Helicopter down outside Leicester City Football Club

[Overdrive]

Insider....twas not a military gig....but involved guarding some very high value items.

What is magic about the aircraft being shut down and unattended for 3.5 hours?

SASless - no magic! I think the thread looked to be going in a "Left unattended/not properly pre-flighted" direction, due to some of the stated (but not confirmed) timings.

My previous post there by the way is void as its target post was removed.

[Apate]

Extract from report into AS355 TR Control Failure over Cardiff G-SAEW, Flown by an ex student of mine:

Errr....he was ex-Navy!

[Uplinker]

As a fixed wing pilot, can I ask; how do helicopter pilots inspect the tail rotor and its drive mechanism on the walk around? Are there inspection doors along the tail boom to enable inspection of every shaft joint? I don’t recall seeing any in the helis we used to use for TV work. (Bolkow 105, Augusta 109, Twin Squirrel).

Given that the tail rotor seems to be so critical, why is there only one?. Would it not be safer if there were two separately driven tail rotors, or would that be overkill?

[Prawn2king4]

Extract from report into AS355 TR Control Failure over Cardiff G-SAEW, Flown by an ex student of mine:

DB... can you tell me the point of this illustration?

[DOUBLE BOGEY]

Errr....he was ex-Navy!


And your point being what exactly?

[DOUBLE BOGEY]

Extract from report into AS355 TR Control Failure over Cardiff G-SAEW, Flown by an ex student of mine:

DB... can you tell me the point of this illustration?

Maybe that at least one of our Brethren has survived a similar situation. Its the technique he employed that is certainly of interest to me.

[[email protected]]

And it is the technique I would try to employ in the same situation.

[ShyTorque]

Extract from report into AS355 TR Control Failure over Cardiff G-SAEW, Flown by an ex student of mine:

He used to post here as "Roofus".

[Apate]

And your point being what exactly?

How was he your "ex-student?"

We're not turning into Walter Mitty are we?

[DOUBLE BOGEY]

How was he your "ex-student?"

We're not turning into Walter Mitty are we?

Apate, As I recall I facilitated his AS355 Line Training. I could be wrong. Apologies if I used the word "Student" which I guess is reserved more for ab-initio.

I am certainly not claiming any Kudos from his outstanding act of airmanship. However, fwhy is this such an issue for you?

[Apate]

Apate, As I recall I facilitated his AS355 Line Training. I could be wrong. Apologies if I used the word "Student" which I guess is reserved more for ab-initio.

I am certainly not claiming any Kudos from his outstanding act of airmanship. However, fwhy is this such an issue for you?

Simples, you made a statement that I know not to be true.

You recollection regarding his AS355 Line Training is also incorrect!

[DOUBLE BOGEY]

Simples, you made a statement that I know not to be true.

You recollection regarding his AS355 Line Training is also incorrect!

I think you are mistaken. As I recall 2 pilots came from CHC/Brintel to Police Ops with Veritiair as somewhat of an experiment to convince good P2s to remain in the Company group by giving them something interesting to do.. I completed the Line Training for both in Cardiff. I am almost 100% sure he was one of the pilots. Unless of course I am having a fantasy moment!!! Always possible at my age. I sent you a PM.

[industry insider]

Overdrive

SASless - no magic! I think the thread looked to be going in a "Left unattended/not properly pre-flighted" direction, due to some of the stated (but not confirmed) timings.


Not at all, it would have made no difference here as any issues were probably not visible. It just seemed a short time to me but I am comparing it with larger types.

[John R81]

Uplinker.

Things that need to be checked regularly are either on the outside (control linkages to the tail rotor, the blades themselves, as examples) or can be inspected either through a specific hatch or other access (for my 120, opening the rear cargo hatch allows me to stand inside the tail and check the battery and look down the inside of the tail boom, for example). Typically the tail rotor drive shaft itself does not require daily inspection and so it (and any bearings along its length) are not made easily accessible. They are, however, subject to specific inspection regimes defined by the manufacturer and performed by mechanics. Where regular inspection is needed (some Rotorway designs have a belt-drive) inspection is facilitated.

There will be many flight-critical parts of a flixed-wing that are similarly treated (looking for corrosion on wing spars is not facilitated by a myriad of little hatches that you can open in a walk-around).

I have not seen a "twin tail rotor" design, and I am not imaginative enough to speculate how that might work; sorry

[Uplinker]

Thanks for your reply. I think I would want to be able to inspect all the bearings and CV couplings along the drive shaft to the tail rotor.

Take your point about wing spars, but corrosion is a slow process, amply covered during C and D checks, and the wing spars do not have moving parts requiring lubrication because they do not rotate during flight! The couplings and bearings driving a tail rotor do, and can potentially break up quickly. I think I would want to check there was no visual evidence of bearing or joint stress, given how critical the tail rotor seems to be.

Twin tail rotors would simply be two shafts running down the tail boom, each driving a tail rotor, so there would be one rotor on each side of the tail boom instead of just one on one side.

[MurphyWasRight]

As a fixed wing pilot, can I ask; how do helicopter pilots inspect the tail rotor and its drive mechanism on the walk around? Are there inspection doors along the tail boom to enable inspection of every shaft joint? I don’t recall seeing any in the helis we used to use for TV work. (Bolkow 105, Augusta 109, Twin Squirrel).

Given that the tail rotor seems to be so critical, why is there only one?. Would it not be safer if there were two separately driven tail rotors, or would that be overkill?
Most likely dual tail rotors would result in an overall reduction in safety given that some tail rotor failures such as a departed blade can cause other damage. The complexity and added weight would also have a detrimental effect. One comparison are the chances of surviving a (single) engine failure at takeoff in single and dual engine light aircraft, I have heard they are roughly the same.

IF the cause of this accident was a servo loop failure/run to a stop it is not clear that a second tail rotor would have helped, especially given the extremely limited time to take action.

Question for those who know: Would an immediate shutdown of both engines have stopped the tail rotor or is the transmission coupling such that it would continue to spin as long as the main rotor was spinning?

[chopjock]

Twin tail rotors would simply be two shafts running down the tail boom, each driving a tail rotor, so there would be one rotor on each side of the tail boom instead of just one on one side.

And contra rotating, it's not difficult to do, I designed and built one for a ROV. It worked great when one of the gearboxes failed...

[Langball]

Thanks for your reply. I think I would want to be able to inspect all the bearings and CV couplings along the drive shaft to the tail rotor.

Take your point about wing spars, but corrosion is a slow process, amply covered during C and D checks, and the wing spars do not have moving parts requiring lubrication because they do not rotate during flight! The couplings and bearings driving a tail rotor do, and can potentially break up quickly. I think I would want to check there was no visual evidence of bearing or joint stress, given how critical the tail rotor seems to be.

Twin tail rotors would simply be two shafts running down the tail boom, each driving a tail rotor, so there would be one rotor on each side of the tail boom instead of just one on one side.
Example of why it is important to inspect the tail rotor drive shaft : http://www.aaiu.ie/sites/default/files/report-attachments/11816-2009018_EISBM_DUNSHAUGHLIN-0_0.PDF

[ShyTorque]

Question for those who know: Would an immediate shutdown of both engines have stopped the tail rotor or is the transmission coupling such that it would continue to spin as long as the main rotor was spinning?







The tail rotor shaft is geared directly to the main rotor transmission, not to the engines.

[industry insider]

Does the AW169 have a any HUMS or condition monitoring system?

[Non-PC Plod]

Yes it does

[CloudHound]

From the CAA website Tail rotor inspection (http://publicapps.caa.co.uk/docs/33/20181119EASAAD20180250E.pdf)

Specifically ATA 64 – Tail Rotor – Tail Rotor Flight Control System – Inspection
Since that AD was issued, EASA decided to require an inspection of the TR duplex bearing, as additional precautionary measure.

[arizona]

When reading AAIB Bulletin S/1 2018 a few things catch my attention:

1). According to the bulletin the general wind direction was north-westerly with a strength of 10 to 12 kt at the surface and around 25 to 30 kt at 1000 ft.
2). The helicopter began a climb on a rearward flight path while maintaining a northerly heading.
3). The climb then paused. (somewhere above 320 ft)
4). Then the helicopter entered an increasing right yaw contrary to the pilot’s left pedal command
6). The helicopter reached a radio height of approximately 430 ft before descending with a high rotation rate.

— The climb out procedure was performed in downwind, (the heading was northerly, but the climb was rearward) so maybe the helicopter was following the same airmass with zero TAS so to speak?.. Somewhere above 320 ft the climb paused. Why?.. Did the helicopter reach a condition that sometimes is referred as ”settling with power” / climb in its own downwash? No descend was reported… If so, did the pilot manage to escape the situation with an unusual procedure by applying max power/max collective?-- The helicopter entered an increasing right yaw and reached 430 ft radio height before going down with high collective and power applied?

[Exo.]

3). The climb then paused. (somewhere above 320 ft)
4). Then the helicopter entered an increasing right yaw contrary to the pilot’s left pedal command
6). The helicopter reached a radio height of approximately 430 ft before descending with a high rotation rate.

The wind direction for the take off couldn't have been much better, with it fore port side, it mitigates the likelihood of reduced tail rotor effectiveness while maximising forward wind component for a forced landing prior to tdp.

The report doesn't indicate the point at which the climb paused. It references undercarriage retraction at 320', and that maximum rad alt height was 430'. Aside from there being a pause, and that left pedal was applied to no effect, somewhere in the mix, we really know no more.

[John R81]

21 November, EASA issued AD No. 2018-0252-E, which now supersedes AD 2018-0250-E (of 19 November). AW169 & 189 (similar TR flight control systems).

"The incorrect installation of the TR servo-actuator, if not detected and corrected, depending on the flight condition, could possibly result in loss of control of the helicopter." Since the initial ASB from Leonardo, there has been further instruction to look at the duplex bearing.

Still an interim measure, but alters the inspection of the duplex bearing and servo-actuator within 5 flight hrs or 24 hrs.

1. Inspect TR duplex bearing in accordance with the accomplishment instructions - Part I of the ASB
2. Accomplish a breakaway torque check of the TR Duplex bearing in accordance with the accomplishment instructions - Part II of the ASB
3. Accomplish an inspection and reinstallation of the TR servo actuator castellated nut in accordance with the accomplishment instructions - Part III of the ASB

(CAA circulated 22 November, which I read this morning).

[GrayHorizonsHeli]

Where is the duplex bearing in this assembly?

[atakacs]

I'd say that the plot thickens...

Can we assume that it was still the factory issued one on this aircraft?

[RVDT]

Where is the duplex bearing in this assembly?

Ignore the fact that this is actually a 365 but the principal is near enough -

Item 16 is the Duplex bearing and Item 14 is where the nut is roughly.

https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/577x519/as365_fenestron_303fa550efba051649a5d1c6fed4507ca620d1bd.jpg

I will let you work out the rest for what happens if the bearing fails and what could possibly happen to the nut.

[helimutt]

Rather than concentrating too much on the rad alt height, couldn’t this 100’ anomaly be discounted due to the terrain below radalt changing rapidly due to the stadium contour below? Bar alt would give steady climb then level off if indeed that was the case.

[[email protected]]

14 is where the pitch change arm and the TR servo attachment is on a 365.

Gray - what they call a duplex bearing is what allows you to superimpose horizontal pitch change movement onto the rotating drive shaft - you probably already knew that but I though I'd clarify just in case.

[ShyTorque]

Crab,

My understanding of a "Duplex" bearing is that it's actually two ball (or roller) bearings sitting side by side in one case. They are used in various applications such as vehicle wheel bearings. They provide an alternative to having two separate bearings with a space between them and are used where lateral "rocking" or "run-out" needs to be minimised and also provide load sharing between the two races. They need to be carefully installed to avoid damage, more so than simple ball or roller bearings. If they are knocked in too hard (such as to overcome too tight an interference fit), they can suffer damage.

Are you sure you're not mixing this up with what is known as a "linear" bearing? A car gear change assembly I rebuilt had a small linear bearing in the housing. It allows the gear linkage's horizontal shaft to slide fore and aft (as when going from 1st to 2nd or 3rd to 4th) and rotate from side to side, as when going from 2nd to 3rd. These have rows of small ball bearings sitting in longitudinal slots in a tubular, centre cage which holds them in alignment. I ditched the rusted out one in my gear change and because I had great difficulty in finding a suitable replacement I fitted a plain, sintered bronze bush instead.

[dClbydalpha]

The bearing is to allow rotation of the hub around the static control rod. It's a duplex bearing because it has two races precisely matched to maintain alignment when the shaft is moved horizontally.

In my personal opinion it looks like the inspections are checking both ends of the control rod for anything that may cause it to jam or break.

[gevans35]

The bearing is to allow rotation of the hub around the static control rod. It's a duplex bearing because it has two races precisely matched to maintain alignment when the shaft is moved horizontally.

In my personal opinion it looks like the inspections are checking both ends of the control rod for anything that may cause it to jam or break.

I would hazard a guess that there was a major mechanical failure that caused severe damage to many components and they are looking for the initial cause. Were there not eye witness reports of "gear crunching" noises?

[GrayHorizonsHeli]

14 is where the pitch change arm and the TR servo attachment is on a 365.

Gray - what they call a duplex bearing is what allows you to superimpose horizontal pitch change movement onto the rotating drive shaft - you probably already knew that but I though I'd clarify just in case.

I'm well aware of the function yes. I've overhauled alot of 120 and 130 TGB's.
Its likely a similiar position to it enough that I can relate it to the 169 at least.
They use a double row ball bearing that seems undersized for it's function. Only seen a couple failures though, but nothing catastrophic.
I just wasnt sure where it was located and how it was connected to the nut issue previously.

I am not familiar with the 169 directions of rotation, or left/right hand threads for that nut. but can anyone say for certainty what could happen if the bearing seizes? will it rotate that shaft and un-thread the nut with enough force to overcome the torque, and safety locks?

[RVDT]

GHH,

I am not familiar with the 169 directions of rotation, or left/right hand threads for that nut. but can anyone say for certainty what could happen if the bearing seizes? will it rotate that shaft and un-thread the nut with enough force to overcome the torque, and safety locks?

If you look back through the message thread there are a few pics of the arrangement. The locking in the pic implies a RH threaded nut and the directions of rotation would imply your theory could be correct.

Normally the servo piston is keyed in some way to stop this happening but you can't tell from the IPC drawing. Having a split/cotter pin and locking kind of implies that this is the only locking. If it was keyed you wouldn't go to that length.
Similar things have happened before but not through design issues.

Once the feedback link is disconnected the servo would invariably motor to the end of full travel and stay there.

Pic at #682 and IPC at #630

[dClbydalpha]

I am not familiar with the 169 directions of rotation, or left/right hand threads for that nut. but can anyone say for certainty what could happen if the bearing seizes? will it rotate that shaft and un-thread the nut with enough force to overcome the torque, and safety locks?

Surely the issue is that if the bearing seizes then your tail rotor pitch control is now jammed. Whereas if it is just the nut coming off then it means that control is lost.

[RVDT]

Surely the issue is that if the bearing seizes then your tail rotor pitch control is now jammed. Whereas if it is just the nut coming off then it means that control is lost.

The bearing is seized in the rotational sense as it isolates the static and rotational part of the control. Then the rotational torque is transferred to the static part of the shaft and could unwind the nut as the torque is now applied
to the static shaft.

The nut connects the static pitch change to the feedback lever. The feedback lever cancels the pilot valve input once the piston reaches the desired position. Standard servo actuator behaviour.

Without feedback the pilot valve input will not cancel.

[dClbydalpha]

The bearing is seized in the rotational sense as it isolates the static and rotational part of the control.


I may have misunderstood, but doesn't the duplex bearing also help transfer the horizontal load from the static to the rotating part?

If the bearing fails then a number of possibilities then exist. Losing the nut at either end is one, but a misalignment of the static rod leading to it jamming is another.

The servo may run hard over, or may run to neutral depending on how the feedback control is connected. I don't have any details of the servo to be able to say with certainty.

[nodrama]

Tired of listening to 'I'm not familiar with the AW169' and the utter rubbish. The duplex bearing is at the end of the grey bit, at the output end of the red bit. Knock yourselves out guys -
https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/1059x632/capture_7051ca1c3ffbc0aa67f09429ca6a6ade4f433990.png
https://cimg4.ibsrv.net/gimg/pprune.org-vbulletin/690x669/2_d36046ae82a17c737ed222af8452b940e6d53214.png
https://cimg5.ibsrv.net/gimg/pprune.org-vbulletin/907x671/3_efd50ffa059bac562751f64f591734ad440098c0.png

[DOUBLE BOGEY]

Tired of listening to 'I'm not familiar with the AW169' and the utter rubbish.

NODRAMA - For a man with that handle you certainly like to make a "Drama" out of other peoples honest statements.

[GrayHorizonsHeli]

Tired of listening to 'I'm not familiar with the AW169' and the utter rubbish. The duplex bearing is at the end of the grey bit, at the output end of the red bit. Knock yourselves out guys -
https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/1059x632/capture_7051ca1c3ffbc0aa67f09429ca6a6ade4f433990.png


regardless of experience on type and you're oh so helpful "red bit and grey bit" descriptions, I'm sorry my experience isnt helpful to you in anyway and I'm only triggering you.

[noooby]

NODRAMA - For a man with that handle you certainly like to make a "Drama" out of other peoples honest statements.

Sorry but I'm with NODRAMA. Posting a pic of the 365 system, which is about as different as you can get from the 169 and labelling the duplex bearing as item16 is not helpful at all.

Looking at your 365 diagram, number 21 is about the closest you could get to where the duplex bearing is.The duplex bearing is outboard of the tailrotor and is nowhere near the hydraulic part of the servo.

Thank you for posting accurate diagrams NODRAMA.

[nodrama]

Where is the duplex bearing in this assembly?
It’s at the outboard end of the red bit 😉 inside the pitch change spider. That’s all folks.

[RVDT]

For those that can read English.

Ignore the fact that this is actually a 365 but the principal is near enough -

Where it is physically is irrelevant - it performs that same function.

The issue that is in the debate is the consequences of a failure of the duplex bearing.

Thanks for the pics nodrama.

As you seem to be conversant with the type can you inform us as to how the pitch change shaft is locked against rotation at the servo end apart from the clamp up and a split pin and a bit of lockwire .

[tigerfish]

Why all this pointless armchair speculation/discussion? The only authoritative body here is the AAIB. Why not wait for their report! I suspect there are many who are seeking to say
"I told you so!"

TF

[GrayHorizonsHeli]

Hardly believe it's pointless.
if you fly one or fix one, its proactive discussion.
but if you would rather wait a year or two be my guest...no one is forcing you to participate in the discussion.

[atakacs]

Agreed
It is actually one of the main interests of this forum IMHO. Sometimes it certainly gets irritating but on average I find these discussions usefull.

[Gustosomerset]

Why all this pointless armchair speculation/discussion? TF
Because this forum is for armchair speculation/discussion. Whether or not it's all pointless is a matter of opinion...
As for, "I told you so"...here's my theory. Somehow the aircraft was released from the factory in 2016 with an 'incorrect installation of the TR servo-actuator'. In practice, probably meaning that the cotter pin and lock-wire were never fitted, allowing the nut to work its way loose over time until the final (as it turned out) pedal input allowed it to fall off, resulting in complete loss of tail rotor control.
This is entirely pointless (but, I believe, harmless) speculation from my armchair based, almost entirely, on the preceding 42 pages of discussion.

[sycamore]

If you look at p11,#216,pic of tail rotor,it appears that the t/r `spider is intact,and appears to have a `device` on the outside of the `spider`...perhaps `nodrama` can comment if this appears normal..? nut /bolt still attached...?

[rotorcraig]

If you look at p11,#216 (https://www.pprune.org/rotorheads/614822-helicopter-down-outside-leicester-city-football-club-11.html#post10295925),pic of tail rotor
Link is here (https://www.pprune.org/rotorheads/614822-helicopter-down-outside-leicester-city-football-club-11.html#post10295925)

[nodrama]

In my opinion, all that one can tell from that photo is that there has been a catastrophic, and very sad, helicopter crash. That's what the AAIB are for.
That 'device' is nothing more than a cover, or cap, for where the duplex bearing is housed.

[sycamore]

nodrama, thanks for clarification...

[GrayHorizonsHeli]

How robust is this spider and duplex bearing?
separate component or integral with the TGB?
what's its overhaul life and inspection schedule?
greased bearing or oil supplied from TGB?
c'mon nodrama, if you're experienced, share and educate.

[arizona]

The nice thing, with a tail rotor failure in hoover or low speed is that you DIRECTLY and FIRST OF ALL goes for an autorotation by lowering the collective all the way to the lower stop. You don’t need to think if a cotter pin, a nut or all the tail rotor blades are missing. You -bang!- go for autorotation. If this accident was a tail rotor drive failure you had to land immediately, even if in trees, but here you had a few options like open areas where you probably could make a survivable ”landing".

Look at tailrotor failures like this: If you hesitate to go in autorotation you are toasted, but if you go in to a autorotation and it shows wrong, you can always change your mind. To survive a helicopter, always think negative for a positive outcome.

The different types of tail rotor failures and necessary actions taken are well known and are described in all helicopter manuals

[ShyTorque]

Arizona, your answer is too simplistic and unfortunately doesn't take into account that not all tail rotor malfunctions involve a loss of drive/thrust.
In the situation where the tail rotor continues to be driven but pitch control is lost and the tail rotor goes to maximum positive pitch, entering auto-rotation might be the last control input you ever made.

[Lonewolf_50]

Shy, I am reminded of an old aphorism (maybe from Einstein) that a great many complex problems have simple, clear, and wrong answers. (ref to what you responded to) :cool: I'm with you; an incorrect response to a particular malfunction can make things worser faster.

[Echo Romeo]

Blimey, this is still rumbling on.......

[hihover]

Crab, simply as a matter of interest and as I have easy access to a Level D 139 FFS, I had a go yesterday at using the technique discussed by you and DB. Even in the sim, and I knew it was coming, I very quickly became disorientated. I even found it difficult to take my hand off the collective to kill the engine mode switches. I felt I had no control whatsoever of the helicopter and was just along for the ride. No real thanks to me, the sim helicopter did remain level and I eventually put the collective down quickly. I got away with a successful level, spinning landing but I was very surprised at how quickly I lost all references. I carried out this exercise from 20 ft. From 400 feet I have absolutely no doubt it would be impossible. IMHO and as we seem to all agree, those poor guys didn't stand a chance.

[DOUBLE BOGEY]

Crab, simply as a matter of interest and as I have easy access to a Level D 139 FFS, I had a go yesterday at using the technique discussed by you and DB. Even in the sim, and I knew it was coming, I very quickly became disorientated. I even found it difficult to take my hand off the collective to kill the engine mode switches. I felt I had no control whatsoever of the helicopter and was just along for the ride. No real thanks to me, the sim helicopter did remain level and I eventually put the collective down quickly. I got away with a successful level, spinning landing but I was very surprised at how quickly I lost all references. I carried out this exercise from 20 ft. From 400 feet I have absolutely no doubt it would be impossible. IMHO and as we seem to all agree, those poor guys didn't stand a chance.

Hi-Hover, take a look at the Cardiff Police TR Malfunction. In the report the Commander stated he got totally confused by the image outside the windscreen so he settled on looking out the side window. This seemed to provide enough references for him to lower the collective, power on, and accept a trade of between ROD and yaw rate. He states that as he saw the surface appearing he pulled the collective all the way up to cushion the impact. Power on throughout. However, I am not sure if the references in any SIM would facilitate this or indeed of the flight loop allows for a reasonable attempt. Different type (355) also.

[hihover]

Sure, I agree, simulation has limitations, it was a different type from your student, and not even a 169. My point is simply that, after 40 years and 14,000 hours in the game, and being ready for the exercise, and having been in a helicopter with a tail rotor failure, I found it quite impossible to accept the rate of yaw with any form of control. I would need much more spare capacity.

[Gustosomerset]

Meanwhile...back to pointless speculation. Nodrama states that 'The aircraft that is the subject of this thread had flown less than 400 hours.'. Apologies if this has been stated elsewhere and I've missed it but what maintenance inspections would it therefore have been subject to since being built in 2016? - and would any of these inspections specify detailed examination of the tail rotor parts that are the subject of the AD notices as discussed above?

[[email protected]]

HiHover - I didn't say it would work but compared to a zero speed auto to EOL from 400' it might be the lesser of two evils:ok:

Perhaps the entering auto but keeping the engines running and using them to cushion the touchdown - if you can judge it - might be a refinement of the idea.

Either way I am in no rush to try it out!

[hihover]

You could be right mate. I was just astounded at how little control I had in trying to keep her level and reduce the yaw rate.

[SASless]

The beauty of the Sim is you can try it over and over and crash repeatedly....and at some point call it a day and head for a Pint on wobbly legs!

the IAF lost a Bell 212 (or 412 maybe) after a Pilot had picked up way too heavy a load on the Hoist....and came to our Facility to see if they could replicate the event.

I played with the Lateral CG Parameters to arrive at their calculated data and after about a dozen or so attempts we actually got to where we could fly out of the situation by using some very un-unorthodox control techniques which were quite counter-intuitive.

A large police department aviation unit came to us and wanted to practice autorotations and tail rotor failures at various HOGE heights....and we found that to be very interesting as well.

Whether the actual aircraft would react exactly as the Sim (or conversely....) was not the point of the exercises but determining what actions improved the situation was.

One thing for sure....we could prove certain actions were very bad juju and immediately made certain situations absolutely uncontrollable.

Get back in your box of tricks and experiment.....just do not activate your "Crash Over-Ride" as even a Sim can hurt you if left to its own.

[hihover]

:) No thanks SAS, I’ve seen enough. Unfortunately, when faced with an issue that requires immediate action, we revert to basic mode and we generally deliver whatever our survival instincts think we should. I can’t think of a worse position to suffer a catastrophic tail rotor issue.

[chopjock]

hihover
I can’t think of a worse position to suffer a catastrophic tail rotor issue.
And they were put in that position by the regulations in the name of safety...

[DOUBLE BOGEY]

:) No thanks SAS, I’ve seen enough. Unfortunately, when faced with an issue that requires immediate action, we revert to basic mode and we generally deliver whatever our survival instincts think we should. I can’t think of a worse position to suffer a catastrophic tail rotor issue.

HiHover, what is your view of the Landing Gear being raised prior to the TDP?

[212man]

HiHover, what is your view of the Landing Gear being raised prior to the TDP?

How do we know that it was?

[Non-PC Plod]

hihover

And they were put in that position by the regulations in the name of safety...

Hmm.... Be interesting to know what regulation you think would ensure a safe profile to get out of a football stadium with a tail rotor failure! Somehow ,I dont think you can avoid exposure to a low speed/high torque environment.

[SASless]

Hihover......if you have free access to a Sim....and care not to utilize its great potential for learning and exploration of how a helicopter reacts to unusual events....please understand I find that attitude discouraging.

You have a golden opportunity and seem to be throwing it away.

Take the Sim out of motion....and run the scenario that way....so you only have the Visual and Instruments operating while the Sim sets happily on the blocks.

[DOUBLE BOGEY]

How do we know that it was?

212Man from AAIB Special Bulletin:

The helicopter then began a climb on a rearward flight path2 while maintaining a northerly heading. Gear retraction started as it passed through a height of approximately 320 ft. The climb then paused. Heading changes consistent with the direction of pedal movements were recorded initially, then the helicopter entered an increasing right yaw contrary to the pilot’s left pedal command. The helicopter reached a radio height3 of approximately 430 ft before descending with a high rotation rate.

[hihover]

SAS....:):):) I have a job. Please feel free to be discouraged by me maintaining my employment. :):) When I have time I will have another look.

[212man]

212Man from AAIB Special Bulletin:

The helicopter then began a climb on a rearward flight path2 while maintaining a northerly heading. Gear retraction started as it passed through a height of approximately 320 ft. The climb then paused. Heading changes consistent with the direction of pedal movements were recorded initially, then the helicopter entered an increasing right yaw contrary to the pilot’s left pedal command. The helicopter reached a radio height3 of approximately 430 ft before descending with a high rotation rate.

Yes, I've read the report (in fact it was me that posted the link here first). But, where in that paragraph does it state what the TDP height was?

Earlier a poster, who would appear to be familiar with the type, posted this:

The TDP is 115ft + the height of the obstacle in your takeoff path so I’d guess a minimum of 250-300ft.

[hihover]

Chop, take-off profiles are designed to give us the best chance when dealing with an engine failure, and they actually work very well if you stick to the profile when you lose an engine. Sadly, it would be impossible to write a similar profile for the catastrophic tail rotor event because control is effectively lost. It is not lost when an engine fails.

DB. Not sure why the gear was retracted, I haven't flown the 169 and know very little about it. In truth I don't believe it would have made any difference. Clearly, if he had suffered an engine failure before TDP he would have had to be quite slick with the gear but from that height I'm sure he felt he had plenty time to put it down again if required. Personally, I raise the gear when it is of no further use to me (rather than 200 feet on climb out). I normally operate from a 10,000 ft taxiway so I have the option to reject for quite a while.

[DOUBLE BOGEY]

Yes, I've read the report (in fact it was me that posted the link here first). But, where in that paragraph does it state what the TDP height was?

212Man, I am assuming that as forward flight transition was not attempted prior to the Gear being raised that his calculated TDP had not been reached or indeed missed as events transposed.

I think though, that the salient point is should retractable landing gear be raised prior to achieving a sensible amount of forward airspeed and height. Obviously prior to TDP the gear should of course remain available for a reject. In big helicopter world, as you know, we generally do this at Vy+200 feet. More to avoid distraction between VTOSS and Vy and often to conform to a critical performance profile.

However, if we accept that the landing gear is able to absorb some considerable energy in a high ROD touchdown, then maybe the Vy+200 feet (both conditions having to be met), is a good compromise should the RFM not specify for the operating conditions.

I guess what I am really alluding to is would the outcome in this instance have been improved if the gear remained down to absorb the initial energy at impact.

[212man]

I guess what I am really alluding to is would the outcome in this instance have been improved if the gear remained down to absorb the initial energy at impact

Yes, I agree with you that even if they had reached TDP, the logic behind early retraction is not clear, although another poster suggested it has a low gear limit airspeed (which I queried and can only assume is related to it being electrically actuated, not hydraulically). I also agree that the outcome of a heavy landing on the undercarriage is likely to have been different to what actually happened. However, when you look at the Kenya Police AW139 accident where they all survived from a similar rate of descent - also with the gear retracted in a hover - and the relatively low level of obvious structural deformation in the early photos taken before the fire really took hold, in this incident - I shudder to think what was going on inside on the ground.

[chopjock]

Hmm.... Be interesting to know what regulation you think would ensure a safe profile to get out of a football stadium with a tail rotor failure! Somehow ,I dont think you can avoid exposure to a low speed/high torque environment.

You are obviously correct, however the profile used was designed to mitigate a one of two engines failing, but increases the exposure time on the absolute dependance of the one and only tail rotor. I can't help but think a diagonal forward acceleration from the far downwind corner and a zoom climb at the upwind corner would have resulted in less exposure time and less stress to the tail rotor. This obviously puts more reliance on the engines, but they are very reliable these days and there are two of them!

[DOUBLE BOGEY]

I can't help but think a diagonal forward acceleration from the far downwind corner and a zoom climb at the upwind corner would have resulted in less exposure time and less stress to the tail rotor. This obviously puts more reliance on the engines, but they are very reliable these days and there are two of them!

Chopjock, the profile you propose would require reject distance available if the helicopter is above the OEI-IGE Hover Mass. You would also need to prove you could clear the stadium OEI after TDP on the remaining engine. Its really difficult to make this work when the distance available are so short. That's why the Rearwards Profile (VTOL Helipad) was conceived. To overcome these issues.
Having said that, the fact that this, and other TR malfunctions have occurred leading to loss of the aircraft and sometimes occupants, at low speeds means you are making an argument. How does the likelihood of a TR event compare to an OEI before TDP?

[ShyTorque]

Chopjock,

As always you think you know better than those who actually do the job. However, most of us tend to operate the aircraft in accordance with the manufacturer's certificated performance procedures, even those of us not operating for public transport (where it is usually mandatory). It is sometimes a condition of the aircraft insurer (and therefore a requirement / condition of maintaining one's employment status with the operator) and it is always a condition of a CAA congested area written permission.

[TeeS]

Yes, I agree with you that even if they had reached TDP, the logic behind early retraction is not clear

Hi 212man, it is a long time since I flew retractable undercarriage aircraft but I always thought/assumed that the gear was retracted as soon as possible after TDP in order to minimise drag during the acceleration to Vtoss/Vy, I do wonder whether this was another hang over from fixed wing A/C where the target speeds are higher and so the drag had a bigger influence. Obviously those operating single pilot might feel there would be more important considerations during a busy stage of flight!

Cheers TeeS

[chopjock]

Chopjock,

As always you think you know better than those who actually do the job. However, most of us tend to operate the aircraft in accordance with the manufacturer's certificated performance procedures, even those of us not operating for public transport (where it is usually mandatory). It is sometimes a condition of the aircraft insurer (and therefore a requirement / condition of maintaining one's employment status with the operator) and it is always a condition of a CAA congested area written permission.

Just like what I said earlier...
And they were put in that position by the regulations in the name of safety...

[JimL]

Chopjock, you are the ultimate laxative.

You continue to put forward your ridiculous views in the face of logic and weight of informed opinion against you. It would not surprise me if you were the twin (or alter ego) of the other fool on this board.

Jim

[SASless]

Now there is a thought......ONLY two Fools in this Forum!:D

[DOUBLE BOGEY]

Now there is a thought......ONLY two Fools in this Forum!:D

Don't forget me!

[chopjock]

Chopjock, the profile you propose would require reject distance available if the helicopter is above the OEI-IGE Hover Mass. You would also need to prove you could clear the stadium OEI after TDP on the remaining engine. Its really difficult to make this work when the distance available are so short. That's why the Rearwards Profile (VTOL Helipad) was conceived. To overcome these issues.


Except no consideration at all about longer exposure time and total dependance on the tail rotor...

[212man]

Hi 212man, it is a long time since I flew retractable undercarriage aircraft but I always thought/assumed that the gear was retracted as soon as possible after TDP in order to minimise drag during the acceleration to Vtoss/Vy, I do wonder whether this was another hang over from fixed wing A/C where the target speeds are higher and so the drag had a bigger influence. Obviously those operating single pilot might feel there would be more important considerations during a busy stage of flight!

Cheers TeeS

TeeS, no the certification and so hence RFM profiles assume that the gear is left down until Vy and 200 ft agl. for a conventional accelerating departure as it is assumed that the pilot/crew are too busy concentrating on achieving the first segment climb at Vtoss to be operating any other controls. For a procedure where the TDP is already above 200 ft I would assume that retraction would take place at Vy.

and less stress to the tail rotor

Chopjock - what stress are you talking about? This is a tail rotor on a Part 29/CS29 certified modern machine, not something on a balsa wood aeroplane with a wind up propeller. Do you seriously think that a gentle climbing imanoeuvre is somehow placing any components in jeopardy?

[TeeS]

Hi Chopjock, so you have backed your helicopter into the corner, accelerated as fast as possible towards the opposite corner (135m away according to Google Earth and that might just allow you to get to 25-30kts before you hit the far stands) and then 'zoom climbed' at a moment when you guess that your angle of climb will just avoid the stadium roof. Thankfully your one engine (one of my two engines) hasn't stopped because the engines are rather reliable, the climb continues until you get to the top of the stadium roof whereupon you continue an acceleration/climb to a suitable height. If at any time an engine stops (except in the first few seconds of acceleration or on reaching a safe height), the tail rotor quits its job, the rotor head lets go or you just overestimate your guess at what climb angle you were going to achieve, you stand a good chance of killing your passengers. How on earth can you argue that you have reduced any exposure times!!
Cheers
TeeS

[DOUBLE BOGEY]

Chopjock, don't think that the exposure to TR failure poor outcomes is restricted to zero speed hovering. TR Malfunction risks are present throughout the envelope and decrease significantly with speeds above Vy, However, in a twin, TR drive failure anywhere in the envelope turns you effectively into a SEH as an autorotation is generally required. Unless you have a Fenestron behind you whereby your options are significantly increased.

To illustrate my rather awkward point. If we all agreed to depart in accordance with your suggestion (Lets call it the "Corner" profile), losing the TR control or drive anytime in the first 70 knots would see you in a smoking heap. Up to 100 kts in the climb out your arse would be eating the seat cushion and the prospects of survival still 50/50. There are no easy alternatives.

That's why others on this thread (and you have managed to stimulate JimL into a frenzy), think you somewhat lacking in the old common sense department. However, try not to get defensive and think carefully about what I have written. This is nothing new. TR Malfunctions and Drive Failures present limited prospects of success wherever they occur.

[chopjock]

TeeS
How on earth can you argue that you have reduced any exposure times!!

How long does it take to VTOSS using CatA profile V a "Corner" profile?

[TeeS]

TeeS, no the certification and so hence RFM profiles assume that the gear is left down until Vy and 200 ft agl. for a conventional accelerating departure as it is assumed that the pilot/crew are too busy concentrating on achieving the first segment climb at Vtoss to be operating any other controls. For a procedure where the TDP is already above 200 ft I would assume that retraction would take place at Vy.


Thanks for that 212man, I hadn't come across that as a certification spec - every day is a learning day :-)

[TeeS]

TeeS


How long does it take to VTOSS using CatA profile V a "Corner" profile?

Well I would be able to calculate that pretty accurately for the VTOL profile as the test pilots have produced the graphs but I can't compare it to your 'Corner' profile as that would depend on your gut feeling about when to stop your acceleration in order to miss the stadium roof and hence what speed you would reach and therefore what your rate of climb would be.

Cheers
TeeS

[Thomas coupling]

Chopjock is a drone operator.
{and has flown as a passenger in an R22 once}.
So be gentle.

[ShyTorque]

TeeS


How long does it take to VTOSS using CatA profile V a "Corner" profile?

Surely you don't think that Vtoss is in any way related to the tail rotor?

Here's an example of a pilot trying to fly your proposed technique:
https://www.youtube.com/watch?v=N6vWraEkVkY

[chopjock]

Surely you don't think that Vtoss is in any way related to the tail rotor?


I was giving a time period in which to answer a question.


Here's an example of a pilot trying to fly your proposed technique:
https://www.youtube.com/watch?v=N6vWraEkVkY

That was all too slow, nothing like "my technique". No back up, no firm acceleration and no zoom climb...
I wonder how in that video it would have turned out if that Bell had attempted a Cat A departure profile?

[ShyTorque]

If he had attempted a Cat A departure profile and had insufficient power to fly away, which is exactly what happened, he would have been able to fly safely back down to the departure point.
Maybe the point that Cat A departure profiles are designed for exactly this reason has so far escaped your understanding.

[chopjock]

https://www.youtube.com/watch?v=YEQaC95v0OM

Like this...

[TeeS]

You've got it at last Chopjock, that is a perfect demonstration of a class 1 profile (it wasn't called that then). From the hover the pilot flying accelerates the aircraft and the pilot monitoring calls 'groundspeed' when he sees the ground moving past the window and then 'airspeed' when the ASI just begins to fluctuate. At that point, the pilot flying adjusts the attitude and power to climb almost vertically whereupon the pilot monitoring calls '20 feet', '40 feet' and 'rotate' as the radalt shows 50'. At that point, the pilot flying initiates a further acceleration and is committed to fly away - before this point, in the event of an engine failure (or other emergency) the take off is rejected.

This profile allows the S61 to depart from a very short runway providing class 1 performance in the same way that the vertical profile provides. It does not allow a departure with obstructions that would be found in a stadium environment!

Cheers

TeeS

[chopjock]

Ok a Class 1 profile takes about 12 to 15 seconds of "exposure" time to Vtoss. A Cat A profile takes about 30 seconds or more. If neither can survive a tail rotor failure during this time, then which is the lesser of the two evils? I would suggest a shorter exposure time...

[SASless]

Laddy....if you think 15 seconds means squat considering the rarity of a Tail Rotor failure.....you ought to stay home in bed all day and pray a Police Helicopter doesn't join you.

[chopjock]

Laddy....if you think 15 seconds means squat considering the rarity of a Tail Rotor failure.....you ought to stay home in bed all day and pray a Police Helicopter doesn't join you.

Multiply by how often did they do this and was it after 15 seconds when this tail rotor failed?

[SASless]

JimL......There cannot be two such on the Forum.....the odds against that are just too far out there for it to be possible.:ugh:

[ShyTorque]

Ok a Class 1 profile takes about 12 to 15 seconds of "exposure" time to Vtoss. A Cat A profile takes about 30 seconds or more. If neither can survive a tail rotor failure during this time, then which is the lesser of the two evils? I would suggest a shorter exposure time...

Again, you're missing the point. The one you linked to showing a S-61 is a "clear area" profile. It needs a specified minimum distance of clear area to operate from. If there is sufficient distance available (such as existed at Penzance before Tescos built a supermarket on it), it would be flown. If not, from a shorter and confined area surrounded by high obstacles (such as Leicester City ground), it can't be flown because the required obstacle clearance or stopping distance can't be guaranteed and a landing following any loss of power would almost certainly end up in the stands.

[chopjock]

Again, you're missing the point. The one you linked to showing a S-61 is a "clear area" profile. It needs a specified minimum distance of clear area to operate from. If there is sufficient distance available (such as existed at Penzance before Tescos built a supermarket on it), it would be flown. If not, from a shorter and confined area surrounded by high obstacles (such as Leicester City ground), it can't be flown because the required obstacle clearance or stopping distance can't be guaranteed and a landing following any loss of power would almost certainly end up in the stands.

I understand what you are saying. The linked S61 video shows a large helicopter in a large (ish) space with a clear reject area, so scale it down to a smaller helicopter in a smaller space and a smaller reject area. Risk flaring into the stands in 15 seconds of exposure v spinning out of control at 400 feet with no airspeed and 30 seconds + of exposure?

[SASless]

Care to explain why the size of your chopper is improved by a lack of its length?

[industry insider]

Not really relevant but I think the specific S-61 profile shown was called an oblique in Bristow days? I think we operated Cat A or Cat A with dump on the 61 providing we could dump down to Cat A weight with the contents of the centre tank.

[horizon flyer]

On reading the posts on gear retraction and minimum forward airspeed for aerodynamic control why is gear retract based on height should it not be on airspeed i.e. minimum airspeed for control if TR failure then it's down to help cushion the landing.

I believe autorotates are difficult to judge especially at night how about a belly air bag, triggered by under carriage compression rate or a G sensor and why has no one thought of anyway of adding anti torque at low speed if TR failure. Maybe a smallish parachute fired sideways to slow the spin triggered by a yaw rate and time sensor. Just a bit of brain storming.

[bront]

Chop, more people have crashed doing your corner departure (because it relies on a performance guess rather than tested data) than crashes from TR failures doing Cat A take offs.

[DOUBLE BOGEY]

Chopjock is a drone operator.
{and has flown as a passenger in an R22 once}.
So be gentle.

Ahh! now it make sense. TC you must be lovin the last bunch of posts. We all on Chopjocks little hook!

[DOUBLE BOGEY]

CHOPJOCK, I am not going to trash your ideas because sometimes, lack of doctrinal thinking and dogma can be the mother of invention. However, I sorry but in this instance you are "droning" up the wrong tree.

You see the TR is there to stop the torque applied to the rotor head, spinning the fuselage once the undercarriage friction with the ground is not longer present. To make the TR efficient, it is placed a long way down the tailboom to give some leverage. This provides the added advantage of "Keel" surface. That is to say the surface of the fuselage and tailboom combined, when pushed through the air at speed, provides an element of anti-torque reaction all on its own. However, this "Keel" surface is not really effective until the forward airspeed is well above 70-80 KIAS. Now I am not a Test Pilot so my numbers here are just a guess but you can get the idea. The more power you apply to the rotor head, the more "torque" is trying to spin you.
So armed with this new knowledge, you can understand that what you propose, scuttling across the football pitch then zooming over the stand, you will still be below the speed where any benefit from the keel surface is not present. Added to the requirement to apply lots of power, to accelerate quickly, and you can understand that a TR failure anytime during your procedure would leave you, at worse case, moving forward across the ground, spinning like a an angel with a broken wing towards the obstacles in front of you. This set of circumstances is extremely difficult to resolve.

In recognition of the critical role the TR plays, especially in the low speed, low height part of the flight envelope, the certification requirements for the TR system are very comprehensive and the design failure rate much lower than other systems that may for example, be duplicated. This is why this accident is so significant if indeed it proves to be some kind of design fault.

This problem, the heavy reliance on the TR, has plagued designers and performance systems since Sikorsky donned his pork pie hat. It is nothing new.

Many factors come into play when a TR malfunction occurs and the actions of the pilot are just one aspect. Design characteristics, Power Setting, airspeed, height, wind direction etc all have a influence on the outcome.

I understand your concern, hanging around in the low speed envelope increases the exposure to the impact of a TR failure and you are certainly correct in this assumption. However, your proposal does not solve the problem as the exposure remains critical throughout your proposed take-off envelope and marginally improves as speed (>80 KIAS guess) and height is gained.

If you are a Drone Pilot and you do it for a living, big respect. I had a drone for my Daughters Wedding this year and it and the operator were awesome.

I hope this post helps you to understand why so many are perplexed by your proposal.

[Hughes500]

TC

He is actually a helicopter pilot, having done his PC's for the past 7 years or so ( he is actually a very good handling pilot, including a successful landing with a stuffed tail rotor in an Enstrom and an engine failure in a Hu 369) He is a single engine pilot so is asking why one has to climb quite so high when in a single you wouldn't. I suppose I can understand what he is saying, but he may have a hypothetical point are the manufacturers concentrating too much on engine failures rather than the plethora of other problems.

[[email protected]]

I think TC knows that - it was a throwaway remark based on some of the ideas Chopjock has come out with.

He could at least have googled Cat A/PC1 profiles to understand them before posting......

[DOUBLE BOGEY]

TC

He is actually a helicopter pilot, having done his PC's for the past 7 years or so ( he is actually a very good handling pilot, including a successful landing with a stuffed tail rotor in an Enstrom and an engine failure in a Hu 369) He is a single engine pilot so is asking why one has to climb quite so high when in a single you wouldn't. I suppose I can understand what he is saying, but he may have a hypothetical point are the manufacturers concentrating too much on engine failures rather than the plethora of other problems.

If he has survived a TR and an Engine failure he deserves some respect. Certainly experienced more have I have in the last 35 years.

[Non-PC Plod]

Chopjock,

To reiterate what Double Bogey is saying - you seem to be hung up on the idea that getting VTOSS will somehow protect you from TR failure, and that being under this speed is somehow "exposure".
The reality is you are ALWAYS "exposed" to TR failure - you just may be better positioned if you are say above 1000' AGL with 100 kts airspeed and a clear area in front of you. Anything else, and its not going to be your day.

[chopjock]

Chopjock,

To reiterate what Double Bogey is saying - you seem to be hung up on the idea that getting VTOSS will somehow protect you from TR failure, and that being under this speed is somehow "exposure".
The reality is you are ALWAYS "exposed" to TR failure - you just may be better positioned if you are say above 1000' AGL with 100 kts airspeed and a clear area in front of you. Anything else, and its not going to be your day.

I was referring to VTOSS simply as a timing reference point for exposure comparison. I am aware if refers to single engine safety speed and will have little "keel" effect, but gaining forward airspeed sooner must be better than backing up with none for longer.

[JimL]

Vtoss is a term that is only used in Category A procedures; it is used establish the Take-Off Distance Required. The definition of TODRH is:

The horizontal distance required from the start of the take-off to the point at which VTOSS, a selected height and a positive climb gradient are achieved, following failure of the critical engine being recognized at TDP, the remaining engines operating within approved operating limits.

Note.— The selected height stated above is to be determined with reference to either:

a) the take-off surface; or

b) a level defined by the highest obstacle in the take-off distance required.

The minimum climb performance required at Vtoss is 100ft/min.

Jim

[Aucky]

I was referring to VTOSS simply as a timing reference point for exposure comparison. I am aware if refers to single engine safety speed and will have little "keel" effect, but gaining forward airspeed sooner must be better than backing up with none for longer.

So Chopjock, the pilot flies your ‘corner’ departure, just clears the stadium roof then 30seconds in at 60kts and 300ft still at high power has a tail rotor failure, he’ll be performing some form of emergency night autorotation onto dark roads and houses vs some form of vertical descent potentially into the lit stadium. Not a lot in it, both horrendous situations.

However had an engine failed instead which statistically is 1000 times more likely according to JimL’s certification stats, he would have a) (corner departure) collided with the stadium, or b) (Cat A) landed back onto the take-off spot without incident.

[SASless]

The AAIB Report tells us when the failure occurred and generally how it happened so we can roughly guess as to the rate and extent of pedal travel and time from liftoff that the Pilot lost control of Yaw.

Perhaps applying that information to Dash and Climb method that is being debated.....to get a good guess at what point the loss of yaw control would occur and thus make an assumption as to what forward speed and height the aircraft would have be at.....would be an interesting comparison to what did happen.

My bet is it would have be at a low height at a fairly low airspeed and probably. not clear of obstacles.....which would be a very bad situation to deal with.

We have to include the take off to a hover, the time required to hover back as far as possible, the time to point the aircraft to the takeoff heading, then accelerate towards the obstacles, hit some speed at or slightly above Translational Lift speed, then rotate the nose upwards, and climb until reaching a safe height above the obstacles.....then lower the nose and accelerate to Vy.....and of course all of the pedal movements that would be required in the process.

What if the yaw control failure happened while in the hover maneuvers prior to start of the take off run?

Anyone care to take a whack at that?

[Hughes500]

SAS

I think preying to whatever God tickles your fancy is the correct answer for almost ever version being discussed

[SASless]

I would like to have a go at this in a Sim....and see how different scenario's would play out.

Two key issues to me are reaction time and what the initial Pilot action was....and how the aircraft was moving and reacting to the response made.

If the Collective had been put full down almost immediately.....and the nose lowered....would there have been a different outcome (not necessarily a pretty one).

Likewise....had the landing gear been down the whole time....how would that have affected the outcome of any less than controlled impact.

Early on in this thread I asked the question.....do we cause ourselves problems with all the various profiles as compared to the days when we flew the aircraft based upon all engine performance.

That was meant to ask a question similar to what Chopjock is going on about today.

I posed the question by asking if an old fashioned Confined Area Takeoff (Towering Takeoff) would not have been better than the Profile flown by the Pilot in the accident.

I was suggesting the Stadium is just a confined area....and if we did not bind ourselves to the now required profiles....would the outcome have been better in this case.

I reckon the Towering Takeoff method would in some regards not be that much different than the Rearwards Takeoff that was performed except for going vertical and losing visual contact with your Rejected Landing area.

The height required would be less than that reached by the aircraft....but not much.

The basic fact remains....this was a very difficult failure to define quickly (as to cause of the rotation) but the Pilot response would be too reduce Collective and lower the pitch attitude to obtain some forward airspeed.

If the Tail Rotor had a full stroke input....at zero airspeed (or nearly zero), the failure would very difficult to deal with.

One thing for sure is none of us....none.....ever want to be in the situation that this Pilot was that night.

What is important now is to learn as much as possible from this tragedy.

Discussions like this one is part of that process.

We may poke fun at each other on occasions...and even take objection to some things said....but in the end discussion, debate, and even argument can be productive.

[Non-PC Plod]

SALESS,

"If the Collective had been put full down almost immediately.....and the nose lowered....would there have been a different outcome (not necessarily a pretty one)."

I think - maybe, .....but what you are asking is probably beyond any reasonable expectation of human performance. I tried what I think is the closest approximation available in the 169 simulator, which is the TR driveshaft failure in a similar position. But - knowing exactly what was going to happen and when, in the daytime, over a clear runway, I still was too slow at the first attempt, and crashed. I managed it second attempt. You would have to be superhuman to react quickly enough in the real (unexpected) situation.

[SASless]

Another method would have another pilot/person apply full pedal and hold it at full extension.....and see how that worked.

[Michael Gee]

REMINDER

Intrim AAIB report 14 Nov
Extract Ongoing investigationThe cause of the apparent loss of yaw control has yet to be determined.

[AMDEC]

The AAIB Report tells us when the failure occurred and generally how it happened

Unless I missed something the AAIB report does not tell that a failure occurred...

[pilotmike]

SAS

I think preying to whatever God tickles your fancy is the correct answer for almost ever version being discussed

Preying on anybody or anything would appear to be a waste of time. How exactly does one prey on God - an imaginary figure? Pray do tell.....

[500e]

Preying on anybody or anything would appear to be a waste of time. How exactly does one prey on God - an imaginary figure? Pray do tell.....

Sarcasm perhaps @pilotmike ?

[SASless]

Unless I missed something the AAIB report does not tell that a failure occurred...

Yes you missed something....other posters have quoted the AAIB several times.

The AAIB states it is continuing to investigate the apparent loss of tail rotor control.

They do not categorically state there WAS a Tail Rotor Control failure that caused the accident.

Combine the AAIB statement with the Video's that are available and one can easily assume a Tail Rotor malfunction of some sort occurred.

Did the AAIB not state in its Report that the aircraft was responding correctly to Yaw Inputs by the Pilot until the onset of the yawing?

Then add in to your consideration of the AD:s that were issued re the Tail Rotor Control Assembly and ask yourself why that happened very shortly after the accident.

Yes....I would say you have missed something.

[AMDEC]

...with no advanced warning, he felt the nose of the helicopter drift right. He initially corrected with left pedal; however, the nose continued to drift right.

This is very similar to what AAIB is describing, but with other words taken from one of the many investigation reports addressing unanticipated yaw events. A failure cannot be ruled out but is not mandatory to explain the accident.

[Pittsextra]

Quick question without wishing to draw any conclusion but picking up on what AMDEC has highlighted - without a physical failure/restriction would a LTE event look somewhat different that what is seen in the videos?

[[email protected]]

You would expect it to recover in the descent rather than keep yawing through several revolutions

[Non-PC Plod]

Another method would have another pilot/person apply full pedal and hold it at full extension.....and see how that worked.

Agree - it would be interesting to see if that was even worse than having zero rotor thrust -Just wondering whether you can get negative pitch on the tail rotor at extreme pedal inputs?

[ShyTorque]

Agree - it would be interesting to see if that was even worse than having zero rotor thrust -Just wondering whether you can get negative pitch on the tail rotor at extreme pedal inputs?

Yes, of course you can get negative pitch! Any helicopter has to be able to turn in both directions, both at high power and in autorotation. As I stated a long way back in this thread, the tail rotor of a Puma helicopter, as one example, can go between +35 degrees to -17 degrees of pitch.

That is why a tail rotor control failure can be far worse than a tail rotor drive failure.

[chopjock]

That is why a tail rotor control failure can be far worse than a tail rotor drive failure.

Perhaps there should be an explosive bolt or a quick release dog clutch on the tail drive shaft then...

[malabo]

None of you guys have ever tried this??? With training aircraft like the B47 we'd do lots of simulated tail-rotor failures from 500' by jamming in full right pedal, pointing out the pedal position was not going to move again until on the ground, and then slowing down until keel effect was lost and it snapped around. Let it go around a time or two (any more and the student would lose horizontal plane reference and bad things would happen) then roll throttle off to stop rotation, bottom collective to hang onto rpm, and pitch for some airspeed. From 500' you'd get about 10 mph per 100, so you'd have enough to flare and skid it on full-auto. It would veer a little to the right but not bad, at which point you'd point out to the student that the pedals hadn't moved. Most pedal goes to counter right yaw, in an auto it's almost full right. 212's weren't much different, but all this stuff used collective throttles.

Who knows whether a sim is going to replicate reality or not. You have pilots chatting with programmers on what they think will happen, nobody has hard data. And after all, the only failure we're concerned about now is an engine failure (but only if you have two), anything else is as JimL describes beyond the probability where it is worthy of much attention. On the S76 we tried jamming right pedal in a hover and having the copilot chopping throttles but it didn't work all that well with experienced trainers and there was low probability of success with a plebeian crew, hence the new world order of just lowering collective. We never did zero airspeed simulated by right pedal tail rotor failures at altitude either, too hard. So we always trained with enough speed for keel effect where the lowering of collective had some merit an gave time to bring the throttles back.

Discussions on takeoff profiles again, comparing the statistical exposure time. Jungle ops we started out verticalling some distance above the canopy before transitioning to forward flight, but gave up on that due to what felt like a lifetime of exposure for that extra hundred feet and the reality that you weren't going to find your way back down that keyhole anyway. Judgement call based on wide consideration of environmental factors, something we can't do anymore - each specific case has to be in the RFM.

In this one, I saw nothing wrong with Eric's departure profile. In the absence of any RFM or regulatory Class 1 guidance I would have flown the same. Raising the gear, I don't know if it had any significance or not. The S76 has a priority valve to deal with extra demand on hydraulics during gear retraction. The 139 deals with it by bolting on an extra hydraulic pump.

[212man]

Perhaps there should be an explosive bolt or a quick release dog clutch on the tail drive shaft then...
so you’re worried about exposure to TR problems in a Cat A profile but now you’re advocating sticking explosive bolts in the assembly? You couldn’t write this stuff!!!!

[chopjock]

so you’re worried about exposure to TR problems in a Cat A profile but now you’re advocating sticking explosive bolts in the assembly? You couldn’t write this stuff!!!!

Yes, if full un commanded pedal is worse than loss of drive, perhaps a "drive disconnect" switch could be an option on the way down.

[nodrama]

Malabo
Just to eliminate your query on landing gear vs hydraulics. The landing gear on an AW169 is moved by electric motored actuators. No hydraulics involved at all.

[RVDT]

Another EASA AD issued with expansion -




Emergency Airworthiness Directive

AD No.: 2018-0261-E

Issued: 30 November 2018



Required as indicated, unless accomplished previously:

Repetitive inspection:

(1) Within 10 flight hours (FH) after the effective date of this AD, and, thereafter, at intervals not exceeding 10 hours, inspect the slippage marking of the castellated nut installed on the back-end of the TR servo actuator in accordance with the instructions of Part I, and inspect the roughness and breakaway force of the TR duplex bearing in accordance with the instructions of Part II, of the applicable ASB at intervals not exceeding 30 hours.



Part Removal and Send to Leonardo:

(5) From the effective date of this AD, within 2 days after removal of a TR duplex bearing, if part of the corrective actions as required by paragraph (2) or (3) of this AD, as applicable, send the TR duplex bearing and the collecting containers of the grease to Leonardo for in-shop inspection. This can be done by using the instructions of the applicable ASB.

[[email protected]]

Yes, if full un commanded pedal is worse than loss of drive, perhaps a "drive disconnect" switch could be an option on the way down. so you want to add to the complexity of the TR system, adding weight and more potential points of failure, into something that already has a low probability of failure in any form, just to guard against an even more remote type of failure.............c'mon chopjock- get real!

[chopjock]

so you want to add to the complexity of the TR system, adding weight and more potential points of failure, into something that already has a low probability of failure in any form, just to guard against an even more remote type of failure.............c'mon chopjock- get real!

If you have an un commanded stuck full pedal and you don't have easy throttle control or perhaps a bird strike on a tail blade causing severe vibration with tail rotor imminently about to fall off etc and high up in the cruise, how do you maximise your chances to land safely? Presumably this is when you do not want a turning tail rotor? Just for discussion ...

[Thomas coupling]

I love you choppy - you should have your own show. The last dozen posts have all been about real professionals tying themselves up in knots trying to win you over. :=
For someone who actually poles one of these contraptions, your technical and practical insight and knowledge of rotary wing operations is excrutiatingly lacking.
You've been like this for years on Pprune and I for one love watching proper pilots froth at the mouth trying to convince you .
Great show choppy - love it.

[chopjock]

crab
so you want to add to the complexity of the TR system

I think they are all doing it wrong. It would be much simpler if there was an electric motor out on the boom and simply switch it off when not needed...

[ShyTorque]

Chopjock, the RAF SAR Wessex which crashed into the lake in Wales with cadet passengers on board had a tail rotor dis-connectable coupling failure.
First of all you complained that a tail rotor drive system is too unreliable but now you want to make it more complicated and therefore by definition less reliable...what sort of logic is that?

[Nadar]

I think they are all doing it wrong. It would be much simpler if there was an electric motor out on the boom and simply switch it off when not needed...


Everybody relax, the above statement should make it perfectly clear that he's just trolling you.

[John R81]

ADNo.: 2018-0261E


“For the reasons described above, this AD requires repetitive inspections of the TR servo actuator’s back-end castellated nut slippage marking, and of the roughness and breakaway force of the TR duplex bearing and, depending on findings, accomplishment of applicable corrective action(s). This AD is still considered to be an interim action and further AD action may follow.”

part 1 every 10 flight hrs
part 2 every 30 flight hrs

[SamYeager]

send the TR duplex bearing and the collecting containers of the grease to Leonardo for in-shop inspection

Just an ordinary SLF here and I wonder if anyone can explain why the grease containers are also required?

[dClbydalpha]

Just an ordinary SLF here and I wonder if anyone can explain why the grease containers are also required?
If things are wearing unexpectedly then it may leave traces in the lubricant that can be found by analysis.

[GrayHorizonsHeli]

unlike oil, where you can run a chip plug and filters to collect debris, grease holds onto those particles.
if a bearing has started to fail, those chips/flakes will continue to beat around the bearing. collecting the grease for inspection will help determine whats breaking down. whether its the bearing cage, the rollers themselves or the races. It may even determine that it's a foreign material that shouldnt be in there to begin with.

[SASless]

Or even if there was no grease or the wrong grease....or other kinds of contamination.

[ethicalconundrum]

Just an ordinary SLF here and I wonder if anyone can explain why the grease containers are also required?

The scientific name is 'tribology'. It studies lubricants, and effectiveness. If there is a failure of a lubricant, the examination of the fluid(grease) and contaminants may point to a specific failure mode.

[SamYeager]

Thanks to all that answered my query.

[ROC man]

Further info on AAIB website released today at 1400

[David Thompson]

Further info on AAIB website released today at 1400
AAIB Special Bulletin here ; https://assets.publishing.service.gov.uk/media/5c090ab1e5274a0b64c8a2f4/S2-2018_G-VSKP.pdf .

Already reported on the BBC .

[212man]

AAIB Special Bulletin here ; https://assets.publishing.service.gov.uk/media/5c090ab1e5274a0b64c8a2f4/S2-2018_G-VSKP.pdf .

Already reported on the BBC .
so clearly the EADs were not based on speculation

[gipsymagpie]

If I read this right, the yaw went full travel to the right. Since the limit of right yaw is usually set for the ability to yaw right in autorotation, it would probably mean it was still yawing right even in auto with engines shutdown (normal recovery actions for a tail rotor drive fail in the high hover). Thoughts?

[Slowclimb]

“Sufficient force and torque had been applied to the castellated nut on the actuator end of the control shaft to friction weld it to the pin carrier and to shear the installed split pin. The observed condition of the duplex bearing and the increased torque load on the castellated nut that remained on the spider end of the shaft is consistent with rotation of the tail rotor actuator control shaft. Whilst the shaft was rotating and a yaw control input was applied, the shaft “unscrewed” from the nut, disconnecting the shaft from the actuator lever mechanism, and causing the nut to become welded to the pin carrier.”

So... what provided the force and torque which welded the actuator-end nut and broke split pin? Failure of the duplex bearing locking the inner shaft to the outer shaft?

[[email protected]]

My reading of this is that the duplex bearing was overheating, binding and then seizing, causing rotation of the control shaft which undid the castellated nut and removed the pitch control of the TR allowing the servo to go to the 'full right pedal' position.

I don't care how many times you practice in the sim, that one is pretty unrecoverable from that height and speed.

The question is - is this an assembly issue (bearing end nut over torqued) or a maintenance issue? The aircraft was quite young.......

[Slowclimb]

GHH,



If you look back through the message thread there are a few pics of the arrangement. The locking in the pic implies a RH threaded nut and the directions of rotation would imply your theory could be correct.

Normally the servo piston is keyed in some way to stop this happening but you can't tell from the IPC drawing. Having a split/cotter pin and locking kind of implies that this is the only locking. If it was keyed you wouldn't go to that length.
Similar things have happened before but not through design issues.

Once the feedback link is disconnected the servo would invariably motor to the end of full travel and stay there.

Pic at #682 and IPC at #630

Looks like you guys were right.

[Speed of Sound]

The rent-an-idiot David Learmount has just been on the BBC News saying that the report reveals that the control cable between the pedals and the tail rotor had snapped!

Really?

Why would somebody agree to an interview about a very short bulletin without taking five minutes to read and understand it?

Looking at the photos and CAT scans of the bearing and carrier, I tend to agree that the bearing failed for some reason and the actuator linkage became disconnected as a result of this failure.

[DType]

Looks like it should have been a left hand thread, although that might just have postponed disaster.

[GAPU]

My reading of this is that the duplex bearing was overheating, binding and then seizing, causing rotation of the control shaft which undid the castellated nut and removed the pitch control of the TR allowing the servo to go to the 'full right pedal' position.

I don't care how many times you practice in the sim, that one is pretty unrecoverable from that height and speed.

The question is - is this an assembly issue (bearing end nut over torqued) or a maintenance issue? The aircraft was quite young.......

​​​​​​From the Special Update "The observed condition of the duplex bearing and the increased torque load on the castellated nut that remained on the spider end of the shaft is consistent with rotation of the tail rotor actuator control shaft."

I read that to mean that they think the increased torque on that nut was caused by the rotation of the control shaft - for the reasons you gave. Maintenance issue or defective bearing ??

[Jetscream 32]

What sober and horrible reading the S2 bulletin is - if only they had a few more minutes to transition into forward flight.... RIP!! Horrible sad loss for all involved

[Onceapilot]

My reading of this is that the duplex bearing was overheating, binding and then seizing, causing rotation of the control shaft which undid the castellated nut and removed the pitch control of the TR allowing the servo to go to the 'full right pedal' position.

I don't care how many times you practice in the sim, that one is pretty unrecoverable from that height and speed.

The question is - is this an assembly issue (bearing end nut over torqued) or a maintenance issue? The aircraft was quite young.......

The report seems to be a little confusingly worded. I agree, the duplex bearing at the spider end seems to have failed and then spun the shaft. Result, the castleated nut at the input end of the shaft has spun, with the shaft, against the pin-carrier, until the nut friction-welded to the pin-carrier and then the split pin was sheared and the nut unscrewed and disconnected the input. That is the way I read it. There does not seem to be any prejudice against the assembly in that. The duplex bearing failure looks to be the culprit from my interpretation. :sad:

OAP

[ShyTorque]

So, if my understanding is correct, it appears that the duplex bearing seized solid, causing what should have been a non-rotating component to become physically welded and locked to the rotating part of the shaft, so that it also rotated, unscrewed then threw off the locking nut, overwhelming its thread locking methods. Effectively, the pilots yaw pedals were no longer connected to the tail rotor pitch change mechanism and the servo went to full travel, pushing the tail rotor to a position where it would give maximum right yaw.

Unfortunately, JetScream32, transitioning to forward flight would not have made any difference. The aircraft would still have yawed hard right because the tail rotor made it do so as hard as it physically could. Shutting down both engines and entering autorotation would not have stopped it from yawing right. As I wrote earlier in this thread, the closest equivalent failure on a twin engined aeroplane would have been a full right rudder deflection with asymmetric thrust pushing it in the same direction, making it worse. If the "outer" engine of said aeroplane were shut down to remove the asymmetric thrust, the rudder was still at full deflection.

Irrecoverable and horrible to contemplate.

[sycamore]

A balance-spring bias on the servo unit towards left pedal( power input) would have been useful...

[Echo Romeo]

I despair with Learmonts comments.

[Otterotor]

And that is why blade designers strive to get the overall c.g. of the blade forward of quarter chord or the center of lift, such that if one loses pitch link input on one or all of the blades they will drive themselves to zero (or neutral) pitch. If the c.g. is aft of quarter chord the blade drives itself to maximum blade angle which is a divergent condition and does not help a disastrous situation. If other control failures are present such that the blade is driven to some adverse pitch angle, the c.g. effect may be over-ridden. Otter

[bumpthump]

I despair with Learmonts comments.

At least he stopped short of saying “The cable that broke is simply secured a by solderlesss nipple. Just like a 1963 BSA” From a mechanical engineering perspective, he was getting there.

[[email protected]]

Otter - in this case the TR servo was allowed to drive the pitch change mechanism all the way to full travel which would easily oppose any aerodynamic backloads..

[[email protected]]

Sycamore - the Lynx had such a spring bias unit but it wouldn't have helped as it would, if designed in the same way, have gone past top dead centre and pushed to the right instead.

[[email protected]]

OAP - the nut at the duplex bearing end was clearly overtightened - hence my comment about assembly problems. It may be that the TR pitch change mechanism has to be taken apart and greased regularly which ,might make the over-tightening a servicing error instead.

If the nut is over-tightened and impinges on the duplex bearing, you have a probable cause for the excess heat and seizing of the bearing.

[ShyTorque]

If the nut is over-tightened and impinges on the duplex bearing, you have a probable cause for the excess heat and seizing of the bearing.

Hence my earlier comment about the requirement for careful installation with this type of bearing.

[Nige321]

OAP - the nut at the duplex bearing end was clearly overtightened - hence my comment about assembly problems. It may be that the TR pitch change mechanism has to be taken apart and greased regularly which ,might make the over-tightening a servicing error instead.

If the nut is over-tightened and impinges on the duplex bearing, you have a probable cause for the excess heat and seizing of the bearing.

Hang on, doesn't the report suggest the overtightening is from the spining shaft...?
ie Wind-up in the shaft has pulled the spider nut down and unwound the servo end?
Page 7.
Am I miss-reading this?

Sufficient force and torque had been applied to the castellated nut on the actuator end ofthe control shaft to friction weld it to the pin carrier and to shear the installed split pin. Theobserved condition of the duplex bearing and the increased torque load on the castellatednut that remained on the spider end of the shaft is consistent with rotation of the tail rotoractuator control shaft. Whilst the shaft was rotating and a yaw control input was applied, theshaft “unscrewed” from the nut, disconnecting the shaft from the actuator lever mechanism,and causing the nut to become welded to the pin carrier.

[Grummaniser]

OAP - the nut at the duplex bearing end was clearly overtightened - hence my comment about assembly problems. It may be that the TR pitch change mechanism has to be taken apart and greased regularly which ,might make the over-tightening a servicing error instead.

If the nut is over-tightened and impinges on the duplex bearing, you have a probable cause for the excess heat and seizing of the bearing.
As I understand it the over tightening of the nut at the spider end of the shaft was likely caused by rotation of the shaft as the duplex bearing got tighter, not necessarily by over torquing during assembly. In other words the over tightening was a result, rather than a cause of the duplex bearing failure

[gevans35]

There was no grease, in it's original form, in the bearing. Was that likely to be cause or effect?

[ShyTorque]

Having looked again at the diagrams, it appears to me that the castellated nut assembly should be locked on the threaded end at the end of the shaft and it should be able to freely rotate with it. The duplex bearing should allow this. However, the duplex bearing seized, or partially seized and produced an extreme overheat. This resulted in the locking nut welding to the carrier and it then it unscrewed as the shaft continued to rotate. The carrier, which is vital for the pilot to input yaw demands, was then no longer connected.

[Echo Romeo]

There was no grease, in it's original form, in the bearing. Was that likely to be cause or effect?
if that is the case, then I suspect it might.

[gevans35]

if that is the case, then I suspect it might.

That's what it says in the report.

Either it was burnt off by an overheated bearing, or it was the cause of the over temperature and seizure?

[Just This Once...]

Shy, to my eye the control shaft is not designed to rotate at all. That it did so is without a doubt and the effect of the shaft rotating pushed the rest of the mechanical system beyond design limits.

[airsound]

Forgive me - not an engineer, but an observer of the scene. And I have read Bulletin S2. It's quite hard to understand...

However, I think what is confusing is that the bearing seized up at one end, and the nut came off at the other. Anybody else think that's the case?

airsound

[ShyTorque]

Shy, to my eye the control shaft is not designed to rotate at all. That it did so is without a doubt and the effect of the shaft rotating pushed the rest of the mechanical system beyond design limits.

You might be correct - that was my initial thought, as above. I do find the report difficult to understand without looking at the actual parts in question.

[Just This Once...]

If the control shaft had been keyed rather than a plain rounded shaft the outcome with binding at the TR end would have been different. Of course, this could have moved the failure point elsewhere if still not detectable to vib sensors or feet.

The design of the locking nut and carrier assembly looks totally appropriate for a rod that will only ever act in a push/pull manner. What the designer appears to have missed during failure analysis is that binding at the TR end would transmit the rotational force of the tail rotor onto the linear control rod inside the powered shaft. The rod was neither designed to freely rotate in this failure mode or mechanically prevented from spinning. It does not appear to be monitored by sensors either, so this control rod may have been spinning for a while.

[[email protected]]

Nige 321 and Grumanniser - yes, having re-read it, that is what the report says so the question is why did the bearing fail?

Airsound, as I understand the system, you have two shafts, one inside the other - one just goes in and out (control rod for pitch change) which is the inner one and the other spins around it to transmit the drive from the TR driveshaft to spin the TR.

The duplex bearing allows the inner (control) shaft to move longitudinally in the outer (driving) shaft so that you can superimpose pitch change inputs onto the spinning TR.

If the bearing seizes, the driving shaft starts to turn the control shaft and that is what broke the split pin and allowed the castellated nut to undo.

Presumably the extra drag on the driving shaft is what caused the spider end nut to tighten too much.

[Rigga]

I agree with JTO...The shaft was able to rotate with the rotor head enabling the nut to be compromised and shear the spilt pin to twist off. The issue with the bearing appears to be a lack of grease which was either burnt off or just not there when installed - that is not clear in the report.

In my experience similar TR pitch shafts have been mounted in splined guides to stop rotation.

I understand why the Actuator went to full travel and I'm unsure if a neutral default/detent setting is possible for an input failure mode.

[rlsbutler]

I was always the sort of pilot that was likely, at any one time, to earn an “E” for Aircraft Knowledge. Reading the latest Special, I can see what the author is saying, but I lost track early on and am happy to take his word for it. That perhaps explains my immediate and intense admiration of the inspection team as I read the first paragraph of “Findings from the technical investigation”. The aircraft is presumably one they were not familiar with. By definition, what they get from speed-reading the appropriate manuals is not what has actually happened. Add to that the muck-and-corruption in which the wreckage has settled. Just a few of the many components in front of them went wrong before the crash, while almost everything else was ruined after it. This work is not as brave as bomb-disposal work, but by heavens these experts have to be very intelligent, alert and cool. The hangar and laboratory work that follows is as impressive in its own way.

I have read many NTSB, several USAF and a couple of BEA reports (including the Orly Concorde). All of these tend to lose concentration or to have a clear angle. One or two of the USAF reports are clearly politically driven and miss the point they should be driving for. Year by year, the AAIB reports get to the point with diligence and logic whether the object of the investigation is a dinged gyrocopter or something very much more significant. How long shall we be provided with such a service ?

[gevans35]

The issue with the bearing appears to be a lack of grease which was either burnt off or just not there when installed - that is not clear in the report.

They're still investigating the "initial cause and exact sequence of the failure".

I have heard of bearings being stripped of grease by over aggressive cleaning, pressure washer?

[Pozidrive]

Forgive me - not an engineer, but an observer of the scene. And I have read Bulletin S2. It's quite hard to understand...

However, I think what is confusing is that the bearing seized up at one end, and the nut came off at the other. Anybody else think that's the case?

airsound

I can imagine one end tightening up as far as it can, then shearing the weakest point at the other end (the split pin).

[GrayHorizonsHeli]

it makes sense to me how things transpired once familiar with the parts and terminology.
what happened to the grease. Was the grease non existent or minimalat manufacture, leading to a failure 300hrs later? or did the bearing start failing and from the heat burn away the grease?
if the system doesn't get a redesign, whats the resultant inspection and remedy to prevent this from happening again?

[Onceapilot]

Gentlemen, in view of the report limitations, the future identification of the cause of the failure of the duplex bearing will be very important. Without specific experience of this type, I would venture the following thoughts. After the duplex bearing became effectively locked and the shaft rotated at T/R speed, I would guess that (with an input axial load) the castleated nut would friction weld to the pin-carrier in a short time (probably measured in seconds) then, followed almost immediately by the unscrewing of the nut and T/R runaway. The question must be, how to mitigate this process in the event of a duplex bearing failure.

OAP

[Tony Mabelis]

Sounds like a 'MOD', fitting a grease nipple might not be a bad idea, if it already has one, then a maintenance schedule revision to lube it more often.
Tony

[Onceapilot]

Sorry to be pedantic. The report refers to the "castleated nut" that unscrewed from the input end of the shaft and, the "locking nut" on the duplex bearing end (that remained in place). The illustration also includes a depiction of the "locking nut" at the duplex bearing end as, "castleated nut". Obviously, these possible confusions should be reviewed, IMO.

OAP

[Onceapilot]

What the report is saying is that the duplex bearing failed, for whatever reason, and stopped doing its job of allowing the spider to rotate freely around a stationary rod. The rod then started rotating with the spider. Because the rod would rotate anti-clock wise, it tightened the big nut up as far as it’s cotter pin would allow, and then the big nut would continue to rotate with the rod. At the smaller nut end, the effect would be the opposite. The locking wire and cotter pin would prevent the nut from undoing, and the smaller nut would turn with the rod. This caused friction against the carrier, eventually welding the smaller nut to it. The nut is now part of the fixed servo link. The rod would be turning against a cotter pin and locking wire, which it seems, from the report, that it sheared, and wound itself out of the nut.

The pitch change rod would now be no longer attached to the servo link to provide feedback of its position, and the servo would continue to move in the direction of its last input.

Agree, been here since my #944, the words used in the report just make it tricky. As far as the direction of movement of the servo input, it will depend upon the forces acting after the disconnection of the input nut and it's contribution (now removed).

OAP

[airsound]

nodramaMaybe I can help clear some of the confusion.....the duplex bearing failed, for whatever reason, and stopped doing its job of allowing the spider to rotate freely around a stationary rod. The rod then started rotating with the spider. Because the rod would rotate anti-clock wise, it tightened the big nut up as far as it’s cotter pin would allow, and then the big nut would continue to rotate with the rod. At the smaller nut end, the effect would be the opposite. The locking wire and cotter pin would prevent the nut from undoing, and the smaller nut would turn with the rod. This caused friction against the carrier, eventually welding the smaller nut to it. The nut is now part of the fixed servo link. The rod would be turning against a cotter pin and locking wire, which it seems, from the report, that it sheared, and wound itself out of the nut. You did. (clear up confusion). Thank you!

airsound

[Slowclimb]

Well, EASA EAD 2018-0250-E, back on 21 November required the duplex bearing on all 169s and 189s to be inspected before they flew again.

Finding out exactly what the inspection is looking for requires access to Leonardo AMPs. I'd dearly like to know what they were looking for.

[n5296s]

I don't care how many times you practice in the sim, that one is pretty unrecoverable from that height and speed.

I'm curious how this could possibly be recovered given any amount of height/speed. As I understand, even entering autorotation wouldn't help - the TR is in full right pitch and you'll just rotate all the way down. I suppose with an infinite amount of good luck it MIGHT be possible to flare and touch the ground with this going on and survive, if the cabin integrity remained. But who could realistically fly an auto while rotating at what, once every couple of seconds?

[Feathers McGraw]

I can't see clearly from the diagrams in the SB, but I think that if you have the actuator rod able to rotate on the duplex bearing then setting the correct torque on the castellated nuts is going to be difficult during assembly or maintenance.

Is there a method to do this or is it a matter of getting the nut to a position where the split pin can be fitted. That does not guarantee the final position of the nut, so any torque value would be at best vague.

[mercurydancer]

I am not an engineer or a pilot, but involved in healthcare risk management.
In summary, there is no pilot error, he was faced with an impossible situation.
Now, is the remaining situation a design error, a maintenance error, a build error or a combination of all three?

[Pozidrive]

I am not an engineer or a pilot, but involved in healthcare risk management.
In summary, there is no pilot error, he was faced with an impossible situation.
Now, is the remaining situation a design error, a maintenance error, a build error or a combination of all three?

You could add manufacturing and materials quality.

[Speed of Sound]

I can't see clearly from the diagrams in the SB, but I think that if you have the actuator rod able to rotate on the duplex bearing then setting the correct torque on the castellated nuts is going to be difficult during assembly or maintenance.


Figure 5 in the SB seems to show a square-section end to the actuator rod which could be used to prevent the rod from rotating while the nut is torqued.

[funfly]

....and a surprise conclusion after the conjecture in this thread. This must have quite a relevance to many other Rotaries in operation not just the same model.

[jimjim1]

Figure 5 in the SB seems to show a square-section end to the actuator rod which could be used to prevent the rod from rotating while the nut is torqued.

I had just prepared the screenshot -

https://assets.publishing.service.gov.uk/media/5c090ab1e5274a0b64c8a2f4/S2-2018_G-VSKP.pdf
Fig 5.https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/645x512/aw169_shaft_end_990a96a627934c30850aea4e0e7a3fc803e3264e.png

[Pozidrive]

I had just prepared the screenshot -

https://assets.publishing.service.gov.uk/media/5c090ab1e5274a0b64c8a2f4/S2-2018_G-VSKP.pdf
Fig 5.https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/645x512/aw169_shaft_end_990a96a627934c30850aea4e0e7a3fc803e3264e.png

Well spotted. And now you've mentioned it, Figure 2 also appears to show flats at the other end, slightly obscured behind the lock wire.

[Planetary]

The torque on the castellated nut retaining the shaft in the duplex bearing has no effect on the function of the bearing. It's an opposed angular contact set-up, with solid inner races. The torque applied to the nut will have no effect whatsoever on the internal bearing clearances. Any over-torque of the nut would serve only to stress the shaft, most likely in the waisted section immediately below the thread, and that doesn't seem to have occurred give the photos above.

[ShyTorque]

I'm curious how this could possibly be recovered given any amount of height/speed. As I understand, even entering autorotation wouldn't help - the TR is in full right pitch and you'll just rotate all the way down. I suppose with an infinite amount of good luck it MIGHT be possible to flare and touch the ground with this going on and survive, if the cabin integrity remained. But who could realistically fly an auto while rotating at what, once every couple of seconds?

As I wrote earlier, once the aircraft fuselage begins spinning in yaw, the original trimmed position datum for the cyclic no longer exists because it (the fixed swashplate) is also rotating.

[Overdrive]

" Year by year, the AAIB reports get to the point with diligence and logic whether the object of the investigation is a dinged gyrocopter or something very much more significant. How long shall we be provided with such a service ? "


In an age of replacing technical excellence and genuine quality with mitigation and perception management (for both financial and political "Gain") across much of business and life, I understand and share your concern.

[Ahernar]

https://www.youtube.com/watch?v=9XMfhKxwo2c

This made me think that with the right software and servo's you can fly down a helicopter without a working tail rotor . Biggest problem will be for the soft to select the place to land or allow some form of human input , but from a technical point of view what i seen above means automation can fly a TR-less heli indefinitely

[[email protected]]

n5296s - it could be survivable from a low hover or a situation where you are so close to the ground that an immediate landing could be made (probably still roll over though).

In the high speed cruise it might not be so dramatic until you slow down so perhaps a very high speed running landing might be possible, drooping the Nr to reduce the TR power (again you are probably going to roll over)

Other than those two extremes I don't believe there is any solution - they were extremely unlucky.

[FlimsyFan]

https://www.youtube.com/watch?v=9XMfhKxwo2c

This made me think that with the right software and servo's you can fly down a helicopter without a working tail rotor . Biggest problem will be for the soft to select the place to land or allow some form of human input , but from a technical point of view what i seen above means automation can fly a TR-less heli indefinitely

Ah, I’m a mechanical man, not a software guy, so you’ll have to excuse my ignorance that software can now overcome the established laws of physics.

[Agile]

Software cannot overcome the laws of physics but it can

diagnose the problem in millisecond,
analyze 100s of possibilities in a few millisecond more,
apply the scenario that would maximize the chance of survival.

The key to that is more sensor distributed throughout the aircraft and a huge onboard database

Did you hear about the self recovery button on the cyclic of the H160 I think this a sign of things to come

[Uplinker]

.....This made me think that with the right software and servo's you can fly down a helicopter without a working tail rotor . Biggest problem will be for the soft to select the place to land or allow some form of human input , but from a technical point of view what i seen above means automation can fly a TR-less heli indefinitely

You seem to be missing a fundamental point: A helicopter turns its main rotor by having an engine, (or engines), pushing round - via a small gear wheel - a large horizontal ring gear which is connected to the main vertical shaft that drives the main rotor.

Newton’s third law states that ‘every action has an equal and opposite reaction’, so the turning force that the engine(s) are applying to turn the ring gear against the forces from the main rotor also pushes back on the engine(s). On the ground, this reaction force is not enough to move the helicopter when it is sitting on its skids/wheels, but in free air the reaction force pushes the engine(s), and therefore the helicopter to which they are bolted, in the opposite direction to the turning of the main rotor. The engines are effectively trying to ‘climb round the main ring gear - similar to a motorbike doing a wheelie.

To compensate for this, a tail rotor is fixed at the end of a long boom to apply a side force to oppose this reaction torque. The tail rotor has to produce a variable force to allow adjustment for varying main rotor torque, so it is made variable pitch, controlled via foot pedals. This also allows the pilot to yaw the aircraft when required, similar to the rudder of a conventional aircraft.

Having a tail rotor pushing sideways, will push the whole helicopter bodily sideways, (in the air), so the main rotor is usually offset sideways from the vertical by a small amount so a component of the lift force pushes in the opposite direction to the tail rotor, and stops the sideways drift.

When a helicopter loses its tail rotor drive, it will start turning in yaw due to the reaction to the engine torque on the main ring gear. If it has sufficient forward airflow past its tail fin, the heli will be able to fly straightish, but on slowing and landing, will yaw round and round. I am not a heli pilot but I understand the recovery is to idle the engine(s), which removes the turning torque, and gives the pilot a chance of an auto-rotation landing. This video of a ?Wessex lifting an air-con unit from a building roof and losing tail rotor drive illustrates all of this very clearly.

https://m.youtube.com/watch?v=5kXUZQYFu18

[Bell_ringer]

The key to that is more sensor distributed throughout the aircraft and a huge onboard database


Sensors can also provide bad data, to systems which will make the wrong decision as a result.
You can push a problem upstream, but it does not get rid of it.
If a servo doesn't know when it should stop, it doesn't matter whether it was someone's foot or a digital system that gave it the instruction to initially move.
You can't engineer out failures, the more underlying complexity, the more failure points are introduced.

[dClbydalpha]

Software cannot overcome the laws of physics but it can

diagnose the problem in millisecond,
analyze 100s of possibilities in a few millisecond more,
apply the scenario that would maximize the chance of survival.

The key to that is more sensor distributed throughout the aircraft and a huge onboard database

Did you hear about the self recovery button on the cyclic of the H160 I think this a sign of things to come

I'm a big proponent of the capabilities of avionics. But at this point it has lost a major effector. While the auto-stab may have a mode that can help level, it could also help relieve torque, but what would it use to counter a yaw forcing, possibly even beyond the design limits? I suspect nothing, at least nothing that would change the game significantly for the pilot.

[robdean]

Software cannot overcome the laws of physics but it can

diagnose the problem in millisecond,
analyze 100s of possibilities in a few millisecond more,
apply the scenario that would maximize the chance of survival.
The key to that is more sensor distributed throughout the aircraft and a huge onboard database


One of the most recent heavy jet crashes had its root in a faulty AOA sensor being acted upon by automated systems. Something which has become very clear in recent years is that the interaction between automated systems decision making and pilot initiative is very complex, not least given the added factors of *multi-mode* automation and the cognitive barriers to a pilot avoiding being dangerously 'behind' an automated system. The decision tree in a comprehensive automated system would be dauntingly complex, subject to human error in its creation and subject to sensor malfunction in use: yet it would have to be significantly more robust and reliable than existing systems or would offer no net gain. Even then, pilots would need to defer instantly to the automation... unless the automation was malfunctioning, in which case the pilot would need to instantly fight or disable the automation. Which brings us full circle.

There is surely a case for more intelligent and intuitive safety systems, but this accident seems to have arisen from an unforeseen failure mode leading to a malfunction which in all likelihood was simply not recoverable. You can’t automate 'thinking of what you haven’t thought of' nor program a computer to fix in real time any catastrophe which sets the laws of physics in fundamental opposition to accident survivability.

[Ahernar]

Not going to argue much around this but there are no laws of physics preventing tailles rotor flight . You have to equalise the rotor drag with drag from the body of the aircraft . If this requires 130 rpm , 130 rpm will be necessary ,if that's not humanly pilotable does not mean some autostab will find it impossible too .Most problems will be around oscilations and vibrations in the rotor disk but for an emergency descent it should work

[Slowclimb]

Well, EASA EAD 2018-0250-E back on 19 November required inspection of the duplex bearing on both 169s and 189s before they flew again. So hopefully there are no duff duplex bearings in service anymore.

I’d dearly like to know what the inspection was specifically looking for, as that would shed light on why the bearing failed. Knowing that would probably require access to Leonardo AMPs. Anyone know?

[ShyTorque]

Looking at the photo evidence in post#983 above, a thought that crossed my mind - would a simple left handed thread on the shaft have prevented this tragedy?

Having recently changed a rear wheel bearing, even my old car has a left handed thread on the left hand side stub axle, to prevent the hub nut coming loose in service.

[rotorspeed]

For those with a good engineering understanding of this failure, would the tail rotor have been harder to turn by hand than normal, pre-flight, given the breakdown of this bearing? Obviously not easy to reach on a 169, but is on most smaller types.

[ShyTorque]

Not going to argue much around this but there are no laws of physics preventing tailles rotor flight . You have to equalise the rotor drag with drag from the body of the aircraft . If this requires 130 rpm , 130 rpm will be necessary ,if that's not humanly pilotable does not mean some autostab will find it impossible too .Most problems will be around oscilations and vibrations in the rotor disk but for an emergency descent it should work

130 rpm (Nr) is well below the normal operating range of modern helicopters. Most run around 275 - 350 or even higher for smaller machines, with a minimum in flight of 90% of that figure for aerodynamic reasons. Going lower would cause the blades to stall and loss of control of the rotor disc.

[Pozidrive]

For those with a good engineering understanding of this failure, would the tail rotor have been harder to turn by hand than normal, pre-flight, given the breakdown of this bearing? Obviously not easy to reach on a 169, but is on most smaller types.

With the length of the blade and the weight of the rotor, I think it would be very difficult to detect any tightness or "grittiness".

[Pozidrive]

Looking at the photo evidence in post#983 above, a thought that crossed my mind - would a simple left handed thread on the shaft have prevented this tragedy?

Having recently changed a rear wheel bearing, even my old car has a left handed thread on the left hand side stub axle, to prevent the hub nut coming loose in service.

Interesting, but your hub nut still has a split pin? It was once the practice to use LH threads on the left of vehicles, including wheel nuts. I think it was all a bit theoretical and the fact it isn't done any more suggests it wasn't really necessary.

I don't know which way this tail rotor turns relative to the control shaft, but with similar threads the nut at one end would tend to loosen and the other would tighten. One LH and one RH would both try to tighten, or loosen - depending on the rotation.

[RVDT]

# I will let you work out the rest for what happens if the bearing fails and what could possibly happen to the nut.

Apparently so - it will be interesting to see how the design complies with CS-29.

I am surprised that the attachment of the lever mechanism to the control shaft relies purely on clamp up with the nut. No key or index?

Maybe there should be a weak point in the shaft so that the integrity of the feedback is maintained so the servo does not go to full travel?

I am amazed that things like this can still develop in this day and age. Somehow folk have lost sight of the details.

[nodrama]

Looking at the photo evidence in post#983 above, a thought that crossed my mind - would a simple left handed thread on the shaft have prevented this tragedy?

Having recently changed a rear wheel bearing, even my old car has a left handed thread on the left hand side stub axle, to prevent the hub nut coming loose in service.

I'm not sure that it would have done. The smaller nut would have tightened up against its cotter pin and still possibly spun with the pitch rod. It may have still welded itself to the pin carrier. It wouldn't unwind, but could strip the threads or break the pin carrier if the turning force was great enough. If the pin carrier failed and detached from the servo feedback link (as it did on VSKP- according to the report, it was found detached and lying in the TRGB cowling), the resulting loss of TR control would be the same.
Or, the tightening of both nuts on each end may increase the clamping effect on the rod and stop it from spinning, despite the failed bearing. Difficult to say really, and one for the design engineers.

[SASless]

Uplinker,

In your video....that was a Sikorsky S-58T that had the tail rotor drive failure.

The Pilot controlled the aircraft and was maneuvering towards a safe landing area.

He maintained power until he thought he could make an authoritative landing and cut the throttles to Flight Idle.

After the accident I had the opportunity to talk with him about the crash....and he was very bothered that he was unable to land the aircraft safely.

He did his best....but the odds were against him...but he flew the machine all the way to the ground and did not give up trying to land it.

The failure we see in the 169 under discussion was a very different kind and far more critical than that shown in your video.

It was a loss of thrust in the video and just the opposite (apparently) in the 169 crash.

The 58T was a drive shaft failure.....where the 169 was far more complicated a failure.

[Sky Sports]

If the duplex bearing seized for an unknown reason, why hasn't the type been grounded like the Puma's were when the Main rotor heads came off for an unknown reason?

[Lonewolf_50]

If the duplex bearing seized for an unknown reason, why hasn't the type been grounded like the Puma's were when the Main rotor heads came off for an unknown reason? Because it may be a lot simpler to inspect, check for correct function of, and replace perhaps, a bearing than it is to dismantle a main transmission when the Company in question was damned sure (based on its public utterances) that someone hadn't attached one of the three transmission mounting pins correctly ... which wasn't even the problem.
Put another way, comparing the two accidents the way you are trying to is a category error.

[sycamore]

Would the HUMS `indicators` to show an increase in t/r vibration due to the duplex bearing`s ongoing failure...?

[Uplinker]

Uplinker,

In your video....that was a Sikorsky S-58T that had the tail rotor drive failure. . Thanks, I wasn’t sure, hence my question mark.


The failure we see in the 169 under discussion was a very different kind and far more critical than that shown in your video.
It was a loss of thrust in the video and just the opposite (apparently) in the 169 crash.
The 58T was a drive shaft failure.....where the 169 was far more complicated a failure.

Fair enough. As I stated, I am not a heli pilot, but I was trying to explain to Ahernar why single rotor helis need a tail rotor, because they seem to think a (single rotor) heli can fly and land without one.

[Ahernar]

130 rpm (Nr) is well below the normal operating range of modern helicopters. Most run around 275 - 350 or even higher for smaller machines, with a minimum in flight of 90% of that figure for aerodynamic reasons. Going lower would cause the blades to stall and loss of control of the rotor disc.

Ofc , i was meaning the rpm of the cabin !!!
@ Uplinker
You lose TR , you spin faster and faster until at certain rpm the drag from the spinning body of the heli counterracts the rotor drag . Unfortunately no human can control a helicopter spinning like that , but a computer surely can ,go watch that rocket gently land after spinning from several km height . I'm not expecting the control system to be able to keep it right afterward it lands so the machine will be lost but the landing would be gentle and very survivable ( except some bad cases of fire / bad terrain ) .

[[email protected]]

Ahernar - I think you place too much faith in computers - an autopilot in a helicopter only has the same controls as the pilot to control its attitude. Yes it works faster and constantly but the sensors required to make them work have limits too.

And btw the rocket in your video didn't land, it crashed.

[Nadar]

@Ahernar How do you think humans deal with a 130 RPM fuselage? I think you might as well crash to the ground...

Since there is a lot of speculation about all kind of (unrealistic or irrelevant) ways to counter this, I'm surprised that emergency tail rotor recovery systems hasn't come up. From what I understand the idea here is to have some alternative way to create anti-torque trust at least for a limited time to allow landing. There's a patent from 2005 (https://patents.google.com/patent/EP1787907A2/en) here for example, and I'm sure there are other potential solutions. Whether such a system would be able to overcome the trust from the TR going above "full" pitch is a question of course.

I'm more interested in finding the cause of the failure than to speculate about possible solutions without fully understanding the problem, but since it seems to be a returning topic it might be a better starting point than making automated systems for killing people while still in the air.

[skadi]

And btw the rocket in your video didn't land, it crashed.

It did not crash, it ditched very softly ... :)
The high CG and the low lift of the landing legs prevented it from staying upright.

skadi

[ShyTorque]

Interesting, but your hub nut still has a split pin?

No, it doesn't have a split pin. The outer edge or lip of the nut is turned to a thinner profile and once the nut is torqued up the lip is knocked with a hammer and punch into a longitudinal groove on the end of the stub axle. From personal experience, this isn't quite as secure a method as a castellated nut and split pin. However, a car can also be driven in reverse so it is important to secure it in both directions. Unlike a tail rotor, which only runs in one direction.

[ShyTorque]

Ofc , i was meaning the rpm of the cabin !!!
@ Uplinker
You lose TR , you spin faster and faster until at certain rpm the drag from the spinning body of the heli counterracts the rotor drag . Unfortunately no human can control a helicopter spinning like that , but a computer surely can ,go watch that rocket gently land after spinning from several km height . I'm not expecting the control system to be able to keep it right afterward it lands so the machine will be lost but the landing would be gentle and very survivable ( except some bad cases of fire / bad terrain ) .

If the cabin rotates at 130 rpm, this will result in a net loss of 130 rotor rpm. This would almost certainly result in the same outcome - it is a loss of almost half the rotor rpm needed to sustain lift and the integrity of the disc would probably be lost, especially as the aircraft would be descending at a high rate.

[dClbydalpha]

If the cabin rotates at 130 rpm, this will result in a net loss of 130 rotor rpm. This would almost certainly result in the same outcome - it is a loss of almost half the rotor rpm needed to sustain lift and the integrity of the disc would probably be lost, especially as the aircraft would be descending at a high rate.
Out of interest where does the 130rpm come from?
The only other time I've seen data from an event like this it was about 120 deg/sec.

[Ahernar]

Out of interest where does the 130rpm come from?
The only other time I've seen data from an event like this it was about 120 deg/sec.

An exageration from my belly ( sort of worst case scenario for the control system ) . I would have to calculate real rpm values . The airframe coud be further designed to lower this value , with an enlarged fin aft , maybe even a little pirotechnically launched drogue chute (that's my new ideea , lol ).

[ShyTorque]

(that's my new ideea , lol ).

Actually, neither of those are new ideas. The late Ray Prouty discussed both in his books, years ago.

[Onceapilot]

There will be a reason for the bearing failure that probably is the main cause of the subsequent accident. Questions about the overall design of the input assembly might come into it. However, there will be great effort made to identify why the bearing failed and, the mitigation of that process. It might be anything from: a major redesign, an improved bearing, improved QC of the bearing, revised servicing or an improved inspection.

OAP

[ShyTorque]

Let's hope so!

I'm sure that all the RW pilots reading this are very concerned that the failure of a relatively small but highly critical item such as this bearing could bring down an aircraft and that the design wasn't in any way "fail safe". I certainly am.

[gevans35]

Let's hope so!

I'm sure that all the RW pilots reading this are very concerned that the failure of a relatively small but highly critical item such as this bearing could bring down an aircraft and that the design wasn't in any way "fail safe". I certainly am.

'Twas always thus. Years ago, I was fortunate to notice the clutch light flicker a couple of times during run-up on an R22 and shut down. It turned out to be the upper sheave bearing breaking up. Not a good failure to have.

[m25]

Can anyone explain why the "big nut” holding the damaged bearings onto the shaft tightens up in this scenario? Looks from the photo to be a right handed thread and I would have thought it would tend to be undone a little if the bearings failed, assuming the appropriate direction to undo the small nut at the other end....?

[ShyTorque]

As someone who has been flying helicopters for a living for almost four decades, the prospect of this type of tail rotor control failure has always greatly concerned me.

I've been writing about it on this forum and elsewhere (initially when I flew RAF helicopters) since the mid 1980s. Some would say I "bang on" about it, but imho with good reason. This situation is a helicopter pilot's worst nightmare. Tail rotor control is completely lost and the tail rotor pitch then goes to either full power at "push" or "pull". I don't know of any present helicopter where this situation is recoverable by the pilot, whatever his level of skill and experience.

The answer is to mandate that by design that all helicopter tail rotor control systems are equipped with a centreing device for the servo valve mechanism, to prevent a servo runaway in the event of the pilot's flight yaw controls becoming disconnected, such as now appears to have occurred here. With such a device fitted, the servo is automatically biased to move to a fixed, central position enabling controlled flight to still be available and a safe landing of some sort to be carried out.

Some aircraft manufacturers, such as Sikorsky, having been fitting such devices to their aircraft for many years. Other manufacturers, as appears to be the case here, obviously still do not.

These devices are not difficult to design. Most consist of a simple spring mechanism. The simplest one I've seen (Super Puma) consists of something that looks very much like a bicycle tyre pump, with a spring either side of a piston, sitting inside a cylindrical body. The body of the device is attached to the aircraft structure and the "piston rod" is attached to the servo control valve arm. The piston rod moves in and out of the body under the influence of the normal movement of the servo valve mechanism, such as when the pilot moves the yaw pedals. As it moves it compresses one spring and relaxes the other and is "invisible". However, in the event of a pilot's control input disconnection (e.g. a cable break), the springs equalise to bring the piston to a central / neutral position, so that the servo valve cannot not run away. The Sikorsky devices I'm familiar with are slightly more sophisticated and consist of quadrant arms controlled by springs, but the devices all work in a similar way.

Had such a device been fitted to this 169, the outcome might have been very different.

[helimutt]

As ShyTorque says, it is probably one of two things we, as helicopter pilots, would worry about failing the most, save for a rotor blade departing in flight. Ok so three. But a main gearbox failure (we don't carry a spare) or a tail rotor failure (for which we can only practice in the sim, and then its what the computer modellers/engineers and designers 'think' will actually happen).

My personal worry is a tail rotor failure, everything else we are pretty much well trained for, but in this day and age of modern technology, the 169 design does appear very simple. Too simple?

[Pozidrive]

...The answer is to mandate that by design that all helicopter tail rotor control systems are equipped with a centreing device for the servo valve mechanism, to prevent a servo runaway in the event of the pilot's flight yaw controls becoming disconnected, such as now appears to have occurred here. With such a device fitted, the servo is automatically biased to move to a fixed, central position enabling controlled flight to still be available and a safe landing of some sort to be carried out...

This wouldn't help if the failure was beyond the servo, i.e. the shaft to the tail rotor. There is also the down side of adding weight and complexity.

[Uplinker]

Perhaps a way of removing hydraulic power to the actuator in an emergency as well?

The TR pitch control linkage of this aircraft looks insanely complicated and convoluted to my untrained eye - so many linkages, pivots, rose joints, cranks and other components: they must be a nightmare to maintain and keep working smoothly and check during the pre-flight inspection?

Seems bizarre to have only a single yaw control system, which if it fails will guarantee a crash.

[CloudHound]

NOTAR technology was meant to solve this failure mode, wasn't it? Having flown an MD-600 and a Jet Ranger I couldn't 'feel' any difference.

Is there a reason why the design hasn't found wider adoption?

[Ahernar]

Actually, neither of those are new ideas. The late Ray Prouty discussed both in his books, years ago.
Probably yes but not in conjunction with computerised control .
Just for fun - kind of amazing - 2.20
https://www.youtube.com/watch?v=Jm0JQf2mmm8
Anyway i just read that fly by wire for helicopters is being researched in the military so if it's capable of coping with TR loss we will find someday .

[ShyTorque]

This wouldn't help if the failure was beyond the servo, i.e. the shaft to the tail rotor. There is also the down side of adding weight and complexity.

A) Maybe not but what is the point of that statement?

B) The size and weight of such a mechanism (they do exist!) are very small, possibly adding not much more than couple of kilos, if as much as that. All we're talking about here is a device to spring bias the hydraulic servo control valve to a central position, i.e. to replicate the pilot's normal pedal input.

Can I suggest you read the following publication? Section 3 is of relevance to this accident:

https://publicapps.caa.co.uk/docs/33/CAPAP2003_01.PDF

[RVDT]

The answer is to mandate that by design that all helicopter tail rotor control systems are equipped with a centreing device for the servo valve mechanism, to prevent a servo runaway in the event of the pilot's flight yaw controls becoming disconnected, such as now appears to have occurred here.

Really? Go back and read the report.

[ShyTorque]

RVDT, Your point isn't clear. I've read the report, thanks, more than once. It would help if you would explain what your disagreement is based on, rather than asking a very vague question and giving an order..

[nodrama]

I think what RVDT is pointing out is that it wasn't 'a servo runaway'. That would be when a servo malfunctions and moves uncommanded. The hydraulic systems and servo were operating correctly, as far as has been revealed, and the servo was just doing what it had been commanded to do before the feedback link stopped working.

[RVDT]

Shy take a look at how the thing works. It was not the input that failed but lets say the "output" to cancel the input via the feedback mechanism.

It's pretty simple - when all 3 points attached to the lever line up with the servo valve in the neutral position nothing happens as the control shaft moves and cancels the input request.

As the feed back end became detached - nothing to cancel the input - servo motors to the stop and in this case full right pedal?

BTW - Sikorsky has been fitting centreing as the result of accidents or incidents not unlike this one - not pro-actively.

The design and acceptance of the AW models may be under review shortly. If not I will be very surprised.