A Century of American Submarine Propeller Design
Edward Monroe Jones, Ph.D.
As the paddle wheel was replaced by the screw type propeller in the 19th century the 20th century brought submarine propulsion design from the bladed propeller to the ducted impeller propulsor. That journey was marked by milestones in mathematical applications, feats of engineering, and trial-and-error experimentation. It is worthwhile to review this fascinating aspect of American submarine history.
In the latter part of the 19th century pioneering naval engineers such as William Froude, David W. Taylor and Stefan Drzewiecki determined the basic behavior of propellers.
In its simplest form, as viewed from the perspective of John Holland and Simon Lake, a submarine's propeller was made up of a rotating hub with radiating paddle-shaped blades angled to the axis so as to bite into seawater and thereby push the water rearward with resultant vessel forward advance. The straight paddle shaped, two-bladed propeller was replaced by elliptically shaped blades in two and three bladed variations. As submarine designs improved, so did propeller design and both men attacked the problem of blade efficiency by mathematical formulae and experimentation. The shape of a blade's pressure side, or face, changed from a flat surface to a variously- curved shape depending on the desired rotation speed. Propeller efficiency was plotted on a graph where pitch variation was compared to ship's advance through the water. Graphic plots illustrated a propeller's slip ratio or wasted thrust.'
John P. Holland concentrated his efforts on a single screw submarine and by the turn of the 20th century was in negotiation with the Navy Department to build the first United States submarines. His major competitor was Simon Lake who also competed for Navy submarine construction contracts. Lake introduced certain worthwhile concepts in submarine design which included twin screws and variable pitch propeller blades. Holland stuck to his single, centerline propeller concept despite Lake's argument that such a design meant poor maneuverability and unwanted torque.'
The Navy Department saw merit in Lake's arguments and insisted that Holland redesign his submarine to include twin shafts and propellers. In frustration, some of Holland's initial designs submitted to the Navy Department included triple screws since he still believed that a single, centerline screw would provide the greatest efficiency.' On April 11, 1900 the U.S. Navy bought Holland's final design which became known as the USS Holland. It incorporated his first concept of a single propeller extending aft along the centerline axis of the submarine.The Holland in dry dock showing its control surfaces aft of the 3 bladed centerline screw. Naval Historical Center
Subsequent submarines included the Fulton and the Adder series, all of which incorporated the single centerline propeller design. Simon Lake continued to compete with Holland and produced a reliable adjustable pitch propeller which continued to be a concept ahead of its time. He switched to a fixed pitch design because of cost and manufacturing complexity. Lake's boat performed less favorably than those of Holland and he blamed the performance on compromised propeller design.' He introduced a twin, four bladed propeller with exaggerated blade tip width and won the Navy Department's contract to build the USS Protector. The succeeding Viper class of submarine was equipped with Simon Lake's variable pitch propellers. These were used to advantage when maneuvering in restricted waters. The recently organized Electric Boat Company built the Octopus which was the Navy's accepted twin screw submarine.' At the same time, improvements were made to propeller shaft glands and thrust bearings. Improved thrust bearings were constructed of segmented pivotal shoes which butted against a revolving thrust collar.
The twin screw with three bladed propellers became the standard design of the U.S. Navy through the 1920s. It adopted a four bladed screw design with improved blade curvature efficiency prior to the Second World War. At the close of the Second World War the Navy acquired the German Type XXI advanced-design submarine. The Navy's new Tang class, fast-attack submarine appeared in the 1950s and was the result of concepts taken from the German design including retention of the basic twin screw. The Navy's first nuclear powered submarine, the USS Nautilus (SSN-571), retained the fast-attack hull design and twin screw, four bladed propeller. The submarine was a truly revolutionary phenomenon. It could out-run ASW surface ships with ease, but its major problem was noise. BuEng and civilian engineers renewed their efforts to solve problems of propeller efficiency and noise-producing cavitation.
Some basic concepts may serve to outline the problems. The purpose of the propulsor is to develop thrust to overcome resistance to motion of the submarine. The delivered power of the propulsor is defined as thrust times speed.' Propeller efficiency is output energy divided by input energy. Less energy is expended if a large mass of water is given a small change of velocity. Hence, for propulsor efficiency there is benefit in having a large diameter propeller.' Also, when the propulsor is close to the hull, certain effects arise. As water passes around the hull it changes velocity. At the bow it comes virtually to a standstill, then accelerates around the hull shape. As the stern-form reduces in diameter the fluid again slows. Therefore, at the stern there is an area surrounding the tail of the hull where there is slow moving water with an accompanying efficiency advantage in placing the propeller in the low-velocity wake. The initial conclusion is that a large diameter propeller located close to the submarine's stern can be expected to give higher efficiency; however, the wake is limited to a slim region around the stern, so a very large propeller diameter may extend beyond the wake resulting in rather poor overall efficiency. There is a limit to the hydrodynamic gain with a large diameter propeller. Also, a large propeller introduces an augmented drag which requires a greater input thrust to overcome. By moving the propeller farther astern the associated loss of efficiency can be reduced, but to do so may lose the advantage of the wake. To make the problem even more complicated, the wake is not uniform because the hull has upstream appendages such as the sail and these cause interfering wake turbulence.'
The above considerations represent simplified explanations of the complex problems faced by naval engineers. Experimentation was needed to test engineering principles. This was tentatively achieved using tanks such as the David Taylor Basin.
Although efficiency can be largely defined in terms of axial acceleration of the fluid, the action of a screw type propeller also imparts rotational motion or swirl to the downstream fluid which constitutes a waste of energy. The design of a large diameter propeller leads to the requirement for high torque on the shaft and low speed of rotation. These requirements not only pose problems for the propulsion machinery design, but also require that the hull resist the torque reaction of the propeller.
One way of combating large propeller torque is by use of coaxial contra rotating propellers. The combined action of the double propellers cancels out the rotational energy loss so that a greater propulsor efficiency can be obtained. However, there is a substantial penalty of complexity in the design of the coaxial shafting with its bearings and the need for elaborate gearing or secondary power source.9
Another method of combating torque is the use of stator blades mounted on the hull to introduce a counter-swirl to that produced by the propeller. While such mountings produce the desired straight thrust, the drag of such stators reduces the advantage while introducing undesirable engineering problems.
It may be seen by the description above that propeller design and placement entails compromises and engineering innovation that could best be tested by a vehicle such as the Albacore (SS-569)'°. The resulting hull shape and propeller design of USS Barbel (SSN-580) and USS Skipjack (SSN-585) incorporate the findings from experimentation by Albacore. American submarines produced in the 1970s and 80s had advanced propellers that were not only more efficient, but quieter in terms of cavitation.
Cavitation is the formation of gas bubbles in a flowing liquid where the pressure of the liquid falls below its vapor pressure. It may be visualized as the formation of low pressure resulting from fluid acceleration around propeller blades.' The faster the blades move, the lower is the downstream pressure. As it reaches vapor pressure the fluid evaporates and forms small bubbles of gas. When the bubbles collapse they typically cause very strong local shockwaves in the fluid which are audible to sonar. The noise produced by this cavitation is somewhat higher in frequency than the machinery noise in the interior of a submarine. For example, a five bladed propeller turning at 300 RPM has a blade rate of only 25 Hz, whereas electrical machinery such as a turbo generator often spins at some multiple of 50 or 60 Hz.12 Nuclear submarines are particularly vulnerable to sonar detection.
The problem of machinery noise, reduction gear whine and cavitation was attacked in the Navy's submarine silencing program and was incorporated in the Thresher design. Engineering efforts to reduce cavitation focused on the shape of the propeller. Submarine propeller engineers attempted to improve propellers by skewing rearward the tailing edge of each blade. A mean hydrodynamic pressure was used in calculations by assuming normal operating and transit depths.' Large scythe-shaped propellers alleviated the problem.
Rotational force in the slipstream vortex distributes rearward-trailing bubbles radially. Vapor-filled bubbles collapsing within this corkscrew-like trail do so in rhythmic patterns that reflect the particular construction of the propeller and correspondingly, the probable type of submarine housing the propeller. Experimentation revealed that blades could be radially pitched in three dimensions to include a rearward bend. This variance gave a propeller a complex shape resembling an open umbrella.
The three-dimensional skewed propeller with up to seven blades had the positive effect of producing thrust at lower rotational speeds. Lower spindle speed meant minimized blade vibration with a commensurate reduction in cavitation. Manufacturing such a complicated propeller was a challenge for American industry and could only be accomplished using computer-controlled milling machines. This type of propeller became America's standard nuclear-powered submarine propulsive device in the latter decades of the 20th century. The design was held in great secrecy by the Navy, but Toshiba sold propeller milling machinery with accompanying computer programming to Kongsberg Ltd. of Norway which in turn sold it to the Soviets. That government used the computer data to rapidly begin a program to imitate research of the US Navy. As a result American submarines were faced with Soviet opposites difficult to detect."
While cavitation bubbles collapse violently and thereby produce unwanted noise emission, air filled bubbles collapse at a slow rate and produce virtually no noise. Air bubbles introduced alongside a metal surface softens the sonar reflection. Fluid density is essentially unchanged in an air bubble cloud. It remains that of water, but the rigidity is that of air. The result is that the speed of sound in a cloud of air bubbles in water is a factor of almost 10 times slower than in water. Sounds within a submarine hull which would otherwise propagate for a long distance are reflected back into the hull and eventually dissipated therein. In the case of a propeller, the leading and trailing edges of the screw having small holes that emit air bubbles can dampen cavitation bubbles. Noise generated by submarine propeller blades is substantially reduced when forced air bubbles mix with cavitation bubbles.
Such a system is called the Prairie-Masker and was used on some GUPPIES in the 1960s. It was particularly effective when the submarine was snorkeling since it is necessary to either pull air from the surface or to use compressed air needed for other purposes within the submarine. Keeping a set of small holes in propeller blades clear from fouling while running submerged for lengthy periods can present an additional problem.'USS Barbel purges its Prairie-Masker with water spewing from its propeller blades. The Ralph Chatham collection.
A submarine propeller project of the 1990s was the innovative idea of using an external ducted propeller with multiple blades. This idea resembled in many respects the primary intake stage of an aircraft jet engine. Despite the difference between air compressibility and the non-compressibility of seawater, engineers examined the possibility of using a multi-bladed fan within a ducted ring to produce greater thrust. This type of propulsor has the advantage of a power-efficient slow rotation speed with accompanying reduction in cavitation. The circular duct which surrounds the impeller is shaped so as to accelerate fluid velocity through the impeller blades. The rotating element is protected by the ducted ring. Disadvantages include the cost of the propulsor installation, difficult mounting requirements, higher maintenance and greater weight at the extreme stern of the submarine. The American submarine Seawolf (SSN-021) and the Virginia class submarines are equipped with ducted impeller propulsors.' The Royal Navy's Trafalgar and Astute class submarines are equipped with pump-jet type propulsors similar to those of American design. Wikipedia reports that both the French and Russian navies also have submarines with impeller type propulsors.17"
One must ask what the next generation of submarine propulsive device might be. In the case of Tom Clancy's Red October a mysterious "caterpillar" propulsor is located inside the fictional Soviet submarine. Internal propulsor's may have a place in the future of American propulsion design since anticipated advantages might include better noise emission control and well-anchored stator blades that could produce smooth rearward thrust at the submarine's stern. In the meantime the Virginia class submarine will continue to represent the best in American propulsion design.Notes
- Dommett, W. E., Submarine Vessels, 6th Edition, Isaac Pitman and Sons, London, 1914, p. 17.
- Friedman, Norman, US Submarines Through 1945, Naval Institute Press, Annapolis, MD., 1995, p 22,39, 46.
- Burcher, Roy and Ryotell, Louis, Concepts in Submarine Design, Ocean Technology Series 2, Cambridge University Press, Cambridge, England, 1999, p 113, 114, 120.
- McKellar, Mark W., USS Albacore, A Revolution by Design. http://www.hazegray.orenavhistialbacore.htm
- Friedman, p 141
- Andersen, Paul; Kappel, Jens; Spangberg, Eugene; Aspects of Propeller Developments for a Submarine, Engineering Seminar, Trondheim, Norway, 2009.
- Interview with Ralph Chatham regarding Barbel's Prairie-Masker device and Jane's Underwater Warfare Systems, 2010.
- Friedman, p 142 and Burcher, p 117. 17 http://en.wikipedia.org/wiki/pumpjet.