sea level pressure.

The altitude adjustment in A-36 of 2001 is used for all agents to adjust concentration for altitude.

3-7.1.2 Discharge Time. The agent discharge shall be substantially completed in a normal 10 seconds or as otherwise required by the authority having jurisdiction. This period shall be measured as the interval between the first appearance of liquid at the nozzle and the time when the discharge becomes predominantly gaseous. This point is distinguished by a marked change in both the sound and the appearance of the discharge. 4-7.2.2 Enclosure Integrity Acceptance. All total flooding systems shall have the enclosure examined and tested to locate and then effectively seal any significant air leaks that could result in a failure of the enclosure to hold the specified Halon 1301 concentration level for the specified holding period. The currently preferred method is using a blower door fan unit and smoke pencil. If quantitative results are recorded, these could be useful for comparison at future tests on the same room with a door fan. A-3.3.3 The design of total flooding Halon 1301 systems only beneath the raised floor of EDP facilities when the occupied space above the raised floor is not similarly protected by a total flooding Halon 1301 system does not meet the intent of this standard. Such a design does not comply with the definition of a total flooding system or with this chapter. A-3-5.2 Leakage of Halon 1301 through Enclosure Openings. Halon 1301 discharged into an enclosure for total flooding will result in an enclosure for total flooding will result in an air/agent mixture that has a higher specific gravity than the air surrounding the enclosure. Therefore, any opening in the walls of the enclosure will allow the heavier air/agent mixture to flow out of the enclosure, being replaced with lighter outside air flowing into the enclosure through the same opening. The rate at which agent is lost through openings will depend on the height and width of the opening, the location of the opening in the wall, and the concentration of agent in the enclosure. Fresh air entering the enclosure will collect toward the top, forming an interface between the air/agent mixture and fresh air. As leakage proceeds, the interface will move toward the bottom of the opening. The space below the interface will contain essentially the original extinguishing concentration of agent, whereas the upper space will be completely unprotected. The rate at which the interface moves downward increases as concentrations Appendix B – NFPA Standards Excepts Page 68 of 109

Level 1 Training Manual of agent increase, so that simply injecting an overdose of agent initially will not provide an extended period of protection. All of Appendix B 9.5 NFPA 12 for CO2 2-6.1 General. The venting of flammable vapors and pressure buildup from the discharge of quantities of carbon dioxide into closed spaces shall be considered. Venting of flammable vapors is covered in 2-2.1.4. The pressure venting consideration involves such variables as enclosure strength and injection rate. 2-6.2 Pressure Relief Venting. Porosity and leakages such as at doors, windows, and dampers, though not readily apparent or easily calculated, have been found to provide sufficient relief for the normal carbon dioxide flooding systems without need for additional venting. Record storage rooms, refrigerated spaces, and ductwork have also been found to need no additional venting when tested under their average system conditions. 2-6.2.1 For very tight enclosures, the area necessary for free venting shall be calculated form the following formula. Assuming the expansion of carbon dioxide to be 9 ft.3/lb (0.56 m.3/kg) will give satisfactory results. X= Q Where X = Free venting area in in. Q = Calculated carbon dioxide flow rate in lbs./min. P = Allowable strength of enclosure in lbs/ft2 For SI Units XM = 23.9 QM XM = Free venting area mm2 QM = Calculated carbon dioxide flow rate in kg/min. PM = Allowable strength of enclosure bars, gauge. 2-6.2.2 In many instances, particularly when hazardous materials are involved, relief openings are already provided for explosion venting. These and other available openings often provide adequate venting.
PM
2

1.3

P

Appendix B – NFPA Standards Excepts Page 69 of 109

Level 1 Training Manual

10 Appendix C – Sample Enclosure Integrity Test Specification
The first section of this guide discusses issues to be addressed by the designer/specifier, on behalf of the client, at the conceptual stages of the planned project. Further sections provide sample performance and prescriptive specifications, which specifically impact enclosure acceptance, particularly when using the new NFPA 2001 Enclosure Integrity Test or ISO 14520. These sections are to be integrated into other readily available standard Clean Agent specifications, which cover other areas such as References, System Description, Submittals, Qualifications, Warranty and Service, Products and Piping. The general objective of this document is to ensure that new Clean Agent protected enclosures are built so they can be accepted using the 12A & 2001 Enclosure Integrity Test. It is important to note however that most of the recommendations will also help ensure trouble-free acceptance of enclosures using the conventional discharge test. In addition, many of these recommendations can be applied to CO2. This material must be carefully edited for each installation to ensure that the completed specification is appropriate to the particular installation, to avoid conflicting requirements, to ensure consistent numbering, and to ensure that the particular specification is placed in the appropriate contractor's contract. Items shown in bold are sample specifications. Items shown in regular type are either instructions or background information for the specification writer. Items in parentheses are generally options to be included or not depending on system requirements. This document has been written assuming that year 2000 NFPA 2001 CLEAN AGENT STANDARD is the principal governing code or standard. If this is not the case, the appropriate substitutions should be made. This document is not to be considered a formal or informal interpretation of NFPA 2001. If any aspect of NFPA 2001 is not clear to the reader, he is advised to contact NFPA directly for a formal interpretation. This document reflects only the views and opinions of Retrotec based on our experience. This document itself carries no authority. Retrotec makes no warranty either expressed or implied that these sample specifications are appropriate for use in their current form. While many of the examples in this document refer to computer facilities, this should not be taken to mean that the guidelines are not appropriate for other types of hazards. 10.1 General Enclosure Design Guidelines If Clean Agent has been determined to be the most appropriate extinguishing Agent for the installation, a number of items should be reviewed prior to designing the enclosure and system and developing the specifications. Appendix C – Sample Enclosure Integrity Test Specification Page 70 of 109

Level 1 Training Manual

Slab To Slab Walls or Solid Ceiling In order to retain the Clean Agent for the full retention time, the enclosure must be tightly built. In order to be accepted using the Enclosure Integrity Procedure, each zone's enclosure must either have perimeter walls which extend from slab to slab as per NFPA 75, or must have a solid (drywall) ceiling. Interior walls between rooms within one Clean Agent zone should stop at the suspended ceiling level, so that the Clean Agent protected zone has a common ceiling plenum. In addition it must be possible to "pneumatically" connect all the rooms within the zone, preferably with permanently installed transfer air openings or an uncarpeted common subfloor. Past experience has shown that slab-to-slab construction is not generally necessary for an enclosure to retain the Agent. It is however preferred, if the enclosure is to be accepted using NFPA 2001 Appendix C. Slab-to-slab walls or a solid ceiling are not specifically required by NFPA 2001 although they are in NFPA 75 and by many local authorities. An extremely thorough visual inspection using a smoke pencil and door fan may satisfy some authorities. Its non-quantitative nature is a drawback however. If the client originally intended to install a lay-in tile ceiling with a common ceiling plenum to the rest of the building, the benefits of either slab-to-slab or solid ceiling construction should be described and they are: 1. Greater overall fire protection of the hazard will be obtained through having at least a one-hour fire rated separation surrounding the enclosure. Compartmentation is considered one of the key first elements in effective fire protection. 2. Greater environmental control (humidity, dust and temperature) and lower ongoing maintenance costs will be provided by a tight enclosure. 3. Greater protection from smoke contamination originating outside the hazard is obtained with a tight enclosure. 4. The increased cost of providing additional drywall and dampers will be offset by lower maintenance costs, possibly lower initial acceptance test costs if a discharge test is not performed and also reduced costs to maintain an acceptably tight enclosure over time.

Appendix C – Sample Enclosure Integrity Test Specification Page 71 of 109

Level 1 Training Manual 5. Authorities Having Jurisdiction now require periodic Enclosure Integrity Testing to ensure continued performance. Slab-to-slab walls make the enclosure easier to "re-accept" it quantitatively in the future. Avoidance of Attached Volumes The Enclosure Integrity Test requires that an unrestricted return path be present during the test between the door fan and each leak in the enclosure. This is generally obtained by opening doors (or ceiling tiles) between rooms and spaces adjoining the Clean Agent protected enclosure while the test is being conducted. The resulting surrounding network of rooms and corridors is known as the relief path. If a doorway or ceiling space is not available to pneumatically connect an adjoining room to the rest of the relief path during the fan test, the leakage from the Clean Agent protected room into that room may not be accurately measured. This adjoining room is then referred to as an attached volume. Attached volumes can be avoided by ensuring that doors, hatches, or common ceiling plenums exist between all adjoining rooms and corridors outside the protected enclosure. Note that attached volumes should be avoided in any case as they may indicate a restriction of egress options from such rooms in the case of an emergency. If leakage from the Clean Agent protected enclosure to an attached volume is less than half the leakage from the attached volume into the relief path, or if the barrier between the two spaces has only a small portion of the enclosures total leakage, little effect will be made on the test's accuracy. Unfortunately, this is a subjective evaluation and difficult to administer in the field. Do not design a common ceiling plenum between the attached volume and the protected enclosure. A slab-to-slab wall should be installed between the protected enclosure and all adjoining rooms and corridors. Do not design an unprotected room to be built within the envelope of the protected enclosure. Leakage into an unprotected space completely within the protected enclosure will not be measured by the Enclosure Integrity Test. Such spaces should be included in the Clean Agent protected space. Penetration Planning Achieving and maintaining a high degree of tightness is facilitated by having the location and design of certain penetrations, specifically for cables, planned in advance. The installation of round pipe sleeves or other engineered sealable openings is recommended. Sufficient extra capacity should be installed to handle expected future expansion. Sealing openings Appendix C – Sample Enclosure Integrity Test Specification Page 72 of 109

Level 1 Training Manual between cables within bundles is a very common and difficult problem to solve once all cables are in place. Document Passageways Many data processing centers include a "pigeon hole" system to control the flow of documents into and out of the room. These generally have glass or acrylic doors or flaps. If such a system is installed, it is recommended that it not penetrate the enclosure perimeter walls, as excessive leakage will occur. A Clean Agent protected vestibule should be included in the room design or the design should include an electrical or mechanical means to close such an opening. HVAC Dampers All ducts leading into or out of the space must be mechanically dampered, even if the air handler serving them will be shut down and the ducts terminate at ceiling level. Dampers should be smoke rated. "Un-closeable" Openings Un-closeable openings must be avoided. Very few such openings are actually un-closeable. For example, most conveyor belt systems can be designed so that the belt is divided at the opening and spring loaded "guillotine" type dampers can be installed to seal off the opening upon discharge. Other "pass-throughs" can be similarly sealed. A perfect seal may be impossible to obtain, but a reasonably tight seal is usually possible. The key issue is to identify and accommodate such openings at the design stage, not after a room fails an Enclosure Integrity or Discharge test. Location of Dedicated HVAC units Whenever possible, dedicated HVAC units for room cooling, dehumidification and dust control should be placed entirely within the protected enclosure. Duct and access panel leakage often cause excessive Clean Agent loss. If this is not possible, specify that the additional measures needed to ensure a tight system be the responsibility of the HVAC contractor. Minimum Protected Height Even if a Clean Agent protected enclosure is designed and built to be as tight as possible, a certain degree of leakage must be expected to occur. The leakage mechanism is somewhat as follows. During the retention period, the agent/air mixture, being heavier than air, will generally leak out lower openings. Air will enter through openings high in the room at the same rate to replace it. If air-moving devices in the room are shut down, this incoming air tends to collect at the top of the room. The upper level of the Clean Agent mixture descends over time. This boundary layer between the original agent/air mixture and the infiltrated air is known as the descending interface. Appendix C – Sample Enclosure Integrity Test Specification Page 73 of 109

Level 1 Training Manual However, if any air moving equipment is left on during the retention period (blowers, air conditioning units or UPS equipment), the incoming air becomes completely mixed with the original agent/air mixture. This causes the average concentration throughout the room to decay. This phenomenon is known as mechanical mixing. If a descending interface forms, the allowable height to which it can descend in 10 minutes is a crucial factor. This minimum protected height is usually where the upper probe would have been placed during a discharge acceptance test. The minimum protected height is best defined as the highest combustible item in the room. Design the room and its equipment (cable trays are the most common problem) so that all combustibles are kept below the 75% level (measured from the floor slab). The 75% level is an NFPA 2001 guideline, and allows for a reasonable amount of Clean Agent leakage (up to 25% of the room volume) while not severely restricting the equipment design. If the minimum protected height is be set above 85% of the protected enclosure height, more reliable and extended protection can be provided near ceiling level by intentionally designing in continuous mechanical mixing. If his is done the minimum protected height is irrelevant, as the same concentration will be present everywhere in the enclosure. A discussion of the mechanical mixing option is necessary in this Enclosure Integrity Testing document even though it enters into system design, as it significantly impacts on the ability to accept the enclosure. Mechanical mixing is most often achieved by running the in-room air conditioners upon discharge. There are however three drawbacks to this approach: the AC unit may be what is on fire; sub-floor AC units will invariably accelerate the Clean Agent loss as they push the mixture out any leaks in the sub-floor (or ductwork outside the protected enclosure); and the mixture may inadvertently be blown into unprotected ceiling voids. Mechanical mixing can also be achieved by installing a separate air handling system, which activates upon discharge to continuously circulate the Clean Agent mixture up from floor level to ceiling level. Better control of the distribution can be obtained. A separate system is not ideal either, as it must be periodically checked and maintained in order to ensure it is still operational (much the same as for a Clean Agent purge system). The final selection is up to the designer. The "descending interface or mechanical mixing" decision must be made early in the design process as it may affect the determination of required Agent quantity, and, therefore, the required cylinders and piping layout. If mechanical mixing will occur, and a 6% initial concentration is used, only 16.6% of the room volume could be lost before a 5% concentration is reached. If mechanical mixing will occur, longer retention times will be achieved using a higher initial concentration. For example, if a 6.7% initial concentration is Appendix C – Sample Enclosure Integrity Test Specification Page 74 of 109

Level 1 Training Manual achieved, up to 25% of the room volume can leak out before 5% is reached. This may be an appropriate design choice for either a small room or one, which has features, which make achieving an airtight seal difficult. An extended discharge design is another option. Small rooms (say up to approximately 5,000 cubic feet) have historically been the most difficult to pass using a discharge test. There appears to be two reasons for this. One is that Clean Agent is more likely to be lost during the initial discharge, especially if there is an unprotected ceiling void above. This appears to be reduced if a "soft" discharge is used. Contact Clean Agent equipment manufacturers for more guidance. The predominant reason appears to be because small rooms have much less favorable surface to volume ratios. For example, a 10,000 cubic foot room has ten times the volume of a 1,000 cubic foot room, but has only three times the wall area. Relatively speaking, the small room has to be much tighter to retain the agent. As the Room Integrity Test is even more stringent than the discharge test, this can make small rooms difficult to accept if they aren't practically airtight. Summary If all air moving equipment will be shut down in the event of a fire, the Minimum Protected Height (e.g. 75% of room height) and Minimum Initial Concentration (e.g. 6%) should be specified in the bid request documents. Refer to Section 3.00 for a discussion of why it is necessary to specify a minimum initial concentration to design for if an Enclosure Integrity Test is to be used for acceptance. It is recommended that the Minimum Protected Height be no higher than 75% of the room height, especially if the enclosure volume is less than 5,000 cubic feet. If a mechanical mixing design approach is taken, the HVAC contractor and Clean Agent installer must be informed of the mechanical and control requirements. It is recommended that a higher Minimum Initial Concentration (e.g. 6.75%) be specified if the enclosure volume is less than 5,000 cubic feet. Note that the new Year 2000 NFPA 2001 Enclosure Integrity Procedure models the "mechanically mixed" retention period. 10.2 Enclosure Integrity Specifications (For the General Contractor) On new installations, it is recommended that the General Contractor (GC), if one is present, be made responsible for overall room tightness. The GC in turn would then require that all his subcontractors perform the necessary sealing, which relates to their work. Any work being done on the installation by second level contractors (e.g. cable pullers) not operating under the GC must also be subjected to this requirement under their contracts. If the Clean Appendix C – Sample Enclosure Integrity Test Specification Page 75 of 109

Level 1 Training Manual Agent system is being installed as a retrofit, one contractor must be made responsible for sealing existing holes. If no building contractor is involved in the retrofit, the Clean Agent installer may be able to arrange for this service. The prescriptive specifications give guidance on what must be sealed, while the performance specification determines whether the job was done right. In order to pass the Enclosure Integrity Test, the contractor may have to seal items, which are not specifically described, in the prescriptive specifications. Item 2.02C covers just about every possibility. Enclosure Integrity Performance Specification Enclosure leakage shall be eliminated to at least the degree necessary to enable the Clean Agent protected enclosure to pass a test conducted in accordance with the Year 2000 NFPA 2001 Enclosure Integrity Procedure. Variables of interest are listed in Article 6, APPROVAL/ACCEPTANCE OF ENCLOSURE INTEGRITY. It is possible to calculate in advance using NFPA 2001 Appendix C what the maximum allowable Equivalent Leakage Area would be for the enclosure. If this is done, the performance specification could be even more specific. Enclosure Integrity Prescriptive Specifications The following items cover enclosure leakage in a general fashion, and should be placed in the General Contractor's specification. He should then repeat those appropriate to specific subcontractors in their specifications. If the client or AHJ requires that the materials and techniques used must produce a one or two-hour fire rated enclosure, this must be specified. Because historically the walls and roof of unprotected ceiling voids above suspended ceilings have not had to be well sealed to retain agent, existing building practice, if retained, will produce enclosures where large leakage areas will be measured, resulting in unacceptably low predicted retention times. It is recommended that where possible the walls and roofs of unprotected ceiling voids be sealed as tightly as the protected enclosure below. If this is not possible or practical (in a retrofit for example), it is generally possible to accept the enclosure using the B-2.6.2 Suspended Ceiling Leakage Neutralization Method. It is recommended however that every attempt be made to seal the ceiling void first. 1. The perimeter walls of the protected enclosure shall extend from the structural floor to the structural floor above, or the roof. Alternately: The ceiling of the enclosure shall be ( ) inch drywall (plasterboard), mudded, taped and painted per Article ( ). Access panels shall be provided as indicated on the plans (# ).

Appendix C – Sample Enclosure Integrity Test Specification Page 76 of 109

Level 1 Training Manual Possible addition: The ceiling system shall be constructed with an upper deck in order to provide a walk on surface for servicing above ceiling utilities. (Specify construction details if desired.) 2. Where an under floor space continues out of the Clean Agent protected area into adjoining rooms, airtight partitions shall be installed under the floor directly under above-floor border partitions. These partitions shall be caulked top and bottom. If a removable floor tile extends under a doorway over such a partition, it shall either be: permanently sealed in place; installed with a flexible seal between it and the wall below; or the tile shall be discontinued at the doorway with a permanent airtight ledge created up to which the floor tiles abut. If adjoining rooms share the same under floor air handlers, then the partitions shall have dampers installed of the same type as required for ductwork. 3. All holes, cracks, or penetrations leading into or out of the protected area shall be sealed. Pipe chases and wire troughs shall be sealed around both the outside and inside at a point where they pass through the envelope of the protected zone. All walls shall be caulked around the inside perimeter of the room where the walls rest on the floor slab and where the walls intersect the ceiling slab or roof above. 4. Porous block walls shall be sealed slab-to-slab to prevent gas from passing through the block. Multiple coats of paint may be required. 5. All doors shall have door sweeps or drop seals on the bottoms, weather stripping around the jambs, latching mechanisms and door closer hardware. In addition, double doors shall have a weather-stripped astragal to prevent leakage between doors and a coordinator to assure proper sequence of closure. 6. Windows shall have solid weather-stripping around all joints. Glass to frame and frame to wall joins shall be sealed. 7. All floor drains shall have traps designed to have water or other compatible liquid in them at all times. 8. All unused and out-of-service ductwork leading into or from a protected area shall be permanently sealed off (air tight) with metal plates caulked and screwed in place at the point where they breach the envelope of the protected zone. 9. All ceiling tiles shall have a weight of at least (xx) pounds per square foot. Lightweight vinyl coated acoustic tiles shall not be used. The possibility of ceiling tiles being displaced during a discharge should be addressed at the design stage. Possible options include tile clipping, nozzle deflectors, lowering the nozzles a Appendix C – Sample Enclosure Integrity Test Specification Page 77 of 109

Level 1 Training Manual certain distance from the ceiling and ensuring proper nozzle location. Contact Clean Agent equipment manufacturers for guidance. 10.3 Clean Agent System Specifications (For the Clean Agent Contractor) This section covers only the issue relating to the Clean Agent system design, which has an impact on the Enclosure Integrity Test. A complete specification should cover the appropriate Year 2000 NFPA 2001 articles and features of particular interest to the client. Note: Examples use Halon because it is generic - substitute the proper values using Halon as a guide. Past practice in the industry has been to specify the minimum concentration (usually 5% for Halon) which has to still be in the room at the end of the required retention period (usually 10 minutes). The typical Clean Agent installer, knowing that some enclosure leakage is usually present, prudently designs for l0% to 20% more Agent on the initial discharge. Experience has shown this to usually be sufficient to deal with either the room volume lost if mechanical mixing takes place, or the dilution of the upper layer of the Clean Agent column if a descending interface forms. The descending interface's dilution is caused by the infiltrating air, which enters high in the room. After l0 minutes, the top l0% or so of the Clean Agent column usually contains only half its initial concentration. The vast majority of "5%" systems are, therefore, installed to provide between 5.5% and 6% initially. While a small amount (10% to 20%) of extra Agent is generally needed, more is rarely better (if mechanical mixing doesn't occur). The greater the initial concentration, the denser the Clean Agent mixture becomes, the more the column of Clean Agent weighs, and the faster the descending interface will drop. If a room is leaky the answer is generally to make the enclosure tighter, not to add more agent. The following from NFPA 2001 Appendix A confirms this: "The rate at which the interface moves downward increases as concentrations of Agent increase, so that simply injecting an overdose of Agent initially will not provide an extended period of protection." It is important to note however that this is only true if a descending interface forms during the retention period. If mechanical mixing occurs, more Agent will extend the retention period. The essential point to be made is this: If a discharge concentration test will not be performed, it is necessary to specify the initial Clean Agent concentration required (usually 6% for Halon) that would normally ensure the desired minimum concentration is maintained (usually 5% for Halon). Otherwise, the marketplace will compel the Clean Agent installer to Appendix C – Sample Enclosure Integrity Test Specification Page 78 of 109

Level 1 Training Manual install only enough gas to reach 5%. This is a distinct departure from past practice, which has generally been to only specify the minimum. If mechanical mixing will occur in a small room, it is prudent to specify an even higher initial concentration (e.g. 6.5% to 6.75% to retain 5% for Halon). These ratios can be adjusted accordingly if the client requires more than the most commonly required minimum of 5%. The system shall be designed and installed to provide a (6) % halon concentration throughout the protected enclosure upon discharge, as calculated in NFPA 2001. The protected enclosure extends from the floor slab to (the slab above) (the suspended ceiling). The following rooms are considered to constitute the Clean Agent protected zone: ( ). 10.4 HVAC Specifications (For the Mechanical Contractor) Ductwork Ductwork in service with the building air handling unit shall have gasketed low leak agent/smoke type dampers with flexible seals (option: conforming to UL-555S "Standard for Leakage Rated Dampers For Use in Smoke Control Systems", Class I leakage rated). Rigid metal-to-metal blade seals shall not be used. Dampers shall be spring-loaded or motor-operated to provide near airtight shut-off. (Option: The dampers shall be of the spring close, motor open type.) The dampers shall be installed as close as possible to the duct's point of entry into the room. All duct joints between the damper and the duct entry point shall be sealed. The gap between the damper frame and the duct wall shall be sealed. A minimum 6" square access panel shall be installed to permit internal inspection of the damper. Alterations to air conditioning, heating, ventilating ductwork and related equipment shall be in accordance with NFPA 90A, Standard for the Installation of Air Conditioning and Ventilating Systems, or NFPA 90B, Standard for the Installation of Warm Air Heating and Air Conditioning Systems, as applicable. It is recommended that whenever possible, any in-room air conditioning units be shut down upon discharge to reduce the possibility that they will expel the mixture from the sub-floor. Ideally, the Clean Agent protected enclosure will be a "dead" room from a static pressure standpoint by the time the Clean Agent discharges. If the dampers are truly tight, and the in-room air conditioning units are shut down, close to zero pressure is usually achieved. Occasionally, however, significant imbalances exist in the building HVAC system, which could Appendix C – Sample Enclosure Integrity Test Specification Page 79 of 109

Level 1 Training Manual increase the leakage of Clean Agent from the enclosure. If a significant static pressure is uncovered during the Enclosure Integrity Test which is not solved by improving damper seals or sealing leaks, it may prove to be necessary to have that zone of the building's air handlers shut down in addition to closing the dampers. 10.5 Approval/Acceptance of Clean Agent System The following article covers only the acceptance of the Clean Agent system, which is the Clean Agent installer's responsibility. Adequate enclosure integrity is confirmed in section 6. Historically, the vast majority of discharge test failures have been caused by lack of enclosure integrity. Nonetheless, if a discharge test is not being carried out, it is essential that other aspects of the system installation be verified and tested per NFPA 2001. 1. The contractor shall carry out the acceptance tests described in NFPA 2001 year 2000 edition, section 4-7 in the presence of the AHJ or its representative. 2. The contractor shall provide a test report. After the tests are completed and the system has been accepted, the system shall be brought to full operating condition. 10.6 Approval/Acceptance of Enclosure Integrity In most instances, the Clean Agent contractor is required to provide the Enclosure Integrity Test, although it is possible to separate it from the overall contract and obtain bids from other parties. The following wording assumes that the Clean Agent contractor is providing the test, and these articles are placed in his contract. It is important to note that while the Clean Agent contractor is often responsible for providing the Enclosure Integrity Test, he should not be responsible for the sealing unless very specifically stated in his contract. This usually is only possible on some retrofits. Article 11.6.1 may be used if the enclosure will have a raised floor and/or a suspended ceiling, and if the specifier wants to have the Clean Agent contractor involved in inspecting the General Contractor's sealing work. This task could be done with the inspecting authority. The intent is to confirm that all possible leaks have been sealed at the earliest and easiest stage of construction. 1. Prior to the installation of the (raised floor) (and) (suspended ceiling) the Clean Agent contractor shall depressurize the enclosure to at least -5 pa (-.02" w.c.) with a door fan unit and inspect the enclosure using a smoke pencil. The inspection shall be done in the presence of the owner's representative and the General Contractor. Uncalibrated fans and/or the building return air-handling system may be used if needed to create the pressure differential required. Temporary sealing of un-closeable openings is permitted if this is needed to obtain the pressure differential required. Examples of such openings are doorways without doors, ducts without dampers and unsealed cable trays i.e. those Appendix C – Sample Enclosure Integrity Test Specification Page 80 of 109

Level 1 Training Manual openings which are ultimately to be sealed prior to contract completion. Measurements need not be recorded. A written report and plan view of the enclosure identifying the location and nature of leaks uncovered shall be submitted. 2. Upon completion of the enclosure by all trades involved (e.g. doors and dampers installed, all penetrations sealed), the Clean Agent contractor shall conduct an Enclosure Integrity Test in conformance with NFPA (12A 1994) & 2001 Appendix C, in the presence of the owner's representative and the General Contractor. Variables of interest are listed in Article 11.6.4 below. Acceptable deviations from the Procedure are listed in Article 11.6.5 below. Should the test be unsuccessful, an inspection shall be conducted and a report and plan view of the enclosure identifying the location and nature of leaks uncovered shall be submitted. If more than one test is required, additional tests shall be at the expense of the contractor(s) whose deficiencies are responsible for the test failure. 3. Upon successful completion of the test conducted in article 6.02, a final Enclosure Integrity Test per Year 2000 NFPA 2001 shall be conducted in the presence of the AHJ or his representative. The contractor shall provide a test report, including a copy of the recorded measurements. Adequate notice shall be given to the AHJ or its representative to enable either or both to attend. If the enclosure's leakage has increased since the successful completion of the test specified in 11.6.2, and this leakage causes the enclosure to fail this test, the enclosure shall be inspected to uncover the source of this leakage. If retesting is required it shall be conducted at the expense of the contractor (s) whose deficiencies are responsible for the test failure. 4. Variables of Interest A. (MECHANICAL MIXING) (DESCENDING INTERFACE) The Clean Agent leakage shall be modeled assuming that (mechanical mixing) (descending interface formation) takes place during the retention period. B. (MINIMUM PROTECTED HEIGHT) (MINIMUM ALLOWABLE CONCENTRATION) The Minimum Protected Height shall be xx.x feet from the floor slab. Use this if a descending interface will form. Note: This is not the height to be used in the calculation of total extinguishing Agent required. Appendix C – Sample Enclosure Integrity Test Specification Page 81 of 109

Level 1 Training Manual The Minimum Allowable Concentration shall be (x)%. Use this if mechanical mixing will take place during the retention period. C. RETENTION PERIOD The minimum retention period shall be (10) minutes. 5. Acceptable Deviation From Year 2000 NFPA 2001 Procedure ONLY TO BE INCLUDED WITH THE PERMISSION OF THE AUTHORITY HAVING JURISDICTION OR THEIR REPRESENTATIVES. The following optional items address possible refinements of the testing procedure. They have not been reviewed or accepted by the 2001 Technical Committee as of the date of this document. Determination of Height of Protected Enclosure Very high or very deep spaces in the room may be ignored when determining the Height of Protected Enclosure provided that the volume of the space represents less than 15% of the zone's volume. If the space is very deep (e.g. dropping below floor level) a qualitative leak inspection must confirm that insignificant leakage exists in the space. The wording in paragraph 4-7.2.3 states that the enclosure shall be "examined or tested" to ensure tight construction. Since most AHJs prefer a quantitative test over a subjective examination, the door fan test is likely to be most often requested. However, this paragraph does not specifically state that leakage measurements must be taken or that the test be used to predict retention time. It is up to the AHJ to determine: a) What procedure to follow (possibly Appendix C), b) Whether the enclosure has to pass a retention time prediction, c) How to test and accept enclosures that are outside the scope of Appendix C, d) What testing is required in addition to that specified in section 1-7.4 in order to accept the system. 10.7 Warranty If Maintenance Service is requested as part of initial installation, add this article to other standard system checks. Appendix C – Sample Enclosure Integrity Test Specification Page 82 of 109

Level 1 Training Manual

12 months after the acceptance of the enclosure, the Clean Agent contractor (or other testing agency if appropriate) shall conduct an Enclosure Integrity Test in conformance with Year 2000 NFPA 2001 Edition Appendix C. Variables of interest are listed in article 6.04 above. Acceptable deviations from the procedure are listed in article 6.05 above. Should the test be unsuccessful, an inspection shall be conducted and a report and plan view of the enclosure identifying the location and nature of leaks uncovered shall be submitted. Furnished by: Retrotec Energy Innovations Ltd., 1015 Ironwork Passage Vancouver BC, Canada V6H 3R4 e-mail [email protected] Phone: (604) 732-0142 FAX: (604) 737-0152

Retrotec provides this information as an industry service. The material is not copyrighted and may be copied and redistributed without restriction. Constructive criticism and comments on this document are welcome in order to improve future revisions.

Appendix C – Sample Enclosure Integrity Test Specification Page 83 of 109

Level 1 Training Manual

11 Appendix D – Enclosure Integrity Verification Form

Enclosure Integrity verification form
Building Room Test # Testing technician Witness Date and time of test
Check off __ yes __ no __ yes __ no __ yes __ no __ yes __ no __ yes __ no __ yes __ no Screen tab Name on Computer Screen “View” button will display the current certificate “View” button will display the current certificate “View” button will display the current certificate Elevation What to look for

Home

Is the One Year Calibration Certificate up to date?

Home

Is the Five Year Calibration Certificate up to date? Does the technician have the correct level of training? See Level 1-A, page 15 Is it correct within 1000 ft.? See Level 1-A, page 12 This is used to re-calculate the design concentration. It must be re-measured, was it? See Level 1-A, page 12 Was the temperature expected during a discharge within 10F or 5 C? It may differ from the temperature at the time of test. See Level 1-A, page 12 Was it re-measured from floor slab to highest combustible? Enter the maximum agent height from lower slab to highest point that is flooded with agent. See Level 1-A, page 13 Do you agree with the time shown? See Level 1-A, page 13

Home

Building/Room

Building/Room

Net protected room volume

Building/Room

Room operating temperature

__ yes __ no __ yes __ no

Building/Room

Maximum agent height

Building/Room

Minimum agent retention

Appendix D – Enclosure Integrity Verification Form Page 84 of 109

Level 1 Training Manual
Check off __ yes __ no __ yes __ no __ yes __ no __ yes __none __ yes __ no __ n/a __ yes __ no __ yes __ no __ yes __ no __ yes __ no Total Leaks Total Leaks Total Leaks Enter untested values Name on Computer Screen Agent weight

Screen tab

What to look for Were you able to confirm the agent weight or volume? See Level 1-A, page 14

Agent/Test

Agent/Test

Agent volume

Only used for INERGEN in North America

Agent/Test

Initial Concentration

Does the concentration meet the specification? See Level 1-A, page 14 Remove all temporary tape or get sufficient assurance it will be replaced with a permanent seal. If untested values were entered, do you agree with their validity? See Level 1-A, page 16. It would be unusual to have untested values.

Smoke

Did you see the smoke movement test at the doorway? See Level 1-A, page 17

Total Leaks

Test both directions:

Was the enclosure tested in both directions?

Total Leaks

Static pressure

Did you observe the static pressure measurement at the time of the door fan test?

Temperature during Was the temperature within 100F or 50C of that test(0F) recorded? The NFPA Procedure requires a measurement if the difference is greater than 18 Temperature during 0F or10 0C test(0C)

__ yes __ no Total Leaks Range for room pressures:

Was the room pressure reading within the range specified?

Appendix D – Enclosure Integrity Verification Form Page 85 of 109

Level 1 Training Manual
Check off __ yes __ no Total Leaks Blower range Name on Computer Screen

Screen tab

What to look for Was the range used verified? A picture of the range will pop up on CA2001 when the room pressure is entered or see the range section a few pages ahead. Were the correct room pressures recorded with the door fan running? Were the pressures checked across each wall or was there sufficient return path from the door fan to the enclosure leaks to ensure the pressure was the same across all enclosure boundaries. Was the flow pressure accurately recorded while the door-fan was running without pop ups in CA2001 warning of the pressure being too low? Applies only to multipoint tests and here the value must be 6% or less for the test to be good Must be between 0.45 and 0.9 to be acceptable applies only to ISO tests Must be 99% or higher on multipoint tests applies only to ISO tests Must be 0.07 or less - applies only to ISO tests Were the Total Leaks measured first and was the smoke was neutralized at ceiling openings before readings were taken? The resultant leakage area must be entered in the Lower Leak tab, NOT the Total Leak tab (common mistake). Were the Total Leaks measured first? The resultant leakage area must be entered in the Lower Leak tab, NOT the Total Leak tab (common mistake). Are you comfortable with the method of estimation?

__ yes __ no

Total Leaks

Room pressure

__ yes __ no ISO only ISO only ISO only ISO only

Total Leaks

Flow pressure:

Total Leaks Total Leaks Total Leaks Total Leaks

Error %: Slope n: Correlation: Standard Error:

__ yes __ no

Lower Leaks

Flex-duct test

__ yes __ no __ yes __ no __ yes __ no

Lower Leaks

Plastic on ceiling test

Lower Leaks

Estimated

Retention

Mixing during retention

Would this occur? See Level 1-A, page 24

Appendix D – Enclosure Integrity Verification Form Page 86 of 109

Level 1 Training Manual
Check off __ yes __ no __ yes __ no __ yes __ no Name on Computer Screen No mixing during retention

Screen tab

What to look for Would conditions be calm enough after discharge for no mixing of agent and incoming air? Is there an extended discharge. If so, what is the quantity and duration? Have the enclosure set up as it would be just prior to agent discharge. Do you think the smoke direction recorded is the same as what it would be during the retention period? Was the height above the lower slab to the equipment being protected, properly measured? Only applies if there is no mixing but it is useful to have it recorded either way. Do you agree that the value recorded will prevent re-ignition at the end of the retention period? Only applies if there is mixing but it is useful to have it recorded either way. This is the retention time given all the variables input into CA2001 so far. If this is greater or equal to the “Minimum agent retention” specified on the Building/Room tab, the room PASSES.

Retention

Retention

Extended discharge

Retention

Smoke

__ yes __ no __ yes __ no __ not sure __ pass __ fail

Retention

Minimum protected height

Retention

Minimum concentration

Retention

Time,t

Note any other concerns you had about the test ->

Appendix D – Enclosure Integrity Verification Form Page 87 of 109

Level 1 Training Manual

12 Appendix E – Glossary of Terms
60 Gauge The gauge used to measure the Room pressure. Full scale is 60 Pascals (or 0.24” W.C.). It is usually located on the left and connects to the door panel by a red tube. 250 Gauge The gauge used to measure Flow pressure. Full scale 250 Pascals (or 1.004” W.C.). It connects to the blower via quick connects and a clear tube. Agent Short form for Clean Agent fire suppressant. Can be an inert or a halocarbon. Agent/Air Interface See sharp interface - the vertical distance through which the agent concentration goes from that discharged to 0. AHJ Authority Having Jurisdiction. ASHRAE The American Society of Heating, Refrigerating and Air Conditioning Engineers. Developers of standards and technical guidance relating to HVAC/R issues. BCLA (Below Ceiling Leakage Area) Leaks below a suspended ceiling. Hole in the floor or lower leaks. Often assumed to be one half of the Total Hole in the Room (ELA). Below Ceiling Leakage = Lower Leak See Lower Leak. Blower As used in the text, this term means the Retrotec Infiltrometer fan unit that both flows air and provides a FLOW PRESSURE signal from which flow is measured. Sometimes it is called a fan. Ceiling neutralization = Flex-duct test = BCLA (Below Ceiling Leakage Area) test Center Panel A red ABS plastic molded sheet, which goes between the upper and lower panel to fill the gap.

Appendix E – Glossary Of Terms Page 88 of 109

Level 1 Training Manual Compartmentation = Compartmentalization

Conditioned Space An area or volume that is normally air-conditioned or heated (i.e. inside the thermal envelope). Even though supply ducts may not discharge directly into these spaces, they are conditioned if their temperature follows indoor temperature closer than outdoor. Continual Mixing Airflow activity within the test room that is sufficient to maintain an equal concentration at all locations and prevent the formation of distinct zones of air and agent/air mixture, i.e. no interface develops. Continuous Discharge = Extended Discharge See Extended Discharge. Descending Interface Agent leaks out during the retention period and air leaks in the upper part of the enclosure to replace the lost volume. Typically it is assumed that as the agent leaks out, an interface will occur between the layer of agent on the bottom and the layer of air on top. Since this interface drops with time, it is called a descending interface and is the most commonly assumed leakage regime. The other regime is Continual Mixing. Depressurization The process of creating a negative pressure in the house by blowing air out of the house. Air is drawn in from outside to replace it, showing up as "geysers" when checked with an air current tester. Door Fan = Blower Door = Blower Door Fan = Infiltrometer ( reg’d tm of Retrotec) A test instrument that fits into an open doorway in order to pressurize and enclosure. The result is a measurement of the hole size. Dropped Ceiling See suspended ceiling Dynamic discharge pressure This is a combination of the peak pressure during the actual discharge and the velocity pressure associated with streams of agent hitting walls or ceilings, thereby trying to force its way out of the enclosure during this brief period. Enclosure=Room In this manual, this word is used to mean the volume that is protected with clean agent. It could also mean the above ceiling space if it is not protected since for all intents the Appendix E – Glossary Of Terms Page 89 of 109

Level 1 Training Manual enclosure boundary is at the fire barrier and suspended ceilings do not represent a fire barrier unless they are fitted with special fire rated tiles. Envelope The surfaces composed of floor and walls and floors that separate the test volume from volume surrounding the test volume ELA= Total Leak = Whole room leakage= Equivalent Leakage Area=Total Leak Equivalent Leakage Area (ELA) In layman's terms, the ELA is the size of hole we'd have if all the building's cracks and holes could somehow be brought together. Also called: Whole Room Leakage and includes Leaks through the ceiling and below the ceiling (BCLA). In CA2001 we measure this in units of ft2 or m2 at a reference pressure in Pascals (Pa). In Engineer’s terms: the equivalent size of hole required in a flat plate to give the same flow rate having a discharge coefficient of 0.61 and taken at the Reference Pressure. This ELA is sometimes called the EqLA or Canadian ELA because it was first used in the Canadian CGSB air leakage standard for houses. This ELA enjoys worldwide acceptance by most testers, even in the US. This ELA should not be confused with another ELA that is often called the EfLA or Effective Leakage Area. It is very unfortunate that both these ELA’s have the same acronym of ELA. The EfLA was developed for the US ASTM Standard and is smaller than the EqLA by at least a factor of 0.61 because it uses a discharge coefficient of 1.0. This EfLA is sometimes called the LBL or Lawrence Berkley Labs ELA because it was developed there and is used in the LBL natural airchange model that enjoys wide usage- apart from that usage, the EfLA is not used very much but the existence of both can create huge problems that are totally lost on some users. Expander Mechanism The red metal mechanism attached to the door panel that enables the panel system to expand sideways into the doorframe. It has a mechanical ratio of 5:1. Extended Discharge = Continuous Discharge An optional method to maintain concentration whereby after the initial discharge an extended discharge takes place with the intention of maintaining the original concentration more or less by injecting a continuous stream of agent for an extended period (usually 10 to 20 minutes). Retrotec CA2001 software will calculate the amount of extended discharge required. Extender Panel Appendix E – Glossary Of Terms Page 90 of 109

Level 1 Training Manual An optional molded plastic panel, which can be temporarily attached to the main door panel to fit up to 48" wide doorways.

False Ceiling See suspended ceiling. Also, can be called T-bar ceiling or Lay-in ceiling. Flex-duct ceiling neutralization A door-fan test method that uses a second blower connected through the suspended ceiling. The second blower takes care of the upper room leaks with the above the ceiling blower. The first blower takes care of the Lower Leaks. The flow through both blowers is adjusted till there is neutral flow across the ceiling that is determined by smoke puffed into gaps. The Lower blower measures the Lower Leaks. Flow Pressure The pressure difference between inside the blower and the surrounding air read off of the Infiltrometer's 60 and 250 Pa flow gauges. It is used by the computer to calculate the airflow through the blower. Height of Interest The highest point in the room requiring protection for the duration of the specified retention time. In the NFPA procedure it’s called the “height of interface from floor”, in the software it’s called the Minimum Protected Height. Hole in Floor (BCLA) All Below Ceiling Leakage Area (BCLA) is assumed to be in the floor to get worst-case leakage rate. Room Pressure The pressure difference created by the blower between inside and out, read off of the Infiltrometer's 60 Pa gauge. This gauge is labeled "Room/House Pressure". HVAC Heating Ventilating and Air conditioning system. Infiltrometer A name used and registered by Retrotec to describe their door-fan equipment. Often called blower-door. Large or Main Panel Refers to the panel with the 20" diameter hole intended for sealing the doorway. Lay in (Tile) Ceiling Appendix E – Glossary Of Terms Page 91 of 109

Level 1 Training Manual See suspended ceiling.

Leakage A general term used to describe holes or the area of holes or leakage through holes in or around an enclosure. See also Total Leaks and Lower Leak(s). Leakage Area This is the same as “Leakage” but express in sq.ft. or sq.m. Lower Leak = Below Ceiling Leakage A lower leak is any leak below the ceiling. Leaks in the walls and floor are counted as Lower leaks where agent will leak out. Lower Leak also refers to the Lower Leak tab of CA2001 where the Lower Leak would be measured. Lower Leaks Leaks attributed to air that flows in from below. If the room were filled full of water, more water would leak from these leaks. All leaks below the ceiling are assumed to be Lower Leaks. Includes wall and floor leaks. Negative Static Pressure A room pressure that is independent of the door fan that will cause test smoke to flow into the room. Pa See Pascal. Pascal Often shown as “Pa”. A very small metric unit of pressure. Equal to 1/249th of an inch of H2O. There are 249 Pascals in 1" Water Column (the pressure required to push water up 1" in a tube). One Pascal = 0.000145 psi. Peak Pressure = Vent Pressure When the system is discharged, there is brief period at the ten-second mark where a maximum peak pressure is created in the room. For inerts, this is where the flow rate is near maximum. For halocarbons this occurs at the end of the discharge where the cooling effect of the agent is reduced and it starts to expand. CO2 starts to increase in volume towards the end of the discharge because the cooling effect caused by the rapid flashing of the agent at the nozzles is eliminated. Positive Static Pressure Appendix E – Glossary Of Terms Page 92 of 109

Level 1 Training Manual A room pressure that is independent of the door fan that will cause test smoke to flow out of the room.

Pressurization This is the process of creating a positive pressure in the house by blowing air into the house. Air is pushed out through all the leaks, causing the smoke to move away from the operator when checked with an air current tester. Protected enclosure This term describes the total space that is flooded with clean agent upon discharge. This includes above ceiling spaces only if that volume is intentionally flooded with agent. This includes adjacent rooms if they are discharged at the same time. Protected enclosure boundary This term describes floor, wall and surfaces that define the protected enclosure. Reading A set of simultaneous Room Pressure and Flow Pressure readings. Sometimes referred to as a data set or test point because it is plotted as one point on a graph. Reference Pressure The pressure at which the ELA is calculated, usually at the test pressure. Return Path Space (Relief Zone) The volume around the tested room that the Infiltrometer blows into (under room depressurization) or out of (under room pressurization). The flow from the Infiltrometer must be allowed to return to the point of leakage in the room through the return path space. Room = Enclosure Sharp Interface The height at which the agent concentration is considered to go form that discharged to 0. The boundary between the agent mixture below and the pure air above. Smaller Panel Refers to the smaller sliding panel used to seal the doorway. It's permanently attached to the large panel. Suspended Ceiling = T-bar ceiling = False ceiling = Dropped ceiling = Lay-in ceiling Common ceiling type found in most computer rooms and offices. Tiles lift up to expose space above. Appendix E – Glossary Of Terms Page 93 of 109

Level 1 Training Manual

Total Leaks = Whole room leakage = (ELA) Total Leak includes floor, wall and ceiling leaks. It also refers to the Total Leak tab in CA2001 where the total leak is measured. T-bar Ceiling See suspended ceiling. Upper Panel Cover This covers the 22" diameter hole (20” for the older 900 series models) in the upper panel and has 2-calibration holes cut to precise size at the factory. The panel comes out with a quick pull. Upper Leaks Leaks attributed to air that flows in from above. If the room was filled full of water, no water would leak from these leaks. Also called the Hole in the Ceiling. Vent Pressure = Peak Pressure Whole Room Leakage (ELA)= Total Leak = Whole room leakage Includes leaks through ceiling and lower leaks.

Appendix E – Glossary Of Terms Page 94 of 109

Level 1 Training Manual

13 Appendix E – CA2001 Demo Example
If you already have a version of CA2001 either from CD or from a download, this building should be preloaded. If it is not, you can import 10 Northtown City Hall test.txt. This example shows screenshots from the software for you to follow. Test: Northtown City Hall This NFPA Retention Time test took place on a small computer room. The first four screens show how easy it is to enter the data to get a Retention Time. The next screens show how to enter the door fan data to get the Leakage Area. The Field Calibration, Wind Losses and Venting test screens are also shown. These are followed by the computer report, then the written report with pictures. 13.1 Home tab

Appendix E – Page 95 of 109

Level 1 Training Manual 13.2 Building/Room tab Choose a “New Building” and input the data shown.

13.3 Agent/Test tab

Note that this Technician is certified to Level 4, the highest Level. In general, testers need a minimum of Level 2 for a single-door fan test and Level 3 for a double-door fan test. Appendix E – Page 96 of 109

Level 1 Training Manual 13.4 Total Leaks tab Since you may have no ability to enter real door fan test data, just click the “Enter untested values” button, then put “.12” into the “Leakage area, ELA” box. Hit the “Calculate/Save” button and move to the Retention tab.

13.5 Retention tab The result is 19.1 minutes.

Appendix E – Page 97 of 109

Level 1 Training Manual

You have now successfully completed a Retention Time calculation using a directly entered Leakage Area value. Now, go back to the Total Leaks tab in your program. Enter the door fan data supplied on the next page. You should get the same Leakage Area and Retention Time result. 13.6 Total Leaks tab Input these “Test” values. You should arrive at the same result as previously.

The room pressure was entered as 17 but displayed as 17.4 here and in the printout. This resulted from the readings being automatically corrected according to the Calibration Certificate that can be viewed any time by hitting the “View” button on the Home tab.

Appendix E – Page 98 of 109

Level 1 Training Manual This Retention Time test is now complete, but we can look at the other tabs to see what is going on. 13.7 Field Cal tab This Field Calibration procedure allows the operator to test his equipment as required by NFPA2001 at any time. Witnesses may ask for this test to be performed to test both the equipment and the operator. 15% is the maximum allowable error. Here, a room is tested, then a 144 sq inch hole is added and the room re-tested. We should see a 144 sq inch difference, though the actual result will never be exactly 144 sq inches. In this case, 133 sq inches was measured which translates into a -7.4% error, which is acceptable.

The Field Calibration Report is produced when this button is pressed.

Appendix E – Page 99 of 109

Level 1 Training Manual 13.8 Wind Losses tab There are not any Wind Losses in this case because the enclosure has no surfaces exposed to the outdoors. We did a wind calculation anyway to show what the losses might be if this building were exposed to the outdoors.

Most enclosures have no walls facing the outdoors. Some have one or two walls facing the outdoors, but most of the leaks are within the building, making Wind Losses small. In the few cases where most of the leaks are in walls exposed to the outdoors, these losses can easily exceed the NFPA or ISO maximum losses.

Appendix E – Page 100 of 109

Level 1 Training Manual 13.9 Venting tab

The venting test was conducted at a higher room pressure than that of the Total Leakage tests and it indicates a correspondingly higher leakage. The vent test should be run at the highest possible pressure to get the most realistic reading. In this case, the maximum allow able room pressure (by design) is 478.9 Pa or 10 lb/sq ft, but the predicted pressure is only 149Pa which will not cause an overpressure problem.

Appendix E – Page 101 of 109

Level 1 Training Manual 13.10 Saved Tests tab

Notice the new test that has been added. This file can now be saved outside the CA2001 program using the Export Building button.

Appendix E – Page 102 of 109

Level 1 Training Manual The computer will ask you where you want to save this file and by what name.

When you use this feature to import files from another user, you not only receive all of the user’s data associated with the test, but you also receive the complete equipment and technician Calibration Certificates. You will see the information exactly as the originator of the test saw the data, analysis, and results.

Appendix E – Page 103 of 109

Level 1 Training Manual 13.11 Calibrations and Reports Both Calibration Certificates are also available for inspection. This is what you see when you hit the “View” tab.

Gauge readings are corrected automatically according to this certificate.

Appendix E – Page 104 of 109

Level 1 Training Manual

Note that this certificate shows this technician’s certification. Blower readings are corrected automatically according to the N, K, K1, K3 and K4 values on this certificate.

Appendix E – Page 105 of 109

Level 1 Training Manual .

Appendix E – Page 106 of 109

Level 1 Training Manual

Note that the statement on this page of the report shows this Technician is certified.

Appendix E – Page 107 of 109

Level 1 Training Manual Level 2 Certification is the minimum required for Technicians performing tests with a single blower. Technicians using two blowers require a minimum Level 3 Certification. Level 4 is for ISO and venting tests.

This graph on page 3 is a nice visual but does not mean a lot for single point tests. Its real value lies in multi-point tests under ISO. Appendix E – Page 108 of 109

Level 1 Training Manual

13.12

Field Calibration Report

Appendix E – Page 109 of 109