Submarine Research Center (SRC)
Note: The following comments regarding bulletin 25 has been received from a reader in Canberra, Australia: "Kockums did not build the COLLINS submarines for Australia. Kockums built two sections only for the lead boat, and the rest of that boat and the class were entirely built in Australia. The contract for the design and construction of the COLLINS class was awarded to the Australian Submarine Corporation, which sub-contracted the design work to Kockums, and the early fabrication work on the two sections in question to Kockums." Also a spelling note on Wartsile which should be spelled Wartsila and DCN for Directions des Constructions Navales. SRC thanks our friend from down under for the corrections.
Note: Bulletin 27 will deal with early U.S. missile submarines.
History of Development
Pre-Albacore submarines utilized propulsion systems that are known as diesel-electric where multiple diesel engines drove generators which in turn drove electric DC motors that shafted through the stern to dual propellers. Twin propellers have an advantage in maneuvering and are suited to submarines that spend most of the time on the surface. Nuclear propulsion gave submarines the ability to remain submerged for long periods and the emergence of the single, large screw meant slower rotation speeds and thus less cavitation.
The U.S. Navy equipped a few submarines with contra-rotating screws and accompanying contra rotating turbines within turbines. Propeller designs changed in response to the need for ever quieter submarines.
An interesting note comes from several sources who observed that when a nuclear power full bell is answered in a surfaced single screw, fast attack submarine (i.e., Scamp), the boat tends to rise in the water much the same way as a speed boat tends to lift its bow when at full throttle. It is referred to as riding the step and adds a few more knots to the surface speed.
Propulsion by means of something other than an external screw was initiated in the 1940s through the Navy's interest in reducing submarine noise emission. An exclusive interest in active sonar gave way to a predominant interest in passive sonar when it was discovered that German submarines detected American ships long before active sonar detected them. Keeping pace with improvements in passive sonar technology were the efforts by the Navy to reduce noise emission by every means possible.
Cavitation presented both the fundamental cause of noise emission and the most formidable problem to be solved. Propeller designers were convinced that noise coming from submarines came not from propellers but from other sources. Their reasoning stemmed from their confidence in theoretical propeller designs. It was found that cavitation was not on the propeller's blades, but rather a helical string of cavitation in the water trailing behind each tip of the blade. It became known as "tip vortex cavitation."
An intensive study followed using "wind tunnel" techniques. The first facility was called the Anechoic Flow Facility which accurately measured blade interaction within its medium. It was discovered that some noise emission was the product of boundary layer pressure fluctuations not dissimilar to an aircraft wing's stalling characteristic.
By the 1970s researchers had a reasonably good understanding of the physics related to noise radiation by open propellers. In 1978 researchers began to examine alternative propulsive systems with a view to incorporating such an innovative system into the forthcoming Seawolf. They went to work on the P5168 propeller design. They concentrated on the optimization of the blade surface and the curve of the intersections. Through precise measurements they were able to improve the tip design to reduce vortex, however the job was far from complete.
ONR began serious work on fluid mechanics involving fixed and moving boundaries within closed vessels. Would it be possible to reduce or eliminate vortex cavitation by controlling the blade tip boundary? It used the Model Design and Fabrication facility in Fulton, Maryland to test certain internal screw designs. In 1989 the Navy turned the Y-12 National Security Complex in Oak Ridge Tennessee into a propulsor development center.
In 1993 the facility delivered to the Navy its first prototype propulsor. The propulsor (as later modified into two alternative designs from the prototype) was approved by the Senate in 1998 through the Defense Authorization Bill. The Navy subsequently selected the design that best met the cavitation limitation goals. That design was wedded to the Seawolf hull design.
General Description of a Propulsor
Because of security it is impossible to describe the propulsor that drives the Seawolf and Virginia class submarines, however, an examination of propulsor design can give us a fair idea of what the system involves.
Water is, as a working hypothesis, incompressible. This means that we cannot compare a multi-stage, jet engine compressor (within a compressible medium such as air) with a marine propulsor even though certain similarities are striking. In its simplest form, a marine propulsor has an intake, an impeller (propeller, screw) and a nozzle all of which are inside the body of the submarine.
The intake of a ducted marine propulsor acts as a ram. The faster the propulsor moves through the water the greater the ramming effect to the impeller. A wide mouth intake funneled to the impeller maximizes the ram effect. A screen, grate and/or shredder guard against fowling the impeller blades.
The impeller has axial blades, each one of which is shaped into a curve not unlike an airplane's wing. The impeller can have three to as many as twenty four blades each with a pitch that drives water from the intake side to the discharge side of an enclosure called the pressure chamber. As the impeller turns, suction on the intake side reduces water pressure while water pressure on the discharge side is increased. The water in the pressure chamber is discharged through a nozzle at the rear of the propulsor driving the submarine forward.
The nozzle effects the discharge stream by being either constricted which accelerates a narrow channeled discharge or dilated which decelerates the discharge, but correspondingly increases the volume of discharge. Manipulation of nozzle orifice size in conjunction with impeller speed, controls the water pressure in the chamber.
Since cavitation is a function of water pressure, impeller speed and orifice size manipulation can ride a cavitation threshold curve that is responsive to ambient water pressure. Since cavitation occurs at the blade tips the bubbles spin in a helical extremity along the walls of the pressure chamber.
They are trapped by anti-torque vanes and imploded within the chamber. The shape and length of the chamber in combination with controlled water pressure determine the purity (lack of cavitating bubbles) at the nozzle.
Lower relative pressure on the intake side of the impeller reduces the effectiveness of the impeller. The pressure difference on each side of the impeller can only be minimized by increasing the pressure on the intake side since to reduce the pressure on the discharge side means a commensurate reduction in thrust. When the ram effect is insufficient a first stage impeller can be added to increase water pressure going into the main thrust impeller.
The impeller transports water in a straight-line rearward movement called thrust. It also spins the water as a reaction to the impeller's rotation. The spinning occurs on the intake as well as discharge sides of the impeller. This is torque. The most effective impeller maximizes thrust and minimizes torque.
Since torque is unavoidable the propulsor has within its discharge chamber radial, anti-torque vanes that quiet the rotational movement of the water. They are feathered at the center of the chamber and helically pitched at the chamber wall to counter flow torque. Water discharged at the nozzle is effectively non-rotational, pure thrust.
Propulsors attend to problems not found in normal marine screws. Cavitation in a submarine screw is controlled by rotational speed and blade shape/curvature. In a contained vessel an impeller (or more than one impeller in series) can be designed to produce a smooth rearward discharge and thereby reduce the noise emission of the propulsion plant. Astern movement can be achieved by reversing impeller rotation, however, as with the standard screw, efficiency is low.
Propulsors can be driven by electric motors such as in directional pods to help with maneuvering or by direct drive steam turbines. In the case of nuclear submarines a direct drive turbine can turn one or more impellers in series or propulsors in parallel. Problems of ducted marine propulsors include tip-gap size, blade tip geometry, blade loading and inflow structure.
The most significant problem is tip leakage vortex where high pressure water on the discharge side of the impeller tends to surge backwards into the lower pressure intake side. Another problem is sub-visual cavitation where the spinning blades create microscopic bubbles that collapse in the helical down flow.
Aircraft jet propulsion design now incorporates vectored thrust nozzles that swivel to provide directional thrust. Of course, directional or vectored thrust nozzles are used in rocket propulsion. The possibility of Seawolf and Virginia propulsors using vectored thrust is not known, however, a Seawolf model in the David Taylor Model Basin clearly shows rudder and stabilizer appendages which would be redundant in a vectored thrust propulsor.