CEN21LA070

The impairment of both pilots from carbon monoxide exposure due to a degraded muffler. Contributing to the accident was the inadequate maintenance inspection, which failed to detect the degraded muffler.

The instructional flight with a flight instructor and a private pilot receiving instruction had been airborne about 2 hours when the accident occurred. During the final portion of the flight, air traffic control had issued an instrument flight rules (IFR) clearance and established radar contact and voice communication with the flight. When the airplane was about 60 nautical miles west of the intended destination, radar contact was lost, and voice communication became intermittent and was eventually lost. The final transmissions received from the airplane were unclear. The airplane’s flightpath after receiving the IFR clearance was erratic. During the final 5-1/2 minutes of the flight, the airplane made 3-3/4 left turns of varying radius, followed by 1-1/4 right turns. The accident site was located about 0.2 miles from the last recorded position. The airplane was destroyed during the impact with trees and terrain. The airplane’s cabin heating system used a heat shroud over the exhaust system mufflers to draw heat for the cabin. A postaccident metallurgical examination of the airplane’s exhaust system revealed that one of the mufflers had several fractures present before the accident crash sequence. No other anomalies were detected regarding the airplane, engine or systems that would have precluded normal operation. The compromised muffler allowed combustion gases from the engine to enter the cabin heating system. The airplane received a 100-hour inspection, and an annual inspection 10 days before the accident. The 100-hour inspection noted that the inspection was performed in accordance with the airplane service manual. The service manual featured a warning that a very thorough inspection of the exhaust system be accomplished at each 100-hour inspection. Based on the condition of the fractured muffler, it is likely that the heat shroud was not removed to facilitate inspection of the mufflers and the degraded condition of the fractured muffler was not detected during the inspections. The flight instructor’s toxicology testing detected carboxyhemoglobin at 29%, consistent with inhalation of carbon monoxide escaping into the cabin from the cracked muffler. Toxicology testing also detected past use of the stimulant methamphetamine and its metabolite amphetamine, which were detected in his urine but not detected in his cavity blood. Therefore, the identified methamphetamine and its metabolite in the flight instructor were not a factor in this accident. Although a low concentration of ethanol and n-butanol were detected in the flight instructor’s muscle tissue, their absence in liver tissue makes it likely that the identified ethanol was from sources other than ingestion and did not contribute to the accident. The private pilot’s toxicology testing did not detect carboxyhemoglobin; however, the tests reporting limit for carboxyhemoglobin is 10%. Therefore, although the toxicology results did not report a carboxyhemoglobin level above 10% for the private pilot, it is reasonable based on the operational evidence that the private pilot was also impaired and unable to either take control of the aircraft or communicate with controllers. Toxicology testing also identified the sedating antihistamines diphenhydramine and cetirizine in the private pilot’s specimens. While both could cause drowsiness, given the circumstances of this accident and the presence of the flight instructor, the effect of the private pilot’s use of diphenhydramine and cetirizine were not factors in this accident. The operational evidence, the erratic flightpath, and the unclear communication with controllers, along with the compromised muffler due to fatigue cracking support both pilots’ impairment due to carbon monoxide exposure. Furthermore, it is likely that inadequate maintenance inspection failed to detect the compromised muffler, which then allowed carbon monoxide from engine combustion to enter the cabin through the airplane’s cabin heating system during flight.

The flight instructor’s flight logbook was not available for review. He reported flight experience that included 3,200 total flight hours and 45 hours in last 6 months as of his last medical exam dated October 26, 2020. The flight logbook for the pilot receiving instruction showed that he had 154.8 hours total flight experience, including about 98 hours in Piper PA-38 airplanes, as of the last entry dated November 28, 2020. The final three logged flights were in the accident airplane with one flight recorded on November 27, 2020, that lasted 2.2 hours, and two flights on November 28, 2020, that lasted 2.2, and 3.0 hours respectively. The final three flight entries noted simulated instrument flight instruction. Toxicology testing performed by the FAA’s Forensic Services Laboratory on specimens from the flight instructor detected carboxyhemoglobin at 29%. Toxicology was positive for ethanol at 0.032 gm/hg in the muscle tissue but no ethanol was detected in the liver tissue; n-butanol was also detected in muscle tissue but not in liver tissue. Methamphetamine was detected in the urine at 307 ng/mL and amphetamine was detected in urine at 31 ng/mL. Toxicology testing performed by the FAA’s Forensic Services Laboratory on specimens from the private pilot did not detect carboxyhemoglobin in the cavity blood; FAA laboratory’s reporting limit for carboxyhemoglobin is 10%. Toxicology was positive for sedating antihistamine diphenhydramine at 848 ng/mL in the private pilot’s cavity blood and in his urine; the sedating antihistamine cetirizine was detected in his muscle and liver tissues. The non-impairing high blood pressure medications losartan and amlodipine were detected in the private pilot’s cavity blood and urine.   Maintenance records for the accident airplane showed that the airframe and engine had received a 100-hour inspection that was approved on November 20, 2020, by the accident flight instructor who also held an airframe and powerplant mechanic certificate. A separate endorsement for an annual inspection was approved by a different mechanic holding an inspection authorization on that same day. The endorsement for the 100-hour inspection noted that it was completed in accordance with the Piper PA38 Service Manual. The airplane impacted trees and terrain as seen in Figure 3. The debris path was about 230 ft long and oriented on a magnetic heading of 320°. The airplane was severely fragmented, and the main portion of the wreckage came to rest about 160 ft from the initial tree impact point. The engine was located about 230 ft from the initial impact point. There was no fire. FIGURE 3. Photograph showing the main wreckage at the accident site. The cabin heating system used a heat shroud over the exhaust system mufflers to draw heat from the exterior of the muffler for cabin heating. A postaccident metallurgical examination of the muffler and cabin heat shroud system revealed multiple branched cracks in the outer shell of one of the two mufflers (figure 4). The fracture faces of the cracks were covered with oxidation product, consistent with its presence before the accident crash sequence. Metallurgical cross sections revealed oxidative degradation that progressed through the thickness of the muffler shell from the inside surface outwards. Other than the cracked muffler, examination of the wreckage by Federal Aviation Administration inspectors and an investigator from Piper Aircraft revealed no preimpact mechanical malfunctions or anomalies with the airframe, or engine that would have precluded normal operation. Figure 4. Photograph showing the two mufflers from the airplane. On November 30, 2020, at 1213 central standard time, a Piper PA-38-112 airplane, N9879T, was destroyed when it was involved in an accident near Franklin, Arkansas. The flight instructor and the private pilot receiving instruction were fatally injured. The airplane was operated as a Title 14 Code of Federal Regulations Part 91 instructional flight. Automatic dependent surveillance-broadcast (ADS-B) data were obtained for the accident flight, which is overlaid on Google Earth in Figure 1. The airplane departed Country Air Estates Airport (1AR9), Lonoke, Arkansas, at 0931, made a short flight to Carlisle Municipal Airport (4M3), Carlisle, Arkansas, and arrived at 0946. The airplane remained at 4M3 for about 16 minutes before it departed at 1002. The airplane then traveled about 100 nautical miles (nm) north-northwest, turned to a northeast heading and overflew Marion County Regional Airport (FLP), Flippin, Arkansas, and Baxter County Airport (BPK), Mountain Home, Arkansas, before it turned eastbound. The airplane did not land at FLP, and due to a gap in the ADS-B data it could not be determined if the airplane landed at BPK. After it passed BPK, the airplane turned to the east, contacted the Memphis Air Route Traffic Control Center (ARTCC) and obtained an instrument flight rules (IFR) clearance to Walnut Ridge Regional Airport (ARG), Walnut Ridge, Arkansas, which was about 76 nm east of BPK. During the portion of the flight from BPK to ARG, air traffic control initially established radar contact and voice communication with the flight. When the airplane was about 60 nm west of ARG, radar contact was lost, and voice communication became intermittent and was eventually lost. The final transmissions received from the airplane were unclear. The controller attempted to maintain communication with the accident airplane using messages relayed through other aircraft and reverted to receiving acknowledgements to transmissions using microphone key clicks from the accident pilot. Figure 1. Map showing the airplane’s ADS-B track (red), the relative positions of airports, the straight-line course from BPK to ARG (white), and the accident location. The airplane did not follow a straight-line course between BPK and ARG, as it first deviated up to 2.8 nm north of the straight-line course then turned toward south. The airplane then deviated south of the straight-line course by as much as 4.5 nm. During the final 5-1/2 minutes of the flight, the airplane made 3-3/4 left turns of varying radius, followed by 1-1/4 right turns before the end of the recorded ADS-B data. The final ADS-B position was at 1213:36 and the accident site was located about 0.2 nm south southeast of the last ADS-B position, 4.7 nm south of the straight-line course, and about 35 nm from BPK (figure 2). Figure 2. Map showing the final 7.5 minutes of the accident flight path (red), with the last recorded ADS-B position, the accident location, and the straight-line course from BPK to ARG (white). In December 2021, the NTSB made recommendations regarding the use and installation of carbon monoxide detectors in enclosed cabin airplanes equipped with reciprocating engines. Recommendation A-22-1 to the Federal Aviation Administration stated: Require that all enclosed-cabin aircraft with reciprocating engines be equipped with a carbon monoxide detector that complies with an aviation-specific minimum performance standard with active aural or visual alerting. Recommendation A-22-2 to the Aircraft Owners and Pilots Association and Experimental Aircraft Association: Inform your members about the dangers of carbon monoxide (CO) poisoning in flight and encourage them to 1) install CO detectors with active aural or visual alerting and 2) proactively ensure thorough exhaust inspection during regular maintenance. Surface temperatures were consistent with the use of cabin heating for the accident flight. Airmen's Meteorological Information notices (AIRMET) for turbulence and icing were in effect at the time of the accident. In addition, pilot reports of turbulence ranging from light turbulence to occasional severe turbulence were reported in the area where the accident occurred. The PA38 Service manual stated: WARNING: A VERY THOROUGH INSPECTION OF THE ENTIRE EXHAUST SYSTEM, INCL