NTSB CAROL · Event
Event DEN03FA155
Registry · N45BP
FAA Aircraft Registry record.
Make / Model
BOMBARDIER INC BD-700-1A11
Year of manufacture
2006
Engine
ROLLS DEUT BR700-710A220
Seats / Engines
8 seats · 2 engines
Last airworthiness date
20070406
ADS-B equipped
Yes — Mode-S A56FEA
Registrant of record
TVPX AIRCRAFT SOLUTIONS INC TRUSTEE
Source: FAA Aircraft Registry (releasable master file).
Aircraft involved
Probable cause & findings
the first pilot's failure to maintain aircraft control and adequate airspeed. Contributing factors included performing intentional stalls at too low an altitude to afford a safe recovery, the pilot's failure to add power in an attempt to recover, and the flight instructor's inadequate supervision of the flight.
Factual narrative
HISTORY OF FLIGHT
On September 20, 2003, at 1854 central daylight time, a Hawker Siddeley HS-125-700A, N45BP, operated by Starflite Management Group, Inc., of Houston, Texas, was destroyed when it impacted terrain approximately 15 miles northwest of Southeast Texas Regional Airport (BPT), Beaumont, Texas. The airline transport certificated flight instructor, and the airline transport pilot and commercial pilot who were receiving instruction (hereinafter referred to as the instructor-pilot, first pilot, and second pilot, respectively) were fatally injured. Visual meteorological conditions prevailed. An IFR (instrument flight rules) flight plan had been filed for the instructional flight being conducted under Title 14 CFR (Code of Federal Regulations) Part 91. The flight originated at Houston-Hobby Airport (HOU), Houston, Texas, at 1759. According to the Federal Aviation Administration (FAA), the instructor-pilot was preparing the first and second pilots for their FAA Part 135 competency and proficiency checks scheduled to be conducted in the accident airplane the following week. Operator proving tests were to follow shortly thereafter. The operator, Starflite Aviation, had five other airplanes on its air carrier certificate: a Beech 200 King Air, an IA-1121 Commodore Jet, two IA-1124 Westwinds, and a Lockheed L-1329 JetStar. According to FAA documents, the first pilot obtained a CSC (Computer Science Corporation) DUATS (Direct User Access Terminal Service) weather briefing and filed an IFR flight plan, indicating the airplane would fly to Beaumont in 23 minutes at 250 KTAS (knots true airspeed) and at 5,000 feet. The instructor-pilot was listed as the pilot-in-command. An IFR clearance to Beaumont was issued at 1748, and taxi clearance to runway 22 came at 1749. After the 1759 takeoff, the crew contacted Houston Departure Control at 1801, and was handed off to Beaumont Approach Control at 1810. At 1815, the flight was cleared into a practice area, the boundaries being between the 270-degree and 360 degree radials and within 20 DME (Distance Measuring Equipment) miles of the Beaumont VORTAC (Very high frequency Omnidirectional Radio range TACtical Air Navigation). The airplane was to remain between 5,000 and 7,000 feet msl. At 1828, Beaumont Approach Control instructed the crew to turn eastbound because they was approaching the fringes of its practice area and entering Houston airspace. The crew acknowledged and said that they were turning around. That was the last radio contact with the airplane. The closed-loop cockpit voice recorder, that records ambient cockpit sounds for the past 30 minutes, started at 1822:50 and ended at 1854:27. Only the last 13 minutes of the recording, beginning at 1841:14 and ending at 1854:27, were transcribed. Prior to the start of the transcript, several steep turns and stalls had been performed. The second pilot was complimented on his performance. He remarked that he had never flown a jet before and would have to get use to its feel. The instructor-pilot then asked the first pilot to perform a stall in the approach configuration. The first pilot asked the instructor-pilot if he had "ever done stalls in the airplane?" The instructor-pilot replied, "It's been awhile." The first pilot remarked, "This is the first time I've probably done stalls in a jet. Nah, I take that back, I've done them in a (Lear)." The instructor-pilot said he had stalled "the JetStar on a [FAR] one thirty five ride." At 1848:18, the first pilot asked for approach flaps (15 degrees) and for the landing gear to be lowered. The second pilot complied. The stall was performed and the recovery was accomplished. At 1849:57, the second pilot announced, "Five thousand feet, gear's up and locked. Vee two plus twenty, your flaps [are] up." At 1850:22, the instructor-pilot briefed the pilots on the next maneuver, an approach-to-landing stall: "Okay, accelerate back to two hundred knots, five thousand feet. Find the ref[erence] speed. Do the checklist for the ref speed for our weight. We're at uh fourteen and seven is twenty one. Approach to landing stall is next. What did I say, twenty one? One twenty two for the ref. Your power is gonna be back at idle, you know. What else? Flaps. You know, approach flaps, gear. You don't wanna get flaps in there * late. No, no more trimming past, I think its one fifty or one sixty. Recover. Just like a go-around maneuver. Power, positive rate, flaps ten, okay, positive rate. Gear up. Ref plus twenty. Flaps up." The second pilot questioned the configuration: "So this one the flaps don't go all the way to forty-five, they just go to twenty-five?" The instructor-pilot said, "No, full flaps." A discussion followed as to whether the stall should be performed in a turn or straight ahead. It was decided that the stall would be done straight ahead. At 1852:58, the first pilot asked for approach flaps and then for the landing gear to be lowered. The second pilot then reported, "Flaps twenty-five, set." The first pilot said, "Flaps," and the second pilot replied, "Flaps," and there was the sound of a click. The instructor-pilot reminded them, "*** power," and there followed the sound of decreasing power. According to the digital electronic engine control (DEEC) data, power decreased to 37 percent N1. At 1953:57, the stick shaker sounded. Radar-computed ground speed was 192 knots and decreasing rapidly. Altitude was 4,900 feet msl. DEEC data depicted a commanded increase to takeoff power and the instructor-pilot said, "Aww, don't do that now." DEEC data then showed a commanded power reduction to between 30 and 40 percent N1. At 1854:03, the first pilot said, "Gimme flaps." At 1854:08, the second pilot asked, "What do you want me to do?" Ground speed was between 112 and 113 knots. At 1854:10, the first pilot said, "Recover." There was the sound of increasing background noise, and at 1854:17, the second pilot said, "Power up, power up, power. Do something, man." The instructor-pilot said, "Power up." DEEC data did not record any increase in power, even though both engines were operating and controllable by pilot command. The CVR recording ended at 1854:27. Shortly thereafter, during the 1854 time frame, Beaumont Approach Control gave the flight a new altimeter setting. There was no acknowledgement. At 1855, the approach controller advised that he was not receiving the airplane's transponder signal. There was no reply. Repetitive calls were made at 1855, 1856, and 1908. There was no response to either call. Twenty-five witnesses were either interviewed by telephone or in person. The consensus was that the airplane was flying at a low altitude and doing "erratic maneuvers." One witness said that when airplane emerged from the overcast, it "seemed to stop in midair," then it pitched nose down and disappeared behind the tree line. Several witnesses said the airplane was spinning --- some described it as a flat spin --- before it struck the marshy ground. One witness said the airplane fell "like a falling leaf." The accident occurred during the hours of daylight at a location of 030 degrees, 08.44 minutes north latitude, and 094 degrees, 13.19 minutes west longitude, and at an elevation of 12 feet msl.
CREW INFORMATION
According to Title 14 CFR Part 1.1, the "pilot in command" is the person who: "(1) has final authority and responsibility for the operation and safety of the flight, (2) has been designated as pilot in command before or during the flight, and (3) holds the appropriate category, class, and type rating, if appropriate, for the conduct of the flight." Based on this definition, the pilot-in-command was the instructor-pilot, who, according to the CVR transcript, was seated in the jump seat. The instructor-pilot was also listed on the filed flight plan as being the pilot-in-command. The instructor-pilot, age 36, held an airline transport pilot certificate, dated November 19, 2002, with an airplane multiengine land rating, type ratings in the Hawker Siddeley HS-125, Dassault DA-2000, and Lockheed L-1329, and commercial privileges in airplanes single-engine land. He also held a flight instructor certificate, dated January 13, 2003, with airplane single/multiengine and instrument ratings, and a ground instructor certificate, dated May 20, 1988, with a basic rating. His first class airman medical certificate, dated July 18, 2003, contained no restrictions or limitations. He was hired by Starflite in June 2003, and completed company training in July 2003. His flight training record, dated July 22, 2003, contained the following notations: "Flight 1, 2. Good control - smooth - knows flows. Flight 3. Check ride ready. Flight 4. Retrain for check ride failure." According to his resume and personnel records, he had logged the following flight time (in hours): Total time: 5,230 Pilot-in-command: 3,521 Second-in-command: 1,455 Turbine: 3,231 Pilot-in-command, turbine: 1,776 Multiengine: 3,290 Instrument: 367 The first pilot, age 42, held an airline transport pilot certificate, dated March 25, 2003, with an airplane multiengine land rating, type ratings in the Beech 300 and Hawker Siddeley HS-125, and commercial privileges in airplanes single-engine land. He also held a flight instructor certificate, dated February 28, 2003, with airplane single/multiengine and instrument ratings. His first class airman medical certificate, dated June 30, 2003, contained no restrictions or limitations. According to his application for this medical certification, he indicated he had logged 3,800 total flight hours, of which 350 hours were accrued in the previous 6 months. Starflite hired him in August 2003. According to his resume and personnel records, he had logged the following flight time (in hours): Total time: 3,817 Pilot-in-command: 2,684 Turbine: 1,575 Pilot-in-command, turbine: 617 Turbojet: 855 Pilot-in-command, turbojet: 80 Airplane, single-engine: 1,620 Airplane, multiengine: 2,198 Second-in-command: 873 Night: 304 Instrument: 493 Cross-country: 2,730 Instructor: 665 The second pilot, age 27, held a commercial pilot certificate, dated August 19, 2000, with airplane single/multiengine land and instrument ratings. He also held a flight instructor certificate (gold seal), dated April 19, 2002, with airplane single/multiengine and instrument ratings, and a ground instructor certificate, dated January 7, 2002, with an instrument rating. His first class airman medical certificate, dated March 24, 2003, contained no restrictions or limitations. According to his application for this medical certification, he indicated he had logged 2,000 total flight hours, of which 600 hours were accrued in the previous 6 months. Starflite hired him on September 1, 2003. Officials said the accident flight occurred on his first day of training. According to his resume and personnel records, he had logged the following flight time (in hours): Total time: 2,400 Pilot-in-command: 2,200 Airplane, single-engine: 1,100 Airplane, multiengine: 1,300 Night: 1,000 Instrument: 250 Cross-country: 1,600
AIRCRAFT INFORMATION
Hawker Siddeley manufactured N45BP (s/n NA0219, formerly N219TS), a model HS-125-700A, in 1978. It was equipped with two Garrett TFE731-3--1RH turbofan engines (s/n P-08159, left; P-80160, right), each rated at 3,700 pounds of thrust. Starflite documents indicated the program used to maintain N45BP was outlined in FAR 91.409(f)(3). According to the maintenance records, the engines were given a "Garrett pre-purchase evaluation" on June 3, 2003, and "a list of discrepancies [were] provided to the customer." Total airframe time was 9,690.4 hours. The airplane was then placed under "Raytheon's Flexible Maintenance Schedule [FAR 91.409(f)(3)]." The airplane was last inspected in August 2003 (day of the month not given). According to the Daily Aircraft and Engine Log, N45BP flew the day before the accident. At the end of the day, the airframe had accrued 9,780.1 hours total time and 7,098 landings. The left and right engines had accrued 9,359.9 and 9,489.9 hours, and 7,053 and 6,734 cycles, respectively, since new.
METEOROLOGICAL INFORMATION
The following METAR (routine aviation meteorological report) observations were recorded at Beaumont Airport at 1753 and 1853, respectively: Wind, 070 degrees at 9 knots; visibility, 7 statute miles; sky condition, clear below 12,000 feet msl; temperature, 27 degrees C.; dew point, 21 degrees C.; altimeter setting, 29.97. Wind, 090 degrees at 7 knots; visibility, 8 statute miles; sky condition, clear below 12,000 feet msl; temperature, 26 degrees C.; dew point, 21 degrees C.; altimeter setting, 29.98.
AIDS TO NAVIGATION
There were no reported difficulties with aids to navigation.
COMMUNICATIONS
There were no reported communications difficulties.
FLIGHT RECORDERS
The airplane was equipped with a Fairchild A-100a cockpit voice recorder (CVR). It was recovered and sent to NTSB's vehicle recorder laboratory where, on October 9, 2003, the CVR Group convened to audition the tape. The group consisted of NTSB's investigator-in-charge and representatives from FAA, Raytheon Aircraft, and Starflite Aviation. A transcript of the last 13 minutes of the CVR tape was prepared and made part of this report (see EXHIBITS). The engines installed in N45BP were equipped with digital electronic engine controls (DEECs) with non-volatile memory chips. Both DEECs were recovered and, though substantially damaged, were shipped to Honeywell's Engines, Systems & Services (ES&S) Laboratory in Tucson, Arizona, where, on October 8, 2003, data was downloaded and analyzed under the auspices of FAA aviation safety inspectors. Data from the accident flight could only be retrieved from the right engine DEEC (see TESTS AND RESEARCH and EXHIBITS).
WRECKAGE AND IMPACT INFORMATION
The on-site investigation commenced on September 21 and terminated on September 24. Inspection revealed a water-filled crater that looked like the outline of an airplane, including its nose, wings, and forward fuselage. The crater was aligned on a magnetic heading of about 055 degrees. The airframe was extensively fragmented. The main body of wreckage, consisting of pieces of the cabin, aft fuselage, both engines, empennage, and cockpit roof were found within 25 feet of the initial impact crater. The separated landing gear legs were found buried in the mud in the crater. The landing gear was down. Only one flap actuator was recovered. Its extension corresponded to a flap setting of 25 degrees (approach flaps). Smaller pieces of wreckage (overwing emergency exit hatch, seats, seat cushions, and insulation) were strewn up a hill on a magnetic heading of 158 degrees. Both engines' fan blades were gouged and bent in the opposite direction of rotation, and some fan blades were separated in reverse bending. There were rotational scoring marks on both cases, spinners, and spinner supports. Metal spray was evident on both engines' third-stage low-pressure turbine blades. Examination of various cockpit instruments revealed an airspeed indicator that registered 104 knots, an altimeter indicating 4,240 feet, and the flap indicator gauge that showed about 15 degrees down. Both engine Fan Speed gauges read approximately 29 per cent.
MEDICAL AND PATHOLOGICAL INFORMATION
Autopsies were performed on all three pilots by a forensic pathologist at the Jefferson County Morgue, and FAA's Civil Aeromedical Institute (CAMI) performed toxicological screens on the three sets of specimens submitted. According to CAMI's reports, no ethanol was detected in muscle or brain tissue, and no drugs were detected in liver tissue. Carbon monoxide and cyanide tests were not performed.
TESTS AND RESEARCH
National Track Analysis Program (NTAP) radar data, DEEC data, and the CVR transcript were collated. Although NTAP data indicated the last known radar contact was at 1854:13, a primary contact was recorded at 1854:27 (no data tag attached). Time on the CVR transcript was based this primary contact, 1854:27, and worked backwards. Since DEEC data depicted only the last 51 seconds, it was manually transposed to the NTAP plot and CVR transcript. No data from the accident flight was recovered from the left engine DEEC recorder. The right engine data showed that maneuvering power had been between 65 per cent and 70 per cent N1. Then, 51 seconds before impact, the power lever was retarded to idle and remained in that position for approximately 20 seconds. The altitude sensor registered between 4,000 and 6,000 feet. Takeoff power was then applied for 5 seconds. Twenty-four seconds before impact, the power lever was again retarded to idle and remained there until impact. The altitude sensor registered between 4,000 and 6,000 feet. Ten seconds later, or 14 seconds before impact, the altitude sensor registered between 2,000 to 4,000 feet. Seven seconds later, or 7 seconds before impact, the altitude sensor registered less than 2,000 feet.
ADDITIONAL INFORMATION
The following is an excerpt from John Lowery's book on Aerodynamics: "The T-tail design for swept-wing jet transports...has provided a special dilemma. With the horizontal tail mounted high atop the rudder, it would appear that...tail damping in a stall or spin would be strong. Unfortunately this is not the case. For at maximum AOA (angle of attack) the wake of the stalled wing and engine nacelles envelopes the empennage and destroys much of the tail-damping and most of the horizontal stabilizer's effectiveness. This results in a 'locked-in deep stall'...Because the horizontal stabilizer of a T-tail is above the wing wash, there is no telltale elevator buffet prior to stall. Thus to warn of an approaching stall, and to preclude the deep stall phenomenon, a stick or rudder shaker is required." The following are excerpts from the HS-125 Crew Manual: "Intentional stalls are only permitted when...altitude [is] between 10,000 feet and 18,000 feet...Intentional stalling is restricted to flaps 0 degrees or 15 degrees...Speed should be reduced at not more than 1 knot per second. Rapid or violent movements of any control during the approach to the stall should be avoided...Immediately the stall is recognized, prompt and positive recovery action should be taken by forward movement of the control column until normal flight is resumed. Any rolling tendency should be corrected by use of ailerons...There is no natural stall warning or aerodynamic buffet prior to the stall. Stall warning is provided by a stick shaker...The stall occurs suddenly and is defined by the onset of buffet associated with a nose drop..." In addition to the Federal Aviation Administration, parties to the investigation included the Raytheon Aircraft Corporation and the Honeywell (Garrett) Corporation. The wreckage was released to the insurance company on September 23, 2003, The cockpit voice recorder was released to the insurance company on May 20, 2004. The release of the tape recording from the cockpit voice recorder to the operator is pending. The purpose of the flight was for the instructor-pilot to prepare the first and second pilots for their FAA Part 135 competency and proficiency checks scheduled to be conducted in the accident airplane the following week, with operator proving tests to follow shortly thereafter. The first pilot obtained a computer science corporation (CSC) direct user access terminal service (DUATS) weather briefing and filed an instrument flight rules (IFR) flight plan. The instructor-pilot was listed as the pilot-in-command. The airplane took off and proceeded to its designated practice area. According to the cockpit voice recorder (CVR), the pilots practiced various maneuvers under the direction of the instructor-pilot, including steep turns and approaches to stalls. Then the first pilot was asked the to demonstrate an approach-to-landing stall. The first pilot asked the instructor-pilot if he had "ever done stalls in the airplane?" The instructor-pilot replied, "It's been awhile." The first pilot remarked, "This is the first time I've probably done stalls in a jet. Nah, I take that back, I've done them in a (Lear)." The instructor-pilot said he had stalled "the JetStar on a [FAR] one thirty five ride." Flaps were extended and the landing gear was lowered. Digital electronic engine control (DEEC) recorded a power reduction that remained at idle. According to national track analysis program (NTAP) data, the stall was initiated from an altitude of 5,000 feet. The stick shaker sounded and shortly thereafter, the recording ended. The consensus of 25 witness' observations was that the airplane was flying at low altitude and doing "erratic maneuvers." One witness said it "seemed to stop in midair," then pitched nose down. Some witnesses said that the airplane was spinning. Other witnesses said it was in a flat spin. Still another witness said the airplane fell "like a falling leaf." The airplane impacted marshy terrain in a nose-down, wings-level attitude. Wreckage examination revealed the landing gear was down and the flaps were set to 25 degrees. Both engines' compressor/turbine section blades were gouged and bent in the opposite direction of rotation, and there were rotational scoring marks on both cases. No discrepancies were noted. Source: NTSB Aviation Accident Database (Pre-2008 Archive) Retrieved: 2026-02-12
Verbatim from NTSB's published report. Source file
NTSB_2003_DEN03FA155.txt.
Findings + structured fields enriched from FAA avall.mdb.
Full investigation docket on
data.ntsb.gov ↗.
Beyond the agency record
Search this event elsewhere.
Pre-filled searches into the sources where news + community discussion of aviation events lives. External sources are reported, not agency. Treat them as signal that something happened, not as fact about what happened.
Entity-clustered aviation events in the press — last 24 hr + 30-day archive.
Official agency record + docket.
Investigative docket: factual reports, photos, transcripts.
Long-running aviation incident database (Flight Safety Foundation).
Community NTSB synthesis blog — often has photos and witness reports.
Gold-standard aviation incident blog.
Aviation industry news search.
GA pilot forum — informed but rumor-prone.
GA pilot subreddit search.
Tail-number page — flight history (free tier limited).
AOPA Air Safety Institute search.
Mainstream press coverage. Recent events only.
Privacy-preserving news search.
External links open in a new tab. We don't ingest their content; we deep-link search queries.
Related research
What the literature says.
Academic papers and agency reports matching this event's aircraft type or causal vocabulary (stall, go-around, maintenance). Sourced from NASA NTRS, NTSB Safety Studies, FAA CAMI, AOPA Air Safety Institute, Embry-Riddle Scholarly Commons, arXiv, and the Semantic Scholar academic graph.
- Embry-Riddle Scholarly Commons 2023 · Conference paper
The Value of Strong Partnerships to Build a Successful Aviation Maintenance Career Pathway Program for Transitioning Military Service Members
The aerospace industry is competing with other industries for a qualified workforce, and many of those competing industries are investing heavily in creating workforce development pipelines.
- Embry-Riddle Scholarly Commons 2026 · Journal article (IJAAA)
From Reactive to Predictive: A hybrid Trust-Mediated Adoption Framework for Data-Driven Maintenance in Distributed-Authority Aviation Environments
Modern aviation maintenance operates within increasingly data-intensive technological environments, yet the operational integration of predictive maintenance into routine decision-making remains incon…
- NASA NTRS 2026 · Conference Paper
Computational Analysis of Steady State Aerodynamics of Transonic Truss-Braced Wing Configuration in Deep Stall
This study presents a computational investigation of steady state aerodynamics of the Subsonic Ultra-Green Aircraft Research (SUGAR) Transonic Truss-Braced Wing (TTBW) configuration over a wide range …
- NASA NTRS 2025 · Conference Paper
A Training Study to Improve Monitoring During A Go-Around
As part of an FAA program to improve go-around (GA) safety, we were asked to determine if we could improve the performance of the Pilot Monitoring (PM) during a GA maneuver.
- Semantic Scholar 2025 · Article (Applied Sciences)
Decision-Making Framework for Aviation Safety in Predictive Maintenance Strategies
The implementation of predictive maintenance (PM) in aviation presents unique challenges due to strict safety requirements, complex operational environments, and regulatory constraints.
- Flight Safety Foundation 2024 · FSF / AeroSafety World
Go-Around Safety Forum Findings
Foundation Go-Around Safety Forum technical findings — examines why pilots fail to execute go-arounds when criteria are met (stabilized approach gate not met, energy state out of envelope, traffic con…
Browse the full corpus — academia portal ↗