Alternate Design to the Douglas DC-1

[Dynoman]

In a now famous letter sent from Transcontinental and Western Airlines (TWA) executive Jack Frye to Donald Douglas of the Douglas Aircraft Corporation (provided below), Frye attempted to buy the modern Boeing 247, which was a state-of-the-art airline design that ushered in all metal construction, retractable landing gear, and NACA designed cowlings, however United Airlines secured a deal with Boeing that would tie up production of the B-247 until 60 aircraft were built. This gave United a competitive edge over all of the other airlines who were still flying the biplane Curtiss Condors (American Airlines) and Boeing B-80 and the slow Fokker and Ford trimotors (Eastern, American and TWA). The letter from Frye requested a trimotor design monoplane that could seat 12 passengers and cruise in excess of 145 mph. Douglas went to work on designing an aircraft that could meet Fryes expectations, but instead developed a design that would go on to become the legendary DC-3. (The DC-1 being a prototype of the DC-2 and its DST version, the DC-3)

As a fun exercise, using historical data and references, to design an aircraft that meets or exceeds Jack Frye's request for an airliner that can compete against the Boeing 247 and possibly even to exceed the performance of the DC-1/D-2 using era available technologies. This is strictly a fun exercise in conceptual and preliminary design to provide the opportunity to converse, research and design a classic airliner of the early 1930s, an era know for it's luxury airline travel and pioneering aviation achievements.

 

[riggerrob]

Can we start by making it a high wing and installing a nosewheel?
At "first guess" the nose wheel will have to be almost as big as the main wheels to handle soft grass airfields.

 

[Archibald]

Something like the Ju-252 maybe ? But American of course...

Ju-252 - Recherche Google

 

[Dynoman]

Can we start by making it a high wing and installing a nosewheel?

I agree, I think that the elliptical lift distribution of a high wing will allow for a greater useful load and a nosewheel will increase steerability during taxi and reduce the chances of ground looping. Waldo Waterman's 1929 tailless Whatsit successfully used a steerable nosewheel in 1929.

 

[Dynoman]

Ju-252 is a good configuration. The low wing supports shorter main gear and as a result less gear weight. A high wing can have the gear stored in the fuselage or in fairings along the side of the fuselage.

With a high wing aircraft with the wings engines underslung beneath the wing the thrust line can be matched to the longitudinal axis of the aircraft, increasing longitudinal stability.

 

[Dynoman]

The landing gear design could be incorporated in to the engine nacelles. On a high wing, such as a Fokker XX, or for a low wing similar to the DC-2.

 

[Dynoman]

NACA tests NACA cowlings on Fokker Trimotor wing mounted nacelles found that the separation of air between the undersurface of the wing and the nacelle caused excessive drag. A recommendation would be to fair the engine cowling into the lower wing surface.

Engineering Science and the Development of the NACA Low-Drag Engine Cowling

The comparative speed trials proved extremely disappointing. Separate tests on the individual nacelles showed that cowling the Fokker's nose engine gave approximately the improved performance we expected. Cowling the wing nacelles, however, gave, no improvement in performance at all. This was strange, because the wind-tunnel tests had already demonstrated convincingly that one could obtain much greater improvement with a cowled nacelle than with a cowled engine in front of a large fuselage. Some of us started to wonder how the position of the nacelle with respect to the wing might affect drag.

 

[Dynoman]

An arrange similar to the bottom drawing would help to minimize drag around the engine for both high or low wing mountings,

 

[UpForce]

If this is supposed to be a Douglas design (I assume as much as the starting point is Frye's letter to Douglas specifically) I wonder whether a trimotor arrangement could've ever happened? I'm not one to get into the historical weeds (a lot) but looked at other contemporary Douglas designs and when it came to larger sized projects of theirs they used Wright R-1820 Cyclones a lot, almost exclusively it seems.

If that's your preferred engine, then two of those will always satisfy Frye's requirement. OK, they come just a little bit short but that brings up another question. One great advantage of the DC-1's design was said to be its variable propellers. Being only superficially familiar with radials, was there something specific to Cyclones that enabled this or could it have been done with Frye proposed Wasps as well?

 

[riggerrob]

An arrange similar to the bottom drawing would help to minimize drag around the engine for both high or low wing mountings,

The Curtiss cowling was also applied to many WW2 single-engined fighters. The British Royal Navy Fleet Air Arm wired shut the upper cooling flaps - on their F4U Corsairs to improve visibility during landings. The later Grumman F8F Bearcat routed its cooling air and exhaust out the sides of the cowlings (ala. FW 190 and Hawker Sea Fury).

 

[Dynoman]

UpForce you are right that the twin engine monoplane became the dominant configuration for many years for both Douglas and many other airline manufacturers. The increased HP and reliability of engines resulted in two engine configurations, although some trimotors were developed after the DC-1, such as the Junkers Ju-52, Fokker FXX, and the Stinson Model A.

Initially, the concept of this design exercise was to make an era related design that was true to the technology of the time and not necessarily a Douglas focused design. however it would make more sense to look at Douglas, their manufacturing techniques, heritage, and capabilities as the letter was directed at DAC. This of course will make the effort more challenging.

I'll try and find a 1932 description of Douglas' engineering and manufacturing capabilities and create a framework for making the design.

 

[UpForce]

I'm fine either way, Douglas or not. Generally speaking sometimes a more restrictive framework yields more imaginative solutions and I tend to fall on that side of the argument.

My general impression - which is not terribly well founded - about the DC-1 is that it was quite a departure from other Douglas designs. Many of their models up to that point seem at least somewhat agricultural, impractical and ungainly in comparison. It might be that during just those years a new engineering elegance was being enabled and TWA's predicament was mutually serendipitous for those two companies. On the other hand I can't help thinking that by going by Douglas' other designs it will be surprisingly hard to justify coming up with something as elegant as they actually did.

I expect there's no shortage of material on the DC series of planes. The trouble might be finding the best, authoritative sources among these; again, the limitations of Douglas' operation may well be the most instructive. If they could not have, for instance, added a variable pitch propeller function to the Wasps even if they wanted to, then I feel any speculation about an alternative design that doesn't adhere to that reality becomes entirely fictitious. Of course this just applies to my sensibilities and I'm not only appreciative of other designs but also other methodologies.

 

[Dynoman]

UpForce, it appears that the trimotor request from TWA, as per the letter's specifications from Frye, were based on pilot and airline executive questionnaires after the Knute Rockne accident. Frye and his technical consultant, Charles Lindbergh (of Atlantic crossing fame) thought that an engine failure of a twin engine aircraft after takeoff would be dangerous and thought that a trimotor would be a safer design. However, they told Douglas' designer, Arthur Raymond, that if Douglas could prove that it could climb out on one engine after the other's failure they would buy the aircraft from Douglas. So, the trimotor specification was not necessarily a hard spec., and for the purpose of this exercise we should consider other options as well.

The engine choices are also elastic, as the resulting DC-1 did not fly with the spec'd engine. For this design I think that Frye's focus was a safe aircraft design that could fly more passengers further than the B-247. The challenge is to design it with the constraints of the era's technology, and to further challenge designers by using the known engineering capabilities of the Douglas firm. Douglas had an exceptional design team that included "Dutch" Kindelberger and Jack Northrop. They also worked closely with the Caltech, GALCIT wind tunnels, to develop the DC-1 shape.

Aircraft manufacturers often develop their designs from a synergy of previous in-house designs (where they have an existing technology base that supports manufacturing) and industry technological innovations that themselves or others have developed (either in design or hardware). The Boeing B-247 emerged from the Boeing B-9 and the Boeing Monomail. Considering the predecessors of Douglas's DC-1 it would appear that the B-247 had a significant influence on the DC-1.

Below are two large Douglas aircraft whose designs would have preceded the DC-1 and the B-247.

 

[UpForce]

Here's a Douglas factory display model drawing for the DC-1, courtesy of Aviation Archives. Incidentally, to me it seems that current all-electric regional airliner projects are not too dissimilar from Frye's requirement, only nine decades on, from Eviation's Alice (9 passengers) to Heart Aerospace's ES-19 (19 passengers). Funny how that goes.

 

[Dynoman]

Thank you for the DC-1 image, it will be helpful to use in comparison with the proposed design. Looking at historical bomber aircraft performance circa 1930, Frye's performance request was on par with the bombers of the era (many were flying almost as fast as the fighters). Looking at different configurations it would be hard to beat the B-247 or the DC-1 configurations. Both were all-metal, low wing, streamlined, retractable landing gear aircraft. Refining the configuration would be possible, however a revolutionary design would be more difficult. The flying wing concepts would provide greater volume for passengers and could be sized to meet the needs of Frye. Burnelli's twin engine flying wing designs come to mind.
A chronology of Burnelli's designs can be found here: https://www.aircrash.org/burnelli/chrono1.htm

 

[riggerrob]

May I suggest 2 ways to eliminate those complex, curved wing root fairings? Those fairings just fill in an area where diverging airflow - from both the wing and fuselage can turn turbulent.

The first method can be done as a simple, bolt on retrofit. It involves installing a wing root box with vertical walls ... similar to modern jet airliners and business jets.

The second method involves making the fuselage belly flat and extending the fuselage walls straight down until they meet the flat bottom. This continues the DC-1's straight fuselage walls - with curved top and bottom - aesthetic. It also simplifies production of lower bulkheads.

Finally, make the fuselage a constant cross-section where it intersects the wing. A constant cross-section reduces the number of pressure changes at the wing root intersection.

 

[Dynoman]

Riggerrob, what about faired conformal fuel tanks that blend the bottom of the fuselage of a UB-14 design, into the bottom of the wing. This would help to reduce interference drag and strengthen the cantilever wing. Another idea is to wing mount the engines with the nacelles faired into the upper an lower surfaces of the wing. However, I like the engines close to one another, reducing asymmetric thrust in the event of an engine failure and their forward placement may aid in moving the CG forward. Flying wings require a forward CG. This may allow the tail to be resized in to a smaller tail unit.

The increased cabin volume of the flying wing fuselage should be exploited in the design. Below is a depiction of the Burnelli developed Cunliffe Owen OA2 design's seating arrangement (from our own Motorcar).

 

[HoHun]

Hi Dynoman,

As a fun exercise, using historical data and references, to design an aircraft that meets or exceeds Jack Frye's request for an airliner that can compete against the Boeing 247 and possibly even to exceed the performance of the DC-1/D-2 using era available technologies. This is strictly a fun exercise in conceptual and preliminary design to provide the opportunity to converse, research and design a classic airliner of the early 1930s, an era know for it's luxury airline travel and pioneering aviation achievements.

Coincedentally, I'm just reading "Professor Junkers called it 'The Fly'" by Herbert Pohlmann, designer of the Junkers Ju 87. In this book, Pohlmann makes a bit of a case for the flying wing as per Junker's 1909 patent (which wasn't restricted to tail-less flying wings).

In his book, he assesses the qualities of the 1922 "Junkerissime" passenger aircraft (wing body with tail on twin tailbooms, 4 x 700 HP, 60 passengers,. 30 tons take-off weight). His verdict is that the aircraft would have been technologically viable, but not ecnonomically, and that in the 1920s, the passengers probably would have been unhappy with an aircraft that did not give them a good view to the ground, as the aerial view was one of the major attractions of air travel at the time.

I guess that "good ground view for passengers" might be a specific requirement resulting from the more general "luxury airline travel" purpose of the intended design, even if we're into the 1930s by now.

(Burnelli's designs, some of which are also depicted in Pohlmann's books, seem to be almost as limited in this regard as the Junkerissime.)

Regards,

Henning (HoHun)

 

[Dynoman]

HoHun, Northrop saw this as an issue with his flying wing airliner concepts and placed windows at the leading edge of the wing where the passengers could look out and see the scenery. Burnelli's designs for transports also had windows, however his designs were considered high wing aircraft that afforded windows along the fuselage under the wing. I'll work out a sketch from the comments and see what could work. I'm wondering if slanted windows could be used to blend the bottom of the fuselage to the wing and still provide passengers with a view, similar to the Zeppelin's dining room (although on a much smaller scale).

 

[riggerrob]

Riggerrob, what about faired conformal fuel tanks that blend the bottom of the fuselage of a UB-14 design, into the bottom of the wing. This would help to reduce interference drag and strengthen the cantilever wing. Another idea is to wing mount the engines with the nacelles faired into the upper an lower surfaces of the wing. However, I like the engines close to one another, reducing asymmetric thrust in the event of an engine failure and their forward placement may aid in moving the CG forward. Flying wings require a forward CG. This may allow the tail to be resized in to a smaller tail unit.

The increased cabin volume of the flying wing fuselage should be exploited in the design. Below is a depiction of the Burnelli developed Cunliffe Owen OA2 design's seating arrangement (from our own Motorcar).

Dear Dynoman,
Are you suggesting a DC-1 like airplane with a stock tubular fuselage, but a thicker wing center-section?

 

[Dynoman]

I was thinking of something similar to the sketch below. This would allow for windows and to be able to exploit the volumetric efficiency of the blended body design. Its a bit futuristic, however the design could be all metal with metal formers that create the compound curves. This could result in additional weight. The design probably would not be able to compete on speed with the DC-1 design, but its useful load and perhaps field performance may be better.

 

[Dynoman]

A comparison between the DC-1 and the UB-14, according to Wiki, the Burnelli is just as fast, seats more than the DC-1, and has greater range. A further analysis of the UB-14 versus the DC-1 would be an interesting read. I'll try and get some more information on their performance.

 

[elmayerle]

If I remember correctly, the Douglas proposal in response to Frye's letter included at least a couple references to "...as on the Northrop Alpha...". Not surprising since that version of the Northrop Company was partially owned by Douglas and, when Jack Northrop wanted to go do other things, it became Douglas-Elsegundo. Amusingly enough, this building is once again owned by Northrop-Grumman Corp. and has been since at least the early F-18 days if not before.

 

[riggerrob]

I was thinking of something similar to the sketch below. This would allow for windows and to be able to exploit the volumetric efficiency of the blended body design. Its a bit futuristic, however the design could be all metal with metal formers that create the compound curves. This could result in additional weight. The design probably would not be able to compete on speed with the DC-1 design, but its useful load and perhaps field performance may be better.

Imaginative sketch!
May I suggest flattening the top of the fuselage?
Basically make it a single-curvature from one engine nacelle to the other. This will preserve lift over the center-section and simplify construction.
You will still need the V-tail or dual rudders to compensate for the turbulent airflow aft of the engines, fuselage, etc. But V-tails look so much more stylish.

 

[Dynoman]

Riggerrob, good suggestions. I'll modify the drawing and post it. Digging into the UB-14 design I am surprised that the Burnelli concept did not receive a greater amount of attention than it did in the early 1930s. Airline service in the 1930s were characterized by point-to-point, long-thin service lines, using multi-engine aircraft and single-engine feeders. To make that profitable the aircraft needed to be very efficient while flying the trunk lines. The Burnelli appears to have been able to do that with greater room for passengers.

 

[HoHun]

Hi Dynoman,

To make that profitable the aircraft needed to be very efficient while flying the trunk lines. The Burnelli appears to have been able to do that with greater room for passengers.

Pohlmann in the book I mentioned above expresses the opinion that the Junkers wing layout (to which he obviously counted Burnelli's design) only makes sense for larger aircraft. He suggests that the G-38 would have been better as a true Junkers wing, instead of being the compromise design it historically was.

(I'm using "Junkers wing" as a shorthand for aircraft with a profiled center section which houses payload and technical components, regardless of the design being tailless or "tailed", and regardless of whether the wing transitions to the tips smoothly or with a "blended step". It appears this was what Junkers covered with his patents, and I'm not sure that there's established terminology to describe that. "Blended wing-body" perhaps comes close, but I'm not certain whether it's really the same.)

Pohlmann seems to have been a structures man though, and perhaps not so much an aerodynamicist.

Regards,

Henning (HoHun)

 

[Dynoman]

HoHun, I've read the term 'lifting-fuselage' used with the Burnelli design. The UB-14 and earlier Burnelli model's components weren't faired into the wing to form a blended-body, nor flying wing (i.e. tailless), or lifting body (as the Burnelli designs had wings). The proposed high wing design with a blended fuselage allows for minimal interference in the elliptical lift distribution across the aircraft's wing span and the design helps to reduce the wetted area of the aircraft resulting in a reduction in form drag. The drawback maybe in that it has a higher empty weight for a given payload relegating the aircraft to longer haul missions where it would be more efficient.

I'm putting together a direct comparison between competing designs of the circa 1932 era when the Frye letter was written. This would help in creating historical weight fractions that will aid in sizing the aircraft.

 

[riggerrob]

Yes HoHun,
Flying wings - or lifting fuselages - are inefficient for passenger carrying until the fuselage/wing is deep enough to allow passengers to walk around upright. Walking around upright speeds loading and un-loading. It also improves passenger comfort. Keep in mind that only the upper class could afford to fly during the 1930s, and they are better fed than most, so grow to their full potential, with men often more than 6 feet tall. That means a minimum door height of 6 feet. By the time you add structural floors and ceilings, the external depth of the fuselage/wing approaches 8 feet. If you consider a maximum chord to thickness ratio of 18 percent, that means a minimum chord of at least 40 feet, which leads to a total length of 80 feet and a wing span of 120 feet. Now back-calculate those numbers to determine the payload and number of passengers.
For hatch size, I was picturing a span-wise corridor with hatches cut through ribs. The corridor would need to be just ahead of or just behind the main spar to get maximum depth .. say 25 percent mean aerodynamic chord.
Try to preserve as much of the leading edge as possible for windows. Mind you, post-1930s windows will be needed to preserve smooth airflow over window edges.

 

[riggerrob]

Dear Dynoman,
Your blended Brunelli concept has several advantages. Its flared side walls move wing-tip vortexes farther outboard.
Moving engines out onto wings helps free up the forward fuselage for better visibility from the cockpit. Your semi-bubble cockpit also addresses the question of how far apart can you seat two pilots. Modern wisdom limits cockpit width to 6 feet. Sitting close together allows pilots to cross-check each other's instrument panels and flip each other's switches. Since your configuration eliminates blind-spots created by engines ... how about moving pilots farther forward until their rudder pedals are almost at the leading edge?
How much can you narrow the cockpit before it starts messing with airflow over the fuselage/wing center-section?

 

[Dynoman]

Here is a narrower cockpit version of a Burnelli design, a CBY-3.

 

[riggerrob]

Here is a narrower cockpit version of a Burnelli design, a CBY-3.

Well ... yes .. sort of ... but it ruins the boyish good looks of the basic Burnelli concept.
I prefer Dynoman's sketch in post #21. That sketch positions engines far enough outboard that they only impinge a little on visibility from the pilots' seats.
You really don't need a parabolic nose until radar is introduced and that did not happen until part way through World War 2. Most pilots would be happier with a mere instrument panel suspended in an all glass nose (e.g. Heinkel 111 or Boeing B-29).

 

[Dynoman]

Here's another sketch of the concept. I lengthened the fuselage to increase the moment arm of the horizontal stabilizer and vertical stabilizer for more control authority. The wings have also been swept and the engines moved further outboard. It's length is between the UB-14 and the DC-1.

 

[Dynoman]

Burnelli aircraft comparisons.

 

[riggerrob]

Here's another sketch of the concept. I lengthened the fuselage to increase the moment arm of the horizontal stabilizer and vertical stabilizer for more control authority. The wings have also been swept and the engines moved further outboard. It's length is between the UB-14 and the DC-1.

Dear Dynoman,
Your latest sketch has several improvements.
Your cargo version would be better with a nosewheel. ... especially if you want to install a tail ramp. The ideal tail ramp is the same height as a standard truck bed, to minimize the amount of hand-bombing required.
I also preferred your V-tail. It looks better and is less likely to be hit by a half-awake truck driver.

Whether a passenger-carrying Burnelli has a nosewheel or tail wheel is unimportant as long as you have one door close to the ground. On a high-winged Burnelli, the entire cabin floor is close to ground level when you install a nose-wheel. Passengers enjoy easier boarding when the floor is level. A level floor also allows you to install two or three passenger doors per side.

 

[riggerrob]

Dear Dynoman,
As for your control surfaces ...
Sweeping the wings' leading edges will help with roll stability and fits with the very latest (post 2000) Schumann wing now favored by Formula 1 racers and competition sailplanes. While Schumann favors adding a slight sweep to the trailing edges, it complicates control surface installation. Schumann also recommends swept wing tips that are the result of a graceful curve from the leading edge to tip. Think about some of the light planes designed by Dean W. Wilson (Explorer, etc.).
Control surfaces work best when their hinge lines are perpendicular to the airflow.

As for the elevators ... I am not sure if the center portion of the elevator will contribute much because it is in the turbulent wake of the fuselage and propellers. OTOH the outboard portions will definitely be effective at controlling pitch.

 

[Dynoman]

Riggerrob, I agree the trailing edge should have a zero angle of sweep for better control surface effectiveness (i.e. perpendicular to the hinge-line). The DC-1 had about a 14 degree LE sweep, which is approximately what I drew into the sketch. The canted vertical stabilizers are definitely more effective, however I couldn't find any other design that promoted V-tail configurations until the 1940's. A canted design during the era would be cool to see. I'm trying to put together a historical weight fraction chart of the Burnelli aircraft and other era transports to get a better idea of the component weights for sizing.

 

[Nicknick]

I was thinking of something similar to the sketch below. This would allow for windows and to be able to exploit the volumetric efficiency of the blended body design. Its a bit futuristic, however the design could be all metal with metal formers that create the compound curves. This could result in additional weight. The design probably would not be able to compete on speed with the DC-1 design, but its useful load and perhaps field performance may be better.

looks great! Would be nice to see a model of it...

 

[Nicknick]

This rather crappy looking scratch is just to meant to demonstrate the idea behind it. I’m so much used to 3d CAD, that I started to loose my drawing capabilities…

I guess, no one could pretend, that he would have ever built a fully conventional airplane better than the DC-1 at its time, so I tried to find an potential old fashioned looking approach which offer some advantages, which I call, the ultra-high wing double fuselage design. This design allows for an uninterrupted wing spar and an undistributed air flow over the upper wing surface. The turbulence of the propellers and the cooling flow will not interfere with the wings which might help to reduce drag. Other two fuselage designs like the twin Mustang will have four interference points between the wing and the fuselage, which will be reduced to just two which only are only interfering with the under side of the wing. The structural weight can be reduced by reducing the wing bending moment with two fuselages which apply the load out of the middle. The design allows an aisle which is tall enough to stand upright, but you need to be carefully about not bumping your head when seating. Furthermore, the design offers plenty of space for a fully recappable landing gear which was not the case on the DC-1 (and not even in the DC-3). The fuselage would be getting narrower after the wings, to be not in the way of the downwash from the wing. I’ve drawn 3 windows per side of the fuselage for twelve passengers in total which equals to the DC-1.

Unfortunately, storing the luggage might be somewhat more complicate, the wing center section might be an option, despite it would be not very handy to reach. All twelve passenger will have an own window, with quite a good view due to the high wing, even if the view for the center seats would be disturbed by the opposed fuselage.

 

[riggerrob]

This rather crappy looking scratch is just to meant to demonstrate the idea behind it. I’m so much used to 3d CAD, that I started to loose my drawing capabilities…

I guess, no one could pretend, that he would have ever built a fully conventional airplane better than the DC-1 at its time, so I tried to find an potential old fashioned looking approach which offer some advantages, which I call, the ultra-high wing double fuselage design. This design allows for an uninterrupted wing spar and an undistributed air flow over the upper wing surface. The turbulence of the propellers and the cooling flow will not interfere with the wings which might help to reduce drag. Other two fuselage designs like the twin Mustang will have four interference points between the wing and the fuselage, which will be reduced to just two which only are only interfering with the under side of the wing. The structural weight can be reduced by reducing the wing bending moment with two fuselages which apply the load out of the middle. The design allows an aisle which is tall enough to stand upright, but you need to be carefully about not bumping your head when seating. Furthermore, the design offers plenty of space for a fully recappable landing gear which was not the case on the DC-1 (and not even in the DC-3). The fuselage would be getting narrower after the wings, to be not in the way of the downwash from the wing. I’ve drawn 3 windows per side of the fuselage for twelve passengers in total which equals to the DC-1.

Unfortunately, storing the luggage might be somewhat more complicate, the wing center section might be an option, despite it would be not very handy to reach. All twelve passenger will have an own window, with quite a good view due to the high wing, even if the view for the center seats would be disturbed by the opposed fuselage.

May I suggest that you do not need to "neck down" the width of the fuselages towards the top?
"Necking down" adds two more interference points per fuselage. You will get some weird airflow in the channel between the top of the (wide part) fuselage and the underside of the wing.
You will have fewer intersections and less complex airflow if you just extend the fuselage sides straight up to the wing. The smooth airflow aft of the wing, keep fuselage sides vertical and make the top of the fuselage flat. Curve the top of hte fuselage to mimic airflow coming off of the top skin of the wing.

 

[Nicknick]

Youre propably right, but I think the, as you called it, "neck down" approach might work better in a more radical fashion if it is done like on the Catalina flying boats with a more triangular fuselage cross section and a longer neck above it. If I remember it correctly, these must have been quite aerodynamic efficient for a flying boat.

In my drawing I ment the fuselage to be have a declining height after the wing to make room for the downwards airstream, just as you described it.

BTW. I know there are much better plane drwers than me in here...

Historic Air Tours - Come Fly The PBY

Come fly the PBY based in Zepher Hills, Florida in the Winter months, and Martha's Vineyard in the Summer.

www.historicairtours.com