Mikoyan Gurevich MIG-23UB accident near Belleville, Michigan, August 13, 2023
On August 13, 2023 at about 8:10 pm local time, a 1981 Mikoyan Gurevich MIG-23UB, registered N23UB, was destroyed in an accident during maneuvering (low-alt flying) near Belleville, Michigan (Willow Run airport). It was an air show or air race flight under general aviation rules (Part 91). 2 people were seriously injured. The weather was visual conditions (good weather).
The NTSB's probable cause their words, unchanged
A reported partial loss of engine power while maneuvering for reasons that could not be determined.
Source: NTSB aviation accident database, copy made October 5, 2026. Docket and reports at the NTSB.
What the record shows
- Date
- August 13, 2023 · about 8:10 pm local time
- Place
- Belleville, Michigan · Willow Run · map
- Type
- Accident
- Injuries
- 2 people were seriously injured.
- Weather
- visual conditions (good weather)
- Aircraft
- Mikoyan Gurevich MIG-23UB, built 1981
- Registration
- N23UB · registry record · serial 1038107
- Damage
- Destroyed
- Flight
- Air show or air race flight · general aviation rules (Part 91)
The NTSB's narrative final · quoted from the NTSB record
Both the front-seat pilot-in-command (PIC) and the rear-seat pilot-rated observer (PRO) of the Russian-manufactured fighter jet reported that the airplane’s engine did not respond to application of afterburner power during a display pass at an air show. According to the PIC, the engine was operating in a degraded power condition that was insufficient to sustain altitude and airspeed. The PIC maneuvered the airplane back toward the airport while he attempted to restore engine power. The PIC believed that his corrective measures would have restored engine power; however, the PRO activated the crew ejection system before engine power was restored. The PRO reported that he inquired about ejection with the PIC but did not hear a response before he activated the ejection system; however, the PIC reported that he replied “no” in response to the PRO’s ejection inquiry and that he was ejected from the airplane while he was troubleshooting the loss of engine power and still flying the airplane. Although the PIC was completing corrective actions, engine power was not restored by the time of the crew ejection and, therefore, the investigation was unable to determine if his corrective efforts would have been successful. According to the PRO, the ejection system was activated about 350 ft above ground level (agl) and below 200 kts. Both pilots were ejected from the airplane, with the PRO ejected first, followed by the PIC. The airplane continued in a left bank and descended into the ground about 1.7 miles from the runway. The airplane impacted the ground adjacent an apartment building and was destroyed by impact forces and a post impact fire. Other than the PIC’s mechanic, the NTSB was unaware of any independent engine shop or manufacturer within the United States with the appropriate tooling and expertise to perform a teardown examination of the Tumansky R27F2M-300 engine. The PIC’s written statement addressed potential sources for degraded engine performance; however, the NTSB could not validate the PIC’s opinion of the reason for the loss of engine power. Based on photographic evidence, the engine was producing some power at impact; however, the level of power output could not be determined. Due to investigative support limitations and the extend of damage, the investigators were unable to determine the reason for the loss of engine power. The airplane flight manual indicated that during a descent, the minimum safe altitude for crew ejection is calculated based on the airplane’s rate of descent and is equal to the rate of descent multiplied by 5. Based on the 350 ft agl altitude at ejection provided by the PRO, this altitude would be appropriate for ejection at a maximum rate of descent of 70 feet per minute (fpm). Although the airplane’s actual rate of descent was not known, it was likely significantly higher than the 70 fpm limitation at the time of crew ejection. Additionally, video evidence showed that the ejection seat parachutes did not fully deploy until the occupants had descended to just above treetop level. Therefore, the timing of the crew ejection was not premature. Given the serious injuries received by both occupants, crew ejection at a higher altitude may have reduced the severity of their injuries. Likewise, had the crew ejection been further delayed, the occupant injuries may have been more severe.
The complete narrative as the NTSB published it. The NTSB's docket holds the report as a PDF and any photographs, statements and other documents from the investigation.
The factual record from the NTSB's investigation tables, in plain English
What happened, in order
- Unknown or undetermined during maneuvering (low-alt flying) defining event
The NTSB's findings
- Aircraft › Aircraft power plant › (general) › (general) › Unknown/Not determined
Pilot
- Certificate: airline transport pilot, flight engineer, private
- Ratings: multi-engine land; single-engine land; instrument: airplane
- Flight time: 21,296 hours in all; 42 in this make and model; 79 in the last 90 days; 5 in the last 30 days; 14,000 as pilot in command
- Last flight review: March 21, 2023
- Medical certificate: Class 1 (without waivers/limitations)
- Seat: frt
- Injury: minor injuries
Other crew
- Certificate: airline transport pilot, commercial pilot, private
- Ratings: multi-engine land; single-engine land; instructor: airplane multi-engine; instructor: airplane single-engine; instructor: instrument airplane; instrument: airplane; rotorcraft: helicopter
- Flight time: 15,000 hours in all; 12 in this make and model
- Medical certificate: Class 1 (without waivers/limitations)
- Seat: rear
- Injury: minor injuries
The aircraft
- Airframe total time: 1,697 hours
- Last inspection: approved inspection programme, July 17, 2023; 5 hours since
- Seats: 2
- Landing gear: retractable
- Engine: Tumansky R27F2M-300 (turbojet); 890 hours total
- Fire on the ground
The flight
- Departed from: YIP Belleville MI at 8:00 pm
- Flight plan: none
- A second pilot was aboard
Weather at the time
- Light: daylight
- Wind: from 330° at 10 knots
- Visibility: 10 statute miles
- Sky: a few clouds at 6,500 ft
- Temperature: 81°F (27°C), dew point 63°F (17°C)
- Altimeter: 29.98 inHg
- Observation at 3:53 pm from KYIP, 1 miles away
Weather report (METAR): METAR KYIP 131953Z 33010KT 10SM FEW065 27/17 A2998 RMK AO2 SLP146 T02670172=
Injuries
| Fatal | Serious | Minor | None | |
|---|---|---|---|---|
| Flight crew | 1 | |||
| Passengers | 1 |
Documents from the investigation the NTSB's docket: the evidence folder behind the report
7 documents, released by the NTSB on August 27, 2025. View them here, or download them; the NTSB redacts some personal details before release, and this site shows the NTSB's own titles rather than its file names.
| # | Document | What it is | |
|---|---|---|---|
| 1 | Accident Photographs | PDF, 9 pages | View Download |
| 2 | Pilot/operator Aircraft Accident Report, NTSB Form 6120.1 (PIC) | PDF, 29 pages | View Download |
| 3 | Pilot/operator Aircraft Accident Report, NTSB Form 6120.1 - (Observer Pilot) | PDF, 11 pages | View Download |
| 4 | Powerplants - Factual Report | PDF, 5 pages | View Download |
| 5 | Memorandum - Conversation with Observer/pilot (01-30-2025) | PDF, 2 pages | View Download |
| 6 | Memorandum - Conversation with Mig 23 Pilot and Former USAF Test Pilot | PDF, 2 pages | View Download |
| 7 | Mig 23 Flight Manual Excerpt | PDF, 2 pages | View Download |
The same docket at the NTSB · documents without a copy here are fetched from the NTSB when you open them.
Everything on this page comes from the NTSB's public records. The narrative, probable cause and findings are the NTSB's own words; the coded tables behind the report are written out in plain English, with pilots' ages, home towns and medical details left out. The documents and photographs are the NTSB's docket, shown as the NTSB released them. This site's own text never names anyone involved. A preliminary report can change; the final report usually follows one to two years later, and the page is refreshed when it does.
