Air Vanuatu Twin Otter elevator cable snap

[Interlude2]

From the Vanuatu media: An Air Vanuatu (NF) Twin Otter on a domestic flight made a safe landing with 19 pax or pob when the elevator cable snapped just moments before landing at an outer island airstrip (Walaha, Ambae). The cable had been fitted new just 6 months ago. If the aircraft had been at higher altitude and not so close to touchdown the result could have been catastrophic, and in a similar incident in Tahiti several years ago.
NZ aviation accident people are part of the investigation.

[F-16GUY]

If the TwOtter has a conventional trim that is controlled by its own cables, it might be possible to fly it down to a safe landing using the trim only.

[The Ancient Geek]

Indeed.
Non event - just fly it with the trim wheel.

[Nemrytter]

Hardly a non-event.:=

[Basil]

Indeed.
Non event - just fly it with the trim wheel.
Hah! Very good! ;)

[evansb]

RE Tahiti incident: Air Moorea Flight 1121 was a De Havilland Canada DHC-6 Twin Otter which crashed shortly after takeoff from Moorea Airport on Moorea Island in French Polynesia on 9 August 2007, killing all 20 people on board.

It was bound for Tahiti's Faa'a International Airport on a regular 7-minute service, one of the shortest on earth, scheduled 40 to 50 times a day. Frequent takeoff and landing is believed to have been a major factor in the crash, because of wear and tear on the elevator cables, inspected only at fixed time-intervals, regardless of usage.

[Council Van]

In 2003 a Dornier 228 suffered a lightning strike which melted, or probably vaporised, the elevator control rod.

The crew flew the aircraft using electric trim, including a go around, the subsequent landing was not the best resulting in the aircraft being written off and some substantial injuries to the occupants.

ASN Aircraft accident Dornier 228-202 LN-HTA Bodø Airport (BOO) (http://aviation-safety.net/database/record.php?id=20031204-0)

I was reminded of this accident when an FO tried to convince me it would be OK to fly a DO228 through a storm, I politely declined and we went around the weather.

[westhawk]

Frequent takeoff and landing is believed to have been a major factor in the crash, because of wear and tear on the elevator cables, inspected only at fixed time-intervals, regardless of usage.In the case of the Air Moorea DHC6, I seem to recall how it was typically parked on the ramp possibly contributing to cable wear as well. Apparently jet blast from taxiing large aircraft would regularly strike the elevator from behind, causing it to move violently. Additionally, the salty tropical environment is not kind to the steel cables. Closer/more frequent inspection and corrosion protection treatments are more than just a good idea in such an operating environment.

An elevator cable break may be more recoverable in some circumstances and airplane types than others. Though I've demonstrated a trim only (elevator free) landing in a C-172 with good results, I'd be less sure about doing this in a larger airplane with higher control force required and/or slower trim. Particularly if it was unplanned and had not been previously practiced under controlled conditions at a safe altitude.

If near the ground and out of trim, things could easily go south before the effect of cranking the trim wheel affected the pitch attitude. Given the typical delay in trim effect while flying stick/elevator free, PIO may result in some airplane types and loading conditions. Anyone thinking that just because it's relatively easy to control one aircraft in one set of conditions would do well to consider that it might not be so easy in other conditions. It pays to know ahead of time!

[evansb]

Indeed. First off, Fa'a'ā International Airport is relatively small. Unsecured elevator movements while parked in a high hydro-carbon contamination area, (jet blast), may, over time contribute to several problems, one of which is premature control strand failure.

Why the DHC-6 aircrew may or may not have routinely installed a control lock is undetermined. The direction the DHC-6 aircraft were routinely parked on the apron is also subject to investigation.

Most pilots I've flown with, (and taught), know enough to park upwind and install a control lock.

As an aside, in the late 1970's and early 1980's, the BN-2A Islanders were marshalled well away from the big jets at Pape'ete's Faa airport.

[roulishollandais]

Did these DHC6 fly above VNE...:mad: as I often watched with planes used to drop jumpers!!! Enough to destroy cables .

[TowerDog]

Not sure if the same aircraft as the 2007 crash, but a plane I flew years ago crashed in the Pacific somewhere:
I flew this plane on floats in saltwater 1984/85.
All the salt going in the tail section without being rinsed out daily and flap brackets that correoded away to half thickness after a year in the salt made me nervous and I quit the job.
Years later I tracked the tail number and found it on wheels in the Pacific, just after it crashed and killed everybody.
Tail number when I flew it was N-784DL.
DL for Don Lewis, owner of the company.

Edit:
Found it, same airplane that I flew on floats/
https://en.m.wikipedia.org/wiki/Air_Moorea_Flight_1121

[The Bullwinkle]

Indeed.
Non event - just fly it with the trim wheel.
This site really does bring out some knobs!!! :mad:

[Anilv]

If you cant fly it with a trim wheel, get the attitude you desire by having the pax move up and down the cabin.

If the aircraft nose is too high, fill the front seats/ aisle, laps. If the nose is too low get every one in the rear... FO too if it helps!

I remember reading a report of a crash where a similar sized plane crashed due to an alligator(or crocodile) getting loose and every one running to one end of the aircraft to get away from it. Don't know if its true but very believable.

[9 lives]

The OP posted:

when the elevator cable snapped just moments before landing

That is a very different situation from a landing planned with time to allow for an alternate method of pitch control.

If you cant fly it with a trim wheel, get the attitude you desire by having the pax move up and down the cabin.

If the aircraft nose is too high, fill the front seats/ aisle, laps. If the nose is too low get every one in the rear... FO too if it helps!

This would be a really poor idea, re configuring an aircraft in a uncontrolled manner, with undisciplined occupants, who would now no longer be secured in the aircraft is asking for trouble! Too many people move too far, and you're out of C of G limits, out of control, and not going to regain control - and you asked them to move! Not when I'm flying!

Non event - just fly it with the trim wheel.

If you're having to consider flying with the trim wheel, for loss of other control, it's an event! The trim wheel, with practice and time to perfect technique, can be used to set up a landing, but certainly not at the last moments on the approach. The Twin Otter is a more pitch sensitive than some other types. Changes in flap position, or power, can affect pitch of a Twin Otter more dramatically and suddenly than other types, flying by pitch trim alone is not a last minute alternative.


The design requirement associated with this characteristic:

(e) By using normal flight and power controls, except as otherwise noted in paragraphs (e)(1) and (e)(2) of this section, it must be possible to establish a zero rate of descent at an attitude suitable for a controlled landing without exceeding the operational and structural limitations of the airplane, as follows:
(1) For single-engine and multiengine airplanes, without the use of the primary longitudinal control system.
(2) For multiengine airplanes--
(i) Without the use of the primary directional control; and
(ii) If a single failure of any one connecting or transmitting link would affect both the longitudinal and directional primary control system, without the primary longitudinal and directional control system.

Nothing there says "suddenly" or "at the last moment". It also must be practiced.

[peekay4]

I remember reading a report of a crash where a similar sized plane crashed due to an alligator(or crocodile) getting loose and every one running to one end of the aircraft to get away from it. Don't know if its true but very believable.
https://en.wikipedia.org/wiki/2010_Bandundu_Filair_Let_L-410_crash

[BRE]

There are shorter scheduled flights:

Nur zwei Minuten in der Luft: Das sind die kürzesten Linienflüge Europas - FOCUS Online (http://www.focus.de/reisen/flug/einmal-rauf-und-dann-gleich-runter-die-zehn-kuerzesten-linienfluege-europas_id_3885735.html)

[777boyo]

Now defunct (I believe) South Pacific Island Airways lost a Twin Otter from the Pago Pago base on the island of Ta'u in about 1984 due to the same problem on short final. The guys got it down, just, but it slid off the dirt strip into the bush and caught fire. Can't recall if there were pax casualties, but I know the American captain survived and returned to flying not too long afterwards.
I seem to recall (but open to correction) that the corrosion of the elevator cable was attributed to the local pax habit of travelling with badly sealed buckets of fish in saline water which were carried in the aft hold and tended to leak.

[sailor]

Short scheduled flights - Loganair Westray to Papa Westray took less than a minute.

[BlankBox]

...made me curious so did a little light reading...

http://www.dtic.mil/dtic/tr/fulltext/u2/a800185.pdf


...report is a few pages in...past all the chaff...

[EDMJ]

One gets the impression from reading this that this problem essentially only affects Twin Otters? Why? Surely other types, such as Islanders, PC-6 and Caravans, also operate in such harsh environments, and seem to get away with it?

[chuks]

I once had an engineer install a hose incorrectly on a Cessna 404, the hose that connected the turbocharger to the inlet manifold. When it popped off, I lost the engine due to the mixture suddenly going far too rich for the suddenly reduced manifold pressure.

This was in a fully-loaded aircraft off a short strip on a very hot afternoon, when I was very lucky to have the hose pop off during extended climb, not too late to abort or right after lift-off.

When I got back to base, the guilty party told me that all I would have had to do was to lean the mixture to get the engine back, that there was no real problem there if I had simply known what to do. That's the closest I have ever come to slugging an engineer, with his blithe notion that I was going to to sit there and twiddle with a few controls to see what might work, in a machine with a highly negative rate of climb due to gear and flaps down and a windmilling prop on a failed engine.

So there you are when the elevator cable on your Twotter goes "Twang!" but you can sort it out no probs by playing with the elevator trim ... on short final or on take-off? Better yet, simply turn around and ask the SLF to run this way or that. These things I would like to see done in practice!

The Twin Otter uses a control lock for the elevator that locks the pilot's controls, just a rod that runs between the control column and the bottom of the instrument panel. That means that the elevator cables are still exposed to stress from winds and jet blast whether the controls are locked or not. The elevator itself is protected from banging around, as are the ailerons and the rudder, but none of the control cables is protected.

[westhawk]

One gets the impression from reading this that this problem essentially only affects Twin Otters?

The twin otters have been under scrutiny for cable wear and corrosion for quite a while. Service bulletins going back for at least 20 years. One big fatal crash that attracted global attention that I know of. Service difficulty reports caused transport Canada to issue ADs requiring cable inspection and maintenance 20 years ago and mandatory life limits later on. With this type, it's been a known issue for which mitigating action was taken.

It's interesting that reports indicate that the broken cable in the Vanatu incident had only 6 months in service at the time of failure. Investigators should soon know whether it was a manufacturing or installation defect if they don't already.

Other types of aircraft are of course affected by cable wear and corrosion. In fact life limits have been imposed on all primary flight control cables for all Aussi registered aircraft. Though US registered A/C currently only have such life limits if imposed by a specific type AD, the Aussie approach could be in our future too. One source of undetectable corrosion cited by the Aussie CASA is the cable terminal ends that were swaged on during the cable assembly manufacturing process. There has been allot pushback over this issue but failed cable terminals continue to be found worlwide.

The simple fact is, the aviation industry has known about this since the advent of steel control cables! For the most part, periodic cleaning, visual inspection and lubrication (where applicable) has been considered sufficient. More recently, field generated service difficulty reports are indicating that undetected wear and corrosion has become more problematic. Hence the CASA reaction.

Some airplanes require a significant number of man-hours to properly carry out inspection and maintenance activities on cable systems. In some instances the cables must be removed from the aircraft to inspect them properly. Then they must be re-installed and rigged. My belief is that many of these cable system inspections aren't being properly accomplished. Many annual inspections are completed in the time it would take just to properly inspect the cables. Mandatory replacement is probably coming to much of the rest of the world outside Australia!



.

[777boyo]

The Twotter is a magnificently rugged little aircraft - loved every minute of the 1500 hours I spent on it, and have to say that technical problems were really rare. There were several occasions (not technical related) sitting in the left seat of a 777 in the middle of the night staring into the dark somewhere over an ocean with 10 hours to go that I would have willingly gone back to a Twotter if someone would have paid me the same money :). Having said that, the Boeing is a bloody fine piece of kit, too.

[Loose rivets]

Non event - just fly it with the trim wheel.


Luxury!! I dreamed of having a trim wheel when my elevator was solid.

Freelancing on a Shorts out of SEN. I was at about 6,000' in the London TMA.

Thank heavens, nice day and smooth. So smooth in fact that I only became aware of the stiffness of the trim gradually. Then it wouldn't move at all. Gentle pressure on the stick gave the same reaction.

Via my FO I asked London for, and got, a long gentle turn back roughly to SEN with Stansted alerted. I had asked to be routed away from populated areas. Pax strapped in, and our lovely young FA was standing at the door behind me trying to maintain her smile. I briefed on moving the passengers, but stressed that only one must move just a foot or so at a time. My FO went back to eating his sandwiches.

Moving the pax. Okay you, "I wouldn't let that happen" guys. What the hell would you have done?*
I had pressed on the Yoke just about as hard as I dared. After all, if it had gone forward sharply, and then stuck, doing a Neil Williams probably would have done nothing but keep my mind off the metaphorical hole I was in and the smoking hole I would have shortly been in.

Yes, I was gingerly altering the power to make minute changes of attitude but the approach was going to be somewhat more challenging. I had a mild sense of unreality and it made me mindful then, that an order to a passenger might be met with nothing but a blank stare. I thought through a lot of things during that time but despite having thousands of hours in command of medium weight turbo-props I hadn't got my usual bag of tricks open to me. Side-slipping, even on quite large aircraft, had always been a powerful tool for bush-style flying but that always required immediate elevator input, so a non-starter. Just keeping stable was about all I had.

Decent into warmer air freed the controls. I still recall the disappointment of the challenge being stripped away. One is really alive during situations like this.

It seemed the Shed's control run was in the belly. There had formed an emulsion of aviation goo and water - which has frozen over a longish length of the cables. I'd pushed quite hard with the above mentioned caution, but I recall bracing my entire body ready to stop pushing should it suddenly move. While frozen it was solid.


*Flying on trim. There has not been an airline I've worked for where the PF did not use the trim on a moment by moment basis. My first jet transport, the BAC 1-11, had a small gap - 3 degrees I think it was to allow micro switches to cut the Mach trim - and I still recall the clank clank of that hysteresis being taken up. Remember, we hand flew almost everything except cruise - just because we wanted to. When I fist stepped into the ATR for a somewhat impromptu bash at flying it, I reached for a trim wheel that wasn't there. It seemed insane - just as though a major flying control had gone missing.

[funfly]

Flying using the trim might be OK if the elevator is jammed but if the cable has broken it is very possible that the elevator would be able to move around offering a quite different scenario.

[khorton]

Flying using the trim might be OK if the elevator is jammed but if the cable has broken it is very possible that the elevator would be able to move around offering a quite different scenario.

Hopefully the elevator will be moving. The idea is that the elevator moves in response to the trim tab movement, and this causes the aircraft to pitch up or down.

If the elevator is locked completely solid, then you only have the tiny trim tab moving, instead of the much larger elevator, so the aircraft pitch response response will be very minimal. And, to further complicate your life, the pitch movement will be in the other sense of normal response to trim movement - i.e. if the elevator is completely frozen, trim nose up to cause a very slow pitch down. Note: this is referring to an aircraft with a trim tab on the elevator. It does not apply to aircraft where a moveable stabilizer is used for pitch trim.

The Twin Otter AFM has an Emergency Procedure to cover loss of elevator control, so the pilots would hopefully have some idea what to do.

From the DHC-6-300 AFM:

Elevator Control Malfunction

WARNING

AIRCRAFT LONGITUDINAL RESPONSE TO ELEVATOR TRIM CONTROL AND POWER CHANGES WILL BE REDUCED. AVOID LARGE AND/OR RAPID CHANGES IN TRIM AND POWER.
SECTION 3 EMERGENCY AND ABNORMAL PROCEDURES

TC Approved

If movement of the control column does not produce a corresponding change in pitch attitude, longitudinal control can be managed by using the elevator trim. Engine power should be used to control vertical speed and airspeed.

Flaps should be extended or retracted cautiously, in increments of 5° or less.

Application of power will usually result in a nose-up pitching moment, and reduction of power will usually result in a nose-down pitching moment.

Extension of flap will usually result in a nose-up pitching moment, and retraction of flap will usually result in a nose-down pitching moment.

[Chu Chu]

Also, I'd imagine that if the elevator was stuck more than a very small amount nose-up or nose-down, the trim tab wouldn't be able to overcome it.

[rileymcdu]

Another related accident (fatal) from 1994, on the west coast of Canada, near Port Hardy. Elevator cable failed due corrosion. The pilot survived, with very serious injuries.

Transportation Safety Board of Canada - Aviation Investigation Report A94P0215 (http://www.tsb.gc.ca/eng/rapports-reports/aviation/1994/a94p0215/a94p0215.asp)

[westhawk]

Another related accident (fatal) from 1994, on the west coast of Canada, near Port Hardy. Elevator cable failed due corrosion.That's probably the one that prompted TC into action the following year.

Last time I was in Port Hardy, the big issue was how to get thousands of birds off the runway so we could take off. Taxiing down to midfield and blasting them with engine noise and jet exhaust was the locally recommended procedure. It worked too! They even stayed away long enough for us to take off without hitting any.

[RAT 5]

Flying on the trim: I'm sure it would take quite a bit of trial & error and experimentation before I'd make an approach. Is the elevator frozen or is it floating? Which way does the a/c pitch with what trim input? How much trim input = how much change in attitude? What happens with power changes and can the trim control counteract pitch changes caused by power changes? There would be many variables to ascertain and gain confidence with before going near finals. Remember, you still might not know what will happen if you decide to GA. That would also be a manoeuvre to practice at height; and we haven't even discussed pitch changes with flaps yet.
Not quite so easy as first imagined. With a trimming stabiliser, à la B737, some what easier than an elevator & trim tab.

A slight drift about different ways to achieve the same using trim tabs. I was taxying out behind an old DC-9/MD-80 family. The high level stabiliser had elevators, but one was up & one was down. I was totally confused and asked, via ATC, to alert the crew. The reply was "eet's a normal" in accented italian english. Later, in discussion with a type rated pilot he explained, if I can remember correctly, the elevators were floating and pitch control was via the training edge tabs. With airspeed the elevators streamlined and all was good. So I wonder what would have happened if those elevators became stuck. Back to my PPL books about balance tabs & anti-balance tabs etc. etc.

[The Ancient Geek]

If a cable breaks the elevators will be floating and trim will work. This is clearly documented in the POH along with other factors affecting pitch such as flaps. This procedure should be regularly practiced and trained for. We were aware of this and practiced it in Africa but that was a long time ago. I doubt is standards have slipped in the meanwhile but.......

The danger is a cable break at low level with little time to respond, if the aircraft is fully trimmed at all times you have a chance, if you fly with pressure on the yoke it can all go wrong very quickly.

[V1... Ooops]

ISTR from my flying the Twotter many, many years ago that there is an elevator trim / flap / interlink to prevent a recurrence of tailplane overstress during flap retraction.
That's not what the flap-elevator interconnect tab is there for. It is there to reduce the amount of force that the pilot needs to apply to the control column to maintain the same pitch attitude prior to retrimming the aircraft. There is a certification requirement (CAR 3, later SFAR 23) that sets a limit on the amount of force that is needed to maintain pitch attitude following flap extension or retraction but prior to retrimming. Without the interconnect tab, the DHC-6 would not have met that limit.

[V1... Ooops]

...If the elevator is locked completely solid, then you only have the tiny trim tab moving, instead of the much larger elevator, so the aircraft pitch response response will be very minimal.
There was an incident many years ago in Canada in which a DHC-6 pilot encountered that exact situation - the elevator was locked solidly in position. It was not a result of an aircraft failure - the pilot was transporting fuel drums that had snow and ice on top when they were loaded into the aircraft. During the long flight, the snow and ice melted in the warm cabin, the water ran down under the floor and froze around the elevator control cable(s).

The pilot was able to land the aircraft successfully by using the trim tab to control pitch and being very judicious with engine power adjustments.

The AFM extract that you quoted earlier in this thread was added as a result of that incident.

[Jetjock330]

25 Years ago, I flew a KingAir C90 that would trim nose up in the cruise and then, disconnect the autopilot with a violent nose up response after an hour of flight or so. Months went by with this snag, changing autopilots and "Please Report Back". GTFS! Ground Tested, Found Satisfactory was the usual response.

Then one day while hand flying the departure, the trim cable snapped in flight. I reduced speed to find the trim speed and fly it from there on wards to landing.

On the ground the authorities got involved, rightly so, following up on each snag in the Flight Folio, and all those about the autopilot problem.

The trim cable became frayed and was rolling onto the autopilot drive pulley and wasn't able to leave the drive pulley as it had wound up on itself. Hence each flight, it got a little worse and worse and eventually snap.

Lessons learn't back then already, that anything wrong with trims, flight controls, stickiness in full flight control check, go straight back to the hangar!

[RAT 5]

Trimming issues can be easily dismissed. Remember Air Alaska DC-9/MD family that spiralled down onto the Pacific after the trim jack broke due to under greasing in maintenance.

[9 lives]

Trimming issues can be easily dismissed. Remember Air Alaska DC-9/MD family that spiralled down onto the Pacific after the trim jack broke due to under greasing in maintenance.

I remember that accident, and the factors, and they should be dismissed relative to discussion of a Twin Otter trim system - the two systems are entirely different. A failed Twin Otter elevator control or elevator trim control will result in the elevator more or less trailing, unless a force is being applied to one of the surfaces. The DC-9 Trim is an all moving horizontal stabilizer, with the trim actuator at the front, which with a failed trim jack screw will mot likely go divergent to an extreme of the trimming range - probably instantly irrecoverable.

The second flap interconnecting trim tab is, as stated, there to minimize pitch control forces with flap position changes. Twin Otter flaps are hugely effective, and this additional system is required.

Control system linkages are known to freeze (in the low temperature sense) in several types. That's a bit of a design oversight that it's possible. Other than that, the seizing of an elevator control in a certain position is much more rare than a one side linkage failure. In such a case, the elevator will generally float free in one direction, and can be controlled against a deliberate out of trim selected by the pilot. But this is not something that most pilots will diagnose, and correctly set up in seconds, so is not an instant fix. I have flown an aircraft on two occasions where a massive pitch control rigging error resulted in flying a circuit while applying only variations in high "push" forces on the control wheel, as opposed to being trimmed, or "pulling" for some of the desired pitch changes. This is counter intuitive, and requires a careful thought, which many pilots could not be expected to get right on short final.