N440SH

On September 30, 2018, at 1735 eastern daylight time, a Robinson R44 II helicopter, N440SH, operated by Charm City Helicopters, was substantially damaged during a loss of control on the ground following a precautionary landing atop the Longwell Parking Garage Structure in Westminster, Maryland. The commercial pilot was not injured, and there were no ground injuries. Visual meteorological conditions prevailed, and no flight plan was filed for the flight to reposition the helicopter atop the parking garage. The flight was conducted under the provisions of Title 14 Code of Federal Regulations Part 91.

The operator provided 4 to 5-minute helicopter rides as one of the attractions at the Westminster Fallfest and used the top deck of the multi-story parking garage as a helipad, where the passengers were staged. At the completion of each flight, passengers were off-loaded, the next group of passengers was loaded as the helicopter ran, and then the helicopter departed over the down-sloping portion of the top deck.

The pilot provided a written statement to police and a telephone interview to an NTSB investigator at the scene. He stated that the helicopter experienced an engine and main-rotor overspeed at takeoff, and he performed a precautionary landing at the bottom of the parking deck that sloped downward immediately in front of him. After disembarking the passengers, "I attempted to reposition the helicopter to its original spot."

According to the pilot, "I increased the throttle to return to normal engine rpm [and] the aircraft started spinning uncontrollably." As the helicopter rotated around its mast, the owner/operator standing beside the helicopter ducked to avoid the tailboom before it struck a parked automobile. The pilot said he stopped the engine and the rotation of the helicopter by pulling the mixture control.

During subsequent telephone interviews with investigators and a second written statement, the pilot's account of the flight remained unchanged, but the events which occurred after landing differed. He insisted he was "misunderstood" and that his intent was to stop the helicopter's engine after he deplaned the passengers and reposition the helicopter by attaching ground-handling wheels and pushing it back to the top deck. According to the pilot, the engine was idling with the throttle at the idle stop, when it accelerated un-commanded, the helicopter rotated around its mast, and the engine and rotation could only be stopped by pulling the mixture control. In a telephone interview with his insurance adjuster, the details of the flight remained the same, but the pilot reported that after deplaning the passengers, he needed to accelerate the engine to 70 percent to perform a proper cooldown prior to engine stoppage. When he increased the throttle, the engine accelerated uncontrolled, the helicopter rotated, struck the parked car, and came to rest only after he pulled the mixture control.

According to the operator, she heard something "heighten in pitch" as the helicopter departed, and that the "loud whining occurred again" when the helicopter began to spin on the ground after landing and deplaning the passengers.

The pilot held commercial pilot and flight instructor certificates with ratings for rotorcraft helicopter. He also held a private pilot certificate with a rating for airplane single engine land. The pilot did not possess an instrument rating.

His most recent Federal Aviation Administration (FAA) second class medical certificate was issued on May 7, 2018. The pilot estimated he had accrued 700 total hours of flight experience, of which 400 were in rotorcraft, and 200 were in the accident helicopter make and model.

According to FAA and maintenance records, the helicopter was manufactured in 2005, and had accrued 2,216.8 total aircraft hours. The most recent annual inspection was completed December 1, 2017, concurrent with a complete airframe and engine overhaul. The engine and helicopter had accrued 70 hours since that date.

On August 24, 2018, at 2192.1 total aircraft hours, the right magneto was removed and replaced due to "engine rpm fluctuating."

The 1750 weather observation at Frederick Municipal Airport (FDK), 20 miles west of the accident site, included clear skies, calm winds, and 10 miles visibility. The temperature was 24°C, the dew point was 15°C, and the altimeter setting was 30.30 inches of mercury.

The wreckage was examined at the accident site on October 1, 2018 and all major components remained attached to the helicopter. The tailrotor blades were substantially damaged but remained attached in their respective grips. The bottom of the vertical stabilizer also displayed substantial damage, and the tailboom displayed a small dent. The tailboom itself remained attached and appeared intact. The remainder of the helicopter was undamaged. When rotated by hand, the main and tail rotors rotated smoothly.

Control continuity was established from the flight controls to all flight control surfaces, and from the engine controls to the engine. The examination was suspended for a detailed examination to be conducted later.

Overlapping, rotational skid marks that were the same approximate dimension as the R44 landing gear were observed on the parking deck at the landing site. According to the city government, the down-sloping parking ramp was 190 ft long with an elevation change of 10 ft 6 ins at a slope of 5.5 percent. The skid marks placed the toes of the skids on the flat deck, and the tails of the skids on the sloped deck at landing. The result was four small contact patches in each corner, rather than the length of each skid tube in contact with the concrete surface.

A detailed examination and a test run of the helicopter was performed on October 9, 2018. The engine started immediately and idled smoothly at 68%. The rpm governor was turned on at the collective control stick and the engine was accelerated at the throttle twist grip until the governor "captured" control of the engine rpm at 80 percent. Engine rpm then increased, paused twice at intermediate rpm thresholds below normal operating range (101 -102%), when the engine accelerated up to and through the normal operating range past 105%, then slowed slightly, and surged between 96% and 104% un-commanded.

The governor was turned off, the throttle was set at 100% engine speed and operated smoothly at that rpm. The governor was turned back on, and the engine rpm surged again. The governor was tested below the 80% threshold and operated as designed. There were no un-commanded surges below 80%, and the engine rpm was fully controlled at the throttle twist grip below 80%.

The governor was controlled by a governor controller module that received a signal from the right-side engine magneto. The magneto was removed, the points cover was removed, and the tachometer leads were inspected. Inspection revealed that the leads were in close proximity to the points cam and its attachment screw, that the wire insulation was damaged, and the strands were exposed.

A video recording and still photos of the magneto disassembly were reviewed by a Technical Investigator of the Robinson Helicopter Company, who provided a report. According to the investigator and the magneto service manual, the tachometer leads were not properly routed, and were in close proximity to the rotating cam screw head. The tachometer points also displayed contamination and misalignment.

According to the report, "The damage to the wire insulation will allow contact between the wire and rotating screw head creating an intermittent short to ground. This will send an errant signal to the governor control unit which in turn will manipulate the throttle accordingly."