On Saturday, June 26, 1993, at 1450 eastern daylight time, a Boeing-Stearman PT-17, N58212, registered to and piloted by Ronald G. Shelly, impacted the ground while performing acrobatics during an air show at the Concord Municipal Airport, Concord, New Hampshire. The airplane was destroyed by impact and post crash fire. The pilot and passenger were fatally injured. Visual meteorological conditions prevailed, and no flight plan was filed. The flight was being conducted under 14 CFR 91.

The pilot was scheduled to perform the acrobatic maneuvers, which were to be followed by wing walking activity by the passenger (the pilot's daughter).

The pilot was executing rolling maneuvers at low altitude. After completing a left barrel roll, the airplane entered a roll from which it did not recover prior to impacting the terrain.

Mr. Wayne T. Smith, Aviation Safety Inspector (Operations) for the Federal Aviation Administration, was the Inspector-In- Charge for this air show, and he witnessed the accident.

In his report, Mr. Smith stated:

I observed the acrobatic performance and accident from the air show command platform located at the show center. After the aircraft completed a left slow entered a left snap roll....I saw the aircraft lose approximately 50 to 75 feet after completing three quarters of the roll. I could see by the acrobatic smoke that the aircraft was skidding to the right. The aircraft continued its left roll as its wings came level about 25 feet above the ground. The nose then came up sharply while the aircraft continued its roll to the left...I could still hear the aircraft engine and it sounded normal to me. The nose of the aircraft continued smoothly in its arc while the wings continued to roll to the left. The nose came down through the horizon striking the ground at about a 60 degree attitude. The left wing struck the ground almost at the same time. Almost immediately thereafter the aircraft erupted in flames.

Mr. Smith's report also stated:

Earlier that morning, Mr. Ronald G. Shelly, the pilot of N58212, had informed...the air show director, that he did not feel "Up to snuff" and wanted to skip his morning solo acrobatic routine. [The air show director] informed me (after the accident) that Ron Shelly had been complaining about having flu like symptoms four or five days before the accident.

...on the morning of the accident, I spoke with Mr. Shelly and his daughter during a routine ramp check....I spent about fifteen minutes with Ron ...During that time he gave no indications of illness nor did he discuss with me the flu like symptoms he had experienced earlier that week.

Mr. Stanley Segallo, an air show performer, witnessed the accident. In a statement, he said:

I observed the snap roll which followed a slow roll down runway 17. He appeared to be at 85 mph out of the slow roll. He then immediately did a snap roll, still at about 85 mph. That speed is too slow. Normally a snap roll would be done at about 110 mph. When coming out of the snap roll he still had a lot of back stick pressure (That causes too much drag). When he came around through the 180 inverted position the aircraft did not unload to gain airspeed. The aircraft was still stalled while rolling to the left. The aircraft struck the ground in about an 85 degree nose down attitude.

The accident occurred during the hours of daylight at about 43 degrees, 12 minutes North and 071 degrees, 30 minutes West.


The wreckage was examined at the accident site on June 26, 1993, by Mr. Wayne T. Smith, Aviation Safety Inspector for the Federal Aviation Administration. In his report, Mr. Smith stated:

The aircraft was on its nose. It was crushed from the nose back to the pilot's seat. I found the fuel tank to be ruptured and leaking avgas on the engine. The left wings were attached to the aircraft. Its spars were broken and the left wing tip showed crush damage. The aircraft had ailerons on the top and bottom wings. Both were connected with an intercon- necting rod. The lower aileron was connected to the pilot's control stick by a connecting rod. The right wings were burned except for the aileron connecting rod. It was connected to the pilot's control stick.

The fuselage was intact except where it had broken open just behind the pilot's seat. It was twisted and bent to the left. The rudder and elevators were connected to the vertical and horizontal stabilizers....The rudder was connected to the pilot's left rudder pedal by its cable. The cable connecting the right rudder pedal was broken at the clevis where it normally would have been connected to the pilot's right rudder pedal....

The engine was destroyed. The propeller was twisted and showed cord wise scratches....Most of the cockpit was destroyed by fire. The flight instruments were destroyed in the crash.

During telephone conversations on June 27, 1993, between the Safety Board and the FAA, it was decided that certain sections of the right rudder control system should be forwarded to the NTSB for examination.

The wreckage was examined again on July 1st and July 13, 1993, by the Safety Board, the Federal Aviation Administration and the New Hampshire Division of Aeronautics. These examinations focused on the rudder system. Additional parts were removed and sent to the Safety Board's Metallurgical Laboratory.


Mr. Shelly held a Commercial Pilot Certificate, with single and multi-engine, land airplane and instrument ratings. He also held a Second Class Airman Medical Certificate that was issued on February 2, 1993.

He possessed a current FAA Form 8710-7, Statement of Aerobatic Competency, dated February 23, 1993. This form was issued, after an aerobatic evaluation of Mr. Shelly was conducted by an Airshow Certification Evaluator, from the International Council of Air Shows, on February 12, 1993. Mr. Shelly was approved for a Level 1, which involved "No Restrictions" on his performance, including solo acrobatics and his daughter's wing walking.

In his application for these ratings, Mr. Shelly reported that he had performed in eight air shows in 1992. His applications for the previous 2 years also show eight air show performances. In the "Ground Evaluation Notes" written by the most recent evaluator, it stated: "I've observed Ron at several air shows in the past year and have observed the same safe operations I have consistently seen over the past 7 years we've worked together."


N58212 was a vintage 1940 airplane produced as a trainer for the U.S. military. In 1983, Mr. Shelly had the airplane rebuilt and modified. This included the following:

1. Installation of Pratt & Whitney, R985, 450 horsepower engine.

2. Hamilton Standard constant speed propeller.

3. Inverted flight fuel system.

4. Four aileron system.

5. Complete disassembly of fuselage: new cables, entire recovering of airplane.

6. Wing walking stand and smoke system.


The autopsy on Mr. Shelly was performed by Dr. Roger M. Fossum, Chief Medical Examiner for the State of New Hampshire, on June 27, 1993, at the Concord Hospital, Concord, New Hampshire. The autopsy report stated:

The heart is normal size and the coronary arteries follow their usual distribution; However, there is severe atherosclerosis of the proximal left anterior descending artery and its branches with up to 90 to 95 percent focal closure...The myocardium reveals an old myocardial infarction scar of the anterior septum in a subendocardial location.... Microscopic sections of the heart reveal well developed mature scar tissue....The coronary artery reveals essentially complete occlusion by mature atherosclerosis with focal calci- fication.

Toxicology tests for both the pilot and the front seat passenger were conducted by Dr. Dennis V. Canfield, Manager of Toxicology and Accident Research Laboratory, Federal Aviation Administration, on October 22, 1993. The results of these tests were negative for carbon monoxide, alcohol or drugs.

Dr. Charles S. Springate II, Chief Deputy Medical Examiner for the Armed Forces Institute of Pathology, submitted a consultation report, in which he stated:

We received the autopsy protocol, preliminary NTSB investigative information, a videotape of the crash and...a copy of his outpatient record from the National Naval medical center.

Comment: This man's heart disease was certainly severe enough to cause sudden incapacitation at any time. However, there is no way to determine from examination of the heart whether such incapacitation did, in fact, occur.

Dr. Charles A. DeJohn, Medical Officer for the Federal Aviation Administration Aircraft Accident Research Section, conducted an Aerospace Medical Consultation for this accident. The report stated:

It appears that a heart attack may be the most likely explanation for this accident. The pilot had a history of a previous myocardial infarction (MI) as well as severe coronary artery disease (CAD). During the week prior to the airshow he was suffering from fatigue and "flu-like" symptoms, both of which can be symptomatic of heart disease. The abrupt rolling pull-up into unbalanced and eventually uncontrolled flight during the show is consistent with agonal reaction of an individual experiencing the sudden, severe pain of a heart attack. In addition, it appears that for a short period of time during the final phase of flight the aircraft was wings level long enough for an experienced aerobatic pilot to have salvaged an unintentional maneuver and recovered, or at least crash straight ahead to minimize the severity of damage. It does not appear, however, that there was any attempt on the part of the pilot to recover and the aircraft continued its final left spiral into the ground. This suggests that the pilot may have been incapacitated and unable to effect a recovery at the time.

His report continued:

....The principal symptoms of heart disease include dyspnea (difficulty breathing) chest pain or discomfort, cough and excess fatigue. The chest pain is often confused with gastrointestional causes and denial on the part of the patient frequently leads to the conclusion that the constellation of symptoms is due to indigestion, musculo-skeletal aches and pains, or the "flu."

...evidence suggests the possibility that the inflight incapacitation of the pilot may have been responsible for the accident. Although the cause of the incapacitation cannot be determined for certain, there are aspects of the history and the videotape that might explain the events:

* Myocardial infarction

* Kidney stone

* "Flu" symptoms and fatigue

The pilot complained of "flu-like" symptoms and fatigue for a week prior to the accident. While minor illness, coupled with fatigue have been known to be contributing factors in other airshow accidents, they are usually associated with additional causes, such as a stressful schedule causing accumulated loss of sleep (especially the night before), increased alcohol consumption, etc. These elements appear to be lacking here.

The pilot was first hospitalized with kidney stones on July 3, 1989. He was again diagnosed with kidney stones at Bethesda...on May 12, 1993, approximately a month prior to the accident.... kidney stones are marked by severe, sometimes incapacitating pain....there was no evidence of kidney stones reported on the autopsy.

He concluded the report with the following:

In view of the variety of data available, a MI appears to be the most likely explanation for the accident....Unfortunately, definitive post mortem diagnosis of heart attacks is still only experimental. The "markers" used in making the determination are not normally obtained at autopsy, and the methods are not well known. No such information is available in this case, therefore, the conclusions reached must be arrived at by reviewing the medical, pathological, toxicological, video, and accident investigation information.


The NTSB Materials Laboratory, Washington, D.C., examined pieces of rudder control cables, rods and turnbuckles. The Metallurgist's Factual Report stated:

The right-hand cable turnbuckle assembly was separated at the forward end through a severely deformed rod of cable terminal.... Examination with the aid of a low power (up to X) bench binocular microscope revealed that about 40% of the break was flat and contained ratchet marks, indicative of fatigue cracking. The fatigue features were emanating from an origin area located at the inner radius of the bend...and were progressing through the cross section of the rod....A higher magnification examination...revealed that the fatigue cracking originated from corrosion pits. Extensive pitting was evident on the entire surface of the terminal rod....The SAM examination confirmed the presence of the fatigue crack.

Mr. Pete Jones, Manager of Air Repair, Inc. Cleveland, Mississippi, was interviewed by telephone of January 7, 1994. This company specializes in the restoration and rebuilding of vintage airplanes, in particular Boeing Stearmans. In February, 1983, Air Services, Inc. completed the rebuilding of N58212. Mr. Jones stated that his company completely disassembles the flight control system, including the rudder cables. All parts were examined and replaced if necessary.

Mr. Jones stated that he had never heard of a failure of the rudder control rod. He said that he knew of cables failures as the result of "ribbing and binding."

He said that the rudder control system operates under "positive pull stress, and that it is a complete loop system." He also said "there was not enough cable and rod assembly to have had such a bend in the rod prior to impact. The rudder pedals would have been deflected, if the rod had been bent, and the pilot would have noticed such a deflection."


Mr. Gene Littlefield, Chairman of the Safety Standards Steering Committee, of the International Council of Air Shows, reviewed the accident video tape and consulted with other air show performers. In his report he stated:

The video coverage seemed to show a poorly executed left snap roll descending to the ground while continuing to turn left. This happened following a nearly perfectly executed left slow roll.

In examining the "stop action" video, the rudder is clearly visible throughout the maneuver but it did not deflect to the left at the onset of the maneuver as it must, to be a left snap roll. As a matter of fact, the rudder does not deflect in either direction at the onset of the maneuver, it stays absolutely neutral.

Mr. Littlefield's report continued:

The rudder is operational in the left direction, however, as the video shows the aircraft rotates to the left, most likely from "P" factor and torque, then continues left rotation and upon reaching right Knief Edge, left rudder comes into play at full deflection.

The [video] shows a number of interesting items for discussion.

(1) A very high pitch angle for a snap roll.

(This was due to zero rudder input. This presented the bottom of the aircraft to the flight path virtually stopping the forward momentum. The aircraft will eventually rotate left due to "P" factor and torque when the lift is accelerated in this manner. The aircraft at this point did a high lift "half snap" roll to the left and had nearly no forward speed at this point.)

(2) Zero rudder input at the point of snap.

(This appears to be not a snap roll or least not a snap roll in the left direction.}

....Note: My aircraft is nearly identical to the aircraft in this accident including four ailerons etc. So I went to a reasonable altitude and tried to duplicate this maneuver. I did six repetitions and in every case, if you do not input rudder, the aircraft will pitch upward to a 70 to 80 degree angle and then rotate to the left. I tried various speeds from 90 mph to 110 mph and the result was basically the same. I was not able to control the aircraft until I had obtain flying speed. The loss of altitude varied from 200 to 300 feet.

Mr. Littlefield addressed the subject of possible mechanical failure as follows:

....When operating this aircraft, there is constant pressure on the rudder pedals, so it stands to reason that with constant pressure on the pedals the materials aft of the rod ends are both in a straight line and could explain the deformation of both rod ends just by the accelerated weight at impact.... I do not believe that the right rod end separated before impact. Therefore I do not believe that the maneuver was to have been a right snap roll...The rod end could have had a small amount of deformation and still be attached and operating the cable and rudder, but there could only be up to a possible 45 degrees of deformation and still attach it to the rudder bar. Also due to the deformation there could be the pre-existing crack found by your lab. But it also stands to reason that the cable and rod ends were attached to the right rudder and still functional until impact. This again by deduction, the rod end could not be physically attached with the deformation that exists now after impact. The cable is simply not long enough.

In summary, Mr. Littlefield's report stated:


(1) I do not believe that this maneuver was to have been a snap roll in either direction.

(2) I believe that the aircraft was pitched upward inadvertently by one of the occupants.

(3) Not putting in a rudder upon execution of a snap roll could be compared to leaving the throttle at cruise power on landing. It would not happen. This is instinctive at this experience level.

(4) Possible wingwalker entanglement in the controls after the slow roll.

(5) Possible momentary physical problem with the pilot.

The complete aircraft wreckage was released to the Ryan Insurance Company on January 25, 1994.

Use your browsers 'back' function to return to synopsis
Return to Query Page