The BWB Airliner revisited

[steelpillow]

The archetype BWB airliner, which everybody clings to like tigers, is of a tailless compound or cranked near-delta planform of moderate aspect ratio, with multiple engines mounted above and behind. No tailfin as such, but add winglets to taste. It looks so cool, it is both aerodynamically and structurally efficient. But it has real problems, revealed when Boeing submitted a design study to the FAA.

Any aeroplane must balance about its centre of lift or CL (not quite true, but close enough). The CL is a vertical line and can be surprisingly far forward, as the centre section does most of the lifting and its CL is around the mean quarter-chord point (i.e. not far from the leading edge). Putting the engines at the back needs some counterbalance at the front. A BWB struggles to find such a counterbalance; the cockpit, avionics and nosewheel alone are not nearly enough. The conventional solution (such as on the Sud Aviation Caravelle) is a lengthened forward fuselage, but this requires additional tail fin area, something we have just done away with. And putting say all the fuel or cargo or passengers up there means that an empty plane will be out of balance; only so much can be done.

In the engine-out condition, the fin of a conventional plane has a long lever arm to counter the swinging action of the remaining engine/s out to one side. This causes extra drag, perhaps 5% of the lost thrust so that is manageable. But a tailless plane must use a drag rudder - a drag devices at the wing tip, out beyond the working engine/s, to counter the swing. Unless the engine/s are close to the centre line, the extra drag will be around 20-40% of the lost thrust. So each engine must have far more reserve power than a conventional plane, and that means greater weight and cost. Although the weight of the fin is saved, the wings must instead be strengthened to take the increased drag forces, to there is little net gain there.

So here is my idea of how to overcome these problems and build a practical BWB airliner. The only question that remains is, would it end up any more efficient than a conventional toothpaste tube after all? I have added split trailing-edge flaps inboard, because such sleek tailless machines tend to float above the runway and refuse to slow down. I like to think they will be very much needed.



Most airliners put the engines under the leading edge of the wing for very good reason. They receive clean airflow at all angles of attack, and do not interact with the lifting flow above; on the approach to the stall, the engines are sure to deliver of their best. You could put them above the leading edge, but then the clean efficiency of the wing suffers and mechanics find them hard to service without special gear (as they would with the archetype dream). This does require an undercarriage as long and heavy as normal, so the theoretical weight reduction is eaten into. But it's better than falling out of the sky. It also helps greatly with the problem of balance.

Balance is further addressed here by sweeping the wings back a little more and reducing the root chord, so that the lift from the outer sections moves the CL aft to meet the new, forward CG. The downside is that this limits the width of the blending between body and wing, and in consequence makes the the wing roots thinner, increasing their weight.

Engine-out asymmetry is partly overcome by adding split drag rudders to the winglets, giving them maximum lever arm. Were they the sole rudder mechanism, full compensation would require one to create drag equal to minimal cruise thrust, with the engine supplying three times that. That is not realistic, so a modest central fin is still required to make up the difference and provide sufficient control authority. More claimed efficiency gains vanish like smoke in the wind. Still, the drag rudders will provide efficient control during cruise. It is possibly not an optimal solution, but it well illustrates the designer's dilemma.

Passenger evacuation is catered for by the limited root chord, allowing doors next to the leading and trailing edges. The rear doors drop down and the back edge is angled out, to direct a long inflatable slide out and back over the curve of the trailing edge. Short walkways slope down to the LE doors, set weill inboard of the engines. Additional doors are provided for the forward passengers and aircrew.

One more feature to mention is the drooped extension to the outer leading edge. This has been found down the years to help in stability and handling throughout the speed range, with little if any drag penalty. It is important here because, without a tailplane, the tailless wing must be inherently better behaved than its tailed cousins.

[PR19_Kit]

With the addition of the flaps, won't you need a pair of canards to help trim them out? The elevons won't be far aft enough to give a large enough trim effect with that wing shape.

That's why deltas don't usually have flaps. Mind you, some of them don't really need them, the Vulcan being one.

[steelpillow]

With the addition of the flaps, won't you need a pair of canards to help trim them out? The elevons won't be far aft enough to give a large enough trim effect with that wing shape.

That's why deltas don't usually have flaps. Mind you, some of them don't really need them, the Vulcan being one.

That's why they are split flaps, so they act more like airbrakes than lift augmentors. They are primarily there to kill the speed, not prolong the float. But the CL does tend to move forward at high AoA, so a little lift augmentation actually offloads the elevons. An alternative would be to make the tip finlets and drag rudders a lot bigger, but this solution has no drag penalty in cruise.

And yes, Roy Chadwick did it with high area/low loading. But he still had to go back and add outboard leading-edge droop

[Spino]

Now that's an idea!  Reminds me of a military cargo aircraft preliminary design I worked on, had to rule out BWB early-on due to a low cargo floor requirement.  If only it had been a passenger or freight aircraft where that sort of thing is less of an issue.