Sunday, May 01, 2022

Aircraft Structural Failure: Piper PA-28-235, N8991W; fatal accident occurred May 31, 2020 in Carlinville, Macoupin County, Illinois

Daniel Shedd, front left, texted his mother this photograph on May 31, 2020 shortly before he left Creve Coeur Airport with his college friends. The pilot, Joshua Sweers, is at front right. In the back row at left is John Camilleri and the man next to him is Daniel Schlosser. The plane crashed near Carlinville, Illinois. 

Joshua Sweers

Aviation Accident Final Report - National Transportation Safety Board

The National Transportation Safety Board did not travel to the scene of this accident.

Additional Participating Entities: 
Federal Aviation Administration / Flight Standards District Office; Springfield, Illinois
Piper Aircraft; Vero Beach, Florida 
Lycoming Engines; Williamsport, Pennsylvania 

Investigation Docket - National Transportation Safety Board:

Location: Carlinville, Illinois 
Accident Number: CEN20LA201
Date and Time: May 31, 2020, 15:46 Local 
Registration: N8991W
Aircraft: Piper PA28
Aircraft Damage: Destroyed
Defining Event: Aircraft structural failure 
Injuries: 4 Fatal
Flight Conducted Under: Part 91: General aviation - Personal


Position data depicted the airplane in cruise flight on a northeast course toward the destination airport at 5,500 ft mean sea level (msl). Shortly before the accident, the airplane entered a left turn with a gradually steepening bank angle. After completing a 360ยบ turn and returning to a northeast course, the airplane immediately transitioned into a right turn that continued until the airplane again returned to a northeast course. Airplane bank angles reached 50° and 60° in the left and right turns, respectively. As the right turn continued, the airplane entered a descent and the airspeed increased. The bank angle ultimately reached about 110° (right wing down), the pitch attitude reached 63° nose down, and the airspeed increased to over 200 kts during the descent. The maximum computed load factor based on the available data was 4.72 G. The position data ended when the airplane was between 2,000 ft and 2,500 ft msl (1,400 ft and 1,900 ft above ground level). At the end of the data, the airplane was in a steep, spiral dive and about 35 kts above the never-exceed airspeed (VNE). The airplane was established on a southwesterly course away from the accident site at that time.

The wreckage debris path was oriented on a northeasterly course and was about 400 ft long. The wreckage distribution was consistent with a low-altitude inflight break up. In addition, the presence of all airframe structural components and flight control surfaces within the debris path was consistent with the airplane being structurally intact as it approached the accident site. A postaccident examination revealed that the wing structure failed as a result of overstress. No preimpact anomalies with respect to the flight control system were identified. As a result, the steep descent was likely an intentional action by the pilot but for reasons that could not be determined.

The investigation did not have any data from which to determine the flightpath from the final data point to the accident site. However, because the final segment of the flightpath was toward the southwest and the debris path was oriented to the northeast, it is clear that the pilot attempted to pull out of the dive and, in doing so, reversed course. That maneuver, which was initiated from a steep, spiral dive and above VNE, resulted in the pilot inadvertently exceeding the ultimate load factor for the airframe. The excessive load factor caused the separation of the wings and stabilator, and a loss of control of the airplane.

Probable Cause and Findings

The National Transportation Safety Board determines the probable cause(s) of this accident to be:
The pilot’s attempted recovery from a steep descent which inadvertently exceeded the ultimate load factor of the airframe and resulted in a low-level inflight breakup.


Aircraft Center wing box (on wing) - Capability exceeded
Personnel issues Decision making/judgment - Pilot
Personnel issues Aircraft control - Pilot

Factual Information

History of Flight

Maneuvering Aircraft structural failure (Defining event)
Maneuvering Loss of control in flight
Uncontrolled descent Collision with terr/obj (non-CFIT)

On May 31, 2020, at 1546 central daylight time, a Piper PA-28-235 airplane, N8991W, was destroyed when it was involved in an accident near Carlinville, Illinois. The pilot and three passengers were fatally injured. The airplane was operated as a Title 14 Code of Federal Regulations Part 91 personal flight.

According to automatic dependent surveillance – broadcast (ADS-B) and Appareo Stratus position data, the flight departed about 1518 and proceeded on a northeast course toward the intended destination, ultimately climbing to an approximate altitude of 5,500 ft mean sea level (msl). About 1543, the airplane entered a 15° banked left turn which continued until the airplane was on a southwest course. About 1545, the left turn steepened to about 50° and continued until the airplane returned to a northeast course. About 25 seconds later, the airplane rolled out of the left turn and immediately into a 60° banked right turn. The airplane reached an altitude of 5,685 ft and had slowed to about 85 kts during the right turn. The airplane then entered a descent, and the airspeed began to increase. At 1545:54, the right turn steepened to about 110° right bank.

By 1546:07, the airplane altitude decreased to 2,850 ft, the descent rate increased to over 18,700 ft/minute, and the airspeed increased to over 200 kts. During this timeframe, the pitch angle recorded by the Stratus unit decreased from +10° (up) to -52° (down) and decreased further to -63° (down) before the end of the data. The data ended about 1546:10. At that time, the airplane altitude was between 2,000 ft and 2,500 ft msl (1,400 ft and 1,900 ft above ground level). During the timeframe covered by the available data, the maximum computed load factor of 4.72 G occurred about 1546:06.

At the end of the data, the airplane was in a steep, spiral dive and about 35 kts above the never-exceed airspeed (VNE) of 171 kts. The airplane was established on a southwest course at that time and oriented away from the accident site located about 0.15 nm north-northeast from the final data point. The investigation did not have any data from which to determine the flight path from the final data point to the accident site.

A witness reported observing the airplane “going up and down,” doing “dips in the air.” She noted five or six “dips” where the airplane would “come back up” each time. She subsequently observed the airplane enter a “nosedive” and begin “spiraling down.” The airplane was initially heading south when it entered the dive and appeared to be intact at that time. As it neared the ground, the airplane “burst into pieces in the air.”

Pilot Information

Certificate: Private
Age: 35, Male
Airplane Rating(s): Single-engine land 
Seat Occupied: Left
Other Aircraft Rating(s): None
Restraint Used: 3-point
Instrument Rating(s): None
Second Pilot Present: No
Instructor Rating(s): None
Toxicology Performed: Yes
Medical Certification: Class 3 Without waivers/limitations
Last FAA Medical Exam: August 7, 2018
Occupational Pilot: No 
Last Flight Review or Equivalent: December 28, 2019
Flight Time: 93.9 hours (Total, all aircraft), 28.8 hours (Total, this make and model), 38 hours (Pilot In Command, all aircraft), 11.3 hours (Last 90 days, all aircraft), 8.1 hours (Last 30 days, all aircraft), 0.5 hours (Last 24 hours, all aircraft)

Aircraft and Owner/Operator Information

Aircraft Make: Piper 
Registration: N8991W
Model/Series: PA28 235
Aircraft Category: Airplane
Year of Manufacture: 1964 
Amateur Built:
Airworthiness Certificate: Normal 
Serial Number: 28-10571
Landing Gear Type: Tricycle
Seats: 4
Date/Type of Last Inspection: May 1, 2020 Annual 
Certified Max Gross Wt.: 2900 lbs
Time Since Last Inspection:
Engines: 1 Reciprocating
Airframe Total Time: 2235 Hrs as of last inspection
Engine Manufacturer: Lycoming
ELT: C91A installed
Engine Model/Series: O-540-B4B5
Registered Owner: 
Rated Power: 235 Horsepower
Operator: On file 
Operating Certificate(s) Held: None

A weight and balance calculation based on estimated occupant, baggage, and fuel loadings, suggested that the airplane was within the gross weight and center-of-gravity limitations specified by the airframe manufacturer.

The airplane was certificated as a normal category airplane. The applicable limit load factor was 3.8 G’s. The corresponding ultimate load factor was 5.7 G’s.

Meteorological Information and Flight Plan

Conditions at Accident Site: Visual (VMC)
Condition of Light: Day
Observation Facility, Elevation: 3LF,691 ft msl 
Distance from Accident Site: 12 Nautical Miles
Observation Time: 15:55 Local 
Direction from Accident Site: 110°
Lowest Cloud Condition: Clear
Visibility 10 miles
Lowest Ceiling: None 
Visibility (RVR):
Wind Speed/Gusts: 5 knots /
Turbulence Type Forecast/Actual:  /
Wind Direction: 90°
Turbulence Severity Forecast/Actual:  /
Altimeter Setting: 30.26 inches Hg 
Temperature/Dew Point: 24°C / 8°C
Precipitation and Obscuration: No Obscuration; No Precipitation
Departure Point: Maryland Hgts, MO (1H0)
Type of Flight Plan Filed: None
Destination: Charlotte, MI (FPK)
Type of Clearance: None
Departure Time: 15:19 Local 
Type of Airspace: Class G

Wreckage and Impact Information

Crew Injuries: 1 Fatal 
Aircraft Damage: Destroyed
Passenger Injuries: 3 Fatal
Aircraft Fire: None
Ground Injuries: N/A
Aircraft Explosion: None
Total Injuries: 4 Fatal
Latitude, Longitude: 39.241664,-89.915557

The airplane wreckage was located in an open field adjacent to a storage building and a pond. A portion of the left wing passed through the storage building roof and came to rest within the building. The debris path was about 400 feet long. The main wreckage consisted of the fuselage, engine, and propeller. The fuselage was fragmented, and the cabin was compromised. The engine was damaged consistent with impact forces and the engine mount was fragmented. The propeller remained attached to the engine. Both wings, the vertical stabilizer, and the stabilator had separated from the fuselage. All flight control surfaces were located within the debris field. The wreckage distribution appeared consistent with a low-level, in-flight break up.

A postaccident examination revealed that both wings had separated at the root. The wing spars exhibited upward bending adjacent to the fracture surfaces consistent with positive load factors (pitch up) at the time of the separation. The fracture surfaces exhibited a dull, grainy appearance consistent with overstress. The ailerons, including the counterweights, were separated from the wings and located within the debris path. Control cable and control rod separations were consistent with overstress.

The stabilator was separated except for the center spar section which remained attached to the aft fuselage hinge points. The balance weight and mast remained attached to the center spar section, and the stabilator control cables remained attached to the mast attachment points. The cables were continuous to the cockpit area. The vertical stabilizer was separated with exception of the aft spar which remained attached to the fuselage. The rudder remained attached to the spar at each hinge point. The rudder control cables were attached to the control surface and were continuous to the cockpit area. 

No preimpact anomalies with respect to the flight control system were identified.


  1. Friends don’t let friends fly with people with < 100 hours. They have absolutely no idea what they don’t know. In some cases you could make that a lot more time. Condolences to all of the families.

    1. Almost all pilots with less than 100 hours know better than to do 60+ degree steep turns with friends in a fully loaded aircraft. My instructor always told me your #1 job when flying friends is to give them the smoothest and most comfortable flight you can, and never try to show off or scare them.

    2. “Almost” only counts with horseshoes and hand grenades. The difference between what 100 hour pilots know and what they sometimes do is often large. Most survive it, learn from it and become better and wiser pilots. Sometimes the difference is written in blood.

  2. `The above post says it all. My condolences to all involved too.

  3. Low time pilot. Clearly not used to flying with that much extra weight on board. Probably simultaneous CG/control issues. Just not ready to take on that much responsibility. This is truly tragic and my thoughts and prayers go out to the families.

    1. as noted, medical 08/2018, "Last Flight Review or Equivalent: December 28, 2019" likely his PPL check ride, logged "38 hours (Pilot In Command, all aircraft)," accident May 20, 2020. RIP

  4. An aircraft is not like a high powered sports car .. so tragic in so many ways ... RIP

  5. As a non-pilot I have a question. They were on route from St. Louis to Michigan. They were about 30 minutes or so out of SL. Did the wing strut fail spontaneously or was it the result of the pilot "pulling G's" or some other unnecessary maneuver?

    1. Read the report posted here...

    2. General aviation planes are fragile, you can over stress one very easily, both wings separated at the root.

  6. Impromptu aerobatics ? Loaded outside cg/weight limits ?

  7. Sure looks like he was trying to give them a scary ride. He succeeded beyond his wildest dreams. Stressing a 55 year old airplane? Wow

  8. 38 hrs. as PIC. How did this brainchild ever get through flight training, let alone the check ride? He must have gone through a 'puppy mill' flight school.

    So, this know-it-all aviator loaded up three of his good buddies and flew them to their horrifying deaths - after showing off his aviating skills - loosing control, and uncontrollably diving into a 35mph over VNE in-flight breakup. It must have been absolutely terrifying for the passengers.

    Friends don't let friends fly with ... well, the first poster here nailed it!

    1. “Joshua always believed in education. He earned his pilot’s license and planned to pursue a commercial pilot’s license. I just want his love for education to continue on forever", his mother said.

    2. I went to school with him. He was a smart individual. I’m also a pilot and I would not have expected this behavior from him. I’m really conflicted about this report.

  9. The first turn wasn't exciting enough, so the next turn was made tighter.

    If their cell phones had survived and the unlock codes were known, one of the videos being made at the time would reveal how the fun turned to horror.

  10. I'm not seeing aircraft hours anywhere here, only pilot hours. Maybe I'm just missing it. It looks like a bad combination of low-time pilot, heavily loaded, and an airplane built 56 years earlier with a design known for some in-flight break-ups without being pushed all that hard--especially at around the 5,000+ hour mark.

    1. Airframe total time: 2234.86 as of May 1, 2020. You can find it in the NTSB Docket "Maintenance Records - Airframe".

    2. And the Airplane Examination Summary in the docket didn't report any conditions similar to the 2018 Embry-Riddle wing-off event.

  11. "CARLINVILLE, Ill. — On a clear and sunny Sunday, Angela Anderson saw a small plane doing dips in the air. Dillon Wiser pointed to the sky to show his son the plane’s tricks. Nearby, Danette Edwards thought the maneuvers were by an aerobatics pilot or crop duster.

    But the scene soon turned horrific.

    The four-seater aircraft, its engine revving loudly, began a nosedive. It spiraled straight down. The wings came off and the plane broke apart before slamming into a livestock ranch 3 miles southwest of Carlinville.

    Sweers’ mother, Georgeann Ricketts of Flushing, Michigan, told a reporter Wednesday that the agency must have missed the real cause and shouldn’t suggest her son made reckless moves.

    “That is not the personality or actions of my son,” she said.

    Ricketts said her son was cautious with the aircraft, built in 1964, and particular with his safety, including refusing to drink 24 hours before he planned to fly."

    1. That article also includes:
      "I believe they had to blame Josh,” he added. “They had no other choice. They had no other evidence.”

      NTSB doesn't deserve that characterization. NTSB would have noticed that the pilot had performed the same maneuver on the first day of that weekend's trip when the ADS-B track of 5/29/2020 was reviewed.

    2. Per the above article: Joshua Sweers enjoyed skydiving, motorcycle riding and climbing buildings; “lived every single day of his life on full throttle and arms wide open,” his obituary said.

  12. I'm going to play devil's advocate here and point out that the flight profile also looks like that of an emergency descent. Seems a strange place for them to just start doing maneuvers - in the middle of a x-country flight. Autopsy was done on the pilot, but not on any of the three pax. Had the pilot suddenly gone unconscious due to something that wouldn't be found on an autopsy, I imagine this is what it would look like when a non pilot pax tried to take the controls.

    Very likely it was negligence, but the randomness of stopping to do maneuvers where they did makes it suspicious enough that I'll refrain from urinating on this guy's grave.

    1. The pilot had done a similar maneuver during the earlier leg from his home in Michigan on the way down to spend the weekend with the Shedd's. The accident leg was while taking Shedd to Michigan to retrieve a motorcycle. Three of the four were aboard when the maneuver was performed the first day.

      Here is the turn maneuver of 29 May:

      It appears to have been a signature move. Shedd wasn't aboard for that 5/29 demonstration, so it was likely done for his benefit during the accident flight.

      News report with the weekend trip details:

      Later article, states that the pilot "lived every single day of his life on full throttle and arms wide open,”

    2. A sight-see or maneuvering variation of around +/- 10 kts. and total altitude change of around 1,000 ft. lines up with what is seen on the accident flight? I agree with others here that PA28/32 structure isn't renowned for its strength, especially with longevity, anyway, but unless he maintained ADS-B numbers that stable while pulling a lot of G's (kind of hard to do), I'm not seeing a clear connection.

    3. @Rel-poster - You don't see the double turn? The clear connection is that the pilot previously did the same double turn demonstration for two of his victims on the way to pick up his third. The examination of that earlier leg was prompted by reading:

      "Seems a strange place for them to just start doing maneuvers - in the middle of a x-country flight."

      Nothing unclear about acknowledging the pilot's same maneuver - in the middle of a x-country flight on the way down from Michigan with 2 pax aboard. It's Gryder-level deflection to instead think "emergency descent" after detecting the repeated behaviour.

      Realizing victim Shedd wasn't aboard for the first demo is part of understanding the scenario as it unfolded.

    4. I think his or her point is that there is nothing inherently dangerous or taxing on the airframe in the numbers there and I would have to agree.

    5. Noticing that a low-experience pilot performed the signature reversing direction double turn move when pax were aboard was not offered as an assertion that the earlier double turn demo of 5/29 overtaxed the airframe.

      A maneuvering crash like this is reminiscent of the N8849V crash by an experienced pilot who posted videos of performing a signature move. It's fun, but only until you lose control.

    6. Sorry, this was no "emergency descent" Read the witness reports. All of them reported the engine screaming at an extremely high, loud rpm. Every pilot knows that step one for an emergency descent is reduce power to idle. Also, the aircraft made no attempt to descend at all and in fact climbed slightly during the first 2-3 minutes after the initial turn. It wasn't until it's bank angle exceeded 70 degrees that aerodynamics took over and forced a descent.

    7. You're probably right, and I wasn't aware earlier of a similar maneuver during a previous flight. Still, the eye witness reports could also describe a pax having to take controls for some reason. Doesn't seem likely in this case, but until I heard of the similar maneuver profile from a previous flight, it was still suspect for me.

    8. @Kenneth- About the eye witnessed dips - If you scroll down, there is a comment about the docket's Aircraft Performance Memorandum data plots where those dips can be seen, showing up as divergent pitch oscillation that began during the left turn.

      Of the three times the reversing direction double turn signature move shows up in ADS-B data (May 24 practice w/"Steep Turn" YT video, May 29, and the accident day), the aircraft was heaviest on the accident day.

      Makes sense that the response lagged his control inputs more due to the increased weight aboard on that day. Pitch oscillation building up from the pilot over-controlling his inputs and "chasing" the aircraft's response started as dips of increasing magnitude that diverged into loss of control.

  13. "I'm going to play devil's advocate here." You did just the opposite--good advocacy for reason and caution. I also wondered about the supposed aerobatic / video-taking display occurring where it did. I would want to know if there was any reason identified for why it would be done there in the middle of a cross-country segment. The crash site also looks quite like the sort of emergency-landing site that I might have identified in a dire circumstance.

  14. "Steep Turn" video posted by the pilot a few days before the accident:

    1. Steep Turn
      May 25, 2020

    2. That Steep Turn video was just a portion of a right turn made near Woodbridge Elementary School during a 24 May flight recorded in AdsbExchange. The right turn's location is easily confirmed by observing the distinctive ponds visible in the first seconds of the vid.

      A full length video of the entry into the left turn and continuing all the way to exit of the right turn would have provided an insightful visual bank angle reference for comparison of the accident day maneuver against the ADS-B data recorded for the 24 May reversed turn pair.

      Maybe the accident day's extra weight of four on board changed the response such that exiting from the left turn and making the right turn entry in close succession was simply beyond practiced experience.

      Flight track of the 24 May right (and left) turn pair:

      Map-pinned location of the distinctive ponds:

    3. A rough analysis of that steep turn video shows the bank angle was around 60 degrees at times, a full 15 degrees greater than recommended for a private pilot steep turn maneuver.

      This extreme bank angle would impose 2 Gs of force on the aircraft (versus 1.4 Gs at 45 degrees) and increase the stall speed by a factor 40% as well as causing much higher overbanking tendencies.

      Also, based on that ADS-B track above, it looks like there was no attempt made to roll out on the heading the turns were initiated from, which is one of the key elements to achieve in a precise steep turn.

  15. I wonder if Piper has considered a spar strap, like Beechcraft has done for various King Air models. While denying any fault with the design. It was a common modification for Beech 18's, also. Whether or not Piper is technically at fault for this and other wing spar failures, it doesn't look good. The last few years, I have associated "spar failure" with "Piper PA28", and that does not behoove Piper. Maybe they don't care, now that they are Brunei owned.
    I guess the speeds given in figure 4 must be ground speeds, as they are nowhere near the 138 knot maneuvering speed.

    1. I own a 235, the rough air/maneuvering speed is 138 mph or about 119 knots.

      Agree that the data is likely groundspeed.

    2. Keep in mind that a descent straight down would have an apparent ground speed of zero even with calm winds aloft. I recommend looking at the "Aircraft Performance Memorandum" in the docket. It's a great read! Basically they calculated the winds aloft as well as total inertial speed (the combination of ground speed and vertical speed) from the ADS-B data and onboard Stratus logs to determine what the calibrated airspeed was during the descent. They determined that during the dive, the pilot exceeded 200 KCAS and then pulled 4.72 Gs trying to pull out of the dive. That's enough to tear the wings off most standard GA aircraft, not just a Piper.

  16. The docket's Aircraft Performance Memorandum data plots make it easy to see where the witness's description of five or six “dips” occurred. The description of dips where the airplane would “come back up” each time are pitch oscillations.

    The time period of 15:45:00 to 15:45:30 is where the dips are recognizable and they are going divergent, increasing in magnitude at each pitch reversal across the zero pitch line of the chart. That time interval is during the first turn, before the direction changes from a left turn to a right turn.

    Divergent pitch oscillation building up during the left turn suggests over-controlled pilot inputs, "chasing" the aircraft's response. The roll out from the left bank briefly produces a big increase in lift as wings pass thru the zero roll angle in a wings level orientation, and the pilot inputs made to counter the resulting nose high pitch worsens the divergent pitch chase.

    Divergent control chases are unnerving and difficult to recover from.

    See it clearly in the top plot of page 16, here:

  17. I don't believe everything that I see on the internet.
    According to the pilot's obituary, his other hobbies included skydiving, motorcycle riding and climbing buildings.
    His Youtube channel shows steep turns in an aircraft.
    His Facebook page shows zip-lining and bungee jumping.
    I wonder if the mid cross country interruption of his flight was motivated by thrill seeking behavior.

    1. Quote:
      / AnonymousMonday, May 2, 2022 at 8:23:00 PM EDT

      The pilot had done a similar maneuver during the earlier leg from his home in Michigan on the way down... /

    2. After 51 years of flying, I can relate to doing stupid things with airplanes.

  18. A statement my CFI made that's always stuck. 'The two most dangerous words in aviation are 'Watch This''.

  19. Lesson is: Be careful who you trust with your life. I flew one time with a low-time pilot who thought he should buzz an attraction. I was ready to grab the controls at the first sign of stall. Never flew with him again. Later, he told me of taking off with an over-loaded 140, just barely getting airborne. An idiot. Some people get stuck in the idea of flying without comprehending the finality that results from missing the mark.

  20. Josh was a friend and a colleague. At work, he was pragmatic, risk mitigating, and humble. It's sad to learn that a lapse in those qualities led to the deaths of 4 people. He was a truly good guy. Yes, he f'd up- but he's not a villain and I miss my friend. If there's a paradise after this, I look forward to having out cigar together.

  21. you can't fix dumb...,the laws of physics can

  22. Physical event? His buddy trying to fly to save them??