AAIB Bulletin 6/2023

TO REPORT AN ACCIDENT OR INCIDENT

PLEASE CALL OUR 24 HOUR REPORTING LINE
01252 512299
 Air Accidents Investigation Branch
Farnborough House AAIB Bulletin: 6/2023
Berkshire Copse Road
Aldershot GLOSSARY OF ABBREVIATIONS
Hants GU11 2HH
aal above airfield level kt knot(s)
ACAS Airborne Collision Avoidance System lb pound(s)
ACARS Automatic Communications And Reporting System LP low pressure
Tel: 01252 510300
ADF Automatic Direction Finding equipment LAA Light Aircraft Association
Fax: 01252 376999 AFIS(O) Aerodrome Flight Information Service (Officer) LDA Landing Distance Available
Press enquiries: 0207 944 3118/4292 agl above ground level LPC Licence Proficiency Check
http://www.aaib.gov.uk AIC Aeronautical Information Circular m metre(s)
amsl above mean sea level mb millibar(s)
AOM Aerodrome Operating Minima MDA Minimum Descent Altitude
APU Auxiliary Power Unit METAR a timed aerodrome meteorological report
ASI airspeed indicator min minutes
ATC(C)(O) Air Traffic Control (Centre)( Officer) mm millimetre(s)
ATIS Automatic Terminal Information Service mph miles per hour
ATPL Airline Transport Pilot’s Licence MTWA Maximum Total Weight Authorised
AAIB investigations are conducted in accordance with BMAA British Microlight Aircraft Association N Newtons
BGA British Gliding Association NR Main rotor rotation speed (rotorcraft)
Annex 13 to the ICAO Convention on International Civil Aviation, BBAC British Balloon and Airship Club Ng Gas generator rotation speed (rotorcraft)
EU Regulation No 996/2010 (as amended) and The Civil Aviation BHPA British Hang Gliding & Paragliding Association N1 engine fan or LP compressor speed
CAA Civil Aviation Authority NDB Non-Directional radio Beacon
(Investigation of Air Accidents and Incidents) Regulations 2018. CAVOK Ceiling And Visibility OK (for VFR flight) nm nautical mile(s)
CAS calibrated airspeed NOTAM Notice to Airmen
cc cubic centimetres OAT Outside Air Temperature
The sole objective of the investigation of an accident or incident under these CG Centre of Gravity OPC Operator Proficiency Check
Regulations is the prevention of future accidents and incidents. It is not the cm centimetre(s) PAPI Precision Approach Path Indicator
CPL Commercial Pilot’s Licence PF Pilot Flying
purpose of such an investigation to apportion blame or liability. °C,F,M,T Celsius, Fahrenheit, magnetic, true PIC Pilot in Command
CVR Cockpit Voice Recorder PM Pilot Monitoring
DME Distance Measuring Equipment POH Pilot’s Operating Handbook
Accordingly, it is inappropriate that AAIB reports should be used to assign fault EAS equivalent airspeed PPL Private Pilot’s Licence
or blame or determine liability, since neither the investigation nor the reporting EASA European Union Aviation Safety Agency psi pounds per square inch
ECAM Electronic Centralised Aircraft Monitoring QFE altimeter pressure setting to indicate height above
process has been undertaken for that purpose. EGPWS Enhanced GPWS aerodrome
EGT Exhaust Gas Temperature QNH altimeter pressure setting to indicate elevation amsl
EICAS Engine Indication and Crew Alerting System RA Resolution Advisory
EPR Engine Pressure Ratio RFFS Rescue and Fire Fighting Service
ETA Estimated Time of Arrival rpm revolutions per minute
ETD Estimated Time of Departure RTF radiotelephony
FAA Federal Aviation Administration (USA) RVR Runway Visual Range
FDR Flight Data Recorder SAR Search and Rescue
FIR Flight Information Region SB Service Bulletin
FL Flight Level SSR Secondary Surveillance Radar
ft feet TA Traffic Advisory
AAIB Bulletins and Reports are available on the Internet ft/min feet per minute TAF Terminal Aerodrome Forecast
http://www.aaib.gov.uk g acceleration due to Earth’s gravity TAS true airspeed
GNSS Global Navigation Satellite System TAWS Terrain Awareness and Warning System
GPS Global Positioning System TCAS Traffic Collision Avoidance System
GPWS Ground Proximity Warning System TODA Takeoff Distance Available
hrs hours (clock time as in 1200 hrs) UA Unmanned Aircraft
HP high pressure UAS Unmanned Aircraft System
hPa hectopascal (equivalent unit to mb) USG US gallons
IAS indicated airspeed UTC Co-ordinated Universal Time (GMT)
This bulletin contains facts which have been determined up to the time of compilation. IFR Instrument Flight Rules V Volt(s)
ILS Instrument Landing System V1 Takeoff decision speed
Extracts may be published without specific permission providing that the source is duly acknowledged, the material is
IMC Instrument Meteorological Conditions V2 Takeoff safety speed
reproduced accurately and it is not used in a derogatory manner or in a misleading context.
IP Intermediate Pressure VR Rotation speed
IR Instrument Rating VREF Reference airspeed (approach)
Published 8 June 2023 Cover picture courtesy of Marcus Cook
ISA International Standard Atmosphere VNE Never Exceed airspeed
© Crown copyright 2023 ISSN 0309-4278 kg kilogram(s) VASI Visual Approach Slope Indicator
KCAS knots calibrated airspeed VFR Visual Flight Rules
Published by the Air Accidents Investigation Branch, Department for Transport KIAS knots indicated airspeed VHF Very High Frequency
Printed in the UK on paper containing at least 75% recycled fibre KTAS knots true airspeed VMC Visual Meteorological Conditions
km kilometre(s) VOR VHF Omnidirectional radio Range
AAIB Bulletin: 6/2023

CONTENTS

SPECIAL BULLETINS / INTERIM REPORTS

None

SUMMARIES OF AIRCRAFT ACCIDENT (‘FORMAL’) REPORTS

None

AAIB FIELD INVESTIGATIONS

COMMERCIAL AIR TRANSPORT

FIXED WING
None
ROTORCRAFT
None

GENERAL AVIATION
FIXED WING
Aeroprakt A32 Vixxen G-ENVV 19-Jul-22 3
Mudry Cap 10B G-BXBU 12-Aug-21 21
Pitts S-1S G-BOXV 26-Aug-22 67
ROTORCRAFT
None

SPORT AVIATION / BALLOONS

Flight Design CT2K G-CBDJ 24-Mar-22 84
Pegasus Quik G-CGRR 6-Aug-22 102

UNMANNED AIRCRAFT SYSTEMS

None

AAIB CORRESPONDENCE INVESTIGATIONS

COMMERCIAL AIR TRANSPORT

ATR 72-211 G-CLNK 25-Oct-22 109
Boeing 777-300(ER) HL-7782 ∫ 28-Sep-22 113
Boeing 757-256 TF-FIK
∫
Cessna Citation 560XL EC-KPB 1-Jun-22 125

GENERAL AVIATION
None

© Crown copyright 2023 i All times are UTC

AAIB Bulletin: 6/2023

CONTENTS Cont
AAIB CORRESPONDENCE INVESTIGATIONS Cont

SPORT AVIATION / BALLOONS

None

UNMANNED AIRCRAFT SYSTEMS

DJI Mavic 2 Enterprise n/a 7-Aug-22 130
DJI Mavic 2 Enterprise Zoom n/a 5-Dec-22 136

RECORD-ONLY INVESTIGATIONS

Record-Only Investigations reviewed: March / April 2023 143

MISCELLANEOUS

ADDENDA and CORRECTIONS

BB85Z hot air balloon G-ELMR 13-Sep-22 147

List of recent aircraft accident reports issued by the AAIB 149

(ALL TIMES IN THIS BULLETIN ARE UTC)

© Crown copyright 2023 ii All times are UTC

AAIB Bulletin: 6/2023

AAIB Field Investigation Reports

A Field Investigation is an independent investigation in which
AAIB investigators collect, record and analyse evidence.

The process may include, attending the scene of the accident

or serious incident; interviewing witnesses;
reviewing documents, procedures and practices;
examining aircraft wreckage or components;
and analysing recorded data.

The investigation, which can take a number of months to complete,

will conclude with a published report.

© Crown copyright 2023 1 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

ACCIDENT

Aircraft Type and Registration: Aeroprakt A32 Vixxen, G-ENVV

No & Type of Engines: 1 Rotax 912ULS piston engine

Year of Manufacture: 2019 (Serial no: LAA 411-15611)

Date & Time (UTC): 19 July 2022 at 1920 hrs

Location: Newtownards Airfield, County Down

Type of Flight: Private

Persons on Board: Crew - 2 Passengers - None

Injuries: Crew - 2 (Fatal) Passengers - N/A

Nature of Damage: Extensive

Commander’s Licence: See pilot information section

Commander’s Age: See pilot information section

Commander’s Flying Experience: See pilot information section

Information Source: AAIB Field Investigation

Synopsis

On the evening of 19 July 2022, two pilots were flying circuits around Newtownards Airport
in G-ENVV an Aeroprakt Vixxen. After approximately 20 mins of circuits they flew a low
pass parallel to Runway 03 followed by a steep right turn passing over several people on
the ground. Recorded data showed the aircraft passed over the people with 70° angle of
bank at 72 ft above the ground. During this turn the aircraft was seen to descend and hit
the ground.

The investigation could not determine exactly why the aircraft descended in the turn but no
defects could be found with the aircraft or engine. There was evidence that the aircraft’s
electronic displays lost power before the accident and this could have caused a distraction.
However, it was being flown in a manner that exposed the aircraft, the occupants and the
people on the ground to a high risk of an accident.

The investigation identified several shortcomings in the build process and the registration of
the ballistic parachute recovery system, which did not contribute to the outcome. The LAA
and CAA have taken action to address these.

During an inspection carried out immediately before the accident the CAA identified
shortcomings in the aerodrome’s safety management system, which the CAA stated have
now been addressed.

© Crown copyright 2023 3 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

History of the flight

The accident occurred whilst two pilots (one male and one female) were flying circuits
around Newtownards Airport (near Belfast) in G-ENVV, an Aeroprakt A32 Vixxen. It is not
known who was flying the aircraft; the male pilot was seated in the left seat and the female
pilot was seated in the right. The accident occurred after the airport’s normal operating
hours so there was no air/ground radio service. No other aircraft was flying at or near the
airport at the time of the accident flight.

Shortly before the accident flight the two pilots had been flying together in a Van’s RV-8A
aircraft. They landed from this flight at 1840 hrs. Prior to this flight the male pilot had flown
the RV-8A with a different pilot. Another pilot had flown G-ENVV just before the accident
flight. He reported that the aircraft had no problems and was flying well.

Group of Witnesses

Runway 26

Accident Location
Runway 03

Runway 33

Figure 1
Newtownards Airport showing the accident and witness locations

Several people, who were at the airport preparing for an open day due to take place the
following weekend, saw parts of the accident flight. When the accident occurred, they were
stood by a vintage bus on the main apron (Figure 1). Many other people in the local area
witnessed parts of the flight.

Data recorded on the aircraft’s electronic displays showed the aircraft started to taxi from
the apron on the north-west side of the airport at 1859 hrs and took off from Runway 33 at
1901 hrs. Witnesses saw the aircraft taxi out, complete normal power checks then takeoff.
The aircraft flew a left-hand circuit to a touch-and-go back on Runway 33 (Figure 2 – Point 1).
This was followed by another left-hand circuit to a full stop landing on Runway 26 (Figure 2
– Point 2). The aircraft then backtracked the runway and took off again from Runway 33.

© Crown copyright 2023 4 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

Several witnesses reported the aircraft was flying lower and closer to the aerodrome than
they typically saw aircraft flying. It was reported that the aircraft was flying inside Scrabo Hill
(Figure 3) during each circuit. Several people reported that the engine sounded as though
it was struggling during these circuits. After taking off again, the aircraft made a spiral climb
to 1,600 ft above the airport. It then made a spiral descent, completing several orbits to
land on Runway 33 (Figure 2 – Point 3). At 1907 hrs one of the people at the airfield, who
had a handheld radio, heard the female pilot make a comment about the wind over the
radio which the witness thought was referring to the approach and landing the aircraft had
just completed. The aircraft took off again and completed another left-hand circuit back to
Runway 33 (Figure 2 – Point 4). The recorded data showed the aircraft’s angle of bank
was around 30° during most of the turns up to this point in the flight, although it reached
47° during one descending turn.

The aircraft then took off again and flew an ’S’ turn to approach Runway 03 (Figure 2 –
Point 5). During this turn the angle of bank reached 59° with the aircraft at 370 ft agl. The
aircraft then flew a low pass parallel to Runway 03 at approximately 70 ft above the ground.
At the end of the runway the aircraft made a right turn passing over the people stood next to
the vintage bus. Witnesses estimated the aircraft’s bank angle was over 60°. One witness
described the aircraft “travelling more in the direction of its belly rather than forward”. It
was reported that the engine sounded “normal” and was at “full power”. The last data point,
recorded as the aircraft turned right, showed the aircraft was at 72 ft agl, with 71° angle of
bank, travelling at an indicated airspeed of 86 kt and a normal acceleration1 of 1.91 g.

Figure 2
Flight path of accident flight
Footnote
1
‘Normal Acceleration’ is the head to foot acceleration experienced by the pilot.

© Crown copyright 2023 5 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

Seconds after the aircraft passed over the bus it appeared to be out of control and descended
rapidly to the ground. Witnesses reported it happened very quickly and accounts of exactly
what happened varied. However, witnesses described the aircraft “spiralling”, “turning wing
over wing” and “tumbling”. One witness reported that the aircraft “made a steep bank to
its right before it pitched up. The aircraft partially turned over on itself, continued along the
trajectory of its initial banking turn but it was losing altitude”. Another witness described
seeing the aircraft yaw markedly into the turn as it lost control. No one reported any
abnormal noises from the aircraft or its engine. Witnesses also confirmed that the aircraft
did not collide with any other object.

The aircraft struck the ground on the airfield boundary just to the south of the flying club
and airport cafe buildings. Witnesses ran to the accident site and airport staff collected
the airport fire vehicle. They reached the aircraft within a few seconds but there was no
sign of life from the occupants. A fire started on the right side of the fuselage but this was
extinguished by the airport fire crew and the local fire service, who arrived shortly after the
accident.

Accident site

Scrabo Hill

Runway 03

Accident Site

Clubhouse

Figure 3
Accident site location

G-ENVV struck a wooden fence on the boundary of the airfield close to the clubhouse and
its outside eating area, continuing through the fence before coming to rest, inverted, over
a gorse bush (Figure 3). There were no apparent ground marks from impact prior to the
fence line.

© Crown copyright 2023 6 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

Right wingtip Direction

of travel
Nosewheel

Right wing

Tail section

Fence
Engine

Rudder

Left wing

Figure 4
Accident site

Witness marks matching G-ENVV’s paint were found on the fence posts. The right wingtip,
some glazing from the windows, and the nosewheel were detached at the point of impact
with the fence (Figure 4). The rudder became detached at the point of impact with the gorse
bush, and was undamaged by the post-impact fire. The remainder of the right wing, left
wing, tail section, fuselage, engine and propeller were all located at the final accident site.

Recorded information

The aircraft was fitted with two EFIS avionic units that display and independently record
flight, position, engine, and fuel data at a sampling rate of 1 Hz. The recorded data ends
about nine seconds before the accident, based on the last recorded speed and estimated
ground track (Figure 2).

Figure 5 plots some of the data from both EFIS units, starting from when the aircraft was
flying parallel to Runway 03 at about 72 ft (22 m) above the ground, and ending when the
recordings stopped. The close alignment of data between the two units, seen in the figure,
indicates that each unit was sampling data to within a fraction of a second of the other. The
figure also shows that the recording from the left EFIS stopped six seconds before the end
of the recording from the right EFIS. The manufacturer of the units stated that the data is
buffered for six seconds before being stored in memory, implying that the left EFIS stopped
working up to six seconds before the last data point recorded by the right EFIS at time
19:19:32. (The buffer is volatile memory, so its contents are lost when the unit’s power is
turned off or disrupted.)

© Crown copyright 2023 7 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

Figure 5
Last 13 seconds of recorded data

This absence of data indicates there may have been a loss of electrical power to the EFIS
units before impact.

The last few seconds of the recorded data are highlighted by the shaded area in Figure 5
and correspond to when the aircraft was in the right turn with a bank angle greater than 50°.
During this time, the engine speed increased by about 500 rpm [1]; the aircraft’s airspeed

© Crown copyright 2023 8 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

reduced [2]; the pitch attitude reduced from 10° to about 7° [3]; and the normal acceleration
was less than that required to maintain level flight at the bank angles flown [4]. No change
in altitude was recorded; however, the resolution of the altitude data was 5 m (about 16 ft)
so any changes less than this during these few seconds would not be seen in the data.

Aerodrome information

Newtownards Airport is a licensed aerodrome with three asphalt runways. Residential

areas are located close to the airport to the west, north and east. Strangford Lough is to
the south of the airport.

The airport’s operating hours are 0900 hrs to 1700 hrs, during which an air/ground radio
service is provided. Aircraft can fly outside the operating hours by prior arrangement.

The published circuit height is 1,000 ft aal.

Aircraft information

The Aeroprakt A32 Vixxen is a high-wing, two seat light aircraft fitted with a Rotax 912
ULS engine and Kiev 3-bladed ground adjustable propeller. It has an all-moving tailplane
and flaperon flying controls. The construction is largely metal, with a mixture of metal and
fabric‑covered surfaces.

Figure 6
G-ENVV at Newtownards (image used with permission)

Build history

The A32 is supplied as a fast-build kit from the manufacturer to the UK importer, within the
51% owner-builder amateur requirements for a LAA Permit to Fly2. The fast-build kit is
Footnote
2
At least 51% of the physical aircraft build must be completed by the amateur builder.
British Civil Airworthiness Requirements (BCAR) Chapter A3-7 ‘Permit to fly Aircraft - Initial and Continuing
Airworthiness’.
Civil Aviation Publication (CAP) 659 ‘Amateur Built Aircraft’.
Light Aircraft Association LAA Technical Leaflet (TL) 1.02, section 18 ‘Amateur Building Rules’.

© Crown copyright 2023 9 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

close to 49% complete in its supplied state, which precludes significant further work being
completed by anyone other than the amateur owner-builder. G-ENVV was issued with a
Permit to Fly in October 2019.

The aircraft arrived at Newtownards by road, requiring final assembly and avionic systems
installation. Work, including fabric covering and external paint, had been done prior to its
arrival. There was evidence of remunerated subcontracted work towards build completion
having occurred at Newtownards. The address listed for build was not where the aircraft
was finally assembled3. There were three interested parties in the aircraft’s ownership at
the point of build, of which one (the male accident pilot) was registered as the owner for
the initial build. The investigation did not find evidence to substantiate that the owner was
physically involved in the build of G-ENVV to qualify the build within the 51% owner-builder
amateur requirements.

Maintenance history
The aircraft’s airframe and engine logbooks were retrieved from the aircraft but neither
had been updated since the last Permit to Fly revalidation inspection in October 2021, at
451 flying hours.

The A32 has a manufacturer’s maintenance schedule published within its Aircraft
Maintenance Manual, detailing the required maintenance at 50, 100 and 200-hour
intervals. The LAA provides generic maintenance schedules for use when a manufacturer’s
schedule is either not available or is not mandated by the aircraft’s Permit to Fly Operating
Limitations. The LAA stated that the manufacturer’s maintenance schedule was not
mandated for G-ENVV, and the ‘owner’s tailored maintenance schedule’ was declared
within the Certificate of Clearance form. It was stated by those carrying out maintenance
on G-ENVV that a generic LAA schedule was followed. There were no workcards or
maintenance documents to support airframe or engine logbook entries made before
October 2021. There were no logbook entries or maintenance history to verify the status
of the airframe or engine after October 2021.

Data subsequently retrieved from on board flight display systems logged 587 engine hours
at the time of the accident.

Aircraft examination

Structure
The fuselage structure surrounding the passenger compartment was destroyed in the
post‑impact fire.

The aft section of the tail was mainly intact, had separated from the fuselage and suffered
fire damage. There was some fire damage to the left all-moving tailplane but both left and
Footnote
3
Light Aircraft Association LAA Technical Leaflet (TL) 1.02, section 4 ‘Workshop and Storage Facilities’
specifies that the LAA inspector will check the suitability of where the aircraft will be built. Section 7
‘Frequency of Inspections’ details the in-person main inspection stages required for the project. Both require
knowledge of where the aircraft is to be built.

© Crown copyright 2023 10 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

right tailplanes were present and attached. The rudder had detached at final impact and
was largely undamaged. The rudder at its uppermost attachment point to the tail structure
showed evidence of having travelled beyond full deflection to both left and right sides before
detaching from its base plate.

The right wing exhibited impact damage to the leading edge and internal structure
corresponding to contact with the boundary fence. The right flaperon was partially attached
to the wing, but damaged from impact with the gorse bush. The left wing had suffered
significant disruption at its point of attachment with the fuselage and had become detached
during impact with the gorse bush. The left flaperon was still attached. Both wings had
considerable fire damage in the region of the internal fuel tanks.

The nosewheel had detached at the joint with the shock absorber, also corresponding to
impact with the fence.

Flying controls
Flying control surfaces were activated via cable and pulley systems, from a central control
stick located between the seats. Continuity was established between the rudder pedals to
the rudder base plate, and control stick to the tailplane and left flaperon, with the cables free
from restriction and intact after impact. The right flaperon cable was found to have broken
in overload in the area of the wing to fuselage attachment point, consistent with damage
sustained in this area during impact.

The flap lever was found in the ‘up’ position. The lever is held in place with a locking pin,
which is released by pulling the lever laterally to free the pin from its locating hole. There
are three locating holes for each position of flap. Due to the force required to move the
lever, it is likely that the flaps were ‘up’ during flight immediately prior to impact.

The trim control was located between the seats at floor level. Control cable continuity and
free movement from the control lever to the trim tab was established.

Avionics
The aircraft was equipped with flat-screen primary flight display systems. They were
an additional item to the kit purchased by the owner and were installed during the initial
build, except for an Attitude Heading and Reference System (AHRS) which was a later
modification. The AHRS can display information including an artificial horizon, pitch and roll
attitude, airspeed, and altitude. The flat-screen display system is customisable and capable
of displaying comprehensive flight, engine, and navigation information, and is coupled to a
data acquisition unit. Flight logs and aircraft parameters from those flights are stored within
the system, and data from the accident flight was retrieved.

Two battery back-up supply units were installed; one for the flat-screen display system and
one for the AHRS unit. The purpose of the battery back-up is to give up to 40 mins usage
of the instrumentation if aircraft main power fails. The power supplies for the flat-screen
display system were controlled by two toggle switches added to an existing row of switches
on the centre console. Both switches were required to be on for the screens to function.

© Crown copyright 2023 11 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

The AHRS power was controlled from a single toggle switch at the top centre of the same
console. None of the switches were labelled as to their function.

At the accident site, both flat-screen display system switches were found in the on position
and the AHRS switch was found in the off position, however, due to the disruption to the
cockpit area that occurred upon impact, these switch positions cannot be confirmed to have
been the same during flight.

An examination of the electrical system was conducted, but it was too badly damaged by
the post-impact fire to determine system integrity or continuity.

Fuel

G-ENVV was fitted with optional larger fuel tanks, with total capacity of 114 l, 112 of which
is usable. The A32 can be used with fuel meeting three different specifications4. It was
not possible to determine the type of fuel used during the accident flight although witness
accounts refer to the pilot purchasing UL91 fuel in drums that were stored in the aircraft’s
hangar.

It could not be determined how much fuel was on board at the start of the accident flight.
The pilot who flew the aircraft on the previous flight estimated there was 20 litres remaining
after his flight. However, it was not possible to determine if the aircraft had been refuelled
immediately prior to the accident flight. It was reported that the aircraft typically consumed
15-20 l/hr. The length of pipework between the fuel tanks and engine could contain enough
fuel for approximately 40 seconds of flight once the usable fuel had been consumed. Fuel
quantity is displayed on two analogue dials in the central console, each with its own low fuel
warning light. The electronic flight displays did not record fuel quantity.

Fire damage concentrated in the locations of the fuel tanks and fuselage indicated that the
system contained fuel at impact. The A32 has two fuel supply handles, one for each wing
tank. One of the fuel supply handles was retrieved and was in the ‘open’ position, but its
position during flight could not be verified.

There was no fuel remaining for sampling or quantity analysis.

Engine

The Rotax 912 ULS had suffered significant fire damage, and it was not possible to
determine if fuel remained within the carburettors or fuel manifold although recorded engine
data showed it running at approximately 5,100 rpm immediately prior to the accident. The
throttle levers remained connected to the carburettors after the accident.

The spark plug electrode gaps and colouration were within manufacturer limits. Two plugs
were lightly coated with oil, but the engine was found inverted with an accumulation of oil

Footnote
4
Motor gasoline (Mogas), Aviation gasoline (Avgas 100LL), or a fuel that meets a minimum octane of 95 and
has an Anti-Knock Index of 91. The latter specification includes fuel type UL91.

© Crown copyright 2023 12 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

within this cylinder, which probably occurred post-impact. Borescope examination of the
cylinders did not show any areas of excessive wear or build-up of combustion products for
the engine’s estimated hours.

Propeller

All three blades were found at the engine impact point. One blade was still attached to
the propeller hub but cracked at its base, the second was partially attached and the third
had completely detached. The blade damage was consistent with the propeller stopping
within a single revolution upon impact. The Rotax 912 ULS engine is fitted with a clutch to
protect the engine from damage caused by a propeller strike. It is possible that the clutch
disconnecting the propeller from the engine, followed by multiple strikes with the fence and
gorse before hitting the ground, could have slowed the propeller’s rotation. Therefore, the
power of the engine at impact could not be clearly determined.

Ballistic parachute recovery system

A ballistic parachute recovery system (BPRS) is a rocket-deployed parachute, used to

recover a whole aircraft including occupants to the ground in an emergency situation.

G-ENVV was fitted with a Magnum 601 S-LSA BPRS located aft of the luggage compartment
behind the seats. The system comprises a parachute packed into a soft case, launched by
a separate rocket canister. The parachute exits the aircraft through a frangible hatch on the
upper fuselage surface. The system is activated by a pull-handle cable located between
the aircraft’s seats. The pull-handle has a safety pin inserted into it to prevent unintended
operation on the ground, which is required by the manufacturer to be removed before flight.
The BPRS had not been activated by either occupant, and the safety pin was found inserted
in the handle. It is not known if the occupants had inadvertently left the pin in place or if
they routinely flew with the activation pin installed. CAA and LAA guidance5,6 specifies
a two‑stage release control for BPRS to avoid inadvertent operation. BPRS installation
approval for the A32 was based on that for the A22 Foxbat, where a 2.5 mm cable tie is
used for the secondary release, not the activation pin. It is possible that the occupants
regarded the activation pin as the secondary release mechanism or were not familiar with
the two‑stage release guidance for this aircraft type.

External warning placards are required7 to be applied to aircraft fitted with a BPRS, to alert
occupants upon entering the aircraft and emergency responders in the case of an accident.
The presence of a BPRS must also be notified to the CAA (in this case via the LAA) for
inclusion in the central aircraft register to provide safety information to those attending in an
emergency. The requirement to notify the CAA at initial aircraft registration was introduced
at the beginning of 2022 in response to an accident where BPRS was fitted but the CAA
Footnote
5
Civil Aviation Authority (CAA) CAP 482 British Civil Airworthiness Requirements (BCAR) Section S – Small
Light Aeroplanes, Sub-Section K, Issue 7. ‘AMC S 2003 (Interpretive Material)’.
6
Light Aircraft Association (LAA) Technical Leaflet (TL) 3.27 Ballistic Parachutes, Issue 1, 27 March 2020
Section 4.3 ‘Miscellaneous Points’.
7
Civil Aviation Authority (CAA) CAP 482 British Civil Airworthiness Requirements (BCAR) Section S – Small
Light Aeroplanes, Sub-Section K, Issue 7. S 2041 ‘Markings and Placards’

© Crown copyright 2023 13 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

was not informed8. There was no evidence of placards on the aircraft and the presence of
a BPRS was not shown in G-ENVV’s CAA database record. The only reference found to a
BPRS fitted to G-ENVV was within the weight and balance record completed at initial permit
application. No entries were found within the aircraft’s build record, modification record or
Permit to Fly application paperwork, nor subsequent annual Permit to Fly revalidations.
LAA members and inspectors associated with G-ENVV’s build and maintenance did not
appear to have a good understanding of the relevant requirements.

Survivability

The manufacturer of the BPRS specifies a minimum deployment altitude for safe rescue of
200 m, with some documented rescues down to 80 m. This allows sufficient time for safe
parachute opening and aircraft stabilisation during which altitude is lost. G-ENVV was flying
at an altitude of approximately 80 ft (24 m) at the point where departure from controlled
flight occurred. If the BPRS had been activated at this moment, there would not have been
enough time for the parachute to deploy and effectively arrest the aircraft’s descent.

G-ENVV was fitted with four-point harnesses at both seats. There was extensive fire
damage to the harness webbing but the buckles from both seats were found fastened. A
lack of heat and smoke damage to the interior of the buckles indicated that they were both
fastened correctly prior to impact.

Weight and balance

The aircraft had two occupants on board, a partial fuel load and no luggage, and would have
been within its maximum takeoff weight of 600 kg and centre of gravity limits.

Aircraft performance

The Pilot’s Operating Handbook (POH) gives the aircraft’s 1g stall speed at maximum
takeoff weight, with flaps up, as 32 KIAS. The POH does not give the stall speed for level
flight with different bank angles. However, based on data supplied by the manufacturer,
the stall speed in level flight at 70° angle of bank would be approximately 68 KIAS9. A load
factor of 2.92 g is required to maintain level flight with 70° angle of bank.

The aircraft is certified as non-aerobatic and its operating handbook specifies a maximum
bank angle of 60°. It has a maximum positive load factor of +4.0g.

Meteorology

Weather reports are not recorded at Newtownards Airport outside aerodrome hours. Belfast
City Airport (7 nm west-north-west) reported, at the time of the accident, a surface wind from
350° at 6 kt varying between 300° and 020°, visibility greater than 10 km, cloud overcast at
3,200 ft, temperature 16°C and a sea level pressure of 1022 hPa.
Footnote
8
AAIB investigation to Silent 2 Electro, G-CIRK, 23 April 2021 https://www.gov.uk/aaib-reports/aaib-
investigation-to-silent-2-electro-g-cirk [Accessed February 2023]
9
The indicated stall speed at this angle of bank is higher than might be estimated from increased load factor
alone, due to indication errors at increased angle of attack.

© Crown copyright 2023 14 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

A pilot who took off from Newtownards approximately 40 minutes before the accident
estimated the wind at the airfield was from the north-west at 8 - 12 kt but he commented
that the wind increased markedly in the climb.

Pilot information

The female pilot was 44 years old and held a Private Pilot’s Licence which was issued in
October 2020. She held a Single Engine Piston rating which was valid until 31 October 2022
and also held a night rating. She had a valid Class 2 medical.

She had been undertaking training for a restricted instrument rating (IR(R)) and it was
reported that she had been studying to take the ATPL ground exams.

The last entry in her logbook was on 21 October 2021 which was an IR(R) training
flight in a Cessna 172. Her last flight as pilot in command in G-ENVV was recorded
on 13 October 2021. The logbook gave a total flight time of 204.6 hours. Flying club
technical log dockets were found relating to 14 further flights between 23 October 2021
and 19 February 2022 totalling 12.1 hours but these were not recorded in her logbook. A
notepad was also found containing notes about three flights on 7, 14 and 20 May 2022.

An instructor commented that she had been a “competent and knowledgeable student” who
flew with “caution and diligence”. However, he had noticed a change in her flying in the
months preceding the accident. After observing several tight approaches with sharp turns
onto final approach he felt compelled to speak to her about the risks involved in manoeuvring
close to the ground. He spoke to her informally, advising her to be more cautious.

The male pilot was 50 years old and held a National Private Pilot’s Licence which was
issued in June 2018. He had held Microlight and Simple Single Engine Aeroplane (SSEA)
ratings but the Microlight rating lapsed on 31 May 2022 and the SSEA rating lapsed on
31 October 2021. There was no evidence that either rating had been renewed. He had a
valid self-declared medical.

His logbook was in the aircraft when the accident occurred and was significantly fire damaged.
A photograph was found of a page of his logbook which showed that in November 2021
he had accumulated 421 flying hours. His logbook contained five further completed pages
(55 flights) but the logbook was too damaged to read the details of these flights.

Several qualified pilots who knew him well commented that “he liked to push the boundaries”
and “he enjoyed the more exciting side of flying”. It was reported that he was “a very capable
pilot” but he enjoyed “flying low approaches and very tight circuits”. A commercial pilot who
had recently flown with him in G-ENVV commented that he felt “uncomfortable” with the
tight circuit the pilot had flown. He had recently purchased a Van’s RV-8A aircraft and had
been flying aerobatic manoeuvres, although he had no formal training or qualification in
aerobatics.

© Crown copyright 2023 15 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

Organisational information

The airfield is operated by a flying club. The flying club incorporates a Declared Training
Organisation (DTO). Flying club members can fly the club aircraft but some members
also hangar their own aircraft at the airfield. The flying club is managed by a committee of
volunteers.

The accident flight occurred in a privately owned aircraft being flown by two qualified pilots10.
It was therefore outside the oversight of the DTO. It also occurred outside the aerodrome’s
published hours, which limited oversight by the flying club.

The Aeronautical Information Publication entry for the airfield states that flying outside the
licensed hours is allowed ‘by arrangement’. The club website refers to an out-of-hours
indemnity form which should be completed to fly outside licensed hours but the investigation
found that no such form existed. The investigation did not find any record of who was
authorised to fly outside hours or any arrangements by which this was managed.

Some committee members reported that there had been previous reports of low flying and
excessively tight circuits being flown by other pilots outside licensed hours. They reported
that during licensed hours, with a duty instructor on duty and with considerable flight training
activity taking place, there was sufficient oversight, but that outside hours there was no
oversight. Several years previously, the committee had tried to report to the CAA another
pilot who was observed low flying, but they had not been able to provide sufficient robust
evidence, so the CAA was unable to take any action. This resulted in the committee feeling
they were unable to tackle future similar issues.

The flying club standard operating procedures set out the requirement for all incidents to
be reported in writing and posted in the reporting box in the flying club. Despite several
previous incidents being mentioned to the AAIB during this investigation, no evidence was
found of any previous incident reports being made to the flying club. The aerodrome manual
describes the club safety management system (SMS). It states that ‘an aerodrome safety
committee meets a minimum of twice per year to review any safety related issues, accident
and incident reports’. The investigation did not find any evidence that these meetings had
ever happened.

On the day of the accident and the previous day the CAA was conducting an oversight audit
at the airport. The audit finished before the accident and made five Level 2 findings11. One
of these related to the SMS and stated:

Footnote
10
The male pilot’s flying licence was not valid at the time of the accident.
11
A Level 2 finding means it been identified that the Aerodrome is not in full compliance with the aerodrome
licensing requirements set out in either the Air Navigation Order, the Aerodrome Licence, ICAO Annex 14,
Civil Aviation Publication (CAP) 168 or the Aerodrome Manual. Rectifying action must be taken within the
agreed timescales.

© Crown copyright 2023 16 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

‘The safety management system was found not to be operating as described in

the aerodrome manual:

1) Current records could not be found for the committee monthly meeting
or the 6 monthly Airport Safety Group meeting,
2) No evidence of internal audits could be found,
3) The SMS should include reference to “just culture”,
4) Current monthly reports could not be found’.

The auditor also observed a runway incursion during the audit and made the following
finding:

‘The aerodrome does not have effective procedures to prevent runway

incursions. An incursion was witnessed during the audit when an individual
riding a motorbike entered the aerodrome and crossed runways 03/21 and
15/33 to access hangars across the airfield but without making radio calls. An
MOR needs to be filed for this event and adequate procedures implemented
(covering access, driving rules and procedures and RTF procedures) to enable
this activity to be completed safely.’

When asked about the safety culture at the flying club several people reported to the AAIB
that it was poor. It was reported that there was a small group of pilots who often operated
outside normal aviation convention, but this had not been addressed by the committee.

Analysis

Accident flight

The accident occurred whilst the aircraft was making a low altitude steep turn, which
occurred above people and buildings.

There was no evidence of any construction anomaly, failure or malfunction in the airframe
or flying controls that could have contributed to the accident. Data recovered from the
aircraft’s avionics suggested the engine was performing normally during the accident flight.
The exact fuel quantity on board could not be determined. However, the previous pilot
reported there was at least 20 litres remaining after his flight and, if that were the case,
there would have been sufficient fuel onboard for the 20 minutes flight. It is possible that the
steep angle of bank could have caused the fuel supply to the engine to be interrupted but
the fuel pipework was of sufficient capacity that had it contained fuel it could have sustained
the engine for approximately 40 seconds.

An absence of recorded data from the EFIS units indicates they may have lost power before
impact, although it could not be determined why this occurred as battery back-up systems
were installed. Had such a failure occurred, it is possible that it captured the pilot’s attention
and briefly distracted the pilot from the primary task of flying the aircraft.

© Crown copyright 2023 17 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

The aircraft exceeded its 60° bank angle limit in the last 1.5 seconds of recorded flight. It
was not possible to determine if this was intentional.

It was not possible to determine the exact sequence of events which led to the aircraft
losing height and striking the ground. The last recorded data point showed the aircraft had
a normal acceleration of 1.91 g. If sustained, this would be insufficient to maintain level
flight with a 70° bank. It is possible that the accident was caused by over banking and
descending in the turn with insufficient altitude for the pilot to recover, but witness accounts
are more consistent with a departure from controlled flight. The stall speed with 70° bank
and level flight is approximately 68 KIAS. At the last data point the aircraft was flying at
86 KIAS, but the speed was reducing and the application of nose-up elevator intended to
arrest a descent might cause a stalling angle of attack.

The aircraft was flown at 72 ft above people and buildings. The rules of the air12 state
that the pilot in command should ‘not fly in a manner that would endanger either people or
property’ and should ‘not fly closer than 500 ft to any person, vessel, vehicle or structure
unless necessary for taking off or landing’. It is likely the pilot in command was breaching
both rules.

It is not known which pilot was flying the aircraft when the accident occurred. Both pilots
had previously been seen flying steep turns at low altitude. The male pilot had recently
purchased a Van’s RV-8A aircraft and it was reported that he had been flying aerobatic
manoeuvres in that aircraft. The investigation did not find any evidence that either pilot had
been trained in aerobatic flight.

Choosing to fly excessive manoeuvres close to the ground increases the opportunity for
error and reduces the room for recovery, placing occupants and the public at unnecessary
risk.

Pilot’s licence

The male pilot did not have a valid flying licence. His Microlight and SSEA ratings had
expired. It is likely that he would have had sufficient flying hours to revalidate by experience
but he had not had his licence signed by an examiner13. Once the ratings had lapsed he
would have needed to complete a proficiency test to renew the ratings.

It is possible that he was not aware that his ratings had lapsed. The CAA does not provide a
reminder service for when ratings and licenses lapse. However, all pilots must ensure their
licence is valid before flying. If unsure about the requirements advice can be sought from
an instructor, examiner, flying club or the CAA.

Footnote
12
Guidance on the Rules of the Air can be found in the Skyway Code (CAP1535) available at The Skyway
Code | Civil Aviation Authority (caa.co.uk) (accessed 22 November 2022).
13
It could not be determined if he had completed a flight with an instructor which is required to revalidate by
experience but the 55 flights in his logbook suggests he would have achieved the 12 hours required.

© Crown copyright 2023 18 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

Aircraft build and maintenance

It was not possible to ascertain the exact build and maintenance status of G-ENVV due
to a lack of detail within the aircraft’s build, maintenance and logbook paperwork. This is
likely, in part, due to the low level of owner involvement during build completion compared
to third party assistance, regarding the 51% amateur builder requirements. It is possible
that as a result the aircraft did not conform to the requirements to qualify for a Permit to Fly,
indicating this aspect of the process for fast-build kits that would benefit from a higher level
of oversight. The LAA is exploring ways to improve oversight of the build process relating
to the 51% rule, looking to include, but not limited to, more frequent and better targeted
auditing of build projects.

There was nothing to indicate to those attending the accident site that G-ENVV was fitted
with a BPRS. CAA registration of G-ENVV occurred in 2019, before the introduction in
2022 of a point in the registration process to verify that BPRS and compliant placarding
was fitted. There was no point within the LAA Permit to Fly issue or annual Permit to
Fly revalidation process to verify that BPRS and relevant compliant placarding had been
fitted, other than by relying on inclusion in the aircraft’s modification record during the build
process. It also appears there was no reliable way to ensure this information was included
in the CAA’s central aircraft register. Permit to Fly issue and annual revalidations presented
three opportunities at which this could have been identified for G-ENVV. To ensure future
visibility of BPRS installations the following safety actions have been taken:

The CAA has amended form CA1 Application for Aircraft Registration or Change
of Ownership, introducing a field to indicate whether an emergency ballistic
device such as BPRS is fitted.

The LAA has amended form CA3 Permit to Fly Application, introducing a field to
indicate whether an emergency ballistic device is fitted, which is flagged to the
CAA Aircraft Registration Team to subsequently update the GINFO database.

Retrospective installations of BPRS must be notified to the LAA by an application

to install BPRS as a new modification to the aircraft type, or notification that
BPRS has been installed as a manufacturer’s standard option. The LAA will
then notify the CAA of the installation.

There was a lack of knowledge regarding regulations for BPRS installation, placarding
and notification to the appropriate authorities amongst LAA inspectors and members. It is
becoming increasingly popular amongst the general aviation community to fit BPRS, which
are offered as an option or retrofit on many microlight and general aviation aircraft. To
improve awareness among LAA inspectors and members, the following safety actions have
been taken:

The LAA has included the requirements for BPRS markings and notification
of installation to the CAA for inclusion on G-INFO register in articles within the
‘Engineering Matters’ section of the member publication Light Aviation.

© Crown copyright 2023 19 All times are UTC

AAIB Bulletin: 6/2023 G-ENVV AAIB-28491

New LAA Inspector induction briefings given by the LAA Chief Inspector now
include a specific topic explaining BPRS installations and associated aircraft
marking requirements.

Organisation

Members of the flying club committee were aware that low flying was taking place outside
the aerodromes licensed hours in the months before the accident, but no action had been
taken to prevent it. It was reported that the committee had not found an effective way to
manage flying outside licensed hours.

The CAA conducted an audit at the airfield just before the accident which found that the
SMS was not functioning as described in the aerodrome manual. During the investigation of
this accident the AAIB found a lack of safety reporting within the club and received several
reports of a poor safety culture. The CAA audit made several findings which the CAA stated
the aerodrome operator has now addressed.

Conclusion

The accident occurred when the aircraft was flown at low altitude and a high angle of bank
over people and buildings. There was insufficient evidence to determine the precise cause.
No defects were found with the aircraft or engine which could have contributed to the
accident. There was evidence that the electronic displays in the aircraft lost power before
impact and it is possible that this caused a distraction. However, the aircraft was being
flown in a manner which exposed it, the occupants and the people on the ground to a high
risk of an accident.

The investigation identified several shortcomings in the build process and the registration of
the BPRS system. The LAA and CAA have taken action intended to address the registration
of BPRS systems. The LAA is exploring ways to improve oversight of the build process
within the 51% amateur building rules.

Published: 18 May 2023.

© Crown copyright 2023 20 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

ACCIDENT

Aircraft Type and Registration: Mudry Cap 10B, G-BXBU

No & Type of Engines: 1 Lycoming AEIO-360-B2F piston engine

Year of Manufacture: 1980 (Serial no: 103)

Date & Time (UTC): 12 August 2021 at 0920 hrs

Location: Lower Colley Farm, Buckland St Mary,

Somerset
Type of Flight: Private

Persons on Board: Crew - 1 Passengers - 1

Injuries: Crew - 1 (Fatal) Passengers - 1 (Fatal)

Nature of Damage: Aircraft destroyed

Commander’s Licence: Private Pilot’s Licence

Commander’s Age: 69 years

Commander’s Flying Experience: 1411 hours (of which 648 were on type)
Last 90 days - 15 hours
Last 28 days - 4 hours
Information Source: AAIB Field Investigation

Synopsis

The pilot found himself stuck above cloud during a cross-country flight under Visual Flight
Rules. After contacting the Distress & Diversion Cell for assistance he was transferred to
the radar frequency of a nearby airport, at which the cloud base was below the minimum
required for the approach offered. The pilot, who was not qualified to fly in cloud, lost control
of the aircraft during the subsequent descent and the aircraft was destroyed when it hit a
tree. Both occupants were fatally injured.

The investigation found that air traffic service providers did not obtain or exchange sufficient
information about the aircraft and its pilot to enable adequate assistance to be provided.
There was an absence of active decision making by those providers, and uncertainty
between units about their respective roles and responsibilities.

Seven Safety Recommendations are made to address shortcomings identified in the

provision of air traffic services in an emergency.

History of the flight

G-BXBU departed Watchford Farm in Somerset, which was the aircraft’s home base, at
0704 hrs on 12 August 2021 with the pilot and one passenger on board. Their intention was
to fly to St Mary’s on the Isles of Scilly for a day trip before returning to Watchford Farm later
that afternoon. At the time of departure, the local weather was described by witnesses as
clear skies with good visibility.

© Crown copyright 2023 21 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

After departure, the aircraft flew south-westerly as planned towards Cornwall. As the
aircraft passed north of Culdrose, it began a descent to 1,000 ft over the sea before turning
right to head east away from the planned destination. It continued in a north-easterly
direction, passing to the north of Torquay then out over Lyme Bay. While over the sea, the
aircraft reached a minimum of 320 ft momentarily before completing three 180º turns and
two 360º orbits. It then began to fly north from Lyme Regis toward Watchford Farm climbing
to a peak altitude of 8,200 ft amsl (Figure 1).

Figure 1
Aircraft planned route and flight path

At approximately 0905 hrs the pilot called Dunkeswell Radio, using the words “PAN, PAN,
PAN” (indicating urgency), asking about the weather conditions at the airfield and stating
that he was unable to land at Watchford Farm because he was stuck above cloud. The A/G
operator at Dunkeswell replied that the weather at the airfield was poor – the cloud base
was ‘on the deck’ and the visibility was 400 m. He suggested the pilot contact Exeter Radar
or the Distress and Diversion (D&D) Cell on the emergency frequency 121.5 MHz.

© Crown copyright 2023 22 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Figure 2
Flight path during D&D communications
(Note that this is not a comprehensive listing of the RT)

© Crown copyright 2023 23 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The pilot made another PAN call on 121.5 MHz at 0911 hrs, stating he was in “real
trouble” as he was stuck above thick cloud and he didn’t know what to do (Appendix 2).
He finished his radio transmission by stating “i need to divert to somewhere er close
to me where i can land”. Several witnesses in other aircraft who heard the call described
the pilot sounding anxious and stressed. The PAN call was initially acknowledged by two
commercial aircraft in the vicinity prior to a response from the D&D controller who stated “...
your pan is acknowledged...”. In response, the pilot stated his altitude was 7,500 ft and
that he had a fuel endurance of 1.5 hours. The transponder on G-BXBU had not been used
throughout the flight, but at the request of the D&D controller the pilot switched it on and set
the emergency squawk of 7700.

At the time G-BXBU declared an emergency, there was a military jet holding in the vicinity of
Exeter Airport where the jet had departed 28 minutes earlier. The aircraft had experienced
a technical fault after takeoff which was subsequently resolved. The aircraft was holding
to burn fuel and reduce its landing weight, prior to returning to land at Exeter. The military
jet did not declare an emergency at any point. G-BXBU had been seen by controllers on
Exeter’s primary radar but there was no altitude information displayed as the transponder
was not switched on. Exeter ATC were concerned about a potential conflict with the military
jet, which was holding between 3,000 and 4,000 ft. The military jet was not moved clear
of G-BXBU’s primary return despite the lack of altitude information. However, altitude data
recovered from the tablet from G-BXBU showed that the aircraft were sufficiently far apart
to discount proximity as a factor in this investigation.

While G-BXBU’s initial contact with the D&D controller was ongoing, a phone call between
the Exeter Radar assistant and the D&D support controller took place between 0912 hrs
and 0914 hrs (full transcript):

09:12:00 D&D support1: “d and d support”

09:12:01 Exeter assistant: “hello it’s exeter”

09:12:02 D&D support: “yep”

09:12:03 Exeter assistant: “hi, has anyone updated you firstly about the
[military jet]?”

09:12:07 D&D support: “er, no”

09:12:08 Exeter assistant: “ok er just to let you know that [military
jet callsign] is still intending to land at exeter, he’s got a
normal undercarriage indication now”

09:12:16 D&D support: “ok”

09:12:17 Exeter assistant: “and also, has a light aircraft called you in
the dunkeswell area?”
Footnote
1
The ‘D&D support’ controller in this event provided support to D&D controllers equivalent to that provided by
an air traffic control assistant (ATCA) to a civil controller when interacting with civil ATSU’s. The D&D support
controller was not permitted to conduct a radar handover.

© Crown copyright 2023 24 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

09:12:21 D&D support: “yes, we are currently dealing with that

situation”

09:12:23 Exeter assistant: “excellent, he’s right in the way of er of

[military jet callsign] would you, what’s he, what’s his
intentions and his level?”

09:12:29 D&D support: “er, don’t know his level but he is currently
above cloud and wanting to divert to the nearest aerodrome”

09:12:37 Exeter assistant: “well that would be us”

09:12:38 D&D support: “er..”

09:12:40 Exeter assistant: “exeter”

09:12:41 D&D support: “i think…”

09:12:44 Exeter assistant: “he’s basically flown all the way up the coast
and then across our extended centreline twice in front of
er, a [military jet]”

09:12:49 D&D support: “yes”

09:12:50 Exeter assistant: “er.. do you want to put him over to us?”

09:12:54 D&D support talking to D&D controller offline: “exeter are asking
if er, maybe we want to, she asked to put it over to them?.......
they are wondering if er… they want to take over.”

09:13:14 D&D support: “standby, we are just talking to the aircraft”

09:13:15 Exeter assistant: “oh ok, alright” (offline): “he’s working d and d
that aircraft”

09:13:21 D&D support (offline): “exeter are willing to take the aircraft”

09:13:22 Exeter assistant: [unintelligible]

09:13:48 D&D support: “exe.. er we are putting him on an emergency

squawk, is there a frequency that we can put him on to?”

09:13:52 Exeter assistant: “er... one.. one.. hang on”

09:13:55 Exeter assistant (offline): “which one of you wants to work this
aircraft inbound, do you want to [name] or shall [name] take
it? the inbound. for weather. the one that’s been in the way
for the last ten minutes. yeah. yeah?”

09:14:10 Exeter assistant: “yeah, if you put it through one two three
five eight zero”

09:14:14 D&D support: “one two three five eight zero”

09:14:17 Exeter assistant: “and what’s his callsign?”

© Crown copyright 2023 25 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

09:14:18 D&D support: “er, callsign is…”

09:14:21 Exeter assistant: “oh we’ve got it, we’ve got it it’s ok.”

09:14:23 D&D support: “you’ve got...”

09:14:25 Exeter assistant: “yep, alright then.”

09:14:26 D&D support: “alright then, bye.”

09:14:27 Exeter assistant: “thanks cheers, bye.”

The D&D controller understood the D&D support controller to mean that the Exeter Air
Traffic Service Unit (ATSU) had assessed that aerodrome as suitable for a diversion by the
pilot of G-BXBU. He believed Exeter heard the ‘PAN PAN’ call on 121.5 MHz and that the
reason for Exeter’s phone call was solely to offer help to G-BXBU. He was not aware of
their concern of a potential conflict with the military jet, nor that the D&D support controller
was speaking to an assistant. When the D&D support controller told the D&D controller that
Exeter was willing to take G-BXBU, the D&D controller advised the pilot of this one second
later (Appendix 2). The location of the D&D controller and the D&D support controller
was such that the controller could not overhear the conversation with the Exeter assistant
directly. The Exeter assistant did not identify herself as such during the phone call, contrary
to operating procedures2.

The D&D controller informed the pilot of G-BXBU that his aircraft was identified on radar
and operating under a deconfliction service.

CAP 1434 states that a deconfliction service is,

‘only available to IFR flights in Class G airspace. An ATCO will use radar to
provide you with detailed traffic information on specific conflicting aircraft and
advice on how to avoid that aircraft. However, the pilot retains responsibility for
collision avoidance; you can opt not to follow the ATCO’s advice3.’

Although the emergency squawk of 7700 was visible on the radar controller’s screen, G-BXBU
was transferred to Exeter before anyone with controlling authority at that aerodrome had
been made aware the aircraft was diverting in an emergency. There was no formal radar
handover4 from the D&D controller and the suitability of Exeter, in particular the weather
conditions at the airfield, were not discussed at any point by either the Exeter assistant,
D&D support controller or the D&D controller.

Footnote
2
Manual of Air Traffic Services Part 2 (Exeter Airport) states ‘When ATCOs [controllers] use the mediator or
direct lines they shall identify themselves as “Exeter Radar” or “Exeter Tower”, ATCAs [assistants] add the
suffix “Assistant”.
3
Civil Aviation Publication (CAP) 1434 - ‘UK Flight Information Services’.
4
A radar handover is designed to ensure the safe transfer of responsibility of aircraft between ATSU. RA 3233
contains the details required to be included in a radar handover from controller to controller. https://assets.
publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/974521/RA3233_
Issue_3.pdf [accessed April 2023].

© Crown copyright 2023 26 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The Radar South controller at Exeter, who had come on duty at 0900 hrs but was not yet on
frequency, agreed to accept G-BXBU on a separate frequency. At the time, Exeter Radar
North was active and had only the military jet on frequency. There was no discussion as to
the nature of the diversion in the context of the weather conditions at the airport, nor was
consideration of the pilot or aircraft capability expressed.

Figure 3
Flight path and flight level / altitude radar during Exeter communications
(Note that this is not a comprehensive listing of the RT)

When the pilot of G-BXBU made initial contact with the Exeter controller he confirmed his
emergency ‘PAN’ status and stated, “have been diverted”. The controller initially advised
she would give him vectors for an ILS approach for Runway 26 at Exeter. The pilot asked
her to repeat her transmission and in her response the controller advised she would give
vectors for a Surveillance Radar Approach (SRA), instructing him to fly a radar heading
of 220°. The controller recalled that this change in clearance was prompted by input from a
colleague who was in the room and witnessed her communications with G-BXBU. The pilot
asked her to confirm the cloud base at Exeter, to which she replied the visibility was 6 km
and the cloud was broken at 500 ft. The controller commented that she was surprised to be
asked about the weather conditions at this point, as she would have expected the pilot to
have this information before diverting.

At 0914 hrs the controller observed the aircraft descending and not maintaining the assigned
heading. At 0916 hrs the radar track showed the aircraft levelling briefly around 4,000 ft.
Without having noticed this, the controller instructed G-BXBU to descend to 2,600 ft,
which was the minimum safe altitude5, aiming to prevent the aircraft descending below
that.

Footnote
5
The minimum safe altitude in this sector was 2,600 ft.

© Crown copyright 2023 27 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The last radio transmission from the pilot was “descending two thousand six hundred,
you want me on two three zero?”. The last radar return was at 0917 hrs and showed
the aircraft at 2,700 ft.

Several ear witnesses nearby described a loud engine noise prior to an impact.

At 0920 hrs, Devon and Cornwall Police received a report of an aircraft accident. First
responders found that both occupants had been fatally injured.

Accident site

The accident site was a field approximately 1.2 km north-west of Buckland St Mary in
Somerset. The aircraft struck the boughs of an oak tree and then the ground in the northern
end of the field (Figure 4). Around the tree and from the ground impact there were large
amounts of debris scattered on a southerly path. Running east-west midway across the
field was a concrete single track road bounded on both sides by a single strand, wire fence
supported on wooden posts. The wire had been broken and a piece of wire was caught in
the tail wheel. The engine, cockpit instrument panels and rear fuselage were approximately
40 m to the south of the roadway and had been arrested by the wire fence. The left landing
gear wheel was found in a sunken stream at the southern end of the field, approximately
235 m from the tree.

Figure 4
Ground impact marks

The oak tree at the northern end of the accident site was approximately 20 – 25 m tall with a
large swathe cut through it at about 15 m from ground level. Several large boughs had been
broken and some pieces of wreckage were lodged in the tree. The ground to the south
and east contained broken branches and further wreckage, including wing and fuselage
structure, shards of clear plastic from the canopy, and fragments of the propeller (Figure 5).

© Crown copyright 2023 28 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Figure 5
Accident site looking from the north showing wreckage distribution and boundary

To the south of the concrete roadway was the engine and rear fuselage (Figure 6). The
engine was attached to the cockpit instrument panels and rear fuselage by flight control and
electrical cables. The primary fuel tank had ruptured and was empty whereas the auxiliary
fuel tank was intact and still contained a small quantity of fuel.

Figure 6
Engine and rear fuselage

Inspection of the seat belts revealed they were still done up and that all the structural
attachments had either failed in overload or become detached from the structure so that
they were no longer capable of restraining the occupants. The right landing gear leg was
found close to the rear fuselage section and was complete including a small section of wing
spar (Figure 7).

© Crown copyright 2023 29 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Figure 7
Main landing gear wheels

Recorded information

The aircraft’s avionics did not have any recording capability.

A mobile phone was recovered from the accident site by the police and passed to the AAIB.
The contents indicated that it was not used to check for weather and not used in flight.
There was a change of one of its internal settings recorded at 0917:57 hrs, likely associated
with the time of the accident.

A heavily damaged tablet device was recovered from the aircraft. The main logic board
had become twisted with a part of it ripped off. Some of the integrated circuits had been
damaged and some detached. Damage was largely focused on one end of the board and
included a distorted circuit board with the left narrow part detached, a detached chip and a
cracked chip (Figure 8). Despite the damage, a specialist organisation was able to recover
data from the item.

Figure 8
Tablet logic board

The tablet contained an aviation navigation application (“app”), from which route information,
flight path and app settings were recovered. The waypoints and path are shown in Figure 1.

© Crown copyright 2023 30 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

A function enabling the app to access the internet in flight was selected off. Therefore,
there was no access to the latest weather information from within the app once airborne.

The weather information was last updated the day before the accident. The NOTAM
information was last updated approximately 17 minutes before the end of the accident flight.
Given that internet access was off, preventing an update, this indicates that the pilot had
changed which NOTAMs were hidden or unhidden.

The recorded track of the aircraft stopped an estimated 27 seconds before the final impact.
This was during the start of a turn to the left, close to the accident site. The recorded file
showed that tracking was not stopped due to user input. The app company stated that there
was no explanation for the loss of this period associated with the app itself. However, it also
stated that any buffering delays of the underlying tablet operating system were not known.

The data from the tablet was reviewed for other relevant activity outside of the use of the
app. It showed that at 0519 hrs on the day of the accident, weather information for St Mary’s
(Isles of Scilly), Sidmouth, Bodmin and Exeter was checked on a BBC website. The tablet
contained a screenshot of the weather information for Exeter taken at 0519 hrs (Figure 9).
As the information is not intended for aviation use, it provided insufficient cloud information
for aviation use, summarising the weather as “Light cloud and a moderate breeze”.

Figure 9
Screenshot from the tablet – (times are UTC +1)

© Crown copyright 2023 31 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Radar

Primary radar data from airfield installations along the flown route were provided by the
respective airfield ATC units. The operator of enroute radar facilities found primary radar
tracks for large parts of the flight path, aiding the investigation early on and corroborating
track data later recovered from the navigation app.

The aircraft was fitted with an ATC transponder but was not detected by secondary radar
until the D&D Cell asked the pilot to switch the transponder on. The secondary radar
recordings from Clee Hill and Burrington radar facilities were provided. Burrington radar
provided the most complete recording of the end of the flight but stopped approximately
2,000 ft above the accident site. Figure 10 shows the data for the duration of the secondary
radar recording and data from the navigation app over the same period.

Burrington’s radar antennas swept the area every 8 seconds. The next sweep after the
last recorded radar return did not detect the aircraft, either because the transponder
antenna was obscured by an unusual aircraft attitude, or because the aircraft had rapidly
descended below the line of sight of the radar. Such a descent would have required
a loss of approximately 1,700 ft in 8 seconds, equating to a descent rate in excess
of 12,750 ft/min.

Radio transmissions

RT recordings were obtained from the NATS Swanwick facility where the D&D Cell is located
and from Exeter ATC. The recordings included telephone conversations between the two
facilities associated with the aircraft, pertinent extracts of which are provided in the History
of the flight section of this report.

Aerodrome logs and recordings showed no communication with the aircraft other than those
described in the History of the flight section.

CCTV

CCTV from a local farm did not show the aircraft but provided evidence of the visual
conditions before, during and after the accident (Figure 11). The times shown compensate
for errors in the embedded timestamps.

The tree line that is about 360 m from the camera was clearly visible in the recorded image
an hour before the accident but was no longer visible in the period leading up to and after
the accident. Trees about 170 m from the camera also became hazy at about the time of
the accident. Ground level visibility significantly improved over the next hour.

The CCTV camera recordings included audio. Audio from one of the cameras captured
the sound of the aircraft propellor intermittently for about 90 seconds before the aircraft
contacted the ground. The pitch of the audio varied in this period, reflecting a combination
of a higher propellor speed due to airspeed or throttle changes, and distortion of the pitch
by travelling towards or away from the audio recording device.

© Crown copyright 2023 32 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Figure 10
End-of-flight radar, navigation app data and CCTV audio signature data

© Crown copyright 2023 33 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Figure 11
Cropped snapshots from one of the CCTV cameras on a local farm
showing changing visibility in the area

Final descent

The audio characteristics, proximity of the first left turn to the accident site, the altitude
above the terrain and the expected radar line of sight capability in the area, indicate that
the aircraft flew another tight left turn after the end of the radar recording. If there were no
significant changes to throttle settings, the audio indicates that the final turn was associated
with higher speeds than the previous 10,000 ft/min / 200 kt descent, followed by a brief
reduction in speed before impact with the terrain.

Other traffic

There was military jet activity east of Exeter during the later stages of G-BXBU’s flight.
Comparison of the recorded aircraft paths showed that this traffic was not close enough
laterally or vertically to have influenced the accident aircraft directly. The recorded radio
communications with the accident pilot did not suggest the traffic had interacted in any way
that had affected the controllability of G-BXBU.

© Crown copyright 2023 34 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Aircraft information

The Mudry Cap 10B is a low wing aerobatic aircraft predominantly constructed from wood
and powered by a Lycoming AEIO-360 four-cylinder engine with a fixed pitch, wooden
propeller. The pilot and passenger sit side-by-side and it is certified for flight under VFR
only.

The wing is a single piece with a main spar of spruce and birchwood. It is covered with
2 mm thick plywood and fabric. The fuselage is a spruce lattice structure covered with
fabric with the vertical fin an integral part.

There were two fuel tanks, each having a capacity of 20 gallons, located within the fuselage.
The primary tank was forward of the instrument panel and behind the engine firewall. The
auxiliary tank was under the baggage compartment to the rear of the cockpit. The tanks
were constructed from thin gauge aluminium sheet and secured to the aircraft structure by
steel straps.

Each seat was fitted with a five-point harness which was attached the aircraft structure; the
lower fixings to the main spar and the shoulder straps to the upper cockpit structure. In
addition, a secondary lap strap belt was also provided which was secured to the main spar.

Aircraft examination

The aircraft was recovered to the AAIB facilities where it was laid out to confirm that all the
aircraft had been at the accident site. No significant items were missing. Along with the
identifiable structural items, the flight control systems were also laid out and examined to
verify continuity (Table 1).

System Components Breaks Comments

Cables, push pull rods, Push pull rods broken
Ailerons Yes
bell cranks through bending
Push pull rod broken
Elevator Cables, push pull rod Yes
through bending
Elevator trim Cable No
Cables cut during
Rudder Cables Yes
aircraft recovery
Push pull rods, Push pull rods broken
Flaps Yes
bell cranks through bending
Table 1
Flight control continuity

The AAIB determined that the damage to the engine was probably sustained during the
impact and no evidence was found of any anomalies that would have prevented normal
operation.

© Crown copyright 2023 35 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Meteorology

METAR and TAF information is shown in Appendix 1. A TAF was available for St Mary’s
from 0629 but not all relevant en route or alternative aerodromes had begun reporting for
the day. However, the pilot lived 35 minutes from Watchford Farm and while some weather
forecasts for relevant aerodromes had become available by the time the flight departed, it
may not have been practical for the pilot to access them after he left home. There was a
weather forecast for below 10,000 ft published at 0312 hrs by the Met Office (Figure 12).
Although this forecast was valid at 1200 hrs, it was available before G-BXBU departed and
indicated the weather conditions which were expected along the planned route.

Figure 12
Met Office forecast for below 10,000 ft

During the investigation the Met Office provided the following interpretation of forecast
conditions in south-west England:

‘Conditions [in area C2] were expected to be generally 15KM visibilities [sic],
but occasional areas of rain with 7000 m visibilities, isolated (occasional for
upslopes) areas of rain and drizzle or mist with visibilities of 3000 m, and
occasional areas of hill fog. There were expected to be isolated areas of
scattered/broken altocumulus with bases of 8000 ft and tops of 10000 ft or
above, overcast cumulus or stratocumulus cloud with bases 1500-3000 ft and
tops of 5000-8000 ft, occasional areas of scattered or broken stratus with bases
500-100 ft and tops 1500 ft, locally bases of 200-400 ft on upslopes and at
the surface in the hill fog. The freezing level was expected to be at or above
10,000FT.’

© Crown copyright 2023 36 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Approximately 30 minutes before the aircraft departed from Watchford Farm, several
METAR’s and TAF’s relevant to the planned route to St Mary’s were published, indicating a
deterioration in weather conditions, locally and at the planned destination.

The navigation app used for flight planning the previous evening displayed the then most
recently published METAR and TAF weather information for the airfields along the planned
route. There was no evidence the app was used to assess the weather on the day of the
accident flight. The screenshot of Exeter weather which was accessed through the news
website indicated cloudy conditions, with the temperature increasing throughout the day.

It was not possible to reproduce exactly what weather would have been presented on the
app at a particular time. However, it probably generated weather for Yeovilton Naval Air
Base, Exeter Airport, Newquay Airport, Culdrose Naval Air Base, Land’s End Airport and
St. Mary’s Airport. The actual and forecast conditions for these aerodromes available at
0600 hrs, 0630 hrs and 0900 hrs are available in Appendix 1. These weather reports show
a marked deterioration in the conditions as the morning progressed; the extent of the poor
weather was not evident in the early morning reports.

Witnesses who were flying locally at the time of the accident described the weather conditions
as ‘intermittent IMC’ with areas of VMC between 3,000 ft and 4,000 ft. Figures 13 and 14
shows satellite imagery of the cloud cover along the south coast.

Figure 13
Satellite image at 0700 hrs

Figure 14
Satellite image at 0915 hrs

© Crown copyright 2023 37 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Both Birmingham Airport and Gloucestershire Airport were within 90-minutes flying time of
the aircraft’s position during the emergency. Both reported 1-2 octas of cloud and good
visibility; conditions that were suitable for flying a visual approach.

Aids to navigation

Although not required for the planned flight, G-BXBU was fitted with a pressure altitude
reporting Secondary Surveillance Radar (SSR) transponder. The transponder was not
switched on during the flight until requested by the D&D controller, after the pilot had
declared an emergency.

The UK Aeronautical Information Publication (AIP) states that, where fitted, pilots shall
operate the transponder to the full extent of its capabilities. The retention by the UK of the
relevant EU Regulations means that the Standardised European Rules of the Air (SERA)
apply to aircraft operating in UK airspace. SERA.130016 states:

‘1. When an aircraft carries a serviceable SSR transponder, the pilot shall
operate the transponder at all times during flight, regardless of whether the
aircraft is within or outside airspace where SSR is used for ATS purposes.

2. Pilots shall not operate the IDENT feature unless requested by ATS.

3. Except for flight in airspace designated by the competent authority for

mandatory operation of transponder, aircraft without sufficient electrical
power supply are exempted from the requirement to operate the transponder
at all times.’

Aerodrome information

Watchford Farm has two grass strips, 08/26 and 04/22, both of which are 400 m long and
20 m wide. The airfield elevation is 840 ft amsl.

Exeter Airport is an international airport with one runway, 08/26. The landing distance
available on Runway 26 is 2,036 m. The airport has regular commercial traffic and has
various instrument approaches available for both runways.

Personnel

Background

The pilot had a total flight time of just over 1,400 hrs. He held a valid PPL(A) with a valid
Single Engine Piston (SEP) rating issued by the CAA, and his medical was in date. He
had completed a total of 1.5 hrs of instrument flying during his initial PPL training 21 years
earlier.
Footnote
6
The retention by the UK of the relevant EU Regulations means that the SERA apply to aircraft operating in
UK airspace ‘Standardised European Rules of the Air’ Annex: Rules of the Air Section 13, available at https://
www.easa.europa.eu/en/document-library/easy-access-rules,https://www.easa.europa.eu/en/document-
library/easy-access-rules/online-publications/easy-access-rules-standardised-european?page=20
accessed 1 November 2022.

© Crown copyright 2023 38 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The pilot had owned G-BXBU since 2014 and it had been hangered at Watchford Farm
since 2015. He was described by flying acquaintances as a ‘fair weather’ flyer. It was
reported that he did not routinely request an ATC service, nor did he operate the aircraft’s
transponder.

The passenger had no flying experience but was known to have flown previously as a
passenger in G-BXBU.

Pre-flight planning

The day before the accident, the pilot contacted St. Mary’s Airport by telephone to make
a prior permission request for his flight the following day. During that call, the pilot
indicated that he intended to land at Bodmin Airfield should the weather at St. Mary’s
not be suitable.

Air traffic control

Air traffic control assistants

Air traffic controllers may delegate some of their responsibilities, including duties which
are closely associated with the safety of aircraft (such as phone calls regarding flight
data), to adequately trained support staff such as air traffic control assistants (ATCAs).
The responsibilities which can be delegated must not require an air traffic control licence.
The Exeter assistant and D&D support controller were not licenced to make decisions
concerning the diversion of G-BXBU.

Management of emergencies

The Manual of Air Traffic Services (MATS) Part 1 contains procedures, instructions and
information intended to form the basis of air traffic services in the United Kingdom. Section
5 ‘Emergencies’ states that pilots should contact an ATSU as soon as it becomes apparent
that an emergency situation exists7, to allow the ATSU to provide the necessary priority
and assistance as appropriate to the emergency. There are two states of emergency
which are classified and declared as follows:

‘Distress: defined as a condition of being threatened by serious and/or imminent

danger and of requiring immediate assistance. Distress is indicated by the
words “MAYDAY MAYDAY MAYDAY” being spoken on the RTF.

Urgency: defined as a condition concerning the safety of an aircraft or other

vehicle, or of some person on board or within sight, but does not require
immediate assistance. Urgency is indicated by the words “PAN PAN, PAN PAN,
PAN PAN” being spoken on the RTF.’

Annex 10 to the International Civil Aviation Convention (Annex 10) Volume V states that
the emergency channel 121.5 MHz shall only be used for genuine emergency purposes

Footnote
7
‘Manual of Air Traffic Services (MATS) - Part 1’ Section 5 Chapter 1 4.1 – Civil Aviation Publication (CAP) 493.

© Crown copyright 2023 39 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

and when normal channels are being utilised by other aircraft, although the UK has filed a
difference for the purpose of pilots conducting practice PAN radio calls.

Annex 10 Volume II states that aeronautical stations shall guard and maintain a continuous
listening watch on the emergency channel 121.5 MHz during the hours of service of the
units at which it is installed8. Civil Aviation Publication (CAP) 4139 states that this ICAO
requirement is not applied in the UK. ICAO Annex 10 Vol II further states that the station
addressed by an aircraft in an urgency or distress condition will normally be that station
communicating with the aircraft or in whose area of responsibility the aircraft is operating,
until it is considered better assistance can be provided elsewhere.

MATS Part 1 states:

‘controllers shall offer as much assistance as possible to any aircraft that is

considered to be in an emergency situation, including weather information,
availability of aerodromes and associated approach aids’. It further states that
‘before transferring an aircraft, controllers should obtain sufficient information
from the pilot to be convinced that the aircraft will receive more assistance from
another unit.’

The United Kingdom AIP states that distress and urgency communications within the
UK Search and Rescue Region (SRR) are in accordance with standard international
procedures10. It also states that the D&D Cell exercises ‘executive control’ over emergencies
in the London and Scottish FIRs, which encompasses the airspace covering England,
Scotland, Wales and Northern Ireland.

The Distress and Diversion Cell

General

The D&D Cell is a military air traffic unit based at the London Area Control Centre at
Swanwick. The service the D&D Cell provides is a collaboration between military operators
and civil providers – the unit is exclusively operated by Royal Air Force personnel who use
equipment owned by a civil air navigation service provider (ANSP). The service provided
is described in CAP 413 as unique to the UK.11

The minimum requirement to be a D&D controller is to hold a valid area control

endorsement (AC EMerg), meaning they are qualified to control air traffic in an area
environment. Prior practical experience is not a requirement to become a D&D controller.
The D&D Cell commonly receives ‘practice PAN’ calls and the general aviation community
is encouraged to practice these calls during training. Pilots have reported comparatively
higher levels of transmission on 121.5 MHz in UK airspace than in other jurisdictions,
primarily involving ‘practice PAN’ calls.
Footnote
8
‘International Civil Aviation Organisation’ (ICAO) Annex 10, Vol II 5.2.2.1.3 & Vol II 5.3.1.5, Vol V 4.1.3.1.1.
9
CAP 413 – ‘Radiotelephony Manual’.
10
‘UK AIP’ - GEN 3.6.6.1 Search and Rescue, accessed at https://www.aurora.nats.co.uk/htmlAIP/
Publications/2021-12-02-AIRAC/html/index-en-GB.html on 2 December 2021.
11
CAP 413 – ‘Radiotelephony Manual’, Chapter 8.7.

© Crown copyright 2023 40 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The role of the D&D Cell is to provide military and civil pilots with emergency communication
and aid, a position fix service, and a search and rescue alerting service12 within the Scottish
and London FIRs. It achieves this in part by providing pilots with the weather and operational
status of an aerodrome, selecting a suitable aerodrome, and providing a steer toward that
aerodrome. D&D utilises two boards to display all current military aerodrome weather
‘colour codes’13 across the UK in order to select a suitable diversion. It also has access to a
limited number of electronic weather reports from civil aerodromes around the UK; it stated
that 51 of the 558 civil airfields are potentially able to provide electronic weather. Details
from the other 507 are obtained by calling the aerodrome or farm strips on a landline. The
unit may be contacted by civil pilots on the VHF emergency frequency 121.5 MHz and by
military pilots on UHF frequency 243.0 MHz, all day and every day.

The D&D Cell has the facility to detect emergency SSR squawks automatically. It can also
locate an aircraft’s position using VHF Direction Finding14 (VDF), subject to the aircraft’s
position and altitude. The service uses the callsign ‘London Centre’ and the AIP states that
it provides coverage over the greater part of the UK above 3,000 ft15.

According to the structure in place at the time of the accident, when the D&D Cell receives
an emergency call directly on 121.5 MHz, it automatically assumes executive control and
operational control of the emergency. MATS Part 1 states:

‘Once D&D hand the aircraft to another unit they pass-over Operational
Control but retain Executive Control. This means that D&D do not give up all
responsibility for an emergency once the aircraft is working another unit. They
retain responsibility for overall management until the emergency ends.’ 16.

Operational control is control by an ATSU directly issuing instructions and support to

the emergency aircraft, which should be consistent with the executive control objectives
determined by the D&D Cell. The D&D Cell transfers operational control when it completes
a handover of the traffic to another ATSU. Guidance for D&D controllers states they are to
‘verify before handing Operational Control to another agency, that the receiving controller
has been given all the details’.

If pilots experiencing an emergency are already in communication with a military or civil

ATSU, they should request assistance directly from them. Air traffic controllers should
inform the D&D Cell of an aircraft emergency17, at which point the D&D controller assumes
executive control. The D&D controller normally delegates operational control back to the
ATSU, but this may depend on the circumstances of the event. MATS Part 1 also provides
controllers with guidance on selecting the most appropriate controlling agency for managing
Footnote
12
CAP 413 – ‘Radiotelephony Manual’, Chapter 8.5.
13
Military METAR reports also display a colour state according to cloud base and visibility.
14
VDF provides information on the position from which a VHF transmission was made.
15
‘UK AIP’ GEN 3.4 Section 3.2.5 – Emergency Telecommunications Services, accessed at https://www.
aurora.nats.co.uk/htmlAIP/Publications/2021-12-02-AIRAC/html/index-en-GB.html
[accessed December 2022].
16
CAP 493 – ‘MATS - Part 1’ Section 5 Chapter 1 9.1.
17
CAP 493 – ‘MATS - Part 1’ Section 5 Chapter 1 2.5.

© Crown copyright 2023 41 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

an emergency aircraft18. The executive control and associated responsibility for managing
the aircraft held by the D&D Cell persists until the emergency ends19.

MATS Part 1 section 5, 9.2 states:

‘D&D controllers have a detailed knowledge of minor aerodrome availability

within their area as well as a comprehensive database that enables rapid
communication with aerodromes, Aircraft Operators, ATSUs, and the SAR
organisation including Police Air Support Units and the regional emergency
services. The D&D Cell can assist a pilot of an aircraft in an emergency and the
civil ATSU to select the most suitable diversion aerodrome.’

MATS Part 1 also states that the D&D Cell do not have detailed knowledge of the local
airspace, terrain or obstacles surrounding aerodromes, and may seek guidance on local
minimum safe altitudes to serve emergency aircraft. MATS Part 1 further states that ATSUs
should not transmit on 121.5 MHz without the authorisation of the D&D Cell, unless the pilot
in distress calls a specific local ATSU, or if it is apparent that the D&D Cell is not responding
to an emergency transmission.

Responsibility and oversight

The Department for Transport is responsible for the overall provision of the national
aeronautical search and rescue (SAR) operations. The initial response to and coordination
of aeronautical SAR is integrated with maritime response and is fulfilled by HM Coastguard.

Before 2016, SAR helicopters were operated by the military and organisationally the D&D
Cell sat within the military SAR operation. In 2016, SAR operations were transferred to
HM Coastguard and since then the Aeronautical Rescue Coordination Centre based at the
National Maritime Operations Centre in Fareham coordinates all helicopter and fixed wing
SAR assets. These aircraft are operated by civilian contractors.

The responsibilities held by the D&D Cell remained following this transfer of SAR provision
from the military to HM Coastguard. The AAIB was not provided with evidence of any
agreement documenting the responsibilities with which the D&D Cell was tasked under
these new arrangements.

MATS Part 1 is a CAA publication which contains several references to the D&D Cell.
However, as a military unit, the D&D Cell is not subject to oversight from the CAA, despite
providing a service to the civil aviation community in the UK. The responsibility to ensure
the D&D Cell is providing the required level of service, detailed in ICAO Standards and
Recommended Practices, is delegated by the Department for Transport to the Ministry of
Defence. The D&D Cell is therefore subject to operational oversight by the Military Aviation
Authority (MAA).

Footnote
18
CAP 493 – ‘MATS - Part 1’ Section 5 Chapter 1 8.1.
19
CAP 493 – ‘MATS - Part 1’ Section 5 Chapter 1 9.1.

© Crown copyright 2023 42 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The MAA issues Regulatory Articles (RA’s), which provide the framework of policy, rules,
directives, standards, and processes; and the associated direction, advice and guidance
that govern military aviation activity and against which air safety is assessed. RA 331120
details the actions expected of a controller once an aircraft has declared an emergency:

‘Regulatory Article RA 3311 (1) Controllers Emergency Actions

Rationale: Air Systems with emergencies need to be afforded special attention
by controllers RA 3311(1) Controllers shall offer as much assistance as possible
to any Air System that is considered to be in an emergency situation.

AMC 3311 (1) On notification that an Air System is suffering an emergency,

controllers should:

a. Inform the pilot of the most suitable aerodrome, considering weather

conditions (including winds), terrain and obstructions. The pilot can be
offered navigational assistance.
b. Coordinate actions with Distress and Diversion and other Air Traffic
Control (ATC) units as required and alert crash and rescue facilities.
c. Advise other Air Systems of the emergency in progress and, where
possible, keep them off the frequency being used by the Air System in
distress. If possible, avoid changing the frequency of the Air System in
distress once suitable contact is established’.

In contrast to the description of executive control in MATS Part 1, the D&D Cell has described
its executive control as ‘administrative’.

Civil Air Navigation Service providers

Responsibility

Although the D&D Cell assumes executive control of all emergencies declared in the London
and Scottish FIR’s, some aerodromes can also offer civil pilots an effective emergency
communications and aid service on 121.521. These airports are listed in ICAO European Air
Navigation Plan, Volume I Part II. Exeter Airport is included on this list.

ICAO guidance for air traffic management states that when an emergency is declared by an
aircraft, the ATSU should take appropriate and relevant action22, including:

‘Take all necessary steps to ascertain aircraft identification and type, the type of
emergency, the intentions of the flight crew as well as the position and level of
the aircraft; …

Footnote
20
‘Regulatory Article (RA) 3311’: controllers emergency actions accessed at https://www.gov.uk/government/
publications/regulatory-article-ra-3311-controllers-emergency-actions on 2 December 2021.
21
‘UK AIP’ – GEN 3.6.5, accessed at https://www.aurora.nats.co.uk/htmlAIP/Publications/2021-12-02-AIRAC/
html/index-en-GB.html on 3 December 2021.
22
Doc 4444 ‘(16th ed) – ‘Procedures for Air Navigation Services - Air Traffic Management’ (PANS-ATM)
Chapter 15.1.1.2.

© Crown copyright 2023 43 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

… ‘Decide upon the most appropriate type of assistance which can be

rendered’…

…. ‘Provide the flight crew with any information requested as well as any
additional relevant information, such as details on suitable aerodromes,
minimum safe altitudes, weather information;’ …

The CAA defines abnormal and emergency situations (ABES) as situations, including
degraded situations, which are not routinely or commonly experienced and for which
automatic skills have not been developed, and serious and dangerous situations requiring
immediate actions. ABES training is included in refresher training programme for ATCOs
and should include dealing with aircraft emergencies23. Training to prepare controllers to
react to ABES events are outlined in CAP 584 – ‘Air Traffic Controllers – Training’.

Exeter Airport did not have a procedure specifically for dealing with VFR traffic stuck above
cloud, but the ‘Emergencies: general’ aid memoire was available to controllers (Figure 15).

Figure 15
Exeter Airport ‘Emergencies: general’ and ‘Further emergency’ procedures

The controller on Radar South did not utilise a specific ABES procedure or checklist prior to
accepting G-BXBU or during her transmissions with the pilot. She referred to G-BXBU as a
‘weather diversion’ on several occasions, not as an aircraft that had declared an emergency.
Although the controller stated she knew the aircraft had been in radio contact with the D&D
Cell, and the emergency squawk of 7700 was visible on her screen, the status of G-BXBU’s
emergency was not interrogated and no handover from the D&D controller was sought.
There was no verbal acknowledgement that its pilot might require additional assistance to
that of a routine arrival to Exeter.

Exeter Airport follows a unit training plan which details the training and assessment
requirements for controllers.

Footnote
23
CAP 584 – ‘Air Traffic Controllers – Training’.

© Crown copyright 2023 44 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Section 3 of the unit training specifies the required training for ABES. It states:

‘Staff must be able to establish some basic information as early as possible.

When appropriate ascertain:
Nature of the problem
Intentions of personnel (e.g. aircraft crew or emergency services)
Time available
Additional information’

It advises that ATC staff should be aware that aircraft emergencies are not always
announced as such, and that staff should be prepared to act in response to events if it is
thought an emergency is developing, even without a ‘PAN PAN’ or ‘MAYDAY’ call.

Controllers are periodically assessed on a variety of emergency scenarios described in the

unit training plan. In one scenario described in the ‘Aerodrome’, ‘Approach control’ and
‘Approach control surveillance’ sections of the plan, ‘a pilot makes a PAN or MAYDAY call,
or other information indicates an emergency situation’, indicating this is a scenario which
controllers would be expected to manage effectively.

Instrument approaches
The controller initially told the pilot to expect an ILS approach, followed very shortly by a
change of plan to an SRA approach, both of which are instrument approaches. To fly an
instrument approach the aircraft must have appropriate instrumentation, and the pilot must
be trained to fly in IMC and hold a valid instrument rating. The pilot of G-BXBU was not
licenced to fly in IMC and the aircraft was not equipped to carry out an ILS approach.

An SRA is an instrument approach flown by the pilot according to ATC heading and rate
of descent instructions. The controller assesses the aircraft position and height on radar
and, when required, issues corrective headings and descent rates to regain the desired
approach path. The minimum obstacle clearance height for a category A aircraft on an SRA
approach to Runway 26 at Exeter is 788 ft.24

SRA approaches are not part of the PPL syllabus and are not commonly flown by pilots
outside a training environment.

MATS Part 2 contains local operating procedures specific to each ATC unit. MATS Part 2
for Exeter Airport states:

‘Unless otherwise stated inbounds receiving a radar service are to be offered

vectors for an ILS approach. The pilot will request if an alternative type of
approach is required, including positioning themselves to the ILS.’

It is not clear how this applies to VFR traffic.

Footnote

24
Doc 8168 (5TH Ed) – ‘Procedures for Air Navigation Services - Aircraft Operations Volume I’, Flight Procedures-
Section 4 Chapter 1.

© Crown copyright 2023 45 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Visual Flight Rules

Flights conducted under VFR are permitted in VMC by day outside Class A airspace in
the UK. The pilot of G-BXBU ordinarily flew in uncontrolled Class G airspace. He did not
routinely file a flight plan with ATC and he had not done so for the accident flight. There is
no requirement for VFR flights in Class G airspace to file a flight plan. Aircraft operating in
uncontrolled airspace may request an air traffic service25. VFR flights in Class G airspace
are entitled to request a Basic or Traffic Service. IFR flights in Class G airspace are entitled
to request a Basic, Traffic, Deconfliction or Procedural Service. The pilot in G-BXBU retained
responsibility for avoiding collisions and terrain.

The meteorological conditions in which aircraft are permitted to operate under VFR, are
determined by the class of airspace, altitude and airspeed. The applicable VMC minima
while operating outside controlled airspace are described in Table 2 (G-BXBU was flying at
less than 140 kt).

Below FL 100 Below 3,000 ft

5 km visibility As per below FL 100.
1500 m horizontal separation from or
cloud 5 km flight visibility, clear of cloud, insight
1000 ft vertical separation from cloud of the surface
or (if operating at less than 140 kt)
1500 m visibility, clear of cloud, in sight of
the surface
Table 2
VFR Weather Minima outside controlled airspace

The airspace around Exeter Airport is an Air Traffic Zone (ATZ) with a radius of 2.5 nm
centred on the airport, rising to an altitude of 2,102 ft. An ATZ conforms to the class of
airspace in which it is situated, so the Exeter ATZ is considered Class G airspace. However,
a pilot must obtain permission from the ATSU at the aerodrome to fly, take off or land within
an ATZ26.

Instrument flying

Training

The CAA PPL syllabus includes one flight exercise in which students are introduced to basic
instrument flying. The PPL skills test includes simulated entry into IMC, following which the
student must complete a 180° turn. The student must also demonstrate consideration of
the relevant safety factors. The biennial SEP revalidation flight test does not require any
additional training on inadvertent entry into IMC or for pilots to demonstrate recovery from
a simulated entry to IMC.
Footnote
25
CAP 1434 – ‘UK Flight Information Services’.
26
‘’Guide to Visual Flight Rules (VFR) in the UK’, Civil Aviation Authority, available at https://www.bfgc.co.uk/
VFR_Guide.pdf [accessed April 2023].

© Crown copyright 2023 46 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Instruments

G-BXBU was not certified to fly in IMC or icing conditions. It did however have some
instruments installed which would aid IMC flight, such as an attitude indicator (AI). The
AI on G-BXBU was a Horizon Bendix J-8 model (Figure 16), featuring yellow markings
on an entirely black background. In more modern AI’s the symbolic sky and ground have
representative colourings that are more instinctive in helping pilots to determine the attitude
of the aircraft without outside visual reference (Figure 17).

Figure 16 Figure 17
Horizon Bendix J-8 Model Example of modern AI design

Spatial disorientation

A pilot’s spatial orientation, although supported by other senses, relies heavily on external
visual references. An obscured visual horizon, false horizons from cloud tops or ground
lights, or featureless terrain coupled with conflicting information from other senses such as
vestibular and proprioceptive, can lead to spatial disorientation in flight.

There are five primary contributory factors which may lead to a pilot experiencing spatial
disorientation:

Environment – cloud/ poor visibility resulting in little or no horizon

Manoeuvres – turns and spins which disturb the vestibular system

Pilot – training and practice in instrument flying, workload and

distraction

Aircraft – AI size, colour and ease of interpretation

Health – congestion and other physiological factors affect proper

function of the vestibular system

There are two classifications of disorientation – unrecognised (Type I) and recognised

(Type II). Pilots recognising they have become disorientated should transfer to flying on
instruments and believe them. In an unrecognised event, the pilot will feel normal until

© Crown copyright 2023 47 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

seeing the ground in the wrong place, understanding that the instruments look ‘wrong’, or
until ground impact. In both cases, the pilot requires skill to maintain or recover the aircraft
to a safe attitude until the required visual reference can be established.

Human performance

Decision making and workload management

The pilot of G-BXBU was making decisions under increasingly uncertain conditions as he
ruled out his planned destination, home airfield and a chosen diversion airfield due to bad
weather. This led to him seeking external assistance, in this case contacting the D&D Cell
with the intention of finding a suitable diversion airfield.

A person’s ability to make effective decisions is limited by, amongst other things,
knowledge, skill and the ability to process information. Everyone will reach a point where
there is too much information to process, or too many tasks to complete, for them to do so
effectively – known as ‘cognitive saturation’. An increase in workload to the point where
mental capacity is reached, results in a degradation in the ability to process information.
This weighs subsequent decision making towards using prior knowledge or experience
rather than assessing the current circumstances and solving a novel problem with a
novel solution. A pilot with reduced capacity may choose to land or divert to an airfield
with which they are familiar, even if it has comparably poor weather conditions or with
comparably complex approaches available. When there is limited cognitive space to
analyse all options, this previous success and familiarity become important factors in the
decision-making process.

Instrument flying is considered a more difficult cognitive task than flying visually. Research
on pilot mental workload in flight showed that pilots who do not routinely fly on instruments
reported higher levels of estimated mental workload when flying on instruments than when
landing or taking off27. Inadvertent or unplanned flight into IMC will add further stress to
a pilot who is not prepared to do so and is likely to consume the attention of the pilot
to the point were making decisions beyond the immediate task of flying becomes more
challenging. Anything that adds to the pilot’s workload, such as communication from ATC
or preparing for an unfamiliar approach, will increase the likelihood of the pilot reaching
cognitive saturation.

Communication

Effective communication is necessary to achieve safe outcomes. Without normal

bodylanguage cues ordinarily available to assist in the transfer of information, the
standardisation of verbal communications has long been recognised as an effective way to
avoid or mitigate potential ambiguity or misunderstandings in communications. This includes
the phonetic alphabet and standard phraseology28 which are well accepted norms within
the industry. Although often considered in the context of pilot-to-controller or pilot‑to‑pilot
Footnote
27
‘An Analysis of Mental Workload in Pilots During Flight Using Multiple Psychophysiological Measures’, Glenn
F. Wilson in The International Journal of Aviation Psychology 12:1, pp 3-18, 2002.
28
Doc 9432 – ‘ICAO Manual of Radiotelephony’.

© Crown copyright 2023 48 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

interaction, the premise is relevant for all communication containing safety critical content –
including between controllers and assistants.

CAP 584 – ‘Air Traffic Controllers – Training’ states that effective communication in normal
and emergency scenarios, and human factors training, shall be assessed as part of controller
refresher training29. CAP 737 – ‘Flight Crew Human Factors Handbook’ states that although
UK ANSP’s ‘tend to have different views on HF training, but in the main the concept of
CRM and human factors has transferred across to Air Traffic Control’30. CAP 737 also
suggests that exposure of pilots and controllers to each other’s operational environment
can be beneficial, although it is not a requirement for initial or refresher training for pilots
or controller’s. In the past, controllers in the UK received some flight training, but this is no
longer the case.

The Exeter Airport human factors refresher training plan for the unit included team resource
management, fatigue management and stress management. Eurocontrol31 suggests that
a breakdown in teamwork makes it more difficult for an individual or team to identify and
correct weaknesses in monitoring pilot actions, communication between ATC personnel
(including handovers) and between controllers and pilots32 .

Further guidance for controllers can be found in CAP 745 ‘Aircraft emergencies – Controller
considerations’, which provides guidance for controllers to understand the challenges which
may be faced by flight crew during an emergency. It is primarily focused on commercial
multicrew operations.

There is no training or guidance for controllers on stress responses that a general aviation
single-pilot might experience during an emergency, how this may manifest and mitigation
strategies which could be employed.

Checklists

Following checklists is a simple and well-established process, particularly when dealing

with abnormal events. They provide an additional safety barrier where personnel are often
operating outside their normal routines and with elevated levels of stress. Checklists can be
used to ensure critical actions are completed, or that critical information required to inform
the decision-making process is obtained. Their use can free mental capacity to create
novel plans whilst ensuring those critical tasks are not missed. While the use of checklists is
embedded in most aircraft operations, this concept has not transferred to ATC to the same
degree.

Footnote
29
CAP 584 – ‘Air Traffic Controllers – Training’ - Chapter 12 pp 52.
30
CAP 737 – ‘Flight Crew Human Factors Handbook’.
31
Eurocontrol is a pan-European organisation that provides technical and civil-military expertise in air traffic
management.
32
Team Resource Management, Guidelines for the Implementation and Enhancement of TRM, Eurocontrol,
2021.

© Crown copyright 2023 49 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

There is no reference to checklists in MATS Part 1 and ATSU’s are not required to use
them in emergency and abnormal events. Exeter Airport did have a list of procedures and
aide-memoire’s33, including ‘Emergencies: general’. This aide-memoire was available at
controller stations but not used routinely. The D&D Cell similarly stated they did not follow
mandatory checklists in response to abnormal or emergency events.

Analysis

The accident

There was no evidence of any aircraft defects before impact that might have affected its
controllability. It was not possible to determine the speed at impact with the large tree but
the spread of wreckage and distance travelled by some of the larger pieces indicate it was
probably greatly in excess of normal landing speed.

Decision to fly

When the pilot checked the weather information online at 0519 hrs, it indicated that
at Exeter there would be light south-westly winds throughout the morning with less
than 20% chance of rain. Weather forecasts for aerodromes along the planned route
deteriorated throughout the morning and the extent of the poor weather was not reflected
in the weather information available at 0600 hrs.

It was not possible to establish the extent of any additional weather planning the pilot carried
out before departure. The forecasts available when he left home differed significantly from
those that became available before G-BXBU took off.

In the absence of sufficient weather reports earlier in the day, the pilot could have delayed
the flight until all relevant forecasts were available. There was an indication of poor
weather enroute in the Newquay forecast available at 0600 hrs. There were no reports for
the destination until the METAR at 0620 hrs and forecast at 0629 hrs, the latter showing
that low cloud was expected at the time of arrival and, although the forecast conditions
may have been sufficient to operate under VFR, it indicated the weather might deteriorate
close to VFR limits. The conditions were sufficiently poor to merit reconsidering the flight
or having a diversion plan to mitigate the risk posed by the low cloud base if the flight
went ahead.

There were clear skies when the flight departed, which may have reinforced the pilot’s
belief the conditions were suitable for the intended flight. Nevertheless, there was sufficient
ambiguity or indication of poor weather to suggest conditions might not be suitable for VFR
flying.

In-flight decision making

It is likely the marked deterioration in the weather as the flight proceeded towards St Mary’s
prompted the pilot not to continue to his planned destination. His stated intention to divert to
Footnote
33
An aide memoire in this context has the same function and purpose as a checklist.

© Crown copyright 2023 50 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Bodmin demonstrates a degree of contingency planning, but it is not possible to know if he

considered an enroute diversion, because he was not communicating with air traffic control.

It is difficult to create novel plans while under pressure and, in the absence of an obvious
alternative option, it was logical the pilot attempted to return to his home airfield where he
was familiar with the local flying environment and where skies were clear when he departed.

The investigation determined that at the time of the accident there were at least two
aerodromes available with weather conditions suitable for G-BXBU to conduct a visual
approach, offering the possibility of a safe outcome had the conditions been checked.

Instrument flying skills

The pilot had logged the minimum instrument flight training required by the PPL syllabus at
the time of his initial training. It is unlikely this training enabled him to deal with this event as
it was limited in scope and completed more than 20 years ago, and there is no requirement
for pilots to revisit the basics of instrument flying in subsequent licence revalidation checks.
It is possible the pilot had not even discussed the topic of inadvertent flight into cloud in a
training setting since his initial skills test, where executing a level 180° turn is demonstrated.

Planning the response to an abnormal or emergency situation in advance increases the

chance of success, saving time and mental capacity when dealing with the emergency in
flight. Without a plan, experience or recent training to flying in IMC, there was a high risk
that the pilot would become spatially disorientated when trying to conduct an instrument
recovery to a diversion airfield. The simple AI display probably increased the challenge.

Pilots will be better prepared to deal with these factors if they are more aware of them. The
following recommendation is therefore made:

Safety Recommendation 2023-011

It is recommended that the Civil Aviation Authority publish guidance for general
aviation pilots on responding to unexpected weather deterioration, highlighting
the factors affecting their performance and the benefits of planning before the
flight how they will respond.

Transponder

The pilot did not comply with regulations requiring the use of transponders when fitted. Had
the transponder been operating throughout the flight, it would probably have enhanced
Exeter ATC’s situational awareness when concern arose about a potential conflict between
G-BXBU and the military jet which was holding in the vicinity. It is not possible to know what
effect this would have had on the outcome of the accident flight. It is not known if the pilot
monitored the relevant frequencies and if Exeter could have contacted him earlier. When
the pilot did contact the D&D Cell, he turned on his transponder and both the D&D Cell and
Exeter could then see his altitude on secondary radar.

© Crown copyright 2023 51 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Communication

The pilot’s first contact with the Dunkeswell A/G operator was pivotal because the pilot,
who rarely communicated with ATC, began to request, and accept external influence in the
decision-making process for the flight.

On making first contact with the D&D Cell, the pilot immediately declared an emergency
and requested assistance, explaining that he was stuck above cloud and needed to divert
to somewhere nearby. In response to a request by the controller, he confirmed he was
currently above cloud and had 1.5 hrs fuel endurance. This provided sufficient information
to indicate there was no immediate pressure to provide the pilot with a solution. Despite the
potentially very stressful situation, the pilot succeeded in asking for help when it is likely his
ability to solve the problem himself had diminished.

It is not possible to know why the pilot made a ‘PAN’ call in preference to ‘MAYDAY’ when
he declared an emergency on 121.5 MHz but, given his 90-minute fuel endurance, he may
not have considered he was in immediate danger. Whether a PAN or MAYDAY call was
made, it should not have affected the response by ATC to support G-BXBU. Guidance to
controllers considers both states to be an emergency, the response to which (the application
of the ATC emergency procedure) is the same. The CAA definitions distinguish between
PAN and MAYDAY based on the immediacy of the emergency, not on the nature of support
required. When the D&D controller advised that Exeter was willing to accept the aircraft, it
is unlikely the pilot had the knowledge or mental capacity to question this plan, particularly
as he had declared an emergency and shared relevant information about his problem with
the D&D controller. In his transmission to Exeter ATC the pilot said, “I’ve been diverted”, a
passive phrase that suggests it was not something he chose to do himself.

The D&D controller does not appear to have considered what options were available to the
pilot. In order to select viable diversion aerodromes, a controller would need to know at
least the capabilities of the aircraft and pilot to fly in the conditions likely to be encountered.

To provide effective assistance controllers must provide practical guidance that can be
understood by pilots in distress, and an intervention is more likely to be successful if the
controller recognises when a pilot has a reduced capacity to respond.

Pilots might assume that agencies providing emergency assistance to aircraft will check
the weather of potential diversion aerodromes. In this event, controllers do not appear to
have considered whether the pilot and aircraft were capable of diverting to Exeter, and there
was no obvious attempt to match the style of communication to the circumstances. (For
example, whilst not incorrect, phrases such as “deconfliction service” and “ILS approach”
may not have been useful or reassuring to a pilot with his experience or qualifications.) In
part this may be because controllers are not sufficiently aware of the factors influencing
human behaviour under stress, or how to address them.

CAP 745 aims to provide controllers with a flight crew’s perspective on ATC communications
in an emergency and is phrased in a manner that may make it more applicable to
professional pilots. Pilots of light aircraft often operate in a less formal environment and

© Crown copyright 2023 52 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

with less frequency, and much of the information in CAP 745 may not transfer to the general
aviation environment. MATS Part 1 states that ‘calm and coordinated actions are essential’
when dealing with emergencies but does not specifically address pilot stress reactions and
the assistance which might be provided to account for it. Accordingly, the following Safety
Recommendation is made:

Safety Recommendation 2023-012

It is recommended that the Civil Aviation Authority require air traffic controllers
to receive training regarding the human performance characteristics and
limitations associated with stress. This should include the verbal cues that may
indicate that a pilot is operating under high stress, and mitigation strategies to
help controllers deal with such events.

The D&D controller stated that he made assumptions based on the phone call received by
the D&D support controller from the Exeter assistant. The Exeter assistant, who was not a
controller qualified to make decisions about air traffic, did not identify herself as such on the
phone, and the D&D support controller did not check her status when this information was
omitted. The first assumption made by the D&D controller was that the Exeter assistant
was a controller; the second, that the phone call had been instigated because an Exeter
controller had heard G-BXBU make a ‘PAN’ call on the emergency frequency and intended
to offer assistance. However, the practical application of D&D’s executive control is that the
D&D controller would take the lead in making an assessment, then contact an aerodrome
which they deemed suitable, not the other way round.

The D&D controller advised the pilot that Exeter was willing to accept the aircraft, within
seconds of the D&D support controller telling him this (Appendix 2), leaving no time to
assess the suitability or practicality of this suggestion.

The D&D controller did not independently check the weather conditions at Exeter and a
handover between controllers did not take place. The D&D controller stated he did not want
to delay the transfer of G-BXBU to Exeter, based on the understanding that his support
controller was speaking directly to a controller. However, as the aircraft had a stated fuel
endurance of 1 hour and 30 minutes, there was no need to expedite the transfer at the
expense of a full handover.

The phone call to the D&D Cell was made by the Exeter assistant regarding the military jet
holding locally; it was not for the purpose of assisting an aircraft experiencing an emergency.
Whereas the assistant’s interaction with the D&D Cell ultimately resulted in the diversion of
G-BXBU to Exeter, there was no active or informed decision to that effect. It is possible this
was seen by the Exeter assistant as the most efficient way to remove a potential conflict
with the military jet. However, it appears there was no attempt to resolve this potential
conflict by moving the military jet away from G-BXBU, which was more readily achieved
given the two-way radio contact between Exeter Radar (north) and the jet. The assistant
did not receive or request information about G-BXBU or its pilot’s capacity to carry out an
approach in the prevailing conditions.

© Crown copyright 2023 53 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

These misunderstandings appear to have misled the D&D controller to believe that Exeter
had a more detailed awareness of the nature of the emergency and, significantly, that a
controller there had determined that Exeter was an appropriate diversionary aerodrome.

The Exeter assistant described G-BXBU as a “weather diversion” and, in the absence of
information normally included in a radar handover, the Exeter controller may not have known
the seriousness of the situation its pilot faced. She commented that to her knowledge there
were no other VFR flights operating in the area, which she believed was due to the weather
conditions. She may therefore have expected that aircraft locally would have some IFR
capability. These cues may have acted to confirm the controller’s belief that the aircraft
could make an approach in the prevailing weather conditions, and may explain why she did
not consider she was dealing with an aircraft in difficulty. There were several contrary cues:
G-BXBU had been in contact with the D&D Cell, the controller had heard a ‘PAN PAN’ call
on 121.5 MHz earlier, and G-BXBU’s emergency squawk of 7700 was visible on her radar
screen. Whereas the D&D Cell has executive control in these circumstances, the normal
responsibilities of a civil controller still apply. The Exeter controller was entitled to request
a full radar handover from the D&D controller in order to understand the reason for the
aircraft’s diversion, and the ambiguity around the inbound aircraft was sufficient to indicate
more information was required.

The Exeter controller advised the pilot of G-BXBU that he could expect vectors for the ILS
approach for Runway 26. This indicates she did not appreciate the nature of the emergency
the pilot was experiencing or his ability to carry out an approach in IMC. When the pilot said
‘sorry I can’t, can you say again?’, she was prompted by a colleague to instead offer an
SRA approach, which she did without further discussion. An SRA approach did not require
onboard equipment but did require the pilot to be appropriately trained and qualified. The
SRA decision point34 was above the reported cloud base and therefore was unlikely to have
been successful.

The ATC units involved do not appear to have considered what options were available to
the pilot or to have communicated them effectively to each other. In order to select viable
diversion aerodromes, a controller would need to know at least the aircraft type and the
ability of the pilot to fly in the conditions likely to be encountered. The following Safety
Recommendation is therefore made:

Safety Recommendation 2023-013

It is recommended that the Civil Aviation Authority specify the types of information
that air traffic controllers will obtain and record when responding to aircraft in
an emergency to ensure that pilots’ needs are met and reported correctly if
communicated to other air traffic control units.

Footnote
34
The ‘decision point’ is the point at which an instrument approach must be discontinued and a go-around
flown if the required visual contact is not obtained.

© Crown copyright 2023 54 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

It is not possible to know why the pilot began his descent before being transferred to Exeter
Radar. It is possible he was distracted by his interactions with the D&D Cell; and by trying
to act on their requests to select an emergency squawk, maintain height and heading, and
transfer radio frequency; and that he inadvertently descended into cloud. It is also possible
that, given his proximity to Watchford Farm, he was attempting to establish visual contact
with the ground in the hope of conducting an approach to his home base. Alternatively, he
may have begun the descent intentionally: the aircraft had reached an altitude of 8,200 ft
and was approximately 16 nm from Exeter Airport. Assuming a direct path from there to the
runway flown at 90 kt, the aircraft would have needed to begin its descent and maintain a
rate of descent of approximately 750 ft/min to land. If the pilot was attempting to divert to
Exeter, he would have needed to begin a descent at this point.

It is likely, based on the D&D Cell’s transmission to the pilot at 09:13:22, and the pilot’s
comment to Exeter that he had “been diverted”, that the pilot thought there was a plan for
him to divert to Exeter. If he had reached cognitive saturation, he may have felt his only
option was to follow that plan. Controllers did not appear to understand that the pilot must
descend into cloud to continue with the planned diversion to Exeter, or that he was not
equipped to do so. Consequently, he had the undesirable options of complying with the
apparent plan or questioning its suitability. Given the D&D Cell’s publicised expertise and
responsibilities for providing emergency assistance, and the absence of clear alternatives in
a high stress situation, it is understandable that he would not question their plan.

It is also not possible to know what would have happened if the pilot had maintained altitude
and followed the controller’s instruction to follow a heading. Although there is no evidence
of a considered plan to help the pilot, delaying the descent into cloud and additional
communication between the pilot and Exeter ATC might have revealed Exeter’s unsuitability
as a diversion aerodrome. In the event, the worst of the weather had passed before the
aircraft’s fuel would have been exhausted, and remaining clear of cloud until then, might
have enabled visual approaches to Watchford Farm, Dunkeswell or Exeter. There was no
need for an immediate descent.

At the time the descent began, G-BXBU was in Class G airspace and, as a VFR flight, the
aircraft did not require ATC clearance to descend. However, the D&D controller provided a
deconfliction service to G-BXBU, which is only available to IFR traffic in Class G airspace.
The confusion may have been compounded by the Exeter controller’s initial intention to issue
vectors initially for an ILS, followed by an SRA approach. These instructions suggest she
believed she was controlling IFR traffic, which would ordinarily require descent clearance
from ATC when carrying out an IFR approach. Although offering an ILS approach accords
with local procedures, in this case doing so demonstrated a gap in understanding the pilot’s
circumstances.

Witness reports from pilots flying in the area at the time G-BXBU levelled off at 4,300 ft
indicate the pilot may have been attempting to fly the aircraft in a gap between cloud layers.
Shortly afterwards, at 09:16:49 hrs, G-BXBU was instructed to descend to 2,600 ft. Whilst
this was intended to prevent the aircraft from descending below the minimum safe altitude
in the area, it is likely to have introduced further confusion.

© Crown copyright 2023 55 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Checklists

The D&D Cell and Exeter ATC had checklists or procedures for dealing with emergencies,
although neither controller used a checklist in this event. The use of an appropriate checklist
in this event could have prompted an effective handover and ensured the transfer of critical
information. The reported tendency to use procedures, checklists, and aide memoirs as
guidance rather than formally, may have given rise to the impression that some or all of their
contents could be disregarded.

Checklists make it easier to carry out routine or emergency procedures or create novel
plans in foreseeable circumstances, and there is no evidence they are less effective in the
air traffic control environment.

Therefore, the following Safety Recommendation is made:

Safety Recommendation 2023-014

It is recommended that the Civil Aviation Authority encourage the use of

checklists in air traffic management operations when dealing with abnormal and
emergency situations.

Emergency air traffic service provision

As a military unit, the operation of the D&D Cell differs from that of civilian ATSUs, which
do not operate under a military structure. For example, the D&D Cell has a board providing
an immediate visual guide for airfield data, including weather conditions, at all military
aerodromes in the UK. There is no equivalent display for civil aerodromes. Exeter ATC
was monitoring 121.5 MHz and was aware a ‘PAN PAN’ call had been made. However,
its controllers were not familiar with the detail of the emergency and did not consider
what assistance they could provide the aircraft. As MATS Part 1 restricts an ATSU from
responding to a call on 121.5 MHz, except in limited circumstances, it is understandable
that ATSU’s do not actively monitor the emergency frequency for the purpose of responding
to an emergency.

When declaring an emergency, if not already in receipt of a service from an ATSU, a pilot
should do so on 121.5MHz. The transfer of operational control takes place when the aircraft
in difficulty is handed over from D&D to the best-placed ATSU. It is unclear if the transfer of
operational control took place during this event as there was no conversation between the
D&D and Exeter controllers (as distinct from their assistants), but G-BXBU was transferred
by D&D to the Exeter Radar frequency.

If an aircraft declares an emergency on a civil ATSU frequency, the ATSU will then contact
D&D to give pertinent details of the emergency via landline. D&D will then give the ATSU
operational control, although it is not clear who has the authority should the ATSU and D&D
disagree on the best course of action to support the aircraft.

The AAIB investigation has received differing interpretations of D&D’s executive control.

© Crown copyright 2023 56 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Although MATS Part 1 does provide guidance, there is no published equivalent interpretation
from the D&D Cell, adding to the potential for misunderstanding between ATSUs. MATS
Part 1 provides inconsistent guidance: one section stating that controllers should consider
the most appropriate ATSU to manage an emergency and are empowered to transfer
operational control to another ATSU; another that controllers should advise the D&D Cell
of all emergencies, transferring executive control to D&D who then have the authority to
delegate operational to control where they see fit. In informing the D&D Cell, this may
remove the civil ATSU from the decision-making process. The D&D Cell is not necessarily
the best placed unit to make decisions regarding an aircraft in difficulty, but having executive
control creates an authority gradient in the decision-making process when working with civil
ATSUs. This may cause inefficiencies, and distances controllers with local knowledge from
the decision-making process without objective benefit. Whether the D&D Cell’s executive
control is solely ‘administrative’, or has the result that it holds ‘responsibility for the overall
management of the emergency’ it is important the definition of these terms is clear to all
ATCOs who are routinely interacting with the D&D Cell, to avoid any misunderstanding or
misinterpretation. The positions expressed by the D&D Cell and the CAA are not consistent
with the guidance in MATS Part 1, and currently the situation is not settled. Therefore, the
following Safety Recommendation is made:

Safety Recommendation 2023-015

It is recommended that the Civil Aviation Authority determine the effect the
D&D Cell’s executive control has on civil ATCOs and inform civil ATCOs of any
differences in their responsibilities whilst executive control is exercised.

In this event, Exeter controllers would have been aware that the weather at Exeter Airport
was not suitable for a pilot and aircraft not equipped for flight in IMC, and could have
acted accordingly had they received the initial ‘PAN’ call themselves. The complexity
of communication between multiple personnel at both ATC units meant no controller or
assistant had all the available information at any moment as the event unfolded.

The UK AIP states ‘Distress and Urgency communications within the UK SRR are in
accordance with standard international procedures’35. ICAO Annex 10 states that the
emergency frequency should only be used when normal channels are not available, and
that it should only be used for genuine emergencies. It also states that ATSUs shall monitor
121.5 MHz and that an aircraft reporting a distress or urgent condition shall normally address
the station already communicating with the aircraft, or in whose area of responsibility the
aircraft is operating. Aircraft are required to monitor 121.5 MHz if possible.

The provision of a nationwide service on 121.5 MHz, including for aircraft practicing
emergencies, is unique to the D&D Cell and can lead to volumes of communication on the
frequency that discourage some pilots from monitoring it.

Footnote
35
‘UK AIP’ – GEN 3.6 Paragraph 6.3.1.

© Crown copyright 2023 57 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

The involvement of the D&D Cell in this process increased the opportunity for
misunderstanding and did not assist in achieving a safe outcome. Therefore, the following
Safety Recommendation is made:

Safety Recommendation 2023-016

It is recommended that the Department for Transport review the current provision
of emergency communications in the UK to determine if the involvement of a
dedicated emergency air traffic service unit is the most effective way to assist
civil aircraft in an emergency, and publish its findings.

There is no formal agreement between the Department for Transport and the Ministry of
Defence defining the responsibilities of the D&D Cell. This may be a result of the D&D Cell
continuing to provide a service to both the military and civilian aviation after the provision
of civil SAR ceased to be a military operation in 2016. As the D&D Cell were associated
with the military provision of SAR, it is possible their responsibilities were previously defined
in this context. If the D&D Cell continues to provide emergency support to civil aircraft its
responsibilities should be set out clearly. Therefore, the following Safety Recommendation
is made:

Safety Recommendation 2023-017

It is recommended that the Department for Transport specify and publish details
of the emergency air traffic service it requires the D&D Cell to provide.

Military-civil ATC interaction

Operations with a solely military purpose are outside the scope of this investigation.
However, the AAIB has considered the service provided by the D&D Cell to the civil aviation
community on behalf of the State, and four Safety Recommendations are made to the Civil
Aviation Authority in areas that should also be addressed by the MAA. Accordingly, the
MAA has stated that it intends to address the intent of Safety Recommendation 2023-012,
2023-013, 2023-014, and 2023-015 made to the CAA.

Conclusion

The aircraft collided with terrain because the weather conditions deteriorated beyond the
capabilities of the pilot who was not trained or qualified to operate in poor weather. The
forecasts available when the pilot assessed the weather did not accurately reflect the extent
of the poor weather.

The pilot found himself stuck above cloud. When the pilot requested assistance in finding
an appropriate aerodrome to land, the level of ATC support from the D&D Cell and Exeter
ATC was not sufficient to provide the assistance required by the pilot, who was in a state
of distress. A breakdown in communication and teamwork occurred between the D&D
Cell, Exeter ATC and the pilot, which led to miscommunication, incorrect assumptions and
omission of critical information.

© Crown copyright 2023 58 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Following published procedures would likely have allowed either the D&D Cell or Exeter
Airport ATC to establish the unsuitability of Exeter Airport as a diversion aerodrome.

The investigation identified shortcomings in the system in place in the UK to provide

emergency support to aircraft in distress.

Seven Safety Recommendations are made.

Safety Recommendations

Safety Recommendation 2023-011: It is recommended that the Civil Aviation

Authority publish guidance for general aviation pilots on responding to unexpected
weather deterioration, highlighting the factors affecting their performance and
the benefits of planning before the flight how they will respond.

Safety Recommendation 2023-012: It is recommended that the Civil Aviation

Authority require air traffic controllers to receive training regarding the human
performance characteristics and limitations associated with stress. This should
include the verbal cues that may indicate that a pilot is operating under high
stress, and mitigation strategies to help controllers deal with such events.

Safety Recommendation 2023-013: It is recommended that the Civil Aviation

Authority specify the types of information that air traffic controllers will obtain
and record when responding to aircraft in an emergency to ensure that pilots’
needs are met and reported correctly if communicated to other air traffic control
units.

Safety Recommendation 2023-014: It is recommended that the Civil Aviation

Authority encourage the use of checklists in air traffic management operations
when dealing with abnormal and emergency situations.

Safety Recommendation 2023-015: It is recommended that the Civil Aviation

Authority determine the effect the D&D Cell’s executive control has on civil
ATCOs and inform civil ATCOs of any differences in their responsibilities whilst
executive control is exercised.

Safety Recommendation 2023-016: It is recommended that the Department

for Transport review the current provision of emergency communications in the
UK to determine if the involvement of a dedicated emergency air traffic service
unit is the most effective way to assist civil aircraft in an emergency, and publish
its findings.

Safety Recommendation 2023-017: It is recommended that the Department

for Transport specify and publish details of the emergency air traffic service it
requires the D&D Cell to provide.

Published: 27 April 2023.

© Crown copyright 2023 59 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix 1 – Meteorological reports

Available airfield weather at 0600 hrs

METAR TAF
Yeovilton Not available Not available
Exeter Not available Not available
Between 0600 and 1500:
light south-easterly wind, good visibility, scattered
At 0550:
cloud at 4,000 ft
Newquay light south-east
Temporarily between 0600 and 1200
broken cloud at 1,200 ft
(30% chance):
8,00 0 m visibility, broken cloud 1,200 ft
At 0550:
light south-easterly wind,
Culdrose Not available
good visibility,
scattered cloud at 300 ft
Land’s End Not available Not available
St. Mary’s Not available Not available
Table A1
Available METAR and TAF information at 0600 hrs

Available airfield weather at 0630 hrs

METAR TAF
Yeovilton Not available Not available
Exeter Not available Not available
Between 0600 and 1500: light south-easterly wind,
good visibility, scattered cloud at 4,000 ft
At 0620:
Temporarily between 0600 and 1200: 8,000 m
light southerly wind,
Newquay visibility, broken cloud 1,200 ft.
9000 m visibility,
Temporarily between 0600 and 1200
broken cloud at 600 ft
(30% chance):
broken cloud at 600 ft
At 0550:
light south-easterly wind,
Culdrose Not available
good visibility,
scattered cloud at 300 ft
Land’s End Not available Not available
Between 0600 and 1500: south-westerly wind, good
visibility, few cloud at 1,500 ft
At 0620: Temporarily between 0600 and 0900:
light south-westerly wind, broken cloud at 1,200 ft
St. Mary’s
good visibility, Temporarily between 0600 and 0900
broken cloud at 1,100 ft (30% chance): 7,000 m visibility,
broken cloud at 800 ft
Between 0900 and 1200: wind gusting to 25 kts
Table A2
Available METAR and TAF information at 0630 hrs

© Crown copyright 2023 60 All times are UTC

AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix 1 – Meteorological reports cont

Available airfield weather at 0900 hrs

METAR TAF
Yeovilton Not available Not available
Between 0900 and 1700: light southerly wind, good
visibility, scattered cloud at 1, 000 ft.
Temporarily between 0900 and 1400: 6,000 m
At 0850: - light southerly visibility, moderate rain and drizzle,
Exeter wind, 6000 m visibility, broken cloud 700 ft.
broken cloud at 500 ft Between 0600 and 1200
(30% chance):
temporarily 2,000 m visibility, mist,
broken cloud 400 ft
Between 0900 and 1800:
At 0850: light south- Wind becoming easterly
westerly wind, Temporarily between 0900 and1200: visibility 800 m,
Newquay
good visibility, broken at 1,200 ft
broken cloud at 300 ft Temporarily between 0900 and1200 (30% chance):
1,400 m visibility, mist, broken cloud at 200 ft
Between 0900 and 1800: light variable winds,
At 0850: light south-
good visibility, few clouds at 900 ft,
westerly wind, 250 m
scattered cloud at 2,000 ft
Culdrose visibility, fog,
Temporarily between 0900 and 1000
scattered cloud at 0 ft,
(30% chance):
overcast cloud at 200 ft
500 m visibility, fog, scattered cloud at 100 ft

Between 0900 and 1800: south-westerly wind,

At 0850: light south-
good visibility, scattered cloud at 1,500 ft
westerly wind,
Temporarily between 0900 and 1200:
good visibility,
Land’s End broken cloud at 1,200 ft
few cloud at 500 ft,
Temporarily between 0900 and 1200
scattered cloud
(30% chance):
at 3,000 ft
7000 m visibility, broken cloud at 600 ft
Between 0900 and 1800:south-easterly wind,
At 0850: light south-
good visibility, scattered cloud at 2,500 ft
westerly wind, good
Temporarily between 0900 and 1100:
visibility,
St. Mary’s broken cloud at 1,200 ft
few cloud at 800 ft,
Temporarily between 0900 and 1100
scattered cloud
(30% chance):
at 2,300 ft
7,000 m visibility, broken cloud at 800 ft
Table A3
Available METAR and TAF information at 0900 hrs

© Crown copyright 2023 61 All times are UTC

 AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix22- -Communication
Appendix transcript
Communication transcript

Colour key
G-BXBU
D&D Controller D&D Support (DDS)
Exeter Controller Exeter Assistant (ATCA)
Other aircraft on frequency

Time G-BXBU Radio communication with the Mediator line – communication between
D&D Cell 121.5 MHz and Exeter Radar Exeter assistant and D&D support
123.580 MHz
09:10:59 G-BXBU: Emergency frequency PAN PAN
PAN this is Golf Bravo Xray Bravo Uniform
09:11:17 D&D Controller (broken): Bravo Xray Bravo
Uniform, London Centre, PAN acknowledged,
pass your details when ready
Commercial aircraft 1: Golf Bravo Xray Bravo
Uniform, it's [callsign] [unintelligible] ...go ahead
09:11:21 G-BXBU: Er say again

09:11:23 Commercial aircraft 1: Golf Bravo Xray Bravo

Uniform, this is [callsign] go ahead we heard
the PAN PAN call
09:11:29 G-BXBU: Yeah, er I am er, I’ve got a, I've got in
real trouble, I am a, it's a cap ten, two P O B, I
am about eight miles, er east of Exeter, er and
there is very thick cloud and I am above it and
can't get below it, er according to Dunkeswell it
is on the deck. I don't know what to do. I need
to divert somewhere er close to me where I can
land
09:11:54 Commercial aircraft 2: Golf Bravo Xray Bravo
Uniform er this is [callsign] and we will contact
London
09:12:00 DDS: D&D support

09:12:01 G-BXBU: Sorry can you please speak slower

09:12:01 ATCA: Hello it's Exeter

09:12:02 DDS: Yep

09:12:03 ATCA: Hi, has anyone updated you firstly

about the [military jet]?
09:12:04 Commercial aircraft 2: Er Golf Bravo Xray
Bravo Uniform this is [callsign], we copy what
you are saying I'll call London for you
09:12:07 DDS: Er, no

09:12:08 ATCA: OK er, just to let you know that that

[callsign] is still intending to land at Exeter,
he's got a normal undercarriage indication
now

52

© Crown copyright 2023 62 All times are UTC

 AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix 2 - Communication transcript cont

09:12:14 G-BXBU: Er thank you Bravo Uniform

09:12:16 DDS: OK

09:12:16 D&D Controller: … Xray Uniform this is London

Centre on one two one decimal five. Your PAN
is acknowledged, your position is
approximately four miles to the west of Chard.
What is your altitude?
09:12:17 ATCA: And also, has a light aircraft called
you in the Dunkeswell area?
09:12:21 DDS: Yes, we are currently dealing with
that situation
09:12:23 ATCA: Excellent, he's right in the way of er,
of red five, would you, what's he, what's his
intentions and his level?
09:12:27 G-BXBU: Altitude is currently seven thousand
five hundred, and that is the cloud base
09:12:29 DDS: Er, don't know his level but he is
currently above cloud and er wanting to
divert to the nearest aerodrome
09:12:35 D&D Controller: Golf Bravo Xray Bravo Uniform
confirm you are above cloud
09:12:37 ATCA: Well, that would be us

09:12:38 DDS: Er….

09:12:40 ATCA: Exeter

09:12:41 G-BXBU: I confirm I am above cloud at seven DDS: I think...

thousand five hundred. Er I've called
Dunkeswell they say it is on the deck there. I
am really quite anxious and don't know what to
do
09:12:44 ATCA: He's basically flown all the way up
the coast and then across our extended
centre line twice in front of er, a [military jet]
09:12:49 DDS: Yes

09:12:50 ATCA: Er.. do you want to put him over to

us?
09:12:50 D&D Controller: Golf Bravo Uniform roger. Golf
Bravo Uniform what is your endurance?
09:12:54 DDS (Offline discussion): Exeter are asking
if er, maybe we want to, she asked to put it
over to them…
09:12:55 G-BXBU: Er one and a half hours

09:12:59 D&D Controller: Golf Bravo Uniform roger. Golf

Bravo Uniform standby.
09:13:03 G-BXBU: Standing by.

53
© Crown copyright 2023 63 All times are UTC
 AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix 2 - Communication transcript cont

09:13:04 DDS (Offline discussion): They are

wondering if er, they want to take over
09:13:05 D&D Controller: Golf Bravo Uniform

09:13:06 Commercial aircraft 2: Er London [callsign] are

you happy if we come off frequency now.
09:13:11 D&D Controller: [callsign], affirm we have the
aircraft position and [unintelligible] identified we
will carry on and thank you for your help
09:13:14 DDS: Standby we are just talking to the
aircraft
09:13:15 ATCA: Oh OK, alright (offline: He's working
D and D that aircraft)
09:13:18 Commercial aircraft 2: OK copied, good luck

09:13:21 DDS offline: Exeter are willing to take the

aircraft
09:13:22 D&D Controller: Golf Bravo Uniform roger.
Exeter are willing to take you and standby your
steer for Exeter is two three zero range sixteen
nautical miles
ATCA: [unintelligible]

09:13:33 G-BXBU: What's the Exeter radio?

09:13:35 D&D Controller: Golf Bravo Uniform, we will

hand you over. Golf Bravo Uniform squawk
seven seven zero zero
09:13:41 G-BXBU: Squawking seven seven zero zero

09:13:46 D&D Controller: Golf Bravo Uniform, make your

heading two three zero report steady
09:13:48 DDS: Exe.. er we are putting on emergency
squawk, is there a frequency that we can
put him on to?
09:13:52 ATCA: Er.. one.. one.. hang on

09:13:52 G-BXBU: Ah, can you hold the line

09:13:55 ATCA (offline discussion): Which one of

you wants to work this aircraft inbound, do
you want to [name] or shall [name] take it?
The inbound. For weather. The one that's
been in the way for the last ten minutes.
Yeah. Yeah?
09:14:06 G-BXBU: Er, squawking, er Golf Bravo
Uniform, squawking seven seven zero zero.
Can you say again next instruction?
09:14:10 ATCA: Yeah, if you put it through one two
three five eight zero
09:14:14 DDS: One two three five eight zero

© Crown copyright 2023 54

64 All times are UTC
 AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix 2 - Communication transcript cont

09:14:14 D&D Controller: Golf Bravo Uniform roger.

Head, make your heading two three zero,
maintain seven thousand five hundred feet
09:14:17 ATCA: And what's his call sign?

09:14:18 DDS: Er call sign is…

09:14:21 ATCA: Oh we've got it, we've got it it's OK

09:14:22 G-BXBU: Two three zero, maintaining er seven

thousand five hundred
09:14:23 DDS: You've got

09:14:25 ATCA: Yep, alright then

09:14:26 DDS: alright then bye

09:14:26 D&D Controller: Golf Bravo Uniform identified

on radar, deconfliction service
09:14:27 ATCA: Thanks, cheers bye

09:14:35 G-BXBU: Sorry, say again

09:14:36 D&D Controller: Golf Bravo Uniform you're

identified on radar, in a deconfliction service,
Exeter have you iden…, have you on their
radar, contact Exeter frequency one two three
decimal five eight zero
09:14:52 G-BXBU: One two, one two three decimal five
eight zero
09:14:57 D&D Controller: Golf Bravo, Golf Bravo Xray
Bravo Uniform, that is correct
09:15:05 D&D Controller: Golf Bravo Uniform if no
contact on that frequency return to this
frequency one two one decimal five
09:15:12 Ringing

123.58

09:15:27 G-BXBU: Er Exeter, er Golf Bravo Xray Bravo

Uniform, have been PAN PAN PAN, have been
diverted
09:15:30 ATCA: Exeter

09:15:32 DDS: It's D&D support, we've just passed

him over to you, has he come up?
09:15:34 ATCA: Yes I think he's called us now. Yes
he has
09:15:36 Exeter: Golf Bravo Xray Bravo Uniform, Exeter
Radar, roger the er PAN call, and we'll be
vectoring you for the er ILS approach for
runway two six for Exeter

55
© Crown copyright 2023 65 All times are UTC
 AAIB Bulletin: 6/2023 G-BXBU AAIB-27584

Appendix 2 - Communication transcript cont

09:15:37 DDS: Oh OK perfect.

09:15:40 ATCA: OK

09:15:41 DDS: Will you er let us know when he's

landed?
09:15:42 ATCA: Yes will do

09:15:43 DDS: Thank you

09:15:44 ATCA: Cheers

09:15:44 DDS: Bye

09:15:54 G-BXBU: Sorry I can't, can you say again?

09:15:57 Exeter: Golf Bravo Uniform I'll be vectoring you

for the SRA Approach for runway two six at
Exeter. Fly heading two two zero degrees
09:16:21 Exeter: Golf Bravo Uniform fly heading two two
zero degrees
09:16:26 G-BXBU: Bravo Uniform, heading two three
zero. What's your cloud base?
09:16:32 Exeter: Golf Bravo Uniform the weather at
Exeter we've got six kilometres visibility and the
cloud is broken at five hundred feet
09:16:49 Exeter: Golf Bravo Uniform, descend to
altitude two thousand six hundred feet, QNH
one zero one seven
09:17:00 G-BXBU: One zero one seven, you er you
require me to descend to what altitude?
09:17:04 Exeter: Two thousand six hundred feet

09:17:07 G-BXBU: Descending two thousand six

hundred, you want me on two three zero?
09:17:14 Exeter: Affirm, when you are able, fly heading
two two zero degrees
09:17:21 G-BXBU: Two two zero

09:18:12 Exeter: Golf Bravo Uniform stop descent and

maintain altitude two thousand six hundred feet

© Crown copyright 2023 56

66 All times are UTC
AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

ACCIDENT

Aircraft Type and Registration: Pitts S-1S, G-BOXV

No & Type of Engines: 1 Superior XP-IO-360-A1HC3 piston engine

Year of Manufacture: 1984 (Serial no: 7-0433)

Date & Time (UTC): 26 August 2022 at 0904 hrs

Location: Shobdon Airfield, Herefordshire

Type of Flight: Private

Persons on Board: Crew - 1 Passengers - None

Injuries: Crew - 1 (Fatal) Passengers - N/A

Nature of Damage: Destroyed

Commander’s Licence: Commercial Pilot’s Licence

Commander’s Age: 59 years

Commander’s Flying Experience: 1,978 hours (of which 530 were on type)
Last 90 days - 26 hours
Last 28 days - 8 hours
Information Source: AAIB Field Investigation

Synopsis

During an aerobatic practice flight, G-BOXV was seen to enter a climbing vertical rolling
manoeuvre from approximately 420 ft agl. The aircraft yawed right at the top of the manoeuvre
which apexed at approximately 1,100 ft agl. During the right yaw, an uncommanded
autorotative right roll developed and the aircraft entered a steep nose-down spiral dive. As
the pilot attempted to pull out of the ensuing dive, the aircraft experienced an accelerated
stall and a rolling departure to the right. At that point there was insufficient height remaining
in which to effect a safe recovery and the aircraft struck the ground. The pilot was fatally
injured in the accident.

No causal or contributory technical issues were identified during the post-accident

examination of the aircraft.

The investigation found that the entry conditions to the initial climbing manoeuvre gave little
or no safety margin when the aircraft began to dynamically diverge from the expected flight
path at the apex. Entering the manoeuvre with more height and/or speed would likely have
increased the pilot’s chances of avoiding the loss of control and/or being able to recover
from it safely.

Generic guidance for aerobatic pilots is contained in CAA Safety Sense Leaflet 19 –
‘Aerobatics’.1
Footnote
1
https://publicapps.caa.co.uk/docs/33/20130121SSL19.pdf [accessed 7 February 2022].

© Crown copyright 2023 67 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

History of the flight

The accident occurred on the first of two aerobatic practice flights the pilot had planned to
undertake in the overhead of Shobdon Airfield (Shobdon) on 26 August 2022. While the
second flight was to be a rehearsal of the display sequence the pilot intended to fly at a
private event on 28 August, the investigation did not find evidence as to the pilot’s detailed
intentions for the accident flight.

G-BOXV took off from Runway 26 at approximately 0900 hrs. The pilot turned left on
departure and climbed to position the aircraft south of the runway before commencing his
aerobatic manoeuvring. Mobile phone video taken by an eyewitness showed the aircraft
completing three distinct aerobatic manoeuvre combinations before, when approaching the
eastern end of the airfield, it pulled up into a vertical climb from approximately 420 ±50 ft agl
on a broadly easterly heading. While the aircraft did not stay in the video frame for all
the subsequent manoeuvring, it could be seen that the aircraft was rolled left through
approximately 450° as it climbed. The nature of the rolling motion indicated the pilot likely
had some left rudder applied during the roll because G-BOXV’s longitudinal axis was not
closely aligned with the aircraft’s upward flight path (Figure 1).

Figure 1
Pull up into vertical roll
(not to scale)

The aircraft continued climbing until reaching an estimated apex height of 1,100 ±200 ft agl.
At the top of the manoeuvre G-BOXV was banked to approximately 90° right wing low with
the nose 30°- 45° above the horizon. The nose of the aircraft then dropped progressively
lower while the bank was maintained (Figure 2). During the transition from nose-up to
nose‑down, G-BOXV’s nose appeared to fall more due to gravity than as the result of
significant rudder application generating the right yaw.

© Crown copyright 2023 68 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Figure 2
Apex of the climbing manoeuvre
(not to scale)

As the nose dropped through an estimated 45° nose-down, the aircraft began rolling right
and the nose dropped further to approximately 80° nose-down (Figure 3). The aircraft
continued rolling right, passing 360° of roll in approximately 2½ seconds. Audio recording
from the mobile telephone footage corroborated eyewitness evidence that the engine rpm
reduced, likely to idle, shortly after the aircraft began rolling during the descent.

Figure 3
Downwards flight path
(not to scale)

© Crown copyright 2023 69 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Video footage showed the aircraft stopped rolling after approximately 1¼ turns, at which
point it was pointing vertically down, if not slightly inverted. At this stage G-BOXV was
approximately 400 ft agl. As the roll stopped, the aircraft’s pitch attitude started to decrease,
and the pilot appeared to be attempting to pull out of the dive. This pitching motion continued
only briefly, stopping abruptly just before the aircraft diverged rapidly in roll to the right from
about 200-250 ft agl (Figure 3). The divergent right roll continued until the aircraft struck the
ground in an almost vertical attitude, just over two seconds later.

From the apex of the climb to impact with the ground took approximately 9 seconds, and
only 7 seconds from the roll starting as the aircraft’s attitude passed through 45° nose-down.

The airfield fire and rescue crew approached the accident site less than 3 minutes from
the alarm being raised and, shortly thereafter, began fighting what remained of the intense
post‑crash fire. The pilot did not survive the initial impact.

Accident site

The aircraft struck the ground in a recently harvested and drilled crop field approximately
130 m south of the threshold of Runway 26 at Shobdon (Figure 4). An intense post-impact
fire destroyed most of the aircraft.

Ground markings indicated that the aircraft struck the ground at a near vertical attitude,
with the upper wing facing east. It then bounced and came to rest, upright, with the aircraft
pointing in an east-south-east direction. The wings, which were of fabric covered wooden
spar construction, were consumed by the post-impact fire and the fabric that covered the
steel spaceframe fuselage was also consumed. One of the blades of the two-bladed fixed
pitch propeller had cut into the ground and fractured at the hub. This portion of the propeller
remained at the impact location; the other blade remained attached to the hub.

Continuity was confirmed for the aileron and elevator controls. The right rudder cable was
also continuous, but the left rudder cable was found to have fractured close to the pilot’s
seat.

Recorded information

No sources of recorded data were recovered from the aircraft. The aircraft was not tracked
by radar or other aircraft tracking networks.

The location of the cameras that recorded the three videos of the accident used in this
investigation are shown in Figure 4.

© Crown copyright 2023 70 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Figure 4
Sources of video recordings

CCTV cameras captured the lower parts of some of the manoeuvres and the impact with
the ground. The field of view of the CCTV cameras did not extend up enough to capture
higher parts of the manoeuvres flown, including the final accident manoeuvre. Figure 5
shows the final descent captured on CCTV.

A witness in a field to the north-east of the accident site recorded a video using their mobile
phone. The aircraft was not always in frame as the phone was panned but it did capture
most of the final manoeuvre. A difficulty with determining the flight path of the aircraft
from this video is that, during a large part of the manoeuvre, the background was entirely
made of cloud with few features to show how the camera was panning. Software tools
were used to pattern match large areas of cloud to model the camera orientation when
ground features were not in view. The limitations of this, and the assumptions required
for estimating distance to the aircraft when it was in view of only one camera, have been
accounted for in the error margins for the apex height stated earlier in this report.

© Crown copyright 2023 71 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Figure 5
Compound image of cropped CCTV snapshots 0.2 seconds apart until 0.8 seconds
before contact with the ground

The mobile phone video also captured the sound of the propeller (Figure 6). The frequency
of the recorded tone is related to the speed of the propeller but is also affected by doppler
shift due to the aircraft moving towards or away from the recording device. In this case the
frequency shift was predominantly due to the initial speed of the aircraft at the start of the
manoeuvre being largely towards the recording position. With fixed pitch propellers, such
as the one in use, the speed of the propeller is affected by airspeed and the throttle position.

© Crown copyright 2023 72 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

The audio recorded a sharp drop in propeller speed which, after factoring in the time taken
for the sound to travel from the aircraft to the mobile phone, was about six seconds before
impact. No sound of the impact was identified as the audio was swamped at this point by
voices at the recording location.

Figure 6
Spectrum analysis of the audio track of the mobile phone recording showing tone
due to propeller blade noise. Scale halved to reflect propeller rpm rather than blade
passes per minute

Aircraft information

The Pitts Special S-1S is a single seat, light aerobatic biplane built for competition
aerobatics. Its wings are of wooden spar and ribs construction, with the fuselage made of
a steel spaceframe. The wings and fuselage are covered in doped fabric. The aircraft has
conventional flight controls with ailerons positioned on both upper and lower wings.

G-BOXV was built in the USA in 1984 and transferred to the UK register in 1990. The
pilot purchased the aircraft in 2003, at which time it had accrued 230 flying hours.
In January 2019, the original Lycoming engine was replaced with a new
Superior XP‑IO‑360‑A1HC3 engine. The fixed pitch MT-propeller which had been fitted
in 2010 was retained. At the time of the accident the aircraft had flown 666 hours, with
the engine accruing 54 hours. The aircraft had a valid Permit to Fly which had been
revalidated in May 2022.

© Crown copyright 2023 73 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Aircraft examination

The aircraft wreckage was transported to the AAIB facility in Farnborough for further
examination. The fractured rudder cable was removed from the aircraft and examined
in a laboratory. This examination determined that the fracture had occurred because of
loading associated with impact-related airframe distortion in combination with heating from
the post‑accident fire and did not pre-exist the accident sequence.

Examination of the remainder of the aircraft, including the engine, found no indication of
damage that existed before the accident, but the extensive damage, caused by the intense
post-accident fire, meant a complete assessment of the aircraft was not possible.

Survivability

The pilot was wearing a parachute but, with limited time and height available to him from
the point at which the aircraft started rolling after the final manoeuvre apex, abandoning the
aircraft would not have been an option. The forces exerted on the pilot during the impact
resulted in injuries that were not survivable.

Weight and balance

The investigation was not able to ascertain the exact fuel load on board G-BOXV at the time
of the accident. Weight and balance calculations confirmed that the aircraft would have
been within its approved weight and CG envelope with the fuel tank full, empty or at any
level in between.

Aircraft performance

Spinning, autorotation and spiral dives

Spins are preceded by a stall which can be from straight or accelerated flight. Once
the wing has stalled, the phenomenon that develops and sustains a spin is autorotation.
Autorotation can be defined as a self-sustaining rotational motion, initially in roll but may
result in significant yaw depending on the nature of the spin, for example, in a flat spin the
autorotation would be wholly yaw.

The main differences between a spin and a spiral dive are that spins can be erratic as they
develop, and they are associated with a low indicated airspeed and significant yaw. Spiral
dives tend to have higher and increasing airspeeds, low levels of yaw and, with the aircraft
not being in a stalled condition, they are generally smoother than a spin.

Pitts Special spinning characteristics

The following information regarding height loss during spinning had been provided to a
previous AAIB investigation into an accident involving a two-seat Pitts Special aircraft
(G-ODDS2) in 2019.

Footnote
2
AAIB investigation to Pitts S-2A Pitts Special, G-ODDS - GOV.UK (www.gov.uk)
[accessed 9 November 2022].

© Crown copyright 2023 74 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

‘[assuming] that a conventional technique to induce and maintain a spin was

used, ie full rudder and control column held fully back with ailerons neutral…
the manufacturer indicated that a 10-turn upright spin incurred a height loss of
3,400 ft in an elapsed time of 32 seconds. Therefore, each spin rotation takes
approximately three seconds and incurs a loss of 340 ft with a rate of descent
of approximately 6,800 fpm.

The manufacturer advised that, with full opposite rudder deflection and neutral
(or released) control column, it would take approximately 500 ft to stop the
rotation and then another 500 ft to level flight with a 4 g acceleration.’

The G-ODDS report also contained the following observation regarding spin recovery
technique:

‘…if in-spin aileron was maintained during the recovery the aircraft could
potentially enter another spin, possibly inverted, in the opposite direction.’

Pitts Special pilots who spoke to the G-BOXV investigation reported that, in an erect spin,
the aircraft would adopt a “relatively flat” 30°-50° nose-down attitude while the airspeed
would remain “low and stable.” The manufacturer’s information indicated that it would take
500 ft to stop the rotation from a fully developed spin at 6,800 fpm rate of descent. One
pilot reported that, for a single turn spin before the rate of descent had built significantly, he
found it possible to effect recovery to level flight in approximately 500 ft from initiating spin
recovery action.

Meteorology

The weather at the time of the accident was benign. There was good visibility with a distinct
horizon at low level, the wind was calm, and the cloud base was broken3 at 2,500-3,000 ft.

Airfield accident response

Because aerobatic practices in the airfield overhead were not routinely permitted at Shobdon,
the local procedures had not included any requirement for an enhanced level of standby
posture for the on-site fire and rescue assets. While not included in the airfield procedures,
shortly after G-BOXV took off, the Airfield Manager and on-duty Flight Information Service
Officer (FISO) independently thought it prudent to put the fire crew on ‘local standby’ as
they would for a first solo flight. Consequently, the lead fire fighter had already donned
protective clothing and was able to board the response vehicle, parked in front of the ATC
building, within one minute of the alarm being raised. The second firefighter saw the impact
while he was mowing the grass at the western end of the airfield and immediately drove
back to join the rescue vehicle. The fire crew arrived on scene within three minutes of the
accident occurring. In light of this accident, it was decided that the airfield fire and rescue
service would, in future, be brought to immediate readiness for any aerobatic practices in
the overhead as well as for first solo flights.

Footnote
3
Five to seven eighths coverage.

© Crown copyright 2023 75 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

The combined operations team had recently updated the airfield incident response plan.
Their previous experience was that, in stressful situations, standard sequential checklists
were not always easy to follow, and their unidirectional flow pattern meant that, if steps in
the process were missed, the slip was less likely to be caught and rectified. As a counter
to this, they had developed a three-phase matrix response aide-memoire with key action
priorities for each phase forming a circular flow chart (Figure 7).

Figure 7
Overview of airfield incident response matrix (used with permission)

Within each action priority area, amplifying notes were provided as further prompts to
the person coordinating the response. The matrix had been successfully trialled during
a recent simulated emergency at Shobdon. Airfield personnel judged that the revised
incident response matrix as well as their recent training had left them well prepared for the
challenges posed by this accident.

Personnel

The pilot was a commercial pilot’s licence holder with a Flying Instructor rating. He was an
Intermediate Category competition aerobatics pilot and a qualified Upset Prevention and
Recovery Training (UPRT) instructor. Before the accident flight, he last flew G-BOXV in an
aerobatic competition on 21 May 2022 and had flown one further aerobatic practice in the
aircraft on 9 July 2022. Between 9 July and 26 August, the pilot had undertaken 13 UPRT
training flights as the instructor in a Slingsby T67M-200 Firefly aircraft.

The pilot’s Class 1 aviation medical was valid, and the post-mortem could find no evidence
of any chronic or acute medical condition that might have been causal or contributory to the
accident.

© Crown copyright 2023 76 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Other information
Planned aerobatic manoeuvres/sequence

Documentation provided to the investigation indicated that, for his second planned flight,
the pilot intended anchoring his display over a datum south of the runway (Figure 8). During
the accident flight, the pilot began aerobatic manoeuvring south and west of this display
area. The way the manoeuvres progressed suggested the pilot could have been using
them to warm up while re-positioning the aircraft closer to his intended datum.

Figure 8
Primary display axis and datum for the pilot’s second planned flight
(Imagery ©2023 Bluesky, Infoterra Ltd & COWI A/S, CNES / Airbus, Getmapping plc, Maxar Technologies)

The investigation consulted several pilots who had expert knowledge of the aircraft type
and/or knowledge of the pilot’s handling style and approach to flying to try and establish
where the accident manoeuvre diverged from the pilot’s intent.

The AAIB obtained evidence of two different aerobatic sequences the pilot was known to fly,
one for aerobatic competitions and one for displays. Neither written sequence correlated
to the sequence of manoeuvres captured on the mobile phone video of the accident flight.

The closest comparable manoeuvres to that immediately preceding the accident were the
‘Avalanche’ and ‘Reverse half Cuban’ from his display sequence and a 180° vertical roll
followed by a 1½ turn spin from his competition sequence (Figure 9). Annotations on a
copy of the pilot’s competition sequence indicated he used 2,500 ft agl as a target entry
height for the spin and that, in contrast to the accident manoeuvre, it was preceded by
decelerating level and erect flight to generate the required stall conditions for entry. While
all three manoeuvres began with climbing rolls ±45° from the vertical, none of them included
a 450° upwards roll followed by a tight descending spiral.

© Crown copyright 2023 77 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Figure 9
Closest comparable manoeuvres from the pilot’s known aerobatic sequences
(not to scale)

While no supporting documentation could be found, the investigation was provided with
video footage taken on 1 August 2020 of the pilot flying a climbing manoeuvre like that which
preceded the accident. This video showed the aircraft pulling up into the vertical before
rolling left through 450°. This roll, like the one in the accident manoeuvre, was off‑axis to
the left and finished with a steeply banked ‘knife-edge’ over the top. As the aircraft’s nose
dropped below the horizon, the pilot held the pitch attitude at approximately 45° nose-down
with the aircraft inverted for about one second before rolling erect and continuing the 45°
descent. In the August 2020 video, the transition from nose-up to nose-down was more
dynamic than on the accident flight; visually, the aircraft appeared to have more airspeed
approaching the apex and the right yaw looked more positively controlled. Additionally, on

© Crown copyright 2023 78 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

the accident flight the aircraft remained close to 90° angle of bank as the nose dropped,
while for the August 2020 manoeuvre, the bank angle was closer to 120°, thus requiring less
yaw but more pitch to bring the nose down to the desired angle. A simplistic comparison of
the accident manoeuvre with the one from 1 August 2020 is shown at Figure 10.

Figure 10
Comparison of 1 August 2020 manoeuvre and accident manoeuvre
(not to scale)

Background noise from another aircraft on the August 2020 video interfered with G-BOXV’s
engine note, thus making any spectral audio analysis inconclusive. It was not possible to
determine if the August 2020 manoeuvre was flown with a higher power setting than on the
accident flight.

The accident pilot normally flew to a minimum 1,000 ft agl base height during competition
flying and would use 500 ft agl as his minimum height for displays.

UPRT syllabus

Multiple individual manoeuvres would be demonstrated and practised on each of the

three flights comprising the UPRT airborne syllabus that the pilot regularly delivered. A
consolidated list of the individual exercises for each flight is reproduced at Table 1 from
a copy of the pilot’s kneeboard aide-memoire that he used when teaching the course.
Elements with direct read across to the accident manoeuvre are shown in bold.

© Crown copyright 2023 79 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

UPRT Exercise 1 UPRT Exercise 2 UPRT Exercise 3

▪ Steep turns ▪ Energy trading demonstration ▪ Recovery from nose high upsets
▪ Unaccelerated stall ▪ Zero G flight to demonstrate flight by centring controls
▪ Slow flight below VS with no stall symptoms ▪ Stalling in balanced turn
▪ Spiral dive ▪ Jammed controls: ▪ Stalling in skidding turn
▪ Recovery from nose low ▪ Ailerons ▪ Demo ineffectiveness of ailerons
upset ▪ Elevator to counter wing drop
▪ Stalls: nose high/low, ▪ Recovery from nose high upset ▪ Vertical accelerated stall
secondary ▪ Roll/yaw to lower nose and ▪ Practice recovery from upsets
▪ Maximum rate turn recover from nose high upset various attitudes: surprise &
▪ Stall in maximum rate turn ▪ Demo UPRT3 aerobatic startle
manoeuvres ▪ Nose high
▪ Recovery from inverted flight ▪ Nose low
▪ Recovery from incipient spin ▪ Stalls (1G and accelerated)

Table 1
Consolidated reproduction from the pilot’s kneeboard
aide-memoire for the UPRT syllabus

Analysis

At the time of the accident G-BOXV had a valid Permit to Fly and was operating within
the manufacturer’s defined weight and balance envelope. Although extensive fire damage
prevented a detailed reconstruction of the aircraft, examination of the wreckage identified
that, prior to the accident, the primary flying controls were correctly connected and free from
restriction. The engine was in good condition with no indications of low or poor performance.
It is therefore likely that there were no technical issues with the aircraft which affected its
ability to fly normally during the accident flight.

The accident pilot was correctly licenced and qualified for the intended flight. In his role as
an UPRT instructor, albeit in pre-planned training scenarios, he regularly demonstrated how
to recover an aircraft from unusual attitudes, incipient spins, spiral dives, and accelerated
stalls. He was an experienced aerobatic pilot who had successfully flown G-BOXV in
competitions and public displays for many years.

The investigation did not find evidence of the pilot suffering from any chronic or acute
medical issue that might have been causal or contributory to the accident.

Accident manoeuvre

From analysing the video evidence taken in August 2020 and August 2022, and in the
absence of the pilot’s known aerobatic sequences containing any comparable manoeuvres,
the investigation deemed it likely the pilot was attempting a similar manoeuvre to that seen
in the 2020 video. If that was the case, the aircraft departed from the pilot’s intended flight
path as the nose dropped below the horizon at the top of the final climb.

© Crown copyright 2023 80 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

Based on witness testimony, if the pilot had intended to spin the aircraft after the climb,
it is likely he would have planned on doing so higher and from level flight, using a more
controlled and conventional technique. Given the Pitts Special’s spinning characteristics
as explained to the investigation, of low sustained airspeed and comparatively shallow
nose‑down attitude, the investigation determined that G-BOXV entered an autorotative
spiral dive after the final apex, rather than a spin.

Visually, the apex of the accident manoeuvre was less dynamic than that seen on the
August 2020 video. This suggests that the aircraft was slower as it transitioned from climb
to descent. The reasons for this could be one or a combination of the following (when
compared with the August 2020 manoeuvre):

● A larger rudder pedal input during the upward roll leading to greater off-axis
yaw angle and therefore higher resultant drag and faster speed decay.

● A lower power setting leading to faster speed decay in the climb.

● A lower entry speed leading to lower airspeed at the apex.

● A slower rate of roll meaning it took longer to complete 450° roll, thereby
resulting in a higher than intended climb and slower apex airspeed.

The low apex airspeed would have reduced the pilot’s aerodynamic control over the
aircraft’s flight path leaving him less able to positively position the aircraft as it transitioned
from nose-up to nose-down. With little or no observed pitch rate, the aircraft would have
had a low angle of attack and did not appear to be stalled over the top of the manoeuvre.
Being close to 90° angle of bank, the aircraft was yawing right as the nose dropped so
the left (outer) wing would have been moving faster and producing more lift than the
right wing. This resulting aerodynamic asymmetry appears to have developed into an
autorotative right roll.

With the right wing producing less lift, its ailerons would have been unable to generate
enough counterbalancing rolling force, even assuming the pilot had applied full left aileron
to oppose the roll. The autorotative roll developed rapidly and the aircraft’s nose dropped
steeply as it did so. In an un-stalled condition and subject to autorotation, G-BOXV quickly
became established in a steep nose-down spiral dive with the aircraft accelerating despite
the pilot’s apparent selection of idle power.

As the aircraft’s speed increased in the descent, its ailerons would have become more
effective, and the pilot managed to stop the roll after approximately 1¼ turns. However, by
the time the roll stopped, the aircraft was very low and in a steep nose-down attitude. That
the aircraft started to pitch out of the steep dive led the investigation to conclude the pilot
was active on the controls throughout the attempted recovery. With limited height remaining
and a high rate of descent, it is likely that the rapidly approaching ground prompted the
pilot to pull as hard as possible to recover from the dive. The observed sudden reduction
in pitch rate was indicative of an accelerated stall as the g-loading increased. The rapid
roll divergence could have resulted from a residual rudder or aileron input at the point of

© Crown copyright 2023 81 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

the stall. Based on the video evidence, even if the initial pitch rate acceleration had been
maintained, it is unlikely recovery could have been completed successfully in the remaining
height available.

Height considerations

The investigation was not able to determine why the aircraft appears to have been committed
to the vertical climb from below the pilot’s reported minimum base height for aerobatic
manoeuvring.

Based on the manufacturer’s flight trials and pilot reports, the minimum height loss in a
single turn spin followed by an expeditious recovery to level flight would be somewhere
between 500 and 1,000 ft agl. Even from 1,300 ft agl, the upper tolerance of the
photogrammetry‑derived apex, there would have been little or no contingency height for a
single turn spin and recovery if working to an assumed 500 ft base height. The pilot’s notes
on his competition sequence indicated he used 1,500 ft above base height as the target
entry height for a 1½ turn spin. The investigation concluded that the pilot had not intended
to spin after the vertical rolling manoeuvre.

A spin is a stalled manoeuvre with a relatively low nose-down attitude and low airspeed
when compared with the spiral dive experienced by G-BOXV. The steeper attitude and
increasing airspeed, despite the engine being at low power, would have resulted in a higher
rate of descent than if spinning. Based on the time between apex and impact, the estimated
average rate of descent would have been 8,700 fpm from a maximum height of 1,300 ft agl
and 6,000 fpm from 900 ft agl4. Taken from the start of the autorotative roll these figures
would be 1,000-1,500 fpm higher. The investigation thought it unlikely the pilot intended
entering a steep spiral dive after the manoeuvre apex.

Without supporting evidence, the investigation was unable to determine if any of the
aerobatic manoeuvres seen on the accident flight were flown with specific entry and safety
parameters in mind. One expert witness observed that the manoeuvres appeared to follow
a less structured flow than he would have expected from a planned sequence.

Conclusion

The accident occurred after control was lost when an autorotative roll developed as the
aircraft yawed at the top of a vertical climb. Low airspeed during the yaw would have
reduced the aerodynamic control available to the pilot such that he could not prevent the
aircraft entering the subsequent spiral dive. The entry conditions to the manoeuvre gave
little or no safety margin when the aircraft began to dynamically diverge from the expected
flight path. While the pilot was able to regain control of the aircraft, by the time he did so
there was insufficient height remaining in which to effect a safe recovery.

While the investigation could not determine why the pilot was unable to prevent the aircraft
from entering the spiral dive, starting the climb with more height and/or speed would likely
Footnote
4
Based on the photogrammetry-derived apex height of 1,100 ±200 ft agl.

© Crown copyright 2023 82 All times are UTC

AAIB Bulletin: 6/2023 G-BOXV AAIB-28594

have increased the pilot’s chances of avoiding the loss of control and/or being able to recover
from it safely. The investigation was not able to determine what the pilot’s contingency
criteria were, but this accident serves as a reminder that conducting low level aerobatics
comes with inherent risks when manoeuvres, planned or unplanned, do not proceed as
expected.

Generic guidance for aerobatic pilots is contained in CAA Safety Sense Leaflet 19 –
‘Aerobatics.’

Published: 3 May 2023.

© Crown copyright 2023 83 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

ACCIDENT

Aircraft Type and Registration: Flight Design CT2K, G-CBDJ

No & Type of Engines: 1 Rotax 912ULS piston engine

Year of Manufacture: 2001 (Serial no: 7850)

Date & Time (UTC): 24 March 2022 at 1400 hrs

Location: Beccles Aerodrome, Suffolk

Type of Flight: Private

Persons on Board: Crew - 1 Passengers - None

Injuries: Crew - 1 (Fatal) Passengers - N/A

Nature of Damage: Aircraft destroyed

Commander’s Licence: Private Pilot’s Licence

Commander’s Age: 87 years

Commander’s Flying Experience: 2,677 hours (of which 1,621 were on type)
Last 90 days - 20 hours
Last 28 days - 9 hours

Information Source: AAIB Field Investigation

Synopsis

The aircraft was on a flight from Temple Bruer airstrip, Lincolnshire to Beccles Aerodrome,
Suffolk. The approach was described as “unstable”. The aircraft bounced on landing and
probably stalled. The pilot was fatally injured when the aircraft subsequently struck the
ground.

The pilot was familiar with his aircraft and in recent practice, but the landing diverged from
his intended plan. Given that he was 87 years old and recognised that he would likely have
to stop flying in the near future, it is possible that some age-related deterioration in human
performance was a factor in this accident. The investigation highlighted a lack of medical
guidance for both pilots and medical professionals, as well as a cohort of private pilots who
are not subject to an independent professional assessment of age-related deterioration in
piloting ability. Four Safety Recommendations have been made to the CAA, three about the
Pilot Medical Declaration and one about the revalidation of ratings.

History of the flight

The pilot was a member of a small aviation group that flew from Temple Bruer airstrip,
Lincolnshire, which is located just outside the northern Aerodrome Traffic Zone (ATZ)
boundary of RAF Cranwell. On the day of the accident, he planned to fly from Temple Bruer
to Beccles Aerodrome, Suffolk, which he had visited on two previous occasions.

© Crown copyright 2023 84 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

The weather for the route was clear with light winds from the east. The pilot booked out at
Temple Bruer and took off at 1246 hrs, climbing on a south-easterly heading. Three minutes
later the pilot informed Cranwell ATC that he was returning to Temple Bruer with a “slight
problem”. Approximately one minute later he informed ATC that the problem had been
solved and that he was continuing towards Beccles.

The pilot continued on a south-easterly course, operating at altitudes up to 3,500 ft amsl.

At 1350 hrs he contacted Beccles Radio requesting the airfield details; the radio operator
responded that Runway 09 was in use and that the wind was from the east at less than five
knots. The pilot positioned to join downwind then established on a final approach, reporting
both positions on the radio. The radio operator replied to the final call, passing the wind
direction and speed.

Eyewitness accounts described the final approach of G-CBDJ as “unstable” in roll and
pitch initially, then becoming stable before touching down on its main wheels approximately
50 m in from the threshold. The aircraft was observed to bounce to around 10 ft into the
air before touching down again on the nosewheel. It bounced a second time and pitched
markedly nose up, described as being to 45° as it appeared to climb away. On reaching
around 100 ft above the runway, witnesses described the aircraft veering to the left and
rolling to approximately 90° angle of bank before the nose “dropped” and the aircraft fell to
the ground in a field adjacent to the runway. No witnesses interviewed by the AAIB could
recall hearing sounds of the aircraft’s engine increasing power after either landing attempt.
There were no reports of turbulence on final approach on the day of the accident.

The aerodrome Rescue and Fire Fighting Service (RFFS) vehicle arrived quickly at the
aircraft and found the pilot in the left seat with the safety harness in place, breathing but
unresponsive. Recognising that they would not be able to release him safely from the
wreckage, the RFFS crew continued to provide reassurance to the pilot and monitored the
aircraft for signs of fire until further assistance arrived. They established that the aircraft
was fitted with a Ballistic Parachute Recovery System (BPRS). An ambulance arrived at
1404 hrs followed by the Fire Service at 1412 hrs. The Helicopter Emergency Medical
Service (HEMS) then arrived at 1420 hrs. The HEMS doctor determined that the pilot had
succumbed to his injuries and died at the scene.

Accident site

The aircraft came to rest approximately 100 m to the north of Runway 09 and 240 m from the
threshold. Both wings had detached from the fuselage, with the lower surfaces upper most.
The fuselage was on its right side, on top of the wings with severe disruption to the nose
section (Figure 1). The field to the north of the runway was planted with a rapeseed crop
which was approximately 1 m in height. The crop was heavily damaged in the immediate
area of the aircraft and there was a short visible swath cut into the crop (highlighted in
Figure 1 right) that was consistent with the left wingtip being the first part of the aircraft to
make contact with the field.

© Crown copyright 2023 85 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Figure 1
Aerial view of the accident site

Aircraft information

G-CBDJ was a Flight Design CT2K microlight which was built in 2001. It had an
all‑composite construction with a high wing, conventional control surfaces and a tricycle
landing gear. A three-bladed propeller was driven by a Rotax 912ULS engine. The
aircraft was fitted with a BPRS. Prior to recovering the aircraft to the AAIB, the BPRS was
deployed at the accident site with the necessary safety precautions in place.

A Brauniger Alpha Multi-Function Display (MFD) was fitted to the centre of the instrument
panel in the cockpit of G-CBDJ. The MFD used a monochrome Liquid Crystal Display
(LCD) to provide the pilot with the following information: fuel quantity, the aircraft’s indicated
airspeed, altitude, vertical speed, engine rpm, engine oil temperature, water coolant
temperature, exhaust gas temperatures and oil pressure.

The airspeed and vertical speed were presented as analogue indicators; the altitude,
engine speed, temperatures and pressure were displayed as numerical values; and the fuel
quantity as a bar graph.

The MFD provided a visual alert to the pilot if the airspeed or engine parameters exceed
set limits. For example, the alert thresholds for aircraft indicated airspeed for G-CBDJ was
38 kt. If an engine exceedance was detected, or the airspeed was below the set limit, the
associated display readout on the LCD would flash on and off.

If external electrical power to the MFD was lost in flight, a back-up battery installed within
the MFD automatically provided electrical power that enabled the unit to continue to operate
for several hours. The voltage of the integral battery was tested by the MFD each time
external electrical power was applied to the unit. If the voltage was detected as being low,
a warning was presented on the MFD.

Aircraft handling

Pilots experienced on the CT2K informed the AAIB that in common with many types, it
had characteristics that required vigilance from the pilot. For example, if a pilot lands on
the nosewheel, the nose of the CT2K tends to “kick up” markedly, requiring the immediate
application of power to go around. One pilot stated that, “it is a difficult aircraft to land,

© Crown copyright 2023 86 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

especially if it bounces...you have to apply power to catch it...you have to be on your

game”.

The maximum speed to extend the flaps on the CT2K is 62 KIAS and the recommended
approach speed is 55 KIAS.

Recorded information

Recorded information for the accident flight included the aircraft’s position and altitude,
which had been recorded by ground-based radar and equipment that had received
transmissions from an electronic conspicuity device1 fitted in the aircraft. RTF recordings
of communications between the pilot and ATC, that commenced as the aircraft departed
Temple Bruer airstrip, were also available.

When electrical power was applied to the MFD on G-CBDJ after the accident, the integral
battery passed the unit’s built-in test. The MFD had an internal recording function which
monitored engine rpm, altitude and airspeed to determine the start and end of a recording
period. Each recording started once the engine had been running for one minute and the
MFD detected an increase in altitude of about 75 ft, indicating that the aircraft had taken
off. The recording ended once the engine had stopped with an airspeed less than about
27 kt and with no change in altitude detected for a subsequent period of approximately
30 seconds. The MFD recorded flight duration reflected the time that the aircraft had
climbed above 75 ft and the airspeed remained above 27 kt.

Recorded information was recovered from the MFD for the accident flight and the 23 previous
flights dating back to 14 November 2021. This data provided a peak value of airspeed,
altitude, vertical speed, and engine parameters (except oil pressure). It was not possible to
determine at what point peak values occurred during a flight.

A portable tablet computer was also found in the aircraft. This was damaged and no data
was recovered. A member of the pilot’s family confirmed that the device was operating a
navigation application. This provided a moving map with the aircraft GPS derived position,
and a route could be entered between points, such as when flying between airfields.

Summary of recorded data

The pilot of G-CBDJ made initial radio contact at 1244 hrs with Cranwell Zone ATC
to request permission to depart from Temple Bruer into the Military Air Traffic Zone
(MATZ), for a flight to Beccles (Figure 2). The pilot was cleared for departure, with the
aircraft subsequently taking off from Runway 08 at 1246 hrs before then turning onto
a south‑easterly heading. However, at 1249 hrs, the pilot reported to ATC that he had
a “slight problem” and was returning to Temple Bruer, with the aircraft making a left
turn back towards the airstrip. ATC inquired if they could further assist the pilot, who
responded by advising “no, no i’m, i’m fine thank you, it’s just something hasn’t
fired up as it should have done”. About one minute later, at 1250:54 hrs, the pilot
Footnote
1
PilotAware Rosetta.

© Crown copyright 2023 87 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

contacted ATC and advised them “problem solved” and that he was turning back on
course towards Beccles. A family member of the pilot informed the AAIB that the pilot
used this phrase when his navigation application did not activate the planed route, or did
not connect correctly to the PilotAware Rosetta conspicuity device.

At 1258 hrs, G-CBDJ exited the Cranwell MATZ and the pilot advised ATC that he was
changing to “safetycom”2 frequency enroute. This was the last recorded communication
from the pilot; there were no reports of the pilot communicating whilst enroute and the air/
ground radio communications at Beccles was not recorded.

As G-CBDJ approached the town of Spalding, the pilot altered course to fly an almost
direct track to Beccles, during which the aircraft was operated at altitudes of up to about
3,500 ft amsl.

At 1352 hrs, G-CBDJ had descended to an altitude of about 1,000 ft amsl and was joining
crosswind for a right-hand circuit to land on Runway 09 at Beccles. The final recording of
the aircraft was at 1354:58 hrs when it on the final approach and positioned 800 m from the
runway threshold at an altitude of about 430 ft amsl (a height of 350 ft aal), which equated to
a flight path angle of 7° to the runway threshold. The aircraft’s descent rate at this time was
about 450 ft/min and its calculated airspeed, based on a windspeed of 5 kt, was between
60 and 70 KIAS.

The MFD data for the accident flight indicated that the engine had been running for one hour
and eighteen minutes, and that the recorded flight time was one hour and ten minutes. The
engine was started at 1236 hrs and it had stopped between 1355 and 1356 hrs; this was
consistent with the time for the aircraft to have reached the airfield from the final radar
position based on its groundspeed. Table 1 provides the peak values recorded by the MFD
during the accident flight. None of these values exceeded a threshold for an alert.

Engine KIAS Altitude Rate of Rate of Water Oil EGT

rpm kt ft climb descent temperature temperature Sensor 1 /
ft/min ft/min °C °C 2 °C
4,900 110 3,466 1,100 1,200 113 103 814 / 773

Table 1
Peak MFD parameter values recorded during the accident flight

Footnote
2
SafetyCom is a common traffic advisory frequency (135.480 MHz) for use at, or near to, aerodromes that do
not have an assigned frequency.

© Crown copyright 2023 88 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Figure 2
Aircraft track from Temple Bruer to Beccles Aerodrome
© 2022 Google, Image © Landsat / Copernicus

Aircraft examination

The wreckage was recovered to the AAIB for detailed examination. Continuity of the flying
controls was confirmed, along with the engine controls. Witness marks on the fuselage
correlated to the flaps being deployed to their full (landing) configuration at the time of the
accident. There was evidence of over compression on the left side of the nose landing
gear tyre and there was slight deformation of the wheel hub (Figure 3). A detailed engine
teardown revealed that the engine was probably performing normally prior to the accident.
from performing normally.

Figure 3
Damage to the nose landing wheel

With the exception of the MFD, the disruption to the wreckage precluded detailed testing of
the aircraft systems.

© Crown copyright 2023 89 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Pilot information

The pilot held a UK Private Pilot’s Licence, first issued in 1989, and had flown 2,677 hours
of which 1,621 hours were in G-CBDJ. He gained a microlight endorsement in 1991 and
revalidated the privileges of his microlight class rating on 6 April 2021 by providing evidence
of his experience. His aviation skills were well regarded by his peers at Temple Bruer.
They commended his skills in dealing with turbulent conditions at their flying strip, and his
awareness of how his aircraft handled in such conditions.

A member of the pilot’s family informed the AAIB that the pilot was getting to the stage
where he was “slowing down” and that the pilot was “coming to the conclusion that he would
have to stop flying in the near future”.

Post-mortem report

In his post-mortem report, the pathologist found that the pilot died from multiple traumatic
injuries sustained in the aircraft accident. There was no indication of medical impairment
or incapacitation of the pilot before the aircraft struck the ground, but this could not be ruled
out.

Licence revalidation

There are two methods of revalidating a microlight class rating issued before
1 February 2008 for a UK PPL holder; the first is by experience3. The requirement is
that within the preceding 13 months and during the validity of an existing Certificate of
Experience, the holder must have completed at least five hours of flying as a pilot of a
microlight, including at least three hours as Pilot in Command4. This method of revalidation
does not require a training flight or assessment with an instructor. The pilot of G-CBDJ
had flown approximately 62 hours in the preceding 13 months.

The second method, which applies to all microlight class ratings issued after
1 February 2008 and for pilots with a National Private Pilot’s Licence, is that within the
24-month period of validity of a Certificate of Revalidation, the holder must have completed:

● A minimum total of 12 hours flight time5.

● At least one hour of flight training in a microlight aircraft conducted by an

instructor entitled to give flight training in a microlight aircraft.

Pilots holding a UK PPL with a microlight class rating issued before 1 February 2008 may
choose to adopt the NPPL revalidation scheme but there is no requirement for them to do

Footnote
3
This revalidation route also applies to class ratings for Self-Launching Motor Gliders.
4
Up to two hours can be conducted as dual flying instruction as Pilot Under Training with a qualified flying
instructor who has certified that he/she was fit to act as pilot in command.
5
It also requires the following:
• Eight hours flown as PIC.
• Twelve take offs & landings.
• At least six hours flown within the 12 months preceding the expiry date of the current certificate.

© Crown copyright 2023 90 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

so. The BMAA informed the AAIB that it was not possible to quantify either the number of
pilots who renewed their microlight class rating by experience, or the number who could
exercise that renewal method but had opted to comply with the NPPL scheme.

Ageing pilots

In April 2021 the AAIB published a report into a fatal glider accident, G-CFST, involving a
91-year-old pilot6. The report considered data from the CAA that showed the average age
of non-commercial pilots in the UK was increasing. In 2000 it was 43.7 years; by 2018 it
had increased to 52.2 years. After a review of available literature, the G-CFST investigation
found that:

‘Older pilots are not necessarily less-safe pilots and poor decision making
can affect pilots of all age and experience levels. Nonetheless, age-related
deterioration in eyesight, hearing, mobility, memory, cognition and decision
making are recognised as having an impact on piloting ability.’

And that:

‘Although the broad effects of ageing are well known, there is great variability
on how any specific decline will affect an individual pilot and chronological age
is not a reliable metric to predict age-related impairment…while experience,
knowledge, aptitude and wellbeing can offset or delay the effects of ageing,
there will inevitably come a point where the most sensible option for an individual
is to retire from flying as PIC.’

The investigation concluded that:

‘Unless precipitated by an accident or incident, without an objective metric for

making the decision, it relies on individual pilots to be honest with themselves
and for supervisors to be candid enough to reach a shared acknowledgement
that their days as PIC are over. Family, friends and peers can play a part in
encouraging and supporting pilots when that decision has to be made. This is
especially important for pilots not affiliated to clubs or sporting associations.’

Medical requirements for General Aviation pilots

All pilot licences require a medical certificate or declaration of some description. Holders
of a UK PPL or NPPL wishing to fly a UK (G) registered aircraft in UK airspace can apply
for a Pilot Medical Declaration (PMD) using the CAA’s medical online system. The online
PMD was introduced in October 2016 following a public consultation7 in which the CAA
sought the General Aviation (GA) community’s opinion on adopting the Driver and Vehicle
Licensing Agency (DVLA) standard for Group 1 (car) Ordinary Driving Licences (ODL)
as the medical standard for their sector, expanding on a scheme previously available to

Footnote
6
AAIB investigation to Schleicher ASH 25 E, G-CFST - GOV.UK (www.gov.uk) [accessed February 2023]
7
See CAP 1284, UK Private Pilot Licence and National Private Pilot Licence Medical Requirements, CAA,
June 2015.

© Crown copyright 2023 91 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

NPPL holders8. This was part of the CAA’s wider aspiration to make regulation ‘more
proportionate and less burdensome, while still seeking to protect third parties’, and
focused the risk analysis on the probability of serious incapacitation in flight.

The current PMD is an affirmation of a pilot’s medical fitness to fly based on a ‘reasonable
belief’ that they meet the medical requirements for a Group 1 ODL and are not subject to any
disqualifying medical conditions. There is no requirement to consult a General Practitioner
(GP) or an Aeromedical Examiner (AME), and the pilot does not have to hold a driving
licence. Essentially, if you believe you are fit enough to drive to an aerodrome, you may
consider yourself to be fit to fly your aircraft. In addition, the applicant must comply with the
following requirements:

● Only fly an aircraft no greater than 2,000 kg MTOW.

● Must not be taking medication for any psychiatric illness.

Additional medical restrictions apply if the applicant wishes to fly aircraft greater than
2,000 kg but below 5,700 kg MTOW, and an AME must be consulted if the pilot is unsure
about the applicability of a condition, treatment or medication9.

PMD validity

After completing the online process, the PMD is valid until the age of 70 years, with
no upper age limit, unless voluntarily withdrawn for medical reasons. After this, a new
declaration must be submitted every three years. The accident pilot renewed his PMD on
21 December 2021, hence valid until 21 December 2024.

Assessment of ongoing medical fitness

The CAA’s website provides the following guidance for the assessment of ongoing fitness
for PMDs:

‘If you have reason to believe you no longer meet the DVLA Group 1 ODL
[medical] standard, or suffer from any of the specified medical conditions, you
must not fly and must withdraw the declaration…’

The CAA does not provide further guidance on the DVLA’s medical standard for Group 1
driving licences, nor does it provide a link to the DVLA’s website10 which contains
comprehensive advice to drivers. A pilot who intends to make a PMD must actively seek
out information relevant to his personal medical history from DVLA sources.

Footnote
8
The NPPL was established using a declaration of medical fitness by the pilot, which was then countersigned
by their General Practitioner, who had access to the pilot’s medical records.
9
Further information is available at https://www.caa.co.uk/general-aviation/pilot-licences/applications/
medical/medical-requirements-for-private-pilots [accessed May 2022].
10
https://www.gov.uk/driving-medical-conditions [accessed May 2022].

© Crown copyright 2023 92 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Collating evidence following the implementation of PMDs

As part of the implementation phase of the introduction of PMDs the CAA published
CAP 1397 - Comment response document: UK Private Pilot Licence and National Private
Pilot Licence medical requirements, in April 2016, in which it stated:

‘It will be important to collect evidence post-implementation to confirm the safety

analysis assumptions.

A record keeping system will have to be established to monitor the effects of

implementing the new proposal. This will request private pilots submit information
on an annual basis to the CAA documenting such items as: age, type of flying,
hours flown in last year, total hours.’

The CAA informed the AAIB that following the introduction of PMDs, no record keeping
system had been established to collect the annual data necessary to enable the validation
of the system.

In October 2020 the CAA conducted a post-implementation review of the PMD process and
looked at a sample of 800 PMD holders out of a total of 14,400. This sample comprised
400 pilots who had previously had a medical status of ‘unfit’ or had a medical referred,
and 400 who had no previous unfit or referred status. It found that 4% of the 800 pilots
reviewed should not be self-declaring their medical status for various reasons. The CAA
concluded that:

‘If this percentage represents the whole number of PMD holders, the number
made in error is of concern to the CAA. These errors included disqualifying
heart conditions, neurological conditions and drug/alcohol misuse. It is not clear
to us whether this is due to unclear guidance material or a misunderstanding on
the pilot’s part.’

In parallel with this review, anticipating the UK’s departure from EASA at the end of 2020, the
CAA launched a consultation on opportunities for change for the UK’s GA sector11. One of
the initiatives identified was a review of the PMD process in order to ‘enhance the end user
experience and identify opportunities in the context of the simplification and rationalisation
of GA flight crew licensing’.

Consequently, the CAA launched a further public consultation in October 202212 to revisit
the questions originally asked, prior to the launch of PMDs, to provide guidance for future
development of the scheme and to establish whether any changes needed to be made.

Footnote
11
Published as CAP 1985: UK General Aviation opportunities after leaving EASA – a consultation, CAA,
November 2020.
12
CAP 2408, Consultation: Pilot Medical Declaration (PMD) review, CAA, October 2022.

© Crown copyright 2023 93 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

The DVLA
Guidance for medical professionals
The drivers’ medical section within the DVLA deals with aspects of driver licensing when
there are medical conditions that affect, or potentially affect, the safe control of motor
vehicles. It provides a summary of medical guidelines in the publication, ‘Assessing fitness
to drive – a guide for medical professionals’13. This is intended to assist doctors and other
healthcare professionals in advising their patients whether the DVLA requires notification of
a medical condition, and the potential licensing outcome from the notification.

In the publications section, ‘Age-related fitness to drive’, the DVLA provides the following
guidance:

‘Older age is not necessarily a barrier to driving.

● Functional ability, not chronological age is important in assessments.

● Multiple comorbidity should be recognised as becoming more likely with
advancing age and considered when advising older drivers.
● Discontinuation of driving should be given consideration when an older
person – or people around them – become aware of any combination of
these potential age-related examples:
- Progressive loss of memory, impaired concentration and reaction
time, or loss of confidence that may not be possible to regain.
- Physical frailty in itself would not necessarily restrict licensing, but
assessment needs careful consideration of any potential impact on
road safety.
● Age-related physical and mental changes vary greatly between individuals,
though most will eventually affect driving.
● Professional judgement must determine what is acceptable decline and
what is irreversible and/or a hazardous deterioration in health that may
affect driving. Such decisions may require specialist opinion’.

When medical professionals are assessing a patient’s fitness to drive, the DVLA advises
that they should:

● ‘advise the individual on the impact of their medical condition for safe
driving ability.
● advise the individual on their legal requirement to notify DVLA of any
relevant condition.
● notify DVLA directly of an individual’s medical condition or fitness to drive,
where they cannot or will not notify DVLA themselves.’

Footnote
13
Assessing fitness to drive: a guide for medical professionals - GOV.UK (www.gov.uk) [accessed
February 2023].

© Crown copyright 2023 94 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

This process provides the DVLA with a mechanism of enforcement should a patient choose
not to notify the DVLA of a relevant condition that medical professionals have assessed will
preclude them from driving.

The CAA does not produce similar guidance to assist medical professionals (other than
AMEs) in understanding the effects of age or medical conditions on the ability to fly. Pilots
are expected to interpret advice they receive from medical professionals in relation to driving
and apply it to their private flying. The CAA informed the AAIB that ‘pilots are responsible
for ensuring they are up to date with any current requirements’, and that any changes to
regulations or guidance are promulgated through SkyWise14 alerts.

The DVLA reported to the AAIB that it had not been informed by the CAA that the CAA had
adopted the medical standards for Group 1 driving licences and applied them to aviation.
Consequently, there is no process for the DVLA to inform the CAA if it refuses or revokes a
driving licence for medical reasons15, or if a driver voluntarily surrenders their licence.

Driving licences and medical conditions

To understand how medical issues affect driver licensing, the AAIB consulted the DVLA
to obtain data for the number of Group 1 licences that are refused, revoked or voluntarily
surrendered for medical reasons each year. Data from 2020 to 2022 has been excluded
since they are skewed due to the influence of Covid-19. The results are contained in
Figures 4 to 6 and are grouped into two cohorts: drivers under 70 years of age and drivers
of 70 years of age and older.

The total numbers of licences surrendered in each cohort indicates a measure of compliance
with the medical standards for driving (Figure 4).

Group 1 driving licence decisions - surrendered

25,000 23,657 23,484
21,979
20,534
20,000 17,886 18,377
17,523 16,852
16,651
15,191
15,000

10,000

5,000

0
2015 2016 2017 2018 2019

Surrendered under 70 Surrendered 70 and over

Figure 4
Driving licences surrendered
Footnote
14
SkyWise is a CAA website and application that provides news, notifications and alerts to the aviation sector.
15
An application for a driving licence can be refused where the applicant does not have a current driving
entitlement, eg. first applications, renewals after expiry or application after revocation. A revocation occurs
when a driver has a current driving entitlement removed by the DVLA.

© Crown copyright 2023 95 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

The total numbers of licences refused and revoked (Figures 5 and 6) indicate a measure of
intervention made by the DVLA to withhold or withdraw a driving entitlement from individuals
who do not meet the medical standard to hold a Group 1 licence.

Goup 1 driving licence decisions - refused

25,000
20,858 21,408 21,548
20,236
Number of licence decisions

20,000
17,346
14,104 13,420
15,000 12,632 12,901
10,822
10,000

5,000

0
2015 2016 2017 2018 2019

Refused under 70 Refused 70 and over

Figure 5
Driving licences refused

Group 1 driving licence decisions - revoked

20,000 17,384 17,208
Number of licence decisions

15,894 15,110
15,000
11,449
10,000 8,011 7,787 8,254 8,496
6,155
5,000

0
2015 2016 2017 2018 2019

Revoked under 70 Revoked 70 and over

Figure 6
Drivers licences revoked

For the five years of data, the trend is for an increasing number of licenses to be refused,
revoked, and surrendered in both age cohorts. Also of note is that the number of surrendered
licences is broadly similar to the number of licences subject to enforcement action (revoked
+ refused).

Unlike the DVLA, the CAA does not revoke pilot licences for medical reasons. If an
AME‑assessed medical is revoked, or PMD voluntarily withdrawn, the flying licence remains
in place, but its privileges should not be used until a valid medical certificate or PMD is
regained.

© Crown copyright 2023 96 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Analysis

The accident flight

The investigation could not determine the nature of the “slight problem” the pilot initially
reported to Cranwell ATC that caused him to decide to return to Temple Bruer. The pilot’s
response to an offer of assistance from ATC that “something hasn’t fired up as it should
have done” followed by “problem solved”, suggests that the issue was transitory and likely
related to the pilot’s portable computer that he may have been using to assist in navigating
to Beccles. A member of the pilot’s family informed the AAIB that the pilot had previously
used the phrase when the navigation application he used had not worked as expected.
There was no further reference to a technical issue in subsequent radio communications.

The route from Temple Bruer to Beccles was familiar to the pilot and the weather was fair.
Evidence from recorded data and communications with the ground radio operator at Beccles
showed that the pilot joined the circuit from the north-west, establishing himself downwind
in a right-hand circuit for Runway 09, and then onto the final approach. Witnesses reported
that G-CBDJ’s approach at first appeared to be “unstable” in roll and pitch before settling
to a more stable approach profile. The cause of this instability could not be determined,
and no local turbulence was reported. It is possible that the pilot was trying to reduce the
airspeed below 62 kt, which is the maximum speed for selecting flaps. The pilot had flown
a total of 1,621 hours in G-CBDJ, 62 hours of which had been in the preceding 13 months.
Of those, seven hours had been flown over six flights in March 2022. As such, he was in
recent practice.

An aircraft can bounce on landing for many reasons; a hard landing or landing with excessive
speed are two of the more common causes. In either scenario, the most effective response
is to go around. Following the second touchdown on the nosewheel, witnesses reported
seeing G-CBDJ apparently climbing away in an increasingly nose-up attitude, described
as being up to 45°. However, no witness could recall the sound of the engine increasing
rpm. Without the application of sufficient power to climb away, combined with a significant
nose‑up attitude and full flap, a stall was the likely outcome. The aircraft was not fitted with
a stall warner and the MFD provided only a visual indication of airspeed. The witness marks
on the left side of the nosewheel were indicative of the tyre being compressed against a
hard surface, most probably during a bounce on landing. Post-accident examination of
the wreckage did not reveal any pre-accident defects which would have affected either the
landing phase, or a go-around by the pilot.

Pilots who had flown the aircraft informed the AAIB that the CT2K could be “tricky to land”,
and that pilots needed to be “on their game” if things did not go as planned. The accident
pilot was familiar with his aircraft and in recent practice, but the landing diverged from his
intended plan. Given that he was 87 years old and recognised that he would likely have
to stop flying in the near future, it is possible that some age-related deterioration in human
performance was a factor in this accident.

© Crown copyright 2023 97 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Medical

The pilot had a current medical declaration that was valid until 21 December 2024. The
post mortem found no indication of medical impairment or incapacitation of the pilot before
the aircraft struck the ground.

Pilot Medical Declaration scheme

The online PMD introduced in October 2016 adopted the DVLA medical standard for Group 1
driving licences as the standard for pilots of GA aircraft less than 2,000 kg MTOW. Under
the scheme, there is no requirement to consult a GP or an AME, and the pilot does not have
to hold a driving licence, only have a ‘reasonable belief’ that they could. The CAA does not
provide guidance on the DVLA’s medical standard for Group 1 licences, nor does it provide
a link to the DVLA’s website which contains comprehensive advice to drivers. Pilots who
intend to make a PMD must actively seek out information relevant to their personal medical
history from DVLA sources and translate any guidance found to their private flying.

During a post-implementation review of the PMD process in October 2020, the CAA
looked at a sample of 800 PMD holders out of a total of 14,400. It found that 4% of the
800 pilots reviewed should not be self-declaring their medical status for various reasons.
The CAA concluded that if this percentage was applied to all PMD holders, the potential of
576 declarations being made in error would be ‘of concern’. Additionally, the CAA reported
that it was not clear whether this was ‘due to unclear guidance material or a misunderstanding
on the pilot’s part’.

The CAA informed the AAIB that a review of the PMD scheme is underway. However, to
clarify the medical standards required for pilots to make an online medical declaration, the
following Safety Recommendation is made:

Safety Recommendation 2023-007

It is recommended that the UK Civil Aviation Authority provides comprehensive

guidance for pilots on the medical factors that must be considered when making
an online Pilot Medical Declaration.

The DVLA publishes a summary of medical guidelines intended to assist medical

professionals in advising their patients whether the DVLA requires notification of a medical
condition, and the potential licensing outcome from the notification. However, medical
professionals may not be aware if their patients engage in private aviation and there is no
requirement for pilots to declare this. The obligation to take medical advice received on
fitness to drive and translate this to flying activity is placed solely on the pilot. Therefore,
the following Safety Recommendation is made:

Safety Recommendation 2023-008

It is recommended that the UK Civil Aviation Authority provides guidance for

medical professionals to promote awareness of the medical standards required
by the Pilot Medical Declaration scheme.

© Crown copyright 2023 98 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

The DVLA has established a process by which drivers, doctors and other healthcare
professionals are provided with comprehensive guidance on the medical requirements to
hold a driving licence. The DVLA also has a mechanism by which driving licences can be
refused or revoked based on the medical history of the licence holder. The CAA does not
revoke flying licences for medical reasons, but places an obligation on GA pilots to withdraw
their PMD and cease flying if they have reason to believe they no longer meet the medical
standard for a Group 1 driving licence. However, should a GA pilot misunderstand this
requirement, or choose not to comply, there is no means by which these individuals are
visible to the CAA and they might continue to fly.

Figure 7 shows the total number of driving licences that were subject to enforcement action
(refused and revoked) for medical reasons by the DVLA for the years 2015 to 2019.

Figure 7
Driving licences refused plus revoked

The CAA informed the AAIB that it had not consulted with the DVLA prior to adopting
driver medical standards for the PMD scheme. Nor had it established a formal record
keeping system to collect annual data on pilots making PMDs to confirm the safety analysis
assumptions. Collaboration with the DVLA prior to the publication of the original public
consultation document in 2015 would likely have revealed to the CAA the scale of the DVLA
medical-related licensing decisions and the benefits of a feedback process from healthcare
professionals to assist the oversight of licences. Therefore, to augment the CAA’s ongoing
review of the PMD scheme, the following Safety Recommendation is made:

Safety Recommendation 2023-009

It is recommended that the UK Civil Aviation Authority engages with the UK

Driver and Vehicle Licensing Agency to understand their process for managing
medical related driving licence decisions, and ensure that the UK Civil Aviation
Authority’s process for managing the Pilot Medical Declaration scheme is as
effective.

© Crown copyright 2023 99 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Pilot licence revalidation

The pilot held a UK PPL and revalidated his microlight class rating on 6 April 2021 by
providing evidence of his flying experience gained in the previous 13 months. This
method of revalidation is only available to pilots issued with the microlight rating prior to
1 February 2008. The AAIB was unable obtain details of the number of pilots who are
eligible for this category of licence revalidation. However, whilst its size is unclear, this
group certainly represents an ageing demographic that is potentially not being actively
monitored. Most notably, this method of revalidation does not require a training flight to be
conducted with an instructor.

Studies and literature reviewed during previous AAIB investigations suggest that there is
no single reliable metric to predict age-related impairment in GA pilots. Guidance provided
by the DVLA to medical professionals broadly reflects this finding in relation to driving and
concludes that ‘Professional judgement must determine what is acceptable decline and
what is irreversible and/or a hazardous deterioration in health that may affect driving’.

The parallel processes of flying licensing and medical certification should be expected to
provide appropriate oversight of pilots. However, in this case, a self-declared medical that
does not require input from a GP, combined with a method of licence revalidation that does
not require a training flight with an instructor, exposes a missed opportunity for at least
one independent professional assessment of age-related deterioration in piloting ability.
Therefore, the following Safety Recommendation is made:

Safety Recommendation 2023-010

It is recommended that the UK Civil Aviation Authority assesses the continued

appropriateness for holders of UK PPLs with microlight class ratings issued
before 1 February 2008 to revalidate that rating solely by providing evidence of
experience.

Conclusion

The aircraft bounced on landing and probably stalled. The pilot was fatally injured when the
aircraft subsequently struck the ground.

The accident pilot was familiar with his aircraft and in recent practice, but the landing
diverged from his intended plan. Although the post mortem found no indication of medical
impairment or incapacitation, the pilot was 87 years old and had prior to the flight recognised
that he would likely have to stop flying in the near future, it is possible that some age-related
deterioration in human performance was a factor in this accident.

The investigation highlighted a lack of medical guidance for both pilots and medical
professionals who use the Pilot Medical Declaration which is based on the DVLA medical
standard for Group 1 driving licences. It also identified that pilots that have a UK PPL with a
microlight rating issued prior to 1 February 2008 are not required to fly with an instructor for
licence revalidation; such a flight would provide an opportunity for independent assessment
of agerelated deterioration in piloting ability.

© Crown copyright 2023 100 All times are UTC

AAIB Bulletin: 6/2023 G-CBDJ AAIB-28093

Four Safety Recommendations have been made to the CAA, three about the Pilot Medical
Declaration and one about the revalidation of ratings for holders of licences with microlight
ratings issued prior to 1 February 2008.

Published: 20 April 2023.

© Crown copyright 2023 101 All times are UTC

AAIB Bulletin: 6/2023 G-CGRR AAIB-28540

ACCIDENT

Aircraft Type and Registration: Pegasus Quik, G-CGRR

No & Type of Engines: 1 Rotax 912-UL piston engine

Year of Manufacture: 2010 (Serial no: 8541)

Date & Time (UTC): 6 August 2022 at 1355 hrs

Location: Harringe Court Farm, Ashford, Kent

Type of Flight: Private

Persons on Board: Crew - 1 Passengers - None

Injuries: Crew - 1 (Serious) Passengers - N/A

Nature of Damage: Minor damage to trike, wing damaged beyond

repair

Commander’s Licence: National Private Pilot’s Licence

Commander’s Age: 62 years

Commander’s Flying Experience: 306 hours (of which 90 were on type)

Last 90 days - 2 hours
Last 28 days - 1 hour

Information Source: AAIB Field Investigation

Synopsis

During the landing, the aircraft veered to the right and bounced before tipping over onto its
right side. The cause of the accident could not be determined.

Although the pilot was wearing a lap strap, he was not wearing the shoulder strap provided.
Consequently, he sustained serious facial injuries when his head made contact with the
front strut.

History of the flight

The pilot arrived at Harringe Court Farm airstrip, where the aircraft was based, at about
1000 hrs on the day of the accident. He prepared the aircraft for a planned solo flight to
two local airstrips, both of which he had flown to before. The weather in the morning was
described as generally good, although there were reports of some thermal air currents
coming off the hill on which the airstrip was positioned. The pilot reported he was used to
such conditions and, whilst uncomfortable at low level, the thermal effects soon dissipated
during the climb after takeoff.

The pilot had no recollection of the accident flight due to the injuries he received. It was,
however, possible to get a record of his flying activities from information recovered from the
aircraft’s navigation unit and the pilot’s mobile phone.

© Crown copyright 2023 102 All times are UTC

AAIB Bulletin: 6/2023 G-CGRR AAIB-28540

The pilot took off from Harringe Farm to the north at 1234 hrs. Data recovered from the
aircraft and the pilot’s phone, recorded that he conducted a local flight, landing at two other
airstrips, before returning to Harringe Farm at 1345 hrs. A witness at Harringe Farm saw
the aircraft carry out an apparently normal final approach to land in a northerly direction.
The weather at the time was described as good, with just a light breeze. The witness
reported that after touching down, the aircraft bounced to a height of about a metre before
touching down again. On doing so, they described seeing the left rear wheel of the tricycle
undercarriage slowly lift into the air. They expected to see it settle onto the ground again,
but it continued to rise until the aircraft’s right-wing tip caught the ground, bringing the
aircraft abruptly to a halt on its right side.

Members of the public seeing the accident came quickly to assist the pilot, who had been
seriously injured. The emergency services were called and the pilot was transferred to
hospital by air ambulance.

Accident site

The airstrip, orientated 010° / 190°, was on farmland at the top of a small hill. The accident
occurred approximately halfway along the landing strip and a few metres beyond the western
edge. Although there was only minor damage to the trike, the wing had suffered significant
damage to the keel, cross spar and front section of the leading-edge structure which had
folded under the wing. The nose of the wing had also swivelled clockwise and was pointing
to the right of the aircraft.

The pilot’s helmet and damaged headset were located next to the trike; the visor was located
20 m away from the right side of the aircraft. The compass was found next to the cockpit
and had detached from its mounted position on the front strut. Although the front seat lap
strap buckle had been released, there were no anomalies found with the operation of the
buckle or the strap. The front seat shoulder strap had been rolled up and secured with a
plastic tie-wrap close to the pylon behind the rear seat. Two avionic units were mounted on
the top of the cockpit coaming directly in front of the pilot (Figure 1).

Aircraft examination

Examination of the engine controls and brake systems did not show any faults or anomalies.
The front strut was intact with most of the aircraft damage occurring to the wing and the
A-frame, with the right upright and its top knuckle having failed in overload. All the damage
to the aircraft was consistent with it rolling onto its side.

There was evidence of blood inside the nylon sleeve that covered the front strut and on the
lower section of the right upright.

© Crown copyright 2023 103 All times are UTC

AAIB Bulletin: 6/2023 G-CGRR AAIB-28540

A-frame
upright

Figure 1
G-CGRR pictured in flight showing significant features
(Image used with permission)

Survivability

Seat harnesses

The rear seat was fitted with a four-point harness and the front seat with a three-point
harness. The front seat harness consisted of a lap strap and a shoulder strap, although
the pilot stated that he had never used the shoulder strap on this, or any of the other three
microlight types he had flown. The pilot reported that when he bought G-CGRR the front
seat shoulder strap was rolled up and secured with a plastic tie wrap; he had not changed
this arrangement. He commented that the shoulder strap was not particularly long which
normally resulted in a relatively tight fit, making it difficult to use.

Issue 6 of the Quik Pilot’s Operating Manual states that the seat harnesses should be worn
at all times and warns that ‘Failure to put on safety harness and wear front seat or rear seat
shoulder straps could be the cause of injury or death in the event of an accident’.

Safety helmet

The pilot wore an open-faced airborne sports helmet1 fitted with a transparent visor. The
helmet which had been removed during the pilot’s rescue had some minor scratches and dirt

Footnote
1
British Standards Institute BS EN 966:1996 categorises this helmet as a ‘Helmet for airborne sports’.

© Crown copyright 2023 104 All times are UTC

AAIB Bulletin: 6/2023 G-CGRR AAIB-28540

on its right side. The visor had been badly damaged with multiple scrape marks and a wide
vertical line scored from top to bottom to the left of the visor’s centre line. The distortion of
the visor along the vertical line matched the profile of the front strut. The horizontal curved
profile of the visor had bent inwards along this vertical line (Figure 2). A large piece of the
visor had broken away from the upper left quarter.

Top

Vertical
Bottom score line

Figure 2
Damaged helmet visor

Analysis

The cause of the accident could not be determined.

No faults with the aircraft were found during the examination. The damage to the propeller
blades indicated that the engine was running at a low speed. The minor damage to the trike
showed that most of the impact forces were absorbed by the wing.

Although the possibility of the visor being damaged as a result of contact with the right
A‑frame upright was considered, the curved profile of the vertical line could only have been
formed by impact between the visor and the front strut. As the pilot had not worn the
shoulder strap, his upper torso would not have been restrained during the impact and the
visor on his helmet would only have provided limited protection to his face. It is probable
that both these facts resulted in the pilot sustaining serious facial injuries during the impact.

Conclusion

During the landing the aircraft tipped over onto its side. The pilot, who was not wearing
the shoulder strap provided, sustained serious facial injuries when his head struck the front
strut and the right upright during the accident sequence.

Published: 3 May 2023.

© Crown copyright 2023 105 All times are UTC

AAIB Bulletin: 6/2023

AAIB Correspondence Reports

These are reports on accidents and incidents which
were not subject to a Field Investigation.

They are wholly, or largely, based on information

provided by the aircraft commander in an
Aircraft Accident Report Form (AARF)
and in some cases additional information
from other sources.

The accuracy of the information provided cannot be assured.

© Crown copyright 2023 107 All times are UTC

AAIB Bulletin: 6/2023 G-CLNK AAIB-28756

SERIOUS INCIDENT

Aircraft Type and Registration: ATR 72-211, G-CLNK

No & Type of Engines: 2 Pratt & Whitney Canada PW121 turboprop

engines

Year of Manufacture: 1989 (Serial no: 147)

Date & Time (UTC): 25 October 2022 at 2030 hrs

Location: East Midlands Airport

Type of Flight: Cargo

Persons on Board: Crew - 3 Passengers - None

Injuries: Crew - None Passengers - N/A

Nature of Damage: Damage to nosewheel tyre

Commander’s Licence: Airline Transport Pilot’s Licence

Commander’s Age: 61 years

Commander’s Flying Experience: 11,011 hours (of which 63 hours were on type)
Last 90 days - 63 hours
Last 28 days - 3 hours

Information Source: Aircraft Accident Report Form submitted by the

pilot

Synopsis

After landing in a light crosswind, as the aircraft decelerated through 80 kt, it swerved right
and hit a runway edge light, damaging the nosewheel tyre. The operator has taken action
to address aircraft handling during the ground roll in crosswinds.

History of the flight

G-CLNK was operating from Jersey Airport to East Midlands Airport and made a radar
vectored autopilot coupled approach to Runway 09. During the approach, the reported
wind at the airport obtained by the pilots was from 150° at 10 kt. The commander was PF.
He stabilised the approach by 1,000 ft and then established the aircraft with a slight crab
into wind and power set at 25% torque. In the latter stages of the approach the commander
disconnected the autopilot, removed the small amount of crab in the flare and touched down
aligned with the runway. Once all wheels were on the runway, he selected ground idle and,
as the aircraft continued to decelerate along the runway, it began to swerve to the right.
He handed control to the co-pilot and applied left nosewheel tiller to straighten the aircraft.
The commander noticed that the right wing lifted, which he ascribed to the turn to the right
in combination with the crosswind from the right. The commander then applied more tiller
input to the left, aided by the application of left pedal and differential left brake input by the
co-pilot, to which the aircraft slowly began to respond.

© Crown copyright 2023 109 All times are UTC

AAIB Bulletin: 6/2023 G-CLNK AAIB-28756

During the landing roll the commander reported seeing an object on the runway ahead and
to the left. Review of the CCTV subsequently showed a ‘spark’ under the aircraft during the
landing roll.

The METAR for the airport valid at the time of landing reported wind from 140° at 12 kt.

The commander reported the excursion to ATC, and a runway inspection was carried out
which reported a broken edge light. An engineering inspection found damage to the tyre of
the nosewheel, which was replaced.

Personnel

The commander was experienced and who had recently completed his type conversion on
to the ATR 72 after previously flying the BAe ATP. The co-pilot had also recently converted
to the ATR 72.

The commander reported that, during his line training, he experienced little exposure to
crosswinds greater than 10 kt but, since then, had experienced significant crosswinds with
no control issues during landing. He also stated that he would routinely use the tiller on
the BAe ATP during the landing roll at 80 kt and below to maintain aircraft direction, in part
owing to his experience of asymmetrical braking action on the aircraft.

Manufacturer information

In 2014, the manufacturer published a safety note1, which outlined the crosswind landing
technique. In 2016, it issued a Flight Operations Information Notice2 for the ATR 42 and 72.
Both these publications provided guidance and recommendations on ‘aircraft handling
during the landing roll and deceleration’, noting that ‘insufficient aileron input, crosswind will
lift the upwind wing and make the aircraft turn’. In 2018, the manufacturer presented their
analysis of 18 events over a 5-year period between 2013 and 2017 which shared a number
of common characteristics.

The ATR Flight Safety website3 summarises the guidance and recommendations for aircraft
handling during the landing roll and deceleration as follows:

● ‘Review and brief crosswind landing technique for the decrab, flare and
landing roll, prior to the approach (TEM);

● After touchdown, hold the control column nose down to increase directional
efficiency;

● Maintain aileron input into the wind. Gradually increase it as airspeed

decreases;

Footnote

1
ATR, 2014, Safety Note #1 ’Be Prepared for Crosswind landing’.
2
ATR, FOIM 2016/06 Issue 1 ‘Crosswind landing’, dated June 7, 2016.
3
https://safety.atr-aircraft.com/my-product/prevent-runway-excursion-in-crosswind/#1574785943767-
6e2e8ed0-5ba6 [accessed January 2023].

© Crown copyright 2023 110 All times are UTC

AAIB Bulletin: 6/2023 G-CLNK AAIB-28756

● Correct heading deviation smoothly, using the rudder above 70 kt and

progressively the tiller below 70 kt;

● Rudder correction upwind shall be very smooth and progressive. Most of

the time, gently reducing/adjusting the rudder input downwind is enough to
correct heading deviation downwind;

● Use brake to minimize landing roll.’

It provides further guidance on the use of rudder stating:

‘…correction in upwind direction requires less effort than in [the] downwind

direction. An equal effort will have a stronger effect upwind. Just releasing the
downwind input without applying a force in upwind direction will lead to a rudder
deflection in upwind side – allowing heading corrections only through downwind
pedal movements.’

Operator information

Operator’s investigation

Flight data showed that the aircraft swerved abruptly to the right through 10º as it decelerated
through 80 kt. There was ‘little to no aileron input into wind…during the ground roll’ and no
application of rudder during the turn right. However, ‘as the aircraft reaches the 100º track,
there is a sharp input of both rudder and brake which sharply changes the aircraft direction
back to the runway centreline’.

The operator concluded that a ‘lack of in-to-wind aileron was the most likely cause of
the divergent path of the aircraft’ and that the crosswind ‘induced a weather-vane effect
that induced roll and also initiated the uncommanded turn on the landing roll-out’. It also
concluded that recovery occurred as a result of the co-pilot’s application of left pedal and
brake, and not through the use of nosewheel steering. It further considered that the lack of
application of in-to-wind aileron may have been because the PF did not perceive the need
for its application in the light crosswind, where he would have in stronger crosswinds.

Operator training and guidance

The operator identified aircraft handling during the landing roll with a crosswind as a key
threat. Since it used a third-party training organisation for the type rating training of its pilots
on the ATR 72, it arranged that its own instructors would deliver the final simulator session
of the type rating course and additional simulator training would be conducted by its own
instructors on completion of the type rating course. In addition, it arranged that its own
instructors would deliver the final simulator session of the type rating course, as well as the
proficiency checks and skills tests in the simulator. This enabled the operator to deliver its
own instruction on the areas of aircraft handling, which included crosswind landings.

The operator’s documentation advised that the ‘Weathercock effect makes the aircraft turn
into the wind direction’. It recommended to use of rudder to maintain directional control,
holding ‘the control column in nose down position to increase directional efficiency.’ It

© Crown copyright 2023 111 All times are UTC

AAIB Bulletin: 6/2023 G-CLNK AAIB-28756

also cautioned ‘In case of insufficient aileron input, crosswind gusts could lift the upwind
wing, reduce the aircraft ground contact and could make the aircraft turn into the wind
(weathercock effect).’ The guidance advised the use of nosewheel steering below 70 kt for
directional control.

The part B of the operator’s Operations Manual stated:

‘During the landing rollout, maintain wings level using aileron into wind, and
rudder steering; at a suitable speed below 70 kts Captain resumes control (if he
had been PM) of tiller and PLs [prop levers], or passes yoke control to co-pilot.’

Analysis

There was probably insufficient in-to-wind aileron applied during the landing roll both before
and after the handover of control, which resulted in the upwind wing lifting and the aircraft
turning into wind. The initial attempt to correct the turn by nosewheel steering through
the use of the tiller, rather than by use of rudder to maintain directional control also likely
contributed to the runway excursion.

The PF applied insufficient in-to-wind aileron and may not have recognised the need in the
light crosswind, having 63 hours on type and limited experience flying it in similar conditions.
The use of the tiller was probably a reversion to the technique that the commander had used
on the previous type that he had flown. The manufacturer’s guidance indicates that the use
of rudder above 70 kt, instead of the tiller, would have resulted in better directional control.

Conclusion

After landing in a light crosswind, as the aircraft decelerated through 80 kt, it swerved right
and hit a runway edge light. The loss of directional control probably occurred because of
insufficient in-to-wind aileron. The recovery of directional control was delayed by the use of
nosewheel steering through the tiller, rather than the use of rudder.

Safety actions

The operator took the following actions:

• The crew underwent further training in the simulator on the aircraft handling
technique in crosswinds during landing.

• The syllabus for the operator conversion course is being rewritten to

maximise crew exposure to crosswinds.

• The operator advised all involved in training on the ATR 72 to be alert to

and monitor for incorrect crosswind techniques or inappropriate use of
nosewheel steering through the tiller.

© Crown copyright 2023 112 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

ACCIDENT

Aircraft Type and Registration: 1) Boeing 777-300(ER), HL-7782

2) Boeing 757-256, TF-FIK

No & Type of Engines: 1) 2 GE Aviation GE90 turbofan engines

2) 2 Rolls Royce RB211 turbofan engines

Year of Manufacture: 1) 2009 (Serial no: 37643)

2) 1999 (Serial no: 26254)

Date & Time (UTC): 28 September 2022 at 1850 hrs

Location: London Heathrow Airport

Type of Flight: 1) Commercial Air Transport (Passenger)

2) Commercial Air Transport (Passenger)

Persons on Board: 1) Crew - 18 Passengers - 199

2) Crew - 6 Passengers - Unknown

Injuries: 1) Crew - None Passengers - None

2) Crew - None Passengers - None

Nature of Damage: 1) Left wingtip damage

2) Damage to the rudder

Commander’s Licence: 1) Airline Transport Pilot’s Licence

2) Airline Transport Pilot’s Licence

Commander’s Age: 1) 52 years

2) 55 years

Commander’s Flying Experience: 1) 10,561 hours (of which 3,384 were on type)
Last 90 days – 249 hours
Last 28 days – 73 hours

2) 15,500 hours (of which 12,500 were on type)

Last 90 days – 160 hours
Last 28 days – 50 hours

Information Source: Aircraft Accident Report Forms submitted by

both commanders and further enquiries by the
AAIB

Synopsis

Whilst taxing for takeoff the wingtip of a Boeing 777-300 collided with the rudder of a
Boeing 757 which was not fully parked on its stand. The commander of the B757 turned onto
the stand centreline without stand guidance and did not inform the ATC ground controller
that they were not fully parked, contrary to Heathrow Aeronautical Information Publication
(AIP) instructions. The commander of the B777 taxied past the protruding B757 believing
it to be fully parked and that the ATC clearance and green taxiway lights implied the route
was clear. Previous similar incidents have occurred at Heathrow.

© Crown copyright 2023 113 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

History of the flight

A Boeing 757 landed at 1843 hrs on Runway 27L at London Heathrow and taxied to parking
Stand 241 at Terminal 2B. It was dark but visibility was good. As the commander turned
the aircraft onto the stand centreline he saw that the stand visual docking guidance system
(VDGS) was not on and, at 1848:05 hrs, stopped the aircraft approximately 20 m from the
final parking position. The co-pilot called the ground handling agent on the radio and asked
when the VDGS would be activated. The handling agent replied that marshallers were
aware of their arrival and would be there “very shortly”. Several ground staff were waiting
on the stand but they were not qualified to activate the guidance system. As the flight crew
expected the guidance to be activated imminently and had informed the handling agent,
they did not inform the ATC ground controller that they were not fully parked.

Stand 241

Figure 1
Heathrow Airport Ground Map

Concurrently a Boeing 777-300 was taxiing from Terminal 4 for takeoff on Runway 27R. It
crossed Runway 27L and was cleared to “follow the green lights and hold at titan”
which routed the aircraft along Taxiway Alpha then north on Taxiway Lima. As it made the
left turn from Alpha to Lima the co-pilot saw the B757 on Stand 241 and told the commander
it looked like it was protruding from the stand. The commander looked at the B757 but
could not see an anti-collision light, so thought it was fully parked. He thought it was “quite
close” but as ATC had not mentioned a conflict, he felt it was safe to taxi past. He reduced
speed slightly and moved slightly right of the taxiway centreline, and continued to taxi north
on Lima.

At 1853:56 hrs, as the B777 taxied past, the B757 crew felt a sudden jolt. They had
been stationary for nearly six minutes. The B757 co-pilot looked out of his window and
saw the B777 taxiing past, and the B757 pilots realised there had been a collision. The

© Crown copyright 2023 114 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

commander called the cabin crew at the back of the aircraft to check if they were okay. The
crew confirmed they had felt the jolt but they were fine and no one had been injured. The
commander then informed ATC that they thought there had been a collision.

The B777 crew had not felt the collision and continued north on Lima. As they approached
the northern end of Taxiway Lima, ATC instructed them to hold position and informed them
of the potential collision. At the same time a passenger on the B777 told the cabin crew that
they had seen the wingtip hit the other aircraft. The cabin crew passed the message to the
flight crew at about the same time as ATC instructed them to stop.

Just prior to the collision, an airport operator leader vehicle had arrived on Stand 241 to
marshal the aircraft. The driver saw the B777 pass behind but did not realise there had
been a collision. He switched on the guidance system at about the same time as the
collision occurred. After the B757 commander confirmed the crew were okay he taxied the
aircraft forward to the final parking position.

ATC initiated the ground incident procedure and both aircraft were inspected. The inspection
confirmed there was damage to the left wingtip of the B777 and to the rudder of the B757.
The B777 was shut down on the taxiway and passengers were disembarked to busses and
into the terminal. The B757 passengers disembarked normally.

Figures 2 and 3 show the damage to the B777 left wingtip. Figure 4 shows the damage to
the B757 rudder.

Heathrow ground handling

On the day of the accident one of the tunnels used to access the central area at Heathrow
was closed for several hours. During the closure a contraflow was in operation within the
other tunnel. The ground handling agent reported that this caused several of their staff to be
delayed getting to work. As the B757 approached Heathrow, due to these staff shortages
the ground handling agent realised they would not have a dispatcher available to meet the
aircraft and activate the stand guidance. The handling agent reported that it had a prior
agreement with the airport operator that, in these circumstances, they would ask the airport
operator to send a marshaller to the stand. The handling agent made the request to the
airport operator when the B757 entered the Heathrow Terminal Manoeuvring Area, and the
airport operator agreed to send a marshaller when able. The airport operator assigned the
task to one of its leader vehicle drivers, and he proceeded to the stand as soon as he had
finished his previous task, arriving just prior to the collision.

The airport operator reported that it did not have an agreement with any handling agent
to provide marshallers in the event of staff shortages. It would only expect to provide a
marshaller in the event of a failure of the guidance system as specified in the UK AIP entry
for Heathrow. The airport operator also confirmed that other access points were available
to staff so the tunnel closure should not have caused any staff to be delayed.

© Crown copyright 2023 115 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Figure 2
Boeing 777 left wing tip

Figure 3
Boeing 777 left wing tip from above

© Crown copyright 2023 116 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Figure 4
Boeing 757 rudder

Recorded information

The CVR and FDR were recovered from both aircraft. The FDR from both aircraft had data
from the accident and the CVR from the B757 had a recording of the accident, but by the
time the power was isolated from the CVR of the B777 the incident had been overwritten.
The available recordings were used to confirm the history of flight.

Aerodrome CCTV and footage was obtained from the leader vehicle dashcams. This
confirmed the B757 anti-collision light was on and working normally when the accident
occurred. Figure 5 was created by the airport operator and shows the approximate position
of each aircraft when the collision occurred.

Figure 6 shows radio transmission on the ATC ground frequency from the time the B757
stopped short of the stand to the time of the collision. Shortly after they stopped there were
several periods of between 5 – 7 seconds when there was no transmission. The longest
gap during the 6 minutes was 18 seconds.

A review of the CVR from the B757 during the time the aircraft was stationary on stand
did not reveal any attempt to contact the ATC ground frequency or any discussion about
contacting them until after the collision.

© Crown copyright 2023 117 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Figure 5
Approximate positions of the B777 and B757 when the collision occurred
(image used with permission)

Figure 6
Analysis of the ground frequency whilst the B757 was stationary on Stand 241

© Crown copyright 2023 118 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Aerodrome
Aerodromeinformation
information

Taxiway
Taxiway Lima
Limaisisa a‘code
‘code F’
F’ taxiway,
taxiway, suitable for aa Boeing
suitable for Boeing777-300.
777-300.Parking
Parking stands
stands are are
delineated
delineatedbybya adouble
doublewhite
white line.
line. If
If parked aircraftare
parked aircraft arepositioned
positioned within
within thethe white
white lines,
lines, an an
aircraft taxing on the centreline will have sufficient wingtip clearance.
aircraft taxing on the centreline will have sufficient wingtip clearance.

The
The UKUK AIP
AIPentry
entryfor
forHeathrow contains thethe
Heathrow containsBNL guidance
guidance shown
shownininFigure
Figure 7. The AIP
7. The AIPentry
entry is
several pages
is several long
pages andand
long describes
describesmany manyprocedures for Heathrow,
procedures for Heathrow,ofofwhich which this
this is aissmall
a small
part.
part.It describes
It describesthe
theprocedure
procedure flight crew should
flight crew shouldfollow
followif ifthe
thestand
stand guidance
guidance is not
is not active
active
when
when approaching
approachingthe thestand
stand or
or ifif itit fails whilstparking.
fails whilst parking. Holding
Holdingposition
position
onon
thethe centreline
centreline
has
has the advantage that the aircraft overtly blocks the taxiway, reducing the chance of aof a
the advantage that the aircraft overtly blocks the taxiway, reducing the chance
collision, and permitting the aircraft to continue to taxi if required and with permission.
collision, and permitting the aircraft to continue to taxi if required and with permission.

Flight crew must not attempt to self-park if the VDGS is not activated or calibrated for
their aircraft type.
In the event of there being no activated VDGS displayed upon approach to the stand
flight crew should:

 Hold position on the taxiway centre-line.

 Inform Ground Movement Control (GMC) they are awaiting stand entry guidance.
 Contact company to arrange activation.

Note, GMC may request aircraft to ‘report parked’ – this is not an instruction to self-park.

In the event of a failure of the VDGS during parking, flight crew should

 Inform Ground Movement Control (GMC) of a stand guidance failure.

 Contact company to arrange a marshaller.

Figure 7
Figure 7
Text
Textfrom
from the
the UK
UK AIP entryfor
AIP entry forHeathrow
HeathrowAirport
Airport

Stand
Stand 241241cannot
cannot be be seen
seen from
fromthe
theATC
ATC visual control
visual tower
control as it as
tower is obscured by a hotel
it is obscured by a andhotel
andthetheterminal
terminalbuildings.
buildings. TheTheview fromfrom
view the the
ground controller’s
ground position
controller’s in useinatuse
position theattime
theistime
is shown
shown ininFigure
Figure8.8.Figure Figure 9 shows
9 shows an image
an image fromfrom the ATC
the ATC groundground
movementmovement
radar as radar
the as
theB777
B777 taxied
taxied past past Stand
Stand 241.241. A small
A small primary
primary radarradar
returnreturn
can be canseenbe at
seen
the atbacktheofback
the of
thestand
standand andis islikely
likelytotobebethe
theB757
B757 protruding
protruding from
from thethe stand.Primary
stand. Primary radar
radar returns
returns are are
masked
masked once
oncethe theaircraft
aircraftisis on
on stand.
stand. An aircraft
aircraftnormally
normallyshows
showsa a secondary
secondary radar
radar returnreturn
until itsits
until transponder
transponderisisswitched
switched off.
off. Whilst taxiing,aalabel
Whilst taxiing, labelgiving
givingthethe aircraft’s
aircraft’s callsign
callsign (and(and
parking stand for inbound aircraft) is displayed alongside the secondary return. These
labels are suppressed when the aircraft is 8on stand. With the B757 not fully parked it
would normally show as a hollow diamond until the transponder is switched off. However,
as shown in Figure 9, no diamond was displayed for this aircraft. It was not determined
why the diamond was not displayed. ATC reported that the controller would not normally

© Crown copyright 2023 119 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

respond to the small primary returns as they are common across the airport. Similarly,
even if the hollow diamond had been displayed, it is unlikely to have suggested anything
abnormal to the controllers as these are often displayed on stands (as can be seen on
several other stands in Figure 9).

Figure 8
View from the ATC ground controller position (location of Stand 241 circled)

Stand 241 Boeing 777

Hollow diamond returns

on other stands

Small primary radar return

from the Boeing 757

Figure 9
Extract for the ground movement radar as the B777 passed Stand 241

© Crown copyright 2023 120 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Flight crew

Flight crew experience and recency is shown below.

Boeing 777-300 Boeing 757-256

Commander Co-pilot Commander Co-pilot
Total time (hours) 10,561 3,478 15,500 2,907
Time on type (hours) 3,384 2,735 12,500 2,321
Last 90 days (hours) 248 176 116 194
Last 28 days (hours) 73 56 50 59
Start of duty 1835 hrs 1430 hrs
Table 1
Flight crew hours and recency

The B777 crew reported that they were well rested and did not consider fatigue was a
factor in the accident. They reported they departed stand on time and were not under any
abnormal time pressure.

The commander of the B757 reported that he had read the AIP entry regarding stand
guidance in the past. However, his normal practice was to initiate the turn onto the stand
before looking to see if the guidance was activated rather than looking sideways for guidance
whilst still on the centreline. He stated that, in his experience, the guidance was often
switched on as the aircraft turns onto the stand and that starting the turn had never been
a problem; nor had he heard of it being a problem for other pilots. He reported that he did
consider informing ATC that they were not fully parked, but the ground frequency was too
congested and it was not possible for them to make a radio call.

Previous incident

A similar event occurred six weeks before this event, on 16 August 2022. A Boeing 787-900
was parking on Stand 244 when the stand guidance system failed. The aircraft stopped
short of the final position to await a marshaller. As the marshaller arrived and starting to
marshal the aircraft, a Boeing 787-800 was taxing south along Taxiway Lima. The right
wingtip of the Boeing 787-800 collided with the tail of the Boeing 787-900. Initially no one
realised a collision had occurred and the Boeing 787-800 took off without the flight crew
knowing the aircraft was damaged. Figure 10 shows the damage to the Boeing 787-800,
discovered after landing.

The Boeing 787-900 crew did not report to ATC that they were not fully parked. That incident
was not investigated by the AAIB.

Following this event, the Heathrow Air Navigation Service Provider (ANSP) issued a
Safety Alert to remind operators about the AIP entry regarding VDGS. The Safety Alert
was highlighted at the Heathrow Flight Operations Safety Committee and in several other
forums.

© Crown copyright 2023 121 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Figure 10
Damage to a Boeing 787-800 after a similar collision

The AAIB has investigated previous ground collisions at Heathrow. On 15 October 2007 a
collision occurred between and Airbus A340 (4R-ADC) and a Boeing 747 (G-BNLL) and on
23 March 2004 a collision occurred between an Airbus A321 (EI-CPE) and a Boeing 747
(G-BNLK). Both events occurred near the runway holding points. The reports highlighted
that pilots and tug drivers often perceive an ATC taxi clearance to imply the route is
clear of obstructions. Controllers will alert crews if they know of a conflict but cannot do
so if they don’t. AAIB Safety Recommendations were made to enhance the guidance
available to ensure pilots and tug drivers are aware that conflicts may exist whilst taxiing,
and that pilots and tug drivers remain responsible for ensuring safe separation. The
Safety Recommendations were accepted.

Organisational information

The sequence of events that led to this accident began with the VDGS not being switched
on when the aircraft arrived at the stand. During the period between 1842 hrs and
1858 hrs reviewed as part of this investigation five other aircraft reported being unable to
park because there was no stand guidance. However, as these were reported to ATC,
in accordance with the AIP, no other incidents occurred. Heathrow ATC and the airport
operator reported that lack of stand guidance is currently a persistent problem.

© Crown copyright 2023 122 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

Analysis

The collision occurred when a Boeing 757 was waiting to park on stand and a Boeing 777
attempted to taxi behind it. The B777 pilots saw the B757 protruding from the stand but
considered it was safe to continue taxiing because ATC had cleared them to do so, and
because they had green taxiway lights ahead. They also reported that the anti‑collision
light on the B757 appeared to them to be off, suggesting it was fully parked. CCTV
evidence showed the anti-collision light was on.

Heathrow ATC cannot see this parking stand from the visual control tower, and ground
radar did not show the B757 once it had turned onto stand, so ATC had no way to know
the aircraft was not fully parked. The AAIB has previously reported on ground collisions
where pilots thought they had safe separation because they had an ATC clearance. ATC
try to inform pilots if they become aware of a hazard, but they can only do this if they know
about the hazard. Whilst it remains the pilot’s responsibility to ensure sufficient wing tip
clearance exists, it is not possible to see the wingtips of a B777-300 from the flight deck,
as they are 32.4 m outbound and approximately 47 m behind the flight deck. Parking
stands are delineated with a double white line, and if anything is protruding beyond these
lines wing tip clearance cannot be assured. A significant proportion of the B757 was over
the lines and protruding into the taxiway as illustrated in Figure 5.

When the B757 approached the parking stand the guidance system was not switched
on. In these circumstances the LHR AIP entry instructs pilots to remain on the taxiway
centreline and inform ATC. However, the B757 commander reported he normally initiated
the turn before checking for guidance and was not aware that this could cause a problem.
He reported that they did not inform the ATC ground controller due to congestion on the
ground frequency but there was no discussion between the pilots on the CVR about
contacting the ATC ground controller until after the collision and analysis of the ground
frequency recordings suggested sufficient gaps existed to make a call. The rule to stop
on the centreline and inform ATC is the primary barrier to prevent this type of accident.
However, there are different rules and procedures in airports around the world and it
can be challenging for pilots to read all the guidance and remember all the rules at each
airport. The stand guidance rule at Heathrow is a few lines within many pages of text so
is not especially prominent.

A similar incident occurred a few weeks before this accident but, in the previous incident,
one of the aircraft took off without the pilots knowing their aircraft was damaged and
continued its flight to its destination. It was luck that the damage was minor and a more
serious accident did not occur. Following this incident, the ANSP issued a Safety Alert
to remind the major operators at Heathrow about the AIP rule and this was highlighted
in several airport operator forums. After this more recent accident an Aeronautical
Information Circular and a NOTAM have been published highlighting the rule.

Lack of stand guidance when arriving on stand is reported to be a regular problem at

Heathrow. On the day of this accident, it was reported that the problem was exacerbated by
staff shortage caused by the closure of one of the tunnels used to access the central area.
A lack of resources in one part of a system can have a safety consequence in another part

© Crown copyright 2023 123 All times are UTC

AAIB Bulletin: 6/2023 HL-7782 and TF-FIK AAIB-28692

of the system, with the potential to contribute to an accident. In this case, pilots and ATCOs
inherited a problem initially caused by ground staff shortages.

Conclusion

The collision occurred because the commander of the B777 continued to taxi past the
protruding B757, believing it was fully parked and that the ATC clearance and green
taxiway lights implied the route was clear. The commander of the B757 did not follow the
Heathrow AIP instruction to remain on the centreline if no stand guidance is available,
because his normal practice was to look for guidance after he had turned onto the stand.

The initiating event was a lack of stand guidance when the B757 arrived on stand, caused
by ground staff shortages. Lack of stand guidance is a common occurrence at Heathrow
that all parties should continue to work together to address.

Bulletin Correction

Prior to publication two amendments were made to the report.

On page 115 under the section ‘Heathrow ground handling’, the second
sentence of the second paragraph ‘It would only expect to provide a marshaller
in the event of a failure of the guidance system.’ was changed to:

‘It would only expect to provide a marshaller in the event of a failure of the
guidance system as specified in the UK AIP entry for Heathrow.’

On page 124 the final sentence of the conclusion ‘This is a common problem at
Heathrow.’ was changed to:

‘Lack of stand guidance is a common occurrence at Heathrow that all

parties should continue to work together to address.’

The online version of the report was corrected before the report was published on
8 June 2023.

© Crown copyright 2023 124 All times are UTC

AAIB Bulletin: 6/2023 EC-KPB AAIB-28331

ACCIDENT

Aircraft Type and Registration: Cessna Citation 560XL, EC-KPB

No & Type of Engines: 2 Pratt and Whitney PW545C turbofan engines

Year of Manufacture: 2008

Date & Time (UTC): 1 June 2022 at 1420 hrs

Location: RAF Northolt, South Ruislip, Middlesex

Type of Flight: Commercial Air Transport

Persons on Board: Crew - 2 Passengers - 3

Injuries: Crew - None Passengers - None

Nature of Damage: Detached nosewheel and fractured nose gear

forks

Commander’s Licence: Airline Transport Pilot’s Licence

Commander’s Age: 48 years

Commander’s Flying Experience: 3,800 hours (of which 850 were on type)
Last 90 days - 243 hours
Last 28 days - 78 hours

Information Source: Aircraft Accident Report Form submitted by the

pilot and additional enquiries with the operator

Synopsis

Whilst taxiing, after landing, the nose landing gear wheel detached from the aircraft.
Assessment of the wheel assembly identified that one of the conical bearings within the
axle assembly had failed, most likely as a result of corrosion. The cause of the onset of
corrosion could not be determined.

History of the flight

Whilst taxiing after a normal landing the crew heard an unusual noise from the aircraft and
then felt the front of the aircraft drop.

The nosewheel had detached from the nose landing gear (NLG) leg and came to rest in
the grass adjacent to the taxiway, leaving the forks on the NLG resting on the taxiway
(Figure 1).

© Crown copyright 2023 125 All times are UTC

AAIB Bulletin: 6/2023 EC-KPB AAIB-28331

Figure 1
EC-KPB broken NLG fork assembly
(reproduced with permission)

Aircraft information

The Cessna Citation 560XL has a traditional tricycle landing gear. The NLG has a single
strut arrangement with a single wheel attached to the fork (Figure 2). The NLG is steerable
via the pilot’s rudder pedals and is not braked.

The nosewheel assembly on EC-KPB was last replaced in September 2021 and had been
fitted for 337 flying hours and 230 cycles. Prior to being fitted, the wheel assembly had been
reconditioned in accordance with the manufacturer’s component maintenance manual.
Grease was applied at the time it was fitted. According to the Aircraft Maintenance Manual
(AMM) no maintenance was required to be carried out on the wheel since it was fitted, and
none was carried out.

It was not possible to establish whether the bearings were new or re-used when they were
fitted during the wheel reconditioning, but they would have been inspected at the time. It is
likely that they were of similar usage; however, this could not be confirmed.

© Crown copyright 2023 126 All times are UTC

AAIB Bulletin: 6/2023 EC-KPB AAIB-28331

Figure 2
Cessna Citation 560XL NLG arrangement
(reproduced with permission)

Examination of the wheel assembly

One of the conical bearings in the wheel assembly had failed. Due to the extensive damage
to the bearing, laboratory examination could not determine the cause of the failure; however,
examination of the non-failed bearing identified evidence of multiple lines of material pitting
along the length of the rollers and a band of corrosion pitting around the shoulder (Figure 3).
Similar indications on the cone (Figure 4) and cup were noted. This was indicative of
corrosion pickup between the rollers and races, the linear pitting was particularly indictive
of corrosion whilst the bearing was stationary. As both bearings operated in the same
environment, it is likely the failed bearing would also have been corroded.

Although only a small amount of grease remained on the failed bearing components, as it
had burnt off due to the heat generated during the bearing failure, grease was found within
the wheel cavity. This grease type was consistent with the grease approved in the AMM.
The grease seal, which protects the bearing from ingress of moisture and dirt from the

© Crown copyright 2023 127 All times are UTC

AAIB Bulletin: 6/2023 EC-KPB AAIB-28331

outside environment, and is located outboard of the bearing, was damaged during failure
of the bearing. It, therefore, could not be determined if the seal was functioning correctly
before the bearing failed.

Figure 3
Roller bearings from the non-failed bearing, exhibiting linear and circumferential pitting

Figure 4
Roller bearing cone from the non-failed bearing, exhibiting linear pitting and corrosion
around the track shoulder indicating the onset of corrosion

© Crown copyright 2023 128 All times are UTC

AAIB Bulletin: 6/2023 EC-KPB AAIB-28331

Failure sequence

Examination of the wheel components determined a possible failure sequence as follows:

1. Corrosion pitting in the bearing races caused spalling and then

disintegration of the bearing cage.

2. This caused the rollers to skid and due to a combination of frictional

heating and loose material the bearing seized, causing the axle to spin on
the buckets that support it at each end. This generated more heat.

3. A combination of the heating and sideways loading caused the spacer on

the side of the failed bearing to disintegrate.

4. As the spacer failed, the side load caused the axle to migrate towards it,
hot working the end of the axle, which was splayed by the bucket, forming
a flange.

5. The axle then continued to migrate until it disengaged from the bucket on
the non-failed side, snapping the through bolt.

6. The wheel then twisted out of the forks and separated from the aircraft.

Conclusion

The detachment of the NLG wheel resulted from a failure of the conical wheel bearing
within the hub. The cause of the bearing failure could not be directly established; however,
the non-failed bearing in the wheel assembly exhibited evidence of corrosion. As the
bearings operated in the same environment it is possible that the failed bearing was also
corroded which is likely to have played a part in its failure sequence. The cause of the
onset of corrosion could not be determined.

© Crown copyright 2023 129 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise AAIB-28582

ACCIDENT

Aircraft Type and Registration: DJI Mavic 2 Enterprise

No & Type of Engines: Four electric motors

Year of Manufacture: Unknown

Date & Time (UTC): 7 August 2022 at 1648 hrs

Location: Bangor Train Station Car Park, Gwynedd

Type of Flight: Emergency Services Operations

Persons on Board: Crew - None Passengers - None

Injuries: Crew - N/A Passengers - N/A

Nature of Damage: Damaged beyond economic repair

Commander’s Licence: Other

Commander’s Age: 37 years

Commander’s Flying Experience: 7 hours (of which 4 were on type)

Last 90 days - 1 hour
Last 28 days - 1 hour

Information Source: Aircraft Accident Report Form submitted by the

pilot and further enquiries by the AAIB

Synopsis

The UAS was being used in a police operation over an abandoned building near a public car
park and train station. During hover over the building the aircraft’s motors stopped and the
aircraft fell vertically with no prior warning to the remote pilot. The aircraft struck the roof of
the building and the battery separated. Recorded data indicated that the battery probably
disconnected in flight. This could have been caused by the battery not having been fully
latched prior to takeoff, or the latching mechanism or battery being worn from repeated use
resulting in an in-flight disconnection. The operator has taken safety action to remind their
pilots of the importance of pre-flight checks and checking airframe and battery condition at
their base.

History of the flight

The 1.1 kg Mavic 2 Enterprise UAS was being used to locate suspects during a police
operation. The wind was light, the air temperature was 18°C and there was no precipitation.
The Mavic needed to be deployed quickly but the remote pilot reported checking the battery
and the aircraft prior to takeoff. He decided to take off from near the corner of a car park
(Figure 1) next to an old industrial site to the south-east that had been fenced off. The
pilot was contacted by another officer and asked to fly to and hover over the middle of an
abandoned building on the industrial site.

© Crown copyright 2023 130 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise AAIB-28582

Figure 1
Takeoff and accident site location
(Imagery ©2023 Bluesky, Infoterra Ltd & COWI A/S, CNES / Airbus, Maxar Technologies, Map data ©2023)

About 4 minutes after takeoff another officer joined him who he asked to act as his observer.
The aircraft was hovering over the centre of the building in Figure 1 when, about 6 minutes
after takeoff, the pilot lost signal and video feed on his controller screen. The observer told
him that they had seen the aircraft drop and the noise had stopped. The pilot looked up and
could not see the aircraft.

It was located shortly afterwards on the roof of the building. The battery had separated from
the airframe and they were about 1.3 metre apart from each other.

Recorded information

The recorded data file was downloaded from the controller. No faults were recorded and it
showed the battery state of charge reducing linearly from 98% to 77% when the recording
ended while the aircraft was in a hover at a height of 39 m above the takeoff point. Battery
voltage fluctuations began 4 seconds before the end of recording. The last data point was
recorded 5 minutes and 42 seconds after takeoff.

The data was sent to the aircraft manufacturer for analysis. They stated that there was an
‘abrupt change in battery voltage before the flight record ended’ and that ‘it could be possible
that it was because of the battery disconnection (or loose connection) from the aircraft due
to improper installation of the battery or the battery being swollen (*The battery has been

© Crown copyright 2023 131 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise AAIB-28582

used for about 3 years according to the log file and high temperature during Summer may
have an impact on the battery)’.

They stated that another possible reason for a loose connection was the battery having
been used for three years and ‘probably reaching its end of lifecycle’.

The remote pilot was wearing a video camera (bodycam) which showed him setting the
aircraft on the ground prior to takeoff (Figure 2). The operator and remote pilot reported
that this image shows that the gap between the battery and airframe is larger than it should
be when the battery is correctly installed. This picture was sent to the aircraft manufacturer
and they stated that it was not possible to judge from this image if the battery was partially
attached or not, or whether it was swollen or not.

Figure 2
Image from the pilot’s bodycam prior to takeoff

The video camera also showed the aircraft, as a very small dot, falling from the sky. Another
small dot was sometimes apparent directly above it, which the operator thought was the
separated battery, but it was likely an artefact of the video. When a bird flew across the
camera’s field of view a small dot also appeared behind it. There was no indication that the
aircraft had struck a bird.

Aircraft examination

The aircraft had suffered damage to its underside, with damage to the motor arms and
the camera/gimbal assembly. The battery had black scuff marks (Figure 3). The operator
re‑installed the battery into the aircraft which showed that the scuff marks did not form a line
with any marks on the aircraft’s upper surface (Figure 4). This showed that the marks were
made on the battery after it had separated from the airframe. However, the damage to the
battery was not consistent with it falling from a height of more than 20 m on to a solid roof;
there were no dents. It was also only about 1.3 m away from the airframe, so it is probable
that the battery separated when the airframe struck the roof.

© Crown copyright 2023 132 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise AAIB-28582

Figure 3
Battery from the accident Mavic

Figure 4
Accident battery re-installed in accident Mavic

The operator tested the battery locking mechanism after the accident with the accident
battery, and it functioned normally. The battery was powered on and it provided power to
the aircraft.

The operator also carried out a test with a partially latched battery and were able to power
up the aircraft with no warnings provided to the pilot.

The operator reported that the battery did not exhibit any signs of swelling, and they reported
that the data from the battery showed that its highest recorded temperature was 42.3°C,
which is below the temperature limit of 50°C. The battery had had 46 charge cycles.

© Crown copyright 2023 133 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise AAIB-28582

Aircraft manufacturer information

The aircraft manufacturer has published safety guidelines1 for the Mavic 2 Enterprise which
contains a Pre-flight Checklist which states:

‘Ensure the Intelligent Flight Battery is mounted firmly in place’

In the section ‘Maintenance and Upkeep’ it states:

‘The battery is rated for 200 cycles. It is not recommended to continue use
afterward.’

They have also published battery safety guidelines for the Mavic 2 Enterprise2 which state:

‘Never use or charge swollen, leaky, or damaged batteries.

The batteries should be used at temperatures between -20° and 40°C. Use of
batteries in environments above 50°C can lead to fire or explosion.’

The aircraft manufacturer stated that ‘if the battery was properly handled according to the
guideline, the possibility of abnormal performance of the battery is very low’.

Operator information

The operator concluded that the most likely cause of the accident was that the pilot had not
properly installed the battery and fully engaged the locking mechanism. They considered that
the pilot had likely rushed to set up the aircraft, due to the nature of the urgent deployment,
and not realised the battery was not properly installed.

The operator also noted that the pilot had not placed the aircraft on a takeoff mat to prevent
dirt ingress into the motors and did not wait until joined by an observer before taking off. He
also did not brief the observer on what their role was.

Analysis

The aircraft’s motors stopped and the aircraft fell vertically with no prior warning to the
remote pilot. The recorded data indicates that the most likely cause was a loss of battery
power to the aircraft which instantly cut off the motors and the link to the controller. The
battery state of charge was 77% at the time and there had been no warnings related to
the battery. Both the operator and the aircraft manufacturer concluded that the most likely
cause was the battery becoming disconnected. The damage to the battery was consistent
with it having separated when the aircraft struck the roof of the building, and this detachment
was more likely if it was already loose.

Footnote
1
https://dl.djicdn.com/downloads/Mavic_2_Enterprise_Advanced/Mavic_2_Enterprise_Series_Disclaimer_
and_Safety_Guidelines.pdf. Accessed 17 January 2023.
2
https://dl.djicdn.com/downloads/Mavic_2_Enterprise_Advanced/Mavic_2_Enterprise_Intelligent_Flight_
Battery_Safety_Guidelines.pdf. Accessed 17 January 2023.

© Crown copyright 2023 134 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise AAIB-28582

The operator believed that the pre-flight checks were probably rushed and that the body
worn camera image showed the battery not fully engaged. The aircraft manufacturer could
not confirm whether it was fully engaged, but it is possible that it was not. It is also possible
that the battery was engaged at that time, but that the latching mechanism or battery were
slightly worn from use which led to an in-flight disconnection. There was no indication that
the battery was swollen or had exceeded the temperature limit.

There are no sensors on the battery locking mechanism to detect and warn the pilot that a
battery is not fully latched. For this type of UAS it is important that it is flown in a manner
to reduce the risk to uninvolved third parties if it were to fall vertically, and in this case the
aircraft was being flown over a large abandoned building which reduced that risk.

Conclusion

The aircraft’s motors stopped and the aircraft fell vertically with no prior warning to the
remote pilot. Recorded data indicated that the battery probably disconnected in flight. This
could have been caused by the battery not having been fully latched prior to takeoff, or
the latching mechanism or battery being worn from repeated use leading to an in-flight
disconnection. The operator has taken safety action to remind their UAS pilots of the
importance of pre-flight checks and checking airframe and battery condition at their base.

Safety Action

The operator has shared the learning from this accident with all its UAS pilots
and reminded them of their responsibility to turn on their body worn camera
before they carry out the UAS pre-flight checks so that the checks are captured,
and of their responsibility to take time on the UAS checks that are completed at
a local air base to ensure the aircraft is fit for use.

They also planned to reinforce the briefing of observers, and to carry out routine
checks of the batteries.

© Crown copyright 2023 135 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise Zoom AAIB-28830

ACCIDENT

Aircraft Type and Registration: DJI Mavic 2 Enterprise Zoom

No & Type of Engines: 4 DJI electric engines

Year of Manufacture: Unknown (Serial no: 276DFB5001QU7B)

Date & Time (UTC): 5 December 2022 at 1430 hrs

Location: Garstang, Lancashire

Type of Flight: Commercial Operations (UAS)

Persons on Board: Crew - N/A Passengers - N/A

Injuries: Crew - N/A Passengers - N/A

Nature of Damage: Damage to propellers and body

Commander’s Licence: Other

Commander’s Age: 50 years

Commander’s Flying Experience: 257 hours (of which 35 were on type)

Last 90 days - 26 hours
Last 28 days - 9 hours

Information Source: Aircraft Accident Report Form submitted by the

pilot

Synopsis

During an aerial survey flight, the UAS detected a rapid loss of battery power and initiated
an immediate automatic landing. Whilst descending, its flight behaviour became erratic,
control was lost to the remote pilot, and the UA struck an uninvolved person before hitting
the ground. It was not possible to determine a cause for the loss of battery power or the flight
behaviour. There may have been an opportunity during flight control checks to consider
unexpected battery discharge rate as a reason to abort the flight.

History of the flight

The UAS was planned to fly as part of a survey to monitor progress on a construction
site, to take images along the front and across the site within the property boundaries
agreed with the client. Weather conditions on the day were good visibility, low winds and
moderate temperature.

The flight started from a location towards the rear of the construction site, approximately
300 m from the site frontage. The remote pilot then flew the UA to approximately
head‑height and completed flight control checks, where the battery indicator showed
95% but rapidly dropped to 88%. The pilot continued, manually flying the aircraft at a
height of 50 - 60 m over the construction site towards a point approximately 50 m from
the site frontage to record the imagery.

© Crown copyright 2023 136 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise Zoom AAIB-28830

The pilot received a critically low – landing battery notification and the UA initiated
automatic landing1.

The UA was observed to fly in an erratic manner, and the pilot stated he had limited lateral
control available to enable him to fly it into a clear area. It came close to houses under
construction and stopped responding to control inputs. The pilot described it as then
“appearing to descend too fast, despite the propellers spinning”, before going out of sight
behind some construction materials. The UA struck a site worker on their arm, dropped into
some cement, and fell to the ground. The site worker was not injured.

Figure 1
UA after falling to the ground

Aircraft information

The DJI Mavic 2 Enterprise Zoom is a commercially available UAS. It comprises a UA with
maximum take-off weight of 1100g, and a handheld control unit. The UA is fitted with a
battery unit which has an integral charge level indicator.

Battery management

The UA was fitted with a battery that had been fully charged for the flight. It can be set up
to notify the pilot with an alert at pre-set battery charge levels. This UA was configured
to provide alerts for low battery at 30%, and critically low battery at 20% power
remaining. A battery level of 30% is intended to have enough power left for the aircraft
to Return to Home (RTH) to the last recorded Home Point. However, the aircraft also
self‑determines whether the battery level is sufficient to RTH based upon position
information, meaning that a RTH notification and action can occur independently from
pre-set battery level warnings.
Footnote
1
The aircraft will land automatically if the current battery level can only support the aircraft long enough to
descend from its current altitude. The user cannot cancel the auto landing but can use the remote controller
to alter the aircraft’s orientation during the landing process. (Mavic 2 Enterprise Series User Manual v1.8)

© Crown copyright 2023 137 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise Zoom AAIB-28830

When a critically low battery level alert is triggered, the UA will land after 10 seconds,
or immediately if deemed to be an extremely critically low2 battery level. Both actions cannot
be cancelled by the pilot, but limited directional control should remain available for hazard
avoidance.

The aircraft did not notify the pilot of low battery at 30% remaining, only critically low –
landing immediately before descending. Control of the aircraft was lost to the pilot and he
was unable to avoid hazards despite applying control inputs.

Batteries for this aircraft have a manufacturer’s rating of 200 cycles3; this battery had
43 recorded cycles. The operator normally replaces batteries at 100 recorded cycles.

Aircraft examination

The UA was sent to an authorised repair facility for assessment. No discrepancies could
be found with the battery condition, and the battery’s integral charge level indicator showed
50‑75% charge. The battery appeared to have been seated correctly and no physical
defects were found with the UA that could have caused the loss of control.

The flight log was analysed post-flight but there was no data that showed a loss of power
or flight control.

UAS Operation Regulations

UAS operations within UK airspace are legislated by three main pieces of regulation4
alongside UK CAA policy and guidance UK CAA CAP722 Unmanned Aircraft System
Operations in UK. Flight operations are further categorised using a risk-based approach
and the incident flight was being operated within the ‘Specific’ category where the UAS
operator is subject to additional oversight by the CAA due to a higher level of risk associated
with the operations. To comply with the additional oversight, the operator of this UA had an
operations manual5 and was issued with a CAA operational authorisation for Pre-Defined
Risk Assessment (PDRA) UKPDRA01 which enables Visual Line of Sight (VLOS) operations
within 150 m of any residential, commercial, industrial or recreational areas for UAS with a
Maximum Take-Off Mass of less than 25 kg. The operator was also the remote pilot, and he
held a valid Operation Authorisation - Permission for Commercial Operations (PfCO).

Regarding operational conditions and limitations of safe distances between the UA and
people, structures, and objects, UKPDRA01 states:

‘No flight within 50 metres of any uninvolved person6, except that during take‑off
and landing this distance may be reduced to 30 metres. Any overflight of
uninvolved people must be kept to a minimum.’
Footnote
2
Less than 6% battery charge (Mavic 2 Enterprise Series User Manual v1.8).
3
Mavic 2 Enterprise Series Disclaimer and Safety Guidelines v1.6, dated 01/2021.
4
Regulation (EU) 2019/947, Regulation (EU) 2019/945, The Air Navigation Order (ANO) 2016.
5
Operations Manual, Version 4.6, dated 31/01/2022.
6
An uninvolved person is someone not directly under control of the UA pilot, as defined within CAP722 and
ANO 2016.

© Crown copyright 2023 138 All times are UTC

AAIB Bulletin: 6/2023 DJI Mavic 2 Enterprise Zoom AAIB-28830

The operator’s operations manual sets out flight parameters and safe operating distances
from people, property and hazards in accordance with CAA CAP722 and UKPDRA01.
The flight was planned to operate within these restrictions and a pre-flight plan and risk
assessment were carried out by the pilot. The construction worker who was struck by the
UA was not under the control of the remote pilot and was classed as an ‘uninvolved person’.

Analysis

The flight was planned in accordance with applicable regulation and guidance for safe
distances between the UA and uninvolved people and structures. Due to the erratic flight
behaviour of the UA during its automatic landing, coupled with loss of flight control, the UA
subsequently breached the required safe distances and then struck the construction worker.

The battery was fully charged prior to the flight, was within the manufacturer’s recommended
number of cycles, and no defects were found to have affected its charge. During flight the
battery percentage detected by the UAS dropped at an unexpectedly high rate that resulted
in an automatic landing without notifying the pilot at the pre-set charge thresholds.

The first indication of an abnormal battery discharge rate was during flight control checks.
The pilot chose to continue with the planned flight as it would have been achievable using
the indicated 88% battery level. Subsequently, the UAS continued to detect a high discharge
rate which triggered the automatic landing.

After the flight the battery integral charge level indicator showed 50-75% charge. It was
not possible to determine the difference between this charge level and the battery power
displayed on the handheld controller during the flight. Flight log data did not explain the
discrepancy in battery level.

Discussion

The UAS measured a high loss of battery power in a short space of time, leading to an
uncontrolled landing where the distance between the UA and uninvolved people and
structures was compromised. This sequence of events resulted in the UA striking an
uninvolved person. It was not possible to determine a cause of the UA’s detecting a loss of
battery power or its flight behaviour.

Whilst the displayed level of power remaining was sufficient for the planned flight, there may
have been an opportunity during flight control checks for the pilot to consider the abnormal
battery discharge rate was likely to continue, and to abort the flight.

© Crown copyright 2023 139 All times are UTC

AAIB Bulletin: 6/2023

AAIB Record-Only Investigations

This section provides details of accidents and incidents which
were not subject to a Field or full Correspondence Investigation.

They are wholly, or largely, based on information

provided by the aircraft commander at the time of reporting
and in some cases additional information
from other sources.

The accuracy of the information provided cannot be assured.

© Crown copyright 2023 141 All times are UTC

AAIB Bulletin: 6/2023 Record-only investigations reviewed: March - April 2023

Record-only investigations reviewed: March - April 2023

10 Mar 2020 Squarecraft Cavalier G-AZHH Rufforth East Airfield, North Yorkshire
SA.102-5
The aircraft overshot the runway during landing, damaging the engine,
cowling and propeller. The owner has since modified the brakes on the
aircraft.

9 Dec 2022 Grob G115A G-GPSX Wolverhampton Halfpenny Green

Airport
The aircraft bounced twice during landing before the student pilot, who was
performing her second solo flight, initiated a go-around. She pulled back
on the control column more than intended, causing the aircraft to stall. The
student pilot sustained minor injuries and the aircraft was damaged beyond
economic repair.

1 Mar 2023 Rockwell N6081F Gloucestershire Airport

Commander 114
Following a bounced landing, the nose gear collapsed and the aircraft
veered off the runway onto the grass.

2 Mar 2023 Thruster T600N 450 G-CBGV Freshwater Farm, Isle of Wight
In the final stages of the approach the engine stopped and could not be
restarted. The pilot carried out an emergency landing in a nearby field.
During the landing, the nose of the microlight dug in and the aircraft became
inverted, damaging the forward fuselage.

3 Mar 2023 Zenair CH 750 G-CLYN Goodleigh, Devon

During the roll-out and at low speed the aircraft began to drift sideways
in long grass. The wingtip dropped and touched the ground causing the
aircraft to become inverted.

11 Mar 2023 Avid Hauler Mk 4 G-BWRC East Winch Airfield, Norfolk

During a forced landing following an engine failure, the nose landing gear
and main landing gear collapsed.

27 Mar 2023 Avid Aerobat G-BUON Near White Fen Farm Airfield,
(Modified) Cambridgeshire
Shortly after takeoff the engine stopped. The pilot managed to carry out
an emergency landing in a field but, during the landing, the aircraft struck a
depression, breaking the nose landing gear and flipping the aircraft onto its
back. The pilot attributed the engine failure to the fuel cock not being fully
open which had been overlooked during pre-takeoff checks.

© Crown copyright 2023 143 All times are UTC

AAIB Bulletin: 6/2023 Record-only investigations reviewed: March - April 2023

Record-only investigations reviewed: March - April 2023 cont

1 Apr 2023 Cessna F152 G-CIUU North Weald Airport, Essex

The instructional circuit took place in crosswind conditions. During the
landing flare the aircraft yawed to the left then veered right on touchdown,
departing the runway onto soft ground. The aircraft sustained damage
to the right wingtip fairing. The instructor reflected that he should have
intervened earlier to correct the yaw before touchdown.

2 Apr 2023 DHC-1 Chipmunk 22 G-BBNA Husbands Bosworth Airfield,

(Lycoming) Leicestershire
After touchdown, control of the aircraft was lost and it left the runway and
ran into a hedge. The pilot received only minor injuries but the aircraft
suffered substantial damage to the wings and engine.

2 Apr 2023 Quik GT450 G-CFGD Watnall Airfield, Notthinghamshire

The aircraft encountered a gust before touchdown causing it to bounce.
The pilot began a go-around but the aircraft veered left and struck a
hedge.

3 Apr 2023 Ikarus C42 FB UK G-CBVY Wingland Airfield, Spalding,

Lincolnshire
The pilot considered that after touchdown he was too fast to stop within the
remaining runway length and so initiated a go around. The aircraft was left
of the runway centreline and as power was applied the left wingtip struck
a vehicle parked just off the runway in a marked parking area. The aircraft
spun through 180° and came to a halt.

13 Apr 2023 X’air Hawk G-CHIW Perranporth Airfield, Cornwall

The nose landing gear collapsed on landing and the aircraft came to a halt.

16 Apr 2023 Ikarus C42 FB80 G-CDMS Popham Airfield, Hampshire

On the approach to land, the student pilot rounded out too high and then
immediately corrected by lowering the aircraft nose. The aircraft touched
down on the nosewheel, which broke free from its leg and the propeller
struck the ground.

27 Apr 2023 Bolkow BO 208C G-AVLO Brighton City (Shoreham) Airport,

Junior West Sussex
Following an approach that the pilot thought was normal, the aircraft
bounced on touchdown. The pilot elected to continue with the landing, but
the aircraft bounced two more times and, on the final touchdown, the nose
landing gear collapsed and the propeller struck the ground. The pilot had
been concerned after the first bounce that the high nose attitude might have
led to a stall if he went around, so he decided to continue with the landing.

© Crown copyright 2023 144 All times are UTC

AAIB Bulletin: 6/2023

Miscellaneous
This section contains Addenda, Corrections
and a list of the ten most recent
Aircraft Accident (‘Formal’) Reports published
by the AAIB.

The complete reports can be downloaded from

the AAIB website (www.aaib.gov.uk).

© Crown copyright 2023 145 All times are UTC

AAIB Bulletin: 6/2023 G-ELMR AAIB-28781

BULLETIN CORRECTION

Aircraft Type and Registration: BB85Z hot air balloon, G-ELMR

Date & Time (UTC): 13 September 2022 at 1730 hrs

Location: Deighton, North Yorkshire

Information Source: Aircraft Accident Report Form submitted by the

pilot and subsequent information submitted by
passengers

AAIB Bulletin No 2/2023 refers

After publication the AAIB became aware of further details of the injuries to passengers
on the flight. As a result the report has been updated to take account of this further
information.

Bulletin header, Injuries

Corrected text:
Injuries: Crew - None Passengers - 1 (Serious)
3 (Minor)

Original text:
Injuries: Crew - None Passengers - 1 (Serious);
1 (Minor)

Page 104, final sentence of Synopsis

Corrected text:

During the landing sequence four of the passengers were injured.

Original text:

During the landing sequence two of the passengers were injured.

Page 105, final sentence of History of the flight

Corrected text:

One of the passengers sustained a serious head injury with others reporting neck and limb
injuries.

Original text:

One of the passengers sustained an injury to their neck and another reported similar injuries
later.

© Crown copyright 2023 147 All times are UTC

AAIB Bulletin: 6/2023 G-ELMR AAIB-28781

Page 105, Analysis

Corrected text:

Having realised that ground contact short of his planned field was unavoidable, the pilot
gave the landing instructions to the passengers. The lateness of the instructions and
the background noise of the burners may have meant the passengers did not hear the
instructions clearly or in time. The balloon touched down heavily, before bouncing and
coming to rest in a ditch. Four of the passengers were injured.

Original text:

Having realised that ground contact short of his planned field was unavoidable, the pilot
gave the landing instructions to the passengers. The balloon touched down heavily, before
bouncing and coming to rest in a ditch. Two of the passengers were injured, possibly due
to not being in the correct landing position throughout the sequence.

Page 105, Conclusion

Corrected text:

Four passengers on the flight were injured during the landing sequence, possibly due to
being unable to adopt the correct landing position in time.

Original text:

Two passengers on the flight were injured during the landing sequence, possibly because
they did not maintain their briefed and demonstrated landing position.

The online version of this report was corrected on 8 June 2023.

© Crown copyright 2023 148 All times are UTC

 AAIB Bulletin: /2023

TEN MOST RECENTLY PUBLISHED

FORMAL REPORTS
ISSUED BY THE AIR ACCIDENTS INVESTIGATION BRANCH

1/2015 Airbus A319-131, G-EUOE 1/2017 Hawker Hunter T7, G-BXFI

London Heathrow Airport near Shoreham Airport
on 24 May 2013. on 22 August 2015.
Published July 2015. Published March 2017.

2/2015 Boeing B787-8, ET-AOP 1/2018 Sikorsky S-92A, G-WNSR

London Heathrow Airport West Franklin wellhead platform,
on 12 July 2013. North Sea
Published August 2015. on 28 December 2016.
Published March 2018.
3/2015 Eurocopter (Deutschland)
EC135 T2+, G-SPAO 2/2018 Boeing 737-86J, C-FWGH
Glasgow City Centre, Scotland Belfast International Airport
on 29 November 2013. on 21 July 2017.
Published October 2015. Published November 2018.

1/2016 AS332 L2 Super Puma, G-WNSB 1/2020 Piper PA-46-310P Malibu, N264DB
on approach to Sumburgh Airport 22 nm north-north-west of Guernsey
on 23 August 2013. on 21 January 2019.
Published March 2016. Published March 2020.

2/2016 Saab 2000, G-LGNO 1/2021 Airbus A321-211, G-POWN

approximately 7 nm east of London Gatwick Airport
Sumburgh Airport, Shetland on 26 February 2020.
on 15 December 2014. Published May 2021.
Published September 2016.

Unabridged versions of all AAIB Formal Reports, published back to and including 1971,
are available in full on the AAIB Website
http://www.aaib.gov.uk

© Crown copyright 2023 149 All times are UTC

 Air Accidents Investigation Branch
Farnborough House AAIB Bulletin: 6/2023
Berkshire Copse Road
Aldershot GLOSSARY OF ABBREVIATIONS
Hants GU11 2HH
aal above airfield level kt knot(s)
ACAS Airborne Collision Avoidance System lb pound(s)
ACARS Automatic Communications And Reporting System LP low pressure
Tel: 01252 510300
ADF Automatic Direction Finding equipment LAA Light Aircraft Association
Fax: 01252 376999 AFIS(O) Aerodrome Flight Information Service (Officer) LDA Landing Distance Available
Press enquiries: 0207 944 3118/4292 agl above ground level LPC Licence Proficiency Check
http://www.aaib.gov.uk AIC Aeronautical Information Circular m metre(s)
amsl above mean sea level mb millibar(s)
AOM Aerodrome Operating Minima MDA Minimum Descent Altitude
APU Auxiliary Power Unit METAR a timed aerodrome meteorological report
ASI airspeed indicator min minutes
ATC(C)(O) Air Traffic Control (Centre)( Officer) mm millimetre(s)
ATIS Automatic Terminal Information Service mph miles per hour
ATPL Airline Transport Pilot’s Licence MTWA Maximum Total Weight Authorised
AAIB investigations are conducted in accordance with BMAA British Microlight Aircraft Association N Newtons
BGA British Gliding Association NR Main rotor rotation speed (rotorcraft)
Annex 13 to the ICAO Convention on International Civil Aviation, BBAC British Balloon and Airship Club Ng Gas generator rotation speed (rotorcraft)
EU Regulation No 996/2010 (as amended) and The Civil Aviation BHPA British Hang Gliding & Paragliding Association N1 engine fan or LP compressor speed
CAA Civil Aviation Authority NDB Non-Directional radio Beacon
(Investigation of Air Accidents and Incidents) Regulations 2018. CAVOK Ceiling And Visibility OK (for VFR flight) nm nautical mile(s)
CAS calibrated airspeed NOTAM Notice to Airmen
cc cubic centimetres OAT Outside Air Temperature
The sole objective of the investigation of an accident or incident under these CG Centre of Gravity OPC Operator Proficiency Check
Regulations is the prevention of future accidents and incidents. It is not the cm centimetre(s) PAPI Precision Approach Path Indicator
CPL Commercial Pilot’s Licence PF Pilot Flying
purpose of such an investigation to apportion blame or liability. °C,F,M,T Celsius, Fahrenheit, magnetic, true PIC Pilot in Command
CVR Cockpit Voice Recorder PM Pilot Monitoring
DME Distance Measuring Equipment POH Pilot’s Operating Handbook
Accordingly, it is inappropriate that AAIB reports should be used to assign fault EAS equivalent airspeed PPL Private Pilot’s Licence
or blame or determine liability, since neither the investigation nor the reporting EASA European Union Aviation Safety Agency psi pounds per square inch
ECAM Electronic Centralised Aircraft Monitoring QFE altimeter pressure setting to indicate height above
process has been undertaken for that purpose. EGPWS Enhanced GPWS aerodrome
EGT Exhaust Gas Temperature QNH altimeter pressure setting to indicate elevation amsl
EICAS Engine Indication and Crew Alerting System RA Resolution Advisory
EPR Engine Pressure Ratio RFFS Rescue and Fire Fighting Service
ETA Estimated Time of Arrival rpm revolutions per minute
ETD Estimated Time of Departure RTF radiotelephony
FAA Federal Aviation Administration (USA) RVR Runway Visual Range
FDR Flight Data Recorder SAR Search and Rescue
FIR Flight Information Region SB Service Bulletin
FL Flight Level SSR Secondary Surveillance Radar
ft feet TA Traffic Advisory
AAIB Bulletins and Reports are available on the Internet ft/min feet per minute TAF Terminal Aerodrome Forecast
http://www.aaib.gov.uk g acceleration due to Earth’s gravity TAS true airspeed
GNSS Global Navigation Satellite System TAWS Terrain Awareness and Warning System
GPS Global Positioning System TCAS Traffic Collision Avoidance System
GPWS Ground Proximity Warning System TODA Takeoff Distance Available
hrs hours (clock time as in 1200 hrs) UA Unmanned Aircraft
HP high pressure UAS Unmanned Aircraft System
hPa hectopascal (equivalent unit to mb) USG US gallons
IAS indicated airspeed UTC Co-ordinated Universal Time (GMT)
This bulletin contains facts which have been determined up to the time of compilation. IFR Instrument Flight Rules V Volt(s)
ILS Instrument Landing System V1 Takeoff decision speed
Extracts may be published without specific permission providing that the source is duly acknowledged, the material is
IMC Instrument Meteorological Conditions V2 Takeoff safety speed
reproduced accurately and it is not used in a derogatory manner or in a misleading context.
IP Intermediate Pressure VR Rotation speed
IR Instrument Rating VREF Reference airspeed (approach)
Published 8 June 2023 Cover picture courtesy of Marcus Cook
ISA International Standard Atmosphere VNE Never Exceed airspeed
© Crown copyright 2023 ISSN 0309-4278 kg kilogram(s) VASI Visual Approach Slope Indicator
KCAS knots calibrated airspeed VFR Visual Flight Rules
Published by the Air Accidents Investigation Branch, Department for Transport KIAS knots indicated airspeed VHF Very High Frequency
Printed in the UK on paper containing at least 75% recycled fibre KTAS knots true airspeed VMC Visual Meteorological Conditions
km kilometre(s) VOR VHF Omnidirectional radio Range
AAIB Bulletin 6/2023

AAIB Bulletin 6/2023

AAIB Bulletin 6/2023

TO REPORT AN ACCIDENT OR INCIDENT

PLEASE CALL OUR 24 HOUR REPORTING LINE
01252 512299

AAIB
Air Accidents Investigation Branch