Location: Lethbridge, Alberta
Status: Investigation Complete
Reports Posted: 2012-12-10
Flight Safety Investigation Report (FSIR):
Epilogue:
During an air show practice at Lethbridge County Airport, CF188738 experienced a loss of thrust from its right engine while conducting a high angle of attack (AOA) pass at 300 feet (ft) above ground level (AGL). Unaware of the problem but feeling the aircraft sink slightly, the pilot selected maximum afterburner on both throttles in order to overshoot from the manoeuvre. The aircraft immediately started to yaw right and continued to rapidly yaw/roll right despite compensating control column and rudder pedal inputs.
With the aircraft at approximately 150 ft AGL and about 90 degrees of right bank, the pilot ejected from the aircraft. The aircraft continued in a tight descending corkscrew to the right prior to hitting the ground nose first.
The ejection system worked flawlessly, but the pilot was injured when he landed firmly under a fully inflated parachute.
The investigation revealed a number of factors that contributed to this occurrence. The engine malfunction was likely the result of a stuck ratio boost piston in the right engine main fuel control (MFC) that prevented the engine from advancing above flight idle when maximum afterburner was selected. The large thrust imbalance between the left and the right engines caused the aircraft to depart controlled flight and the aircraft was unrecoverable within the altitude available. The subtle nature of the engine malfunction that was not detected by the pilot when the overshoot was attempted.
In response to this occurrence, the Royal Canadian Air Force (RCAF) expedited the implementation of a program to upgrade all CF188 MFCs. Additionally, the RCAF made changes to the conduct of the CF188 air show routine by increasing the high AOA pass altitude from 300 feet AGL to 500 feet AGL, improving the air show training program and ensuring that both engines of the CF188 air show aircraft have upgraded MFCs.
FLIGHT SAFETY INVESTIGATION REPORT (FSIR)
| FILE NUMBER: | 1010-CF18738 (DFS 2-2-3) |
| DATE OF REPORT: | 26 September 2012 |
| AIRCRAFT TYPE: | CF188 Hornet |
| DATE/TIME: | 231610Z July 2010 |
| LOCATION: | Lethbridge County Airport (CYQL) Alberta |
| CATEGORY: | "A" Category Occurrence |
This report was produced under authority of the Minister of National Defence (MND) pursuant to section 4.2 of the Aeronautics Act, and in accordance with A-GA-135-001/AA-001, Flight Safety for the Canadian Forces.
With the exception of Part 1, the contents of this report shall only be used for the sole purpose of accident prevention. This report is released to the public under the authority of the Director of Flight Safety (DFS), National Defence Headquarters, pursuant to powers delegated to him by the MND as the Airworthiness Investigative Authority (AIA) of the Canadian Forces.
SYNOPSIS
During an air show practice at Lethbridge County Airport, CF188738 experienced a loss of thrust from its right engine while conducting a high angle of attack (AOA) pass at 300 feet (ft) above ground level (AGL). Unaware of the problem but feeling the aircraft sink slightly, the pilot selected military power (MIL) on both throttles to arrest descent. The aircraft continued to sink and the pilot selected maximum afterburner (AB) on both throttles in order to overshoot from the manoeuvre. The aircraft immediately started to yaw right and continued to rapidly yaw/roll right despite compensating control column and rudder pedal inputs. With the aircraft at approximately 150 ft AGL and about 90 degrees of right bank, the pilot ejected from the aircraft. The aircraft continued to yaw/roll right with its nose descending in a tight right descending corkscrew prior to hitting the ground nose first. The ejection and seat-man separation worked flawlessly, but the pilot was injured when he landed firmly under a fully inflated parachute. After landing, the parachute shroud lines became entangled around the pilot's left leg. The parachute re‑inflated before the pilot could release the Koch fittings and he was then dragged several hundred meters before it could be released. After the Koch fittings were released, first aid was administered to the pilot, who was subsequently transported to the Regional Hospital.
The occurrence was caused by a number of factors. The engine malfunction was likely the result of a stuck ratio boost piston in the right engine main fuel control (MFC) that prevented the engine from advancing above flight idle when maximum AB was selected. The large thrust imbalance between the left and the right engines caused the aircraft to depart controlled flight and the aircraft was unrecoverable within the altitude available. Contributing to the occurrence was the subtle nature of the engine malfunction that was not detected by the pilot when the overshoot was attempted.
1 FACTUAL INFORMATION
1.1 History of the Flight
1.1.1 CF188738 flew from Dawson Creek, British Columbia, to Lethbridge County Airport (CYQL), Alberta, on Thursday 22 July 2010. The pilot was scheduled to conduct an air demonstration during the Alberta International Air Show to be held on 24-25 July 2010. On Friday 23 July 2010, an air show practice flight was scheduled. Prior to flight, a performers’ brief was conducted by the air display director (ADD). The brief covered standard operating procedures (SOP) for items such as weather, show lines, emergency procedures, crash fire rescue locations, communications and schedule. The practice show was scheduled for an 1800 coordinated universal time (UTC) (1200 local) takeoff. Due to low clouds in the local area, the pilot elected to practice the low show.
1.1.2 At 1804 UTC, the air show airspace was turned over to the CF188 pilot who took off from runway 23 to conduct the low show. The first six and a half minutes of the flight were uneventful. Following a high speed right-hand (RH) steeply climbing turn to the south, the pilot reversed the turn to the left and established a flight-idle descending left-hand (LH) turn in order to reposition at 300 ft AGL for the slow-speed high AOA pass along the 500-ft show-line, parallel to runway 23 (see Annex A Figures 1 – 3). Setting up and executing the high AOA pass required significant pilot concentration and skill. The pilot’s attention was divided between cross-checking the flight information on the head up display (HUD)1, the velocity vector position on the radar display (selected on the right multi-function display (MFD)) and looking outside the aircraft to the left and to the right to ensure the aircraft was aligned with the show line2. As the aircraft reached approximately 150 knots calibrated airspeed (KCAS), the pilot advanced both throttles to sub-MIL3 power, without referencing the engine display (the normal method of power modulation for CF188 pilots), to arrest the deceleration and capture 25 degrees AOA. Upon reaching 25 degrees AOA, the pilot felt the aircraft sinking slightly. Thinking it was caused by surface wind induced turbulence, the pilot advanced the throttles to MIL power (See Annex A, Figure 4). The pilot felt that the aircraft was still sinking and noted that the aircraft was yawing back and forth slightly. Suspecting something was not right, the pilot decided to abort the manoeuvre (between two and four seconds after selecting MIL power) and selected maximum AB4 on both throttles to initiate a climb away (see Annex A, Figure 5) as he would have done at the end of the show line to reposition for the next manoeuvre. At no time did the pilot observe any aural or visual cues indicating a malfunction of any of the aircraft’s systems. Shortly after selecting maximum AB, the aircraft began to yaw to the right. The pilot tried to correct with compensating control column and rudder pedal inputs but the yaw rate increased rapidly and the aircraft started to roll to the right. The pilot briefly pushed forward on the control column in an attempt to reduce the AOA; however, this was ineffective and when the aircraft reached about 90 degrees of right bank at 150 ft AGL, he elected to eject from the aircraft (see Annex A, Figure 6). The aircraft continued to yaw/roll to the right in a tight right descending corkscrew. At approximately 1811 UTC it impacted the ground nose first in an inverted right wing down steep dive at approximately 25 degrees AOA (see Annex A, Figure 1 and Figure 7).
1.1.3 The pilot ejected from the aircraft on a trajectory parallel to the ground, facing approximately north. Following seat-man separation and parachute deployment, he was reoriented towards the south east (downwind) and had three seconds under a fully inflated parachute (5 seconds from initiating the ejection sequence). The parachute was still undergoing post-opening oscillations when the pilot landed in a non-optimal position. The pilot did not release the survival pack prior to touch down.
1.1.4 After landing, the parachute shroud lines became entangled around the pilot’s left leg. The parachute was subsequently re-inflated by the wind and he was dragged feet first towards the south-east (away from the crash site). The pilot had significant difficulties releasing the upper Koch fittings because the parachute risers had rotated down towards his feet thereby hiding the fittings. The pilot untangled the shroud lines from his leg and rotated onto his stomach while continuing to be dragged, now head first. At the same time, members from the Sky Hawk Parachute Demonstration Team were chasing the pilot in their vehicle and, as they managed to catch up with the parachute, the pilot was able to release the right Koch fitting. One of the Sky Hawks jumped from the vehicle and assisted the pilot by deflating the parachute and other Sky Hawks members performed immediate first aid. Paramedics arrived shortly after and transported the pilot to Chinook Regional Hospital in the city of Lethbridge.
1.2 Injuries to Personnel
|
Injuries |
Crew |
Passengers |
Others |
Total |
|
Fatal |
0 |
0 |
0 |
0 |
|
Serious |
1 |
0 |
0 |
1 |
|
Minor |
0 |
0 |
0 |
0 |
|
Total |
1 |
0 |
0 |
1 |
Table 1: Injuries to Personnel
1.3 Damage to Aircraft
1.3.1 Field Examination
1.3.1.1 The aircraft was destroyed as a result of the ground impact, subsequent explosion and post-crash fire. After striking the ground, the aircraft broke into four major sections (see Annex A, Figure 8). The first main piece of wreckage was the forward fuselage (from radome to the main fuel tank at fuselage station Y350). The forward fuselage was upright but heavily damaged from the impact and post-crash fire. The second main piece of wreckage was the right wing. It was separated from the centre fuselage. The third section was the centre fuselage with the left wing attached. The fourth section was the aft fuselage, which included the engines. All pieces of wreckage had extensive post-impact fire damage.
1.3.2 Laboratory Examination
1.3.2.1 Left Hand Engine
1.3.2.1.1 Based on photographic and video evidence gathered early in the investigation, the LH engine appeared to have been operating correctly at the time of the accident. For this reason, the LH engine was not completely disassembled and was used as a reference for comparison against the RH engine. The LH engine rotated freely, had sustained the least mechanical damage of the two engines and had sustained considerably less fire damage from the accident. The LH engine still contained considerable fuel and oil.
1.3.2.2 Right Hand Engine
1.3.2.2.1 The fan module was found to be in good condition with no obvious pre- impact indications as to a cause of the suspected engine failure. All inlet guide vanes (IGV) were either melted or broken away. The IGV actuator was found in the closed position. After removal of the melted IGVs from the base of the fan section, the engine rotated freely. The upper half of the fan section was removed and all three stages of Fan rotors and stators were found to have varying degrees of foreign object debris (FOD) damage with no missing blades.
1.3.2.2.2 Minor FOD damage was present in all stages on the leading edge and trailing edge blades in the high pressure compressor (HPC) Module. Minor random tip curling was observed as well, but did not follow a uniform pattern.
1.3.2.2.3 The combustor section was borescoped and found to be in very good condition. There were no signs of burn-through or cracking in the combustor. The injectors showed no signs of improper spray patterns.
1.3.2.2.4 The high pressure turbine (HPT) blades were found to be in excellent condition with no damage noted. The low pressure turbine (LPT) blades were also found to be in excellent condition with some heat discoloration on the rearward side.
1.3.2.2.5 The AB module was penetrated by part of the aircraft structure after the crash, resulting in a gaping hole in the lower RH side of the AB (looking from the rear). The flame sensor was also partially torn from the AB. Inspection of both the flame sensor and the AB igniter revealed no obvious defects. All AB fuel nozzles were observed to be in place with no apparent pre-occurrence damage. The flame holder and mixer were also in good condition.
1.3.2.2.6 The engine-mounted alternator engine was inspected to determine its condition. A visual inspection of the internal components of the alternator revealed significant heat damage as well as the displacement of the Nomex wire shield and several wires from the windings. Electrical continuity testing was attempted on the engine mounted alternator; however, the testing was inconclusive due to heat damage sustained by the wiring. The cannon plug was isolated and no shorts were detected between cannon plug pins.
1.3.2.2.7 The MFC, S/N BECA571, was removed and sent to the United States Navy (USN) repair and overhaul facilities (Jacksonville, FL) for a strip-down and discussion of observations and potential failure modes with the engine and MFC original equipment manufacturer (OEM). The MFC had sustained minimal mechanical damage and extensive heat damage. With the exception of the compressor discharge temperature compensation servo mechanism, which was driven to the maximum temperature condition, the positioning of the servo mechanisms within the MFC were consistent with an engine that had completely shut down. However, this condition was also observed with the LH engine which had been operating at full power at the time of impact. The physical evidence suggested that the impact forces were low enough that the engines received limited damage and were able to spool down normally following the occurrence. The inspection also showed the presence of wear of the inner aluminum sleeve of the ratio boost piston. The maximum wear depth was 0.006 inches (see Annex A, Figure 9).
1.3.2.2.8 The variable exhaust nozzle (VEN) sustained some damage to the actuation system. This appears to have occurred during ground impact as video recordings of the flight showed the VEN actuating during earlier portions of the flight.
1.3.2.2.9 In consultation with engineers from the OEM, it was determined that functional testing of the electrical control assembly (ECA) would not be worthwhile due to the extensive heat damage and some post-crash mechanical damage.
1.3.2.2.10 Compressor discharge pressure (P3) air lines were visually checked and pressure tested with no observation of significant damage.
1.3.2.2.11 On completion of electrical checks the fan speed (N1) sensors were found unserviceable due to fire damage. It was not possible to infer the serviceability state prior to the occurrence.
1.3.2.2.12 Fluid samples were difficult to obtain as most of the samples had leaked, burnt off or boiled off. Only limited oil samples were recovered from the VEN. A fuel sample was extracted from the incident aircraft right wing. The sample was found suitable for the laboratory investigation apart from contamination with fire suppression foam. The fuel filter was found to contain some small spherical polystyrene balls of various sizes (500 micron and smaller). The source of the polystyrene balls was undetermined but it is likely that it occurred during the breakup of the aircraft fuel system during ground impact which permitted the direct ingestion of debris into the engine fuel system. Engine oil samples were taken from the right hand engine accessory gear box and oil filter. Analysis of the samples found them to be degraded due to extreme heat exposure but they were otherwise unremarkable.
1.4 Collateral Damage
1.4.1 The aircraft crashed beside runway 23 just north of the 500 ft show line west of runway 12/30 (see Annex A, Figure 8). The crash site had burn areas and some fuel and oil contamination to the soil. 4 Wing Cold Lake personnel and Aerospace and Telecommunications Engineering Support Squadron (ATESS) personnel from Trenton completed the final debris collection. 4 Wing undertook an environmental assessment. There were no claims against the Crown.
1.5 Personnel Information
|
Current |
Yes |
|
Valid Medical Category |
Yes |
|
Total flying time (hours (hrs)) |
1697.1 |
|
Flying hrs on type |
1239.1 |
|
Flying hrs last 48 hrs |
2.1 |
|
Flying hrs last 30 days |
39.4 |
|
Flying hrs last 90 days |
48.5 |
|
Duty hrs last 24 hrs |
8.0 |
|
Duty hrs on day of Occurrence |
4.0 |
Table 2: Personnel Information
1.5.1 The pilot was an experienced CF188 pilot on his third CF188 flying tour. He was posted to 425 Tactical Fighter Squadron in 2008 and was a Section (four-plane) Lead, an Instrument Check Pilot, and a Post-Maintenance Test Pilot. In October 2009, he was selected to be the 2010 CF188 air demonstration pilot. Table 2 provides additional information about flying time and duty hours.
1.6 Aircraft Information
1.6.1 General
1.6.1.1 CF188738 was a single seat twin engine fighter aircraft. The aircraft had been declared fully serviceable at the time of launch from Lethbridge County Airport. It had accumulated a total of 4,176.7 airframe hours at the time of the occurrence.
1.6.1.2 When both engines are operating normally and the thrust of each engine is matched by throttle position, there is no thrust-induced yaw. Should one engine fail, the presence of a second engine enhances overall aircraft safety by providing an independent thrust source with an associated independent electrical system and hydraulic system. However, because of the lateral displacement of the engines from the longitudinal axis of the aircraft, the failure of a single engine will displace the thrust axis laterally. This will induce a yawing moment towards the failed engine proportional to the thrust difference between the two engines. To maintain aircraft control, the yawing moment must be counterbalanced by aerodynamic forces produced by the airframe and flight control surfaces. Although the airframe and flight control surfaces are optimized throughout the flight envelope for the symmetrical thrust situation, they provide acceptable stability and control in the asymmetric thrust situation within a greatly reduced flight envelope.
1.6.1.3 The aircraft maintenance records (not including engine-related records – see Section 1.6.3) were checked with no anomalies noted. Periodic no. 1 and supplementary no. 1 inspections had recently been completed (see Table 3 for details). Upcoming scheduled inspections were a supplementary no. 2 inspection (between 4,235.7 and 4,255.7 airframe hrs) and a periodic no. 2 inspection (between 4,413.3 hrs and 4,493.43 hrs).
|
Airframe (AF) Hrs at Time of Occurrence |
4176.7 AF hrs |
|
Periodic no. 1 |
Completed at 4153.3 AF hrs |
|
Time Flown Since Periodic Inspection |
23.4 AF hrs |
|
Supplementary no. 1 |
Completed at 4145.7 AF hrs |
|
Time Flown Since Supplementary Insp. |
31.0 AF hrs |
|
Last Aircraft Weighing |
5 February 2009 |
Table 3: Aircraft Hours and Inspections
1.6.1.4 The aircraft stores configuration consisted of a basic aircraft with no external stores and no bullets or ballast (orange plastic weighted simulated bullets) in the gun. The weight and balance report listed the operating weight as 24,704 pounds (lbs). The aircraft was fuelled with Jet A1. At the time of the occurrence, fuel on board was approximately 7,200 lbs and the aircraft weight was approximately 31,900 lbs.
1.6.2 Flight Control System
1.6.2.1 The flight control system (FCS) was designed to provide both stability and controllability. Stability, the measure of the aircraft's resistance to external disturbing forces, provides a predictable and steady platform for accomplishing various weapons-delivery tasks. Controllability, the measure of the ease of changing the aircraft's speed, direction, and acceleration, provides the means to fly the aircraft aggressively. The FCS achieves both stability and controllability by monitoring aircraft motion and pilot input, applying pre-programmed control laws, and then commanding control surface movement to provide the responsiveness and manoeuvrability of an agile fighter and the steady platform of a ground attack aircraft.
1.6.2.2 The primary flight controls are the ailerons, twin rudders, differential/collective leading edge flaps, differential/collective trailing edge flaps, and differential/collective stabilators. Hydraulic actuators position the control surfaces. Stick and rudder feel are provided by spring cartridges. Although there is no aerodynamic feedback to the stick and rudder pedals, the effect is generated by the flight control computer's (FCC) scheduling of control surface deflection as a function of flight conditions versus pilot input. Electrical input to the hydraulic actuators is provided via two FCCs loaded with FCS software version 10.7. The FCS employs two modes of operation: auto flaps up (AFU) mode and powered approach (PA) mode. The AFU mode is entered when the flaps switch is selected to the AUTO position (UP) and is used for normal up and away manoeuvring flight. The PA mode is entered when the flaps switch is selected to the HALF or FULL position (DOWN) and is used for the approach and landing phases of flight5. At the time of occurrence the aircraft was operating in the AFU mode.
1.6.3 Engine Information
1.6.3.1 General
1.6.3.1.1 The CF188 is powered by two General Electric (GE) F-404-GE-400 engines. The F404-GE-400 turbofan engine consists of a three-stage low pressure compressor (LPC), also called the fan, driven by a single-stage LPT, and a seven-stage axial flow HPC, driven by a single-stage HPT. The engine is a modular construction, consisting of six major engine modules (fan/LPC, HPC, combustor, HPT, LPT, and AB) as well as an accessories assembly. Details related to the engines installed on the CF188738 are presented at Table 4.
|
Position |
Serial Number (S/N) |
Installed at |
Maintenance due at |
Inspection |
||
|
Engine Flight Hrs (EFH) |
AF hrs |
Min (AF hrs) |
Max (AF hrs) |
Type |
||
|
Left |
376171 |
4655.7 EFH |
4053.3 |
4323.3 |
4383.3 |
Periodic 3 |
|
Right |
376289 |
4508.5 EFH |
4134.0 |
4415.7 |
4475.7 |
Periodic 4 |
Table 4: Engine Hours and Inspections
1.6.3.1.2 The RH engine was of particular interest due to evidence of power loss at the time of occurrence (see Annex A, Figure 5). Before being installed in CF188738 it was installed in aircraft CF188787 as the LH engine. On 27 April 2010 it was removed from CF188787 and installed in CF188738 as the RH engine to provide better engine performance matching with the installed LH engine.
1.6.3.2 Engine Loss of Thrust
1.6.3.2.1 A review of design and maintenance publications associated with the F404 engine revealed multiple failure modes that can result in an un-commanded loss of thrust. The most common failure modes included the loss of electrical power due to a failure of the engine mounted alternator, an internal failure of the ECA, failure of the N1 transmitters, blockage of the P3 airlines, hung stall, and other failure modes within the MFC.
1.6.3.3 Prior Engine Flameout Events (S/N 376289)
1.6.3.3.1 A search of the CF188 Maintenance Signal Data Recording Set (MSDRS) identified that on three separate occasions between March and June 2010, the occurrence engine experienced a series of flameout-low idle6 / flameout-roll7 back events during start; twice while installed in CF188787 and once while installed in CF188738. While flameout-low idle events exhibited different symptoms than flameout-roll back events, the cause of these malfunctions was not well understood and they tended to be dealt with in a similar manner. In all three cases the engine was unresponsive to the throttle inputs. The maintenance actions taken to address these flameout-low idle / flameout-roll back events were varied.
1.6.3.3.2 On 3 March 2010, while CF188787 was deployed to Salina KS, the pilot completed troubleshooting using the Start Anomalies section of the Pilot’s Checklist, which appeared to resolve the issue, and the mission was flown without further incident. The pilot did not report the engine start issue on a CF3498.
1.6.3.3.3 On 6 March 2010, also while CF188787 was deployed to Salina KS, Director Aerospace Equipment Program Management Fighters and Trainers (DAEPM (FT)) message FT23010 262213Z Feb 10 (see paragraph 1.6.3.7.10.4) was used as a reference to troubleshoot the subject engine. At the point where MFC replacement was the next step in the procedure, maintenance crews changed the fuel specific gravity setting on the MFC to 0.78 for both the LH and RH engines. Again a CF349 was not raised to document this event. The engine did not experience any further issues of this nature during the remainder of the deployment and when the aircraft returned to 3 Wing Bagotville the specific gravity was set back to 0.81.
1.6.3.3.4 On 10 June 2010, CF188738 transited from Bagotville, QC to Quebec City, QC for an air show and flew five flights without incident between 10 and 12 June 2010. Then on 13 June 2011, the RH engine experienced a flameout-low idle / flameout-roll back on start. The pilot shut the aircraft down, proceeded to the spare aircraft and completed the air show. Following the air show routine, the maintenance crew chief suggested that the pilot reattempt to start the occurrence RH engine to see if the problem was still evident. The engine was started twice and during each start, it experienced a flameout-low idle / flameout-roll back. During one of the start attempts, the crew chief directed the pilot to advance the throttle to achieve approximately 75% N2 RPM . The engine initially responded by reaching 75% N2 RPM9 but started to decrease to between 65 to 70%. The pilot attempted to advance the throttle further, but the RPM kept decreasing. Concerned about damaging the engine, the pilot shut the aircraft down. Following these restart attempts, the crew chief communicated with 3 Air Maintenance Squadron (AMS) and was advised to change the specific gravity setting on the MFC for both engines to 0.78. Troubleshooting procedures directed in the first line maintenance document (C-12-188-270/NH-000) were not carried out. While a CF349 was not raised for the flameout-low idle / flameout-roll back issues, a CF349 was raised for both engines to change the specific gravity settings to 0.78. The specific gravity settings were changed on both engines and the CF349 signed off. The RH engine was restarted and run for a few minutes at various throttle settings and the engine functioned normally. Both the pilot and crew chief believed that changing the specific gravity had resolved the issue. The engine did not experience any further flameout-low idle / flameout-roll back issues prior to the occurrence.
1.6.3.4 Electronic Control Assembly
1.6.3.4.1 The ECA is a modular, solid-state unit powered by the engine mounted alternator and cooled by fuel from the main fuel pump. The ECA has 18 electrical modules that compute, schedule and control engine operation. It also transmits signals of the fan inlet temperature (T1), exhaust gas temperature (EGT or T5), VEN throat area (A8) and N1 to the cockpit instrument and mission computer systems. At high power settings the ECA modifies the fuel schedule of the MFC as a function of T1, maximum N1, P3 and maximum EGT (or T5). Failure of the ECA can result in engine roll back to safe mode (VEN closed and N2 RPM reduced to produce approximately 50% full MIL power); however, some type of indirectly related caution (e.g. Inlet Temperature caution, which would cause an “Engine Right” (or Left) voice alert to be activated) is typically triggered.
1.6.3.5 Variable Exhaust Nozzle
1.6.3.5.1 The VEN system is a self-contained, hydraulically actuated, electrically controlled system, providing a controlled A8 for the exhaust gases from the turbine and AB. The VEN is scheduled in response to movement of the throttle and A8 is controlled to provide the required thrust and fuel efficiencies while maintaining EGT (or T5) within limits. When the throttle is positioned less than flight idle, A8 is at its maximum. When the throttle is advanced above flight idle, up until the MIL power stop, A8 is scheduled with throttle position (advancing throttle results in decreasing A8, modulated by ambient pressure (P0) and T1). At MIL power and above, A8 is scheduled with throttle position, and T1 while being trimmed with T5 and can again achieve its maximum area (A8).
1.6.3.6 Main Fuel Control
1.6.3.6.1 The engine MFC, located on the accessory gearbox and mounted on the aft end of the main fuel pump, is a hydro-mechanical droop type fuel control. Its main function is to provide regulated fuel flow to the combustor for core engine operation. Below MIL power, N2 is a function of throttle position, P3 and HPC inlet temperature (T2.5) and N1 is a function of N2 and T1. At MIL power and above, N2 is essentially constant; however, it is limited by T5, N1 and P3, while N1 is controlled by a signal from the ECA and is a function of T1.
1.6.3.6.2 The MFC incorporates a fuel density adjustment unit that allows technicians to set the specific gravity corresponding to the fuel loaded into the aircraft. The scale ranges from 0.65 to 0.85 grams per millilitre (g/ml) and can be adjusted in 0.01 g/mL increments.
1.6.3.6.3 When the Canadian Forces (CF) initially procured the CF188 aircraft, engines were delivered with either 4075T76G01 (G01) or 4075T76G03 (G03) configuration MFCs. Over the years, in order to address a number of reliability issues, multiple design changes were incorporated across the fleet, initially converting all G01 configuration MFCs to G03 and later, all G03 configuration MFCs to G04.
1.6.3.6.4 Following engineering investigations on G04 configuration MFCs, wear found at the interface between the ratio boost piston and the servo link. During operation of the throttle, a servo link actuates the ratio boost piston. It was found that wear was occurring on the aluminum skirt of the ratio piston (see Annex A, Figure 9). The aluminum skirt was introduced during conversion from G03 to G04. A wear condition (between 0.0001 inches and 0.014 inches in depth) present during engine operation can, and has caused, the ratio boost piston to stick during throttle transients, which in turn can reduce the engine flameout margin or otherwise cause the engine to become erratic or unresponsive during acceleration / deceleration. Engineering change proposal (ECP) G404-E-91 (introduced as one of the modifications during conversion of the MFC from G04 to G08) was designed to address the ratio boost piston wear issue with material changes and ratio piston redesign. G08 MFCs exhibit approximately one-third the incidence rates as G04 MFCs. At the time of the occurrence, the RH engine MFC was in a G04 configuration.
1.6.3.6.5 A number of CF MFCs returned to third line maintenance were found with the ECP G404-E-32 ratio boost piston (implemented when upgrading from G03 to G04) worn out of limits with an average time of 1061.7 engine operating time (EOT). The occurrence MFC (S/N BECA571) had accumulated 2702.8 EOT, the third highest time accumulated on a G04 ratio boost piston.
1.6.3.6.6 In 2008 a review of the historic data relevant to flameout-low idle / flameout-roll back was carried out with a plan to reduce future occurrences. The study revealed that if the MFC is not replaced following a flameout-low idle / flameout-roll back event, then the chance of a repeat event is high. The study could not identify a single common failure mechanism and found that several different MFC parts, or a combination of parts, might degrade in service resulting in low idle / roll back induced flameout cautions.
1.6.3.6.7 The review also showed that after the introduction of JP-8 fuel in 2001, there was a marked increase in the number of flameout-low idle events. On 19 December 2003, DAEPM (FT) issued a message directing the specific gravity setting to be changed from 0.82 to 0.81 and subsequently the number of flameout-low idle events dramatically declined back to pre-2001 levels. The introduction of JP-8 and the reduction of the specific gravity from 0.82 to 0.81 did not have a notable effect on flameout-roll back events.
1.6.3.6.8 A January 2010 review of the flameout-low idle / flameout-roll back data from 2005 to 2009 showed that the issue was independent of the fuel type used and the location of operation.
1.6.3.6.9 Specific Gravity
1.6.3.6.9.1 The MFC incorporates a specific gravity adjustment lever used to set the specific gravity associated with the fuel used. According to C-12-188-PCM/MB-001 WP009 00 the specific gravity setting for the various fuels is:
a. JP-8 (NATO F-34) and JP-8+100 (NATO F-37) specific gravity = 0.81 g/mL
b. Jet A-1 specific gravity = 0.80 g/mL
1.6.3.6.9.2 The fuel specifications for both JP-8 and Jet A-1 allow for density variation between 0.78 and 0.84 g/mL measured at 15 degrees C.
1.6.3.6.10 Specific Gravity Adjustment History
1.6.3.6.10.1 Since 2005, there had been a total of 60 engine flameout-low idle / flameout-rollback events. Of these, 37 occurred away from the main operating bases (MOB) (Cold Lake, AB and Bagotville, QC) with a notable increase in the rate of occurrences after 2007. Of the 60 events, 31 resulted in replacement of the MFC. On 15 July 2008 a review of third line maintenance reports for MFCs rejected for flameout-low idle / flameout-rollback events failed to determine a root cause. Analysis of the fuel involved in these 60 flameout-low idle / flameout-rollback events indicated that Jet A-1 was not the primary problem.
1.6.3.6.10.2 During a 425 Squadron deployment to Salina KS in June 2009, several flameout-low idle / flameout-rollback incidents occurred. As the problem was not limited to one engine and the fuel in use was Jet A-1, the F404 technical service representative (TSR) recommended that the fuel specific gravity setting be changed to 0.78. This recommendation was carried out and for the remainder of the deployment there were no reported flameout-low idle / flameout-rollback events.
1.6.3.6.10.3 On 24 February 2010, 425 Squadron requested a blanket clearance from DAEPM (FT) to pre-emptively set the specific gravity of all aircraft deploying to Salina KS during the dates 1 to 10 March 2010. The request for a blanket authorization was denied.
1.6.3.6.10.4 Based on a review of GE F404 design and maintenance publications as well as on a discussion with systems experts from GE, modifications to specific gravity setting will bias the total fuel flow (lowering the specific gravity setting below that specified for the fuel used will result in an increased RPM) through the metering valve in the MFC over the entire range of engine speeds and can, therefore, prevent a potential flameout indication by effectively increasing the idle RPM above the flameout threshold when throttles are positioned at ground idle. For this reason, in part, DAEPM (FT) released a message (FT23010 262213Z Feb 10) that provided additional direction, over and above the direction already provided in the maintenance manual (C-12-188-270/NH-000), when troubleshooting flameout events that allowed the modification of the specific gravity in certain situations. It was only applicable at Salina KS for the period of the deployment from 1 to 10 March 2010. In particular it stated that if a flameout caution occurred and troubleshooting leads to replacement of the MFC, then the specific gravity for that engine should be lowered one increment, but not to exceed 0.78. If the specific gravity is at 0.78 and the problem re-occurs, then the MFC is to be replaced. All actions were to be reported by the TSR.
1.6.3.6.10.5 On 28 July 2010, DAEPM (FT) rescinded the earlier message from 26 February 2010. It was stated that changing the specific gravity setting was not a proper troubleshooting procedure and could mask real problems with the MFC. Furthermore, it was found that the specific direction in the 26 February 2010 message was not always followed and in some cases the specific gravity was simply being set to 0.78 by default after transit from the MOB to Salina, KS, and refuelling with Jet A-1, even though the aircraft had not experienced a flameout-low idle / flameout-rollback event.
1.6.3.7 Engine Mounted Alternator
1.6.3.7.1 The engine-mounted alternator is made up of a rotor and stator with three separate windings; one powers the ignition exciter for main and AB ignition, one powers the N2 signal for ECA operations and one powers the signal for N2 cockpit indications. A subsequent review of the historical F404 engine records revealed that movement of the Nomex wire shield had previously been identified as a cause of engine flameouts and led to the incorporation of a “Caution” in the CF188 first line technical publications. The “Caution” states that a Nomex wire shield not firmly secured may result in loose stator wires contacting the alternator rotor and locknut, causing engine flameout. Failure of the engine mounted alternator can result in engine roll-back to safe mode (VEN closed and N2 RPM reduced to produce approximately 50% full MIL power thrust). In this mode some type of indirectly related caution (e.g. Inlet Temperature caution, which would cause an Engine R/L voice alert to be activated) is typically triggered.
1.6.3.8 Fan Speed Sensors
1.6.3.8.1 The engine incorporates two fan speed sensors. Both sensors must be serviceable for the system to function properly. Failure of one of the sensors will result in engine roll back to safe mode (VEN closed and N2 RPM reduced to produce approximately 50% full MIL power). In this mode, some type of indirectly related caution is typically triggered (e.g. Inlet Temperature caution, which would also activate an engine related voice alert).
1.6.3.9 Compressor Discharge Pressure (P3)
1.6.3.9.1 In the early 1990’s, three USN flameout-low idle / flameout-roll back events, with symptoms similar to those described at paragraph 1.6.3.3, were investigated and found to have been attributed to a leak in the P3 line due to a crack where the P3 line affixed to the MFC. As a result of the investigation, an engineering program, EPD No. 404M81, was proposed to improve the P3 system hardware (lines, brackets and clamping) to reduce system vibratory response. The engineering program recommended to add a fluorescent penetrant inspection (FPI) to the technical orders when the component was sent to third line maintenance.
1.6.3.9.2 EPD No. 404M81 evolved into ECP G404-C-67 “MFC Manifold Redesign” and introduced a new MFC/P3 manifold P/N 5122T38G01. According to the F404 Control Approval Form dated in the July 1996 timeframe, the CF did not approve this ECP. Furthermore, the recommendation for FPI inspection of the manifold did not make it into the USN or CF technical publications and the current process is to perform a white light inspection. At the time the CF had conducted a return-on-investment analysis of this ECP and determined that a fleet wide implementation was not warranted given the lack of significant CF MFC manifold failures.
1.6.3.9.3 Since 1988, the CF has had three events where the MFC/P3 manifold was determined to be the cause. Of the three events, two were due to a P3 leak and one for a fuel leak. Since 2006 there have been a total of 23 MFC/P3 manifolds rejected at third line maintenance for scratches/dents. None were rejected for cracks.
1.6.3.9.4 Based on this history, the CF still considers that the original return-on-investment analysis still stands.
1.6.3.10 Hung Stall
1.6.3.10.1 C-12-188-270/NH-000 WP006 01 states that when the Mission Computer (MC) detects an engine stall condition, “if the N2 RPM rolls back to 55-70%, while EGT rises or remains above 550° C, then the stall is a hung stall. Also, during a hung stall the engine will not respond to throttle movement, except for retarding throttle to IDLE or OFF.”
1.6.3.11 MFC Idle Arm
1.6.3.11.1 External idle and maximum speed adjustments establish MFC governed idle and MIL power N2 speed limits.
1.6.3.11.2 A worn idle adjustment lever arm can result in flameout-low idle / flameout-roll back events even though an actual flameout has not occurred; however, it will not likely inhibit engine response to throttle inputs. Direction to replace the idle adjustment lever whenever an MFC was upgraded from a G04 to a G08 was provided by the CF during the week of 19 – 23 July 2010.
1.6.4 Caution and Advisory System
1.6.4.1 Cautions and advisories are normally displayed to the pilot on the left MFD. A few cautions are redundantly displayed on the Caution Light Indicator Panel located on the right instrument console. A yellow Master Caution light, on the upper left part of the instrument panel, illuminates when any caution is displayed. The Master Caution light goes out when it is pressed (reset). A momentary audio tone is initiated whenever the Master Caution light illuminates.
1.6.4.2 Some cautions will set a maintenance status panel (MSP) code associated with the particular failure condition. MSPs are recorded on the advanced memory unit (AMU) (see paragraph 1.11.2).
1.6.4.3 For certain critical cautions, voice alert transmissions are also sent to the pilot's headset. When a condition occurs that triggers one of the critical cautions, the voice alert system provides a message to the pilot’s head-set. The message is repeated twice, for example “ENGINE RIGHT, ENGINE RIGHT.”
1.6.4.3.1 The “ENGINE RIGHT, ENGINE RIGHT” voice alert will be triggered for a number of engine related cautions. In particular, it will be triggered for an N1 or N2 over-speed condition, EGT too high, an engine flameout and an engine compressor stall. These engine-related cautions will also set an MSP.
1.6.5 Ejection System
1.6.5.1 The Martin Baker (MB) naval aircrew common ejection seat (NACES) SJU-17 is a ballistic catapult/rocket system that provides the aircrew with a quick, safe, and positive means of escape from the aircraft.
1.6.5.2 The seat survival kit, which fits into the seat bucket, is a contoured rigid platform which contains an emergency oxygen system and a fabric survival rucksack. Two yellow manual deployment handles are mounted on the aft surface of the kit and pulling either handle enables the aircrew to deploy the raft and survival package after man/seat separation, which remains attached by an 8.2 meter (m) drop line. The seat survival kit (SSK) weighs approximately 34 lbs. Conducting parachute landings with the seat kit attached may cause injury.
1.6.5.3 Additional information can be found in the NACES Ejection Seat AOI Advance Change Notice 10-01 dated 21 October 2010.
1.7 Meteorological Information
1.7.1 The flight was operated under visual metrological conditions (VMC).
1.7.2 Special Weather Report taken at Lethbridge County Airport, Lethbridge Alberta at 1815 UTC on 23 July 2010. Winds - 320 degrees True at 12 knots gusting to 18 knots, Visibility - 30 statute miles, Cloudiness - Broken cloud layer based at 2,500 ft AGL, and a second broken cloud layer based at 9,000 ft AGL, Remarks - the first cloud layer comprised of 5/8 in cumulus cloud and the second cloud layer comprised of 1/8 in alto-cumulus cloud.
1.7.3 Weather Report taken at Lethbridge County Airport, Lethbridge, Alberta, at 1800 UTC on 23 July 2010. Winds - 320 degrees True at 16 knots, Visibility - 30 statute miles, Cloudiness - broken cloud layer based at 2,500 ft AGL and a second broken cloud layer based at 9,000 ft AGL, Temperature – 18 degrees Celsius, Dew Point – 11 degrees C, Altimeter setting – 30.04 inches of mercury, Remarks – the first cloud layer comprised of 6/8 cumulus cloud and the second layer comprised of 1/8 alto-cumulus cloud.
1.7.4 Terminal Forecast for Lethbridge County Airport, Lethbridge Alberta sent at 1738 UTC on 23 July 2010 and valid from 1800 UTC on 23 July 2010 to 0600 UTC on 24 July 2010. Winds – 310 degrees True at 15 knots gusting to 25 knots, Visibility – greater than 6 statute miles, Cloudiness – broken cloud layer based at 2,000 ft AGL and a second broken cloud layer based at 9,000 ft AGL becoming between the period 1800 UTC and 2000 UTC on 23 July 2010 Cloudiness – broken cloud layer based at 2,500 ft AGL.
1.8 Aids to Navigation
1.8.1 Not applicable. The aircraft was navigating visually in the vicinity of the airfield.
1.9 Communications
1.9.1 The CF188 has two ultra-high frequency (UHF)/very high frequency (VHF) radios. At the time of the occurrence the aircraft had one radio tuned to the ADD's VHF frequency (123.3 MHz). The ADD acted as both the safety monitor and the announcer for the CF188 demonstration. The other radio was tuned to the flight services station (FSS) VHF mandatory frequency (MF) (121.0 MHz). Normally the pilot would have been tuned to the ADD’s discrete frequency for the show (125.5 MHz); however, this was not a show day so the pilot flew the show on the MF. The ADD reported that he did not hear any radio chatter on the MF during the show. The pilot’s last radio call was made for takeoff and receipt of the airspace clearance to perform the practice show.
1.10 Aerodrome Information
1.10.1 CYQL is a certified aerodrome located at N49 37 49 W112 47 59, approximately 2 km south of Lethbridge, Alberta. The airport is within a class “E” control zone and is serviced by a Nav Canada FSS which provides aircraft advisory services. The airport has two intersecting runways, taxiways and a ramp area for aircraft parking. The main runway is 05/23 (6,500x200 feet asphalt, 054/234 degrees magnetic). The other runway is runway 12/30 (5,500x150 feet asphalt, 124/304 degrees magnetic). At the time of the occurrence, the airport was set-up for the air show that was scheduled to begin the following day (see Annex A, Figure 1 for the layout of the airport). Runway 23 was the active runway and was also used as the show line reference.
1.10.2 At the time of the occurrence, operational control of the aerodrome came under the special flight operating certificate (SFOC) issued by Transport Canada to the Alberta International Air Show. As a condition of the SFOC, the air show authorities were required to produce an emergency procedures manual (EPM). This manual provided the structure for the emergency response plan in the event of accident during the core operating hours of the air show event.
1.11 Flight Recorders
1.11.1 Cockpit Voice Recorder (CVR) / Flight Data Recorder (FDR)
1.11.1.1 The CF188 was not fitted with a CVR or an FDR.
1.11.2 Advanced Memory Unit
1.11.2.1 The CF188 does record certain aircraft parameters on a non-crashworthy AMU, which is a programmable computer serving as a non-volatile solid state memory device. The AMU has two computer memory card slots. A personal computer memory card international association (PCMCIA) mission card is inserted in one of the slots and is used to transfer initialization or mission specific data to the various avionics systems. The PCMCIA maintenance card is inserted in the second slot and collects maintenance data during normal aircraft operations. The PCMCIA maintenance card can be removed from the AMU and downloaded following a flight. The AMU and the mission computers were recovered from the wreckage and forwarded to the National Research Council (NRC) Flight Recorder Playback Centre for analysis. Approximately 2 ½ minutes of the occurrence flight, in addition to the previous 10 flights, were recorded on the PCMCIA maintenance card; however, the accident, which occurred approximately seven minutes after takeoff, was not recorded on the card.
1.11.3 Mission Computer Non-Volatile Memory
The CF188 has non-crashworthy MCs that record up to 45 seconds of flight information in non-volatile memory. The recorded information includes engine and flight parameters, flight control inputs and control positions and cautions, warnings and advisory information. Data from the non-volatile memory chips in the mission computers was recovered and analyzed and provided flight data for the five seconds immediately preceding ground impact. Table 5 presents the relevant engine data.
|
LH Engine |
RH Engine |
|
|
power lever angle (PLA) (degrees) |
130 deg (full AB) |
130 deg (full AB) |
|
N2 (% RPM)10 |
94 |
65 |
|
N1 (% RPM)11 |
98 |
38 |
|
Fuel Flow (lbs/hour) |
7872 |
704 |
|
EGT (degrees Celsius) |
800 |
360 |
|
VEN, A8 (% fully open) |
68 |
0% |
Table 5: Data Recovered from Non-volatile Memory
1.11.3.1 The recovered engine data indicated that during the last five seconds of flight the LH engine was selected to maximum AB and was operating at or close to that condition while the RH engine was selected to maximum AB but, with 65% N2 RPM, the engine was operating within the ground idle range and below the flight idle range12.
1.11.3.2 Flight control positions, recorded approximately four seconds prior to ground impact, indicated that the aircraft was operating at or above 25 degrees AOA with moderate left lateral control (aileron and differential horizontal stabilators) deflection and full left rudder deflection.
1.11.3.3 There were no relevant cautions, warnings, advisory or MSP information recorded.
1.11.4 Head Up Display
1.11.4.1 The HUD video camera can also provide flight information as it can record internal and external cockpit transmissions as well as the information displayed on the HUD. The HUD video camera was turned on during the air show practice but the tape was destroyed during the accident and no data was recovered.
1.12 Wreckage and Impact Information
1.12.1 The aircraft wreckage trail was oriented approximately north-northwest with the nose of the aircraft to the south-southeast, pointing to the south-southeast, and the tail section positioned to the north-northeast, also pointing south-southeast. The wreckage was dispersed over an area approximately 1,000 ft long and 279 ft wide. Apart from the major sections of the wreckage, other components were distributed throughout a fairly compact debris field.
1.12.2 The ejection seat and aircraft impacted the grassed infield approximately 60 m north of the edge of the debris field. The sequencer was intact and was removed from the seat for further analysis of seat performance. The canopy landed on the southeast corner of the same intersection.
1.13 Medical
1.13.1 The pilot received serious injuries during his parachute landing. Initial medical response by the air show paramedics was delayed due to a misunderstanding by the paramedics on their need to have clearance to the runway. They were able to reach the pilot minutes after the Sky Hawks team members had reached him. The local hospital ambulance and crew arrived approximately 10 minutes after the ejection and appropriate measures were taken for the pilot to be transported to the Chinook Regional Hospital in Lethbridge.
1.13.2 The hospital declined to collect toxicological samples. Samples for toxicology were taken that evening by the Cold Lake flight surgeon who had arrived with the Cold Lake initial response team. The samples were shipped to the Armed Forces Institute of Pathology in Bethesda, Maryland, for analysis and yielded negative results.
1.14 Fire, Explosives Devices, and Munitions
1.14.1 Fire
1.14.1.1 There was no evidence of a pre-impact fire. Upon impact, the aircraft broke apart and burst into flames when the Jet A-1 fuel was expelled and exploded. Two minutes after the crash, the fireball had mostly spent itself. Only localized fires remained where the main major wreckage sections came to rest.
1.14.2 Explosive Devices
1.14.2.1 Canopy
1.14.2.1.1 No anomalies were noted with any component of the aircraft canopy. The canopy was jettisoned and all pyrotechnics for the ballistic removal of the canopy were expended. The majority of the canopy fragments were found within a six foot radius of the main canopy frame. The internal canopy jettison handle was recovered from the wreckage and all escape systems aircraft and seat mounted pyrotechnics were expended.
1.14.2.2 Navy Aircrew Common Ejection Seat (NACES)
1.14.2.2.1 The ejection seat pyrotechnics were all contained with the ejection seat, with the exception of the catapult, the outer tube (which remained in the aircraft, as designed), and the inner tube (which broke off the seat at ground impact). The ejection seat was extensively damaged during the ground impact. The main beam was snapped in half and all pyrotechnics were expended. The center pull handle was up and all sears had been extracted with the exception of the manual override handle (MOR). This is a normal function and the MOR would only have been deployed if the pilot had experienced difficulty in the ejection that required him to manually override the parachute deployment system. Since this was not the case, the handle was found down and locked. No anomalies were noted with any component of the NACES system.
1.14.3 Munitions
1.14.3.1 There were no munitions on board the aircraft and no ballast was used in the gun, per the Aircraft Armament State Record (CF338).
1.15 Survival Aspects
1.15.1 Ejection Sequence
1.15.1.1 The pilot initiated ejection in a good body position and exited the aircraft cleanly (see Annex, A Figure 6). The head forward posture is a normal event due to the g-forces experienced at ejection. The parachute deployed immediately and had not yet achieved a steady state descent when the pilot hit the ground during the front swing of a pendulum motion. The low altitude of the ejection did not allow sufficient time to steer the parachute or deploy his SSK and the pilot landed firmly while the chute was still undergoing post-opening oscillations. The pilot became entangled in the parachute lines after contacting the ground, which caused him to be dragged feet first. As a result, the pilot had difficulty locating and disconnecting his parachute release fittings (Gen 2 Koch fitting). Because he was being dragged feet first, the parachute shroud lines were pulled forward, rendering the Koch fittings inaccessible until the pilot was able to free his feet from the shroud lines. Once free of the lines, he was flipped over onto his stomach and dragged head first. The pilot was finally able to manage a complete disconnect from the parachute after being dragged approximately 500 m. The Aerospace Engineering Test Establishment (AETE) Crew Systems specialists classified the ejection as unsuccessful / survivable due to the extent of the pilot’s injuries sustained during his parachute landing. Unsuccessful / survivable is defined as an aircrew injury that results in capture (war time), or returning to flying status at a later date (peacetime), or cannot return to flying status due to injuries (peacetime).
1.15.2 Aviation Life-Support Equipment (ALSE)
1.15.2.1 Seat Survival Kit
1.15.2.1.1 A detailed examination of the SSK was conducted. The hard shell and content bag showed signs of abrasion. The emergency oxygen system activated upon ejection and the system was recovered with approximately 500 pounds per square inch (psi) remaining. The oxygen communication service line, attached to the aft left corner of the outer hard shell, showed minor damage. Some damage to the content of the SSK was also noted.
1.15.2.2 Emergency Transmitters
1.15.2.2.1 The CF188 is equipped with a “bail-out tone” device that automatically transmits an audible tone on 243.0 MHz if the ejection seat is activated. The tone was recorded on the air traffic control (ATC) tape on this frequency.
1.15.2.3 PCU-56
1.15.2.3.1 The pilot was wearing a medium harness, which was cut off by the first responders to administer initial first aid. The chest strap was secured and the recommended securing procedure was used. The harness was in remarkably good condition.
1.15.2.4 G-Suit
1.15.2.4.1 The pilot was wearing a medium regular G-suit, which was in fair condition with definite signs of fading. Minor holes were noted in the outer shell. The right leg had significantly more grass stains and drag abrasions than the left leg. The lower right pocket was intact and the pilot's visor cover was in it. There was minor scorching on the pen pocket located on the lower right pocket.
1.15.2.5 Life Preserver Survival Vest (LPSV)
1.15.2.5.1 The LPSV (MSV 980) was cut off the pilot and removed by the first responders. It showed significant signs of being dragged on the front right side. The gate snap hooks on the lower portion of the LPSV were installed but the right one was damaged and no longer retained the D-Ring. The pocket containing the oxygen regulator showed signs of having been dragged; however, the oxygen regulator was clean, undamaged, and properly connected to the vest. The oxygen hose was routed through the large gun pocket outer flap. This configuration is not normal. The pilot stated the routing of the oxygen hose through the gun pocket was deliberate and was done in an attempt to reduce the interference of the hose and the left arm. The pilot indicated that the approved routing resulted in loose hose getting in the way of the left arm while making throttle inputs. Although this issue had already been identified by the CF, a resolution to the problem was not forthcoming and the pilot felt that the safety of the aircraft was in jeopardy. Independent of this investigation, work is ongoing to resolve these issues and a prototype solution is currently being trialed. It is estimated that a fix will be implemented fleet-wide by the end of July 2013. The Carbon Dioxide cylinder that was installed was a 28-31 gram cylinder whereas the approved configuration calls for a 35 gram cylinder. The discrepancy was brought to the attention of the Senior Design Engineer and has since been resolved.
1.15.2.6 Helmet 190A and Oxygen Mask
1.15.2.6.1 The helmet worn by the pilot had the night vision goggle (NVG) mounting brackets installed. The clear and dark visors were found in the vicinity of where the pilot landed. The dark visor was cracked at the bridge of the nose up to the rubber seal. Two distinct rub marks were present on the left aft side of the helmet. The helmet had a Zeta Liner in lieu of the approved thermoplastic liner. The Zeta liner was not authorized for use in any ejection seat equipped aircraft other than for the CF188 when the pilot wears the joint helmet-mounted cueing system (JHMCS). Furthermore, the Zeta liner was not installed per the original equipment manufacturer instructions. The Zeta liner was subsequently authorized for use with all 190A aircrew helmet users (all fleets). There was no sign of damage on the high altitude / low profile (HA/LP) mask and oxygen hose.
1.15.2.7 Flying Boots
1.15.2.7.1 The pilot was wearing non-issue boots with a steel toe. The boots were in very poor condition and had previous damage on them. The size and manufacturer are unknown.
1.15.3 Emergency Response
1.15.3.1 The EPM developed for the Alberta International Air Show identified three levels of response to occurrences. In the case of this occurrence, it required a Level 3 response which entailed a full and immediate response by all agencies, with additional resources from off the airport being recruited. As per the EPM, the Air Show Fire Boss was the command authority for all aircraft fire fighting and rescue on the airport during the operating hours of the air show.
1.15.3.2 The on-site emergency response was activated via communication on the emergency channel of the air show radio net. Emergency vehicles from Health, Safety and Environment (HSE) Integrated Fire/Ambulance Services were immediately dispatched to the accident site. Simultaneously, a call was made to the Lethbridge Fire and Rescue Services requesting dispatch of manpower and equipment to the emergency area.
1.15.3.3 The first personnel on site were experienced parachutists from the Sky Hawk Parachute Demonstration Team. They assisted the pilot by collapsing his chute, stabilizing him and removing all of his ALSE gear to enable the assessment for injuries and administer first aid. HSE Integrated Fire/Ambulance Services and Lethbridge Emergency Response personnel were not trained in how to deal with parachutes.
1.15.3.4 The Emergency Response Plan detailed that “The Emergency Response Services (ERS) Fire Command will be in charge of all aircraft, fire fighting and rescue operations for all aircraft emergencies on the County of Lethbridge Airport property; however, the ERS must obtain permission to enter the air box, from the Air Boss (Radio Frequency 125.5 MHz).” Since the emergency vehicles were not equipped with aviation band radios that enabled communication with the Lethbridge FSS, they drove to the taxiway and waited for airport escort vehicles. Air show officials were finally able to communicate through the air show emergency channel and facilitate the deployment of emergency vehicles. This sequence of events delayed the response by approximately two minutes.
1.15.3.5 On the day of the occurrence, HSE Integrated Fire/Ambulance Services were at half staff because it was a practice/media day with limited audience on site. The emergency crews were not wearing their personal safety equipment and had to suit up prior to deployment. As a result, it took approximately nine minutes from the time of the crash to when crews began fighting the fire. The Lethbridge Fire Services arrived on the scene about 11 minutes after the occurrence and assisted HSE Integrated Fire/Ambulance Services crews in extinguishing all fires.
1.15.3.6 Once the ambulance crew from HSE Integrated Fire/Ambulance Services reached the accident site, the pilot was transferred to them. He was later transferred to the Lethbridge Emergency Medical Services for transportation by ambulance to the hospital.
1.16 Test and Research Activities
1.16.1 Test Activities
1.16.1.1 The Directorate of Flight Safety (DFS) investigation team proceeded to 4 Wing Cold Lake to evaluate the high AOA manoeuvre and recovery procedures following a single engine failure in the advanced distributed combat training system (ADCTS) simulator. The ADCTS simulator is a dynamic (ie: simulated flyable), non-motion based, CF188 weapon systems training device. The aircraft flight model and engine thrust model have not been validated for accuracy of representation to an actual CF188 aircraft with F404-GE-400 engines.
1.16.1.2 The ADCTS was programmed with similar initial aircraft and environmental conditions as on the day of the occurrence. A number of different thrust loss scenarios and recovery techniques were investigated. From a mid-power setting (sub-MIL), it was found that following engine failure initiation it took about nine seconds before the engine flameout caution and audio warning were activated.
1.16.1.3 On 26 August 2002, a CF188 experienced a RH engine flameout shortly after takeoff as a result of a failed ECA. A review of the engine data revealed that with both throttles set to 88 degrees PLA (sub-MIL) and N2 RPM ~ 92%, the N1 RPM, fuel flow, N2 RPM, TDP, EGT, CDP and T1 all suddenly drop, with the VEN going to full closed, followed after one second by an Inlet Temperature caution. Four seconds later the N2 RPM dropped below 60% and the FLAMEOUT caution was activated. The total time from the loss of thrust event to the activation of the FLAMEOUT caution was approximately five seconds.
1.16.1.4 It was also found that by maintaining 25 degrees AOA with the left throttle above minimum afterburner and the right throttle at flight idle the pilot could not prevent the aircraft from departing controlled flight; however, when the pilot initiated recovery actions by reducing and maintaining AOA at approximately 15 degrees within two seconds of initiation of the engine failure and simultaneously selecting full afterburner and left rudder, the pilot was able to accelerate and recover the aircraft while losing up to 350 ft of altitude before beginning to climb.1.16.2 Modeling and Simulation Activities
1.16.2.1 Unbalanced thrust from an aircraft’s engines will produce a yawing moment that is dependent upon the thrust imbalance and the lever arm of the force. The aircraft’s natural tendency to counteract yaw will typically be weakest at low speed (high AOA) and increase as the aircraft accelerates. If maximum rudder deflection is unable to counteract the yawing moment, the yaw will increase and eventually there will be a loss of directional control. The airspeed where maximum rudder deflection does not counterbalance the thrust asymmetry is called the minimum controllable airborne airspeed (Vmca)13. If the aircraft is operating below Vmca when a thrust imbalance occurs, the pilot must either reduce power on the operative engine(s) or increase airspeed by lowering the nose of the aircraft in order to regain directional control. Both of these options will result in a loss of altitude.
1.16.2.2 A modeling and simulation effort was conducted to identify Vmca under various flight and environmental conditions. Bombardier Aerospace Engineering Services (BAES) was requested to investigate the open and closed loop response characteristics of the CF188, loaded with FCC software 10.7, to asymmetric thrust conditions during the high AOA flight regime using the all-purpose real-time engineering simulator (ARES) CF188 Engineering Flight Simulation tool. The first simulation scenario was an attempt to reproduce the CF188 occurrence. A sensitivity analysis was then performed to obtain the minimum controllable airspeed at different thrust asymmetries followed by open and closed loop analyses of the aircraft response to identify aircraft recovery performance and handling qualities. The Canadian recovery method, defined as the simultaneous reduction of AOA to below 12 degrees and selection of maximum AB while using rudder pedal and lateral stick inputs as required to maintain wings level and zero yaw, was used for the closed loop analyses14.
1.16.2.3 Notable Findings
1.16.2.3.1 With the RH engine failed15 (off), increasing the LH throttle to a mid-power setting (approximately 85%), without any other pilot input to the flight controls, only induced about 1 to 2 degrees of right yaw. Further, increasing the LH throttle to MIL power and the application of full left rudder (achieved with less than full left rudder pedal because of the flight control laws), resulted in a right yaw rate of approximately four degrees per second. Increasing the LH throttle to maximum AB resulted in a right yaw rate of approximately 21 degrees per second.
1.16.2.3.2 Rudder deflection (in AFU mode) was found to vary as a function of AOA, where maximum rudder deflection (30 degrees left or right) occurred at 19 degrees AOA regardless of airspeed. Below 19 degrees AOA, maximum rudder deflection was not achievable, regardless of rudder pedal input, and decreased approximately linearly so that at 10 degrees AOA, only 12 degrees of rudder deflection was possible with full rudder pedal input. Therefore, without taking into consideration any other factors, the best AOA to fly the aircraft to achieve the lowest Vmca at a given power setting was at 19 degrees. This does not mean that 19 degrees is the Vmca, but that Vmca will occur at 19 degrees AOA.
1.16.2.3.3 The AOA to achieve maximum rudder deflection (19 degrees AOA) does not guarantee the aircraft will be able to maintain level flight or accelerate. The total force in the horizontal and vertical directions is also a function of power setting and AOA and there is an optimum point that balances the need to accelerate the aircraft longitudinally while minimizing the rate of descent/maximizing the rate of climb.
1.16.2.3.4 When assessing the Canadian recovery method, it was found that reducing the AOA from 25 degrees to 12 degrees while selecting MIL or maximum AB resulted in a compromise between the various factors; the need to accelerate, the need to minimize descent and the need to maintain directional control. At 12 degrees AOA, the aircraft was operating below Vmca; however, the longitudinal acceleration was such that the airspeed increased rapidly above Vmca before the aircraft had reached more than approximately 8 degrees of yaw after which the yaw decreased. Similarly for the descent rate, initially reducing AOA to 12 degrees resulted in an immediate sink rate; however, as the airspeed rapidly increased, so long as the AOA was maintained at 12 degrees, the sink rate declined and quickly reverted to a climb.
1.16.2.3.5 The altitude lost during recovery was found to be strongly dependent on power setting used (MIL or maximum AB) and pilot reaction time (time between engine failure onset and recovery initiation). Table 6 presents the relevant data.
|
Serial |
Pilot Reaction Time (seconds) |
Throttle Setting(Left/Right) |
Altitude Loss(feet) |
|
1 |
0 |
MIL / Off |
392 |
|
2 |
0 |
MIL / Idle |
369 |
|
3 |
0 |
maximum AB / Off |
168 |
|
4 |
0 |
maximum AB / Idle |
150 |
|
5 |
1 |
maximum AB / Idle |
230 |
|
6 |
2 |
maximum AB / Idle |
342 |
|
7 |
3 |
maximum AB / Idle |
662 |
Table 6: Altitude Lost During Recovery
1.17 Organizational and Management Information
1.17.1 CF188 Air Demonstration
1.17.1.1 The responsibility for providing the CF188 air demonstration pilot and aircraft alternates yearly between 3 Wing Bagotville and 4 Wing Cold Lake. The occurrence year's CF188 air demonstration was 3 Wing Bagotville's responsibility. Maintenance and support crews were provided by either 3 Wing Bagotville or 4 Wing Cold Lake, depending on the location of the air show. For Lethbridge, 4 Wing Cold Lake provided the maintenance and support crew.
1.17.1.2 The process for developing the CF188 air demonstration routine was undocumented. The air demonstration routine for the occurrence year was based on the previous year's routine, with modifications developed by the CF188 air demonstration pilot in consultation with the previous year’s CF188 air demonstration pilot. Analysis of the modifications was carried out by the current CF188 air demonstration pilot and evaluated in the ADCTS simulator and the aircraft. The manoeuvre package16 was then updated to reflect the modifications. A formal technical analysis for the air demonstration routine was never carried out by the Technical Airworthiness Authority (TAA) nor was a record of airworthiness risk management (RARM) completed.
1.17.2 Pilot Selection
1.17.2.1 The B-GA-100-001/AA-000, Chapter 11, Air Show Safety Criteria, Paragraph 6 required that unit Commanding Officers (CO) ensure that all personnel participating in an air show are sufficiently experienced and that they are current and properly trained for the mission and for the specific manoeuvres they are to perform. 1 Canadian Air Division Orders, Volume 2, 2-010, Air Displays, Paragraph 25, detailed the minimum pilot qualifications to become the CF188 Air show Demonstration Pilot and the selection process17:
a. The minimum qualification requirements were 500 hrs CF188 time, Element Lead and written recommendation by the individual's CO; and
b. Applications were to be forwarded to 1 Canadian Air Division Headquarters Winnipeg: Attention A3 FTR RDNS by 30 October.
1.17.2.2 According to the Air Demonstration Operations Order (3350-01 Spl Evt Coord, Operations Order CF188 Demo Team 2010), signed by the 1 Canadian Air Division Commander on 2 December 2009, 3 Wing Bagotville was required to select the 2010 CF188 demonstration pilot. Submission of applications by potential candidates was made through the pilot's CO.
1.17.3 Pilot Training
1.17.3.1 The Air Demonstration Operation Order stated that the 2009 demonstration pilot would deploy to 3 Wing Bagotville for two weeks to carry out training. Instead, initial training consisted of discussions and work-up ADCTS simulator training that was carried out at 4 Wing Cold Lake where the pilot was deployed for a period of time. The pilot returned to 3 Wing Bagotville to continue his training when the previous year’s show pilot became unavailable to assist. A decision was then made to have the 2008 CF188 air demonstration pilot (located in Bagotville) provide the requisite instruction.
1.17.3.2 There was no formalized manual of flying training and no training plan applicable to the CF188 air demonstration nor were there any documented currency requirements. In 1999, after recognizing there was a lack of information on how to plan, train and execute the CF188 air demonstration, the two CF188 air demonstration pilots consolidated their knowledge into a single document. This was titled "Flying the CF188 Demonstration Show" (known within the community as the Air Demonstration Handbook) and was dated 14 July 1999. It was never formalized by the CF as an official publication. This document was passed year-to-year, unchanged, to follow-on CF188 air demonstration pilots and was used by the occurrence pilot as a guide to flying the air show season.
1.17.3.3 Air show training for the 2010 season commenced in January and consisted of briefings, one-on-one discussions and ADCTS simulator usage. The Air Demonstration Handbook and instructor experience represented the key knowledge sources. The occurrence pilot conducted twelve ADCTS simulator training sessions between 15 January and 26 April 2010. The ADCTS simulator sessions consisted of executing the air demonstration routine and conducting air demonstration specific emergency procedures.
1.17.3.4 The air demonstration pilot flight training followed a build-up approach. The first two flights were introductory flights to allow the instructor an opportunity to refresh his air show demonstration skill-set. The instructor occupied the front seat and the occurrence pilot occupied the aft seat. After the two introductory training flights, the remaining flights were conducted either dual (with the instructor in the rear cockpit) or solo, depending on the nature of the flight. Flight training started with learning and practicing individual manoeuvres above 1,000 ft AGL, and then continued by combining the individual manoeuvres into the complete show routine. The occurrence pilot then performed one solo flight above 1,000 ft AGL, one ADCTS simulator sortie and one solo flight above 800 ft AGL prior to the air demonstration check flight. The check flight was flown dual at 700 ft AGL. Following the check flight, the occurrence pilot was authorized to carry out air demonstration training flights down to the minimum show altitude of 300 ft AGL. The occurrence pilot flew 18 air demonstration training flights between 4 May 2010 and 20 May 2010. During these flights, the occurrence pilot practiced the four types of air show routine identified in the manoeuvre package and also incrementally lowered the altitude of the show from 1,000 ft AGL down to 300 ft AGL.
1.17.4 Approval
1.17.4.1 The draft manoeuvre package was submitted to the Officer Commanding Fighter Standards and Evaluation Team (FSET), which was then reviewed and approved via e-mail. On 21 May 2010, the CF188 show was flown for the 1 Canadian Air Division Deputy Commander Force Generation (D Comd FG). Following the flight, a comprehensive briefing and question and answer period was conducted between the D Comd FG and the CF188 demonstration pilot. The final manoeuvre package, 1250-1 (2010 CF188 Demo Pilot) dated 4 May 10 and included as Annex F to the CF188 Demo Team 2010 OP Order, was then signed by the D Comd FG on behalf of the 1 Canadian Air Division Commander, granting the CF188 demonstration pilot authority to fly the profiles outlined in the manoeuvre package within Canada and the United States. The approval was in accordance with the process detailed in 1 Canadian Air Division Orders, Volume 2, 2-010. The pilot had performed the CF188 air demonstration a total of 16 times at various North American air shows prior to the occurrence.
1.17.5 Record of Airworthiness Risk Management
1.17.5.1 Prior to this occurrence, no RARMs were issued pertaining to the potential loss of engine performance as a result of an MFC failure. Following this occurrence, on 19 October 2010, a RARM (RARM-CF188-2010-009) was opened to assess the risk associated with a failed or unresponsive engine due to an MFC failure. The latest version of this RARM (Version Number 3) was assessed as Acceptable Level of Safety (ALOS) for all hazard scenarios considered and was signed off by DAEPM (FT) 2 on 3 June 2011.
1.17.5.2 The RARM described in detail the MFC manufacturer, the engine control system, operation of the MFC, MFC configuration history, reliability issues, and the maintenance plan to upgrade the MFC from G04 configuration to G08. The RARM then described the relevant hazard conditions, various hazard scenarios and the estimated hazard probability of each hazard scenario. Of note, the RARM did not include the air show as a specific hazard scenario for consideration.
1.17.5.3 The risk control plan accepted the risk as it was currently assessed to be ALOS. An engineering analysis was carried out to determine if a life threshold should be instituted dictating when MFCs are to be converted from a G04 to G08 configuration. This was done and it was determined that the time limit would be 3,000 hours since the last overhaul. The inherent risk of air show demonstration manoeuvres was said to be resolved by carefully selecting engines and MFCs of the latest configuration for the show aircraft.
1.18 Additional Information
1.18.1 The Air Demonstration Handbook
1.18.1.1 Chapter 3 Manoeuvres, paragraph 311, described how to perform the high AOA pass. “Though it is not stated anywhere, use of the radar display on the right DDI with no de-clutter selected will give the pilot a horizon line and a velocity vector that is still available and accurate (within 6 degrees of the horizon) at any AOA. This can enable the pilot to fly level at 25degrees AOA much easier than using the HUD as the HUD display is normally in error by 1½-2 degrees due to the high alpha and turbulent air around the AOA probes, ie: with 25degrees AOA the waterline must be placed at 26 ½-27 degrees nose up to maintain level. Of note is that turbulent air around the pitot tubes will cause them to indicate an altitude that is 150-200 ft higher than you actually are, therefore the pass should be flown at a minimum of 450 ft indicated”.
1.18.1.2 Former Canadian CF188 air demonstration pilots interviewed during the investigation indicated they were uncomfortable flying the high AOA pass at 300 ft AGL and said they tended to fly it higher, around 450 to 500 ft AGL. It was felt that 300 ft AGL was too low to safely affect recovery.
1.18.1.3 Chapter 5 Emergencies, paragraph 503, discussed how to deal with engine problems during the show routine. Subparagraph 2 stated “The biggest problem area for an engine problem would be during a high alpha pass. Do not select afterburner on the good engine because you do not have the rudder authority to control it at 25° AOA. You need to unload the jet then use afterburner once the alpha is off the jet. At 300 feet AGL a miss of the ground would occur within 50 feet. This is a good emergency for the simulator.” Interviews with a number of former CF188 air demonstration pilots indicated slight variations with their recovery procedures; however, in general, the procedures all consisted of simultaneously reducing AOA to below 12 degrees and selecting maximum AB while using rudder pedals and lateral stick inputs to maintain wings level and zero yaw18. During practice sessions in the simulator they indicated they were not able to consistently recover the aircraft before impacting the ground and felt 450 feet AGL was a better altitude.
1.18.2 US Navy High AOA Pass
1.18.2.1 Following the occurrence, the US Naval Air Systems Command (NAVAIR) was approached to ascertain what analysis was performed for the US Navy F-18 (models A through D) air show routine, in particular the high AOA pass, and to determine the guidance provided to air show pilots about how to conduct and recover from emergency situations.
1.18.2.2 USN F-18 air demonstrations have been cleared by analysis, simulation and flight testing. The process for approving an air demonstration clearance is the same as any other NAVAIR clearance; however, there has never been flight testing for the high AOA pass.
1.18.2.3 The high AOA pass is allowable within the limits of the naval air training and operating procedures standardization (NATOPS) manual, but air show clearances contain AOA and altitude limits with emergency recovery procedures in the event of an engine failure. Limits and emergency procedures were based on a simulator study conducted to assess pilot response of engine failures during the manoeuvre. The high AOA pass was evaluated in a high fidelity simulator in 199719. It found that 200 ft altitude loss was typical for both MIL power and maximum AB recoveries, but that there was a significant degree of variation between individual recovery attempts. Some MIL power and maximum AB recoveries lost as much as 350 ft.
1.18.2.4 The recovery technique used during the USN simulation study was to simultaneously select MIL power on both engines, apply opposite rudder to keep sideslip at zero, and reduce AOA to less than 15 degrees. Maximum AB was selected, if desired, only when AOA was less than 15 degrees. Engine failure scenarios tested were instantaneous, exponential and roll-back.
1.18.2.5 As a result of this evaluation, the USN high AOA pass was limited to 25 degrees AOA and 500 ft AGL.
1.18.3 SSK Deployment Doctrine Change
1.18.3.1 At the time of the occurrence, the CF doctrine regarding the release or retention of the SSK following ejection was predicated on the environment into which an aircrew was parachuting. In wooded terrain, aircrew were directed to retain their SSK to prevent entanglement of suspended equipment in the trees. A message from 1 Canadian Air Division dated 121957Z JUL 11 has since rescinded this requirement and directs aircrew to release the SSK under all circumstances.
1.18.3.2 A Defence Research and Development Canada (DRDC) report titled On Retaining the Seat Pack after Ejection when Landing in Trees (DRDC CORA TM 2010-238 November 2010), indicated “seat pack retention as the biggest landing injury risk factor, noting that five of eight reported major landing injuries involved retained seat packs.” Two factors were described as contributing to decreased injury when the seat pack is not retained.
a. “A deployed seat pack means the ejectee’s feet are still more than 6 m up when the PSP equipment hits the ground. With lanyard tension gone, the last second of fall sees the parachute further decelerate the ejectee toward a new slower fall speed. The already stopped seat pack mass and the lower descent velocity reduce the momentum (product of mass and speed) that the ejectee must dissipate through the impulse of ground contact during the PLF;” and
b. "To do a proper PLF, the ejectee should land on the balls of the feet with knees slightly bent and toes pointed 45 degrees to one side of the direction of travel, falling onto the outside of the calf and thigh and then rolling diagonally across the back. This proves easier to do safely without 15 – 20 kg hanging at the buttocks.”
1.18.3.3 An automatic deployment unit (ADU) automatically releases the SSK after a set time following ejection. Martin-Baker ADUs can be pre-set to operate from between three and six seconds. Specific timing is adjusted to ensure that at the extreme edge of the ejection seat envelope, the seat pack will be deployed prior to the aircrew reaching the ground.
1.19 Useful or Effective Investigation Techniques
1.19.1 Nil.
2 ANALYSIS
2.1 General
2.1.1 The following discussion addresses the technical aspects surrounding the failure of the RH engine, the departure of the aircraft from controlled flight during the high AOA pass and a number of other aspects of the CF188 air demonstration.
2.2 Technical Analysis
2.2.1 Engine Failure Conditions
2.2.1.1 The LH engine functioned normally throughout the flight until ground impact whereas the RH engine functioned normally for most of the flight. At some point after the pilot retarded both throttles to fight idle to decelerate and reposition for the high AOA pass the RH engine ceased to respond to throttle inputs but it continued to run at flight idle with no cautions activated. When both throttles were subsequently advanced, the RH engine failed to respond and remained at the flight idle RPM.
2.2.2 Possible Failure Modes
2.2.2.1 A review of design and maintenance publications associated with the GE F404 revealed multiple failure modes that could cause the RH engine to not respond to the throttle inputs in a manner similar to what was experienced in this occurrence. These modes include loss of electrical power to the RH engine due to failure of the engine mounted alternator, failure of the ECA, failure of a fan speed sensor, failure of wires or connections in the electrical wiring harness, a cracked P3 line/manifold, a hung stall condition or an MFC internal malfunction. Additionally, a disconnected throttle linkage would cause the engine to fail to respond to throttle movement; however, the VEN functioned normally in response to throttle inputs which indicates the throttle linkage was connected properly.
2.2.3 Unlikely Failure Modes
2.2.3.1 In the case of a failure of the ECA, engine mounted alternator or a fan speed sensor, or during a hung stall condition, the Caution and Warning system will generally provide an audible Caution indication and display an engine related Caution on the MFD. Since there were no cautions or warnings recorded by the MC, it is highly unlikely that any of these failure modes contributed to this occurrence.
2.2.3.2 A cracked P3 line/manifold produces symptoms identical to those experienced in this occurrence; however, since 1988 there have only been two confirmed cases in the CF where a leaking P3 line was implicated. Since 2006, 23 P3 manifolds have been rejected at third line maintenance but none were rejected because of cracks. The occurrence P3 lines were visually checked and pressure tested with no significant damage noted. The rarity of cracked P3 lines and the condition of the P3 lines on the occurrence engine suggest it is unlikely that a cracked P3 line contributed to this occurrence.
2.2.4 Indeterminate Wiring Failure Modes
2.2.4.1 A failure of the wires or connections in the engine electrical wiring harness could cause symptoms similar to those experienced in this occurrence; however, it was not possible to assess the serviceability of the wires and connections because of the extent of the post-crash fire damage to the aircraft.
2.2.5 MFC Failure Modes – Idle Adjustment Lever
2.2.5.1 A worn idle adjustment lever arm can result in flameout-low idle / flameout-roll back events; however, a worn idle adjustment will not likely inhibit engine response to throttle inputs.
2.2.6 MFC Failure Modes – Ratio Boost Piston
2.2.6.1 The MFC has had a long history of reliability issues and, over the years, the OEM introduced a number of ECPs to resolve these issues. In particular, an aluminum skirt introduced during the upgrade from a G03 to a G04 configuration was found to be wearing through movements of the ratio boost piston. During OEM laboratory tests it was noticed that very small wear patterns on the aluminum sleeve could cause the ratio boost piston to stick during quick throttle transitions that in turn could reduce the engine flameout margin or otherwise cause the engine to become erratic or unresponsive during acceleration / deceleration. The behaviour of the RH engine on 13 June 2010 and immediately prior to the accident is consistent with a stuck ratio boost piston.
2.2.6.2 Furthermore, the occurrence G04 configuration MFC was the third highest time G04 MFC in the CF and through technical investigation was found to have wear on the aluminum skirt to a degree that could cause the ratio boost piston to stick (see Annex A, Figure 9).
2.2.6.3 Therefore, considering all of the known factors deriving from the technical investigation, the investigation team concluded that the most likely reason the RH engine failed to respond to pilot inputs during the high AOA pass was because of an excessively worn ratio boost piston that became stuck as the pilot advanced the throttle from flight idle to sub-MIL.
2.2.7 Historical MFC Malfunction Modes
2.2.7.1 The MFC malfunction history primarily was due to problems with worn/out-of-tolerance MFC components. In particular, two components that have been shown to be problematic are a stuck ratio boost piston during throttle transients and a worn idle adjustment arm. A stuck ratio boost piston can reduce the engine flameout margin or otherwise cause the engine to become erratic or unresponsive during acceleration / deceleration. A worn idle adjustment arm can result in flameout-low idle / flameout-roll back events even though an actual flameout has not occurred. Superficially, the two issues display similar symptoms.
2.2.7.2 Identifying the specific cause of an MFC malfunction can be very difficult because the original problem is frequently not repeatable during troubleshooting activity. Shutting the engine down and restarting it removes and reapplies fuel pressure to the MFC. It is possible that this repositions out of tolerance/worn components, releasing a stuck ratio boost piston.
2.2.8 Specific Gravity Adjustment History
2.2.8.1 Lowering the specific gravity has been used successfully in the past to resolve flameout occurrences on start due to a flameout-low idle / flameout-roll back situation, but it can also mask more serious MFC problems since it is possible that the act of shutting the engine down and restarting can temporarily resolve the issue as discussed at paragraph 2.2.7.2. In those cases, the success of the procedure may have been inappropriately correlated with the specific gravity change, not the shutdown/restart event. This could have strengthened the belief that engine issues of this nature could be resolved by simply lowering the specific gravity and over time that could have become, to some degree, the cultural approach to resolving these types of issues. This belief was solidified when the technical authority temporarily sanctioned a procedure that included reducing the specific gravity setting.
2.2.8.2 For the occurrence engine/MFC, this belief may have played out as follows. The squadron had experienced flameout-low idle / flameout-roll back issues with a number of aircraft on a previous deployment to Salina, KS, where the TSR’s recommendation to lower the specific gravity appeared to prove successful. During the next deployment to Salina, KS, when flameout-low idle / flameout-roll back issues were again experienced with a number of aircraft, the technical authority granted the squadron temporary authority to deviate from the approved maintenance procedures and follow a procedure that included reducing the specific gravity setting. This appeared to prove successful in all cases where it was applied. In particular, during that deployment, the occurrence MFC/engine experienced two flameout-low idle / flameout-roll back events. For the first event, the pilot appeared to resolve the issue by employing the troubleshooting procedure outlined in the pilot’s checklist. For the second event, maintenance immediately lowered the specific gravity of both engines to the minimum prescribed. This appeared to resolve the issue. Then, approximately five months later when the aircraft was deployed to Quebec City, QC, to conduct an air demonstration, the occurrence MFC/engine experienced much more serious symptoms and would likely have led to replacement of the MFC had the approved troubleshooting been performed; however, because maintenance technicians may have been operating under the belief that lowering the specific gravity was an acceptable alternate procedure, they reset the specific gravity for both engines to the minimum prescribed for that fuel without following the approved troubleshooting procedures. This again appeared to resolve the issue and, as a result, a critical opportunity to remove a malfunctioning MFC from the aircraft was missed that resulted in a repeat malfunction during the high AOA pass at the Lethbridge air show accident.
2.3 Air Demonstration
2.3.1 Occurrence High AOA Pass
2.3.1.1 During manoeuvring flight, power modulations in the CF188 are applied by feel, rather than by looking directly at the engine instruments. The pilot continuously cross references the aircraft’s performance instruments (e.g. airspeed and altitude displayed in the HUD) to determine the appropriateness of any inputted power setting and will make additional throttle adjustments, if required, to home in on the proper throttle setting for the given manoeuvre. Unless there are overriding reasons to believe otherwise, the pilot will normally rely on the aircraft’s caution and warning system to first inform him of engine malfunctions.
2.3.1.2 During the occurrence high AOA pass, the RH engine continued to operate normally at idle regardless of throttle position, therefore, no RH engine-related cautions or warnings activated. Throughout the manoeuvre, the pilot assumed the RH engine was responding to throttle inputs. In the final stages of the high AOA pass, the pilot was concentrating on maintaining precise aircraft control by cross referencing the HUD, the RADAR display and the outside scene. He was not cross referencing the engine display that is located down by his left knee. During this stage of the high AOA pass, it is unreasonable to expect a pilot to continuously cross check the engine displays because the action of looking down into the cockpit would significantly detract from control of the other flight parameters.
2.3.1.3 Turbulence caused transient aircraft motions in pitch and yaw. These masked the thrust imbalance induced yawing moments and the initial sink actually associated with the RH engine malfunction. In the short time available, without the aid of the aircraft’s caution and warning system to positively alert the pilot of an engine-related issue, the pilot could not identify that a RH engine malfunction had occurred.
2.3.1.4 Aware there was something amiss, but having no idea what the specific problem was, the pilot initiated a maximum AB power overshoot to climb away from the ground and assess the situation. The large thrust imbalance, while below Vmca, generated a right yawing moment that exceeded the aircraft’s inherent yaw restoring moment. As a result, the aircraft rapidly departed controlled flight, the pilot ejected and the aircraft impacted the ground nose first in an inverted right wing down steep dive at approximately 25 degrees AOA.
2.3.2 High AOA Pass Considerations
2.3.2.1 General
2.3.2.1.1 The high AOA pass is flown within the two-engine flight envelope but the aircraft is operating outside the single engine flight envelope because it is operating below both MIL and maximum AB Vmca. The proximity to the ground and the directional control issues resulting from an engine failure at high AOA make timely identification of an engine failure a critical factor for successful recovery. There needs to be sufficient altitude available to allow the pilot to accelerate from below Vmca to above Vmca while minimizing the directional control problem. Of particular note, execution of the recovery procedure for an engine malfunction during the high AOA pass is predicated on the pilot actually identifying an engine malfunction.
2.3.2.2 Minimum Altitude
2.3.2.2.1 The USN performed a simulator evaluation to ascertain the safe minimum altitude for the high AOA pass. As a result of this evaluation, the USN established a 500 ft AGL minimum safe altitude to perform the 25 degree high AOA pass. The Canadian air demonstration high AOA pass was flown at the minimum authorized air show altitude of 300 ft AGL. There was no formal analysis or testing to support this as a safe altitude to perform the manoeuvre. The modelling and simulation effort conducted for this investigation identified that the altitude lost during recovery is non-linear in relation to the time taken between engine failure and the onset of recovery actions. Acknowledging the limitations of the modelling and simulation effort, it was found that initiating the MIL power Canadian recovery method (detailed in paragraph 1.16.2.2) within two seconds after the loss of one engine could result in a loss of approximately 350 ft of altitude. This was similar to the results of the USN simulator evaluation. If the pilot delays an additional second, the modelling and simulation data indicates that the altitude lost will be approximately 650 ft.
2.3.2.2.2 Based on the analysis of the modelling and simulation effort the investigation concluded the aircraft likely would have impacted the ground during any recovery attempt from the 300 ft AGL high AOA pass with this or any similar engine malfunction.
2.3.2.3 Recovery Procedure Considerations
2.3.2.3.1 Given the time it could take for the engine RPM to decrease from high RPM to below 60% RPM and trigger an engine caution (up to five seconds in an actual flameout event), it is likely that a pilot will not receive an engine malfunction indication with sufficient time to execute the proper recovery procedures. Furthermore, a pilot who is not alerted to an engine malfunction will likely delay execution of the correct recovery procedure until an alert is provided. Delayed execution of the proper recovery procedure increases the altitude lost during recovery and could result in the aircraft impacting the ground even after the proper procedure is carried out. While the USN performs the high AOA pass at 500 feet AGL, the investigation team’s analysis concluded it is not certain that this provides sufficient altitude for recovery if the pilot delays executing the recovery procedure.
2.3.2.4 With insufficient engine failure cues, the pilot might execute an inappropriate recovery procedure as this occurrence demonstrates and executing the wrong recovery procedure leads to the loss of the aircraft through a yaw-induced departure from controlled flight at low altitude.
2.3.2.5 Since execution of the correct procedure is predicated on alerting the pilot and since there is no guarantee that the pilot will be alerted, it is concluded that air demonstration pilots should be trained to respond to unusual situations during the high AOA pass as if they had experienced an engine failure. Implementation of this procedure eliminates the requirement for the pilot to identify an engine failure before executing the proper procedure.
2.3.3 Supporting Aspects
2.3.3.1 Technical Analysis
2.3.3.1.1 The investigation did not identify any previous CF technical analysis of the air demonstration.
2.3.3.1.2 Following the occurrence a CF188 fleet RARM was issued to specifically address the potential loss of engine performance as a result of an MFC failure, but the analysis did not include the air demonstration profiles as specific hazard scenarios. Even though the RARM did not include air demonstration profiles as specific hazard scenarios, it stated that the risk of an MFC failure during air demonstration manoeuvres could be resolved by carefully selecting engines and MFCs of the latest configuration (G08) for the air demonstration aircraft. Although it was not stated in the RARM, this mitigation procedure would also have to apply to any aircraft provided as a spare to the air demonstration aircraft.
2.3.3.1.3 This occurrence highlights the fact that the air demonstration is a unique operational mission that requires more formal technical and operational analysis than is present in a RARM to identify potential hazard areas (for more than just engine failure scenarios) and provide suitable mitigation procedures. The loss of engine performance as a result of an MFC failure constitutes only one potential engine failure mode that will have an impact during an air demonstration sortie. The risk from other modes of engine failure, as well as the failure of other critical systems such as flight controls and avionics, have not been formally identified and mitigated. Without a formal technical and operational analysis, a number of critical hazards that could lead to an accident are possibly being missed. The investigation concluded that a formal air demonstration RARM should be produced. This RARM should consider all the relevant technical and operational factors and provide guidance for normal and emergency situations that could be experienced during the various phases of the air demonstration similar to the air demonstration clearance produced by NAVAIR for the USN.
2.3.3.2 Training Manual
2.3.3.2.1 There is no formalized publication applicable to the CF188 air demonstration. In 1999, the two air demonstration pilots recognized this deficiency and elected to consolidate their knowledge into a single air demonstration manual titled “Flying the CF188 Demonstration Show.” While this document provided a source of information to later air demonstration pilots, the information has not been validated for correctness and it has never been formally accepted by the CF. One example of an error in the manual pertains to the high AOA pass. The manual states, “Of note is that turbulent air around the pitot tubes will cause them to indicate an altitude that is 150-200 ft higher than you actually are, therefore the pass should be flown at a minimum of 450 ft indicated.” This statement implies that the high AOA pass can be safely flown between 250 ft AGL and 300 ft AGL, which is below the authorized altitude for the air demonstration. Additionally, the DFS-led modelling and simulation activity and the USN air show clearance effort indicate that the aircraft can lose more than 300 ft AGL during a properly executed recovery. From this analysis, it is clear that 300 ft AGL is not a suitable altitude to conduct the high AOA pass if safe recovery of the aircraft is expected following an engine failure.
2.3.3.2.2 Furthermore, since the manual has not been updated since it was drafted in 1999 it is likely that valuable information and techniques learned since then have been lost because they have not been documented. Validating the accuracy of the information in the air demonstration manual and officially accepting it as a CF publication would enhance the quality and safety of CF188 air demonstrations by ensuring the information is up to date and correct.
2.3.3.3 Training Syllabus
2.3.3.3.1 There is no formalized training plan applicable to the CF188 air demonstration. The type and degree of training is directed by the instructor and is based upon the air demonstration training the instructor received during his tenure as the demo pilot. Because the training is not formally documented, it is likely that over time the training syllabus will vary and could result in insufficient or inappropriate training taking place. The investigation team concluded that implementing a formalized training plan for the CF188 air demonstration will eliminate this hazard.
2.3.3.4 Use of ADCTS simulator
2.3.3.4.1 The ADCTS CF188 flight model has not been validated for use where aircraft handling qualities and performance information is required. Therefore, the ADCTS should not be used as a definitive tool to determine safety parameters or aircraft response and handling techniques following system failures.
2.4 ALSE Issues
2.4.1 SSK Deployment
2.4.1.1 The occurrence SSK employed a manual deployment mechanism that required the pilot to reach behind the lower back with either hand and pull one of the two release handles. This allows the survival contents to fall free on an 8.2 m line that is attached to the pilot’s harness so that it strikes the ground before the pilot, which reduces the suspended weight and lowers the descent rate immediately prior to the pilot landing. Research studies conclude that landing with the SSK is a major contributor to parachute landing injury (see paragraph 1.18.3).
2.4.1.2 The time between initiating ejection and parachute landing was approximately five seconds. In this accident there was insufficient time for the pilot to recover from the ejection sequence and chute opening shock to allow him to reach behind and manually deploy the seat pack. A MB ADU can be adjusted to release the seat pack in as little as three seconds following ejection. Had the SSK been fitted with an ADU set to the minimum (three seconds), there would have been sufficient time for the system to release the SSK contents prior to the pilot landing. Without a system such as the ADU the pilot landed with the 34 lb pack still attached, which exacerbated his injuries.
2.4.1.3 Historically, the CF has taught aircrew to retain the SSK contents if landing in trees to prevent the survival contents snagging trees during descent. A study conducted by DRDC showed that the likelihood of injury caused by seat contents snagging trees was significantly less than the likelihood of injury when the SSK contents were retained. As a result of this analysis, a new CF doctrine has been established to always deploy the seat pack, regardless of the presence of trees; however, without an automatic deployment unit, pilots will likely have difficulty releasing their SSK contents following low altitude ejections which will increase the chances of landing injury.
3 CONCLUSIONS
3.1 Findings
3.1.1 The MFC has a long history of issues. Numerous ECPs have been implemented over the years in an attempt to resolve these issues. (1.6.3.7)
3.1.2 ECP-G404-E-91 (introduced as one of the modifications during conversion of the MFC from G04 to G08) was designed to address the ratio boost piston wear issue with material changes and ratio piston redesign. (1.6.3.7.4, 2.2.6.2)
3.1.3 At the time of the occurrence, the RH engine MFC was in a G04 configuration. (1.6.3.7.4)
3.1.4 The occurrence MFC had accumulated the third highest time on a G04 ratio boost piston. (1.6.3.7.5, 2.2.6.2)
3.1.5 The occurrence ratio boost piston was found with wear on the aluminum skirt to a degree that could cause the ratio boost piston to stick. (2.2.6.2)
3.1.6 A stuck ratio boost piston can reduce the engine flameout margin or otherwise cause the engine to become erratic or unresponsive during acceleration / deceleration. (1.6.3.7.4, 2.2.6.1)
3.1.7 There were no engine-related cautions or warnings displayed prior to the occurrence. (1.1.2, 1.11.6)
3.1.8 Other engine failure mechanisms would likely induce a caution or warning to be displayed. (2.2.3.1)
3.1.9 The occurrence engine had experienced three separate instances of flameout-low idle/flameout-roll back events, one within a month of the occurrence. (1.6.3.3)
3.1.10 It is possible that MFC internal mechanical failures related to ratio boost piston wear can be temporarily resolved by shutting the engine down and restarting. (2.2.7.2)
3.1.11 Concurrent lowering of the specific gravity, in response to a flameout-low idle/flameout-rollback event, may have led to an association that the resolution of the problem was the specific gravity change, not the engine shut down and restart. (2.2.8.1)
3.1.12 Some maintenance personnel may have been bypassing normal troubleshooting as prescribed in the CFTO by simply adjusting the specific gravity following a flameout-low idle/flameout-roll back event. (2.2.7.4)
3.1.13 During the high AOA pass, the aircraft was operating below Vmca. (1.16.2, 2.3.1)
3.1.14 The RH engine was unresponsive at idle when the pilot increased the throttles to reduce deceleration and capture the high AOA speed. (1.11.3.2, 2.3.1.2)
3.1.15 Inherent indications of RH engine failure were masked by turbulence. (2.3.1.3)
3.1.16 The pilot was unaware that the RH engine had failed to respond to throttle inputs. (2.3.1.3)
3.1.17 The pilot selected maximum AB on both engines to commence an overshoot. (1.1.2, 1.11.3.2, 2.3.1.4)
3.1.18 The large thrust imbalance while the aircraft was below Vmca caused a large right yaw to develop. (1.16.2, 2.3.1.4)
3.1.19 The aircraft departed from controlled flight. (1.1.2, 2.3.1.4)
3.1.20 No formal RARM or other analysis had been carried out for the CF188 air demonstration. (1.17.1.2, 1.17.5.1)
3.1.21 There is neither a formalized manual of flying training, a training plan, nor documented currency requirements applicable to the CF188 air demonstration. (1.17.3.2)
3.1.22 The ADCTS CF188 flight model has not been validated for use where aircraft handling qualities and performance information is required. (1.16.1.1, 2.3.3.4.1)
3.1.23 The SSK had a manual deployment mechanism only. (1.6.5.2, 2.4.1.1)
3.1.24 There was insufficient time for the pilot to manually release the SSK contents prior to the parachute landing. (2.4.1.2)
3.2 Cause
3.2.1 The engine malfunction was likely the result of a stuck ratio boost piston in the RH engine MFC that prevented the engine from advancing above flight idle when maximum AB was selected. The large thrust imbalance between the left and the right engines then caused the aircraft to depart controlled flight; the aircraft was unrecoverable within the altitude available. (2.2.6)
3.2.2 Contributing to the occurrence was the subtle nature of the engine malfunction that was not detected by the pilot when the overshoot was attempted. Even if the pilot flew a suitable recovery procedure, it is likely that there would have been insufficient altitude to recover the aircraft prior to ground impact. (2.3.1)
3.2.3 Possibly contributing to the occurrence was a maintenance culture of incorrectly associating resolution of MFC-related issues with the lowering of the specific gravity to the minimum allowed for the fuel type used. This may have resulted in CFTO-prescribed troubleshooting procedures not being conducted when these types of issues arose. As a result, an earlier engine malfunction that would likely have resulted in the RH MFC being changed may have inadvertently been allowed to continue in service. (2.2.7, 2.2.8.1)
4 PREVENTIVE MEASURES
4.1 Preventive Measures Taken
4.1.1 The Lethbridge International Air Show Association took the following steps for the air demonstration at the Alberta International Air Show:
a. All emergency vehicles had the required equipment for direct communication with the Air Boss on an aviation band radio;
b. Personnel with air-side vehicle operations qualification were assigned to each emergency vehicle; and
c. CFR vehicles were positioned forward of the crowd line to minimize response time.
4.1.2 The Lethbridge International Air Show Association re-introduced the requirement for air demonstration crews to brief the Alberta International Air Show Emergency Response crews on the unique nature of their aircraft and how to extract them from their aircraft in the event of an emergency.
4.1.3 1 Canadian Air Division made the decision to cease performing the CF188 air demonstration for the 2010 season until suitable preventive measures were implemented.
4.1.4 1 Canadian Air Division removed the high AOA pass from the CF188 air demonstration for the year 2011/2012 (RDIMS # 1097867).
4.1.5 Message FT23030 281016Z Jul 10 was issued to re-affirm that if an engine flameout caution occurs, troubleshooting is to be carried out as per C-12-188-270/NH-000 and that deviating from specific gravity settings detailed in C-12-188-PCM/MB-001 is not authorized.
4.1.6 RARM CF188-2010-009 (Engine Unresponsive or Lost Due to MFC Failure) identified that the following maintenance actions had been taken:
a. All MFCs will be converted to G08 configuration once inducted into third line maintenance;
b. A seal refresh program has been implemented for MFCs repaired at third line;
c. The Idle Adjustment Lever (p/n 2555121 is being replaced on all MFCs inducted for deep repair at Third Line if showing any sign of wear; and
d. A new Component Improvement Program Engineering Program Development was tabled with the USN to investigate MFC reliability and lead to a “get well program” for the MFC.
4.1.7 Record of operational risk management (RORM)-CF188-2012-001, 9 February 2012 (Re-introduction of the High Alpha Pass into the CF188 Air show) was signed by the Commander 1 Canadian Air Division on 29 February 2012 authorizing the reintroduction of the high AOA pass into the CF188 air demonstration for the 2012 air show season with the following mitigating procedures that reduce the risk from high to low:
a. The manoeuvre will be flown at 500 ft AGL along the 1000 ft show line vice the previously established 300 ft AGL minimum altitude along the 500 ft show line;
b. Recovery will be as follows:
i. Simultaneously select MIL power on both engines, opposite rudder to keep sideslip at zero;
ii. Reduce AOA to max 15 degrees;
iii. Select maximum AB, if desired, only when AOA is less than or equal to 15 degrees;
iv. For MIL or max AB recoveries, climb with AOA max of 15 degrees;
c. Given the fact that standard aircraft cautions and warnings will not provide adequate indication of a loss of thrust in a timely manner, a recovery will immediately be initiated if any of the following occur:
i. AOA reaching 27 degrees without indications of positive corrections (minor transients up to 27 degrees are allowed provided immediate corrections are applied and the aircraft responds);
ii. Altitude reaches 450 ft AGL with corrections applied and aircraft not responding;
iii. Un-commanded heading change of +/- 15 degrees;
iv. Landing Gear Warning tone remains on after positive corrections have been applied;
d. The recovery procedure will be flown the same regardless of the reason for its initiation (normal end of the pass or abnormal situation). By conducting this type of recovery during the normal operation of the aircraft, the aircrew will develop the procedural memory motor skills to apply in any potential abnormal situations; and
e. A CF188 air demonstration manual captures all the above comments, as well as applicable information in the emergencies chapter to highlight potential complications associated with the manoeuvre in the event of a loss of thrust, and the requirement to immediately recover if anything seems abnormal.
4.1.8 Director General Aerospace Equipment Program Management (DGAEPM) is in the process of upgrading MFCs from G04 to G08. It is expected that all MFCs will be upgraded by August 2015.
4.1.9 1 Canadian Air Division has directed that the CF188 demonstration aircraft will be equipped with G08 MFCs. Should the main or backup demonstration aircraft not be equipped with G08 MFCs, then the high AOA pass will not be flown.
4.1.10 1 Canadian Air Division has validated, updated and officially published the draft CF188 air demonstration manual as the Fighter Demo Manual. The Fighter Demo Manual includes training and currency requirements.
4.2 Preventive Measures Recommended
4.2.1 It is recommended that 1 Canadian Air Division formally assess all the manoeuvres of the CF188 air demonstration routine to identify areas of elevated risk and to incorporate risk mitigation procedures if required. (2.3.3.1)
4.2.2 It is recommended that DGAEPM upgrade the NACES SSK with an ADU. (2.4.1)
4.3 Other Safety Measures Recommended
4.3.1 It is recommended that 1 Canadian Air Division ensure that there is an air demonstration operational and technical clearance, an air demonstration training program and an air demonstration standard manoeuvre manual for all aircraft that conduct air demonstrations.
4.4 DFS Remarks
4.4.1 As this occurrence highlights, many factors must combine at the right time to result in an occurrence. While it is not possible to identify them all, we must do everything within our ability prepare for their eventuality, such as by following standard rather than ad hoc troubleshooting, for example. However, we must also protect our personnel from undue hazards through the analysis of our procedures to eliminate risk, such as avoiding vulnerable flight profiles at altitudes that preclude a safe egress. When unforeseen events unfold, they can do so very quickly so be prepared to act accordingly.
4.4.2 The good news is that, in this case, our technical and operational systems were responsive to the need for change and, as a result, we now have a much safer operation.
J.C.Y Choiniere
Colonel
Director of Flight Safety
End Notes
1. HUD flight information consists of heading, pitch and roll attitude, airspeed, altitude, vertical speed and AOA.
2. At 25 degrees AOA, the forward view was obscured by the nose of the aircraft.
3. MIL is selected by advancing the throttle to the MIL power stop. At this throttle position the engine core is operating at maximum power without AB selected.
4 AB is selected by advancing the throttle beyond the MIL power stop. When AB is selected it can be smoothly modulated between minimum and maximum AB.
5. Refer to the CF188 Aircraft Operating Instructions for further details about the FCS.
6. “Flameout-low idle” is when the HPC speed (N2) decreases below 60% revolutions per minute (RPM), which activates the FLAMEOUT caution, and stabilizes at a sub-idle RPM.
7. “Flameout-roll back” is when N2 decreases below 60% RPM, which activates the FLAMEOUT caution, and does not stabilize at a sub-idle RPM but continues to decrease.
8. The CF349 is the Aircraft Unserviceability Record that pilots and/or technicians use to record system faults
9. N2 is displayed to the pilot in the cockpit as a percentage of the maximum allowed N2 measured in RPM.
12. According to the CF188 Aircraft Operating Instructions, Ground idle N2 RPM – 61 to 72%, flight idle N2 RPM – 68 to 73%
13. Mil-STD-3013 defines Vmca as the airborne minimum controllable airspeed with maximum thrust at which an engine can fail and control can be maintained for a specified altitude, weight, and configuration.
14. The Canadian recovery method was based on discussions with former air demonstration pilots.
15. The occurrence engine was operating at flight idle; however, the difference in thrust between off and flight idle is minimal.
16. The manoeuvre package documents the various manoeuvres and restrictions that will be flown during the Air Demonstration.
17. 1 Canadian Air Division Orders were updated on 21 Dec 11 and the CF188 Air Demonstration pilot qualifications were moved to Volume 2, 2-010, Air Displays, Paragraph 35.
18. This procedure was titled the “Canadian Recovery Method” for the purposes of conducting the Modelling and Simulation work as described at paragraph 1.16.2.
19. A limited set of simulator evaluation data was provided by NAVAIR to DFS. The specific data has been omitted from this report for proprietary reasons.
Annex A
Figures
Figure 1: Top view of the airport as it was set up for the air show.
Figure 2: View along intended flight path (approximately along the 500 ft show line).
Figure 3: Both throttles set to flight idle and both VENs wide open.
Figure 4: Both throttles set to MIL power and both VEN fully close.
Figure 5: Left engine selected to maximum AB and engine responded appropriately. Right engine selected to maximum AB but engine remained at approximately flight idle RPM. Flight control positions corresponded to the pilot inputs required to counteract right yaw and right roll. In the CF188 at high AOA, right roll is a natural aircraft response to right yaw.
Figure 6: Early phase of the ejection sequence. The canopy can be seen exiting the figure on the left hand side.
Figure 7: Ground impact. In this figure, the aircraft is yawing rapidly to the right and rolling to the right about the center of gravity and the direction of flight is as indicated because the aircraft is at approximately 25 degrees AOA. The depicted location of the center of gravity and the various axis of motion are approximate and are presented for illustration purposes.
Figure 8: Ground impact layout.
Figure 9: MFC ratio boost piston. The ratio boost piston from the occurrence MFC is presented to the left of the figure and the views to the right look down as depicted. The top right image is of a ratio boost piston from a representative aircraft (CF188720) and the bottom right image is of the occurrence ratio boost piston (CF188738). In both images, the wear pattern is repeated every 120 degrees around the circumference (not visible in the photos). The wear pattern of the occurrence ratio boost piston appears to protrude radially inwards; however, this is an artefact of the photograph. The wear pattern actually protrudes radially outwards into the metal.
Annex B
Abbreviations
|
A8 |
VEN throat area |
|
|
AB |
afterburner |
|
|
ADCTS |
advanced distributed combat training system |
|
|
ADD |
air display director |
|
|
ADU |
automatic deployment unit |
|
|
AF |
airframe |
|
|
AFU |
auto flap up |
|
|
AGL |
above ground level |
|
|
AETE |
Aerospace and Engineering Test Establishment |
|
|
AF |
airframe |
|
|
ALOS |
acceptable level of safety |
|
|
ALSE |
aviation life support equipment |
|
|
AMS |
air maintenance squadron |
|
|
AMU |
advanced memory unit |
|
|
AOA |
angle of attack |
|
|
AOI |
aircraft operating instructions |
|
|
ARES |
all-purpose real-time engineering simulator |
|
|
ATC |
air traffic control |
|
|
ATESS |
Aerospace and Telecommunications Engineering Support Squadron |
|
|
BIT |
built-in test |
|
|
BAES |
Bombardier Aerospace Engineering Services |
|
|
CF |
Canadian Forces |
|
|
CO |
Commanding Officer |
|
|
CVR |
cockpit voice recorder |
|
|
CYQL |
Lethbridge County Airport |
|
|
DAEPM (FT) |
Director Aerospace Equipment Program Management Fighters and Trainers |
|
|
D Comd FG |
Deputy Commander Force Generation |
|
|
DFS |
Directorate of Flight Safety |
|
|
DGAEPM |
Director General Aerospace Equipment Program Management |
|
|
DRDC |
Defence Research and Development Canada |
|
|
ECA |
electrical control assembly |
|
|
ECP |
engineering change proposal |
|
|
EFH |
engine flight hours |
|
|
EGT |
exhaust gas temperature |
|
|
EPM |
emergency procedures manual |
|
|
ERS |
Emergency Response Services |
|
|
FCC |
flight control computer |
|
|
FCS |
flight control system |
|
|
FOD |
foreign object debris |
|
|
FPI |
fluorescent penetrant inspection |
|
|
FDR |
flight data recorder |
|
|
FSET |
Fighter Standards and Evaluation Team |
|
|
FSS |
flight service station |
|
|
ft |
foot/feet |
|
|
g |
normal acceleration |
|
|
g/ml |
grams per millilitre |
|
|
GE |
General Electric |
|
|
HPC |
high pressure compressor |
|
|
HPT |
high pressure turbine |
|
|
HSE |
Health, Safety and Environment |
|
|
HUD |
head up display |
|
|
IGV |
inlet guide vane |
|
|
JHMCS |
joint helmet-mounted cueing system |
|
|
KCAS |
knots calibrated airspeed |
|
|
lbs |
pounds |
|
|
LH |
left-hand |
|
|
LPC |
low pressure compressor |
|
|
LPT |
low pressure turbine |
|
|
M |
meter |
|
|
MB |
Martin-Baker |
|
|
MC |
mission computer |
|
|
MF |
mandatory frequency |
|
|
MFC |
main fuel control |
|
|
MFD |
multi-function display |
|
|
MIL |
military power |
|
|
MOR |
manual override handle |
|
|
MSDRS |
Maintenance Signal Data Recording Set |
|
|
MSP |
maintenance status panel |
|
|
N1 |
fan speed (in RPM) |
|
|
N2 |
HPC speed (in RPM) |
|
|
NACES |
navy aircrew common ejection seat |
|
|
NATOPS |
naval air training and operating procedures standardization |
|
|
NRC |
National Research Council |
|
|
NVG |
night vision goggle |
|
|
OEM |
original equipment manufacturer |
|
|
P0 |
ambient pressure |
|
|
P3 |
compressor discharge pressure |
|
|
PA |
powered approach |
|
|
PCMCIA |
personal computer memory card international association |
|
|
PLA |
power lever angle |
|
|
PSI |
pounds per square inch |
|
|
RARM |
record of airworthiness risk management |
|
|
RH |
right-hand |
|
|
RORM |
record of operational airworthiness risk management |
|
|
RPM |
rotation per minute |
|
|
SFOC |
special flight operating certificate |
|
|
S/N |
serial number |
|
|
SOP |
standard operating procedures |
|
|
SSK |
seat survival kit |
|
|
T1 |
fan inlet temperature |
|
|
T2.5 |
HPC inlet temperature |
|
|
T5 |
exhaust gas temperature |
|
|
TAA |
Technical Airworthiness Authority |
|
|
TSR |
technical service representative |
|
|
UHF |
ultra high frequency |
|
|
VEN |
variable exhaust nozzle |
|
|
VHF |
very high frequency |
|
|
USN |
United States Navy |
|
|
UTC |
coordinated universal time |
|
|
VMC |
visual meteorological conditions |
|
|
Vmca |
minimum controllable airborne airspeed |
|
