Safety Mat­ters

Safety Mat­ters and Safety Briefs are based on the AAIB Bul­letin UK Air­prox Board re­ports, with ad­di­tional ma­te­rial from the US Na­tional Trans­porta­tion Safety Board

Pilot - - CONTENTS -

Slingsby spin fatalities, a cliff-hanger, and a Chi­nook close en­counter

Wrong spin re­cov­ery ac­tion?

Air­craft Type: Slingsby T67M MKII Fire­fly Date & Time: 30 April 2016 at 0938 Com­man­der’s Fly­ing Ex­pe­ri­ence: PPL, 215 hours, three on type Last 90 days: 34 hours Last 28 days: 14 hours The pi­lot and pas­sen­ger were stu­dents on the Tu­cano phase of their RAF fly­ing train­ing, hav­ing flown Grob Tu­tors dur­ing their ini­tial train­ing. Af­ter take­off from Full Sut­ton they tracked to­wards the Cas­tle Howard es­tate, where wit­nesses saw the T67 per­form a loop. On the down­ward half of the ma­noeu­vre the air­craft ap­peared to en­ter a spin, dur­ing which the en­gine was ini­tially heard to cut out, then seemed to restart be­fore cut­ting out again. The wit­nesses lost sight of the air­craft as it de­scended be­hind the roof of an out­build­ing, then heard the sound of an im­pact.

Other wit­nesses re­ported a ‘corkscrew­ing’ mo­tion be­fore the Fire­fly crashed into a ploughed field in a steep nose-down at­ti­tude. It had come to rest up­right. The en­gine was par­tially buried, at an an­gle around 50º to the hor­i­zon­tal. There was se­vere dis­rup­tion to its struc­ture. Both wings had suf­fered se­vere lead­ing edge dam­age, and their left and right fuel tanks had split open, with no fuel left in ei­ther. The fa­tally in­jured pi­lot and pas­sen­ger were each found to be hold­ing the top part of their re­spec­tive con­trol col­umns, which had bro­ken off in their hands. The pi­lot was hold­ing his stick with both hands, the pas­sen­ger with just his right hand.

The Ppl-rated pi­lot had started fly­ing in Novem­ber 2009, and flew Grob Tu­tors un­til Novem­ber 2012. In July 2014 he started train­ing as a RAF stu­dent pi­lot, com­plet­ing his ba­sic course on Tu­tors in Fe­bru­ary 2015 af­ter ac­cu­mu­lat­ing 152 hours, dur­ing which time spe­cific spin train­ing was un­der­taken. In Novem­ber 2015 he started train­ing on the Shorts Tu­cano and had com­pleted 63 hours in the air­craft, 21 hours in a flight sim­u­la­tor, and had car­ried out spin train­ing in the Tu­cano in the early stages of his course. He had recorded four flights in the T67 — two dual check flights and two as PIC. The in­struc­tor who cleared him to fly the T67 in Jan­uary 2016 had not car­ried out spin train­ing, so the pi­lot had not been taught type-spe­cific spin re­cov­ery on the air­craft.

The pas­sen­ger started ba­sic train­ing in Au­gust 2015 as a RAF stu­dent, ac­cu­mu­lat­ing 59 hours on the Tu­tor dur­ing which he would have car­ried out spin­ning and spin re­cov­ery train­ing. He had logged 5.7 hours on the Tu­cano, but had not yet car­ried out spin train­ing on that type. He did not hold a civil­ian li­cence.

Three months prior to the ac­ci­dent the pi­lot and an­other stu­dent pi­lot had flown to­gether in a T67 and recorded their ma­noeu­vres on cam­era. The footage showed them shar­ing the fly­ing and per­form­ing aer­o­bat­ics. This col­league re­called that, fol­low­ing a re­verse stall turn that was ob­vi­ously not go­ing to be com­pleted, the ac­ci­dent pi­lot took the re­cov­ery ac­tion of ‘throt­tle to idle and con­trol col­umn cen­tralised’. He then put the throt­tle to full power and the en­gine

re­turned to full power quickly, but not smoothly. Af­ter the flight, the pi­lot re­marked that the rea­son he used full throt­tle was that he re­alised that for an in­cip­i­ent (spin) re­cov­ery he should have left the throt­tle at full power. Clos­ing the throt­tle was part of the in­cip­i­ent spin re­cov­ery ac­tion for the Tu­cano on which he was then train­ing.

When com­pared with the spin re­cov­ery ac­tion for the Grob Tu­tor and Tu­cano, the Slingsby T67 dif­fers in that the con­trol col­umn must be moved pro­gres­sively for­ward to ef­fect spin re­cov­ery. Ini­tially on mov­ing the stick for­ward the rate of spin ro­ta­tion will in­crease be­fore com­ing out of the spin. Ap­pli­ca­tion of Tu­tor or Tu­cano spin re­cov­ery tech­nique by plac­ing the stick in a cen­tral po­si­tion is not the cor­rect re­cov­ery ac­tion for the Slingsby.

The AAIB re­port con­cludes: ‘De­spite the ex­ten­sive dam­age to the air­craft it can be de­ter­mined that there had been no struc­tural fail­ure or loss of con­trol sur­faces in flight. The ground marks and dam­age to the air­craft sug­gest the outer por­tion of the right wing lead­ing edge hit the ground si­mul­ta­ne­ously with the pro­pel­ler spin­ner, sug­gest­ing a slight left yaw. The clear im­print made by the right and left wing lead­ing edges on the ground at im­pact showed no ev­i­dence of rolling or spin­ning of the air­craft at this point. Ex­am­i­na­tion has found that there was con­ti­nu­ity and op­er­a­tion in the cor­rect sense of the fly­ing con­trol sys­tem and that there was no pre-ac­ci­dent fault or con­trol fail­ure of the aileron, rud­der or el­e­va­tors.

‘From wit­ness ev­i­dence, the air­craft ap­pears to have in­ad­ver­tently en­tered a spin from some form of loop­ing ma­noeu­vre shortly af­ter the apex of that ma­noeu­vre. From the recorded radar data, the al­ti­tude at that point was prob­a­bly be­tween 3,500 and 4,000ft, which should have pro­vided suf­fi­cient height for the spin to have been stopped and the air­craft pulled out of the dive if spin re­cov­ery ac­tion had been taken cor­rectly and promptly. The same wit­ness also de­scribed hear­ing the en­gine fal­ter­ing dur­ing the de­scent but no phys­i­cal ev­i­dence has been found to sug­gest an en­gine prob­lem. It is pos­si­ble the wit­ness was hear­ing the ef­fect of rapid open­ing or clos­ing of the throt­tle, cou­pled with the mask­ing ef­fect on the en­gine sound as the air­craft ro­tated dur­ing the spin. Both pi­lot and pas­sen­ger had been taught spin re­cov­ery in the Grob Tu­tor. The pi­lot had un­der­gone a Slingsby T67 check flight… with the club CFI which was a sin­gle cir­cuit and did not cover aer­o­bat­ics or spin re­cov­ery.

‘The spin re­cov­ery ac­tion in the Grob Tu­tor, and in the Tu­cano, re­quires that fol­low­ing the ap­pli­ca­tion of op­po­site rud­der, the con­trol stick is cen­tralised. This is taught by us­ing both hands in the Tu­cano and there is no re­quire­ment to move the con­trol stick for­ward. In the Slingsby T67, the con­trol stick should be moved pro­gres­sively for­ward un­til the spin stops and with an aft cen­tre of grav­ity the flight man­ual em­pha­sises that the pi­lot must be pre­pared to move the con­trol stick fully for­ward. As the stick is moved for­ward the spin’s rate of ro­ta­tion ini­tially in­creases.

‘The air­craft ap­peared to have de­scended in a spin. How­ever, air­craft at­ti­tude and ground marks at im­pact sug­gest that it had started to re­cover, al­beit too late to avoid hit­ting the ground. An in­struc­tor who reg­u­larly spins this air­craft type states that if the cor­rect re­cov­ery ac­tion is taken, the air­craft will come out of the spin. Given that the air­craft may have been in the process of re­cov­er­ing from the spin in the very last mo­ments of the de­scent, it is pos­si­ble that an in­cor­rect spin re­cov­ery tech­nique was used, as the re­quire­ment to move the con­trol stick pro­gres­sively for­ward is a crit­i­cal el­e­ment of the spin re­cov­ery ac­tion in the Slingsby T67. This was not a re­quire­ment for spin re­cov­ery in the Tu­tor or Tu­cano, air­craft on which the pi­lot had pre­vi­ously re­ceived spin train­ing. It is pos­si­ble that if the pi­lot ini­tially adopted the tech­nique ap­pli­ca­ble to those air­craft, the spin re­cov­ery would have been de­layed.’

Un­ex­plained turn on take­off

Air­craft Type: Ze­nair CH 601 XL Zo­diac Date & Time: 22 Au­gust 2015 at 0820 Com­man­der’s Fly­ing Ex­pe­ri­ence: French ULM Pi­lot Cer­tifi­cate, 670 hours, 64 on type Last 90 days: 40 hours Last 28 days: 15 hours Hav­ing flown in to Sandown Air­port from France the pre­vi­ous day, the pi­lot was de­part­ing for the re­turn trip. Af­ter lift­ing off, the air­craft im­me­di­ately started to turn left, then pitched up and mo­men­tar­ily reached a steep nose-high at­ti­tude, be­fore ad­just­ing to a more nor­mal climb at­ti­tude. The pi­lot, who had not in­tended to turn left, later re­ported that he re­duced power at a height of about 30ft to keep the air­craft track­ing straight ahead. Once the track ap­peared to straighten, he reap­plied power. The flight con­tin­ued in a nose-up at­ti­tude and in a banked skid­ding turn to the left. It straight­ened mo­men­tar­ily, lost height, and ap­peared to en­ter a stall and in­cip­i­ent spin to the left.

The air­craft, which had been air­borne for a to­tal of twenty sec­onds, crashed

into a grass field 300m to the north of the Rwy 05 thresh­old and caught fire. The pi­lot, who was se­ri­ously in­jured, was helped out and given emer­gency med­i­cal as­sis­tance be­fore be­ing trans­ferred to a lo­cal hos­pi­tal.

There were a num­ber of eye­wit­nesses to the ac­ci­dent, and there was also a video record­ing of the en­tire flight, to­gether with still pho­to­graphs taken from along­side the run­way. The dis­place­ment of the flight con­trol sur­faces could be seen in both the pho­to­graphs and the video footage. At the start of the take­off roll, the ailerons were in the neu­tral po­si­tion and ap­prox­i­mately half flap was de­ployed. The air­craft ap­peared to ac­cel­er­ate nor­mally along the run­way and di­rec­tional con­trol was main­tained. It bounced into the air briefly once or twice, be­fore lift­ing off af­ter a ground roll of ap­prox­i­mately 350m. It im­me­di­ately started to bank left. Right aileron and right rud­der con­trol de­flec­tions could be seen, but the left turn con­tin­ued. The ini­tial climb to a height of about 80ft was steep, and then the climb rate re­duced, although a nose-up pitch at­ti­tude was main­tained through­out. Right rud­der and right aileron in­puts were ap­par­ent through­out the re­main­der of the flight but the air­craft con­tin­ued turn­ing left, ap­par­ently fly­ing slowly. As it reached al­most the re­cip­ro­cal of the orig­i­nal run­way head­ing there was a marked loss of height, the left wing dropped and it en­tered an in­cip­i­ent spin.

The AAIB com­mented: ‘There could be a num­ber of pos­si­ble rea­sons why the air­craft made an un­in­tended turn to the left, as it lifted off the ground. How­ever, there was in­suf­fi­cient ev­i­dence to iden­tify any par­tic­u­lar fac­tor. Once the air­craft had turned away from the run­way track, the pi­lot’s op­tions were lim­ited by the ob­sta­cles in his path. The nose-high at­ti­tude prob­a­bly led to a re­duc­tion in air­speed, thereby re­duc­ing the ef­fec­tive­ness of the flight con­trols and ul­ti­mately lead­ing to a stall… There was in­suf­fi­cient ev­i­dence to es­tab­lish why the air­craft had be­haved as it did and why the ac­ci­dent oc­curred.’


Air­craft Type: Europa XS Date & Time: 10 Au­gust 2016 at 1015 Com­man­der’s Fly­ing Ex­pe­ri­ence: PPL, 918 hours, 438 on type Last 90 days: 49 hours Last 28 days: 11 hours When the air­craft ar­rived over­head the un­li­censed grass air­field at Hol­lym, York­shire, the wind­sock showed sur­face wind from 310° at less than 10kt, so the pi­lot de­cided to land on Rwy 32, which he had last used four years pre­vi­ously. The run­way’s first 150m is bounded by a fence on its left side and a cliff, down to the sea, to the right. There­after, the left side of the run­way opens out onto the main grass air­field and par­al­lels the cliff.

Dur­ing its land­ing roll the air­craft slewed to the ex­treme right side of the run­way. Its tail­wheel went over the edge of the cliff and, at low speed, it fell tail-first off the cliff into the sea. The ac­ci­dent was not witnessed by any­one, so the emer­gency ser­vices were not ini­tially alerted, but af­ter a few min­utes the pi­lot re­trieved and ac­ti­vated his per­sonal lo­ca­tion beacon and af­ter some thirty min­utes was able to make his way up the cliff to a nearby house. Help then ar­rived and he was trans­ferred to hos­pi­tal for treat­ment to mi­nor in­juries

This Europa was a monowheel ver­sion, with re­tractable out­rig­ger wheels mounted on each wing. Fol­low­ing the ac­ci­dent, the right out­rig­ger was found on the run­way near the cliff edge. It had failed be­low its at­tach­ment to the wing. The out­rig­gers are de­scribed in the air­craft’s owner’s man­ual as ‘ex­ceed­ingly

strong and pli­able so there is no need to be overly con­cerned about turn­ing sharply or rough field op­er­a­tion. The out­rig­gers will, if nec­es­sary, bend through 90º de­grees. How­ever, they are not de­signed to take sig­nif­i­cant ver­ti­cal loads and, on rare oc­ca­sions, do fail.’

The pi­lot later com­mented that the last time he had landed at Hol­lym, Rwy 32’s mown strip was bounded on both sides by sig­nif­i­cant fal­low ground, pro­vid­ing some mar­gin for er­ror. Re­view­ing the air­field en­try in com­mer­cially avail­able flight guides had ap­peared to con­firm to him that this was still the case. At the time of the ac­ci­dent the run­way was re­ported as be­ing fif­teen me­tres wide, with a sketch show­ing a clear area on each side. How­ever, coastal ero­sion in the area av­er­ages two me­tres per year, with the cliff on the right of the run­way be­ing un­der­mined by the sea and then col­laps­ing in stages. This had re­sulted in the run­way be­ing moved to the left, un­til it abut­ted the fence, while the sea con­tin­ued to erode the ground to the right of the run­way.

The CAA’S CAP 793 Safe Op­er­at­ing Prac­tices at Un­li­censed Aero­dromes rec­om­mends that for air­craft with MTOWS of less than 2,370kg, min­i­mum run­way width should be 18m. Since this ac­ci­dent the air­field op­er­a­tor has in­formed the pub­lish­ers of var­i­ous flight guides that run­way width is now 10m and re­quested that they in­clude a warn­ing to ‘Be­ware cliffs on the east side of the air­field’.

He­li­copter/ model air­craft con­flict

A RAF Chi­nook he­li­copter was tran­sit­ing to Col­erne at low level when the han­dling pi­lot saw ris­ing ground ahead and started to climb, where­upon a crewmem­ber called that he’d seen a white, fixed-wing model air­craft ap­prox­i­mately one me­tre in length pass just above them. The lo­ca­tion was noted and the grid ref­er­ence passed onto op­er­a­tions staff. No model air­craft fly­ing had been pub­lished by NOTAM in the area and no other crew mem­bers saw the model air­craft. The Chi­nook pi­lot was op­er­at­ing un­der VFR in VMC and not in re­ceipt of a ser­vice but lis­ten­ing out on the UHF low-level com­mon fre­quency. The model op­er­a­tor could not be traced.

The UK Air­prox Board noted: ‘All model op­er­a­tors are re­quired to ob­serve ANO 2016 Ar­ti­cle 94(2) which re­quires that the per­son in charge of a small un­manned air­craft [a model air­craft] may only fly it if rea­son­ably sat­is­fied that the flight can safely be made, and the ANO 2016 Ar­ti­cle 241 re­quire­ment not to reck­lessly or neg­li­gently cause or per­mit an air­craft to en­dan­ger any per­son or prop­erty. A CAA web­site pro­vides in­for­ma­tion and guid­ance as­so­ci­ated with the op­er­a­tion of Un­manned Air­craft Sys­tems (UASS) and Un­manned Aerial Ve­hi­cles (UAVS) and CAP722 UAS Op­er­a­tions in UK Airspace pro­vides com­pre­hen­sive guid­ance.

‘In the low level en­vi­ron­ment some risks are un­avoid­able if re­al­is­tic low fly­ing train­ing is to be con­ducted and a good look­out re­mains the only mit­i­ga­tion against this sort of Air­prox. The model op­er­a­tor was within his rights to be fly­ing his model and it was po­ten­tially the low fly­ing and use of ter­rain mask­ing [by the Chi­nook] that was a fac­tor in pre­vent­ing ei­ther the model op­er­a­tor or the Chi­nook’s crew hav­ing suf­fi­cient time to take any avoid­ing ac­tion. As to what ef­fect a model air­craft with a one-me­tre wing­span would have if the two had col­lided re­mains un­known.

‘The Chi­nook did not ap­pear on radar record­ings… [which was] un­sur­pris­ing con­sid­er­ing the Chi­nook’s al­ti­tude. It was re­gret­table that the model air­craft op­er­a­tors could not be traced, be­cause with­out their nar­ra­tive it was not pos­si­ble to un­der­stand the sit­u­a­tion from their per­spec­tive… The speed and low al­ti­tude of the Chi­nook was an oper­a­tional mea­sure de­signed to pre­vent de­tec­tion and this may have worked by deny­ing the model air­craft op­er­a­tor the chance to fly his model away from the Chi­nook. The model air­craft op­er­a­tor was en­ti­tled to use Class G airspace and out­side the bound­aries of a club or na­tional event [it was] un­likely that a NOTAM would be raised. Ul­ti­mately, the air­craft flew into con­flict with each other and… with the model pass­ing over the Chi­nook and nei­ther of its pilots see­ing it, col­li­sion seemed to have been avoided by prov­i­dence alone.’ The Board as­sessed the in­ci­dent as a Risk A.

Although mit­i­gated through mod­i­fi­ca­tions and ad­di­tional pi­lot train­ing, Monowheel Europas have suf­fered a num­ber of loss-of-con­trol land­ing ac­ci­dents

The XL is the ul­tra­light ver­sion of the CH 601, a pop­u­lar kit­plane (this is not the ac­ci­dent air­craft)

The Slingsby T67 Fire­fly is one of the few aer­o­batic train­ers avail­able in re­cent years and is used by many clubs around the coun­try

Model fly­ers: be­ware Chi­nooks fly­ing ‘nap of the earth’ and Chi­nook crews: be­ware un­sus­pect­ing model fly­ers!

Newspapers in English

Newspapers from UK

© PressReader. All rights reserved.