32
$\begingroup$

I'm talking about this sort of thing (this is the fictional airplane from Casino Royale), rather than two (or more) engines per wing in individual nacelles:

enter image description here

If we ignore the weird 'drop tanks' on the outer pylons, what makes this design inferior to one with individual nacelles? I can think of some highly speculative pros and cons for both layouts:

Pros for shared nacelles:

  • Shared nacelle and pylon potentially means less frontal and wetted area per engine, which could maybe reduce overall drag.
  • Overall, this scheme seems to allow the 'center of thrust' for all the engines on one wing to come closer to the centerline of the aircraft. In some designs like the A380, where a massive vertical fin is needed to maintain yaw authority in an engine(s)-out situation, it seems like this could save quite a bit of weight and drag by reducing the size of the vertical fin.

Cons for shared nacelles:

  • Obviously there is an increased risk that an uncontained engine failure like the one on an A380 a few years ago could cause cascading failures in its 'neighbor'. I'm not sure how to quantify this risk, but it seems significant.
  • Additionally, a structural failure in a pylon/nacelle (extremely rare but it does happen IIRC) will affect two engines instead of one.
  • Maintenance will be probably be more difficult, since you'll have to mess with two engines just to get at one.
  • I'd expect that due to the structural requirements of supporting two engines on one pylon, you probably wouldn't save much weight and you might actually end up with a heavier aircraft, all other things being equal.
  • Maybe you'd lose some efficiency in the interaction between the exhausts of both engines in a nacelle?
  • It might turn out that you can't fit two large high-bypass turbofans into one nacelle without choosing between flow separation issues and greatly increased frontal area.

This is all layman speculation, though. In reality, from an engineer's point of view, is the choice of one layout over the other obvious? Why or why not?

$\endgroup$
  • 1
    $\begingroup$ It appears that you have answered your question in itself. $\endgroup$ – Farhan Feb 3 '15 at 21:57
  • 7
    $\begingroup$ I wouldn't say never. B52s have been rocking siamesed engines for quite some time: fas.org/nuke/guide/usa/bomber/b52-981974c.jpg and the B36 was doing it even earlier i00.i.aliimg.com/wsphoto/v3/343762004/… EDIT: must admit, those are probably not High bypass turbofans on either bomber, though the B52 may have been upgraded over it's lifespan $\endgroup$ – FreeMan Feb 3 '15 at 21:58
  • 6
    $\begingroup$ I think there will be issues with the mass flow at take off. At take-off the engines are sucking in air from a greater area than their frontal surface. If two engines are next to each other these two areas will overlap $\endgroup$ – ROIMaison Feb 3 '15 at 22:14
  • 1
    $\begingroup$ Slightly off topic, but the mounting of both the engines and the pods are oddly off center. It's really rather annoying now that I've seen it. $\endgroup$ – FreeMan Feb 4 '15 at 2:37
  • 1
    $\begingroup$ The VC-10 seems to prove your rule. upload.wikimedia.org/wikipedia/commons/1/1f/… $\endgroup$ – Hugh Feb 4 '15 at 4:41
34
$\begingroup$

Short answer

It's the vastly different flow conditions from static to cruise which demand a separate placing of high-bypass-ratio jet engines. They would produce less thrust and more drag when paired.

Why were there paired engines at all?

The early jets had their engines mounted directly beneath or in the wing, and comparisons between separately mounted and twinned engines showed a slight advantage for the latter due to lower wetted surface area and less impact on the wing.

Arado built two four-engined prototypes of their Ar-234 jet, one with separate engines (V6, see directly below) and one with paired engines (V8, further down). The V8 became the prototype of the C version of the Ar-234.

Arado 234 V6

Arado 234 V8

However, with the increasing airflow of high-bypass-ratio engines the interference between both will turn pairing into a disadvantage. In cruise, only the central stream tube flowing towards the engine will be ingested, and the rest will spill over the intake lip. Placing a second engine directly next to the first will block the flow of spilled air on that side and will increase spill flow on the opposite side. This will most likely cause massive separation there if the intake is not heavily modified, leading to a noticeable drag increase. Also, the now asymmetric flow in the intake would reduce the efficiency of the fan - it needs very homogenous flow over the full cross section in today's engines.

Conversely, at low speed the engine sucks in air from all around and will face competition from a second engine, such that both will not be able to ingest as much air as when mounted separately. The consequence of pairing would be reduced thrust during take-off.

The initial disadvantage of separate engines, their collective impact on wing aerodynamics, is now greatly reduced by mounting them on pylons, so they are ahead and below the wing.

$\endgroup$
  • 1
    $\begingroup$ How on earth did you know about these planes. Don't tell me it was one of those ... oh yeah, those Arado planes... $\endgroup$ – Burhan Khalid Jan 7 '16 at 5:07
  • 3
    $\begingroup$ @BurhanKhalid: Easy. My father had a copy of William Green's Warplanes of the Third Reich, and I read it front to back, several times. Like I did with Bill Gunston's Encyclopedia of Russian Aircraft, which I bought myself. $\endgroup$ – Peter Kämpf Mar 25 '16 at 23:01
  • $\begingroup$ The reduced thrust at low speeds would seem to not only affect take-off (which I suppose can be managed, if the desire to do so is there), but also on go-around, which is... am I allowed to say "harder"? The aircraft is already in a high-drag configuration then (lots of flaps, possibly spoilers and friends, landing gear down, ...) and is slowing down for landing, and then you want as much power as possible as quickly as possible to climb back out. That feels like it would, at best, require some pretty serious changes to at least current aircraft designs. $\endgroup$ – a CVn Nov 8 '17 at 12:13
  • $\begingroup$ So would it be better to simply use one huge engine instead of two smaller ones? $\endgroup$ – dalearn Jan 28 '18 at 15:28
  • $\begingroup$ @dalearn: Generally yes. Only when the available engines are too small will more than one be placed on each wing. $\endgroup$ – Peter Kämpf Jan 28 '18 at 16:42
14
$\begingroup$

A big part of this is probably related to maintenance, as you observed. Currently, engines are generally taken straight off of the pylon. If they were paired on a pylon, They would either have to be removed together (making them essentially half as reliable) or attached in a different way.

With modern jet engines, only the largest aircraft make sense to have 4 engines, because two-engine aircraft are more efficient. This means planes like the 747 and A380 would be candidates for this design. There is actually a structural benefit of placing an engine further out on the wing. When in flight, the weight of the engine helps to relieve the bending moment on the wing. As you observed, this makes engine-out control more difficult. The placement of two engines on a wing is a compromise between structure and engine-out control.

This would also affect the amount of power the plane needs. Twin-engine aircraft must have enough power to take off if an engine fails after V1. This means that they need to continue a takeoff on 50% power. Aircraft with four engines must meet the same requirement, but this means they need to continue a takeoff with 75% power. However, pairing the engines makes it much more likely that a failure in one will affect the other. This means that the plane may be required to fly on 50% power, which will make the plane even less efficient.

Another potential issue is reverse thrust. Currently, engines can use locations around the circumference of the nacelle to eject this air. With the engines combined, each one would only have part of the circumference, so this could create flow issues.

Cases of failed engines or thrust reversers would put torque on the pylon, requiring additional weight for strength.

There could be some benefit of combining engine systems, but it would be at the cost of reduced redundancy.

This would reduce the area of the outside of the nacelle, but increase the frontal area. Considering two 120-inch diameter engines, combining them by connecting straight across on top and bottom, you decrease the perimeter from 750 inches to 615 inches, but you increase the frontal area from 22600 in^2 to 25700 in^2. Bringing the profile in between the engines (like in the photo above), will decrease the frontal area, but also increase the perimeter.

$\endgroup$
  • 1
    $\begingroup$ Nice point the reverse thrust, I missed it! $\endgroup$ – Trebia Project. Feb 3 '15 at 22:19
  • $\begingroup$ Why not use one big reverser across the backs of both engines? $\endgroup$ – Sean Mar 26 '18 at 21:38
4
$\begingroup$

For an engine to be efficient from a propulsive point of view, there is the need to "move a high quantity of flow with the same energy". Which means, for a given core create the bigger fan possible. So, from propulsion perspective, that design is less efficient than the A380 like configuration. You can see here more information about propulsive efficiency.

There is no so much saving in wetted area, but there will be some interference between both engines in the exhaust reduction on top the efficiency. Also the junction between both engines will create a grow of boundary layer, making the junction more draggy. I can't quantify the total drag, but is not clear for me that netly the drag saving in wet surface compensates or not the increase in viscous (boundary layer) increase in drag and the propulsion integration.

An engine is basically an aspiration machine which tries to absorb the flow arround the engine, this will be less important at cruise, but at take off both engines will compete for the air around, making them to produce less thrust. Critical engine failure take off is usually a dimensioning condition for engine size... so we have less efficiency in the most critical condition. From a global airplane perspective this makes the engine more overdimensioned.

Your point about the vertical fin is correct.

Concerning the risk of uncontained engine failure, the engines are usually design to avoid this situation. I do love this video about the test made.

In terms of structure, take into accoutn that in the condition of one engine failure you will need to include yaw momentun of having one engine operationally working and the other not. So very likely the pylon will be heavier that 2 pylons (although less draggy).

Also, this configuration doesn't make any sense, is usually cheaper to include a trim tank than using external, draggy, fuel pods.

As a example that the trade-off joining engines are not really making the business just look to all blended wings airplanes, all of them are having their engines on the upper part of the airplane separated between them (like X48).

As general rule I would say "fewer engines as bigger as possible".

$\endgroup$
  • $\begingroup$ From the video: "Just imagine what would happen if anything threatened the integrity of one of those fuel tanks, like a catastrophic engine failure that sends shards of metal tearing through them. That's the nightmare that haunts Rolls Royce engineers and they're about to destroy an engine to make sure it never occurs." Oops. Imagine what could happen indeed. QF32 was very fortunate that it ended as well as it did. $\endgroup$ – reirab Feb 4 '15 at 16:51
3
$\begingroup$

Joining engines has been attempted in the past in the straight-turbojet days -- the Iluyshin Il-62 and the Vickers VC-10 used this configuration, and so did the Lockheed JetStar business jet; this was done due to the poor thrust output of early turbojets. Modern turbofans crank out enough thrust that you don't need to pair up engines -- in fact, one proposal was floated for the USAF to replace the paired turbojets on the B-52's engine pylons with single RB-211 turbofans (thus turning an eight-turbojet-engine bomber into a four-turbofan-engine one). This was rejected due to the upfront costs; however, the fuel savings would have been quite significant due to both the ability to use half as many engines and the improved specific fuel consumption of the turbofans.

Another issue with paired engines on a pylon is safety -- LOT 5055 demonstrated this in tragic fashion when one of its Soloviev D-30 turbofans suffered an uncontained failure, causing an engine fire and severe damage to the other engine on that side; had the Il-62 used a more traditional layout, such an incident would have been much less problematic as the fire and damage pattern would have been much more confined.

$\endgroup$
3
$\begingroup$

Some of what I have read about the "re-engineing discussion" on the B-52 is not so much on the merits of updating to more modern engines (better fuel economy, spare parts availability, etc.) but on the implications of "rudder authority".

If the BUFF went from 8 engines to 4 then the implications of a loss of 1/4 or 2/4 engines on the same side would implicate that the current B-52 rudder and horizontal stabilizer might not be adequate to land the plane safely in a 3/4 or 2/4 engine situation. To wit: replacing eight engines with four might require a huge (expensive) re-engineering of the rudder and vertical stabilizer.

$\endgroup$
2
$\begingroup$

Pairing engines like that makes it more likely that the failure of one engine will propagate to the other. This has happened on, for example, the F-18 where one engine had an uncontained failure, resulting in bits of engine (I think they were turbine blades) destroying the other.

Also, when you've got wing-mounted engines, spreading them out reduces wing bending moments, which tends to reduce wing weight.

$\endgroup$

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.