SEO Title
Final Report: Wing succumbed to fire
Subtitle
<strong>FAIRCHILD/SWEARINGEN SA-226-TC METROLINER, MONTREAL, QUEBEC, JUNE 18, 1998</strong>–Often layers of events come together to create catastrophes.
Subject Area
Channel
Teaser Text
<strong>FAIRCHILD/SWEARINGEN SA-226-TC METROLINER, MONTREAL, QUEBEC, JUNE 18, 1998</strong>–Often layers of events come together to create catastrophes.
Content Body

FAIRCHILD/SWEARINGEN SA-226-TC METROLINER, MONTREAL, QUEBEC, JUNE 18, 1998–Often layers of events come together to create catastrophes. Pilots may handle individual layers successfully, but combined those layers are often insurmountable. Nine passengers and two pilots died after the left wing of a Metroliner failed during an emergency landing at Mirabel/Montreal International Airport (CYMX). The pilots, attempting an emergency landing less than 25 minutes after departure, were unaware a fire was burning in the wheel well and had  focused their attention instead on the left engine and flight control difficulties.

Propair Flight 420 departed Dorval/Montreal International Airport (CYUL) at 0700 EDT, on an instrument flight plan, destined for Peterborough, Ontario. During the takeoff run the crew noticed the Metroliner pulled to the left and that considerable right rudder was required to keep the airplane on the runway centerline. ATC cleared the crew to climb to 16,000 ft on departure. Climbing through 12,500 ft, the crew advised ATC of a hydraulic problem and requested to return to Dorval. Dorval’s weather at the time of departure was 600 ft overcast, with three miles visibility in fog and northeast winds at 4 kt. The temperature and dewpoint were both 17 deg C.

ATC granted the pilots’ request immediately and cleared the crew to reverse course and descend to 8,000 ft. About 70 seconds later they initiated the turn and said that, for the moment, they did not have an emergency. Seconds later they realized the controls were not responding properly. They had the first indication of trouble in the left engine in the ensuing minutes, followed by illumination of the left wing-overheat light. The light went out within 30 seconds but (the CVR revealed) not as a result of the crew’s accomplishing any checklist items.

Two temperature sensors for the overheat alarm reside in each wing; one is located in the upper outer part of the wheel well near the air intake duct, the other on the alternator wiring harness on the inboard wing leading edge. If overheating is detected in the leading edge the wing ovht light for that side will flash. If the overheat condition is detected in the wheel well the light illuminates and remains on. Temperatures must exceed 350 deg F in the nacelle or 450 deg F in the cabin air conditioning ducts for the light to illuminate. According to the aircraft flight manual (AFM), a steady light requires the bleed air to be shut off on the affected side and the landing gear extended. If the light goes out, the gear can be retracted. After three minutes with the light on, the engine on the affected side must be shut down.

The AFM makes no mention of the possibility of a fire in the wheel well. It does say that with the gear lowered and one engine shut down, the aircraft will not maintain a level attitude.

At 0718:12 the left engine appeared to be on fire, so the crew shut it down; the captain then took over the controls. Again, the pilots noted the flight controls’ lack of responsiveness. At 0719:19 they advised ATC the left engine was secured. ATC suggested they divert to Mirabel instead of Dorval, and the pilots agreed. Less than two minutes later the crew told ATC they had flames coming out of the “engine nozzle.” Their preparations for an emergency landing included a review of the emergency gear extension procedure.

The crew was having increased difficulty controlling the Metroliner, and by this time they had full aileron trim set. They called ATC at 0723:10 to say that the left engine fire was out; three minutes later they reported the fire had started again. On short final for Runway 24L at Mirabel, the crew lowered the landing gear but saw only two green lights in response. As they crossed the threshold, the left wing failed upwards, the aircraft rolled more than 90 deg to the left and crashed, inverted on the runway. According to the report, the aircraft immediately caught fire and continued to slide for another 2,500 ft before it came to rest on the left side of the runway. Firefighters, positioned near the runway threshold, quickly brought the fire under control. Despite their efforts, no one made it out of the wreckage.

Through autopsies and toxicological testing, investigators found that both pilots and most of the passengers had died of impact forces. The autopsies showed the pilots were free of drugs and alcohol. Two passengers apparently survived the initial impact but died of extensive burns. One passenger died “as a result of inhalation of combustible products…a lethal dose of hydrogen cyanide causing rapid cerebral asphyxia.” According to the report, “incomplete combustion of synthetic foam normally gives off hydrogen cyanide.” The aircraft met Part 23 regulations which (unlike Part 25) do not require a protective layer, to retard combustion.

Both pilots were 35 years old and held airline transport pilot certificates. The captain had 6,515 hr total time and 4,200 hr in type, while the first officer had 2,730 hr total time and 93 hr in type. On the day before the accident, the captain reported for duty at 1035 and left his office at 1935. He reported for duty at 0545 on the day of the accident. The first officer worked an 11-hour day, finishing at 1930 the day before the accident. He too reported for work at 0545 on the accident day. Reports show each pilot slept well and appeared fit for flight.

Propair operates a fleet of turboprop aircraft across Quebec, including the King Air 100 and 200, Fairchild Swearingen Metro SA-226, Gulfstream GI and Cessna 425 Conquest. The company retains 100 full-time employees and makes more than 3,500 flights per year.

Investigators calculated the normal takeoff distance, based on the aircraft’s weight and the conditions of the day, to be 1,800 ft. Both the manufacturer and FlightSafety (FSI) confirmed the investigators’ estimates from the AFM, with FSI using an SA-226 simulator. Observers of Flight 420’s takeoff roll said the aircraft lifted off between the A1 and A2 taxiway on Runway 24. Depending on the actual brake release point, which could not be confirmed, they figured the Metro used between 4,300 ft and 5,400 ft to get off the ground. They determined the takeoff run lasted 27.5 seconds and, using time versus acceleration calculations, verified the distance as slightly more than 4,000 ft.

According to the Transportation Safety Board (TSB) report, investigators found several databases documenting problems with SA-226 and SA-227 Metroliner brake systems, several involving brake overheating and fires. Two such occurrences they researched in detail because of the similarity to the Propair accident. “On July 27, 1988, a Peninsula Airways Fairchild SA-227-AC Metroliner III lost hydraulic pressure, the left wing overheat warning light came on, the tires burst and the left wheel well sustained severe fire damage. The crew was able to make an emergency landing at Anchorage International Airport, Alaska. On February 10, 1990, a Perimeter Airlines Swearingen SA-226-TC Metroliner II lost hydraulic pressure, the left wing overheat warning light came on, the tires burst and the left wheel well sustained severe fire damage. The crew shut down the left engine and was able to make an emergency landing at Winnipeg International Airport, Manitoba.”

In Propair Flight 420’s case, investigators determined the overheating of the left landing gear brakes–which were stuck on–during taxi and takeoff caused the fire in the nacelle after gear retraction. The fire then spread within the wing structure, ultimately causing the wing to fail.

The SA-226 is equipped with two parking-brake valves, one for each set of brakes in the cockpit. Previous owners of this aircraft complied with AD 92-01-02, which set out to correct the problem of residual brake pressure caused by the parking brake control cable. Fairchild Facts newsletter published an article in 1993 about the Metro II parking brakes, saying, “Before taxiing or beginning the takeoff roll, ensure that the parking brake is fully released. Just moving the parking brake control knob to the off position is not enough to ensure release of the brakes. As stated in the AFM, the system requires pressure be applied to the brake pedals to fully release the parking brake since some residual pressure can remain even with the knob in the off position. Taxiing and taking off with brake partially engaged can result in any or all of the following:

• Increased power necessary to taxi.

• Longer or much longer takeoff rolls (possibly longer than available runway).

• Hot, burned or seized brake components.

• Possible fire in the main gear well.

• Tire failure on takeoff or the next landing.

• Overheated hydraulic fluid streaming overboard from the vent located near the nosewheel well.

• Some pilots make it a habit to press the brake pedals firmly to ensure that the parking brake is fully released before they add power and get the aircraft moving.
TSB investigators found Metro II crews didn’t trust the parking brake and made a habit of using wheel chocks for engine start. They were unable to figure out whether this crew used wheel chocks for the start. They did conclude, through numerous tests, “that residual pressure as low as 50 psi in the brake system could cause dragging and subsequent overheating of the brake components during the takeoff run.” Overheating of the brakes and in turn the brake lines caused the fluid to expand, thereby increasing pressure in the system. The TSB felt it was unlikely the crew would notice a braking pressure of 50 psi during takeoff.

According to the report, the AFM for the SA-226 did not provide information on “the possibility of brake overheating, precautions to prevent brake overheating, the symptoms that could indicate brake problems or actions to take if overheated brakes are suspected.” In response to the TSB’s recommendations, Fairchild released “FAA-approved airplane flight manual changes” that expand the “wheel well and wing-overheat-light on” information in the emergency procedures section.

Investigators also found the hydraulic fluid used in this aircraft was not pure MIL-H-83282 as required. Instead, they found a mixture of two types of hydraulic fluid. The combined fluids had a lower flashpoint, prompting earlier ignition. While the aircraft maintenance manual said the two hydraulic fluids used were compatible, it did not mention that mixing them would reduce their fire resistance. The TSB issued a recommendation and Transport Canada issued an airworthiness notice titled “Hazards of Incorrectly Identifying or Mixing Aircraft Fluids,” with the hope of preventing the mixing of hydraulic fluids in the future.

In conclusion, the TSB found that the extreme heat generated by the friction of the left dragging brake caused a brake seal to fail, allowing brake fluid to leak onto hot brake components and (because of the lowered flashpoint) ignite, causing the fire in the wheel well. They found many incidents and a few accidents involving SA-226/227 landing gear-failures, flat tires, wheel fires and loss of control on the ground. Despite changes to the AFM and bulletins sent out to pilots of these aircraft, the TSB stated “flight crews are still not provided with an unambiguous alert of a dragging brake condition caused by residual hydraulic pressure in the brake system.” Without such information, crews are unable to rectify the situation in a timely manner and the TSB considers there is still a high risk of fire “with possible ensuing loss of life and property.” Therefore, in issuing the final report on Propair Flight 420, the TSB recommended “Transport Canada, the [FAA] and Fairchild explore options for SA-226 and SA-227 aircraft to be equipped with a brake pressure warning indicator for each main wheel brake system.”

Expert Opinion
False
Ads Enabled
True
Writer(s) - Credited
Publication Date (intermediate)
AIN Publication Date
----------------------------