A. A Team Effort
The design of the NC series was fundamentally a team effort. While certain individuals made singular contributions to the concept or design, no one person could be credited for designing the entire vehicle. Ideas from all members were given equal consideration, chosen on the basis of merit alone. As Hunsaker writes, “no one man can be said to have designed these craft, although the Chief Constructor of the Navy, Admiral Taylor was at all times responsible”. Even the name of the series, NC – “N” for Navy and “C” for Curtiss – bears out the unified nature of this project. Hunsaker goes on to describe the design as the “organized result of what we had learned from previous experience, what we could deduce as to the future by application of aeronautical engineering theory and methods, and what we could learn from foreign practice.” Today, this is commonplace, but in the early 20th century, these “best-practices” techniques were newly being applied to aircraft design. After Taylor and Hunsaker’s initial studies into the feasibility of the aircraft that was to be known as the NC, the development and design team was quickly completed with Glenn Curtiss and Navy Commanders George Westervelt and Holden Richardson. Each of these men had significant and unique experience and were some of the best and most promising aeronautical engineers of the day (Fig. 8).[ii]
Glenn Curtiss, the founder of the Curtiss Aeroplane and Motor Company, was the first to successfully build a seaplane and remained at the forefront of seaplane design. His company had worked with the Navy since the beginnings of naval aviation in 1911, and had a proven record of innovation. Furthermore, while his America was never able to attempt transatlantic flight as intended, the Model H series of flying boats that followed were some of the most successful designs of the war. Curtiss was one of the pioneers of aircraft design and manufacturing in the Unites States; as Taylor said: “The Curtiss Engineering Corporation is the only firm in position to undertake this development at the present time”. Curtiss was therefore chosen as the contractor to build and integrate the Nancies and his company was responsible for many of the design details.
Commander George Westervelt was a naval constructor, specializing in structural design. He too had a history with aviation and seaplanes. In 1914, Westervelt was stationed in Seattle, Washington, overseeing shipbuilding efforts. He became interested in aviation in general and seaplanes in particular, along with a local lumber supplier and boat builder. This man decided that he wanted to purchase a seaplane, and asked Westervelt to recommend a type. After researching the available aircraft, Westervelt could find none to recommend, and so the man offered to build two aircraft if Westervelt could come up with a good design. While he had never attempted to design an aircraft before, Westervelt agreed to the challenge, enlisting as much help as possible from every source he could find. Eventually, he settled on a design, and as promised, two were built (Fig. 9). This was the lumber-man’s first foray into aircraft manufacturing, but would not be his last; William Boeing decided to become a manufacturer of airplanes. Based on this experience, CDR Westervelt was placed in charge of aircraft inspection and construction by RADM Taylor. One of Westervelt’s early contributions to the effort was suggesting a name for the new flying boat; his proposal was “DWT” for David W. Taylor. On further consideration, it was decided that Taylor would not take kindly to this, and the NC designation was adopted.
Commander Holden Richardson, also a naval constructor, was another one of Taylor’s prodigies. Richardson became involved with naval aviation in 1911 as the Navy’s first engineering and maintenance officer for aviation. In 1912, Richardson translated Gustave Eiffel’s aeronautical research data and began the process, along with Taylor, of designing and building the Navy’s large scale wind tunnel, which was the foundation of the Navy’s newly established Aeronautical Laboratory. A few years later, Richardson used this tunnel to evaluate the government’s first in-house designed aircraft, the 82-A or Richardson Seaplane. While an accomplished aerodynamicist, he had unparalleled expertise in the design of flying boat hulls and seaplane floats, and this is what he was brought to the NC design team to do. In addition to being an accomplished engineer, Richardson was also a pilot. He was naval aviator number thirteen, and the first engineering test pilot. CDR Richardson served as a test pilot on the NC project, and he personally tested the sea-worthiness of his hull design while sailing the 200 nm to Ponta Delgada as pilot of the NC‑3.
B. Designed for Combat
Per Taylor’s direction, the NC series was to be a serviceable, combat-ready, flying boat. This required building a vehicle that could withstand the rigors of combat deployments, protect the crew, complete its intended mission in non-ideal conditions, be maintainable and repairable in theater, and be built in accordance with standard Navy practice. These requirements all tend to have the undesirable side effects of adding weight, cost, and complexity. For a vehicle that was already the largest and heaviest flying boat ever built, these challenges were compounded. Examples of these considerations are: multiple redundancies on flying wires and landing wires in order to maintain integrity in the event that wires were cut by enemy fire; extra factors of safety on critical components, where needed, to survive maneuvering and rough landings; control rigging hidden under a hinged leading edge to allow ease of inspection as well as reduced drag; and, of course, the ability to carry and deploy weaponry.
Taylor’s requirement that the Nancy handle foul weather in the water put additional requirements on the overall design. The entire vehicle, but especially the hull, would need to be robust enough to take a pounding from the ocean and then continue with the mission. Corrosion was also a significant consideration in the design of a vehicle intended to serve its useful life in salt water. Wood and fabric treatments were well developed by that point, though the fabric covering on the wings would need to be changed every six months to one year. Improvements were necessary in order to protect the highly stressed and weight-optimized metal components. A new process of electro-galvanization prior to painting was developed for steel. This was considered a significant advance in corrosion protection. Aluminum was also used for certain components in the design. Given that this was the first time aluminum was used in quantity on heavier-than-air aircraft, and due to its reactive nature to salt water, methods of protection were necessary before it could be utilized. The coatings developed were very successful and used later for strength members of dirigibles.10
C.The Incredible Hull
Designing a flying boat as large as the NC series required re-envisioning what the shape of the aircraft should be. Flying boats of the day were built on a single hull extending from the bow to the tail surfaces, directly supporting all of the vehicle’s components. Due to the unprecedented size of the Nancy, though, Curtiss had a different idea. His concept was a shorter hull with the tail supported not by the hull itself but by a system of booms, struts, and outriggers anchored to the hull and the upper wing (Fig. 10). While somewhat unusual looking, this highly visible aspect of the design helped to meet the sea-worthiness requirements by keeping the tail as high above the waterline as possible. This would allow for operation in higher seas without the waves hitting the tail. Curtiss’ concept not only allowed the tail to be mounted high up, but also saved a significant amount of weight.
Starting with Curtiss’ idea, Richardson went about designing the shortened hull. He based the design on his previous work with seaplane floats, and the result was unlike anything that had been seen before. In fact, it was so out of the ordinary that it was made the subject of ridicule by many of the world’s experts in aircraft and flying boat design. Many found it ungainly, and were not shy about expressing their doubts. As one distinguished British visitor opined, “The hull of this machine was examined, and is the design of a naval constructor. The machine is impossible, and is not likely to be of any use whatever.” Even CDR Towers found the design, at first, odd looking, and stated openly that he did not like it. Richardson, for his part, was undeterred. He was designing a flying boat hull with unprecedented buoyancy and planing requirements that had to be operable and safe in adverse seas while being as light as possible. It was not an ordinary problem, and required an extraordinary solution.
Previous designs relied on the width of the hull to achieve planing at reasonably low speeds, often adding sponsons or pontoons to the sides of the hull to get the necessary lift. Richardson realized that this additional width would add both weight and drag, neither of which could be afforded in the design. Furthermore, the added width would be destabilizing in heavy seas, making the requirement of navigating through rough seas impossible. Instead, he designed the hull to plane with speed rather than width which was a radical departure from the standard practice. Another important feature of flying boat hulls is the “step,” which both reduces the drag and can provide stability while planing. Richardson’s novel design included a single step and also makes use of the stern of the hull as a second step, providing a stable platform while planing and allowing the pilots to better control the aircraft while at speed on the water. This concept was yet another significant improvement in the design of flying boats (Fig. 11).
The structure of a hull is of equal importance to its shape. Being part boat, part airplane, a flying boat must be able to withstand the loads imposed by the sea while at the same time remaining light enough to fly. For a vehicle as large as the Nancy, this challenge is magnified. These opposing requirements were successfully managed by careful selection and distribution of material. W. L. Gilmore, a Curtiss engineer, is given credit for much of the structural design of the hull. The keel of the hull is built up from spruce while the bottom planking is laid up from two plies of cedar separated by a waterproofing barrier of muslin set in marine glue. Ash girders braced with steel wire provide longitudinal strength. As designed, a bare hull weighs 2,800 pounds with a displacement of 28,000 pounds, an incredible-for-the-time ten-to-one ratio. Hunsaker described the Nancy’s hull as having an “easy flaring bow so that it can be driven through a seaway to get up the speed necessary to take the air and a strong V-bottom to cushion the shock of landing on the water. The combination of great strength to stand rough water with the light weight required of anything that flies was a delicate compromise, and it is believed that a remarkable result has been obtained in this design.”10
While small by current standards, the hull was spacious for 1918. There were six compartments separated by bulkheads, and originally watertight doors for survivability in the event of damage from battle or heavy seas, as shown in Fig. 12. Narrow passageways along the side of the hull allowed the crew of six to move between these compartments, and all but the two pilots could remain below decks and out of the weather if desired. The aircraft commander even had enough space to lie down on the planking that made up the floor of his compartment at the front of the airplane (the airplane commander also served as the navigator). In order to permit inspection and maintenance of the engines, topside hatches and non-skid walkways were incorporated to allow the engineers to move about. A “tunnel” on the aft deck was provided for an engineer to crawl through, under the centerline pusher propeller, and they used linesman’s belts to secure themselves to the aircraft while moving about in flight (Fig. 13).
Curtiss and Richardson’s unusual design was vindicated by its performance. It permitted the Nancies to get-away at weights even above the originally designed maximum gross weight, and remained stable on the water and in the air. Further, it proved rugged and sea-worthy beyond what any could have imagined during the transatlantic flight. When the NC‑1 and NC‑3 put down in the ocean near the Azores, the seas were rougher than anticipated at up to fifteen feet. Both aircraft suffered damage upon landing, but would not have been able to take off again regardless, due to the sea conditions. During the overnight saga of the NC‑3, the seas were in excess of thirty feet and very steep, with gale force winds blowing, according to the first-hand accounts of Towers and Richardson. This is the equivalent to sea state eight conditions, and well beyond the sea-keeping capabilities of any other flying boat of the day. The design saved the crews of both the NC‑1 and NC‑3, and permitted the crew of NC‑3 – including Richardson himself – to sail safely, if not comfortably, to port in unbelievably difficult conditions.19,
D.Experimentation, Analysis, and Testing
With the availability of the Experimental Model Basin and Experimental Wind Tunnel at the Washington Navy Yard, and Curtiss’ own smaller facilities, the team had unprecedented access to cutting-edge experimental facilities. In 1917 and 1918 Dr. A. F. Zahm, head of the Navy’s Aerodynamics Laboratory, conducted one wind tunnel test of hull designs and two tests of the complete aircraft in the Navy’s large wind tunnel, as well as a special stability test. These tests validated the aerodynamic design of the vehicle and were used to tune the performance and handling with evaluations of tail size and incidence, control surface balancing, and overall stability. In 1917 and 1919, three tests of the hull were conducted in the model basin by Richardson and Naval Constructor William McEntee. During these tests, three different hull designs were tested before the final shape was decided upon, then fine tuned for best trim and performance. Richardson’s earliest design had two steps with an upward curvature of the keel. The first modification removed the curvature, and the second modification removed second step creating the final shape with the unique stern that functioned as a step. Through testing, it was found that without these modifications, the Nancy would not have gotten off the water. Finally, in late 1918 and 1919, seven tests were conducted in Curtiss’ smaller wind tunnels to assess design changes and final configurations. The entire design was thoroughly analyzed for lift, drag, and power required for flight; control authority and power; stability; and hull hydrodynamics and stability (Fig. 14). These tests provided the basis for the team’s confidence, prior to construction of the first prototype Nancy.
As the aircraft was originally intended for wartime use, the design and test schedule was highly compressed. From Taylor’s initial idea in August of 1917, it took just over one year to complete the prototype aircraft and on October 4, 1918, the NC‑1 flew for the first time with CDR Richardson as the test pilot. Initial tests of the NC‑1 proved the design to be sound with performance that exceeded expectations. The handling of the aircraft was excellent without requiring too much effort on the part of the pilots. There had been concern that an aircraft as large as the Nancy would need servo assistance on the controls, but due to the careful design of the control surfaces for aerodynamic balance and fine-tuning of the tail size and incidence in the wind tunnel, the aircraft flew without much effort and was very stable. To further improve handling qualities, the center of lift was determined through wind tunnel tests and the vehicle was balanced so as to collocate it with the center of gravity, as shown in Fig. 14, right.
Soon, the NC‑1 was being exercised at high gross weights, even beyond the design maximum, and over extended ranges. One of these flights took the NC‑1 on a trip to Washington, D.C. where it docked on the Anacostia River at the Washington Navy Yard. It was here that RADM Taylor saw the aircraft that he envisioned for the first time. It was also decided to attempt to set a record for the most people carried aloft while the NC‑1 was still in test. On November 25, 1918, 51 people (one being a stowaway, hiding himself in the hull for hours wanting to be a part of the record setting flight) were crammed into the hull and the NC‑1 easily lifted off (Fig. 15). This bested the record of 40 persons set just prior in a Handley-Page bomber.
The initial design was for a maximum gross weight of 22,000 pounds with three engines. During flight tests of the NC‑1, the structure was determined to be capable of carrying more weight if more power was available. Consequently, the decision was made to configure the NC‑2 with a fourth engine for testing (Fig. 16). This extra engine, while not explicitly necessary for flight, offered the advantages of additional redundancy in addition to greater range and payload, which was especially important for an aircraft intended to operate over vast expanses of open ocean carrying as much fuel, equipment, and weaponry as possible. The fourth engine brought the maximum gross weight up to 28,000 pounds, 12,000 of which are payload. This is a useful weight fraction of 43%, an incredible achievement. For comparison, the land based Handley-Page V/1500, or Super-Handley, while heavier, had a lower useful weight fraction of 41%. Different configurations for the three and four engine installations were tested before the final configuration was decided upon. As with the design of the rest of the vehicle, results and performance, rather than preconceived notions or personal preference, guided the process.
E.Structure
The structural design of all the various components had to be carefully engineered to carry the massive loads while remaining light enough to fly. The wings, struts, spars, tail booms, fitting, wires – everything that went into the build of the vehicle – needed to be carefully considered. While the designers utilized the standard RAF 6 airfoil for the wings, the ribs and structure had to be built to handle the enormous weight and load requirements for the 28,000 pound flying boat. In some cases, the structure had more in common with bridges than with typical aircraft construction. George Westervelt, having been assigned by the Navy to oversee final design and construction of the Nancies, was also responsible for the structural design and testing of all the various parts of the aircraft, and personally directed the build-up of the wing. As he did when designing his first aircraft for Boeing, CDR Westervelt gathered as much information as possible on the methods that other engineers had used to build wings for large aircraft. He traveled to England and met with Sir Frederick Handley-Page, who, after much discussion, gave Westervelt a sample of the rib used in his Super-Handley night bomber. Westervelt ended up basing his design on this rib.
Metal fittings were a challenging design problem to keep the amount of material used to a minimum. Each fitting, having unique load bearing requirements, was analyzed individually in order to ensure that it met the structural design requirements while remaining as lightweight as possible. The result were pieces that were, literally, the work of a jeweler (Fig. 17). This attention to detail at all levels exemplifies the commitment to excellence that the entire team exhibited throughout the course of the design, construction, and testing of the NC flying boats.
In addition to the aerodynamic and hydrodynamic testing, significant experimentation and analysis was done on the proposed structures of the wing, tail, and riggings, and to determine the best materials to use. Load testing rigs, able to simulate the forces exerted in flight, were used for testing wing rib designs to failure (Fig. 18). Booms were tested for strength in compression and bending, and components of different materials were tested for best performance. Through the course of this process, many different concepts were tested for the variety of load bearing components. For some of these, there were collegial disagreements over which design would be best. The result would remain objective based on testing and engineering analysis, however there would be friendly bets placed on each design as to which would be optimal. This light-hearted competition fostered both ingenuity and application of solid design principles.19
F.Construction
Significant advances in construction were necessary in order to build the unprecedented Nancy flying boats. The aircraft was simply too large and complex to be built by a single manufacturer, especially given that the original intent was to produce the aircraft in quantity for combat use. It was decided to break the construction up into components and sub-contract the build to manufacturers who could fabricate the specialized pieces. Curtiss would be responsible for the overall construction and integration of all the parts, and the Navy, with Westervelt as its representative, would retain overall authority over the build. This method of construction, while standard today, and common for ships of the day, was new for aircraft and required significant coordination and precision in design in order for all the pieces to fit together and work as required. The following major components of the NC flying boats were built by the different companies shown in Table 1.
These companies had significant expertise, but in areas not necessarily related to aircraft manufacturing. For example, Unger Brothers was a maker of fine silverware and jewelry, Locke Body Company was a high-end automobile coach-builder, and Pigeon Fraser Hollow Spar Company built masts and spars for racing yachts.11 There was concern early in the process that the components would not fit or be serviceable, but those fears were quickly allayed during the first build of the NC‑1. These companies were able to quickly adapt their specialties to the unique requirements of aircraft manufacturing and the assemblies all fit together very well (Fig. 19).
The Curtiss Company needed new manufacturing and assembly facilities to support the number and size of these aircraft. A factory was built for the purpose in Garden City, and in the course of one evening, the entire staff moved from Buffalo picking up immediately where they had left off the previous night. The variety of subcontractors and their geographic diversity, relative to the transportation options of the day, required logistical solutions to uncommon problems. The completed wings panels had to be moved from downtown Manhattan through Long Island to Garden City for assembly. These 12-foot by 45-foot structures were delicate and could not be moved quickly, or easily, through the narrow and rough roads. The only trailers available to move the large sections were built for moving theatrical sets, and there were only a couple in the entire city. It was decided that they would be moved in the middle of the night when there was minimal traffic, or witnesses, and this strange caravan would slowly make its way out of the city whenever wing sections were completed and ready for installation.
G.Engines and Power
Prior to 1917, very large aircraft were impractical due in large part to the lack of suitable engines. None of the available powerplants had the combination of power and lightness required for practical use in a large airplane. The first engine that offered this performance was the Rolls Royce V-12 Eagle, which was being used to power the large British bombers. Curtiss was also developing the K-12, an advanced, powerful engine made from lightweight materials and incorporating a gear reduction system to improve power and efficiency. While promising, the K-12 was ahead of its time and would not become a viable engine. Even though Taylor explicitly directed the use of the United States Motor, consideration was given to these alternatives if the preferred engines were not developed in time.18 As the NC design progressed, so did the Liberty engine, as the motor was to be known. It was a serviceable powerplant by the time the NC‑1 was ready for engine installation. Multiple versions of the Liberty were under development, each with progressively better performance, but all were based on the same 27 liter, 45° V-12 block. The first version of the Liberty was known as the low compression Liberty but these were quickly superseded by the high compression, or “Navy Liberties”. These engines produced 400 horsepower and weighed 850 pounds. A geared version was being developed that promised much greater efficiency, but it was too far from completion to be considered for use in 1918 or 1919. Through the course of the war, Liberty engines were built by many manufacturers, including Buick, Cadillac, Ford, and Lincoln, though the Nancies used engines built by Packard.
The initial design of the Nancy used three low compression Liberty engines in a tractor configuration, with the engines installed in nacelles between the wings. When it was determined through testing that engine performance was a limiting factor, it was decided that adding a fourth engine would be beneficial for performance and safety in the event of the all-to-common engine failures. The NC‑2, originally built with three engines similar to the NC‑1 (the centerline engine on the NC‑2 was a pusher though), was modified to operate with four high compression Liberty engines, installed in tractor-pusher “twin-tandem” pairs between the wings. This “NC‑2T” retained the center nacelle for the pilots, as shown in Fig. 16. When the NC‑3 and NC‑4 (Fig. 20) were built, a compromise arrangement was tried where a tandem pair was mounted along the centerline and single tractors were mounted in nacelles on the wings, as with the NC‑1. The pilots were then moved to a cockpit in the hull. This configuration increased the efficiency of the propellers as only one would be operating as a pusher in the wash of another, and provided a further measure of safety by decreasing the likelihood of dangerous unintended yaw from differential thrust in the event of engine loss. This would be the final configuration and the NC‑1 would eventually be converted to it as well.
In addition to the new engines, advances were made in the delivery of fuel and oil. The fueling system consisted of a set of nine interconnected 200 gallon aluminum fuel tanks in the hull (Fig. 21) and a single, 90 gallon gravity feed tank in the upper wing. Fuel was moved to the gravity tank by flow powered pumps (Fig. 22) which then fed the engines. There were manual pumps in the event that they were needed. The use of aluminum in the fuel and oil tanks, and through their respective distributions systems, was the first large scale application of this material in heavier-than-air aviation. Each 200 gallon fuel tank weighed only 70 pounds, saving a total of 630 pounds compared to the equivalent steel tanks.
H.Equipped for Success
The vehicle itself was not the only development in aviation technology. The equipment installed and used on the transatlantic flight was cutting edge, and some was being tested for the first time. The Nancies were equipped with a full assortment of avionics. The cockpit had airspeed gauges, altimeters, compasses, pitch attitude and angle of bank indicators, and engine performance and status gauges (Fig. 23). Up front in the navigator’s compartment, the aircraft commander had a specially designed sextant that could be used without a horizon for sighting, a drift indicator, compass, and a table under the deck for all the necessary maps and charts. The real innovations were in the radio compartment, though. The radio operator had access to 75 mile short range and 300 mile long range radio sets, and there was an intercom system allowing the crew to speak with one another and even allowed the commander to speak over the radio. There were two sets of antennae for use depending on whether the boat was on the water or in the air; one fixed between the wing struts and one trailing unit that could be reeled in before landing (Fig. 24). These radios allowed the Nancies to communicate with each other and with the ships strung out across the Atlantic.
Radio was not used for communication alone; for the first time it would be used over a long distance for navigation. The Nancies had radio compasses, or radio direction finders, that the radio operator would tune to a transmitter to determine the aircraft’s relative bearing to the location of the transmitter. Ships strung out across the Atlantic were equipped with these transmitters to provide a beacon for the aircraft to follow. The radio compass worked well while installed on the NC‑2 with its twin-tandem engine configuration, providing good bearings out to sixty miles. Unfortunately, there was insufficient time to fully test the installation with the final engine configuration and the interference created by the centerline engines significantly reducing the radio compass’ effective range. The compasses and gauges were self-illuminating for visibility at night, however these needed to be “recharged” regularly by flashlight. Powering all this equipment were batteries and a wind-powered generator located in the slipstream of the centerline propellers. The result was a better equipped aircraft than had ever before flown, and it needed to be, in order to find its way across the ocean.