N198LTCavalon Gyro AUTOGYRO CAVALON · 2017

On October 30, 2018, about 1448 eastern daylight time, an experimental amateur-built AutoGyro Cavalon gyroplane, N198LT, was destroyed during collision with a power pole, wires, terrain, a residence and a post-crash fire following a forced landing in Sebring, Florida. The commercial pilot and the pilot-rated passenger were fatally injured. Visual meteorological conditions prevailed, and no flight plan was filed for the flight which departed Sebring Regional Airport (SEF) about 1440 and was destined for Manatee Airport (48X), Palmetto, Florida. The personal flight was conducted under the provisions of 14 Code of Federal Regulations Part 91.

The purpose of the accident flight was for the pilot to take the passenger back to his home airport. Witnesses stated the pilot flew the accident gyroplane earlier in the day for about 2 hours, serviced it with fuel, then he and the passenger departed on the accident flight.

Radar information from the Federal Aviation Administration (FAA) revealed that a radar target identified as the accident gyroplane was acquired at 500 ft, and its track depicted a climb to about 1,000 ft and a cruise profile along a heading of about 280°. The radar track traversed the southern border of Lake Jackson in Sebring. About 1446, the radar data depicted that the gyroplane began a descent from about 1,000 ft over the southwestern shore of the lake. The descent followed the shoreline to the northwest, and the radar target disappeared at 900 ft, at which time the gyroplane's groundspeed was 90 knots and its ground track was about 0.4 mile southeast of the accident site. From where the descent began, a golf course was located about 25° and 0.5 mile to the left of the gyroplane's ground track.

SEF was not tower-controlled, but a commercial website that monitored the common traffic advisory frequency recorded a mayday call from the gyroplane at 1448. During the next 18 seconds, part of the gyroplane's registration number and a second mayday call were transmitted before the sounds associated with impact were heard. During the audible portions of the transmissions, sounds consistent with an engine increasing and decreasing in rpm were heard.

A witness who was driving southbound on the highway parallel to the shoreline of the lake stated that the gyroplane was traveling northwest about 300 ft above ground level "with very little airspeed" and appeared to be turning to the east. The gyroplane then "entered an autorotation," and, when it had descended to about 150 ft, "the nose of the aircraft dropped immediately turning toward [the] east then back toward the north." The gyroplane descended from the witness's view; afterward, he saw a large fireball.

The pilot held a commercial pilot certificate with a rating for rotorcraft-gyroplane. He held a private pilot certificate with ratings for airplane single engine land and instrument airplane. His most-recent FAA 2nd class medical certificate was issued August 25, 2017. A review of the pilot's FAA Examiner Designation and Qualification Record revealed he had accrued 4,010 total hours of flight experience, 2,715 hours of which were in "rotortype" aircraft.

According to FAA records, the gyroplane's airworthiness certificate was issued August 17, 2017. Airframe logbooks were not recovered, but photographs of maintenance entries revealed the most recent condition inspection was completed October 4, 2018 at 16.6 total aircraft hours.

Postaccident interviews, e-mails, and collected documents revealed that the owner/builder purchased the gyroplane kit from the German kit manufacturer through AutoGyro USA, Stevensville, Maryland, along with a "build-assist" agreement with Cloud9 Helicopters, West Palm Beach, Florida, an independent dealer of AutoGyro USA products. The owner/builder had no experience building or flying aircraft of any kind. His intention was to build the gyroplane and then learn to fly it. Cloud9 Helicopters agreed to perform the build-assist in exchange for 75 flight hours in the completed gyroplane in order to demonstrate the product to prospective customers for future sales.

After an experienced factory mechanic could not be arranged by the CEO of AutoGyro USA, Cloud9 Helicopters employed a mechanic who had not previously assembled an AutoGyro product, which conservatively required 3 weeks for assembly by an experienced builder. Later, an experienced AutoGyro builder, who could provide only one week's help was brought in to temporarily assist with building efforts. Upon arrival, the experienced builder directed that the gyroplane be disassembled because components had been installed incorrectly and in the wrong sequence. He found all the parts, components, and hardware had arrived with the kit, but they were packaged haphazardly, and considerable time was expended as the contents of the boxes were inventoried and organized for assembly. Further, common consumables (adhesives, lubricants, etc.) were not on hand and had to be purchased. According to the owner/builder, he "spent a lot of time going to Home Depot." When asked about his involvement with the construction of the gyroplane he said, "I put on stickers and wheel covers, nothing critical."

After the airworthiness certificate was issued, the gyroplane was test flown for the phase one testing period (40 hours). The test pilot hired to complete the flight testing stated that the test plan was modeled after FAA Advisory Circular 90-89A, Amateur-Built Aircraft and Ultralight Flight Testing Handbook. As part of the plan, the test pilot performed engine run-ups, system checks (pre-rotator, etc), high-speed taxi tests and short "hops" over the runway about 20 ft above the ground. The test pilot stated that during the testing the cyclic control exhibited a "sharp left pull" and that he explained the issue to maintenance personnel. The test pilot also stated that multiple exchanges with maintenance personnel and AutoGyro USA failed to resolve the issue. The test pilot further stated that he "grounded" the gyroplane at that time due to "improperly rigged flight controls" and indicated that he would not fly the gyroplane again until it was fixed. At the time that the test pilot ceased the test plan, the gyroplane had accrued 16.6 hours.

According to the owner/builder, the CEO of AutoGyro USA and the president of Cloud9 Helicopters were "not available" in the months following to complete the fly-off. In March 2018, the CEO of AutoGyro USA resigned his position and went to work for a competing gyroplane manufacturer, and Cloud9 helicopters became an independent dealer for the same manufacturer. Cloud9 Helicopters no longer had incentive to complete the project and the owner/builder had lost interest, so in September 2018, the owner/builder contacted AutoGyro USA to broker the sale of his gyroplane. Cloud9 Helicopters agreed to complete the condition inspection and the phase one flight testing in advance of an airshow that began November 1, 2018. At the time the agreement was made, which was on or about September 18th, 2018, the gyroplane had not accrued any additional flight hours since the previous flight testing had ceased. According to the President of Cloud9 Helicopters, he and 3 other pilots completed the flight tests between October 4 and 29, 2018.

On October 29, 2018, the accident pilot, who was the chief operating officer for AutoGyro USA, took possession of the gyroplane from Cloud9 Helicopters. His inspection of the accident gyroplane and its logbooks revealed that the hobbs meter displayed exactly 40.0 hours, the carburetors leaked, and completion of the 40-hour flight test period had not been documented in the maintenance logs. The president of Cloud9 Helicopters then made the log entry and added that the gyroplane "needed a rotor balance." The pilot then pointed out the carburetor leaks to the mechanic who had assisted with the build and completed the most recent condition inspection. The mechanic went for tools to address the leaks and returned to find the pilot had departed in the gyroplane.

The accident pilot then flew the gyroplane from North Palm Beach County Airport (F45), West Palm Beach, Florida to SEF. During the flight, the pilot noted that the cyclic control vibration was "excessive," and the airspeed indicator was inoperative. In text messages to the president of AutoGyro USA, the pilot stated, "…carbs leaking fuel, airspeed indicator not working, [and] can mix eggs with this control stick," and "I see how roughly this was put together." The accident pilot attached a video to the message to showing the control stick shaking during the flight. The president of AutoGyro USA stated that, after landing, the pilot and a colleague examined the gyroplane and found the airspeed indicator "disconnected." They retorqued the carburetor float bowls, and "wondered if the 40 hours Phase 1 testing had been done." The two further attempted to balance the rotor system through trial and error by adding and then subtracting weight on either side of the rotor head, which was unsuccessful on each attempt and subsequent test flights. AutoGyro USA offered to ship the necessary rotor balancing equipment to the accident pilot, who rejected the idea.

The AutoGyro Cavalon build and assembly instructions, the 25- and100-hour condition inspection checklists, the maintenance manual instruction, and the AutoGyro Pilot Operating Handbook provide instructions for inspection of the flight control connections.

At 1515, the weather recorded at SEF included clear skies, 10 statute miles visibility, and winds from 030 degrees at 6 knots. The temperature was 28°C, and the dew point was 13°C. The altimeter setting was 30.07 inches of mercury.

According to FAA records, the gyroplane's airworthiness certificate was issued August 17, 2017. Airframe logbooks were not recovered, but photographs of maintenance entries revealed the most recent condition inspection was completed October 4, 2018 at 16.6 total aircraft hours.

Postaccident interviews, e-mails, and collected documents revealed that the owner/builder purchased the gyroplane kit from the German kit manufacturer through AutoGyro USA, Stevensville, Maryland, along with a "build-assist" agreement with Cloud9 Helicopters, West Palm Beach, Florida, an independent dealer of AutoGyro USA products. The owner/builder had no experience building or flying aircraft of any kind. His intention was to build the gyroplane and then learn to fly it. Cloud9 Helicopters agreed to perform the build-assist in exchange for 75 flight hours in the completed gyroplane in order to demonstrate the product to prospective customers for future sales.

After an experienced factory mechanic could not be arranged by the CEO of AutoGyro USA, Cloud9 Helicopters employed a mechanic who had not previously assembled an AutoGyro product, which conservatively required 3 weeks for assembly by an experienced builder. Later, an experienced AutoGyro builder, who could provide only one week's help was brought in to temporarily assist with building efforts. Upon arrival, the experienced builder directed that the gyroplane be disassembled because components had been installed incorrectly and in the wrong sequence. He found all the parts, components, and hardware had arrived with the kit, but they were packaged haphazardly, and considerable time was expended as the contents of the boxes were inventoried and organized for assembly. Further, common consumables (adhesives, lubricants, etc.) were not on hand and had to be purchased. According to the owner/builder, he "spent a lot of time going to Home Depot." When asked about his involvement with the construction of the gyroplane he said, "I put on stickers and wheel covers, nothing critical."

After the airworthiness certificate was issued, the gyroplane was test flown for the phase one testing period (40 hours). The test pilot hired to complete the flight testing stated that the test plan was modeled after FAA Advisory Circular 90-89A, Amateur-Built Aircraft and Ultralight Flight Testing Handbook. As part of the plan, the test pilot performed engine run-ups, system checks (pre-rotator, etc), high-speed taxi tests and short "hops" over the runway about 20 ft above the ground. The test pilot stated that during the testing the cyclic control exhibited a "sharp left pull" and that he explained the issue to maintenance personnel. The test pilot also stated that multiple exchanges with maintenance personnel and AutoGyro USA failed to resolve the issue. The test pilot further stated that he "grounded" the gyroplane at that time due to "improperly rigged flight controls" and indicated that he would not fly the gyroplane again until it was fixed. At the time that the test pilot ceased the test plan, the gyroplane had accrued 16.6 hours.

According to the owner/builder, the CEO of AutoGyro USA and the president of Cloud9 Helicopters were "not available" in the months following to complete the fly-off. In March 2018, the CEO of AutoGyro USA resigned his position and went to work for a competing gyroplane manufacturer, and Cloud9 helicopters became an independent dealer for the same manufacturer. Cloud9 Helicopters no longer had incentive to complete the project and the owner/builder had lost interest, so in September 2018, the owner/builder contacted AutoGyro USA to broker the sale of his gyroplane. Cloud9 Helicopters agreed to complete the condition inspection and the phase one flight testing in advance of an airshow that began November 1, 2018. At the time the agreement was made, which was on or about September 18th, 2018, the gyroplane had not accrued any additional flight hours since the previous flight testing had ceased. According to the President of Cloud9 Helicopters, he and 3 other pilots completed the flight tests between October 4 and 29, 2018.

On October 29, 2018, the accident pilot, who was the chief operating officer for AutoGyro USA, took possession of the gyroplane from Cloud9 Helicopters. His inspection of the accident gyroplane and its logbooks revealed that the hobbs meter displayed exactly 40.0 hours, the carburetors leaked, and completion of the 40-hour flight test period had not been documented in the maintenance logs. The president of Cloud9 Helicopters then made the log entry and added that the gyroplane "needed a rotor balance." The pilot then pointed out the carburetor leaks to the mechanic who had assisted with the build and completed the most recent condition inspection. The mechanic went for tools to address the leaks and returned to find the pilot had departed in the gyroplane.

The accident pilot then flew the gyroplane from North Palm Beach County Airport (F45), West Palm Beach, Florida to SEF. During the flight, the pilot noted that the cyclic control vibration was "excessive," and the airspeed indicator was inoperative. In text messages to the president of AutoGyro USA, the pilot stated, "…carbs leaking fuel, airspeed indicator not working, [and] can mix eggs with this control stick," and "I see how roughly this was put together." The accident pilot attached a video to the message to showing the control stick shaking during the flight. The president of AutoGyro USA stated that, after landing, the pilot and a colleague examined the gyroplane and found the airspeed indicator "disconnected." They retorqued the carburetor float bowls, and "wondered if the 40 hours Phase 1 testing had been done." The two further attempted to balance the rotor system through trial and error by adding and then subtracting weight on either side of the rotor head, which was unsuccessful on each attempt and subsequent test flights. AutoGyro USA offered to ship the necessary rotor balancing equipment to the accident pilot, who rejected the idea.

The AutoGyro Cavalon build and assembly instructions, the 25- and100-hour condition inspection checklists, the maintenance manual instruction, and the AutoGyro Pilot Operating Handbook provide instructions for inspection of the flight control connections.

The gyroplane's impact damaged power lines, a power pole, and a residence. Photographs revealed that the pole was fractured into three sections and that the power lines were severed and entangled with the wreckage. The residence was consumed by the postcrash fire. The power pole was replaced and the powerlines were repaired before the wreckage examination.

The wreckage path was oriented along a heading of about 290° and was about 39 ft long. The initial impact point was about 25 ft beyond the power pole. During the wreckage examination, all major gyroplane components were not accounted for at the accident site due to the extent of the fire damage. Identifiable components of the gyroplane included the rotor system, the engine, and the main landing gear cross-tube. The wreckage path ended at the engine, which was partially embedded in the residence. Landing gear wheels and tires were scattered outside the residence. The remainder of the gyroplane was consumed in the postcrash fire.

The rotor system was found between the initial impact point and the residence and remained attached to the pylon structure. The rotor system was largely intact and displayed signatures consistent with impact and heat exposure. Spiral striations consistent with wire contact and signatures consistent with electrical arcing were visible on the rotor blades. The rotor blades were secure in their grips, and the pitch and roll push/pull cable controls and pre-rotator drive were attached to the rotor head.

All external engine accessories were destroyed by fire. The flywheel was melted, and the remaining slag precluded rotation of the crankshaft. The propeller hub remained attached, and the composite blades appeared to be uniformly severed at their roots before fire exposure.

The engine cylinder heads were removed, and the pistons were removed from their connecting rods to obtain visual access to the crankshaft and connecting rods. The signatures observed were consistent with normal wear and lubrication.

The cockpit side of the roll control rod end bearing was found attached to the impact and fire-damaged aluminum roll control bellcrank. The rod end bearing was found separated from the push/pull cable end. The inner and outer pieces of the cable end were found separated from the inner push/pull cable. The cable end (threaded portion) and rod end bearing for pitch control were separated from the inner push/pull cable. The braided end of the pitch control cable was corroded but less so than the rest of the cable. The outer braided sheath for pitch control was separated from inner cable. The rod end bearing for pitch was not attached to the control horn. The bolt and nut were not installed or found after sifting through the wreckage. The exposed ends of both the pitch and roll cables appeared pulled from their respective sheaths and rod ends. The main flight control tube was corroded but intact.

The pitch control horn had two welded tangs (left and right); When assembled, the rod-end bearing to the pitch-control cable was placed between the tangs, and a bolt passed through one tang, the rod-end bearing, and the other tang and secured with a threaded locking nut. As found, the right-side tang was bent inward toward the left-side tang. The bolt holes and the tangs appeared intact and undamaged. The pitch control horn and the pitch control rod end bearing were further examined by the National Transportation Safety Board's Materials Laboratory. The Materials Laboratory found that the holes for the bolt were corroded but did not exhibit any deformation or signs that the rod end had been pulled from the pitch horn and that the damage was consistent with the rod end not having been attached to the pitch horn at the time of the accident.

According to the functional test flight – Cavalon protocol, rotor vibration qualitative assessment, the test pilot was asked to respond affirmatively to "rotor T&B [track and balance] OK?"

During a postaccident interview, the AutoGyro USA operations manager was asked if there were acceptable ranges for vertical and lateral vibration. The operations manager replied,

If the aircraft has any vibrations the balancer should be put on and the process of documenting the correction should begin. To my knowledge none of this was done to this aircraft. At the least, [the pilot] complained to me about the poor balance (stick shake) of the rotor system on that aircraft when he picked it up from Cloud 9. I remember there was a video of that aircraft stick shake. From that video and my experience, I can say that vibration level was unacceptable for delivery.

The operations manager of AutoGyro USA was also asked about aircraft controllability with a pitch control cable failure or disconnection. The operations manager stated that the Cavalon gyroplane remained controllable through the electric/pneumatic trim system and engine power adjustments. The operations manager cited the emergency procedure in the AutoGyro Pilot Operating Handbook, which stated the following regarding a flight control malfunction:

In case of a flight control failure the gyroplane can be controlled with the remaining primary and secondary controls, including power and trim. An immediate reduction of power, respectively speed may be necessary to avoid pitch oscillations…or other effects affecting dynamic or static stability. Navigate to a suitable landing site with wide and shallow turns and approach against the wind.

The Office of the District Medical Examiner, Winter Haven, Florida, performed an autopsy of the pilot. His cause of death was multiple fractures and visceral lacerations.

Toxicology testing performed at the FAA's Forensic Sciences Laboratory found that the pilot's specimens tested negative for drugs and ethanol.