Crash-Cork Airport

[757_Driver]

This particular accident has many human factor and operational factors that are of serious interest to those of us that actually do this for a living, and I include ATC, dispatchers, crewing, ops departments etc etc in that.
May I suggest that those of you who DON'T do this for a living refrain from continually writing utter rubbish on this thread. For instance if you don't understand what an EU-OPS approach ban is, or where you get the RVR's from that make that decision then keep quiet, and stop posting the same crap over and over again.
Alternatively can we have a forum that armchair observers, trolls and spotters cannot post in.The official report (for those of us that can actually wait for it) will be very useful reading, but there is also much to be gained by discussing this with fellow professionals.

[freespeed2]

Between 0930 and 1000, the RVR improved to more than 550m.

EICK 101030Z 09007KT 1800 R17/P2000 R35/P2000 BR FEW001 SCT002 BKN003 06/06 Q1010 BECMG VIS 5000
EICK 101000Z 09008KT 0400 R17/0600N R35/0450N FG BKN001 05/05 Q1010 NOSIG
EICK 100930Z 08005KT 050V110 0300 R17/0375N R35/0350N FG BKN001 04/04 Q1010 NOSIG

matspart3; You make a good point.
The 0930hr METAR for EICK was below limits, the 1000hr METAR was above.

I was flying around the time of the accident, although I was working Shannon centre. An aircraft was holding at EINN for wx to improve to Cat I limits. The controller advised him that the rvr had improved, and the pilot elected to start descent for an approach. About 3 minutes later the controller advised that the RVR had dropped below limits again, and the aircraft returned to holding. It subsequently diverted to EIDW.

The point I'm making to the non-professionals is that METARS are issued every 30 minutes. During the interim there is no prohibition on the RVR changing from that published in the METAR! ATC will advise changes to the pilot as they happen. The ATC guys in EICK are some of the best in the business at this, and they know the requirements.

Any suggestions that the crew approached below RVR limits based on the METARs is wildly speculative, uninformed and unfair. The ATC tapes / CVR will confirm or refute the situation regarding RVRs at the time. Until then, I think it should be left alone, as it adds nothing to the informed debate.

[overun]

My first, and l hope, my last post on this tragic topic is would you mind awfully deleting your Marvel Comic graphics 757 ?

Rabc.
ln the face of offence you persisted in a gentlemanly fashion.
Well done !

[fdr]

Perhaps somebody needs to have a read.

Perhaps somebody needs to have a read.

RAAAAAAABC

I agree with that.

So how about all the Nintendo pilot's laying off the subject until the accident report has been done by those with all the information, and the competency to analyse the information, and the processes that minimise the chance for making outrageous nonsensical allegations that are offered as fact rather than the fiction they are.

How would you feel as the NOK to have the media spouting the BS that is placed on this forum where the crew are unable to defend themselves from the inaccuracies of this BS.

[PBL]

So, we are now up to page 21.

Just in case there is anyone here interested in analysing the accident - that is, what actually happened to the airplane, or might have happened, here is my summary of the available information. The links to posts are permalinks, but the numbers displayed are current sequence numbers, which of course may change.

jbsharpe gave the link to the IAA announcement Irish Aviation Authority - Aircraft Crashes at Cork in his post #41.

4015 gave the link to an article in the Daily Mail with many pictures in his post #50.

Super VC-10 referenced the first report in Flightglobal Manx2 Metroliner crashes in Cork: fatalities reported in his post #64.

aterpster posted the approach charts for Cork, EICK, in #83. Direct URL is http://www.terps.com/charts/Cork.pdf.

sevenstrokeroll gave info on the Garrett TPE-331 engines with which the plane was equipped. Here is a summary I posted to a private group:

Originally Posted by PBL

There is a negative-torque-sensing system on the Garrett TPE331 turboprops. They are single-shaft, so it is necessary to feather the prop partially if the prop starts driving the turbine during flight, to prevent enormous and probably dangerous asymmetrical drag, and there is a mechanical system which does this. One may complete feathering manually, or there is auto-feather, depending on the installation, and AFAIK there is no word yet on which for EC-ITP.

During normal shut-down, it is necessary to lock the props in fine pitch manually, to avoid excessive drag on start-up. During an abnormal shutdown, as obviously occurred here, lack of oil pressure leads to partial feather if it's possible.

The summary was endorsed by a colleague of almost two decades who flew Metroliners. sevenstrokeroll's original posts are #284, #287, #289, and #291 . Ron Herb contributed some more in #290 as did Templer in #299.

I gave the reference in #134 to David Learmount's later article in Flightglobal/Flight International. It seems to have been discovered by others here, three days later, which goes to show just how many people are truly interested in reading what is posted here. To repeat: PICTURE & GRAPHIC: Crashed Metroliner in Cork identified as EC-ITP.

The Blimp posted a useful picture of the wreckage lying next to the runway at #189.

Machaca posted an enlightening photo of the wreckage at #266. I suggested that it might be worth thinking about how the prop got bent that way in #320. Irish Steve was the first to note that a blade from the left prop was in the feathered position, from the original picture posted by The Blimp.

wozzo posted an article from the Irish Examiner giving some details of what investigators said at a news conference in his post #330

Someone also posted a link to a collection of photos in thejournal.ie at In Photos: The Cork plane crash tragedy TheJournal but right now I am missing who it was (apologies!).

I still think it is worth thinking about how the left prop got that way in Machaca's picture. I don't mean in general - I mean a plausible detailed mechanical sequence. For example, how long does it take to feather a prop in fine pitch on an TPE-331 engine producing normal power? A longer or a shorter time than it takes to slide a couple of hundred yards on landing and flipping over?

(And for all this continuing speculation about people busting minimums - does anyone here know they didn't have vis? No, no one knows it. Indeed, no one knows it but the pilots, and they are sadly no longer with us. Note that reports say that the fog was lifting, indeed was gone within a short period after the accident.)

[andrasz]

Originally Posted by PBL

Just in case there is anyone here interested in analysing the accident - that is, what actually happened to the airplane, or might have happened...

My last comment, then I will shut up:

A little physics 101:

Spinning any object requires torque (that is, rotary force). Producing torque on any object involves producing an equal and opposite torque on the producer. The effect of these forces will be proportional to the size and mass of the object and the producer. In case of a prop aircraft engine, the force required to spin the propeller will produce a twisting force on the engine itself around the axis of the engine, opposing the direction of the propeller spin. The torque is low when the propeller is at a constant rpm, as only the air resistance needs to be countered. It will shoot up exponentially when accelerating the prop, and will be at its maximum when applying full power from idle and the engine spools up at maximum fuel burn.

As the engine is bolted to the airframe, the negative torque on the engine will be transferred to the aircraft to the centre of mass (which is very close to the bisection point of the roll and pitch axes, where the fuselage and the wing meet). A clockwise spinning prop (viewed from front) will produce an anticlockwise twist on the engine and airframe. This translates to a starbord engine producing an upward lift on the centre of mass, while a similarly placed port engine will produce an equal downward force on the centre of mass. (Iolar's linked photo clearly shows both props on a Metro spinning clockwise). This is torque balance, engine acceleration will have no effect on the aircraft as the two equal and opposing forces balance out.

In an engine out scenario (or single engine aircraft), the torque imbalance needs to be countered. On fixed wing aircraft this is the job of the ailerons, on helicopters this is done by the tail rotor. If for any reason the torque remains imbalanced, the aircraft will rotate around the axis of the torque, the engine.

In present accident, the roll to the right is perfectly consistent with the port engine producing little or no torque on the airframe, while the starbord one continuing to do so. This scenario will twist the aircraft right around the axis of the starbord engine, and is also consistent with right wingtip strike and the inversion, as well as the relatively undamaged state of the port propellers.

All I'm saying is that what little facts we know are consistent with a mishandled port engine out at go-around scenario. They are however inconsistent with a low visibility CFIT type accident.

[Matt101]

Originally Posted by andrasz

All I'm saying is that what little facts we know are consistent with a mishandled port engine out at go-around scenario. They are however inconsistent with a low visibility CFIT type accident.

Actually given the statement from the AAIB that the aircraft appeared to be in normal working order prior to impact, at this time, such a statement is in pure contradiction of what little facts we have.

You may in fact be proved correct in the future but for now.....

[rlsbutler]

Thanks PLB and Adrasz

Thanks PBL for pulling all those contributions together.

I share the occasional irritation when a contributor opens an argument afresh that has been well covered several pages earlier – but well understand how a newcomer cannot face 21 pages of preparatory reading before he chips in. I will not be as conscientious as PBL about giving post references, as I presume to offer two more worth-thinkings-about:

The first of these concerns the signs that, somewhere near the runway threshold, the aircraft had no power at its port engine yet it impacted to the right of the runway. For me (Canberras and Meteors) the normal asymmetric disaster finishes with a roll in the direction of the dead engine. Andrasz’s very persuasive contribution makes the story appropriately more complicated for someone like me. I am used to the concept of torque-stall, with its heavy penalties for the odd P-51 Mustang on go-round. I am surprised to think that a civil airliner’s offset engine, acting to lift the fuselage at a (shortish) distance, would have so disastrous an effect.

So I continue to wonder if we may be looking at an ordinary over-control situation, either where the wrong engine is treated for failure or where the wrong rudder is used ? Someone brought up the possibility of fuel starvation. Although a subsequent post cast doubt on the speed with which an engine would respond to transient starvation, I do not see how that engine response should not be immediate and extreme. Suppose the pilot with coarse control manages the first engine rundown but, in so doing, induces fuel starvation in the opposing engine ? Metroliner veterans have made the point that aileron control is habitually coarse on this type. Would not the pilot be spontaneously as rough with his feet as he has learned to be with his wrists ? So (1) port engine runs down (2) full (excessive) right rudder applied and port engine feathered (3) starboard engine runs down (4) aircraft does rudder-induced flick to the right. With Andrasz’s torque-stall, when the starboard engine spools up at the last moment as the fuel slops back its way, we may have more than enough to explain what starts as a riddle.

The second topic concerns the suggestions (an investigator’s initial observation even) that the aircraft progressed laterally along the runway to settle on the grass. Not only does the condition of the undercarriage make that improbable, but the usual dynamics would have caused the aircraft to pitch onto its back and leave the wreckage facing the other way. The thread has recent observations about the extent to which the nose seems to have penetrated the grass surface. These suggest not just that there was a vertical speed vector but that the aircraft was on the attitude of an inverted dive. That fits any theory that sees the tragedy being played out in the air over the runway – even if the aircraft was flying close enough to the ground to cause a wingtip to strike. We could take into account the evident distortion of the nose undercarriage by speculation about a “wheelbarrow” type of landing – except that there is so little distance down-runway from threshold to abeam the wreckage.

[SATCOS WHIPPING BOY]

Even more speculation in the above post which will no doubt go on to feed the media.

Cork air disaster: Spotlight turned on why crash plane didn’t divert - Republic of Ireland, Local & National - Belfasttelegraph.co.uk

more pictures of the wreckage being lifted. and more speculative comments by the press no doubt.

apparently diverting is now a tradition

"...... (pilot) appeared to depart from an aviation tradition under which pilots divert to another airport if poor visibility thwarts a second landing attempt..."

[widebody69]

The second topic concerns the suggestions (an investigator’s initial observation even) that the aircraft progressed laterally along the runway to settle on the grass. Not only does the condition of the undercarriage make that improbable, but the usual dynamics would have caused the aircraft to pitch onto its back and leave the wreckage facing the other way.

Investigators to reconstruct remains of stricken aircraft | Irish Examiner

Quote: "....The AAIU probe, led by air accident inspector Leo Murray, has already established that the tip of the Fairchild Metroliner’s right wing hit the runway as it landed in fog on Thursday morning. The plane then turned onto its roof and ran almost 200 metres along the runway before veering into a muddy verge where one of the engines caught on fire..".

The above was confirmed directly by the investigators over the weekend, I think the implication is that the aircraft flipped immediately and travelled along the runway inverted. The investigators have confirmed there was a debris trail for 180-190m before the aircraft left the runway.

[netstruggler]

The torque is low when the propeller is at a constant rpm, as only the air resistance needs to be countered. It will shoot up exponentially when accelerating the prop, and will be at its maximum when applying full power from idle

I think your physics is a little rusty.

1. The relationship between torque and angular acceleration is proportional rather than exponential.
2. The torque required to accelerate the prop is almost certainly negligible compared to the torque required to counter air resitance (assuming the prop is generating thrust).
3. There's no reason for torque to be low at constant RPM nor for it to be a maximum when accelerating from idle speed.

[Avenger]

Andrasz

I'm confused, in every twin engine I've flown, inc TPs and jets if the Port Engine fails, the A/C yaws and rolls to Port, Rudder, not Aileron is used to control torque...The right wing hit the runway, suggests roll to the right, either Starboard Engine failure, or more likely, overbanking too close to the ground.

Your comment:

"In present accident, the roll to the right is perfectly consistent with the port engine producing little or no torque on the airframe, while the starbord one continuing to do so. This scenario will twist the aircraft right around the axis of the starbord engine, and is also consistent with right wingtip strike and the inversion, as well as the relatively undamaged state of the port propellers".

Can't agree with this one bit...

[BALLSOUT]

ANDRASZ.
What you say is incorect. If you suffer a failure on the left side of a multy engined aircraft it will yaw to the left. If you fly them yourself you must know (dead leg, dead engine) meaning if the left has failed you will need to be pushing with the right rudder pedal, to prevent the aircraft rolling to the left. this is also achieved by the rudder and not the ailerons, which are the equivilent to the tail rotor on a helecopter. Torque caused by the rotation of the propeller is not a big facter in all of this, it is more to do with the imbalance of the power, ie no longer central to the central axis of the airframe, and all of the extra lift being generated over the right wing from the prop on the live engine lifting the right wing.
torque has more of an influence on a normal take off roll on something like a tail wheel aircraft. It would still however be counteracted by the rudder and not the ailerons.

[andrasz]

BALLSOUT

Please don't confuse YAW caused by loss of thrust/increased drag with ROLL due to torque. As some mentioned before, under normal conditions where the size/weight of the prop is insignificant compared to airframe weight, the torque effect is negligible. However the Metro has a relatively large prop compared to it's size/weight...

[Irish Steve]

I think we're in danger of becoming too focussed on the engine issues to the exclusion of a number of other factors.

A lot of people here will have seen videos of the tragic crash at Narita of the MD11 that lost a wing due to spar failure. (If not, a google search for MD11 crash Tokyo will provide a salutory view of how if can go badly wrong very quickly, and this was not in poor visibility)

The result was a sudden and uncontrollable roll due to the remaining wing still being able to produce significant lift, which took the aircraft inverted at a massive relative speed.

It also turned the aircraft, significantly, as it was rolling.

From the very clear images during the removal of the aircraft, it was clear that AT SOME STAGE. (Please note my emphasis here, we don't know the exact timescale) the starboard wing outboard of the flaps has become detached from the aircraft. All I will say is that it seems unlikely that the activities of the emergency services or the subsequent removal of the aircraft were responsible for this damage, but we do not have definitive information at this time.

If the departure of the wing panel occurred at the time of tip impact, as the aircraft was still flying, the roll and probably yaw factor would be valid in terms of the eventual aircraft position and attitude.

We do know that the starboard wingtip made ground contact, and it's being suggested by a survivor that this happend before the gear was on the ground.

If that is the case, the critical information that we don't yet have is that if the starboard wing failed, and was lost at that point, then there would have been significant roll input from the asymetric lift factor that would not have been in any way controllable, and it would have rolled the aircraft rapidly, and the subsequent ground contact inverted would explain the upward displacement of the engines from the correct thrust line given the speed of the impact.

I don't know the exact structure of the engine controls in the Metro, but given that the PF would have been "hands on", very much so, it's altogether possible that involuntary movements of the throttles could have been precipitated by the upset of the aircraft, and we have no way to know what the result of that was. It is very possible that one or both throttles got moved suddenly and possibly in the direction of feather simply as a result of involuntary hand or arm movement.

Before anyone jumps down my throat, or goes for the jugular, I do have significant light twin flying experience, and on occasions, when flying solo and in the cruise at safe altitude, I deliberately left the fuel supply on the tip tanks until one of them became exhausted, and then dealt with the scenario, as it was a quick recovery and didn't present a significant risk, but there was no way to predict the exact timing, so it was a better "check" than deliberatly setting "zero thrust".

What was apparent and got the attention was that the failure was sudden, and caused significant upset, and if not caught almost instantly, had the potential to cause a lot more problems.

Add in the problems of poor weather, being close to the ground, and all the other possible scenarios that have been suggested here, and I'm going to have to say that I'm not sure that I'd have been able to get into Cork in the scenario that's been outlined.

That said, I'm not sure I would have tried that hard either, but with the apparent alternate also having problems, as was the case, I would have been in a hard decision scenario, and they are never easy.

I would want more information in a number of areas before I comment in more depth, and some of that information won't be available for some time to come, possibly not before the final report.

[CarbHeatIn]

From today's Irish Times:

Department of Transport Air Accident Investigation Unit member Paddy Judge said the team was examining the Flightline BCN Metroliner flight data recorder and cockpit voice recorder.

Mr Judge cautioned against drawing premature conclusions as to what caused the crash after several media reports suggested the aircraft flipped over after its right wing hit the main R17 runway at Cork Airport on approach.

He said, “at some stage it is correct that the right wing hit the runway, but we need to be cautious about what was the initiating sequence because this entire event happened in a matter of seconds”.

[cats_five]

Originally Posted by widebody69

Investigators to reconstruct remains of stricken aircraft | Irish Examiner

How common is a reconstruction?

[Genghis the Engineer]

Originally Posted by cats_five

How common is a reconstruction?

Fairly common; it basically means that all the bits will be laid out in order on the floor - if really necessary they may start re-assembling on scaffolding. It helps a great deal in working out which bit is which, whether anything's missing, and whether any particular bits of failed structure seem to be inconsistent from the rest (by which really I mean, that the broke before, rather than after, it hit the ground).

It's important to do something like this early on, if they can - because the investigators are in all likelihood still at that early part of an investigation where they are trying to gather together and document as much evidence as possible whilst it's still fresh.

G
Not an air accident investigator, but have worked with many of them.

[sevenstrokeroll]

I am genuinely confused. And I've flown the metroliner for a living.

Torque is being thrown around pretty lightly. I think some of you are confused by "P" Factor and torque. I really think that sort of thing is more of a concern with a single engine plane.

Torque is how we measure power output in this type. Jets use EPR, or N1 . Piston planes use Manifold Pressure or even just RPM.

And I am almost sorry I even brought up Negative Torque Sensing. The pilots and engineers who actually know this stuff can understand. Those who have not been exposed to this type of engine/prop are confusing the issues.

IF YOU MEAN Engine failure and subsequent Assymetric Thrust...say so. Don't throw around torque unless you know what you are talking about. We aren't talking about P51 mustangs on takeoff roll and use of rudder.

Overbanking near the ground to maneuver to the center of the runway, if initally to the left of centerline, might cause right wing tip contact. The right wingtip , as I have mentioned, is an extension of about six feet and not as strong as the rest of the wing.(simplest terms I can think of). It might have hit the ground and torn off.


"Feathered" position of port prop may or may not mean anything....we will have to see.

I do want to say that detecting and reacting to an engine failure in this type while on instruments is much more difficult than while in Day VFR flight. One might react 'instinctively" by use of aileron intially, however rudder is the only thing that will properly counter an engine failure (albeit with a very slight bank for optimal performance). Indeed the slip/skid indicator becomes quite useful here. And the saying : dead foot dead engine, might be replaced by "the dead engine has no ball"s!

[Robert Jan]

Flightline BCN no longer operating flights for Manx2 following crash