Airprox over Central Scotland
 

Announced on several news channels today, the Airprox Board has listed a near miss between two aircraft to the north west of Pitlochry in Scotland on 23 June. I know this thread has been started on Spotters Corner, but perhaps the implications of a potential disaster on this scale merit this post?

I won't quote the embellished press reports, just point folks in the direction of the Airprox report here :

http://www.airproxboard.org.uk/docs/....09Reports.pdf

Curiosity got the better of me and I can see that the two aircraft were Lufthansa B748 D-ABYC enroute EDDF-KIAD and BA B744 G-BNLM enroute EGLL-CYVR.

One thing to add to the report is that press reports indicate that there "were up to" 1000 souls onboard both aircraft, though in reality likely in excess of 800.

Definitely not dodgy third world airlines then

And probably communicating in impeccable ICAO English as well. :E

Just read the UKAB report, then the story on the Sky News app.
Sky's spin department have exceeded even their own very high RPM on this one! :D

No mention in the article of TCAS.

Degree of Risk C, ATC, pilots, visual sightings and TCAS all kept well apart...

Worthy of looking into into why. Newsworthy? Not sure...

Fairly routine loss of separation exacerbated by crews failing to follow fairly simple instructions, failing to utilise spatial awareness, failing to follow SOP's (advising TCAS RA) & then failing to operate AFS (I would imagine. . or cr@p hand flying? ) to provoke a 2nd alert.

All in all a bit shabby for 1st world legacy crew I would suggest.

It is not 'newsworthy' except for the over excitable tabloid reporting.

However, from a professional standpoint when a controller gives 'avoiding action' in Class C airspace the instruction should be treated in the same way as a TCAS RA - do not second guess, chat amongst yourselves on the flight deck, take a look out of the window to see if it's necessary - react at once and turn. From the controller point of view - ALWAYS expect your instruction to be ignored and watch carefully and reissue the instruction immediately it is apparent that it is being ignored. (I speak from experience in the same airspace). Unfortunately, the track smoothing algorithms can hide a turn on the controller display until 2 or 3 updates - that is a long time.

This has all the hallmarks of crews only half listening to their current controller and already settling into 'oceanic mode'. Perhaps..... As there were two 'avoiding action' instructions close together I think the first PF didn't fully hear them picked up half the second heading and thought it was a repeat of the instruction to his aircraft and turned onto it. The second aircraft PF heard the heading for the first and was not sure it was not for them - then heard their callsign in the second instruction and took the heading heard first. This is always going to be a problem with crews where PNF talks and the PF acts. It is obvious that both the PNFs read back correctly yet watched PF turn to a different heading.

All the masterly pilots on here who never make any mistakes - should think about this one.

I suppose two heavy jets mixing up instructions to turn left or right is fine, as long as they don't mix up instructions to climb or descend when TCAS kicks in.

I find it concerning when the pilots' "misunderstanding" (by both crews) is given as the "cause" when the root cause was 2 aircraft at the same FL on a converging course. The "RT was clearly stated on the VHF recording" which I suspect means a recording of what was being transmitted, prior to it going over the airwaves, maybe not clearly received, maybe blocked by Box 2 or I/c or Cabin chit chat etc.

SOPs likely permit at that stage only 1 pilot on the Flight Deck, maybe using speakers / microphone, when compared to busier climb/decent phases. I think you are correct... when ATC order a Go Around it is "C/S Go Around I repeat Go Around", no doubt with good reason...

<<The "RT was clearly stated on the VHF recording" which I suspect means a recording of what was being transmitted, prior to it going over the airwaves,>>

Recording takes place simultaneously with the transmission going over the "airwaves".

Thanks HD... my point was that the Airprox board seem to guarantee the quality of that recording as that as heard by each crews' ears...

I'm obviously missing something here, if the recording was made at the 'sending' end how do you know what was heard at the other end? Without a recording of the audio output from the receiving end that is?

I have no experience of working en route, so I shouldn't really comment. I just find it a bit strange that the a/c were converging and cleared to same level without being on headings. Especially in this day and age when we are constantly reminded to be "defensive".

The final numbers in terms of separation certainly don't make it headline news, but how the incident developed and how the holes in the cheese are always there makes it interesting and educational read.

"He also noted that, during simulation training, avoiding action was only practiced as a result of a TCAS alerts, and not as a result of ATC instructions; this was an important consideration as to their potential familiarity with receiving, assimilating and actioning such RT instructions in a timely manner."

What?? I'm not having a go at either crews actions, but this statement by the board member surprises me. Avoiding Action instructions should be expected at any stage of flight surely, and from my experience precede/prevent many TCAS alerts. After all a change of heading so you don't hit something is not exactly the most challenging of tasks to assimilate and action.

Apologise for the thread creep here, but this would appear to be the correct course of action following an "Avoiding Action" command from ATC. http://www.ead.eurocontrol.int/eadba...2006-08-17.pdf

2.2.2 Responses to ‘Avoiding Action’ instructions should not be made via autopilot vertical or lateral flight path control selectors

because, due to the design of this equipment, the turn, climb or descent that the Air Traffic Controller has requested and expects to see

actioned will be initiated too slowly to serve the needs of the moment.



It's maybe my gashness, but I'll be honest in that I have never responded to an "avoiding action" instruction by disconnecting the automatics as this would (I think) increase, what can become, a high workload situation. Likewise, I've never seen anyone else do it.

I often wonder if ATC expect us to respond as per the above AIC.

Terrifying report over Scottish Skies..CNN
 

CNN is currently airing his "terrifying report" by non other than "Aviation expert" Richard:} Quest!!!

It all boils down to..CRM says Mr. Quest! I guess ATC/PILOT CRM is the new hot topic...or is it crew fatigue,Automation,inexperience AND CRM?

ATC recordings are recorded "off air" so the signal has been transmitted and received but they are only a record of what was transmitted and managed to travel between one of the transmitter stations and one of the receiver stations, usually a couple of miles or so. They are a good record of what the controller transmitted but only that, you need the aircraft voice recorder to be sure of what reached the aircraft and its relationship to other distractions.

Converging aircraft means holes are lining up in the proverbial cheese. Two aircraft then making incorrect turns to increase rate of convergence is blowing a couple of huge new holes in the cheese.
Yeah yeah, 3 miles separation is great, but the 747 is still one of the fastest heavies and two of them swinging towards each other will gobble that distance up in no time, even if they're not approaching head-on.
Certainly would like to know reason for initial convergence path but much more disturbed by lack of insight into reason four experienced personnel on two flight decks were lulled into making a potentially catastrophic change of direction.

Was the word "say again" or "confirm" ever used by these convergent crew? It might well have been another Tenerife..or worse still..another Zagreb!

Trackdiamond, I guess it depends how you view "worse still". 577 souls perished at Tenerife, 176 over Zagreb.

Having just read the report it makes interesting reading. Amazed that both sets of crew took each others avoiding instructions from ATC. That's really not great PR for pilots at any level or qualification.

Yes TCAS was there as another level of protection so the risk is lower than that currently being hyped by the media. But to the layman you can understand a certain level of apprehension when aircraft instructed to avoid then increase their convergence!

Alas, for all the 'anonymising' of the report, the Airpox board does slip up a little by referring on one occasion to the 'B748'. Narrows down the operators somewhat.

Wow , how many "avoiding action" instructions have you received? I've got around 12000 hours in busy airspace and I've never received one. If I ever do I hope to follow my company SOPs and disconnect and fully expect my FO do the same!
Hopefully you are confusing a routine turn from ATC to ensure separation with a full on "Avoiding action!..." call. Time to get your head into that manual and clear that one up before your next flight perhaps.

I am going to jump in here with little knowledge.
I just find it interesting that we have two aircraft on a converging course that it requires ATC to require both aircraft to take evasive action.

My first thought would be that ATC should have only needed to vector, change altitude, or slow down one aircraft from the converging course long before the situation degraded to the point of requiring both to evade.
Hence, there would only be one set of instructions.
Again, I do not know.

May one guess that the CVR was not pulled - due to them being far from dest? So we'll never know what the crews actually heard. Only that could narrow down the cause.

Better to be talking about this than a prang. For both FC to pick up the wrong side of the TX is a VERY unusual circumstance. So unusual that it may not happen again for a significant time. Given the news, it will help everyone to think about it. Thus, a good outcome.

I would hope that I'd notice a TCAS return, inside 10nm, same level, getting closer, and start asking questions...

VFD, you CANNOT use speed in this situation.

As for issuing instructions to both aircraft, that doubles your chance of one doing it correctly, sadly not what happened in this scenario.

Capn Bloggs

Like he said :bored:

I was always taught from a very early stage that if you hear the words 'avoiding action' or 'immediately' associated with any ATC clearance then you disconnect the autopilot and 'just do it'. I have never practised it in recurrent sim sessions however which is a shame as it would take less than a minute to do as part of a LOFT excercise.

ontimeexceptACARS...Tenerife runway vs Zagreg mid air
 

I read you mate! But some walked away from that Tenerife cockup (PanAm)..a mid air collission is more severe..only in this case smaller crafts were the subject...Trident and. DC9...I was therefore referring to the severity rather than absolute numbers perished.You have a point though..relative speaking!

In terms of absolute comparisons..ofcourse it resembles the Tenerife disaster..similar types and categories involved...conflicting course, R/T cockups, possible CRM deficiencies...only major difference is..it took place is first world airspace with supposedly first class airline pilots..with ELP 6 transmissions no doubt! Its like we have not learn anything after over 3 decades! I think it is a good idea to incorporate a more realistic ATC/Traffic scenario rather than have SFI mimicking ATC..to have multi dimentional scenarios with conflicting ATC instructions for the crew to react to...to sharpen situation awareness...and during failures management phase at that!Also to have provisions in CVRs to be able to extract data during any requested time frame during the flight captured.

I am amazed that a standard safety call from ATC is not practiced in simulators. And from comments here avoiding action is often second guessed by flight crew - whose lives are at risk due to their delay.

Most deconflictions are taken by planning the aircraft transits through the sector and the crew are unaware as they just follow the clearance they have received. However, when something goes awry the controller may have to act quickly and expects the aircraft to turn at once, there is no time for collaborative decision making and gentle FMC half rate turns.

TCAS is a really useful short term safety system but it will only see 'cooperating' aircraft that are transmitting/transponding on 1090/1030MHz. There have been mid-air collisions with one aircraft not transmitting. The controller almost always has a far better 'picture' of what is going on around the aircraft than the limited situation awareness of the flight crew who should not rely on TCAS to see everything. Time for second guessing is after the avoiding action is completed

We did ATC avoiding action scenarios a few years back in the sim. Our OPS manual now reflects that the AP is to disconnected in order that an increased bank of turn can be achieved. Our airline is all airbus and hence if you left in the AP at high cruising levels then all you would get is a gentle 1g turn. Interestingly enough we were briefed that a number of other airlines across Europe refuse to allow their pilots to disconnect the AP so to plan on an ATC avoiding action event to become a TCAS RA.

eglnyt
 

Right, thanks for that info, it's a very sensible way to ensure the record of what was transmitted is accurate and valid.

I appreciate it's probably not a factor in this case*, but it seems to me that if that was the evidence referred to in the report to imply the transmissions were received clearly then, as you say, without the aircraft voice recorder it's stretching it a bit!

*That said, something caused four people in two crews to make similar errors about what they heard...so there has to be a common factor relating to comms surely?

Not watching the TCAS/not doing anything about it? Happens elsewhere too:
although in this case, "The controller immediately instructed the flight crew of the A330 to descend to FL350. The crew acknowledged and advised that they could see the traffic on their TCAS". Bit late for that...

Not much of a turn! ;)

Sometimes you have to do what makes you feel safest. I was in a climb one day as a brand new FO on a B737 and at 3700 ft was told to level at 4,000 ft because of opposite direction traffic at 4300 ft. Knowing I couldn't do that at our rate of climb looked out the window and saw no aircraft so accelerated climb verifying no aircraft rather than pushing nose down into blind area. 20 seconds later passing about 5,000 ft approach said your traffic just passed below you at 4,000 ft. I told the captain what I was going to do of course but blindly following last second instructions from ATC may not always be the safest thing to do.

This happened in the earlly 80's prior to TCAS.

Lucky there wasn't someone above you at 5,000 I guess.

Usual rubbish from our experts who don't fly long-haul, or don't fly aircraft
 

TCAS never instructs turns.

TFN disaster involved full loss of life.

TCAS shows targets at long ranges.

The jeering at an airprox between 2 legacy carriers is being made by idiots and is misplaced. 2 so-called third-world carriers would have been much likelier to not rectify the original mistakes, and even bump into each other.

From the report I'm a bit confused about the Sector management. It seems to me that initially the Sector was only manned by one controller. Is this SOP?
Überlingen and Zagreb spring to mind, everything chugging along normally, no problem for one controller, then it all goes wrong

Whilst recognising the need for anonymity in the report, it may have been useful to see if there was any similarity between the callsigns for these two aircraft.
Frequently I hear ATC instruct for example "BirdSpeed 231 ... " followed by "BirdSpeed 321 ..."
If the two companies alluded to earlier are the ones concerned they use a mixture of numeric & alpha numeric callsigns.
Also I'm not sure if one of the operators (if the hypothesis is true) readback method of callsign preceding the confirmation of understanding of the instruction causes any confusion to:
a) ATC
b) Other aircraft in the vicinity.
When this is heard on R/T it is almost as though it is a confirmation that because the callsign comes before the instruction it is deemed as a matter of fact that the instruction has been understood as the callsign was returned correctly. Well, my simple brain hears it like that!

I suppose now the Daily Mail have found out about Airprox they will report every incident reported such as:

Fighter pilot used weapons radar to avoid colliding with passenger jet | Mail Online

Sensationalism at it's best (or worst which ever way you look at it)

Even luckier I didn't level at 4,000 ft because we were cleared to 7,000 ft.