Apache Tail Rotor Failure

[megan]

An interesting account, page 46 in the attachment.

http://www.raf.mod.uk/rafcms/mediafiles/3EEA9B62_5056_A318_A88B016F60697008.pdf

[crunchingnumbers]

Thanks for sharing a great article.

Really interesting in terms of conditioned/learned behaviour versus situational. Kudos to the crew under extremely stressful conditions and thankful for their safe return.

After all that I might have been driven to do a Phd in Behavioural Psychology too. :ok:

[Lonewolf_50]

I did not notice any mention of the failure mode, in terms of detail, other than the picture in the article. (Nice link, thanks). Is there a publicly accessible article (maybe in a mechanics circular?) that discusses what failed, how and why? (As I've never flown an Apache, I am not sure if this is a class problem or a one off failure of a section of the Tail Rotor Drive system).

[mixing lever]

Failure of a TR hanger bearing happened to a AH-64A at Fort Rucker in 1999 while at a 400' hover. Aircraft spun into pine trees, both pilots recovered with minimal injuries.
Israelis had an AH-64A lose its TR in flight and was able to land at very high speed and got it stopped upright. No injuries.
There was another complete loss of AH-64D tail rotor in Afghanistan in 2007 (only the TR gearbox remained). The pilot was able to continue flight to a runway. He realized on first landing approach that he was not setup well enough and went around. Second approach resulted in more or less controlled touchdown and roll out. No injuries.
The Apache may be one of the ugliest helicopters made, but it is incredibly sturdy.
Kudos to the aircrew in original post.

[krypton_john]

I'm amazed he could arrest the descent and perform the go-around without spinning!

[[email protected]]

It's all about speed John - if you have enough airflow over the tail (and if your tail area is big enough) you can provide enough anti-torque thrust to allow continued flight.

You have to fly cross-controlled with a fair amount of sideslip and be careful which way you turn but it allows you to assess the situation and determine the minimum speed you can accept for landing.

[krypton_john]

Yeah I know in theory, but maintaining level is one thing, arresting a descent is a lot more thing!

Even with some speed... it just surprised me. That tail is a wing but even so...

[John R81]

Training tail rotor failure in the 120 that's how it goes. Make a dummy approach and bleed speed until she just starts to yaw, then fly away - adding speed and power, then repeat the approach just above the yaw speed you found for run-on landing, steer with the throttle when the skids touch down.


I can also say that if you don't arrest the yaw quickly (and lose the aerodynamic wing effect of the tail) then despite having several tries we could not fly-out back to straight / level, and could not add power to climb without losing out to the torque. We were left almost broadside-on to the direction of flight, high drag preventing speed gain. With me at the controls, then unless I can find height to drop the lever and get straight again (fly off a cliff, perhaps?), it looked like I would either land at 50knt whilst sideways (and roll to destruction) or try so hard that she would get into a true spin (into the ground / trees / whatever was underneath at the time). As this was just training, we terminated the exercise and used a significant amount of right boot to bring her back into trim.




Interested to hear if anyone has managed to recover from such a spin by flying out under power without using the tail rotor.

[[email protected]]

JohnR81 - that's why you shouldn't do a low dummy approach to assess the breakaway speed - you should do it at height so you can dive the speed back on if you get too slow.

Once you have an idea of how the aircraft handles you can make a dummy approach.

A crosswind from the power pedal side makes a big difference as well.

[Hughes500]

Crab

Why would you want a cross wind, tried it with various types and have found that nose into wind is much better 3 big reasons slower forward speed and better airflow over your vertical fin and if you touch th ground you won't roll over once nose straight.
John you are correct to do it low to the ground ( with a safety pilot ready to heave in the pedal if it goes wrong)
. I find the best technique is as follows ( in a Hughes product ) fly ac into your area at best rate of climb speed ( 269 is 41 kts 369 is 62 kts ) descending ( start with pedals in cruise setting to get used to what is going to happen). the ac should be yawed to the left. At about 10 ft level the ship from the descent with cylic only. The ac is now flying yawed to left at 5 ft with direction into wind. As you slow the ac with a combination of gentle power and rear cylic ( careful to keep ac level ) the ac will gently start to yaw right with the combination of power and slowing. Once straight you can't use any more power or slow the ac down without right yaw ( John's break away point ). Only option now is too slowly close the throttle to maintain direction. This will also have the effect of slowing rrpm. Which means the ac will stop flying and drop onto ground in a straight line. DO NOT lower collective let ground friction slow you down. Taught this to all my PPL's, once they are confident we try with more and more right pedal.
Some examples in a 369 with 15 kts of wind full fuel and 2 pax you can land the ac with the pedals stuck in cruise flight with no throttle adjustment. Stick more right pedal in at about 50 kts setting then a very slight closure of throttle.
Health warning to the Eurocopter boys US helis ( Hughes stuff ) are left pedal machines you need right boot

[Paul Cantrell]

Hughes500: Why would you want a cross wind, tried it with various types and have found that nose into wind is much better

Interesting because I've done several thousand in Robinson, Enstrom, and Bell and I tend to like the results more with the crosswind than a headwind for most of them.

I use three different methods depending on the situation:

1) A fairly fast running landing with a level off a few feet above the surface and a deceleration until the nose starts to swing due to torque. (i.e. a pretty standard tr/fail technique).

2) An aggressive deceleration about 10 feet above the surface, use power to stop all ground motion (kind of turns into a sideways flare) and then a hovering auto (I like this one for landing on surfaces that won't allow sliding, like a swamp)

3) An autorotation to touchdown using a bit of torque at the bottom to get the skids to align.

I like the left crosswind on the first two, headwind on the last (makes it easier to zero out the groundspeed for touchdown).

For a low inertia system, I tend to like the first two more... In the R22 #1 was good if you had pavement to slide onto, #2 was better for uneven surfaces.

[[email protected]]

Hughes 500 - what you describe isn't a TR drive failure drill - it is a tail rotor control failure - ie stuck pedals in different positions which is a different situation.

A crosswind gives you some anti-torque force (ie what you are lacking in a TR drive failure) and gives you lower run-on speeds before breakaway as a result.

DO NOT try to assess breakaway speed at low level with a real TR drive failure.

BTW - techniques you can get away with on light singles or twins may not be appropriate to a heavy twin like the subject of this thread.

Much simulator time is devoted to TR failures of all kinds on bigger aircraft because it is not usually safe to practice realistically on the real aircraft.

[Arm out the window]

2) An aggressive deceleration about 10 feet above the surface, use power to stop all ground motion (kind of turns into a sideways flare) and then a hovering auto (I like this one for landing on surfaces that won't allow sliding, like a swamp)


Good one, Paul - that's my favourite, and allows you to get little or no forward speed on touchdown. Beats a fast run-on any time in my book but takes a bit of finesse of course.

[rans6andrew]

er..... I don't understand. If you enter autorotation, it removes all of the torque from the main rotor and thus the tail rotor becomes largely redundant (just a bit of opposite due to drag in the gearbox). If you have a tail rotor failure why would you not just autorotate into any decent sized field? This has got to be a lower risk choice than trying to figure out the speed at which the slipstream fails to provide sufficient and then attempting to judge an approach to run on at just above said speed. Plus, autorotations should have been well practiced during training. I don't come at this with no previous, I have 86 hours in R22b.

Help me here.

[Rotate too late]

Help me here

Mate,
Tools in the box.....as NE said in the piece, your experience, feel for the aircraft and situational awareness will give you the information you need to make a decision. Then you just have to fly the bird to the ground!! Easy! :ok:

(OK the last bit was a joke !!)

[Tourist]

er..... I don't understand. If you enter autorotation, it removes all of the torque from the main rotor and thus the tail rotor becomes largely redundant (just a bit of opposite due to drag in the gearbox). If you have a tail rotor failure why would you not just autorotate into any decent sized field? This has got to be a lower risk choice than trying to figure out the speed at which the slipstream fails to provide sufficient and then attempting to judge an approach to run on at just above said speed. Plus, autorotations should have been well practiced during training. I don't come at this with no previous, I have 86 hours in R22b.

Help me here.
Autos are great in toy helicopters, not so much in the big ones.

[nigelh]

I think the point is that you may wish to fly ,say 5-10nm to an airport or flat ground and then set up an auto ...in which case you will have to be able to maintain speed and height . In a single i would find a nice place and auto but in my 109 i would fly to an airport or big empty road and do a running landing . ( If you had a passenger you could get them to slowly wind throttles off as you start to touch down ...maybe only if they were a pilot ..! )

[[email protected]]

Lonewolf50 - apparently the cause was an incorrectly torqued anti-flail mount on the TR drive shaft that came loose, slid down to the next TR bearing mount and jammed against it then gradually sliced its way through the TR driveshaft itself.

Rans6andrew - as Tourist says, an EOL in a 1 to 2 tonne helicopter (probably 1500 to 1800 ft per min RoD) is pretty straightforward with room for error - they are designed that way as they are single engine - there is enough energy in the rotor system to arrest the RoD and forward speed and still allow for cushioning the touchdown. Very few will train in their light singles at MAUM as it starts to make things trickier.

Now increase the mass to 10, 15 or 20 tonnes and the EOL (probably 2000 - 2500 ft per min RoD) would be a much dodgier bet, even in an Apache with undercarriage and stroking seats designed to absorb over 25 G of vertical impact. The margin for error would be very small and a powered running landing to a decent surface starts to look the more survivable option.

For those practicing some of these 'theoretical' techniques such as the zero speed landing - have you empirically assessed your practice pedal position as truly being representative of zero TR thrust? And do you really take the EOL to the ground with that fixed pedal position without cheating?

[nigelh]

I have done zero thrust on pedals with instructors and never felt it was really how it would be in real life . I think you would have to put in opposite pedal to get the real feel .

[Hughes500]

Crab

Even you can't stop the tr when flying so the only way to simulate is to progressively use more and more pedal with the student. So you arrive at the situation where you are landing with full right pedal ( for US machines ) otherwise how else do you teach ???????? . The technique I describe is similar to what the AH64 did.
Although most RFM's suggest an EOL if drive failure !

[[email protected]]

I should have added that small singles tend to have relatively small fins or stabilisers and skinny tail booms so there is very little surface area to provide anti-torque thrust.

Compare that to something like a an AS365 with a massive fin, vertical stabilisers and a thick tail boom.

Nigel - exactly my point - a neutral pedal position doesn't mean zero TR thrust.

[Lonewolf_50]

Lonewolf50 - apparently the cause was an incorrectly torqued anti-flail mount on the TR drive shaft that came loose, slid down to the next TR bearing mount and jammed against it then gradually sliced its way through the TR driveshaft itself.
Thanks.
Hadn't realized how robust the spec was on Apache.
even in an Apache with undercarriage and stroking seats designed to absorb over 25 G of vertical impact.

[MightyGem]

even in an Apache with undercarriage and stroking seats designed to absorb over 25 G of vertical impact.
The Apache landing gear/seat design was designed to meet impacts of up to 42ft/sec(2520ft/min) ROD and allow the crew to walk away.
http://citeseerx.ist.psu.edu/viewdoc/download?rep=rep1&type=pdf&doi=10.1.1.214.8052

A bit like doing an EOL in a Lynx and just not bothering to cushion the touchdown at the bottom. :eek:

[krypton_john]

Crab - good point re auto characteristics of medium to heavy twins vs light singles.

No idea about the AH64 but s/e AH1s and Hueys of course auto like butterflies! (as do 212s)

[Hughes500]

Crab

What do RFM's say about drive failure in the big stuff ? EOL? or run on landing ? Obviously it appears AH64 is best at run on as it seems that they don't do well in on EOL's

[[email protected]]

Hughes - I think it varies from type to type, The 212 is an EOL for a full drive failure whereas a 365 (massive fin) is a fast running landing.

The Lynx and Sea King were EOLs if you couldn't get a sensible speed for running landing.

The problem with the EOL in a bigger helo is trying to get to zero groundspeed - or you risk landing with forward speed but not straight in which case you could roll over.

There is certainly no magic bullet solution and I think we all fervently hope we are not put in the position of having to try.

Back in the 90's a Wessex had a TR drive shaft failure in North Wales - some colleagues of mine were involved - the failure happened at 60 kts and despite lowering the lever and pushing the nose down, they were unable to regain forward controlled flight and were faced with chopping the throttles in a spinning helicopter descending over a lake.

[MightyGem]

That will be this one. Unfortunately, three cadets in the back died in the accident.
https://www.youtube.com/watch?v=cDj6oYaYrR8

[The Cryptkeeper]

Hello all,

I was Nick's SQHI prior to this deployment on Herrick (I was in the process of leaving the mob at the time) so maybe I can shed some light on how TR drive failures are taught on the UK Apache fleet. It was most definitely a TR drive shaft failure (it severed mid way up the tail boom) - the TADS footage of the aircraft in flight from his wing is quite eerie with the tail rotor stationary!

In the simulator a TR drive shaft failure is given after setting up at a suitable height for autorotation with an airfield ahead - as soon as TR drive is lost the EPLs (engine power levers = throttles) are retarded to idle and the HP enters autorotation. In order to maximise the effect of the tail fin the teaching is to maintain a high run on speed. This is, of course, an ideal.
The terrain in and around the Helmand River is rough, dusty and full of wadis and other hazards. Plainly a high speed run on isn't an option!

I've flown multiple aircraft types in the AAC and I must say that one of the strengths of the training was instilling in aircrew the idea that there are many ways to achieve the same result whether that be gunnery, sensors or indeed emergencies. The FRCs tell you what you should do but, as was the case here, Nick's previous experience in Helmand and in the simulator led him and Charlie down a different path. I have to say that they handled it very well indeed and to do a low level (400ft agl) double engine off into a dusty, uneven desert environment from the front seat with minimal recent "hands on" was an exceptional feat!

To give those of you that fly smaller, lighter helicopters a flavour of performance in Afghanistan typical AUM on lift from Camp Bastion was around 19,000lbs including RCEFs (jettisonable external fuel tanks), stores booms carrying HF missiles and rockets, and around 550 rounds of 30mm carried internally in the magazine.

And the ruggedness and crashworthiness of AH is pretty awesome! :)

[Lonewolf_50]

Crab

What do RFM's say about drive failure in the big stuff ? EOL? or run on landing ? Obviously it appears AH64 is best at run on as it seems that they don't do well in on EOL's For a 20,000 pound helicopter, SH-60B, the procedures here are (https://quizlet.com/9661965/sh-60b-emergency-procedures-flash-cards/)close to what I remember. If you've got altitude and speed needed for an auto, an auto it is. If not, you're doing a form of a cut gun ... it can get squirrely. I recall from Blackhawk sim training that some forward momentum was desirable at the bottom, how much being situation dependent.


Don't have any of the old manuals so take my recall with a grain of salt.

[sycamore]

For those who may care to `search`,`Tail-rotor failure...have you had one....? `may provide some more info on t/r failures,that guys have had.Threads from about 2010 but still relevant.
The real `biggie` is a loss of tail rotor and gearbox,and the change in Cof G,to somewhere well in front of your feet....!

[MightyGem]

My former Chief Pilot had a TR drive failure in a Twin Squirrel many years ago. He managed together in back to the airfield and make a run on landing.
https://m.youtube.com/watch?v=z3oVw6_0MEU