Post-maintenance stall tests raise safety concerns

 - September 13, 2006, 7:27 AM

An incident on May 4 this year in which a Hawker 800A lost more than 10,000 feet during a post-maintenance stall test underscores the potential risks of stalling swept-wing jets. The Hawker recovered after what the pilot said was a four- to five-g pullout.

The airplane had undergone maintenance at Duncan Aviation in Lincoln, Neb., before the flight, according to NTSB investigator-in-charge Tim Sorenson. Raytheon Aircraft owned the Hawker, having taken it as a trade on a sale, a Raytheon Aircraft spokesman told AIN. The Hawker had undergone a pre-purchase inspection, though neither Raytheon nor Duncan Aviation would comment on whether Duncan mechanics had done any work on the Hawker’s wing.

The Hawker maintenance manual requires the stall test after installation of a wing leading edge assembly, installation of two or more TKS wing-distribution panels on one side or replacement of either one-piece titanium leading-edge section. According to the maintenance manual, “the airplane must be test flown by a pilot familiar with the stall identification system and stall characteristics of the 800 series and the 800XP series.”

During the stall test, there were two pilots and four passengers on board. The pilot told the NTSB, “The flight was entering a stall series in accordance with the test-flight procedures.” In the existing conditions and at the aircraft’s weight, calculated stick-shaker speed was 115 knots and stick-pusher 107.5 knots, followed by aerodynamic stall at 105.5 knots. “As the airplane slowed through [approximately] 126 knots [indicated airspeed],” the pilot told the NTSB, “it abruptly rolled off/ dropped the right wing and the nose fell rapidly.”

The pilot said that the airplane rolled right and left five to seven times, entered a cloud layer at about 12,000 feet msl then exited the clouds at about 10,000 feet and “descend[ed] vertically. I neutralized the ailerons with the yoke and began a higher than normal back-pressure pullout, experiencing [approximately] four to five gs. The aircraft responded, and we stopped the descent somewhere below 7,000 feet.”

The NTSB preliminary report includes this statement about the pilot’s use of the Hawker’s autopilot: “He noted that, although the autopilot was on as required by the test procedure, he was holding the control wheel and felt ‘no vibration or abnormal indication’ before the event.”

The Hawker maintenance manual doesn’t include the stall procedure, only the requirement to perform the test according to the “Flight Manual Section 4.10.” For the Hawker 800A, Subsection 4.10, page 9 lists the stall procedure. Item number two of the procedure says, “The autopilot must be disengaged.”

A temporary revision (No. 28) in the 800XP flight manual includes the same instruction regarding the autopilot’s being disengaged and specifies that the airplane must be 10,000 feet above clouds. Both the 800A and 800XP manuals require at least 10,000 feet minimum altitude. AIN asked Raytheon if there is a similar revision in the 800A flight manual, and the Raytheon spokesman responded, “There is one out there. I do not know if it was in this aircraft’s flight manual.”

During the May 4 stall test, the Hawker began the maneuver at 17,000 feet, 5,000 feet above the 2,000-foot-thick cloud layer through which it descended before recovery.

The two pilots and four passengers received minor injuries, according to the NTSB. After recovering from the stall, the pilots landed the Hawker using no flaps “without further incident.”

A Duncan Aviation spokeswoman said that the company didn’t know enough to comment on the incident. “We are helping with the investigation,” she said. “At this point, we can’t release anything.” AIN learned later from another source that Duncan mechanics inspected the Hawker after the incident and found no damage.

Hawker Stall Accident History

An accident during stall maneuvers on a Hawker training flight Sept. 20, 2003, near Beaumont, Texas, resulted in three fatalities. An instructor-pilot was preparing two pilots for Part 135 flight checks. At 5,000 feet, after doing steep turns and approaches to stalls, the instructor asked one to demonstrate an approach-to-landing stall. The NTSB report noted the pilots’ conversations, recorded on the cockpit voice recorder. “The first pilot asked the instructor-pilot if he had ‘ever done stalls in the airplane.’ The instructor-pilot replied, ‘It’s been a while.’ The first pilot remarked, ‘This is the first time I’ve probably done stalls in a jet. Nah, I take that back, I’ve done them in a [Learjet].’”

Witnesses reported seeing the Hawker performing erratic maneuvers, seeming to stop in midair and pitching nose down. Probable cause of the accident, according to the NTSB, was “the first pilot’s failure
to maintain aircraft control and adequate
airspeed. Contributing factors included performing intentional stalls at too low an altitude to afford a safe recovery, the pilot’s failure to add power in an attempt to recover and the flight instructor’s inadequate supervision of the flight.”

Former type-rating examiner Derek Jordan, who has 10,500 hours in Hawkers, has stalled the type many times, and on one occasion learned that anomalies do occur. Jordan told AIN that while he was performing stalls during a training flight in a Hawker 3B, “the aircraft flipped and spun from 16,000 feet down to [the] recovery height of 2,700 feet. Mine was a flat spin with between 60 and 80 knots indicated airspeed. [There were] violent aileron/elevator movements, which were difficult to hold. I operated airbrakes, various flap settings and power before normal spin recovery finally got me out of the spin.”

While conducting the wing-maintenance stall check in November 2000 in a Hawker 700, he wrote, “I needed six flights to correct the flick [roll at the stall]. An engineer came with us on the second flight to witness the violent flick. He came again when it was nearly fixed to see the improvement. It seems strange to me that [the Lincoln Hawker] stalled so far above the expected stall speed [20 knots]. There must have been something else that caused the roll. I have never used the autopilot and, whatever type of aircraft, have always aimed for a decrease of one knot per second.”

Raytheon, Duncan Aviation and NTSB investigator Sorenson did not tell AIN why there were four passengers on board. It’s not clear whether FAR 91.407 applied
to the Hawker stall test flight. This rule prohibits carrying people other than crewmembers “in an aircraft that has been maintained, rebuilt or altered in a manner that may have appreciably changed its flight characteristics or substantially affected its operation in flight until an appropriately rated pilot with at least a private pilot certificate flies the aircraft, makes an operational check of the maintenance performed or alteration made and logs the flight in the aircraft records.”

The stall test requirement in the Hawker maintenance manual does not address this issue, although the rule could be interpreted to apply, given that the manual titles the section requiring the stall test as “Correction of unacceptable stall handling characteristics.” A Raytheon maintenance expert consulted for this article said he never allows any passengers to fly on a required post-maintenance flight test.

Hawkers aren’t the only business jets requiring post-maintenance stall tests. Most Learjets require stall tests after the replacement of wing leading-edge sections.