Why wasn't the Airbus A330 designed to stop its fans (with some kind of 'handbrake')?

To clarify, I am specifically talking about the aircraft that flew Air Asia flight D7237.

During the flight the aircraft suffered from a fan blade separation event and the the result was that the engine became unbalanced.

This was the result of the incident the plane was able to continue flying for over an hour and a half with the "moderate" vibrations and able to land safely (where the passengers were rewarded with a $20 voucher)

It's my understanding that the engine was immediately shut down, and that the cause of the vibrations was the fanblades freely turning as the plane moved very fast through the air (similar to blowing very hard into a turned off household fan)

But to me, it seems the whole situation could have been avoided if there was simply some kind of "brake" or 'locking' feature that prevented the fanblades from turning freely in the wind.

I feel as though there wouldn't exist another scenario where a plane could possibly experience more severe, prolonged vibrations than a windmilling event, and the windmilling event itself could be prevented if there was simply a way to lock the fanblades from moving freely.

The reason I say A330s in particular is because I am not aware of the features of every other jetliner, although I am aware some propeller engines can feather their blade parallel to the direction of airflow to prevent this.

  • 6
    $\begingroup$ Weight, extra maintenance, low probability of actually using such a ststem...a long list of reasons exists as to why such a system was not implemented. $\endgroup$
    – acpilot
    Jul 20 '17 at 19:08
  • 1
    $\begingroup$ As you can see, windmilling did not impact the overall safety of the flight. Why add weight and cost for something that isn't needed? $\endgroup$
    – GdD
    Jul 20 '17 at 20:24
  • 4
    $\begingroup$ humorous alternative interpretation of the title when reading it on the hot network questions list: "Why wasn't the Airbus A330 designed to stop people who really like it?" $\endgroup$
    – Michael
    Jul 20 '17 at 21:31
  • 2
    $\begingroup$ You are probably underestimating what it would take to "stop" the fans by a large degree, but aside from that, the amount of drag caused by a "stopped" fan is very much bigger than a fan that is windmilling. The default scenario for ETOPS flights after an engine failure is that the failed engine will continue to windmill, and a lot of engine design work is done to try to make sure that will happen. That said, there has been at least one incident on a 747 where a failed engine did stop windmilling - but losing one engine out of four is less of a problem than losing one out of two. $\endgroup$
    – alephzero
    Jul 20 '17 at 22:50
  • 4
    $\begingroup$ There was an infamous incident with a 747 freighter flying from the far east to the UK (IIRC) where an engine failure caused vibration levels that annoyed the crew. So they made an unscheduled landing somewhere in the third world, and fixed the problem themselves by restraining the fan blades with the webbing used to hold down the cargo(!!!) That worked fine, except the increased drag meant they didn't have enough fuel to complete the flight, so they decided to make another unscheduled stop in Germany to refuel. The German ground crew took one look at the webbing, and impounded the plane... $\endgroup$
    – alephzero
    Jul 20 '17 at 22:56

Complexity and weight, what airlines hate most.

Has it been done before? Yup. A variant of the Convair B-36 Peacemaker was able to stop the windmilling when the jet engines were not needed (that variant had 6 piston engines and 4 jet engines).

enter image description here

The metal petals filling the inlet prevented the jets from turning in the wind as the B-36 cruised for long periods of time with them turned off. The petals retracted to allow the jet engines to be ignited for extra thrust during take-off and high speed dashes. A strut braces the nacelle against swinging from side-to-side.

Scaling this up for the big diameter engines for that very rare engine failure is not cost effective. Instead planes can survive, and are extensively tested for, vibration and flutter.

How rare is very rare?

By the 1970s, advancing technology had set the stage for two-engine, turbine-powered airplanes to safely exceed the 60-min operating restriction [away from the nearest airport]. The result was ETOPS, which began in 1985 with 120-min diversion authority and the requirement for an average engine in-flight shutdown (IFSD) rate of just 0.05 per 1,000 engine-hours. With 180-min ETOPS authority, which followed in 1988, an even more stringent reliability target of just 0.02 IFSDs per 1,000 engine-hours was specified (Boeing, 2003).

  • $\begingroup$ I'm trying to figure out where those petals retract. Do they go up inside that cone somehow? It doesn't look like they'd fit $\endgroup$
    – TomMcW
    Jul 20 '17 at 21:48
  • 2
    $\begingroup$ @TomMcW - here you go 😉 $\endgroup$
    – ymb1
    Jul 20 '17 at 21:53
  • $\begingroup$ Cool. Thx. Kind of ingenious, that $\endgroup$
    – TomMcW
    Jul 20 '17 at 21:55
  • $\begingroup$ @ymb1 The IFSD rates you quote are misleading here. Most IFSDs do not result in any significant vibration levels at all. The airworthiness certification requirements for the type of incident in the OP's scenario is "extremely rare" - which means demonstrating a reliability target of 0.000001 per 1000 engine-hours. (One might argue about how accurate some of the "demonstrations" have been historically, but that's a different question). $\endgroup$
    – alephzero
    Jul 20 '17 at 23:04
  • $\begingroup$ The B36 was designed to fly with non-failed engines shut down, probably on every mission flown. That is a different scenario from unscheduled engine failures. Other aircraft have used different strategies - for example the Nimrod SAR aircraft were designed to loiter with two of the four engines shut down and windmilling, and there was a common oil system between pairs of engines to make sure the windmilling engines were properly lubricated, and their own oil system hadn't cooled down to -80C when you wanted to restart all the engines after 8 or 10 hours loitering at high altitude $\endgroup$
    – alephzero
    Jul 20 '17 at 23:13

Based on general engineering principles, there are multiple reasons why such a locking device could be unhelpful at best, and dangerous at worst:

  • It would be one more system that adds weight and complexity.
  • It would take up additional room that isn't available, and making that room would affect the aerodynamics and the weight.
  • It would be one more system to check and maintain.
  • If it fails, does it fail in a way that causes a running engine to stop, thus creating a dangerous situation, or does it fail in a way that makes it no better than not having it at all?
  • Is it more reliable than the jet engine?
  • There are situations where windmilling is desirable, turning engaging the brake into one more thing the pilots have to decide to do. Is it worth the cognitive load on the pilots?

Engine failures of any kind are rare, and, as was the case with this incident, often quite survivable. It's not worth the design cost, material cost, weight cost, maintenance cost, or training cost to have a turbine locking device.

In this case, the only thing that a brake would have accomplished is reduction in vibration. With one out of two of the engines down, the pilots would still have brought the airplane in for an early landing, and not continued on to their planned destination.

While this flight was not under US jurisdiction, the Rolls Royce Trent 700 engines used on the aircraft still have to comply with FAA regulations, if for no other reason than that A330 aircraft are flown in the United States. In particular, 14 CFR Part 33 is titled "Airworthiness Standards: Aircraft Engines", and says the following in §33.74 Continued rotation:

If any of the engine main rotating systems continue to rotate after the engine is shutdown for any reason while in flight, and if means to prevent that continued rotation are not provided, then any continued rotation during the maximum period of flight, and in the flight conditions expected to occur with that engine inoperative, may not result in any condition described in §33.75(g)(2)(i) through (vi) of this part.

The points described in §33.75(g)(2)(i) through (vi) are:

(i) Non-containment of high-energy debris;

(ii) Concentration of toxic products in the engine bleed air intended for the cabin sufficient to incapacitate crew or passengers;

(iii) Significant thrust in the opposite direction to that commanded by the pilot;

(iv) Uncontrolled fire;

(v) Failure of the engine mount system leading to inadvertent engine separation;

(vi) Release of the propeller by the engine, if applicable;

Since I can not find any documentation to show that the Trent 700 has a rotor locking device, it must be designed to meet these regulations. Vibration is not listed as one of the hazardous conditions. It is true that vibration can be hazardous, but that's more for a running engine, and vibration monitoring systems exist to shut a malfunctioning engine down.

Other regulatory bodies no doubt have similar requirements. I picked the FAA since it is the one I am most familiar with.

  • $\begingroup$ "There are situations where windmilling is desirable", what conditions might that be? $\endgroup$
    – Ksery
    Jul 20 '17 at 21:12
  • 2
    $\begingroup$ @Ksery Windmilling is one way to restart a failed engine in flight if you don't have bleed air available. $\endgroup$
    – TomMcW
    Jul 20 '17 at 21:45
  • $\begingroup$ @Ksery Under certain circumstances, it's possible to windmill start a stopped turbine engine, similar to push-starting a car. In the case under discussion, you can't and don't want to do that, but in emergency situations where it doesn't look like there's engine damage, you might want to give it a shot. See Does windmill restart often work for airliner engines? $\endgroup$
    – Dranon
    Jul 20 '17 at 21:49
  • $\begingroup$ " Significant thrust in the opposite direction to that commanded by the pilot;" Am i correct in reading that its possible for an engine to produce thrust out of the front? $\endgroup$
    – Ksery
    Aug 7 '17 at 4:45
  • 1
    $\begingroup$ @Ksery For a jet, not out the front, no. Thrust reversers deflect gases from the rear. For a prop plane, some propellers have variable-pitch blades that can be rotated to produce reverse thrust. $\endgroup$
    – Dranon
    Aug 7 '17 at 13:38

Yes propellers can feather, a feature which is required for delivering thrust over a wide range of airspeeds. Fixed blade propellers had a problem with either delivering good take-off thrust, or limiting the maximum speed of the aircraft. Variable pitch propellers were invented before WW2 and widely implemented because of the limitations of fixed blade props.

Turbofans have a fan disk imside a shroud, which decelerates the air before it hits the fan blades. This setup can be optimised for cruise comditions, and still delivers good takeoff thrust. The jet engine core delivers part of the take-off thrust directly. The fan is fixed pitch because it can be - a variable pitch mechanism, or a brake for that matter, adds cost and complexity. Most present day airliners have these engines, including the A330.

However, the larger the fan in a bypass turbofan, the more efficient it is, and a large fixed pitch fan starts to have the same issues as a fixed pitch propeller. So now large bypass variable pitch turbofans are being developed, for instance the Rolls Royce Ultrafan.

enter image description here

The variable pitch fan is actually not a new idea, Turbomeca developed the Astafan in the 1960s.

Windwilling of any fan, variable pitch or not, is not bad per se:

  • A windmilling engine can be restarted, priority 1 in an undamaged engine.
  • Windmilling engines have less drag than non-rotating engines, and less asymmetric thrust of the remaining engine(s) to trim away.

If it can be demonstrated that a windmilling engine with a broken fan blade only causes mild discomfort and no further catastrophies, and that the incidence is very rare, then there is no actual problem to solve.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.