The Siren Call of Aerobatics: Cirrus SR22 Crash

21 Jun 13 One Comment

It was the 13th of November, 2011. Two pilots were flying a rented Cirrus SR22, on their way home from the Stuart Air Show at Witham Field in Martin County, Florida. The pilot in the left seat was a 23 year-old private pilot. In the right seat, was a 34-year old commercial pilot whose family says had over 6,000 hours experience. The pilots were cousins and best friends.

About 10 miles south of Witham Field, they saw a Sukhoi SU-29 and an Extra-300 flying in formation, piloted by friends. The Cirrus SR22 joined the formation and the planes proceeded southwest.

ERA12FA068: Sunday, November 13, 2011 in Boynton Beach, FL

Shortly after the flight crossed the northern border of the Loxahatchee National Wildlife Refuge, the pilot of the Su-29 observed the accident airplane’s pitch smoothly increase upward to an angle of about 30 degrees. The airplane then began a roll to the left, and pitched nose-down as it rolled to an inverted attitude. As the airplane descended, it began to roll right, before it impacted the marsh below in an approximate 80-degree nose-down pitch attitude.

The pilot of the Su-29 contacted air traffic control as he orbited around the wreck. It was already clear that neither pilot could have survived.

A man walking his dog saw the three aircraft which flew overhead at a low altitude, less than 1,000 feet above the ground. The FAA confirmed this with radar data, the SR22 was flying at a recorded pressure altitude of 521 feet.

Local news coverage the day of the crash:

It was not immediately clear who was flying the plane. The safety restraints had to be cut from the left-seat pilot but it appeared that the right-seat pilot was not wearing his shoulder harness. However, the right-seat pilot was much more experienced, a commercial pilot with over 6,000 hours of flying experience, including helicopters, corporate jets and single and twin engine aircraft.

He was described as “a really good stick” and an “adrenaline junkie” and was clearly well-liked within the group.

He’d obtained his private pilot certificate in 1996 and then gave it up in 2006, when he submitted a “letter of surrender” to the FAA, abandoning his commercial pilot certificate. The reason for his surrender was “in anticipation of FAA certification action” but the FAA file contained no further details or even any reference to incidents that might have led the pilot to expect action.

In 2008, he obtained a student pilot certificate and over the next two years, he obtained his commercial pilot certificate with ratings for single and multi-engine aircraft, rotercraft helicopters, instrument airplane and helicopter.

The SR22 was rented in his name from 11-13 November for personal use. As a part of the rental agreement, the right-seat pilot agreed that the aircraft would not be used or operated by any other person. This, combined with the pilot’s experience and statements by witnesses, has led the NTSB to state that it was “most likely” that the right-seat pilot was in control of the aircraft.

The Cirrus SR22 is a single-engine four-seater and is currently the world’s best-selling single-engine aircraft. The SR22 is best known for the Cirrus Aircraft Parachute System (CAPS) which lowers the aeroplane slowly to the ground in the event of loss of control or other in-flight emergencies. The accident aircraft, registration N661FT, was operated by Air Orlando Flight School as a rental aircraft.

A Blackberry mobile device found in the wreckage had photographs taken that day. The final four photographs were taken from the aircraft directly before the crash.

Two of the photos depicted two other airplanes flying in a trailing formation off of the accident airplane’s right wing, while one of the later photos shows one airplane in a trailing formation off of the accident airplane’s left wing. The airplanes immediately off of the accident airplane’s left and right wings in the two photos appeared to be the same, though the registration number was not visible in either photo. One of the photos also showed that the left seat pilot was wearing a black t-shirt and that both shoulder restraints were on, while another photo showed that the right seat pilot was wearing a gray t-shirt. Only the right shoulder of the right seat occupant was visible, and the occupant did not appear to be wearing that shoulder restraint.

A witness on the ground saw the aircraft shortly before the crash. He was standing with an acquaintance when they heard several loud and low airplanes.

He remarked to his acquaintance about how close each of the airplanes was flying to the others as they flew from the northeast to the southwest. The airplane that was trailing in the formation began to lag behind when the witness looked away. Just then, the acquaintance remarked, “Whoa, that guy snapped a roll!,” referring to the lagging airplane in the formation.

Another witness said that he’d observed two or three low wing aircraft flying southwest in close formation. He said one of the airplanes did a barrel roll before he lost sight of the formation behind trees.

A barrel roll. In a Cirrus SR22. Let’s just make sure this is clear: the Cirrus SR22 is not an aeroplane meant for loop-de-loops. It is neither certified nor designed for aerobatic operations, nor even for turns where the angle of bank exceeds 60°. For an aileron roll, you pull the aircraft into a 90° bank. I mean, you don’t. I don’t. We wouldn’t do this in a plane which isn’t made for it. But it seemed the pilot of the SR22 that day did.

The investigation confirmed the aerobatic maneuvers when they reviewed the data from the flight recorder. And when they reviewed the flight earlier that day. And when they reviewed the flight from the day before…

On November 11, 2011, two days prior to the accident flight, the airplane departed BCT at 1654 and climbed to an altitude of about 2,000 feet, before beginning a shallow descent to 1,800 feet. At 1658, the airplane began to pitch up and roll to the left, reaching about 30 degrees of nose-up pitch and completing 360 degrees of roll. The airplane then continued to FA44, and landed at 1701.

On the morning of the accident flight, the airplane departed from BCT at 1030 and climbed to an altitude of 1,500 feet. At 1037, the airplane began descending and leveled off at an altitude of about 600 feet at 1048. At 1057, the airplane began pitching nose-up and rolling to the left, reaching 32 degrees of nose-up pitch and completing 360 degrees of roll. The airplane completed a low pass down runway 12 at SUA at a GPS altitude of less than 75 feet and an indicated airspeed of 142 knots before it climbed to about 500 feet, circled the airport, and landed at 1105.

After they departed SUA, they climbed to 1,000 feet but the flight recorder showed that at 17:34, the flight descended, with a GPS altitude of 161 feet and the measured pressure altitude was 0 feet. Over the next few minutes, the GPS altitude varied from a high of 195 feet to a low of 38 feet.

Beginning at 1736:18, while flying at a GPS altitude of 61 feet, the airplane began a roll to the left that reached a maximum bank angle of 66 degrees about 4 seconds later. The airplane then began rolling back to the right, and at 1736:19 reached a maximum right bank angle of 70 degrees, after climbing to a GPS altitude of 308 feet. At that time, the recorded pressure altitude was 109 feet. The airplane returned to a relatively level roll attitude about 4 seconds later.

Then the aircraft descended again, reaching a low GPS altitude of 145 feet (the pressure altitude was an invalid negative number).

The pitch angle then began to increase, reaching a maximum of 27 degrees nose-up at 1736:36, at a GPS altitude of 129 feet, and a pressure altitude of 29 feet. Within 2 seconds, a left roll began that continued past 90 degrees, and as the roll increased, the pitch angle also began to rapidly decrease. As the airplane reached 178 degrees of left roll, the pitch had decreased to 30 degrees nose-down, at a maximum pressure altitude of 353 feet. The airplane then began to descend, and the pitch continued to decrease to 67 degrees nose-down one second later, as the roll transitioned past inverted to 138 degrees of right roll. The final recorded data point, one second later, showed the airplane in a 68 degree nose-down pitch attitude and a 42-degree right roll, at a pressure altitude of 205 feet and an airspeed of 156 knots.

That can be hard to visualise but COPA safety have done an excellent video based on the data. The following shows the aerobatics performed in the SR-22 over the course of these two flights and the accident flight:

One of the pilots of the aircraft said that he’d heard that the right-seat pilot had “rolled the Cirrus” in the past but had never seen him do it. But he’d never logged aerobatic flight hours and was not known to have aerobatic experience. Maybe if he had, he’d have better understood the stresses he was putting the aircraft under.

It’s possible that the lack of a harness was the final straw which caused the right-seat pilot to lose control of the aircraft. At that height, in that aircraft, he had no chance of recovery.

The National Transportation Safety Board determines the probable cause(s) of this accident to be:
The right seat pilot’s decision to attempt a low-altitude aerobatic maneuver in a non-aerobatic airplane.

The parachute was found in a bag near the wreckage. The ballistic recovery parachute system activated on impact. However, even if one of the pilots pulled the handle to activate the parachute, it wouldn’t have made any difference, not at that height.

You can read the full report here: ERA12FA068: Sunday, November 13, 2011 in Boynton Beach, FL


If you are interested in air crash investigations, you’ll probably enjoy my e-book, Why Planes Crash.

PS: If you don’t have an e-book reader and would prefer the book in PDF format, just email me at sylvia@fearoflanding.com and we’ll work something out.

Category: Accident Reports,

One Comment

  • Most light aircraft have enough aileron authority and a PROPERLY flown “barrel roll” does not really impose undue stress on the airframe.
    The difference, in general, between an aircraft approved for aerobatics is – apart from the effectiveness of the controls in all axes – is that is also has redundant strength in case a manoeuvre is NOT flown properly.
    I should not admit to this, but apart from aerobatics with instructors in the likes of Beagle Pup, Tiger Moth, Stampe SV4, and Cap 10, I also have rolled (and looped) Piper Super Cubs, Aeronca Champion, even a Cessna Citation. The latter taught by a retired USAF fighter pilot.
    Was it unprofessional? In retrospect, of course it was. Was it fun ? You bet ! Were the aircraft overstressed? No ! And coming to think of it, neither was I, I guess I just had a big grin on my face. And just for the record: I did not do aerobatics when on a passenger carrying flight.
    I also know of a pilot who rolled a Boeing 707 !! The aircraft concerned was at the end of it’s service life and had been flown by a volunteer crew as part of a hijacking deal with the government of the Netherlands. They had to deliver a hijacker with a large sum of ransom to Libya.
    On return, they rolled the 707. This is on record. Witnessed by ATC and reported by journalists. Boeing and crew survived.
    Of course, one of the key factors is NEVER to attempt aerobatics unless there is sufficient height. We all know that in an emergency, the most useless items in any aircraft are the sky above you, the runway behind you and the fuel you left in the bowser.
    Sufficient height of course depends on the experience of the pilot. I once took an aerobatic lesson with a Russian champion in a Yak 52. He did not allow me to do a looping for the simple reason that he started a few feet above the runway and ended at the same height.
    Most of us never ever will reach his level of skill, for starters it is an extremely expensive hobby.
    One last remark: Before starting any aerobatic manoeuvre: Perform the “HASELL” checks ! It may save your life.

Post a comment:

Your email address will not be published. Required fields are marked *

*
*
*