Birmingham Airport near miss caused by radio mix-up

[jase_31]

A radio frequency mix-up was to blame for a near miss at Birmingham Airport, according to an air accident report.

The pilot of a six-seater private aircraft lost contact with air traffic controllers and passed "over the top" of a stationary passenger plane.

The Socata TBM 850 landed safely after avoiding the Flybe-operated Q400 bound for Guernsey.

The Air Accidents Investigation Branch (AAIB) described the events of 12 January as a "serious incident".

In its report, the AAIB said the pilot of the TBM thought he had selected 118.300 megahertz (MHz) on his radio standby position, but in fact the active frequency was 118.030 MHz.

'High workload'
There was no time for the Flybe plane, carrying about 50 passengers, to taxi clear of the runway before the TBM arrived.

The report said: "The controller's only option was to keep the Q400 obstructing the runway and rely on the TBM pilot either going around or seeing and avoiding the Q400."

It added that the tower controller had not read a recent safety notice which contained advice for dealing with this type of situation, but did alert the other controllers on duty, and the supervisor, of the problem.

The AAIB made four safety recommendations and said the pilot's decision-making process had been "degraded" by a "high workload".

There was no damage to either plane.

BBC News - Birmingham Airport near miss caused by radio mix-up (http://www.bbc.co.uk/news/uk-england-birmingham-15283328)

[IO540]

The report said: "The controller's only option was to keep the Q400 obstructing the runway and rely on the TBM pilot either going around or seeing and avoiding the Q400."

Sounds reasonable, given that he was in the tower and had no obvious way to teleport himself to the TBM and take over its controls.

[Whopity]

Hugh, you are no doubt press. The report is here (http://www.aaib.gov.uk/publications/bulletins/october_2011/socata_tbm_850__n850tv.cfm) Those who read it will be clear what happened and the resulting Safety Notice highlights a discrepancy in SOPsSafety Recommendation 2011-073
It is recommended that the Civil Aviation Authority resolve the conflicting expectations of flight crews and air traffic controllers following a loss of communications during approach.

[172driver]

Sounds reasonable, given that he was in the tower and had no obvious way to teleport himself to the TBM and take over its controls.

Light signals? Would perhaps take too long, but the option is there.

PS: I have not idea if the light gun in a tower is permanently rigged and instantly available.

[BabyBear]

172 driver, read the report, no light available as removal had been approved and test conclude light used from the tower would be ineffective. Interesting reading.

BB

[Pull what]

Yes-failure to read relevant information has a bearing on this incident.

Standard procedure surely is to go back to the frequency you came from if no reply on new frequency

[IO540]

Thank you for the link Whopity.

Interesting report...


The ANSP at Birmingham, in common with several other major UK airfields, had applied to the CAA to remove their signalling lamp approximately six to ten years before the incident. This application was supported by a study which showed that signal lamps had limited effectiveness at those airfields. The CAA acknowledged that they had given permission for the signal lamp to be removed but neither they nor the ANSP could locate a copy of the application, permission or the supporting study.Overall, the report reads a lot less dramatic than the initial posts here which basically suggested that the pilot was some kind of a cowboy. Also, if the aircraft had indeed been impounded once on the ground (as had been reported here) that would seem a massive over-reaction. By who, I wonder?

The Tower controller had logged onto the briefing system after the Safety Notice had been issued but had not seen or acknowledged it. It was subsequently determined that the Safety Notice had been removed in error and thus was not available for the controller to view or acknowledge at that time.After it became apparent that the controller had not read the Safety Notice, the AAIB asked the unit air traffic manager to interrogate the system and provide a list of staff who also had not electronically signed as having read or printed it. The system generated a list that included this controller but also included members of staff who were not licensed controllers and were therefore not expected to have read it. It also listed controllers who for various reasons were no longer at the unit.Interesting pre-occupation by the AAIB with NDB approaches.

The pilot stated that he last flew an NDB approach four years before the incident and could not recall flying one in the TBM 850 even though he was reported to have accomplished two during his TBM course.I can see they could not officially say anything else, but they really ought to know that practically every modern pilot flies NDB approaches using the GPS, either using an overlay or using the OBS mode. Thus, an NDB approach is no different from e.g. a VOR approach (which would be flown using the same method). The actual navaids are (or should be) tuned in, at least.

The lateral accuracy of the final approach suggests he did indeed use the GPS (**)

Apart from his frequency selection mistake, he did not commit any crime by landing without a landing clearance.

ICAO Doc 4444, the UK AIP and the specific instructions for Birmingham, in common with most other UK airfields, require a crew having lost communications on intermediate or final approach to continue visually or by means of an appropriate final approach aid. As the pilot of N850TV had lost communications, albeit through a frequency mis‑selection, this is the situation in which he found himself.Following a loss of communications, ICAO Doc 4444 and the UK AIP require aircraft to continue visually and at least 17 multi-pilot public transport passenger flights have landed without clearance in the UK in the 12 months to March 2011. In those cases the runway was clear. [my bold]

In summary it is likely that air traffic controllers assume that a loss of communications will probably result in a go-around while pilots are expecting that they will land if they are able.The ATC assumption assumes the pilot detects the loss of comms, but that is obviously not always going to happen...

Another little gem:

The ATIS also significantly exceeds the recommended MATS (Part 1) length:)

(**) I have just looked up the ILS 15 for EGBB and wonder if the AAIB report page 2 (shown as page "54" in the PDF, curiously) has a mistake on it. The plate does not suggest the ILS or NDB is offset, but the AAIB diagram shows a ~ 5 degree offset. The Jepp plate shows both the ILS/NDB 15 and the runway 15 to be 148 degrees.

[Contacttower]

Sounds like a case of the pilot getting a bit overloaded and making a simple mistake with the frequencies. However even considering poor visibility and its position on the displaced threshold I do find it hard to understand how he simply didn't seen the Dash 8 sitting there.

[Daysleeper]

I have just looked up the ILS 15 for EGBB and wonder if the AAIB report page 2 (shown as page "54" in the PDF, curiously) has a mistake on it. The plate does not suggest the ILS or NDB is offset, but the AAIB diagram shows a ~ 5 degree offset. The Jepp plate shows both the ILS/NDB 15 and the runway 15 to be 148 degrees.

The NDB ILS DME is not the same as an NDB DME, an easy mistake to make with the way plates are named... for Brum

Jepps plate 11-1 is the NDB to the ILS with DME for Runway 15, thus the NDB gets you to the ILS which is the final approach aid, inbound course 149

Jepps plate 16-1 is the NDB with DME for Runway 15, thus the NDB is the final approach aid, inbound course 154

the Runway is 149.

I accept my plates may be out of date so a 1 degree change to 148 may exist now.

every modern pilot flies NDB approaches using the GPS, either using an overlay or using the OBS mode

Does the G1000 approach database include the NDB approach at Brum? Had a play with a Garmin sim the other day that only had the ILS...

[Pace]

Do you think he would have even noticed a light signal if he didnt even notice a large aircraft on the runway starter strip?
The facts in the report are that he was a low total time and low on type pilot.
He flew for several hours from Italy with three sectors to pick up other businessman as a sole pilot.
The workload on such a mission even on the ground with the sectors would have been high!
Add to that that he was also dealing with his business associate PAX and probably thinking ahead to his business meetings and in all it doesnt make a good mix for NOT making a mistake.
The confusion over the radios, probably seeing the runway between the numbers clear and the overiding desire to land probably lulled him into doing something which he would now kick himself for.
Sadly it was in an N reg with FAA licences and and will probably be heralded as an example of JAA V FAA while such errors are made in every realm of aviation and licence even multi crew.
In future he would be best to have a sit in in the right seat who can take some of the workload on a multi sector flight like that!

Contact Tower
He could have thought that ATC had realised that he had lost his radio and were holding the aircraft clear of lining up on the numbers waiting for him to land and clear?
Crazy maybe but overloaded pilots do get crazy ideas.

Pace

[Pull what]

Sounds like a case of the pilot getting a bit overloaded and making a simple mistake with the frequencies. However even considering poor visibility and its position on the displaced threshold I do find it hard to understand how he simply didn't seen the Dash 8 sitting there.

Sounds like a case of landing without clearance

[IO540]

The NDB ILS DME is not the same as an NDB DME, an easy mistake to make with the way plates are named... for Brum

Jepps plate 11-1 is the NDB to the ILS with DME for Runway 15, thus the NDB gets you to the ILS which is the final approach aid, inbound course 149

Jepps plate 16-1 is the NDB with DME for Runway 15, thus the NDB is the final approach aid, inbound course 154

Interesting; thanks. I guess he flew the ILS inbound. There does not appear to be any actual problem with that. Why is the NDB inbound offset? I cannot see relevant obstacles.

Sadly it was in an N reg with FAA licences and and will probably be heralded as an example of JAA V FAA while such errors are made in every realm of aviation and licence even multi crew.

As always...

[Contacttower]

It also seems that because of his high workload and letting the aircraft get ahead of him (note the speed mentioned at 1.6 DME in the report!) he gave himself a lot less time than he otherwise would have to consider his options.

Standard procedure surely is to go back to the frequency you came from if no reply on new frequency

Well quite; yet to have this happen on final approach but several times enroute I've had to go back to the previous, either because I've heard it wrong or been given it wrong. As the report mentions though the fact that the radio on this aircraft had failed before may have lead him to assume in the confusion of the moment that it had done so again...

As for changing to 7600 I agree it is a bit fiddly on the G1000.

[IO540]

Standard procedure surely is to go back to the frequency you came from if no reply on new frequency

Of course, if you have time.

[172driver]

Read the report now, interesting reading. What strikes me is this:


The trial concluded that, while a white lamp could
be seen at a range of about 1 nm, a red or green filter
would make the lamp impossible to detect

I once flew into Malaga when the tower controllers there tested their lamps (and informed everyone on frequency about it). Both red and green were most definitely visible from about 5 miles out. Would perhaps not have made a difference in this case, but why withdraw the lamps from several airports (according to the report) ?

[IO540]

IMHO, on a typical large airport there are so many lights around the place that one little light in the tower is not going to be noticed. Few pilots are going to be watching the tower anyway when on final approach, and at many airports there is more than one apparent tower (with the extra ones being disused, etc).

[Pace]

I think all the talk about the approach is largely irrelevant other than MAYBE adding a little more loading to a single pilot who had flown from Italy making two pickups enroute with all the extra workload that would have entailed.

One big point not covered is the advisability of allowing an aircraft to enter a point where another aircraft could pass closely overhead.

An aircraft ready to go before another on approach should be lined up from an ajoining taxiway directly onto the runway and onto the numbers! The aircraft should have been instructed to line up not held some way back on a starter extension.

Had the aircraft been sitting on the numbers there would be NO room for confusion which there obviously was in this situation.

Pace

[Cows getting bigger]

I know a few ex RAF runway caravan controllers who are looking for a job. :)

[mad_jock]

Pace at Brum the Q was holding where you would normally start your roll from. Its a displaced threshold due to terrain clearance on approach.

15 can be a bit of a sod as well if they keep you high especially from the north.

And it can be pretty bumpy below 1000ft.

As for the tower its way down the other end with a none restricted GA apron right infront of it which is pretty well lit. I would have though with reduced viz and looking where you were landing you wouldn't see the tower. It would appear to be a hanger structure off on the right next to an apron.

As for not seeing the Q. Well I have similar when they have a light bit of concrete leading into black tarmac. You can't see the aircraft on it even if its CAVOK and you know its sitting there. The aircraft lights are low intensity and the runway lights on 15 are the high intensity CAT II lights.

And this whole business of swapping from the planned approach onto the ILS to an NDB is always fraught with potential for cockup. As a commercial crew you wouldn't be condemned for asking for delaying tactics so you could re-brief.

Lets face it he was just aviating

He had his finger tips on the elevator navigating.

And communicating was just a bit to much.

In someways this out come of landing over the top might be the most safe for all concerned. Because looking at the flight profile I suspect a go-around might have ended with an even worse situation developing.

[172driver]

Few pilots are going to be watching the tower anyway when on final approach, and at many airports there is more than one apparent tower (with the extra ones being disused, etc).

Well, if you suspect you've gone NORDO - as this pilot had - you might. I am not familiar with Birmingham but the lights used in the trial I witnessed definitely stood out.

[mad_jock]

Its a completely different style of airport and the tower is an old style one which is built into the old terminal on the south of the runway on the 33 end.

Malaga tower is huge and distinct along with the fact as well that malaga has only CAT 1 lighting and no center line lights.

Landing in CAT II lighting is something out of close encounters of the third kind. there is a huge mat of banded strips in the touchdown zone the leading lights are stripping away.

He broke through at 600ft more than likely with quite a bit offset last thing he will be focusing on is the old terminal building with a red light shining out the window.

I do have alot of sympathy with the tower controller as well I might add. Sounds like a ****e old day at work at Brum with the wx pants and the ILS out. Back to the days when it was out on the 33 end and all the chaos that they had to put with for months until the work finished.

[Pull what]

So what do you do when it all gets too much on the approach?

a ) Continue and land without clearance
b) Go around

[HeliCraig]

As the report states, the pilot was within rights to do either.

However in a loss-comm scenario you are going to have to land without clearance at some point... :ugh:

[Pull what]

Of course, if you have time. What, just move the selector from one box to the other -that would take less than a second!

[IO540]

Exactly.

In fact you are authorised to do a complete flight and a landing, without any clearances, in a lost comms scenario in the IFR/CAS environment.

There are some limitations in the watertightly-compartmented English airspace e.g. cannot enter CAS on a flight which is OCAS but flight planned to land on a CAS airport, but AFAICT this is one of the variations which ICAO didn't think of.

Speaking of TBMs, some poor chap had a nasty prang here (http://miami.cbslocal.com/2011/10/12/air-plane-lands-on-fla-turnpike-traffic-halted/). I think I have flown with the pilot, too.

[Pull what]

Maybe but this part is very relevant and is basic knowledge for the issue of an RT licence, in the UK anyway!
.
There is no evidence to suggest that the pilot tried to diagnose the apparent radio failure.

[mad_jock]

Aye but you seen those G1000 if you press the wrong button your knackard if you don't know the thing inside out and know how to get back to where you started. Its like getting locked into an overlay on an EFIS machine with no clue how to get your primary data back with all the buttons doing overlay stuff instead of what it says on the button.

When I first started out on the G530 a few times I had to take control from the FO through no fault of their's. It was just so they could put there head down and sort out my kack handed button pushing.

[Pull what]

Yes I know what you mean Jock but if you fly in CAS with public transport flights you expect a minmum level of competency surely?

CAP 413 advises
When an aircraft station is unable to establish contact with the aeronautical station
on the designated frequency it shall attempt to establish contact on another
frequency appropriate to the route being flown. If this attempt fails, the aircraft
station shall attempt to establish communication with other aircraft or other
aeronautical stations on frequencies appropriate to the route.

All he had to do is recontact the approach controller-in fact he wasted time believing that the controller was on the phone!

[mad_jock]

Thats all very well when you have a smidge of spare capacity.

Not with standing single crew and no call from the other seat "wrong frequency you knob!"

I have had some hairy old approaches onto that runway with the ILS dropping out while we were on approach converting to a NDB. And it wasn't pleasant with two crew who had been doing "brum" for a couple of years 5-10 times a week.

Single crew, non native speaker in a new aircraft with systems he was just getting into, he must have been hanging out his backside. I would have needed the AFS to help get the seat cover from between my bum cheeks even if everything had gone to plan and there hadn't been an incident.

[HeliCraig]

Yes I know what you mean Jock but if you fly in CAS with public transport flights you expect a minmum level of competency surely?

All he had to do is recontact the approach controller-in fact he wasted time believing that the controller was on the phone!

Yep... have you seen the time span involved here? Not long. Fortunately he engaged in a little bit of common sense (as someone else has alluded to):

Aviate, Navigate, Communicate. He did the first two, by the book, and landed his aircraft. He didn't have the capacity to the third by the sound of it.

[IO540]

Single crew, non native speaker in a new aircraft with systems he was just getting into, he must have been hanging out his backside. I would have needed the AFS to help get the seat cover from between my bum cheeks even if everything had gone to plan and there hadn't been an incident.I see where you are coming from but you may not be up to date with the level of cockpit automation, and resulting potential pilot workload reduction, in a modern aircraft such as a TBM850 (http://www.peter2000.co.uk/aviation/tbm850/index.html).

So I think your amusing portrait of a white knuckled pilot who is sh*t scared all the way down the approach is way overdone.

I fly relatively little "hard IFR" (most IFR is in VMC) but I have never had the slightest problem flying any approach, and I have a lot less cockpit automation than this TBM has.

It is also pretty difficult to get signed off, and insured, for a TBM without knowing where the knobs are and what they do. The G1000 is opaque to a newcomer, and is perhaps forever beyond some people who can get a PPL, but not to an intelligent pilot who has done the course.

Aviate, Navigate, Communicate. He did the first two, by the book, and landed his aircraft.Exactly.

[mad_jock]

I am sure its a lovely aircraft.

I know when I went from steam driven no auto pilot into a Honeywell EFIS machine it took quite a while for things to settle down until the automation didn't use up more capacity than kicking the AP out getting a full rose up and just flying the sod.

Yep once you get up to brown belt with the kit its easy to press a button twiddle a knob press another two and hey presto your beam bar is locked onto a presudo VORTAC approach it captures you dial in 600ft/min decent and you pop out right where you expect to.

I wouldn't say he was **** scared either that usually comes afterwards.

And I have flown lots of hard IFR not seeing the ground until mins for sometimes weeks at a time 4 flights a day in every bit of ****e wx you can find in Scotland and Scandinavia. I have done it without a AP on steam instruments and with a Honeywell EFIS. But I have always had someone sitting next to me to check what I am doing and also take work load off me. I really do take my hat of to folk that do do it single crew its hard enough work with two of you.

[Pace]

I tend to agree with mad Jock on this one. This Pilot was flying single pilot. Not only that but it was not his only Job, He had a meeting to consider, he had two other businessmen to collect from seperate airports enroute with the extra workload that would have entailed.
He only had 1100 hrs total time with not a lot on type.
English is not his native language.
Yes most of us would have gone back to the last frequency used but consider this.
The mind plays tricks! Under the pressure of landing with very little time before landing and very little time to sort out whats wrong its very easy especially with a system you are not 100% familiar with to jump to the wrong conclusion!
My radios have failed they failed before so thats it they have failed! end of story.
He should have dialed in 7600 but again in the time scale left didnt.
With that amount of time left to touchdown even with 7600 what could ATC do? All they could do was to inform the other aircraft to stay put!
What could the pilot do different believing in all sincerity that his radios had failed?
He could have executed a missed approach and flown a procedural approach?
He could have taken up the hold and tried to sort things out?
He could have just landed! That is presuming as Jock said that he even saw another aircraft sitting way before the numbers.
The numbers is where his focus would be.
He was not the only one at fault in all this IMO although its easy to point the finger at one scapegoat.

Pace

[Red Four]

An interesting read, as ever. AAIB, as seems usual recently, seems to have more to say about the ATC side of things rather than the piloting elements.

A few thoughts that occurred to me reading through it (no angle to grind, just observations):

Even though type of approach not mentioned on ATIS, wouldn't Radar have told pilot what type he was being vectored for in one of first transmissions?

The ATIS did mention ILS transmissions, so non-native english speaker may have well understood this to mean that the ILS was now serviceable again.

Whilst being vectored, the pilot says his speed is going UP to 170, so maybe he speeded up in response to the call to reduce to 180knots - may have left him residually faster than he had planned for.

Pilot readbacks frequency incompletely 'One Eight Three'. Perhaps controller could have re-iterated 'One One Eight Decimal Three' back to pilot in case he did try dialling in on '183'.

With a radio failure evident, did Tower controller initiate a 'Local Standby' in case of any other problems that pilot may have had and hadn't been able to communicate to ATC?

[IO540]

It is very common to hear a readback by a pilot which is either utterly illegible or simply wrong, and ATC dont query it. UK ATC tend to do so but abroad they rarely do. A lot of airliner comms is very sloppy so mistakes tend to be ignored.

On the whole I was impressed with the AAIB report. Some recent GA related ones were of very poor quality especially where an N reg was involved.

[Helen49]

Sounds like the typical Swiss cheese with lots of scope for lots of people to learn from the event!

[2 sheds]

The trouble is that everybody can quote Swiss cheeses but not many people think in practical terms of what it is an analogy for! (Actually, it's a damn' silly analogy - why would anyone ever want to take slices of Emmental and carefully align them - or are you supposed to throw them in the air and see how they fall?)

2 s

[Pull what]

My radios have failed they failed before so thats it they have failed! end of story.

Yes just like Kegworth really-engine has failed think its that one-oh just shut it down then.

A moments thought can prevent a lifetimes regret.

[frontlefthamster]

This does boil down to <aviate, navigate, communicate>, which the poor chap did to the best of his (admittedly, limited) ability.

Swiss cheese is indeed a load of nonsense, but it's kept Jim Reason in fine wines for long enough for him to work out that there's no money to be made in aviation.

Sid Dekker is your man, if you want to learn about 'safety' without having accidents to start with. Google him, then use amazon etc. You'll be pleasantly surprised.