air safety

May 5, 2017

ATSB sounds alarm about certification blind spot in ATR turbo-prop

More safety issues with the infamous Virgin Australia ATR 72 incidents of 2014 have emerged

Ben Sandilands — Editor of Plane Talking

Ben Sandilands

Editor of Plane Talking

Virgin Australia kept this severely damaged ATR turbo-prop in service for five days

Updated: CASA has applied addition oversights and a condition on ATR operators.

The ATSB has issued an alarming interim report into a set of incidents involving a Virgin Australia ATR 72 turbo-prop in 2014 saying a fault in the design had not been taken into account when it was certified as safe to fly.

The report also makes it clear that the inability of Virgin Australia to properly identify the unsafe condition of the airliner which continued to fly 13 sectors over five days after the incident remains under investigation.

The first incident happened on February 20,  2014, as the Virgin flight sustained a pitch disconnect crisis while on descent into Sydney. The pitch disconnect occurred while the crew were attempting to prevent the airspeed from exceeding the maximum permitted airspeed (VMO). The aircraft was significantly damaged during the occurrence.

In its first interim report issued last year the ATSB found to put it in blunter language that the two pilots broke the structure of the T-tail by forcing the controls in opposite direction.

In this report it finds that:

During the continued investigation of the occurrence, the ATSB has obtained an increased understanding of the factors behind this previously identified safety issue. This increased understanding has identified that there are transient elevator deflections during a pitch disconnect event that could lead to aerodynamic loads that could exceed the strength of the aircraft structure.

The ATSB also found that these transient elevator deflections were not identified, and therefore not considered in the engineering justification documents completed during the aircraft type’s original certification process. The ATSB considers that the potential consequences are sufficiently important to release a further interim report prior to completion of the final investigation report.

ATR isn’t represented in Australia, but Plane Talking understands that the maker, based in Toulouse, and partly owned by Airbus, is well aware of the contents of the second interim report and will respond in due course.

The separate issue of the disclosure standards from Virgin Australia, the quality of its oversight of the regional turbo-prop division, and prima facie evidence of an unsatisfactory culture outcome in the cockpit of the flight on descent into Sydney remains extremely important.

It was dealt with in Plane Talking at that time but largely ignored by the PR obedient general media in this post.

The core elements of the ATSB report show that Virgin Australia’s engineering contractor and the airline failed to identify and understand serious damage done to this aircraft in the turbulence event.

The aircraft was then allowed to carry passengers for thirteen sectors in that state before an in-flight crisis five days later approaching Albury from Sydney where it was grounded after landing, and remains to this day, pending repairs if indeed it can be repaired.

These are scandalous disclosures.  No one in the general flying public in this country expects that a contract maintenance organisation could be so bad at its job that it failed to understand and identify the grave safety of flight issues apparent on the Virgin turbo-prop on 20 February.

It is after all, what the maintenance provider is paid by Virgin to do, rather than scratch their heads and release the aircraft back into service.

It’s Virgin’s inescapable legal obligation to ensure that all aircraft are safe before flying. It didn’t ensure the safety of these 13 flights. It’s CASA’s role to enforce and maintain a safe level of oversight on airline operations and ensure that those who carry out aircraft maintenance are competent and effective.

This is a 68 passenger airliner. The situation with the turbo-prop which remained grounded at Albury for months, was one that could have ended, at any time during the five days Virgin Australia had no obvious idea what was going on, in a pile of wreckage and bodies.

Learning, as the industry does today, there was also a blind spot in the certification of the turbo-prop as safe doesn’t build confidence in either the airliner nor the operator in this situation.

The public communication of the seriousness of the ongoing situation in February 2014 was non-existent, and for those media that did make inquiries, totally unsatisfactory.

More recently Virgin Australia has further rolled back its involvement with regional turbo prop services to keep a token force of six of the aircraft operating across parts of it network, once its Queensland ATR base is closed down. Canberra will remain a focus of ATR operations, contradicting the emphasis on E-190 jet flights that prevailed in recent years.

The airline has often said that safety is its first priority and that the ATR operations are safe. Well, they weren’t safe for five days in 2014, the airline was clueless as to what was going on, and the issues with this aircraft appear to be of sufficient concern for the ATSB to issue a global notification to other ATR operators.

Virgin Australia said:

Safety is Virgin Australia’s number one priority, and we continue to liaise closely with the relevant regulatory bodies and the aircraft manufacturer in relation to the Australian Transport Safety Bureau’s ongoing investigation into the ATR’s pitch disconnect mechanism. Virgin Australia has taken a number of risk mitigation measures and is confident the aircraft remains suitable for operations.

CASA said.

CASA continues to audit ATR aircraft operators to ensure appropriate actions have been taken to reduce the likelihood of the aircraft being mishandled in a manner similar to the incident flight.  Flight procedures and pilot refresher training for the ATR aircraft operated in Australia have been amended since the event occurred.

CASA concurs with the European Aviation Safety Agency (EASA), the State of Design, that the aircraft is safe when operated in accordance with documented operating procedures.

CASA will look carefully at the findings of the aircraft manufacturer’s engineering analysis of the issues associated with a pitch disconnect when this work is completed.

If this analysis raises any ongoing safety issues CASA will take appropriate action.  CASA will take into account any measures taken by EASA.

CASA is requiring monthly reports from the air operators of ATR aircraft to make sure the actions that have been implemented continue to be effective.


(Visited 19 times, 1 visits today)


Leave a comment

16 thoughts on “ATSB sounds alarm about certification blind spot in ATR turbo-prop

  1. Dan Dair

    It’s still a shocking tale.
    I have some sympathy with Virgin on this. You pay a maintenance contractor to inspect & maintain your aircraft, so you don’t have to. They bugger it up and then Virgin, not the contractor, is left holding that particular baby.!!

    Also, I’m a bit in the dark with this ‘pitch-disconnect’ thingy.?
    I’m struggling to understand how it can disconnect, but then remain connected when subsequently examined.?
    My assumption was that there was a ‘designed-in’ weak-spot, which would disconnect at a set point above design-loading, which would then allow the intact control-surfaces to return to neutral, thus preventing a ‘stuck’-control-surface generated incident.?
    It appears that I’ve misunderstood.?
    Would someone who knows about these things please enlighten me.?

    1. Giant Bird

      That is the mistake that the cost cutters make time and time again. You can outsource the work but not the responsibility. It is still your liability and your reputation. When they calculate the savings in outsourcing rarely do they allow enough for the oversight. Sometimes when they wake up to this later the outsourcing cost benefit is so small that they would have kept the task in house if it had been costed properly in the first place. And if things go wrong before you put the proper oversight in place the financial cost of damage to the reputation can be so high that the outsourcing will never break even and will for ever cost more than if you had kept the work in house. Outsourcing can work but too often the savings are way too optimistic.

    2. JW (aka James Wilson)

      From the first interim report:
      “In normal operation, the left and right systems are connected such that moving one control column moves the other control column in unison. However, to permit continued control of the aircraft in the event of a jam within the pitch control system, a pitch uncoupling mechanism is incorporated into the aircraft design that allows the left and right control systems to disconnect and operate independently. That mechanism comprises a spring-loaded system located between the left and right elevators.

      The forces applied on one side of the pitch control system are transmitted to the opposite side as a torque or twisting force through the pitch uncoupling mechanism. The pitch uncoupling mechanism activates automatically when this torque reaches a preset level, separating the left and right control systems. When set correctly, the activation torque is equivalent to opposing forces of 50 to 55 daN (about 51 to 56 kg force) being simultaneously applied to each control column.”

      “The pitch uncoupling mechanism can be reset by the flight crew, reconnecting the left and right elevator systems. However, this can only be achieved when the aircraft is on the ground.”

      According to the ATSB’s initial update, dated 10 June 2014:
      “After shutdown the crew completed cockpit tasks including reconnection of the two elevator control systems…”

      1. Dan Dair

        Thanks very much.
        It all makes sense now.!!!

  2. StickShaker

    I thought that pilots were always required to do a pre-flight check of the aircraft before every flight which involves walking around the aircraft to do a visual check for any safety issues. Even blind Freddie would spot something like that.

    1. Dan Dair

      It is therefor a great pity that Blind Freddie wasn’t the aircraft engineer who was sent out to inspect the aircraft after the first ‘hard’ landing.
      To the best of my recollection, Ben reported that it was done at night, in the wind & rain & the inspection was a walk-around on the ground, presumably with a ‘flashlight’.?

      A proper close-up inspection wasn’t done until the day this aircraft was grounded, when the duty pilot noticed something ‘not right’ about the tail, whilst doing the pre-flight inspections.

      Also, I’m pretty sure that there’s a fairing which had already been removed when the photo in this posting was taken, so the damage is a whole lot clearer than it would have been with the fairing in place.?

      1. StickShaker

        Dan – I understand that damage hidden behind fairings may not be readily visible. However, if a duty pilot noticed that something was “not right” about the tail after the aircraft was grounded then why didn’t the respective pilots and engineers operating the previous 13 sectors also notice that something was “not right”.
        I wouldn’t like to be in the witness box at a coroners inquiry protesting (under heavy questioning) that it was a windy and rainy night and the pre-flight inspection was allowed to be compromised due to the weather.

        I have read numerous blogs from veteran US pilots describing how they insist on getting out in the freezing snow and ice to do a thorough pre-flight and how they scowl at younger pilots who prefer to remain in the warm cockpit.

    2. ghostwhowalksnz

      The photo provided is with the fairing removed. Its not what they would see at the time of walk around and the photo is blown up to show in detail

  3. comet

    It’s a bizarre concept that an airline CEO can outsource his criminal liability to another company that he can’t control.

    It seems a risky move.

    1. Dan Dair

      Isn’t that the crux of the problem…..
      A CEO can outsource the work,
      of which they may have no actual hands-on oversight,
      they CAN’T outsource the liability.

      This has come up before, probably in connection with this incident & it’s written into law;
      Whether it’s your fault or their fault,
      ultimately, it’s always your fault.!

  4. Tango

    It seems to me this is in line with the SA 777 tail strike.

    Prudence was not done on either one and both had strong indications of significant impacts.

  5. discus

    It appears that the manufacturer had not identified the risks of uncoupling with both control columns free to move asymmetrically, or that the force required to uncouple would mean that there would be sudden, unavoidable movement in whichever direction a crew member was pushing , pulling or both at the same time, as it was in this case. That is serious if found to completely factual as it derails everything thereafter.

    This being the case, it is likely that the requirements of the inspection in accordance with the aircraft maintenance manual did not involve mandatory removal of the fairings that would cover the vast majority of, if not all of that obvious damage.

    Without pre-empting the investigation’s findings, it seems that there were some problems in the cockpit . Having one pushing, one pulling to avoid an over speed sounds like every poor flying to me. The subsequent write up for engineering is crucial as to what inspections and checks are mandated.

    Given the manufacturer had not identified the possible problem, the engineers may have done everything in accordance with the manuals for what was reported. As an engineer myself and worked in line maintenance for about 30 years, we need to rely on what is reported as being accurate. We are not allowed to deviate from what the manuals tell us. If moderate turbulence is reported and that situation only requires a “General Visual Inspection (GVI)” that is exactly what is done. You lose jobs and or contracts for grounding aircraft and doing inspections etc that are not required by the manufacturer’s manuals for any given report from the flight crew.

    In Ben’s previous post on this incident, it was noted in the comments that the flight crew may not have initially report the turbulence, just the disconnect and it was upon the suggestion of the engineer to write the turbulence up and iirc it was described as “moderate”. In this case it seems a “general Visual Inspection” was required. A GVI really is a walk around. Boeing for example describes a GVI as done from the ground without special stands or equipment.

    Now, if there was obvious damage that was missed the signing engineer , he or she may have serious problems, but if that damage is covered by fairings that are not required to be removed during the required inspection and that inspection was carried out diligently, in accordance with the documentation, that engineer is within their rights to sign off on it.

    As usual, there is a sequence of events , lining up the holes in the Swiss cheese, that allowed this to happen starting with the manufacturer , then the crew’s actions whilst flying, possibly their report to engineering and perhaps the engineers.

    1. derrida derider

      “Having one pushing, one pulling [the stick] to avoid an over speed sounds like every poor flying to me.”
      Yep, it sounds far too much like AF447 for me. Regardless of the other issues, I’d expect the final report to recommend these pilots do some serious work on their CRM.

  6. Tango

    The Asiana crash at SFO is worth reading in depth as well.

    They were above the glide slope so they used the fast decent mode.

    Experienced 777 (and Boeing) pilots call it a FLCH trap.

    Hidden in the fast decent mode is the fact that auto throttle goes off and stays off, all by itself.

    Boeing thinks that is fine. The NTSB disagrees very strongly.

    So you have a manufacture that has deemed it one thing and a safety organizing that disagrees. NTSB cannot force them.

    What I can tell you is that you don’t need stuff like that occurring that you as the pilot did not deliberately turn off. New pilot, used to Airbus that does not do that and one dead person though it could have been the whole planeload.

    Anyone that thinks the mfgs are infallible in their end is mistaken. Its a reason you have a PIC and they are intended to exercise judgment that goes beyond what they have been told.

    Airbus did not have a two engine out procedure for the A320 (0r any other twin). Its not going to happen.

    The good news was Sullenberg was a former military pilot and a lot of quick thinking handled it as well as anyone could have.

    The missed the ditch mode shift as they were too busy getting set up for the landing.

    Now why you have a ditch mode when your don’t expect both engines to stop?

    Its not easy and its hugely complex with an decision having affects on other aspects that can the create a serious and even lethal problem.

    1. Dan Dair

      “Experienced 777 (and Boeing) pilots call it a FLCH trap”
      This is why Boeing &/or the airline issue instruction manuals & why pilots spend so much time in simulators.

      I agree that there is a genuine issue of similar flight systems for both Boeing & Airbus, which have small but very significant differences.
      But this is why line-pilots get the big-money. Most of them can actually remember what make of plane they’re flying.?

    2. ghostwhowalksnz

      NTSB is an investigation organisation only ( air , rail, sea , pipelines etc)
      ‘”to determine the probable cause of transportation accidents and incidents and to formulate safety recommendations to improve transportation safety.”
      FAA is the one that sets and enforces ‘standards’

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details