AC385-0215-R2 #5

Page | 1

January 9, 2015

Mr. Michael O'Reardon, P.E. Birmingham Director of Engineering ICC Evaluation Service, LLC Birmingham Regional Office 900 Montclair Road, Suite A Birmingham, AL 35213 205.599.9800 Re: Proposed Revisions to the Acceptance Criteria for Special-purpose Sprinklers Used with Fixed Glazed Assemblies to Provide an Alternative to a Fireresistance-rated Wall Assembly, Subject AC385-0215-R2 (MO/ST) Dear Mr. O'Reardon, P.E., I would like to bring the following concerns to the attention of the committee; register my objection to modification or deletion of Sections 5.1.8 and 5.1.11; and register my continued objection to the reinstatement of AC385 and any resulting ESRs. Staff comment:

1. Section 5.1.8 is modified to allow removal of the “pony” wall when testing of an assembly without a “pony” wall is conducted in accordance with Section 3.1 of AC385.

AC385 as modified:

5.1.8 Unless the glass has been tested in accordance with Section 3.1 of this criteria for use without a “pony” wall or other protection from combustible materials, all combustible materials must be kept a minimum distance of 2 inches (51 mm) from the face of the glass. such that complete coverage of the glass by the sprinklers is not impeded. The An acceptable means by which this is can be accomplished is by a minimum 36-inch-high (914 mm) knee or “pony” wall at the base of the wall. The means for maintaining clearance the required separation distance shall be set forth in the evaluation report.

Proponent:

REASONS: (5.1.8): The current requirement to separate combustibles from glazing by a pony wall or other means was initiated based on concern that a small fire located immediately adjacent to tempered or heat-strengthened glazing might cause severe localized heating, which in turn might lead to a thermal-shock induced failure of glazing upon application of sprinkler discharge. That concern does not apply to glass ceramic.

311 Camperdown Court, Easley, SC 29642 732-643-1799 Phone 732-982-1001 Fax

AC385-0215-R2 #5

Page | 2

Full-scale testing with glass ceramic conducted at Southwest Research Institute and witnessed by UL repeatedly demonstrated successful outcomes following a 5-minute flammable liquid pool fire exposure with direct flame contact to the surface of the test assembly. The 5-minute exposure is considered to be a conservative delay before window sprinklers would activate and is consistent with early testing that identified thermal shock as a possible failure mode, leading to the current requirement for separation of combustibles from glazing.

Based on this testing, UL has expanded the listing for Tyco’s WS sprinkler to include installations with glass ceramic that do not require a pony wall or other means to protect the glazing from nearby combustibles. Tyco requests that AC385 be updated, as suggested herein, to reflect this new option so that ESR2397 can be revised accordingly.

This portion of the proposal is problematic for a number of reasons. The ‘pony’ wall is a defined entity, one that is visible once installed. Though there is no data or history provided that building occupants and code enforcers are aware of the significance or and the reason for the pony wall through the life of the occupancy, (a condition of occupancy issue Evaluation Services has never addressed), the pony wall does provide a built in method to keep combustibles away from the glazing in a well maintained occupancy setting by setting up a natural barrier. Elimination of the pony wall and reliance on some “acceptable” method buried in the design details including the ESR filed away in the designers office as well as the building department files guarantees that the need to keep combustibles away from the glazing will not be reinforced in the minds of future owners, occupants or maintenance code inspectors. The reference to a five minute flammable liquid pool fire has no relationship to the normal use and occupancy of any building in question unless of course all fires will occur as flammable liquid pool fires of 5 minute duration. This topic is supposedly dealing with an assembly that is equivalent to the requirements of ASTM E119. Where is the connection? The purpose of ASTM E119 is to provide for a standard repeatable laboratory test that can be used to compare construction methods to see if similar results occur and an hourly rating of fire-resistance can be assigned. Where is the 5 minute flammable liquid pool fire in ASTM E119? What are the exact parameters of this flammable liquid test? Are they repeatable in all labs involved in E119 testing? How much flammable liquid was used? What flammable liquid was use? What size are was involved? How close was the pool fire to the glazing? What was the room temperature before and after the burn? Humidity levels? Is there a standard on how to conduct this test for reference and use by others? ASTM E119 is a standard providing for a repeatable test following specific protocols. Yet, this AC and the information provided by the applicant seeking approval as equivalent to ASTM E119 does not follow any standard test procedure. Fire-resistance rated glazing is required to be tested in accordance with ASTM E119.

IBC-2015 703.6 Fire-resistance-rated glazing. Fire-resistance-rated glazing, when tested in accordance with ASTM E 119 or UL 263 and complying with the requirements of Section 707, shall be permitted. Fire-resistance-rated glazing shall bear a label marked in accordance with Table 716.3 issued by an agency and shall be permanently identified on the glazing.

Page | 3

Fire Protection rated-glazing used in fire windows is required to be tested in accordance with NFPA 257.

IBC-2015 716.6 Fire-protection-rated glazing. Glazing in fire window assemblies shall be fire protection rated in accordance with this section and Table 716.6. Glazing in fire door assemblies shall comply with Section 716.5.8. Fire-protection-rated glazing in fire window assemblies shall be tested in accordance with and shall meet the acceptance criteria of NFPA 257 or UL 9. Fire-protection-rated glazing shall comply with NFPA 80. Openings in nonfire-resistance-rated exterior wall assemblies that require protection in accordance with Section 705.3, 705.8, 705.8.5 or 705.8.6 shall have a fire protection rating of not less than 3/4 hour. Fire-protection-rated glazing in 0.5-hour fire-resistance-rated partitions is permitted to have an 0.33-hour fire protection rating.

In all cases in the International Building Code, if the use of a fire-resistance rated glazing or the use of a fire-protection rated glazing product is proposed for use in another assembly required to meet another test standard, it must be tested to that standard. So if I use fire-resistance rated glazing that already meets ASTM E119 in a fire door assembly, the glazing has to be tested as part of that assembly to the standards applicable to fire doors including the hose stream test. The same with fire-protection rated glazing. In this case we have a “wall” constructed of glazing and sprinkler discharge that seeks equivalency to ASTM E119, yet there is no testing of the assembly to the unmodified ASTM E119 and we now have a reference to a flammable liquid pool fire that there is no standard for to ensure repeatability. If an applicant cannot use ASTM E119 glazing in a fire door assembly as part of a sidelight or transom unless it also undergoes the testing under the fire door standard, NFPA 252, how can an ESR allow the use of fire-protection rated glazing in a wall claiming equivalency to ASTM E119 unless that glazing is tested in accordance with ASTM E119 in the same manner as any other construction method? The concept violates the code requirements for the use of glazing materials in addition to the prohibition on using fire sprinkler systems to establish passive fire-resistance rating. A large issue not addressed is radiant heat. An ASTM E119 assembly blocks conduction heat transmittal and also blocks radiant heat transmittal because it is opaque. Fire-protection rated glazing does not block radiant heat. In fact, the ceramic product the proponent is seeking relaxed rules for allows more radiant heat transmittal than other fire-protection rated glazing. Ceramic glass is used as cooktops and fire place doors because of its excellent ability to transmit heat through conduction and radiation. The proponent is asking for the elimination of the pony wall to allow an entire wall of ceramic glazing. Where is the data on radiant heat transmittal that will now be permitted? Does the radiant heat data exist? What standard was used to determine radiant heat transmittal if that radiant heat data exists? The International Building Code specifically limits the use of fire-protection rated glazing to specific locations and amounts because of the radiant heat transmittal that could occur. There is no data addressing this issue and ignoring the topic simply negates the requirements of the IBC without providing for equivalency.

Page | 4

NFPA 80, “The Standard on Fire Doors and Other Opening Protectives” highlights the issue of radiated heat in Annex J. (Attached)

NFPA 80-2013 I.1 Background. Fire windows were originally designed for protecting openings in exterior walls. In such applications, radiant heat transfer was not a significant consideration, since the main function of fire windows was to contain the flames within the building. However, where fire windows are used in interior partitions, users of this standard might need to consider radiant heat transfer during fire. Exiting through corridors and past fire windows could be compromised, and combustible materials on the unexposed side of fire windows could be ignited. The information that follows is a guide to the evaluation of radiant heat transfer through fire windows. Recent revisions to this standard have permitted very large areas of fire protection–rated glazing materials to be used in interior partitions, limited only by the size of the test furnace. Also, recent technological advances in the glazing industry have compounded the problem of radiant heat transfer by making it possible to provide glazing materials with fire protection ratings of 60 minutes and 90 minutes. Historically, fire windows, including glass block, have been limited to a 45-minute rating by the standard fire test, NFPA257, Standard on Fire Test for Window and Glass Block Assemblies. This time limit was predicated on the failure of wired glass at approximately 1600°F (870°C). [1] Some manufacturers also have developed fire resistance–rated glazing assemblies that meet the requirements of a fire resistance–rated wall assembly (currently up to 2 hours). These glazing materials, however, do not transmit excessive radiant heat, since they are re-quired to limit the temperature rise on the unexposed face to 250°F (121°C).

The elimination of the pony wall and increase in area of the permitted glazing cannot be considered without extensive information concerning transmitted radiant heat. Recognizing that the information application would only be for the heat release rate of the fire source used for the data. With ASTM E119 the starting point for radiated heat is zero. This proposal must be measured against that. Since it has long been recognized that water curtains do not stop radiant heat transmittal, this issue effects the entire appropriateness of AC385. Staff comment:

2. Section 5.1.11 is modified to allow use of the assemblies as fire barriers in the construction of exit passageways, horizontal exits, or exit enclosures when testing is without a pony wall or with some other means of protection from combustible materials. Otherwise, the “fire area” in which the assembly is located will need to be fully sprinklered. The proponent is also proposing revisions to require the submittal of documentation to the code official from a registered design profession at addressing performance when the assembly is intended for installation in a required means of egress. As currently specified in Section 5.1.11, use in lieu of firewalls is not permitted.

Page | 5

The ICC-ES staff requests clarification of the term “fire area” as used in the proposed revisions to Section 5.1.11. A definition of “fire area” needs to be added to the criteria that, where applicable, makes reference to provisions of NFPA 13 for special sprinklers and the area of design or protection area of coverage. Section 5.1.1 of AC385 dated October 2013, reproduced as a Condition of Use in Section 5.9 of the evaluation report ESR-2397 dated October 2013, was added to the acceptance criteria upon the recommendation of the Evaluation Committee during the February 2013 hearing. This addition was intended to clarify where the sprinkler-protected fixed glazed assemblies may be used, and is based on language in Section 3.1.8.18 of the Ontario Building Code (OBC).

AC385 as modified:

5.1.11 The assemblies evaluated are not permitted to be used in lieu of firewalls. Where the assemblies are used an alternative to fire barriers used as for exit-passageways, horizontal exits, or exit enclosures, the fire area in which the assembly is located shall be fully sprinklered or the assembly shall use glass that has been tested in accordance with Section 3.1 of this criteria for use without a “pony” wall or other protection from combustible materials. The evaluated systems must be limited to use in fully sprinkled buildings. In addition, the registered design professional shall provide the code official with documentation in accordance with Section 104.10 of the IBC for a code modification or Section 104.11 of the IBC for an alternative method of construction that addresses any anticipated impact on the functionality of the means of egress.

Proponent:

(5.1.11): Although we continue to believe that use of AC385 assemblies should not be restricted based on the presence of a fire sprinkler system in the building or restricted with respect to exit enclosures, because these restrictions were not technically justified when they were added to AC385, we recognize that the Committee is seemingly unwilling to revert to the legacy approach that had been in place for decades prior to AC385’s recent repeal and reinstatement. Accordingly, this proposal attempts to identify a middle ground that maintains an increased level of fire protection, above what had been required by the legacy provisions, but better addresses consumer needs. The proposed middle ground is based on comments that were offered at the Committee’s October hearing. The particular concern that seemed to resonate with the committee at the October hearing was the possibility of a fire in remote area reaching a state of rapid growth or flashover and then instantaneously producing a severe exposure to the glazing, perhaps before window sprinklers would operate and leading to a subsequent thermal shock condition. Although this type of exposure is not contemplated or tested under ASTM E119 for any fire-resistive assembly, the committee nevertheless considered this to be of concern with respect to AC385 assemblies. It was suggested that, if sprinklers are provided in the fire area where the AC385 assembly is located, the possibility of such an event would be mitigated. From the perspective of the AC385 assembly, the assembly will be in an area that is sprinklered, and a fire outside the immediate fire area (as defined and constructed per the IBC), could not expose the assembly without having to pass through both a fire barrier and a sprinklered building area. With this

Page | 6

change, the basic intent of the current AC385 restriction would be retained, with the portion of the building containing the AC385 assembly be sprinklered.

As an alternative, it is proposed to allow the use of glass ceramic, which is now UL listed for use with window sprinklers and is not subject to the risk of thermal shock induced failure from a sudden fire exposure followed by sprinkler discharge.

This proposed change will allow the use of a glazing material that has a recognized affinity for the transmittal of radiant heat to be used in locations that the code specifically forbids any use of fire protection rated glazing because it allows for the transfer of radiant heat. Applicant has submitted no radiant heat testing or documentation as identified earlier in this comment. All proponent has done has provided for the testing of one type of fire protection rated glazing to be used in place of another heat treated glazing material. There no data on radiant heat transmittal provided for either type of glazing and the ceramic glazing proposed allows for increased radiant heat transfer. How can an Acceptance Criteria be proposed that clearly violates code restrictions on the use of fire-protection rated glazing? The application of this acceptance criteria already severely modifies the application of ASTM E119 in manner than does not allow comparison of other methods of construction which is the purpose of having a standard, with this suggestion modification there is simply a voiding of important building code requirements that are supported by NFPA 80. AC385 and the proponent are providing for the allowance of a method of construction that will rely on an active fire protection method that has been identified by NFPA to only operate effectively 87% of the time.

http://www.nfpa.org/research/reports-and-statistics/fire-safety-equipment/us-experience-with-sprinklers “Sprinklers operated in 91% of all reported structure fires large enough to activate sprinklers, excluding buildings under construction and buildings without sprinklers in the fire area. When sprinklers operated, they were effective 96% of the time, resulting in a combined performance of operating effectively in 87% of all reported fires where sprinklers were present in the fire area and fire was large enough to activate them. The more widely used wet pipe sprinklers operated effectively 89% of the time, while dry pipe sprinklers operated effectively in 76% of cases.”

I don’t believe that is an acceptable performance level to issue an ESR on. I included the following concern in my public comment when AC385 was under consideration for re-establishment:

"The current proposal before the committee, as with the previous AC385, will create an unacceptable conflict with current code requirements. "TABLE 716.5 OPENING FIRE PROTECTION ASSEMBLIES, RATINGS AND MARKINGS" which is based upon the technical requirements currently in the IBC/2012, is very specific as to where fire-resistance rated glazing and fire-protection rated glazing can be installed. As currently drafted as a wide ranging alternative approval recognition including for fire barriers as

Page | 7

specified by "Section 1.2 Scope", AC385 permits the use of unrated glazing materials in amounts and locations that even fire protection rated glazing is either restricted in use or not permitted at all. How can you have an alternative solution that violates code provisions approved as equivalent of that prescribed in this code in quality, strength, effectiveness, fire resistance, durability and safety?"

I have attached copies of 2015 International Building Code Tables 716.5 and 716.6. Table 716.5 documents that the current IBC does not permit the use of fire-protection rated glazing materials in fire door assemblies other than as vision panels not exceeding 100 sq. in. of size in the following locations:

Fire walls >1 hour Fire barriers > 1 hour Shaft, exit enclosures and exit passageway walls Fire barriers having a required fire-resistance rating of 1 hour: Enclosures for shafts, exit

access stairways, exit access ramps, interior exit stairways, interior exit ramps and exit passageway walls

Table 716.6 does not permit the use of fire-protection rated glazing materials for fire windows in the following locations:

Fire walls Fire barriers

However, fire-resistance rated glazing materials are permitted in these locations because they are subject to a full ASTM E119 evaluation and listing, i.e., they are considered a fire-resistance rated wall assembly. The code has already determined that any glazing other than fire-resistive rated glazing is restricted, to issue a blanket ES Report for a protection method that counters, or suggests the code language can be countered, would be inappropriate, misleading and create a potential for protection levels below the minimum specified by the IBC. There are few areas of the code where a method of construction is specifically prohibited as it is in Chapter 7 for the use fire-protection rated glazing in certain assemblies. Issuing a report that indicates the prohibition can simply be over ridden by an alternative method submittal is improper. There is a technical basis to limit this application to fully sprinklered buildings. The IBC already does that with the permitted use of wetted glass separating atriums from the remainder of a building.

[F] 404.3 Automatic sprinkler protection. An approved automatic sprinkler system shall be installed throughout the entire building.

Exceptions:

Page | 8

1. That area of a building adjacent to or above the atrium need not be sprinklered provided that portion of the building is separated from the atrium portion by not less than 2-hour fire barriers constructed in accordance with Section 707 or horizontal assemblies constructed in accordance with Section 711, or both. 2. Where the ceiling of the atrium is more than 55 feet (16 764 mm) above the floor, sprinkler protection at the ceiling of the atrium is not required.

404.6 Enclosure of atriums. Atrium spaces shall be separated from adjacent spaces by a 1-hour fire barrier constructed in accordance with Section 707 or a horizontal assembly constructed in accordance with Section 711, or both.

Exceptions: 1. A fire barrier is not required where a glass wall forming a smoke partition is provided. The glass wall shall comply with all of the following:

1.1. Automatic sprinklers are provided along both sides of the separation wall and doors, or on the room side only if there is not a walkway on the atrium side. The sprinklers shall be located between 4 inches and 12 inches (102 mm and 305 mm) away from the glass and at intervals along the glass not greater than 6 feet (1829 mm). The sprinkler system shall be designed so that the entire surface of the glass is wet upon activation of the sprinkler system without obstruction; 1.2. The glass wall shall be installed in a gasketed frame in a manner that the framing system deflects without breaking (loading) the glass before the sprinkler system operates; and 1.3. Where glass doors are provided in the glass wall, they shall be either self-closing or automatic-closing.

Staff comment:

3. Section 3.1.1 is modified to add the following wording: The test assemblies shall be representative of the assemblies to be recognized in the evaluation report. This modification is needed to clarify exclusion of the pony wall when a sprinkler-protected glazed wall assembly without a pony wall is tested in accordance with Section 3.1 of AC385.

AC385 as modified:

3.1.1 Testing: Testing shall be performed based upon ASTM E119 or UL 263 with modifications to take into account the special-purpose fire sprinklers. ULC/ORD-C263.1 shall be consulted as a basis for determining modifications to the ASTM E119 or UL 263 test standards, other than those specified in this section. The testing shall be performed in a room sized to accommodate the test

Page | 9

furnace and water spray. Prior to testing of the special-purpose fire sprinkler-protected glazed wall assembly, test furnace calibrations shall be conducted to establish a natural gas flow rate curve required to generate a time-temperature relationship as specified in ASTM E119 or UL 263. This calibration testing shall consist of a gypsum wallboard assembly, known to provide a minimum three-hour fire exposure, that is installed in a test frame in the location where the glazing is intended to be installed, with no special-purpose fire sprinkler discharge during the calibration test. The fire source shall be a premixed velocity burner system that can be adjusted to generate the desired fire. The fire test shall be conducted on the gypsum wallboard assembly to establish the gas flow necessary to maintain the time-temperature curve specified in ASTM E119 or UL 263. The gas flow that is established shall be used to conduct testing of the glazed assembly with special-purpose fire sprinklers discharge, to establish the ability of the special-purpose fire sprinklers to protect the window under the same natural gas flow conditions established during the calibration test. An engineering analysis establishing this correlation shall also be submitted to ICC-ES. The assembly size shall be as set forth in Section 17 of ASTM E119 or Section 7.2 of UL 263. Tests shall be conducted on assemblies containing both vertical mullion and butt-glazed conditions. Specific details of the test protocol, test chamber, specimen construction and thermocouple location shall be submitted to ICC-ES for review prior to testing. Test assemblies shall be representative of the assemblies to be recognized in the evaluation report.

This portion of AC385 and the suggested modification identifies the flaws in this entire exercise. The requirements are that you use a standard referenced by the code and mandated to be utilized by several sections of the technical language in Chapter 7 of the IBC as modified by another standard. Why do we have standards? If the purpose of ASTM E119 is to provide for a standard test method so we can equate the performance characteristics of one assembly against that of another, how does that occur if you are using another standard to modify the conditions for one of the comparison methods of construction? The suggested modification highlights the problem with AC385. Here it makes it clear that the test assemblies shall be representative of the assemblies to be recognized in the evaluation report, but we don’t know what that assembly is, where it is located, what the occupancy is, and what the fuel load exposures are until we have an actual design of the building and the location where it is to be constructed. That is the fallacy of approving AC385 and any resulting ES report, there is no way to determine if the assembly “complies with the intent of the provisions of this code, and that the material, method or work offered is, for the purpose intended, not less than the equivalent of that prescribed in this code in quality, strength, effectiveness, fire resistance, durability and safety” without knowing where it is going or what type of fire it may be subject to. The last objection I have is the ludicrous proposal to allow the use of this type of wall assembly in unsprinklered areas regardless of the inclusion of ceramic fire-protection rated glazing based upon a five minute exposure to a flammable pool fire. That exposure does not relate in any fashion to s fully involved room or area fire which will subject the entire assembly to elevated temperatures over 2000 degrees. It should be noted that the time temperature curve used for ASTM E119 would have the furnace temperature at 1000 degrees Fahrenheit at 5 minutes of time and at 1700 degrees at 1 hour of time. UL has been doing research that includes comparison of legacy furnishings to

Page | 10

modern furnishings and full scale testing documents room flashover with 3 minutes and 30 seconds with temperatures well over 2000 degrees Fahrenheit. So the anecdotal information proponent has supplied in an attempt to put forth the assemblies ability to withstand a fully involved room fire is insufficient to justify the proposed change. That’s without considering the radiant heat issue that will allow for the spread of fire and/or endanger occupants on the other side of the glazing. It should also be pointed out that the 87% effectiveness of sprinkler systems documented by NFPA would only apply to a completely suppressed fire area or building. The wetted glass arrangement which is the subject of this discussion is a partial system and there is no reliability data. Fire protection concepts and the International Building Code identify that an entire fire area or building must be protected by the automatic sprinkler system for the area to be considered protected. Even where wetted glass is permitted for protecting an atrium the atrium, (the hazard), must be protected by a fire suppression system. The Acceptance Criteria and resulting Evaluation Service Report are not necessary. The code identifies that either Sections 104.10 or 104.11 can be utilized to apply for permission to use alternative methods of compliance. Tyco has a specification sheet, TFP620, which can be supplied along with the complete results of the ASTM E119 testing to be reviewed by the code official in relation to the actual location of the proposed assembly and the actual fire exposure to determine if the specific application of the technology being proposed can be approved. From a logistics standpoint, Evaluation Services should not be in the business of establishing Acceptance Criteria for "Section 104.10 Modifications" which apply wherever there are practical difficulties involved in carrying out the provisions of this code for a specific design; or for "Section 104.11 Alternative materials, design and methods of construction and equipment" which are to be provided for approval as part of a specific design and situation. These types of issues are design specific and not standard applications, by nature and application of the code they do not avail themselves of a one size fits all solution. In the case of an alternative method there needs to be a finding of equivalency in quality, strength, effectiveness, fire resistance, durability and safety. Since each situation is different, including fuel package and risk, there can be no standard approval in the form of an Evaluation Services Report. My request is that none of the proposed modifications be approved. However, my preference is that AC385 be rescinded entirely as not necessary and in conflict with the application of the International Building Code l. Sincerely, Robert J Davidson Fire & Life Safety Consultant.

Copyright 2013 National Fire Protection Association (NFPA). Licensed, by agreement, for individual use and single download via NFCSS All Access on July 1, 2013 to Robert Davidson for designated user Davidson Code Concepts, LLC. No other reproduction or transmission In any form permitted without written permission of NFPA. For inquires or to report unauthorized use, contact [email protected]. This

NFCSS All Access subscription expires on 10/16/2013.

80-72 FIRE DOORS AND OTHER OPENING PROTECTIVES

Intumescent caulk as~ required ~

V. in. dia. countersunk sheet metal screws 6 in. max. from ends and 24 in. max. o.c.; jamb punched and dimpled for screws

25 gauge steel studs

Note: For Sl conversion, 1 in. = 25.4 mm.

Existing wall

\_Door side rabbet

#8 countersunk sheet metal screws 6 in. max. from ends and 48 in. max. o.c.

Hat-shaped spacer at each screw

FIGURE G.I0.2 Single Swing Two-Section Adjustable Frame Wrapping Existing Drywall Anchor System with IY2-Hour Posi­tive Pressure.

Annex H Special Purpose Doors

This annex is not a part of the requirements of this NFPA document but is included for informational purposes only.

H. I Acoustical fire doors and frame assemblies are available in single swings and pairs and are furnished complete with sound seals. These doors are tested in accordance with the standard fire test of door assemblies and also with theASTM E90, Standard Test Method for Laboratcrry Measurement of Airborne Sound Transmission Loss of Building Partitions and Elements, test for airborne sound transmission loss with a Sound Transmission Classification (STC) rating determined by the procedures in ASTM E413, Classifica­tion for Rating Sound Insulation.

H.2 Security fire doors and frame assemblies are available in two types of security applications. The institutional type is for use in detention and correctional facilities where prevention against escape, unauthorized movement among secure areas, and van­dalism by inmates are of primary concern (see NAAMM/HMMA-863, Guide Specifications for Detention Security Hollow Metal Doors and Frames). The commercial type is used to protect the rear en­trances of commercial establishments, interior secure areas, electric-generating stations, data processing centers, and security control areas of office buildings and other building types, includ­ing those impacted by homeland security, where protection of life and assets against unlawful entrance, theft, vandalism, and terror attacks is of primary concern (see NAAMM/HMMA-862, Guide Specifications for Commercial Security Hollow Metal Doors and Frames). Security doors are available with bullet-resistant capabili­ties. These doors are tested in accordance with the standard fire test of door assemblies and additionally might be tested in accor­dance with UL 752, Standard for Safety Bullet-Resisting Equipment, which specifies bullet-resistant ratings for medium-power small

!lJ 2013 Edition

arms, high-power small arms, super-power small arms, and high­power rifles. Hardware is provided with the door assembly in the case of doors tested in accordance with ANSI/UL 752.

H.3 Armored attack-resistant fire doors and frame assem­blies are available in single swing design. These doors are tested in accordance with the standard fire test of door assem­blies. This special type of security door assembly also has bal­listic resistance, including resistance to small arms multiple impact threat, armor piercing, and projectiles, and is resistant to attack by tools that can impact the door.

H.4 Radiation-shielding fire doors and frame assemblies are available in single swing and pairs. These doors are tested in ac­cordance with the standard fire test of door assemblies. Doors and frames are lined with lead and are designed to provide shielding against specific types and intensities of radiation.

H.5 Pressure-resistant fire doors and frame assemblies are available in single swing and pairs. These doors are tested in accordance with the standard fire test of door assemblies. These doors also are designed to resist blasts from explosions, wind forces, and pressure differences between spaces.

H.6 Stainless steel fire doors and frame assemblies are avail­able for applications that range from the high aesthetic and decorative to low aesthetics and high corrosion resistance. Typical applications for high aesthetics are interior designs in high-profile office buildings, art centers, hotels, and embas­sies, whereas examples of high-corrosion-resistance applica­tions are interior or exterior openings in water treatment plants, food processing plants, and public swimming pools. (See NAAMM/HMMA 866, Guide Specifications for Stainless Steel Hollow Metal Doors and Frames.)

Annex I Radiant Heat Transfer

This annex is not a part of the requirements of this NFPA document but is included for informational purposes only.

1.1 Background. Fire windows were originally designed for protecting openings in exterior walls. In such applications, radiant heat transfer was not a significant consideration, since the main function of fire windows was to contain the flames within the building. However, where fire windows are used in interior partitions, users of this standard might need to con­sider radiant heat transfer during fire. Exiting through corri­dors and past fire windows could be compromised, and com­bustible materials on the unexposed side of fire windows could be ignited. The information that follows is a guide to the evaluation of radiant heat transfer through fire windows.

Recent revisions to this standard have permitted very large areas of fire protection-rated glazing materials to be used in in­terior partitions, limited only by the size of the test furnace. Also, recent technological advances in the glazing industry have com­pounded the problem of radiant heat transfer by making it pos­sible to provide glazing materials with fire protection ratings of 60 minutes and 90 minutes. Historically, fire windows, including glass block, have been limited to a 45-minute rating by the stan­dard fire test, NFPA257, Standard on Fire Test for Window and Glass Block Assemblies. This time limit was predicated on the failure of wired glass at approximately 1600°F (870°C). [1] Some manufac­turers also have developed fire resistance-rated glazing assem­blies that meet the requirements of a fire resistance-rated wall assembly (currently up to 2 hours). These glazing materials, how­ever, do not transmit excessive radiant heat, since they are re-

Copyright 2013 National Fire Protection Association (NFPA). Licensed, by agreement, for Individual use and single download via NFCSS All Access on July 1, 2013 to Robert Davidson fo! designated use~ Davidson Code Concepts, LLC. No other reproduction or transmission in any form permitted without written permission of NFPA. For inquires or to report unauthorized use, [email protected]. Th1s

NFCSS All Access subscription expires on 1011612013.

ANNEX I 80-73

154 in. max., drywall; 162 in. max., masonry

L~ I ( 154 in. max., drywall; ) I

162 in. max., masonry

154 in. max., drywall; 162 in. max., masonry

l_ Projected fire window frame,

a;. hr positive pressure

1000 "''" m~.do-ll; II // 1111 // 1111 // II m;o m;o,<:o m~ 144 1n. max., masonry /1 /1 /1

000 1 V. in. min., 12 in. max . ....__ ______ ____. ----r

Notes: 1. The projected window frame can include a door frame that is part of a fire-rated door assembly having a minimum a;. hr rating. 2. Inside angle between frame segments can vary. Frame curvature can be continuous or segmented. 3. For 81 conversion, 1 in.= 25.4 mm.

FIGURE G.l0.3 Projected Window Frame.

quired to limit the temperature rise on the unexposed face to 250°F (121 °C).

Test Method. Because the present fire test standard, NFPA 257, does not require measuring and reporting tem­perature rise on the unexposed face of the glazing material or radiant heat transmission, glazing products tested to this stan­dard have not been required to retard heat transfer. However, these data are required in many European fire test standards. [2] As a result, European building codes place limitations on the use of glazing in fire-resistant partitions inside buildings and require the use of insulating glazing in means of egress as well as where combustibles could be in close proximity. Re­search by Margaret Law, Bsc., Ministry of Technology and Fire Offices Committee, Joint Fire Research Organization, led to the development of such limitations in British building regu­lations. [3,4] This research provides a methodology for calcu­lating safe distances from wired glass windows used to screen room fires from adjacent spaces.

Law's research properly identifies two major concerns for the use of fire protection-rated glazing in interior partitions as follows:

(1) The impact on occupants exiting past the glazing (2) The potential for nonpiloted (auto)ignition of combus­

tibles on the unexposed side of the glazing

Both of these concerns should be taken into consideration by users of this standard when evaluating a specific fire protection-rated glazing material for interior application.

The exiting concern relates mostly to corridor applications where evacuating occupants might pass directly in front of the glazing that screens them from fire. Calculation methods de­scribed in references 1 and 5 can be used to determine the radiant heat flux generated by a fire as well as the incident heat flux on a person located any distance beyond the unex­posed face of the glazing. Safe distances for evacuees then can be determined from Figures 3-10.59 and 3-10.60 in the SFPE Handbook of Fire Protection Engineering, which provide data use­ful in estimating the time to reach pain threshold and the time taken to cause second-degree burns. [5]

Exit Enclosures. Traditional glazing materials have been pro­hibited from being used in fire windows in exit stair enclosures because of the concern of radiant heat transfer. Recently, the model building codes also incorporated requirements for lim­iting the temperature rise on the unexposed face of fire doors opening into exit stair enclosures in order to address the prob­lem of heat transfer (both conducted and reradiated) that could expose evacuating occupants passing doors at each floor landing. Therefore, caution should be exercised when considering glazing materials with fire protection ratings of

II] 2013 Edition ... ..,

Copyright 2013 National Fire Protection Association (NFPA). Licensed, by agreement, for Individual use and single download via NFCSS All Access on July 1, 2013to Robert Davidson for designated user Davidson Code Concepts, LLC. No other reproduction or transmission in any form permitted without written permission of NFPA. For Inquires or to report unauthorized use, contact [email protected]. This

~ NFCSS All Access subscription expires on 10/16/2013.

80-74 FIRE DOORS AND OTHER OPENING PROTECTIVES

1 hour or more in such applications, since they can transmit ex­cessive radiant heat into the exit stair enclosure. However, glaz­ing materials with fire resistance ratings are suitable in such situ­ations, since they have been tested to limit heat transfer.

Irradiation Levels. Addressing the problem of the nonpi­loted (auto)ignition of combustibles stored near a fire window demands an understanding of critical irradiation levels. Inci­dent flux levels for autoignition of various combustible mate­rials have been developed. [3,6] Average values of 30 kW /m2

to 35 kW /m2 normally are used for ordinary (cellulosic) com­bustibles. Lower values have been identified for some syn­thetic materials.

The radiant intensity (heat flux) of the exposing fire de­pends on, among other factors, the type of materials burning (rate of heat release) and the ventilation rate of the enclosing room. For well-ventilated fires in light hazard occupancies, such as offices, schools, institutions, and residences, a peak radiation intensity (output) of 85 kW/m2 has been used by Law to represent a 1-hour fire exposure. [3] Nelson provides a method for determining safe separation distances based on the radiant heat flux incident on a combustible material screened by wired glass (with a transmissivity of0.5) from a fire (with an emissivity of 1.0; called a blackbody). [6] In general, a fully developed compartment fire is viewed as a blackbody and, therefore, is assigned an emissivity of 1.0. To determine the radiation intensity, the following formula can be used:

where: I = radiation intensity (kW /m2

)

e = emissivity o = Stefan-Boltzmann constant

(5.67 X w-n kW/m2-K4 )

T = absolute temperature of the fire (K)

For most situations, the temperature of the compartment fire is the only unknown variable. NFPA 72, NationalFireAlann and Signaling Code, provides some guidance for calculating room temperature based on different fire growth rates. Other methods base the temperature on the standard temperature­time curve used in ASTM Ell9, Standard Test Methods for Fire Tests of Building Construction and Materials. In a fully developed compartment fire assumed to have a temperature of 1600°F (87ooc or 1140 K), the radiant heat flux would be approxi­mately 105 kW/m2.

To determine the incident radiant heat flux on the unex­posed side of the glazing, the following formula can be used:

I, =Fti

where: I, = incident radiation intensity (heat flux) (kW /m2

)

F = configuration factor for the glazed opening t = transmissivity of the glazing material I = radiation intensity of the fire (kW /m2

)

Transmissivity of Y4 in. (6.35 mm) wired glass has been re­ported in the range of 0.4 to 0.6. Many analyses have used 0.5 transmissivity to account for the effects of reradiation by the glazing product Some manufacturers might be able to provide specific heat transfer information relative to their products.

Users should consider the significance of the source radia­tion, the transmissivity of the glazing material, the time of expo­sure, the separation distances, and the configuration of the glazed opening in relation to the target.

!lJ 2013 Edition

Continuing with the earlier example of a fire having a tem­perature of 1600°F (870oC), the incident radiation intensity (heat flux) (Ji) for a window opening having a configuration factor of 0.35 in relation to a combustible target would be approximately 18.4 kW /m2

• The referenced documents pro­vide detailed guidance for this analysis.

1.2 References.

( 1) National Fire Protection Association, Fire Protection Handbook, 20th ed., 2008, Section 18, Chap 1, "Confinement of Fire in Buildings," pp. 18-15 to 18-20, "Protection of Openings."

(2) ISO 3009, Fire Resistance Tests- Glazed Elements, Interna-tional Organization for Standardization, 1976, Amend­ment 1: 1984.

( 3) Law, Margaret, "Safe Distances from Wired Glass Screening a Fire," Institution of Fire Engineers Quarterly, London, 1969.

(4) Law, Margaret, "Heat Radiation from Fires and Building Separation," Fire Research Technical Paper No. 5, Lon­don, 1963.

(5) Society of Fire Protection Engineers, SFPE Handbook of Fire Protection Engineering, 4th ed., 2008, "Thermal Radiation Hazards," pp. 3-313 to 3-315.

(6) Nelson, Harold E., "Radiant Energy Transfer in Fire Protec­tion Engineering Problem Solving," Fire Technology, Vol. 4, No. 3,August 1968, pp. 196-205.

AnnexJ Performance-Based Option for the Inspection, Testing, and Maintenance of Fire Door

Assemblies

This annex is not a part of the requirements of this NFPA document but is included for infonnational purposes only.

J.l This annex provides the option to adopt a performance­based method as an alternative means of compliance for Sec­tion 5.2. Equivalent levels of performance can be demon­strated through quantitative performance-based analyses. This annex provides a basis for implementing a performance-based program acceptable under this option (provided that ap­proval is obtained by the AHJ).

J.2 The concept of a performance-based program is to establish the type and frequency of inspection to demonstrate that the assembly is operational. The goal is to balance the inspection frequency with proven reliability of the assembly. The goal of a performance-based inspection program is also to adjust test and inspection frequencies commensurate with historical docu­mented equipment performance and desired reliability. Fre­quencies of tests and inspections under a performance-based program can be extended or reduced from the once-per-year test requirement in 5.2.4 when continued testing of door assemblies in 5.2.3, 5.2.3.5, or 5.2.3.6, as applicable, has been documented, indicating a higher or lower degree of reliability compared with the AHJ's and the owner's expectations of performance. Addi­tional program attributes that should be considered in the adjust­ment of test and inspection frequencies include the following:

(1) Door maintenance programs (2) Door usage frequencies (3) History of door repairs ( 4) Building condition (5) Consequence of failure

J.3 Fundamental to implementing a performance-based pro­gram is that adjusted test and inspection frequencies should be technically defensible to the AHJ and supported by evi-

One SeaGateSuite 1700

Toledo, OH 43604DIRECT DIAL 419.254.5246

PHONE 419.242.7985 FAX [email protected]

WWW.RALAW.COM

January 9, 2015

ICC Evaluation Servicec/o Michael O’Reardon, P.E.Birmingham Regional Office900 Montclair Road, Suite ABirmingham, AL [email protected]’[email protected]

Re: COMMENTS IN OPPOSITION TO PROPOSEDREVISIONS TO THE ACCEPTANCE CRITERIA FORSPECIAL-PURPOSE SPRINKLERS USED WITH FIXEDGLAZED ASSEMBLIES TO PROVIDE ANALTERNATIVE TOA FIRE-RESISTANCE-RATED WALLASSEMBLY AC385-1014-R1(“AC385 Assembly”)Hearing: Tuesday, February 10, 2015Los Angeles, California

Dear ICC Evaluation Committee Members:

The Alliance of Primary Fire Rated Glazing Manufacturers (“Alliance”) opposes the proposedrevisions to Section 5.1.11 of AC385. (The proponent of these revisions, Tyco Fire ProductsResearch and Development, will be referred to throughout as “Proponent”).

In February of 2013, the ICC-ES Technical Committee formulated and specifically addedSection 5.1.11 to AC385 following extensive debate. Then, in October 2014, Proponentattempted to revise or delete Section 5.1.11. Following a second extensive debate, the TechnicalCommittee declined to modify or delete Section 5.1.11.

Now, Proponent seeks, again, to eliminate restrictions imposed by Section 5.1.11 of AC385.

The proposed revisions to Section 5.1.11 are fundamentally flawed and should be disapproved.

In short, Proponent would change Section 5.1.11 in two ways. First, Proponent says that Section5.1.11’s prohibition on the use of AC385 to protect exit areas should be eliminated whentempered glass is used in an AC385 Assembly if an undefined “fire area” in the building issprinklered. Second, Proponent says that Section 5.1.11’s prohibition on the use of AC385 to

AC385-0215-R2 #5

January 9, 2015Page 2

protect exit areas should be eliminated whenever ceramic glass (as opposed to tempered glass) isused in an AC385 Assembly, even when no other area of the building is sprinklered.

As to that aspect of Proponent’s proposal when tempered glass is used in an AC385 Assembly,Proponent offers no details whatsoever as the “fire area” it proposes adding to AC385. Withoutknowing anything about the proposed “fire area,” the Alliance is (unfairly) unable to analyze orrespond.1

As to that aspect of Proponent’s proposal when ceramic glass is used in an AC385 Assembly,Proponent attempts to justify changing Section 5.1.11 of AC385 on the basis of “tests” done atSouthwest Research Institute (“Southwest”) and the ability of ceramic glass to withstand“thermal shock” better than tempered glass.

The Southwest testing referenced by Proponent does nothing to justify any changes to Section5.1.11. Likewise, the response of ceramic glass to thermal shock provides no basis to reviseSection 5.1.11 as suggested by Proponent.

The Southwest Testing

Although Proponent describes the testing conducted at Southwest as “full-scale,” the testingdone at Southwest falls far short of anything that could be considered a “full-scale” fire test. Infact, Proponent admits that the testing conducted at Southwest consisted of nothing more than “a5-minute flammable liquid pool fire.”2

At a minimum, a “full-scale” fire test commonly refers to exposing a building material to acontrolled fire in a specially designed furnace in accordance with the terms of a specific teststandard.

The “testing” on which Proponent proposes revising AC385 did not involve a controlled fire in atest furnace. Instead, it involved little more than putting a piece of ceramic glass in a pool offlammable liquid and setting it on fire!

Likewise, this “testing” is not based on any known test standard. Instead, the only informationoffered by Proponent about the “test” is that it involved a 5-minute exposure to a ceramic glass.

To be considered safe, the fire testing of building materials must be uniform, repeatable andconducted pursuant to a recognized test standard. This is clearly expressed in the very firstsentences of ASTM E119-14:

1 ICC-ES Staff recognizes this deficiency at p.2 of its December 19, 2014, letter at p.2: “A definition of the ‘firearea’ needs to be added to the criteria … [including] the area of design or protection area of coverage.”

2 Proponent Letter to ICC-ES dated November 11, 2014, p. 1, para 4.

January 9, 2015Page 3

The performance of walls, columns, floors and other building membersunder fire-exposure conditions is an item of major importance in securingconstructions that are safe …. To do this it is necessary that the fire-resistive properties of materials and assemblies be measured and specifiedaccording to a common standard expressed in terms that are applicablealike to a wide variety of materials, situations, and conditions of exposure.3

(Emphasis added).

Numerous deficiencies are evident in the Southwest tests underlying this proposal to reviseAC385. For example: fire testing using a pool of flammable liquid is not repeatable; it does notprovide uniform fire exposure; it does not ensure any particular temperature exposure at anyparticular time; it does not expose the test material to the sprinkler system at issue in AC385 atany time during or after the fire; and, it does not impose any of the fire stresses experienced in afurnace during a standard ASTM E119 test.

The Southwest “testing” referenced by Proponent does not support any revision to Section5.1.11 of AC385.4

Thermal Shock

Thermal shock occurs when different parts of a material are heated or cooled differently. Thestress resulting from the thermal differential can exceed the strength of the material, causing it tofail.

The ability of ceramic glass and tempered glass to withstand thermal shock are different.Because of this, Proponent claims that an AC385 Assembly using ceramic glass should beallowed to protect the exit areas of a building, even if no other areas of the building aresprinklered. The Proponent claims that the ability of ceramic glass to withstand thermal shock issufficient to assure its safe performance in the event it is exposed to forces from backdraft orflash over from fire in an unsprinklered part of the building.

Unfortunately, while ceramic glass may have a higher resistance to thermal shock than temperedglass, it has no greater resistance than ordinary window glass to breakage from the impact forceof a backdraft, flash over or other events experienced in a fire.

Conclusion

The proposal related to the use of tempered glass to protect exit areas in a sprinklered “fire area”lacks any definition that can be properly analyzed or assessed and should be rejected. The

3 “Standard Test Methods for Fire Tests of Building Construction and Materials”

4 If UL finds this testing sufficient to warrant a change to the “pony wall” provisions of its listing, that is a matterstrictly between UL and Proponent. At best, it may support some sort of change to the “pony wall” provisions ofAC385 found in Section 5.1.8. However, this testing does nothing to warrant a change to Section 5.1.11 of AC385.

January 9, 2015Page 4

proposal related to the use of ceramic glass to protect exit areas in unsprinklered buildings alsolacks adequate testing or technical merit and should be rejected.

Proposed revisions to Section 5.1.11 of AC385 should be disallowed.

Respectfully submitted,

/s/ Thomas S. Zaremba

Thomas S. Zaremba

8955828 _1