Ultralight aircraft accidents, experimental aircraft accidents, light sport aircraft accident reports 14

 Ultralight News
Covering the World of Ultralight Aviation





Airfield
By Appt ONLY

AAIB Bulletin No: 7/98 Ref: EW/G98/05/02 Category: 1.3

 After take off from Shobdon the right hand side of the top engine cowl became loose and lifted off its seating, remaining attached by the left side pin and clip. As it lifted it caught the propeller, causing minor damage to the leading edge of the tip. The flapping cowl restricted the pilot's view and caused significant drag, affecting the control of the aircraft. It was not possible to climb and the aircraft remained at about 100 feet on the extended runway centreline. The pilot transmitted that he was going to make a left circuit, but as the turn started, the left wing dropped and the aircraft appeared to stall, crashing into an orchard containing young apple trees. The pilot of a following Robinson R22 witnessed the accident, landed in an adjoining field and after running to the aircraft turned off the battery master switch and fuel cock. Despite considerable damage to the fuselage both occupants sustained relatively minor injuries and were able to walk from the aircraft when the ambulance arrived. 

The engine top cowl is a one-piece construction, secured on either side by an overcentre latch to which an 'R' clip is then fitted as a safety device. The cowl had been opened before flight and, although the latch had been found closed with the clip fitted after the accident, there is a strong possibility that the latch had not been engaged over the corresponding attachment on the bottom cowl. It was reported that in this configuration the top cowl can appear to be flush when given a cursory inspection, but is in fact insecure. 

The prevalence of this type of accident has caused the manufacturer to produce a modification designed to prevent the cowl from opening in this event. The modification became available shortly before this accident and will initially be offered to kit builders. The UK distributor intends to notify other owners through the PFA, and to give additional publicity to the need for careful attention to the security of these latches during the pre-flight inspection by an article in their Newsletter. 

AAIB Bulletin No: 7/98 Ref: EW/G98/05/10 Category: 1.4

 The pilot had flown from Manchester (Barton) Airfield to a microlight strip at Otherton; flight time was one hour and ten minutes. Prior to this flight, he had filled the fuel tank to full (32 litres) which should result in a maximum endurance of two hours forty minutes. The weather was good with a surface wind of 260°/08 kt and a cloud base of approximately 4,000 feet agl. 

On the ground at Otherton, the pilot tried to obtain some fuel but the only fuel available belonged to another pilot who was airborne and the pilot of G-MZIY was also not confident that the fuel contained the correct ratio of oil. Additionally, he checked the fuel tank and considered that it was still more than half full. With the slight head wind he had experienced on his outbound flight, he was confident that he had sufficient fuel for his return flight. This return flight progressed normally until approximately two miles from Barton Airfield, after 57 minutes flying, when the engine stopped. The pilot was in radio contact with Barton and in sight of the airfield; he immediately turned into wind and informed Barton of his intention to make a forced landing. He was unable to make his initial landing site and so selected a smaller rougher looking field. Touchdown was on the rear wheels but, as the nosewheel contacted the soft soil it detached and G-MZIY turned onto its back. The pilot assisted his passenger out of the aircraft before informing Barton Radio that they were not injured.

The pilot acknowledged that the most likely cause of the accident was that he had run out of fuel; he could not confirm if any fuel was still in the tank when the engine stopped because the aircraft had inverted after ground contact. The fuel tank is in the wing root and it is easy to see the fuel level; he had checked that it still contained fuel as he passed the visual reporting point at Thelwall Viaduct some 6 miles from Barton. However, he acknowledged that the aircraft may not have been level when he visually checked the fuel tank contents at Otherton and therefore his assessment of fuel prior to departure from there may have been incorrect.

AAIB Bulletin No: 3/98 Ref: EW/G98/01/06 Category: 1.4

The pilot flew his machine to RAF Boulmer during an afternoon of recreational flying. The weather was fine with a light wind from the south when the pilot landed on the grass manoeuvring area using a south-easterly track. He expected the landing roll to be longer than was the case but otherwise the landing was normal. Whilst on the ground at Boulmer for about an hour the pilot liaised with RAF personnel. He also noticed that some areas of the grass manoeuvring area were soft underfoot due to recent rain and some areas had longer grass than others. Having explored the manoeuvring area, he decided to take off on a southerly heading adjacent to the eastern boundary because this run provided him with the firmest ground. The length of the run was some 400 metres.

As the pilot began the take-off roll the wind was calm. Moreover the aircraft accelerated more slowly than he had expected due to the soft surface. In an attempt to unload the landing gear he pulled back on the stick early at about 43 mph before reaching that part of the take-off run where the grass was longer. The aircraft became airborne and the pilot then kept it low with the intention of accelerating in ground effect. Unfortunately the airspeed had reached only 50 mph as he approached the southern boundary fence so he pulled back on the stick to clear the fence. At that point the airspeed started decreasing and the aircraft probably stalled just before the left wing struck the top of the fence. The aircraft cartwheeled and came to rest in an adjacent farmer's field. Neither occupant was hurt but the engine compartment caught fire as they extricated themselves from the wreckage. The fire was swiftly extinguished by RAF personnel.

In his report the pilot attributed the cause of the accident to his making insufficient allowance for the drag on the landing gear during the early part of the take-off roll.

AAIB Bulletin No: 3/98 Ref: EW/C97/8/8 Category: 1.4

History of the flight

The pilot had intended to fly from Netherthorpe Airfield to Cromer. The meteorological conditions were good with no significant weather or cloud, the visibility was 20 km, there was a light south westerly drift and, although the surface temperature was +28_C, there was no significant thermal activity reported. The pilot removed the aircraft from the hangar, readied it for the flight and completed the pre-flight checks; it had about 27 litres of fuel on board. The take off, from Runway 24, appeared to be normal until at a height of about 150 feet the engine note altered, suggesting a reduction in power. The aircraft appeared to level off briefly before recommencing the climb as the sound of the engine returned to normal for climb power. The aircraft then levelled off again and after a brief period of straight flight it entered a turn to the left at a low altitude, the angle of bank continued to increase until the aircraft dived into the ground. The aircraft crashed about 1,000 metres from the threshold of Runway 24 and 500 metres to the left of the extended centreline. Although the pilot had a hand held radio with him no emergency call was made.

A small group of people from the airfield immediately went to the scene and they were joined shortly afterwards by the airfield fire truck. Elements of the wreckage were then cut away in order to free the pilot whilst one of the rescuers tried to minimise a fuel leak. The emergency services arrived shortly afterwards but the pilot had already died of severe multiple injuries. A post mortem found no evidence of any disease or alcohol which may have contributed to the cause of the accident.

Description of the aircraft

This Spectrum was a two place, tandem seat, three axis trainer with conventional wing, tail and control surfaces of composite construction. It was powered by a Rotax 503 piston engine. An approved major modification had been completed in 1995 to incorporate a three bladed IVO propeller and a second carburettor. The aircraft was fitted with dual controls. The front cockpit controls consisted of a conventional stick and rudder, the throttle was on the left hand side and the engine instruments, ignition switches and choke were in the roof above the instrument panel. The rudder pedals for use by a pilot in the rear seat were situated either side of the front seat and the rear stick was placed just outboard of the front pilot's right hip position. The fuel tank was mounted behind the rear seat and held a maximum of 35 litres of petrol/oil mix (50:1). The pilot had recently purchased this microlight and this was to be his first flight in it.

Pilot experience

The pilot had obtained his Private Pilot's Licence in 1989 and had upgraded this to a Basic Commercial Pilot's Licence in 1991. The last recorded entry in his flying log book was for 10 July 1997 and it is therefore not possible to define the hours flown in the period prior to the accident. However, in the two months prior to 10 July 1997 he had recorded 38 hours of flying. His flying had been in light single engine aircraft and his only recorded flying in microlight aircraft had been in April 1997 when he had completed six flights in a Pegasus AX2000 for a total of 5 hours and 15 minutes. This microlight also had conventional three axis controls albeit with side by side seating for the pilots.

Examination of the wreckage

The aircraft had struck the ground in a nearly vertical attitude with the left wing leading. It was heading roughly north-west but tracking about 045_(M) with little speed over the ground but a high rate of descent. The wreckage travelled about 15 metres to the north east during the impact. It was completely destroyed in the impact, but there was no fire.

Damage to the engine was minor and plenty of fuel was found in the lines to the two carburettors. No mechanical distress or evidence of pre-impact malfunction was found. The propeller is of rather unusual construction, being composite with internal metallic reinforcing. This, together with the very steep impact attitude and soft ground, combined to make assessment of engine power from the propeller difficult. One effect of the very steep impact was to provide a reliable indication of engine speed from the tachometer, this was solidly jammed at 6,500 RPM. Maximum RPM is 6,900. It was concluded that the engine was delivering normal power at impact. There was limited evidence that the engine had not been running for very long before take off, and this could possibly have caused somewhat erratic behaviour until it had completely warmed up.

Other indications from the instruments were that the altimeter subscale was set to 1021 millibars but the altimeter was reading -500 feet. The Air Speed Indicator was reading 58 kt but the nature of damage to the aneroid mechanism made this an unreliable indication.

A large bag weighing 20 lbs whose contents included: a large plastic container and a 1 gallon container, a fuel funnel and some other small items were found in the wreckage, together with a quantity of blue nylon rope which, together with the rear seat belts had been used to secure the load. The 1 gallon container was found beneath the rear seat, in an area where flying control cables and the pitch and roll control mechanisms run. The handle of the container had been broken outwards in a manner which was consistent with it having been tied through the handle with the rope. The damage to the handle included large amounts of permanent deformation and appeared to have been a result of large forces. Tests showed that the forces required to do this type of damage were larger than could be applied manually which would have occurred, for example, if it had been jammed in the area of the flying controls, and it was concluded that this was an impact feature. Due to splitting of both containers, it was not possible to establish if they had contained any fluids before impact, but neither contained any residual fluids on site. Both containers appeared to have been used for fuel at some time.

The baggage, containers, funnel and other small items found were loaded into a similar aircraft. It was found that the bag and large container could be conveniently lashed into the rear seat, but there was no practical means of stowing the other items. The only possible locations were to either side of the rear seat, behind the rear seat or at the pilot's feet. The latter was discounted and behind the rear seat control runs were visible and items would have to be tied to prevent them falling either through the runs or through an aperture at the bottom of the aircraft. It was felt that this was an improbable location, although anything stowed in this area could move under the rear seat. The 1 gallon container was found to fit snugly to the right of the rear seat, without necessarily fouling the rear seat control column, just forward of it. This would have left the funnel and loose items to be stowed. To the left of the rear seat was a similar aperture with, at first sight, no control cables although the controls were just out of sight and unprotected. It was possible to secure the remaining items in this location. This would have permitted movement of any of these items on either side of the seat to provide a distraction due to control restrictions, or even a control jam. Although the location of all these items on the accident flight will never be known, it is difficult to see how they could have otherwise been stowed. With the weight of the pilot and the weights of the bag and containers on board, the weight and balance was calculated and found to be well within limits.

In summary, no technical malfunction could be found which may have caused the accident but the possibility exists for interference between the items on board and the flying controls in all three axes.

Analysis

Shortly after the take-off the engine noise was heard to reduce as the aircraft levelled off briefly. The previous owner explained that it was normal for the pilot to ease back on the throttle at this stage in order to ensure that the fuel tank would provide an adequate fuel flow to the high mounted engine throughout the take-off and climb. The previous owner had briefed the pilot on this requirement. The engineering evidence concludes that the engine was delivering normal power at impact and it is therefore probable that this perceived power reduction after take-off was a deliberate action by the pilot.

The physical evidence indicates that the aircraft struck the ground with a high rate of descent and little forward speed. The eyewitnesses describe the angle of bank increasing throughout the turn until the aircraft dived into the ground. These accounts are consistent with the pilot having experienced a problem in the turn which led to either an aerodynamic stall or an inability to recover from a spiral dive.

The pilot would have experienced unusual pitch attitude cues when seated in the Spectrum since it has a very low coaming. These cues would have been significantly different to those that he would have been familiar with in light aircraft. This may have caused the pilot some minor problems in controlling the airspeed initially and thus his proximity to the stall. Furthermore, the margin of pre-stall buffet in this aircraft is only about 3 kt. However, centralising the controls at the onset of the stall warning results in an immediate recovery and the recovery from a spiral manoeuvre is also straightforward for a pilot of this experience. It is therefore probable that the pilot's failure to retain control of the aircraft during the turn may have been due to a control restriction, or other distraction, and this may have been directly related to the manner in which he had loaded the various items of baggage.

AAIB Bulletin No: 2/98 Ref: EW/G97/10/02 Category: 1.3

The aircraft was taking off from Goodwood on Runway 32. At about 100 feet agl the engine stopped and the pilot decided to execute a forced-landing straight-ahead into a ploughed field. Although the touchdown was normal, the rough nature of the field caused the left mainwheel to partially detach from the axle but this was the only apparent damage.

Examination showed that the section of exhaust pipe outboard of the silencer box had detached and was found lying on the runway. Hot gases from the exhaust had thus impinged on the adjacent gascolator causing a vapour lock in the fuel system. The pilot informed the UK agent for the Pulsar aircraft who advised that he had not heard of any similar occurrences. However, the agent has promulgated this incident in a Newsletter which is circulated to all registered owners of the type and also the Popular Flying Association. The agent also advised that his personal aircraft had been fitted with a heat shield around the exhaust system which would probably have prevented an engine failure of this nature.

AAIB Bulletin No: 2/98 Ref: EW/C95/10/5 Category: 1.4

History of flight

The pilot was on a qualifying solo cross-country flight for the award of a Private Pilot's Licence. The planned route was from Hunsdon Airfield, near Harlow, Essex to Headcorn Airfield in Kent, where the aircraft would be refuelled for the return flight to Hunsdon. There was low cloud in the area and the pilot was briefed to return to Hunsdon if this became a problem; the planned altitude was 1,500 feet initially and then 2,000 feet.

The pilot had programmed his route into his Global Positioning System (GPS) equipment. Data stored in the GPS was used to determine, in part, the history of the flight. At 1032 hrs, the aircraft took off from Hunsdon Airfield and flew a complete left hand circuit before tracking south down the western side of Harlow. At 1042 hrs the aircraft turned left and tracked east, over Junction 7 of the M11 towards the village of Moreton. At 1051 hrs, shortly before it reached Moreton, it made a 180° turn to the left and tracked along the southern edge of Harlow. It then turned to track north and, shortly afterwards, the GPS stopped logging data. The last relevant position logged was at 1058 hrs after the aircraft had started a left turn in a position about 260 metres northeast of the accident site. The final part of the track was confirmed by comparing it with recorded data from Heathrow radar; the time of the final return was also 1058 hrs.

Witnesses were consistent in their observations. They reported hearing a sharp 'crack' and seeing the left wing move up almost to the vertical. Shortly afterwards the right wing did the same and, the aircraft spiralled to the ground with the wings folded together and the trike assembly describing a circle beneath them. The aircraft was in the vicinity of the final GPS/radar contact when the observations were made and one witness called the emergency services while it was still falling; the call was recorded at 1058 hrs. The AAIB was unaware of anyone who saw the aircraft in normal flight.

The pilot survived the impact but died shortly afterwards from his injuries. Post mortem examination revealed no pre-existing medical condition which would have contributed to the accident.

Pilot's flying experience

The pilot started a course of microlight flying training in March 1997. Initial training was in a Mainair Blade aircraft. His first flight in the Gemini Flash IIA, G-MVEP, was on 26 August 1997. He made good progress through the course and went solo on 20 September 1997 after 19 hours. His last flight before the accident was on 26 October 1997, after which he had flown a total of 19 hours dual and 1140 hours solo.

Meteorology

An aftercast was obtained from the Meteorological Office at Bracknell. There was a weak cold front lying from Kings Lynn to Folkestone; it was moving slowly westwards.

Surface wind 100°/10 kt
2,000 feet wind 130°/15 kt
Visibility 9 km
Cloud FEW base 600 feet
SCT base 1,000 feet
BKN base 4,500 feet
Temp/Dew point +10C/+8C
QNH 1027 mb

Global positioning system

The GPS equipment carried in the aircraft worked throughout the accident flight. It uses satellite navigation to establish its position which it then displays on a small screen. The equipment has a logging facility which saves position, time and date into non-volatile memory, on a periodic basis when direction or speed change significantly. The equipment was found to operate satisfactorily after the accident.

The data logged during the accident flight was downloaded onto a computer. As the downloaded data was based on a datum other than that used for Ordnance Survey maps, a correction factor was applied to each latitude/longitude point logged and the results plotted onto a 1 to 25,000 map. The plotted points correlated well with the known positions of the take off and the accident. It was not possible to determine the height of the aircraft during any part of the flight as GPS altitude, although displayed on the screen, is not logged in the non-volatile memory.

From 1058 hrs, the GPS stopped logging data for a period of twenty four minutes. It is considered that this was because the GPS was unable to receive sufficient satellite signals to calculate its position.

It was probably carried in the document pocket on the right side of the aircraft and consequently the accuracy of the recorded data may have been degraded because of satellite obscuration by the pilot or the metal framework of the aircraft. This degradation may have been more noticeable when tracking west as five of the eight satellites in view would have been behind the aircraft.

Further geographical positions consistent with that of the accident site were recorded in the equipment; two points at 1122 hrs, two points at 1229 hrs and a period of uninterrupted operation from 1406 hrs to 1443 hrs. It is likely that, as the aircraft wreckage was moved during the post accident recovery, the GPS may have periodically received adequate satellite signals before it was eventually turned off when the AAIB arrived on the scene.

Accident Site

The microlight was found laying on its left hand side with the wing folded in two. The only significant ground mark had been made by the muffler and did not show any signs of aircraft rotation. The guarded magneto switch was found in the 'off' position. The aircraft was carrying 50 kg of ballast in an appropriate container.

The propeller had disintegrated and the majority of it was retrieved from a 400 metre long trail, however only 45% of the leading edges was recovered. The fabric nose cone was found 275 metres from the accident site along the line of the propeller debris. No vibration damage was apparent on the engine mounts, muffler, air filter or engine casing.

Subsequent examination at the AAIB facility at Farnborough revealed heavy indentations on the lower left hand side of the keel, and wood dust embedded in the left vertical face of the keel fabric in the area over the indentations. The indentations were very similar to keel damage on another wing which was known to have had a propeller strike on the keel.

The following structural damage had occurred before the microlight hit the ground:
The outer 1.33 m of both leading edge tubes had broken in a downwards direction.
The right hand leading edge tube had detached from the nose plate.

The control bar had failed at the inner end of the right hand grip by being forced through the front strut; the front strut had a corresponding slight bend.

The wires on the upper wing did not show any curling typically associated with excessive tension, and the king post was undamaged by compression. The structural damage was consistent with negative 'g' and the trike 'dropping into the wing'. The aircraft had a renewal of its certificate of validity on 30 July 1997 and had flown for a total of 152 hours, and the log book did not contain any record of significant maintenance actions.

The ASI and the altimeter were checked and found to be satisfactory. Detailed examination of the microlight revealed the failure of a muffler spring and an upper side wire ferrule. Whilst these failures had occurred after the loss of control and subsequent structural failure, the details are given below in the interests of flight safety.

Muffler Spring Failure

The hook end on one of the three muffler springs had failed and, although the area on the accident site around the muffler impact mark had been examined closely, the failed end was not recovered. The fracture surfaces were therefore examined to determine whether the failure had been caused by fatigue, and hence the hook could have struck the propeller, or whether the failure had been in overload.

The fracture surface had started from a pre-existing groove in the spring material, and the failure had then propagated to the circumference of the spring. The groove was in a position that could only have been reached by the locking wire used to retain the spring body to the muffler in the event of the spring failing. However, the locking wire fitted showed no similar wear pattern. It was therefore concluded that the wire had been renewed, and that a previous locking wire had worn the groove. The spring had subsequently failed. There was no evidence to support the theory that the spring had failed before the propeller disintegration, however the vibration after the disintegration, or the ground impact could both have produced forces sufficient to overload and break the weakened spring.

Insufficient evidence of this accident has been accumulated to determine the cause with any degree of confidence. The AAIB therefore intend to rebuild the sail, using as many of the original parts as possible, with the initial objective of determining the pitch moment and stall characteristics of this specific wing. The results of these tests will be published when they are available.

AAIB Bulletin No: 11/97 Ref: EW/G97/08/15 Category: 1.3

This accident was reported to the police by a witness who saw an aircraft apparently in some difficulty. The police later found the aircraft in a tree.

The pilot reported that two to three minutes after getting airborne from South Wraxall microlight field, whilst at a height of about 350 feet, the engine lost power. The pilot then attempted to land in an adjacent large field but the right wing struck the foliage of a tree causing the microlight to spin into it. The pilot was wearing a full harness at the time of the accident and was uninjured. He vacated the aircraft and was then assisted by the land owner who was in the field at the time. There was no significant weather, the visibility was good and the winds were light. Subsequent examination of the engine did not reveal any apparent cause for the loss of power.

AAIB Bulletin No: 11/97 Ref: EW/G97/08/17 Category: 1.3

The aircraft was landing after a flight from Farthing Corner airfield. The pilot reported that he made a normal powered approach to the 440 metre long grass Runway 06 at a speed of 50 to 55 mph. The aircraft crossed the threshold at a height of about 15 feet on the centreline. The surface wind was southerly at 3 to 4 mph. During the roundout, at about 6 feet above the ground, the aircraft suddenly lost height, resulting in a premature and moderately heavy landing. The pilot was unable to control the aircraft which veered to the right. The right main wheel left the runway and entered the stubble of the recently cropped adjacent land. The nosewheel dropped into the shallow ploughed trough on the edge of the strip and collapsed, bringing the aircraft to a halt on its nose and left wing tip. A subsequent inspection of the aircraft revealed that the left main wheel tyre had also deflated.

NTSB Identification: FTW99LA007

Accident occurred OCT-10-98 at OSCEOLA, AR
Aircraft: Painton LOEHLE AVIATION 5151, registration: N81029
Injuries: 1 Fatal.

This is preliminary information, subject to change, and may contain errors. Any errors in this report will be corrected when the final report has been completed.

On October 10, 1998, approximately 1345 central daylight time, a Painton Loehle Aviation 5151 amateur-built experimental airplane, N81029, impacted the ground while maneuvering near Osceola, Arkansas. The airplane, which was owned and operated by the pilot, was destroyed, and the student pilot, the sole occupant of the airplane, was killed. Visual meteorological conditions prevailed and no flight plan was filed for the Title 14 CFR Part 91 local personal flight. The flight originated from the Osceola Municipal Airport approximately 10 minutes prior to the accident. A witness reported that he observed the airplane
make a "normal takeoff and [fly] a conservative right hand pattern." The airplane then "flew down runway 19 with the gear down about 70 mph at 100 feet above ground level." The witness commented that "the motor sounded good." Approximately 2 minutes later, the witness observed the airplane southeast of the airport "in a left hand turn 30 degree bank and nose down 30 degrees as if to fly a spray pass or as if to 'buzz' something." The "motor sounded normal" and "the pitch and bank did not shallow out" as the witness observed the airplane descend out of sight behind a tree line. The airplane impacted in a plowed field approximately 1 mile east of the Osceola Municipal Airport. According to local authorities, it appeared that the left wing
tip struck the ground first, then the nose of the airplane, after which the airplane rolled several times.  

AAIB Bulletin No: 8/98 Ref: EW/G98/06/09 Category: 1.3

Approximately 15 minutes after taking off from Cambridge airport for a local flight, the occupants of the aircraft became aware of a strong smell of fuel. An immediate turn-back to the airfield was made, the passenger door was unlatched to aid ventilation and all electrical selections were avoided. When the aircraft arrived overhead the airfield, the fuel cock was turned off and a glide approach was made to grass Runway 23.

Following an uneventful landing, the aircraft was allowed to run onto the southern taxiway in order to clear the runway. As the occupants vacated the aircraft, the pilot noted a flash in the passenger footwell together with a flame travelling along the doped fabric of the right hand side of the fuselage. The aircraft quickly became engulfed by fire and had been damaged beyond repair by the time the Airfield Fire Service arrived.

The pilot's assessment of the event was that a fuel leak had occurred in the supply lines between the tanks in the wings and the fuel cock, leading to an accumulation of fuel on the floor of the cockpit. When forward motion had ceased, the fuel was able to drip onto the exhaust pipe beneath the passenger footwell, and be ignited by the hot carbon deposits commonly found on two-stroke systems. Unfortunately, the damage to the aircraft was too severe to identify the origin of the fuel leak. The pilot did however recollect that he had experienced a transient resistance to rudder pedal movement whilst taxiing the aircraft, and considered the possibility that a rudder pedal had fouled a fuel line which may have become unclipped from an adjacent part of the tubular steel fuselage framework.

This incident was similar to another which occurred, in January 1998, to an Avid Aerobat, G-BUDH, and which was reported in AAIB Bulletin 7/98. That aircraft also suffered a fire following a strong smell of fuel in the cockpit, and a loose 'T-piece' connector in the fuel line was suspected.

AAIB Bulletin No: 8/98 Ref: EW/G98/06/03 Category: 1.3

 The aircraft had taken off on a flight from RAF St Athan to Oaksey Park, near RAF Benson, and was under the control of Cardiff ATC. After routing over Cardiff city at 1500 feet amsl and when mid-way between Cardiff and Newport, the engine began to run roughly. The pilot transmitted a 'Pan' call on the Cardiff RT frequency, requesting a diversion to the nearest airfield. Despite the aircraft's radio transmission being 'almost unreadable' by Cardiff ATC, the pilot was advised that Cardiff Airport was in fact the nearest airfield and given navigational assistance to avoid overflying the city. As a forced landing appeared likely, the pilot elected to remain to the North of the M4 motorway until abeam the Airport, and advised ATC accordingly. A light aircraft flying in the vicinity was requested by Cardiff ATC to follow 'IX'. The latter aircraft steadily lost height until at approximately 500 feet agl the pilot looked for a suitable field in which he could land into wind. However, at about 200 feet agl the engine failed completely and he carried out a forced landing in a field which had a down-slope. The pilot was unable to stop the aircraft before it collided with a fence, but at low speed. He was uninjured and there was no fire. The pilot of the other aircraft passed the position of the forced landing to Cardiff ATC. A South Wales Police helicopter (UKP 32) then flew to the landing site where it was ascertained that the pilot was unhurt; the crew therefore cancelled the call-out of the Emergency Services.

The aircraft was recovered to the UK Rans agent, with the intention of investigation and repair. However, initial examination has revealed no obvious mechanical failure associated with the engine. Further examination of the ignition and fuel systems will take place during the aircraft's re-build. If any significant defect(s) become apparent at that time, they will be reported upon in a future edition of the AAIB Bulletin.

 AAIB Bulletin No: 8/98 Ref: EW/G98/05/06 Category: 1.4

 The pilot had flown his aircraft from his home airfield at Hoghton, near Preston, to Kemble with an en-route stop at Shobdon for fuel. He reached Shobdon after 1 hour 40 minutes and refuelled with 25 litres, which was consistent with his expected fuel consumption of 14 to 15 litres/hour. Whilst performing his pre-flight engine checks, before taking off from Shobdon, the pilot noticed a slight hesitancy and roughness as the engine was accelerated. Consequently, he prolonged his engine checks but was unable to reproduce the rough running. He then flew uneventfully to Kemble, taking 1 hour, and before returning home later in the day, via Shobdon again, he refuelled with 15 litres; this also being consistent with his expected fuel consumption.

 The return flight from Kemble to Shobdon, by an indirect scenic route, was into a freshening northerly wind and took 1 hour 35 minutes. At Shobdon, he refuelled the aircraft with 27 litres. This was slightly more than he had expected, but he was uncertain whether he had refuelled to a full tank at Kemble. The pilot estimated, from his flight time to a point on this first leg of the return flight, that the headwind had been about 10 to 15 mph. He therefore anticipated this wind for the onward flight from Shobdon to Hoghton.

 During the pre-flight engine checks at Shobdon, he had a recurrence of the hesitancy and roughness which was more persistent than previously. As a result, he performed extended engine ground running, during which he replaced the spark plugs and cleaned the air filter on the forward carburettor which was wet with fuel. After further ground running of the engine, which then appeared to be operating normally, he took off for Hoghton. The pilot estimated that the total ground running time at Shobdon, after refuelling, had been about 22 minutes.

 However after take off the pilot observed, from his handheld Global Position System (GPS) unit, that the wind had increased considerably, beyond that forecast or his estimate based on the previous leg, and his ground speed at one point reduced to 30 mph. A re-estimation of the leg time was made (about 2 hours 10 minutes flight time) which the pilot calculated would give him a half hour fuel reserve at his destination, after allowing for the extra fuel used during the ground running. (The aircraft fuel tank capacity was 45 litres).

Two hours and 16 minutes after leaving Shobdon, when the aircraft was over the north west corner of Chorley with the fuel gauge indicating 1/8 contents remaining and an estimated 6 minutes from his destination, the engine suddenly stopped. The pilot selected a landing field just north of Chorley hospital, but as he manoeuvred for the approach he realised that he would land short since he had not allowed sufficiently for the wind. The pilot therefore sideslipped the aircraft to increase his rate of descent and landed firmly on a small grass area in the hospital grounds. The landing was sufficiently hard to break the right landing gear, but the pilot was able to release himself, uninjured, from the aircraft. 

Subsequent inspection of the aircraft by the pilot revealed that there was no fuel in either carburettor and an insignificant amount in the tank. There was evidence some oily fuel staining on the rear fuselage behind the engine.

AAIB Bulletin No: 7/98 Ref: EW/G98/01/03 Category: 1.3

The Avid Aerobat aircraft is a high-winged monoplane with a tailwheel landing gear. The pilot was taking off on grass Runway 27 following an earlier uneventful flight in the morning. Before take off he refuelled the aircraft with 4 Star petrol, conducted normal pre-flight checks and taxied the length of the runway in order to assess its condition. The grass was described as long and wet; the latter part of Runway 27 sloped downwards and terminated in a deep quarry.

The take off was normal but immediately after lift-off the pilot noticed a smell of petrol and decided to land straight ahead. After landing back he realised that the retardation was too low to stop the aircraft before the end of the runway and therefore applied the brakes fully which caused the aircraft to nose over and come to rest vertically nose down. The pilot was wearing a crash helmet, in accordance with his normal practice, which received some damage, but he was uninjured and managed to exit the aircraft rapidly, after turning off the engine ignition and electrical master switches. He took the aircraft fire extinguisher with him, a dry powder type. A fire started very shortly thereafter and rapidly spread to the cockpit and the remainder of the fuselage. The pilot pulled the safety ring on the extinguisher but was unable to depress the trigger and could not obtain any extinguishant to tackle the fire, which burnt out the fuselage and most of the wings.

The pilot believed that a fuel leak from a tee-piece connector associated with the fuel primer had occurred. He reported that this had previously worn and leaked, when the aircraft had accumulated 70 operating hours since new, and had been replaced. The aircraft had subsequently flown a further 26 hours until the accident. The connector was apparently destroyed in the fire. The Popular Flying Association (PFA) had not received other reports of problems with this type of connector, which is widely used in Rotax engine installations.

The reason for failure of the fire extinguisher was not established and it was disposed of, however, the pilot did note that it was about five years old and the accident occurred within one month of its 'use by date'

AAIB Bulletin No: 7/98 Ref: EW/G98/05/02 Category: 1.3

After take off from Shobdon the right hand side of the top engine cowl became loose and lifted off its seating, remaining attached by the left side pin and clip. As it lifted it caught the propeller, causing minor damage to the leading edge of the tip. The flapping cowl restricted the pilot's view and caused significant drag, affecting the control of the aircraft. It was not possible to climb and the aircraft remained at about 100 feet on the extended runway centreline. The pilot transmitted that he was going to make a left circuit, but as the turn started, the left wing dropped and the aircraft appeared to stall, crashing into an orchard containing young apple trees. The pilot of a following Robinson R22 witnessed the accident, landed in an adjoining field and after running to the aircraft turned off the battery master switch and fuel cock. Despite considerable damage to the fuselage both occupants sustained relatively minor injuries and were able to walk from the aircraft when the ambulance arrived.

The engine top cowl is a one-piece construction, secured on either side by an overcentre latch to which an 'R' clip is then fitted as a safety device. The cowl had been opened before flight and, although the latch had been found closed with the clip fitted after the accident, there is a strong possibility that the latch had not been engaged over the corresponding attachment on the bottom cowl. It was reported that in this configuration the top cowl can appear to be flush when given a cursory inspection, but is in fact insecure.

The prevalence of this type of accident has caused the manufacturer to produce a modification designed to prevent the cowl from opening in this event. The modification became available shortly before this accident and will initially be offered to kit builders. The UK distributor intends to notify other owners through the PFA, and to give additional publicity to the need for careful attention to the security of these latches during the pre-flight inspection by an article in their Newsletter.

AAIB Bulletin No: 7/98 Ref: EW/G98/05/10 Category: 1.4

The pilot had flown from Manchester (Barton) Airfield to a microlight strip at Otherton; flight time was one hour and ten minutes. Prior to this flight, he had filled the fuel tank to full (32 litres) which should result in a maximum endurance of two hours forty minutes. The weather was good with a surface wind of 260°/08 kt and a cloud base of approximately 4,000 feet agl.

On the ground at Otherton, the pilot tried to obtain some fuel but the only fuel available belonged to another pilot who was airborne and the pilot of G-MZIY was also not confident that the fuel contained the correct ratio of oil. Additionally, he checked the fuel tank and considered that it was still more than half full. With the slight head wind he had experienced on his outbound flight, he was confident that he had sufficient fuel for his return flight. This return flight progressed normally until approximately two miles from Barton Airfield, after 57 minutes flying, when the engine stopped. The pilot was in radio contact with Barton and in sight of the airfield; he immediately turned into wind and informed Barton of his intention to make a forced landing. He was unable to make his initial landing site and so selected a smaller rougher looking field. Touchdown was on the rear wheels but, as the nosewheel contacted the soft soil it detached and G-MZIY turned onto its back. The pilot assisted his passenger out of the aircraft before informing Barton Radio that they were not injured.

The pilot acknowledged that the most likely cause of the accident was that he had run out of fuel; he could not confirm if any fuel was still in the tank when the engine stopped because the aircraft had inverted after ground contact. The fuel tank is in the wing root and it is easy to see the fuel level; he had checked that it still contained fuel as he passed the visual reporting point at Thelwall Viaduct some 6 miles from Barton. However, he acknowledged that the aircraft may not have been level when he visually checked the fuel tank contents at Otherton and therefore his assessment of fuel prior to departure from there may have been incorrect.

On December 21, 1997, at 1515 central standard time, an unregistered amateur-built Quicksilver GT-500 airplane sustained substantial damage when it impacted the ground during an uncontrolled descent following the failure of the left leading edge wing strut fitting near Bat Cave Field Airport, San Antonio, Texas. The two-place kit-built airplane was being operated as an ultralight vehicle under an exemption to Title 14 CFR Part 103 granted to Aero Sports Connection, Inc., of Marshall, Michigan, to provide a means for single-place ultralight pilot training. The ultralight instructor pilot, who was the owner of the airplane, received serious injuries, and the student pilot, who was the owner's son, sustained minor injuries. No flight plan was filed and visual meteorological conditions prevailed for the local instructional flight that departed from Kitty Hawk Ultralight Field, San Antonio, Texas, at 1415.

In a written statement, the pilot reported that he had taken his "son/student for an orientation flight" and was passing over Bat Cave Field on the way back to Kitty Hawk "when the left leading edge strut bracket broke," and the airplane "entered a violent left hand spin." The pilot further reported that he "was able to deploy the Ballistic Recovery System [parachute] after several tries." Although the parachute slowed the rate of descent "somewhat," the airplane "continued down in a nose down spiral to the left until [it] struck the ground."

A witness, who was at Bat Cave Field, stated he heard a "loud pop" as the parachute deployed, and then saw the airplane descending under a partially inflated canopy. He reported that the airplane was "nose down approx. 45 degrees spiraling left" and "contacted the ground in this attitude" at an estimated speed of "25 to 30 mph."

The FAA inspector who examined the airplane reported that the nose section of the fuselage was crushed aft, the bottom of the fuselage was deformed upward and aft, and the nose landing gear was folded aft. The inspector further reported that the main line fastening the Ballistic Recovery Systems (BRS) parachute to the airplane was wrapped around the propeller drive shaft of the rear-mounted (pusher) engine. According to the inspector, the right wing appeared to be undamaged; however, the left wing was twisted with the leading edge displaced upward. The forward lift strut was detached from the left wing, while the rear lift strut remained attached to the left wing.

Closer examination of the left wing by the inspector revealed that the left leading edge wing strut fitting had separated from the leading edge spar and remained attached to the upper end of the forward lift strut. The cuff-shaped fitting was designed to slip around the tubular spar and fasten to the spar with three bolts inserted first through holes in the front of the fitting, then through holes in the tubular spar, and finally through holes in the rear of the fitting before being secured by nuts. On the accident airplane, the three bolts remained in place in the left leading edge spar with each bolt secured by a nut. Each of the three bolt holes in the front of the fitting was elongated in a direction perpendicular to the longitudinal axis of the wing spar to a point where it appeared that the bolt had torn out along the upper edge of the fitting. The three bolt holes in the rear of the fitting all appeared to be round and undistorted.

Examination of the undamaged right leading edge wing strut fitting by the inspector disclosed that the strut fittings were installed after the wings were covered with fabric. (Review of the "GT 500 Assembly, Maintenance and Parts Manual" confirmed this to be the recommended installation method.) A zipper in the fabric on the lower surface of the wing provided access to insert the fitting inside the wing and rotate it into position. The inspector noted that the rear side of the fitting was not visible without the aid of a flashlight and mirror.

In the section of the Pilot/Operator Aircraft Accident Report (NTSB Form 6120.1/2) entitled "Recommendation (How Could This Accident Have Been Prevented)," the pilot wrote:

Bracket was not installed correctly on initial construction of aircraft.

Three bolt holes were missed completely. This bracket was in an

area that could not be seen even with a very through pre-flight.

Records provided by Aero Sports Connection (ASC), Inc., of Marshall, Michigan, indicated that on March 27, 1996, the pilot was issued an authorization to operate under ASC's "2-Place Training Exemption No. 6080" for a 2 year period. Additionally, two certificates were issued by ASC to the pilot on March 27, 1996. One of the certificates indicated that he was awarded a "Basic Flight Instructor" rating, ASC number BTX000388, and the other certificate stated that he was issued vehicle registration number A10HDN. According to the "Vehicle Registration Request Form" submitted to ASC by the pilot, the registered vehicle was a 1995 Quicksilver GT-500.

During a telephone interview, conducted by the NTSB investigator-in-charge (IIC), and in communications, the Chief Executive Officer (CEO) of ASC stated that in January 1997, ASC adopted a policy requiring its authorized instructors to provide training only in vehicles which have a recorded condition inspection, current within the last 100 flight hours or 12 calendar months, whichever occurs first. The CEO further stated that in December 1996, a letter explaining the new policy and a suggested condition inspection checklist based on Title 14 CFR Part 43, Appendix D, were mailed to all ASC instructors.

Copies of the pages containing entries from the "Ultralight Aircraft & Engine Log" for the accident airplane were provided to the NTSB IIC by the FAA inspector. Review of the entries indicated that the airplane was first flown on February 19, 1995. The section of the log entitled "Ownership Records" showed that ownership of the airplane was transferred to the pilot on June 30, 1995. No entries were found in the log indicating the performance of any condition inspections on the airplane. The pilot did not complete the blocks on the Pilot/Operator Aircraft Accident Report asking for the date and time the last inspection of the airplane was performed or for the airframe total time. An entry in the log dated June 6, 1997, stated, "master switch left on, 24 hrs run up on Hobbs." Based on this entry and the hour meter (Hobbs) reading at the time of the accident (121.7) as reported by the FAA inspector, the total time on the airplane was calculated to be 97.7 hours.

AAIB Bulletin No: 12/98 Ref: EW/G98/07/25 Category: 1.4

The pilot reported that during the final leg of a cross-country flight, whilst flying at approximately 1000 feet amsl, the engine RPM decreased from 5,800 to approximately 3,800 and the unit began to run very roughly. Various throttle positions were tried, none of which improved the situation. After some 30 seconds the engine stopped completely, the height by then being 750 feet agl or less.

In view of this low height, no attempt was made to re-start the engine. The pilot judged that the only safe landing field available was directly beneath the aircraft and accordingly he commenced a steep right-hand turn through 360 degrees to position the aircraft into wind and facing up the local slope. Unfortunately, after straightening out the aircraft, insufficient airspeed and height remained to enable the aircraft to round-out into the uphill flight-path necessary for a landing. The aircraft stalled whilst about 10 feet agl.

Examination of the engine subsequently revealed that a needle in the jet of one of the carburettors had detached and fallen into the jet area. It detached as a result of breaking in the plane of the locating groove after severe machining type wear had been inflicted as a result of continuing rotation of the needle relative to its mounting circlip. This problem has been encountered in the past; it occurs under certain conditions of vibration and resonance. It led to the fatal accident of the similarly powered Colb Twin-Star, G-MWWF, in 1994, see AAIB Bulletin 11/94. As a result of that accident, the AAIB made the following comments and recommendation to the CAA:-

The CAA did not accept this recommendation. Their reasons were stated in their document 'Follow-Up Action On Occurrence Report ' No F30/94.

The modification ultimately engineered by the manufacturer is slightly different from that described above. It has now been available (at minimal cost) for retrospective installation for approximately 18 months. It is installed on current production engines, although the fact that suppliers and builders of kits frequently store engines for extensive periods results in new aircraft continuing to come into use with engines which may have been built before the modification became part of the production standard.

The aircraft involved in this accident is understood to have completed 49 hours total flying at the time of the accident.

When en route from Priory Farm to Shipdham Aerodrome and two miles south-west of Wymondham at a height of 1,300 feet, there was 'a loud thud' and the engine shook violently. The throttle was initially retarded and as all temperatures and pressures were in the normal range eased forward again. The engine again shook violently but continued to run. The pilot closed the throttle and looked for a landing site.

The aircraft was now at about 1,000 feet and established in the glide at 55/60 mph. The pilot selected a field below the aircraft with a mown grass strip and positioned downwind right-hand at about 600 feet. On reaching a position abeam the intended touchdown point at 400 feet he saw two sets of power cables crossing his intended approach path at right angles in the immediate undershoot. He turned in early to clear the cables and as a result, although committed to land, found himself too high to achieve a landing and stop in the distance available. After side slipping to lose height the speed had increased and so he landed the aircraft on an area of cut hay to the right and at the far end of his intended grass strip. The aircraft came to rest having turned through 180_ to the right from the original approach heading.

On examination the spinner was found to be missing and one of the wooden propeller blades was damaged and a piece recovered from rigging on the right wing. The pilot had cleaned the spinner and propeller during the pre-flight preparations and had not detected anything amiss.

Following a normal landing in benign conditions, the stub axle on the right main landing gear failed causing the aircraft to depart the runway.

The aircraft had been operated at consistently high all-up weights by a previous owner and it is probable that the failure was caused by damage accumulated over a considerable period.

The aircraft touched down nosewheel first and the nose landing gear subsequently collapsed. The landing was on Runway 29 at Headcorn and the surface wind was 260°/10 to 15 kt with slight gusts.

On June 9, 1998, at 1900 central daylight time, a Gisclair Murphy Renegade SP amateur-built experimental airplane, N96VG, impacted the ground following a loss of engine power on the initial takeoff climb from the South Lafourche Airport near Galliano, Louisiana. The airplane, which was registered to and operated by the pilot, was destroyed by the impact and a post-crash fire. The private pilot, the sole occupant and builder of the airplane, sustained minor injuries. No flight plan was filed and visual meteorological conditions prevailed for the Title 14 CFR Part 91 local personal flight.

According to the pilot, the Rotax 912UL engine began to "run rough at about 500 feet altitude." The engine rpm dropped from a normal cruise rpm of 5,200 rpm to 4,000 rpm and then to 3,000 rpm. The airplane, which was passing "over a cow pasture with bails of hay every 50 to 75 feet," began to lose altitude. The pilot "tried to maintain altitude in order to find a place to land." The airplane "stalled and the left wing dropped and nosed into the ground." The pilot exited the airplane, and a post-crash fire erupted and consumed the airplane.