Ditching a modern airliner

[fireflybob]

Despite the Hudson event being one of only two controlled, intentional airline jet ditchings ever AFAIK there have been scores if not hundreds of overruns/excursions into water. A urprising number of airfields have runways that project into water at one or both ends. Life jackets are thus far more useful than most people imagine.

Isn't that the salient point that most ditchings these days are unpremeditated, ie runway overuns etc?
Traditional ditching drills come from the days when large piston engine aircraft were halfway across an ocean and were unable to maintain height due engine failure/shutdowns and then had around 30 minutes to prepare the cabin and plan for some sort of controlled ditching. It was probably a bit before my time but I believe the ocean stations vessels located in the Atlantic could lay a flarepath on the sea within 20 minutes or so. I also believe one of the findings of the Hudson report was that there needed to be an abbreviated ditching drill for unpremeditated ditchings which just covered the essentials.
Whilst practice in a simulator may be useful I doubt modern visual systems can truly represent a swell on the ocean.
As regards lifejackets as a passenger I'd still like to have one just in case!

[tdracer]

Originally Posted by Bergerie1

I used to practise deadstick landings in the 747 simulator when there was spare time. Great fun and a good exercise in energy management as a previous poster has said. You could sideslip the sim just like a Tiger Moth to shed excess height. But does anyone on this thread know how much sideslip can be tolerated by the engine pods? And what are the sideslip limits on modern jet transports?

I don't know what the sideslip limits are, but we did a flight test on the 747-8 where the pilot held a 20 degree yaw for 30 seconds, then repeated in for 30 seconds in the opposite yaw. It was pretty obvious the aircraft wasn't happy about it but it did it just fine (and I'm quite sure someone looked at the design limits for yaw before the condition was approved). I was also very, very impressed with the pilot - the first direction he had trouble holding the condition with lots of corrections, but he learned so fast that when we went the other way he held it rock steady with minimal corrections.
Years ago I had a co-worker who thought we should save the cost and weight of the rafts and life jackets as there'd never been a successful ditching of a commercial jetliner. When I first saw the video of the Ethiopian 767 ditching I thought he might have a point - didn't find out until later that the pilot was having to fight off the hijacker while trying to ditch - talk about your bad day at the office.
The problem you'll always run into for simulator training is that it's a finite resource and the regulators don't want to add stuff to the requirements if the probability is less than ~1 per 10 million flights (dead stick landing falls into that category)

[metro301]

Originally Posted by Centaurus

The B767 captain of Gimli glider fame did an exceptional job of forced landing all flaps up on a 7000 ft runway. Afterwards he said if only Air Canada had given him just one practice simulator forced landing from height he would have been much more confident when he did he real one.

Captain Pearson was an experienced glider pilot
https://en.wikipedia.org/wiki/Gimli_Glider

[lomapaseo]

The problem you'll always run into for simulator training is that it's a finite resource and the regulators don't want to add stuff to the requirements if the probability is less than ~1 per 10 million flights (dead stick landing falls into that category)

Don't blame it on the regulators, it's the operators who have to pay for the off-line testing that count the stats to the nearest decmil

[Winemaker]

Re simulator training, since the simulator can't generate more than one g how does the experience translate to actual flying? As an ex formula car racer I've experienced more than 2 g lateral load for extended periods of time; a simulator can't generate the head/neck loading and wind impact of the real deal. When flying the g load is generally vertical through the seat so I guess very short bumps above 1 g could be made. What I'm asking, as SLF, is how effective the simulator is at representing real flying conditions.

[oxenos]

I was on the same conversion course (from Shackleton to Nimrod in 1970 ) as the pilot who ditched the Moray Forth Nimrod 25 years later.On the course we were shown film of Nimrod model ditching trials .
The model, which was of scale strength, and fitted with g recorders, was projected into a large water tank at diferent pitch attitudes, yaw angles, and with different "sea states" It showed that at a reasonable angle and sea state the ditching would be survivable. It also showed that the fuselage would break at a point in line with the trailing edge.
The day after the ditching there on the front page of the papers was an aerial photo of the Nimrod afloat, and broken exactly as predicted.
As I recall (long time ago now) there was also film of other military model ditching trials. The Beverley looked like a duck taking to water.
It makes sense to do such trials for a maritime aircraft, and perhaps for other military aircraft, but I am not aware that civilian aircraft are similarly tested.

[Deadstick126]

Originally Posted by ATC Watcher

Great Story fdr !

Back to the question : Jeff Skiles the co pilot of the US Hudson flight makes an excellent presentation explaining the various and numerous lucks they had that day. ( outside of hitting the birds that is !) and basically you cannot train for luck.
A photo of the aircraft the next morning , with ice plates everywhere on the Hudson and lots of ferries crossing in the background shows already 2 of those lucks.

I had the pleasure of meeting Jeff at a party at Oshkosh a couple of years ago. Everyone was still asking about Cactus, and he was politely answering the same questions for the umpteenth time. He had been selected to fly "Fifi", at the time the only remaining flying B-29, into Oshkosh for the show. When I asked him what it was like to fly"Fifi", his face lit up and he went into an animated description of landing the beast and having only differential braking for steering control, among other hilarious aspects involved in flying to old bird. I'd love to run into him this year and hear about what treasure he flew in this time.

[captainsmiffy]

Just ditched the 380 in the sim.....��

[tdracer]

Originally Posted by lomapaseo

Don't blame it on the regulators, it's the operators who have to pay for the off-line testing that count the stats to the nearest decmil

But the regulators have to approve the operators simulator syllabus. If they mandate a condition must be part of the syllabus, the operator has no choice (engine failure at V1 being an example). With rare exception, the regulators won't mandate training for anything that happens less than once per 10 million flights (ran into that on another issue - it's not that hard to control but the regulators said it was catastrophic because they don't train for it - we said so train for it - they said it doesn't happen often enough to justify the simulator time....)

BTW, an ocean ditching would be easier to 'get right' than a dead stick on a runway since when ditching you basically have an infinite runway length and don't need to worry about getting the touchdown point right. But the sea conditions make the ditching less likely to be successful.

[dogcharlietree]

With all due respect, by flying an approach to ditch has very little to do with visuals (in a simulator).
In my training, we used to practice in the real aircraft by using an arbitrary figure of say 5,000ft agl/amsl as our "sea level".
The main ingredients of a "successful ditching" are wings level, minimum rate of descent (less than 100fpm) and minimum speed just above stall.
If you try that in a simulator, you will see how difficult it is to achieve ALL three conditions. Try it yourself!

And again and with all due respect, again, a deadstick approach is NOTHING like a ditching approach (if it's to be successful).

[Bend alot]

Just a few points on airliners.

Airliners prior to the B747 had pretty short range, so as a result less time over water and often more "back up engines" than twin airliners.
The B747 obviously had a large range so flight time over water increased and in its early days ate lots of engines, with the exception of flying through volcanic ash 4 engines seemed enough.

We now have big twins with massive range, and a change from hub the to hub model to the point to point model. The introduction of ETOPS and various extensions. All of this leading to many more flights spending more flight time over water by taking more direct routes.

So over time we have reduced the number of "back up" engines to one, and dramatically increased the the time spent flying over water.

Being a LAME I understand a bit. Having both engines on a big twin failing at the same time or within the ETOPS limit (has new name I think) is very slim. The reasons could be a number of individual or the Swiss cheese reasons such as crew, engineering, software (upgrades) contamination and commercial pressures.

What has been highlighted over the last few years is un-contained engine failures - on what I class as modern airlines. Recently we have been lucky with these failures, except the poor lady from flight SW1380.

The secondary damage from these un-contained engine failures (QF 32) is what is likely to cause both engines of a big twin to stop producing thrust. Loss of control systems, pressurisation, fuel and electrical items in certain combinations can increase the requirement of ditching.

Good thing aircraft are designed not to have un-contained engine failures.

As for the life jackets, a few visuals need to be shown on consequences of inflating them while in the cabin, the current brief has the same impact as leave your hand luggage behind.

[Sailvi767]

Originally Posted by oldchina

After 17 posts it's strange that no-one has mentioned that Sully's A320 had a ditching checklist.
Problem being that it was written for a descent from altitude and the FO had no time to finish it.

Which is typical of Airbus checklists. They take entirely to much time to run and often have pitfalls you need to be aware of. Compare the engine out procedures on a 767 to the A330 including the overweight landing section. One is 2 checklists, the other War and Peace!

[Centaurus]

It is much more important to invest into teaching staff how not to end up in situation like Gimli

Which is about the same value as saying "I will not teach you how to swim as it is too dangerous and you may drown. Instead I will teach you not to go near the water.."

[Centaurus]

The main ingredients of a "successful ditching" are wings level, minimum rate of descent (less than 100fpm) and minimum speed just above stall.
If you try that in a simulator, you will see how difficult it is to achieve ALL three conditions. Try it yourself!

Most airline pilots would prefer to take the professional advice of the manufacturer which is somewhat different to that which you advocate.
For example from the Boeing 737 QRH (selected edits for brevity).
"Plan a flaps 40 landing unless another configuration is needed. Select VREF 40 (my note: not minimum speed just above the stall)
Ditching Final. Maintain speed at VREF. Flare the airplane to achieve the minimum rate of descent at touchdown.
Maintain 200-300fpm rate of descent until the start of the flare."

It is certainly not difficult to achieve all of the conditions above providing you are competent at basic instrument flying. After all, apart from the instrument flying bit, a PPL should be able to do that for a short field landing in a Cessna 172. A properly maintained Level D Full Flight simulator has that fidelity. It is the pilot that lacks the fidelity if he can't fly that in a simulator.

[dogcharlietree]

Originally Posted by Centaurus

Select VREF 40 (my note: not minimum speed just above the stall)

I can appreciate what the airlines/manufacturers sprout. They have to to protect their butts.
Please remember that the main problem in a ditching is impact forces.
So doesn't the old equation E=mv2 come into play. If you double the speed at impact then you have four time the impact forces.
The airlines say approach at Vref (1.3 Vs), so as my mathematics is not that good, hypothetically if you came back to Vs (which you wouldn't) aren't you nearly halving the impact forces. (Donning flame jacket now).
Also, in my opening thread, I referred to aircraft that had a significant uncontrollable fire onboard. I know I'd be doing some lateral thinking in these circumstances, like the RAF pilot of the Nimrod, mentioned above. And he had the runway in sight. Good decision!

[cats_five]

Originally Posted by metro301

Captain Pearson was an experienced glider pilot
https://en.wikipedia.org/wiki/Gimli_Glider

So was sully, so apparently was Robert piche https://en.m.wikipedia.org/wiki/Air_Transat_Flight_236

[EDLB]

The odds are not that bad for forced landings.

https://en.m.wikipedia.org/wiki/List...quired_gliding

I would rather make a flare at Vref 40 according to the flight manual parallel to the swell and dissipate the remaining energy in the flare. The ocean runway is usually large in every direction so wind direction is of not much concern. If you fly in close to stall speed you might stall prematurely. Without working donkeys often you have lost most of the hydraulics. That the success rate of glider pilots on forced landings is high, should not be a surprise.

[nonsense]

Originally Posted by dogcharlietree

I can appreciate what the airlines/manufacturers sprout. They have to to protect their butts.
Please remember that the main problem in a ditching is impact forces.
So doesn't the old equation E=mv2 come into play.

KE = 1/2 mv^2

Originally Posted by dogcharlietree

If you double the speed at impact then you have four time the impact forces.

Double velocity, energy to be dissipated goes up by a factor of four. Without knowing a lot more, this only tells you the impact forces will be higher, not how much higher.

Originally Posted by dogcharlietree

The airlines say approach at Vref (1.3 Vs), so as my mathematics is not that good, hypothetically if you came back to Vs (which you wouldn't) aren't you nearly halving the impact forces. (Donning flame jacket now).

1.3^2 = 1.69 which is close enough to double for the back of a napkin.

[Vessbot]

You guys guys are talking past each other, approach speed is different from touchdown speed.

Edit: not to mention that with engines out, your steady state descent rate on approach will be at least 2-3 times normal, so the speed loss during the time it takes to flare out of that, will be even greater than normal.

[601]

Select VREF 40

Maintain 200-300fpm rate of descent until the start of the flare.

approach speed is different from touchdown speed.

By the time you flare and touchdown, how much speed will you wash off and what will be your rate of decent at touchdown.
My guess speed a lot slower than VREF 40 and decent a lot less that 200-300fpm.