Skip Ribbon Commands
Skip to main content
Safety Recommendation Details

Quick Launch
Safety Recommendation A-04-035
Details
Synopsis: On June 4, 2002, Spirit Airlines flight 970, a McDonnell Douglas MD-82 airplane, experienced a gradual loss of power in both of its Pratt & Whitney JT8D-219 engines and an activation of the aural stall warning and stickshaker while in cruise flight at flight level (FL) 330 about 20 miles from Wichita, Kansas. When the stall warning activated, the pilots disengaged the autopilot, turned on the engines’ ignition, activated the engines’ anti-ice system, and initiated a descent. The pilots reported later that they shut down the right engine when its exhaust gas temperature (EGT) increased to about 600° Celsius (C) and were able to restart it again on the second attempt at about 17,000 feet. They reported that the left engine recovered on its own shortly thereafter. The flight then diverted to the Wichita Mid-Continent Airport, Wichita, Kansas, where it landed safely. The flight was operating on an instrument flight rules flight plan under the provisions of 14 Code of Federal Regulations Part 121 as a regularly scheduled passenger flight from Denver International Airport (DEN), Denver, Colorado, to Fort Lauderdale-Hollywood International Airport, Fort Lauderdale, Florida. The 2 pilots, 4 flight attendants, and 105 passengers on board were not injured.
Recommendation: TO THE FEDERAL AVIATION ADMINISTRATION: Actively pursue research with airplane and engine manufacturers and other industry personnel to develop an ice detector that would alert pilots of inlet pressure probe icing and require that it be installed on new production turbojet airplanes, as well as retrofitted to existing turbojet airplanes.
Original recommendation transmittal letter: PDF
Overall Status: Closed - Acceptable Alternate Action
Mode: Aviation
Location: WICHITA, KS, United States
Is Reiterated: No
Is Hazmat: No
Is NPRM: No
Accident #: CHI02IA151
Accident Reports:
Report #: None
Accident Date: 6/4/2002
Issue Date: 4/29/2004
Date Closed: 2/11/2015
Addressee(s) and Addressee Status: FAA (Closed - Acceptable Alternate Action)
Keyword(s): Icing, Weather

Safety Recommendation History
From: NTSB
To: FAA
Date: 2/11/2015
Response: In our December 10, 2009, letter to you about Safety Recommendation A-06-59, we said that, if the final rule regarding icing conditions included certification standards that would effectively prevent the buildup of ice accretions disturbing the airflow to an engine, the rule’s publication would constitute an acceptable alternate response. In our December 9, 2011, letter to you about Safety Recommendation A-04-35, we acknowledged the considerable technical challenges associated with the recommended development of an ice crystal detection system that could be installed inside an engine, and we agreed that the implementation of robust engine design standards that would render engines immune from the effects of ice crystals would constitute an acceptable alternate response. Your November 4, 2014, publication of the Airplane and Engine Certification Requirements in Supercooled Large Drop, Mixed Phase, and Ice Crystal Icing Conditions completes these actions. Accordingly, Safety Recommendations A-04-35 and A 06-59 are classified CLOSED—ACCEPTABLE ALTERNATE ACTION.

From: NTSB
To: FAA
Date: 12/9/2011
Response: The NTSB considered how this recommendation related to Safety Recommendation A-06-59, stated below, which was issued to the FAA as a result of our investigation of several high-altitude, dual-engine flameouts that occurred on Raytheon Beechjet 400-series airplanes powered by Pratt & Whitney Canada JT15D-5 turbofan engines. Recommendation A-06-59 is currently classified “Open—Acceptable Alternate Response,” pending issuance of a final rule based on the FAA’s June 29, 2010, notice of proposed rulemaking (NPRM), titled “Airplane and Engine Certification Requirements in Supercooled Large Drop, Mixed Phase, and Ice Crystal Icing Conditions.” Although these two recommendations are worded similarly, there are some subtle but important differences. The icing that caused the problems in the Spirit Airlines flight 970 incident and the Beechjet 400 incident occurred on different parts of the engine. Although both resulted in dual engine flameouts, the events developed differently. In the Spirit Airlines flight 970 incident, the ice built up around the Pt2 probe, causing a reduction in the Pt2 signal that is used to determine the engine pressure ratio (EPR). The incorrect reduction in the Pt2 signal led to an incorrect indication that the EPR was increasing. On the McDonnell Douglas MD-82 (the model aircraft in the Spirit Airlines flight 970 incident), the autothrottle system has an EPR limiting protection to prevent problems with the engine. In the incident, the EPR increased until reaching its limit, when the autothrottle system retarded the throttles. Because the autopilot had been set in altitude hold mode, as the airplane slowed from the retarded throttles, the airplane pitched up to maintain altitude. Finally, the engines stalled and flamed out from the blanked inlets and high aircraft angle of attack (AOA). In contrast, in the incidents that occurred on the Beechjet airplanes, the ice passed through the fan and partially melted from the temperature rise across the fan and began to adhere to the core stators. It is believed that the ice built up in a shape that followed the contour of the stator vane and did not affect the engine’s performance until the pilots changed the throttle position. In each of the Beechjet events, the throttle was retarded, changing the AOA over the stator vanes and blowing the ice off the vane and into the combustor, where the flame was quenched by the ice. Safety Recommendations A-04-35 and A-06-59 each reflect where the icing was occurring in the particular incidents/accidents involved. In the incident that resulted in Safety Recommendation A-04-35, the icing occurred on the Pt2 probe in the inlet. As a result, Safety Recommendation A-04-35 is worded “…would alert pilots of inlet pressure probe icing….” On the Beechjet incidents, the ice was passing through the fan and building up on the core stator vanes. Hence, Safety Recommendation A-06-59 is worded “to develop an ice detector that would alert pilots to internal engine icing….” In its October 24, 2011, letter, the FAA indicated that there are no detection systems in development for installation within an engine’s internal airflow path, and the FAA discourages detectors within an engine’s airflow path. Because considerable technical challenges are involved in developing and implementing ice crystal detection systems, the FAA stated that it will be several years before a durable and well-correlated design is available, and then it will be available for research purposes only. Because of these technical challenges, rather than pursue ice crystal detectors, the FAA is developing robust engine design standards as an alternative so that future engines will not adversely react to the ice crystal environment. Until the FAA issues the final rule, including the revised certification standards proposed in the FAA’s June 29, 2010, NPRM, the FAA is using certification papers that require engine manufacturers to consider the affects of the ice crystal environment during new-engine certification programs. The NTSB recognizes the considerable technical challenges associated with the development of an ice crystal detection system that could be installed inside an engine, and we agree that the implementation of robust engine design standards that will render engines immune from the effects of ice crystals constitutes an acceptable alternative response. Although the FAA indicated that it considers its actions in response to this recommendation to be complete, the FAA also stated that it expects to issue the icing final rule by February 2012. Issuance of that final rule will complete the FAA’s action in response to Safety Recommendation A-04-35, which is classified OPEN--ACCEPTABLE ALTERNATE RESPONSE pending that action.

From: FAA
To: NTSB
Date: 10/24/2011
Response: From J. Randolph Babbitt, Administrator: In the Board's July 20, 2010, letter to the Federal Aviation Administration (FAA), they acknowledged that the FAA does not plan to install ice detectors on new production or existing airplanes. This is partly due to diminishing number of engines in the JT3D and JT8D fleets. The Board asserted that there are at least 400 U.S.-registered airplanes that are equipped with JT8D-200 engines and that without an ice detector, there is no verifiable way for the crew to know whether there is an ice buildup on the engine inlet probes. The Board maintains that an ice detection system is still needed to alert pilots of the threat of high altitude ice crystals. The FAA believes that activation of the engine anti-ice when operating near convective weather is more appropriate than providing an ice detection system for the JT8D engines. The engine anti-ice system activates engine pitot probe heat and prevents the engine pitot probe from icing over, which precludes a potentially erroneous engine pressure signal. Additionally, the MD-80 flight crew operations manual requires engine anti-ice to be on during operations with visible moisture. The FAA has increased the awareness of the icing environment by pointing out the potential for mixed phase icing by issuing Flight Standards Information Bulletin for Air Transportation (FSAT) 04-02. This has significantly reduced the number of events in service. The worldwide engine power loss event rate due to ice crystals is approximately once per year since 200 I with subsequent power restoration in each event. The low rate of occurrence and subsequent outcomes constitute an acceptable level of risk for JT8D powered airplanes. Therefore, the FAA maintains that flight crew adherence to the flight crew operations manual procedures can eliminate these engine pitot probe freezing events and no action to change the airplane or engine design is necessary. On July 20, 2006, the FAA issued Safety Alert for Operators 06-006 and superseded FSAT 04-02 with Information for Operators 08033 (dated May 16, 2008) emphasizing environmental awareness. Additionally, we revised Advisory Circular (AC) 91-74A, Pilot Guide: Flight in Icing Conditions, to include mixed phase icing. Air France, in cooperation with Environmental Canada and Boeing, developed flight crew training on the ice crystal environment and made it available to industry for use. There have been other engine installations that showed susceptibility to ice crystals. In each case, the effects of ice crystals are unique to that engine installation. The FAA evaluates the specific details of each case and implements corrective actions appropriate to the situation. Those actions include: • Design changes mandated by airworthiness directives; • Changes to the airplane flight manual; and • Reinforced awareness of the concern for high altitude ice crystals. The airplane manufacturers have also increased awareness of the operating environment through workshops and publications. We will continue to monitor in-service events and will take mitigating actions as appropriate. For example, the Pratt & Whitney Canada Model JTl5D series engine installed on the Hawker Beechcraft Model 400 Beechjet airplane also showed a susceptibility to ice crystals. Activation of the engine's anti-ice system can prevent power loss from ice crystals on this engine as well. The increased awareness of the operating environment has also reduced the number of events for the JTl5D engines. In our response to a similar safety recommendation, A-06-59, which addresses the dual engine power loss event of a JTI5D powered 400 Beechjet airplane, we described our continued research to better define the ice crystal environment and develop methods to measure it. Our approach to safety recommendation A-06-59 includes monitoring two competing detection systems currently in development that are intended for use on research aircraft. These systems are installed on the aircraft external to the engines and allow characterization of the high altitude ice crystal environment. There are no detection systems in development for installation within the engine's internal airflow path. The FAA does not encourage detectors within the engine's airflow path for safety reasons. Considerable technical challenges exist in developing and implementing external ice crystal detection systems. It will be several years before a durable and well-correlated design will be available for research purposes only. The FAA stated that even though an early warning system in the form of an ice crystal detector would be desirable, it is not a practical solution to preclude future ice crystal ingestion engine upsets as a significant amount of ice can accrete within minutes of encountering an ice crystal environment. Due to the technical challenges, rather than pursue ice crystal detectors, we are driving toward robust engine design standards as an alternative so that future engines will not adversely react to the ice crystal environment. Our approach to address the issues raised by safety recommendation A-04-35 for the dual engine power loss event of the JT8D powered MD-80 is consistent with the approach we have taken to address safety recommendation A-06-59. Since the last update to the Board, the FAA has issued the notice of proposed rulemaking (NPRM) for Airplane and Engine Certification Requirements in Supercooled Large Drop, Mixed Phase, and Ice Crystal Icing Conditions. The NPRM included a proposal that engines, including engine inlet probes, must demonstrate robust designs by complying with revised certification standards, which include the ice crystal environment. We are now in the process of addressing public comments to the NPRM and intend to issue the final rule by February 2012. The FAA also issued papers that require engine manufacturers to consider the effects of the ice crystal environment during recent certification programs. We will continue to issue papers until the final rule with the revised certification standards has been issued. In summary, the FAA has reinforced flight crew awareness of the ice crystal environment and its effect on performance and has also found that there are significant technical challenges in developing an ice crystal detector for operational use. The FAA continues to apply issue papers on current certification projects and in the future will apply upcoming certification standards on new airplane designs. Therefore, the FAA believes that retrofitting the turbojet powered transport airplanes with ice crystal detectors is not practical and that requiring future engine probes to operate without adverse effect in the ice crystal environment fully addresses the intent of the safety recommendation. I believe that the FAA has effectively addressed this safety recommendation, and I consider our actions complete.

From: NTSB
To: FAA
Date: 7/20/2010
Response: The FAA indicated that it does not plan to install ice detectors on new production or existing airplanes, in part because of the diminishing number of engines in the JT3D and JT8D fleets. Although the NTSB is aware that most engines in the JT3D and the JT8D-I through -17AR fleets are no longer in service, the JT8D-200 engines continue to power at least 400 airplanes in the U.S. registry.' The NTSB points out that, without an ice detector being installed on these airplanes, there is no verifiable way for the crew to know whether there is an ice buildup on the engine inlet pressure probes. Such information is important; it probably would have prompted the accident flight crew to activate the engine anti-ice system before entering icing conditions. Although the FAA's Flight Standards Information Bulletin for Air Transportation 04-02, "High Altitude Icing Conditions," emphasizes the need for pilots to maintain vigilance for signs of high altitude icing conditions and warns them of the effect that these conditions can have on airplane and engine performance, an ice detection system is still needed to alert pilots of the threat of high altitude ice crystals. Accordingly, pending the FAA's reconsideration of its position on this issue, our review of the FAA's plan to develop an ice detector that would alert pilots of inlet pressure probe icing, and implementation of a requirement for such devices in newly manufactured and existing aircraft, Safety Recommendation A-04-35 is classified OPEN – UNACCEPTABLE RESPONSE.

From: FAA
To: NTSB
Date: 11/23/2009
Response: Letter Mail Controlled 12/3/2009 3:42:11 PM MC# 2090707 - From J. Randolph Babbitt, Administrator: The following information highlights the research and actions taken by the Federal Aviation Administration regarding this recommendation: Ice Protection Harmonization Working Group, Recommendation In December 2005, the Aviation Rulemaking Advisory Committee's (ARAC) Ice Protection Harmonization Working Group (HWG), supported by the Flight Test HWG, the Powerplant Installation HWG, and the Engine HWG, completed their final report on recommended rulemaking and advisory material related to supercooled large droplet (SLD) and ice crystal/mixed phase conditions. The report included recommendations for a new Appendix14- to CFR part 25 defining an SLD environment and to 14 CFR part 33, addressing ice crystal/mixed phase conditions. Additionally, the report included recommendations addressing 14 CFR part 25 aircraft performance and handling qualities, engine installation effects, ice protection system requirements, and 14 CFR part 33 engine requirements. In March 2006, we received the approved report from the ARAC. A copy of that report was provided in our most recent response to recommendation A-96-54. With the ARAC's recommendations, the FAA drafted revisions to the aircraft and engine icing regulations and policy, to address engine operation in ice crystal conditions. The FAA believes, as indicated in our May 2009 response to the Board's safety recommendation A-06-59, that engines must demonstrate robust designs by complying with revised certification standards, which include the ice crystal environment. While there is an engine icing draft notice of proposed rulemaking (NPRM) in process, our research does not support the need for an inlet pressure probe ice detector and, therefore, the NPRM does not address it. There are two different types of engine inlet probes; PTZ probes mounted in the engine inlet nose dome and conventional PT2 probes mounted on the diffuser wall of the inlets. The engine inlet PT2 probe was the type found on the incident aircraft and is found on only two engine installations. The JT8D engine is installed on DC-9lMD-80, Boeing 727, and Boeing 737-200 airplanes, and the JT3D engine is installed on DC-8, Boeing 707, and Boeing 720 airplanes. Examination of these current airplane fleets indicates that the average age of an airplane with JT8D engines is approximately 27 years old and an airplane with JT3D engines is approximately 37 years old. We recognize that Pn probe icing in the engine inlet nose dome may be caused by impingement of ice/slush from the ground or by snow/water entering the inlet, freezing, and proper engine anti-ice procedures not being used. We now know that unexpected in-flight ice crystal ingestion in high altitude environments can also cause problems with the PT2 probe when flight crews are not vigilant. Over 200 engine icing events were reviewed by the Ice Protection HWG. Of the 11 reported engine events related to PT probe icing, all occurred on the ground with the exception of 2 events. One of those two events occurred in-flight during climb at 22,000 feet, and the other event had no flight condition data. Neither of these event reporls contained meteorological information. Unlike the specific engine inlet nose dome PT2p robe, most of the conventional PT?p. robes are heated, and to our knowledge, none have exhibited adverse effects to ice crystals. In addition, many electronic engine control systems use air data system total pressure inputs for thrust control and scheduling with the engine's pressure probe perCor111ing a back-up function. For ice crystal detectors to be effective, they would have to be mounted at the point of accretion. This can be difficult to predict and varies among engine designs. This creates considerable technical challenges for the development and implementation of ice detection systems that would be effective over a wide range of airspeeds and in different icing environments. The major obstacles to these systems include accuracy, airspeed limitations, cost, and durability. Our research efforts into engine ice crystal detectors will continue, but realistically it will be several years before a durable and well correlated design will be available and generally accepted by the scientific community for use on research aircraft and ground-based simulation facilities. As indicated in the Board's initial letter, the pilots of Spirit Airlines flight 970 had numerous visual cues that should have alerted them to the progressive loss of engine power and impending airplane stall. The Board also noted that despite the guidance provided in the airplane's Flight Crew Operating Manual (FCOM), the infrequency with which high altitude ice crystals impact engine operation may result in flight crews not fully understanding the risk associated with high altitude ice crystals and their effect on flight operations. As a result of these missed cues, despite the FCOM guidance, the FAA issued Flight Standards Information Bulletin for Air Transportation (FSAT) 04-02, High Altitude Icing Conditions. FSAT 04-02 emphasized the need to maintain vigilance for the signs of high altitude icing conditions, the effect these conditions can have on airplane and engine performance, and the need for the appropriate use of the engine anti-icing system. This FSAT has since been superseded by Information for Operators (InFO) 08033, issued May 16, 2008 (enclosed). Given the service history of both types of PTZ probes; the diminishing population of the JT8DIJT3D engines; the technical challenges in developing and implementing a detector capable of reliably recognizing ice crystals; and the InFO providing high altitude icing guidance to operators, we do not believe it is necessary to install engine inlet probe ice crystal detectors, yet to be developed, on new production nor existing airplanes and we plan no further action. I believe the FAA has fully addressed this safety recommendation, and I consider our actions complete.

From: NTSB
To: FAA
Date: 10/21/2004
Response: The Safety Board notes that the FAA is reviewing the inlet probe icing service history for several PT2 mounting configurations as part of an effort to develop the recommended research program. After this review is completed, the FAA will develop a detailed plan to respond to this recommendation. Pending the development of a research plan, active pursuit of a research program to develop an ice detector that would alert pilots of inlet pressure probe icing, and requirement for such devices in newly manufactured and existing aircraft, Safety Recommendation A-04-35 is classified OPEN -- ACCEPTABLE RESPONSE.

From: FAA
To: NTSB
Date: 7/6/2004
Response: Letter Mail Controlled 7/16/2004 10:47:24 AM MC# 2040417 - From Marion C. Blakey, Administrator: The FAA is reviewing the inlet probe icing service history regarding engines with the PT2 probes mounted in the nose dome, as well as today's more conventional PT2 probe mounting configurations. This activity is needed prior to pursuing a new research program for development of a dedicated ice detector for alerting flightcrews to potential engine inlet pressure probe icing conditions. The FAA will complete its review by September 2004 and will provide the Board with its course of action to address this recommendation once the review is completed.