Safety Recommendation A-00-091

We note that, on December 17, 2013, SAE International issued a revision to Aerospace Standard (AS) 8036, “Cargo Compartment Fire Detection Instruments,” that addressed the issues in this recommendation. We also note that, on August 19, 2014, you issued revised Technical Standard Order (TSO)-Cl “Cargo Compartment Fire Detection Instruments,” which incorporated the revisions to AS 8036. Your action satisfies Safety Recommendation A-00-91, which is classified CLOSED—ACCEPTABLE ACTION.

-From Michael P. Huerta, Administrator: In March 20 II. the Federal Aviation Administration (FAA) submitted a request to the Society of Automotive Engineers (SAE) to develop an improved Aerospace Standard (AS) 8036, Cargo Compartment Fire Detection Instruments. The purpose of the revision was to address resistance to false alarms from smoke detectors due to dirt, dust, and moisture.

In May 2011, SAE tasked the A-4 Aircraft Instruments Committee to develop a minimum performance standard for false alarm resistance. AS8036 revision A was released on December 17, 2013.

Subsequent to the release of AS8036 revision A. the FAA made revisions to Technical Standard Order (TSO)-C I. Cargo Compartment Fire Detection Instruments with an effective date of August 19, 2014. TSO-Cle can be found at the following web address: http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgTSO.nsf/O/f29f56acb0ecb Ifl86257d3a0060a0c0/$FI LE/TSO-C I E.pdf.

I believe that the FAA has effectively addressed this safety recommendation and consider our actions complete.

We have previously said that your June 2000 publication of a report titled “Aircraft Cargo Compartment Smoke Detector Alarms Incident on U.S. Registered Aircraft” satisfied the first part of this recommendation (to document the extent of false indications for cargo compartment smoke detectors). We are aware that, more recently, you have been working with SAE International to revise Aerospace Standard (AS) 8036, “Cargo Compartment Fire Detection Instruments” and that you planned to revise Technical Standard Order (TSO)-Cld “Cargo Compartment Fire Detection Instruments,” based on the revisions to AS 8036.

Your letter described some recent delays that SAE has encountered in releasing the revised AS 8036, but that it is working to resolve as soon as possible. Although this recommendation is over 13 years old, we are encouraged that lately you have been making significant progress in addressing it. Accordingly, pending the completion of revisions to AS 8036 and the incorporation of these revisions into the TSO in a timely manner, Safety Recommendation A-00-91 remains classified OPEN—ACCEPTABLE RESPONSE.

-From Michael P. Huerta, Administrator: As mentioned in our previous letter to the Board, dated August 3, 2012, the Federal Aviation Administration (FAA) plans to revise Technical Standard Order (TSO)-C I d, Cargo Compartment Fire Detection Instruments. In March 201 1, the FAA submitted a request to the Society of Automotive Engineers (SAE) to develop an improved Aerospace Standard (AS) 8036, Cargo Compartment Fire Detection Instruments. The revised AS8036 standard will address resistance to false alarms from smoke detectors due to dirt, dust, and moisture.

In May 2011, SAE tasked the A-4 Aircraft Instruments Committee to develop a minimum performance standard (MPS) for false alarm resistance. Since that time, the committee has met regularly to develop the MPS.

The last update to the Board discussed the revision to SAE AS8036 and the corresponding revision of TSO-Cld, which was planned for release by May 2013. The release of revised AS8036 has been delayed for the following reasons: • The working group added and developed two additional criteria after determining the original approach was not adequate; • The initial dust concentration criteria for testing false alarm resistance was determined to be overly conservative and additional development testing had to be introduced; • A second test option needed to be developed after finding the initial test developed for the ambient light condition was not readily available in North America; and

• In February 2013, the working group released a revised draft of AS8036 for SAE approval, which identified issues with one of the testing procedures.

The FAA is awaiting SAE's near-term completion and publication of AS8036, after which we will complete TSO-C l d.

I will keep the Board informed of the FAA's progress and provide an updated response by October 2014.

Thank you for the update on the activities of SAE International, in response to a request from the FAA to revise Aerospace Standard (AS) 8036, “Cargo Compartment Fire Detection Instruments,” and the FAA’s plan to revise Technical Standard Order (TSO)-Cld “Cargo Compartment Fire Detection Instruments,” based on the revisions to AS 8036. Although this recommendation is almost 13 years old, we are encouraged that the FAA has recently been making significant progress on completing the requested work. Accordingly, pending the completion of revisions to AS 8036 and the incorporation of these revisions into the TSO, Safety Recommendation A-00-91 remains classified OPEN—ACCEPTABLE RESPONSE.

-From Michael P. Huerta, Acting Administrator: On March 31, 2011, the Federal Aviation Administration (FAA) submitted a request to the Society of Automotive Engineers (SAE) to develop an improved standard to address resistance to false alarms from smoke detectors due to dust, dirt, and moisture.

Aerospace Standard (AS) 8036, Cargo Compartment Fire Detection Instruments, addresses these standards. As described in earlier correspondence, the FAA's current course of action is to revise Technical Standard Order (TSO)-Cld, Cargo Compartment Fire Detection Instruments, using an improved AS8036. In May 2011, SAE tasked a committee to develop a minimum performance standard (MPS) for false alarm resistance. Since that time, the committee has been meeting regularly to develop the MPS and has identified the tests for smoke detector false alarm resistance that will enable an effective update of AS8036.

The test methods are currently being determined, and success criteria for the tests still need to be defined. After the SAE publishes the updated document, the FAA expects to have the draft revision to TSO-Cld completed by the end of May 2013.

I will keep the board informed of the FAA's progress and provide an updated response by July 2013.

The NTSB notes that the FAA was unsuccessful in acquiring the necessary data from smoke detector manufacturers to develop definitive pass-fail criteria for completing revisions to Technical Standard Order (TSO)-C1d. We also note that the FAA has instead requested that the Society of Automotive Engineers task a committee to develop an improved test standard for Aerospace Standard 8036, “Cargo Compartment Fire Detection Instruments,” to address the effects of dust, dirt, and moisture on false alarms and current minimum performance standards.

We appreciate the FAA’s continued efforts to address this safety issue. Pending the development of improved test standards to ensure that cargo compartment smoke detectors have as low a false-alarm rate as practical, consistent with the need to alert the crew to the presence of a fire before it leads to a catastrophe, Safety Recommendation A-00-91 remains classified OPEN—ACCEPTABLE RESPONSE.

CC# 201100281: - From J. Randolph Babbitt, Administrator: In the Federal Aviation Administration's (FAA) previous response dated July 23, 2009, we concluded that most false alarms on cargo smoke detectors are caused by dust and dirt particles that accumulate in the cargo compartments and interfere with the proper operation of the smoke detectors. We stated that we had begun developing a revision to Technical Standard Order (TSO) C1d, Cargo Compartment Fire Detection Instruments, to include further testing on smoke detector units in dust, dirt, and high humidity environments.

We requested detailed information from the smoke detector manufacturers to develop definitive pass-fail criteria for the TSO. However, the manufacturers refused to share this data, which they consider proprietary. As a result, we are not yet able to complete our revision to TSO-C1d. On March 31, 2011, we sent a task request (enclosed) to the Society of Automotive Engineers (SAE) to task a committee to develop an improved test standard for Aerospace Standard (AS) 8036, Cargo Compartment Fire Detection Instruments. We are requesting a revision of AS 8036 to address the effects of dust, dirt, and moisture on false alarms in addition to current minimum performance standards.

Based on our experience, we anticipate that industry will need at least two years to come to a consensus on a revised test standard. If SAE is unable to provide a revised test standard, the FAA will consider our own revision to the TSO. I will keep the Board informed of the FAA"s progress on this safety recommendation and provide an update by July 2012.

The FAA previously informed the NTSB that the FAA’s William J. Hughes Technical Center had published a report on the number of cargo compartment false alarms, titled Aircraft Cargo Compartment Smoke Detector Alarms Incidents on U.S. Registered Aircraft, 1974-1999, which satisfied the first part of the recommendation. As part of this effort, the technical center’s researchers established a goal of developing more reliable aircraft smoke and fire detection systems to reduce the false-alarm rate of cargo compartments. On July 25, 2001, the NTSB responded that the report documented an alarmingly high false-alarm rate and indicated that the rate appeared to be increasing.

On October 15, 2003, the FAA provided an update of its research on detector technologies to reduce the rate of cargo compartment smoke alarm false alerts. On February 9, 2004, the NTSB responded that the research described appeared to be promising but that the evacuation study that prompted this recommendation had found noteworthy differences in the false-alarm rates of some transport-category aircraft compared to those of other aircraft. The study examined the FAA’s Service Difficulty Reporting (SDR) system and found 30 reports of false cargo smoke indications involving Saab 340s and 15 reports involving Canadair Regional Jets (CRJ) between October 1998 and November 1999.

In comparison, for the substantially larger fleet of all Boeing aircraft, the SDR contained only 16 false smoke indications for the same period. The NTSB indicated that the FAA’s activities to develop and test a system to detect cargo compartment fires and reduce the false-alarm rate without sacrificing early detection of smoldering fires were responsive to this recommendation, but, as part of that effort, the FAA should determine the cause of the higher false-alarm rates in Saab 340 and CRJ aircraft compared to that of the fleet of Boeing aircraft.

Pending the development and adoption of improvements to cargo smoke detection systems to reduce the rate of false alarms, Safety Recommendation A-00-91 remained classified Open Acceptable Response.

In response to the NTSB’s concern about the high false-alarm rate in Saab 340 and CRJ aircraft, the FAA contacted Saab Aerotech and Bombardier, the manufacturers, to inquire about false indication rates, their respective causes, and measures the manufacturers were taking to lower the rates. While neither company explained the reasons for the false alarms, both replaced older ionization detectors with improved optical smoke detectors. The companies reported an order of magnitude decrease in false-alarm rates with the use of the improved optical smoke detectors.

Letter Mail Controlled 8/5/2009 11:38:54 AM MC# 2090501: - From J. Randolph Babbitt, Administrator: In the Board’s previous response, it requested the Federal Aviation Administration determine the cause of the higher false alarm rate in Saab 340 and Canadair Regional Jet (CRJ) aircraft. The FAA contacted Saab Aerotech and Bombardier; the manufacturers of the Saab 340 and CRJ, to inquire about the false smoke alarm/indication rates, their respective causes, and means they were taking to lower the false alarm rate. While neither company explained the reason for the false alarms, they have made changes to improve the reliability and decrease the false alarm rate.

This was accomplished by replacing older ionization detectors with improved optical smoke detectors. The companies reported an order of magnitude decrease in false alarm rates by using the improved optical smoke detectors. In our research to determine the reason for false alarms, we reviewed reports from the FAA Technical Center, service difficulty reports, and data from other government agencies such as the National Institute of Standards and Technology. The results showed that most false alarms are caused by dust and dirt particles that accumulate in the cargo compartments, which interfere with proper operation of ionization detectors more so than the optical smoke detectors.

In general, we have learned ionization detectors respond more quickly to flaming fires, optical sensors respond more quickly to smoldering fires, and a dusty or high humidity environment contributes to smoke detector false warnings. The FAA standards on fire protection of cargo compartments on commercial transport category airplanes are based on quickly detecting small smoldering fires before they grow to a conflagration that would negate the fire protection features within the compartment or damage flight critical systems outside of the compartment. FAA regulations require a means to detect smoke/fire within one minute, but do not specify the type of sensor.

Our goal is to develop objective standards that encourage industry to look at any means to determine the presence of a fire with the hope of diminishing the false alarm rate. We continue to support this effort through the FAA's Technical Center. A decrease in the smoke detector false alarm rate is possible through the use of advanced sensor technologies and a change in testing criteria. These technologies include multiple optical channels, which can discriminate between smoke particles and dust, and microprocessor software, which can identify and disregard false alarms caused by radio frequency and electromagnetic interference.

The Safety Board notes with interest the recently reported results of the FAA research project which showed that a combination of various gas signature detectors would readily detect "flaming fires" with a reduction of false alarms. However, the initial stage of a typical cargo fire is a smoldering fire, and particulates compose the majority of smoldering fire effluents. Particulates readily trigger typical fire detection systems now in use in cargo compartments. In contrast, gas detectors do not detect particulates well. While use of these advanced sensors would decrease the false alarm rate, this would be achieved at the sacrifice of early detection of a smoldering fire.

The FAA believes that fire protection of cargo compartments should be predicated on the quick determination of the existence of a smoldering fire before it can grow to a conflagration. The FAA is working with airframe manufacturers and foreign authorities to develop a system of multiple particulate sensors with varying detection time triggering thresholds which use a voting hierarchy alarm architecture. The FAA believes that such a system may reduce false alarms of typical cargo compartment detector systems.

The Safety Board notes that the evacuation study, which prompted this recommendation, found noteworthy differences in the false alarm rate for some regional "commuter aircraft" compared to other transport category aircraft. The study examined the FAA's Service Difficulty Reporting (SDR) system and found 30 reports of false cargo smoke indications involving Saab 340s and 15 reports involving Canadair Regional Jets (CRJ) for the period from October 1998 to November 1999. In comparison, for the substantially larger fleet of all Boeing aircraft, the SDR contained only 16 false smoke indications for the same period. Thus, the Saab 340 and CRJ have a substantially higher false alarm rate.

The FAA's plan to develop and test a system to detect cargo compartment fires and reduce the false alarm rate without sacrificing early detection of smoldering fires with primarily particulate effluents is responsive to this recommendation. The Board believes that as part of this effort to develop more reliable smoke alarms, the FAA should determine the cause of the higher false alarm rate in Saab 340 and CRJ aircraft compared to the fleet of Boeing aircraft. Pending the development and adoption of improvements to cargo smoke detection systems to reduce the rate of false alarms, Safety Recommendation A-00-91 remains classified OPEN -- ACCEPTABLE RESPONSE.

Letter Mail Controlled 11/5/2003 8:16:20 AM MC# 2030535 - From Marion C. Blakey, Administrator: The Federal Aviation Administration (FAA) has reviewed the material from the William J. Hughes Technical Center's Research Project Description (RPD) Number 558, Aircraft Fire Safety and Cabin Safety, Project 061-110, Fire Research and Safety on Multiple Sensor Technology. The intent of the study was to develop information regarding detector response characteristics using certain gas parametric signatures (CO, CO2, HC, 02, etc.) that newer technology sensors use to detect fires.

The results showed that the use of a combination of various gas signature detectors would readily detect "flaming fires" with a reduction of false alarms. However, the characteristic of the initial stage of a typical cargo fire is a "smoldering fire." The majority of smoldering fire effluents are particulates, which readily trigger typical fire detection systems now in use in cargo compartments. The gas detectors do not detect particulates well. The use of these advanced sensors would likely decrease the false alarm rate but sacrifice early detection of a smoldering fire.

The FAA's position on fire protection of cargo compartments has been predicated on the quick determination of the existence of the fire before it can grow to a conflagration that would negate the fire protection features within the compartment or damage flight critical systems outside of the compartment. The FAA has characterized the "small smoldering suitcase fire" as the type of fire to be detected. The FAA believes that a reduction in false alarms of typical cargo compartment detector systems may be possible by designing a system of multiple particulate sensors with varying detection time triggering thresholds, which use a voting hierarchy alarm architecture.

The FAA is working with airframe manufacturers and foreign authorities on this issue. I will keep the Board informed of the FAA's progress on this safety recommendation.

The Safety Board reviewed the report, Aircraft Cargo Compartment Smoke Detector Alarms Incident on U.S. Registered Aircraft, which documents an alarmingly high false alarm rate on cargo compartment smoke detectors and indicates that the rate of false alarms appears to be increasing. The completion of this report satisfies the first part of this recommendation, to "document the extent of false indications for cargo smoke detectors on all airplanes."

The FAA research program to produce a standard means of testing detectors to demonstrate compliance with regulatory response requirements and to develop reliable aircraft smoke and fire detection systems that reduce the cargo compartment false alarm rate is beneficial and responsive to the recommendation. Given the high false alarm rate found by the FAA and the Safety Board, the FAA should issue appropriate guidance material and certification standards for these alarms once the current research program is completed. The Board asks that the FAA keep it advised on the research program and planned actions after the research results become available.

Pending completion of the recommended actions, Safety Recommendation A-00-91 is classified "Open--Acceptable Response."

Letter Mail Controlled 11/14/2000 4:39:44 PM MC# 2001688 - From Jane F. Garvey, Administrator: In June 2000, the FAA's William J. Hughes Technical Center published a report entitled "Aircraft Cargo Compartment Smoke Detector Alarms Incident on U.S. Registered Aircraft" on the number of cargo compartment false alarms. This effort is part of Research Project Description Number 558, Aircraft Fire Safety and Cabin Safety, Project 061-110, Fire Research and Safety. The intended goal of developing more reliable aircraft smoke and fire detection systems is to reduce the cargo compartment false alarm rate from 90 percent to 10 percent, with an intended completion date of 2002.

The Technical Center report, which documents the number of incidents of cargo compartment smoke detector alarms on U.S. - registered aircraft operating under 14 CFR Parts 121 and 135 from 1974 through 1999, should satisfy the intent of this safety recommendation. A copy of the Technical Center report is available at http://www.fire.tc.faa.gov under "Reports." The aviation industry is constantly improving the reliability of smoke and fire detectors to reduce the number of false alarms.

This is accomplished by improving sensor technology using a combination of different types of sensors (like smoke and particulate, temperature, relative humidity, oxygen, and carbon monoxide) and designing dual path systems. In support of this effort, the FAA is sponsoring a program to produce a standard means of testing detectors to demonstrate compliance with regulatory response requirements and to develop reliable aircraft smoke and fire detection systems that reduce the cargo compartment false alarm rate. Research is being conducted at the FAA's William J.