View the original post on the IUSSCAA message board here

ADM Rickover: “If I could, I'd put the screw a mile behind the submarine.”

By Bruce Rule - Mar 28, 2014


Naval Research Laboratory SECRET Memorandum Report 3467 of March 1977 – declassified by US Government Memorandum 7100-038 of 26 February 2004 – is entitled “Very Low Frequency Acoustic Detection of Submarines.”

The introduction of this document (PDF Url: ADC010105) reads as follows:

(Quote) As a submarine propels itself through the water, it radiates very low frequency acoustic power which is caused by the interaction of the propeller's blades, the water, and the hull. As the propeller rotates, the water pressure on the propeller blades varies due to irregularities in water flow between the propeller and hull. This pressure variation is enhanced at certain frequencies by resonance in the submarine hull. The resulting narrowband radiated power can be detected by spectral analysis of the acoustic signature. These spectral lines are known as "blade rate lines". Blade rate lines occur in the frequency region of about 5 to 20 Hz.(end quote)

The NRL Report also includes the following statements:

(1) Line bandwidths and long term stability for blade rate lines are given in the Standard Threat Assessment. For the C, V and Y class submarines the blade rate line bandwidth is .004 to .07 Hz.

(2) The purpose of this (NRL) report has been to examine the feasibility of VLF detection of submarines. The source levels of auxiliary lines of Russian submarines will probably decrease. Since blade lines are harder to quiet, we may be forced to use detection of blade lines. But as indicated above, there are substantial improvements in detection ranges which may be possible if the acoustic environment is properly exploited.

(3) The Naval Research Laboratory has measured low frequency (VLF) transmission loss and ambient noise at sites in the Atlantic Ocean and Norwegian Sea. The results are used to calculate the array gain necessary to detect present and future classes of Soviet submarines at Atlantic Ocean and Norwegian Sea sites. The results are presented for submarine speeds of interest. The problem of array size needed to obtain necessary array gain and the problem of signal spatial coherence are considered.


The purpose of this posting – primarily for the benefit of active duty personnel - is to examine the influence of propeller design and submarine stern-area configurations on the radiation of blade-rate and thus explain why ADM Rickover made the statement provided as the subject of this posting.

When viewed from astern, a submarine's control surfaces are cruciform with the upper and lower rudders forming the vertical member while the port and starboard stern planes forming the horizontal member.

The flow of water aft along a submarine's hull is disturbed when these appendages are encountered. These effects alter the inflow experienced by propeller blades as they rotate through the areas of disturbed flow aft of the four control surfaces. These changes in flow-field direction and velocity produce changes in thrust developed by the blades. The vibrations induced by these changes in thrust are transmitted by the propeller shaft into the hull where they are radiated directionally (bow-stern directive with bow strongest because of the better “symmetry” of that radiating surface) as narrowband acoustic energy: blade-rate. The term “mechanical” blade-rate is often used to distinguish these hull-radiated signals from cavitation blade-rate which is radiated by the collapse of areas of reduced pressure (bubbles) directly aft of the propeller.

The magnitude (energy level) of mechanical blade-rate signals is strongly influenced by the propeller design: the number of blades and the shape of those blades, and by the distance of the propeller aft of the control-surfaces, the farther be better; hence, ADM Rickover's statement. (Get the propeller as far away as possible from the disturbed flow-field.)

Even numbers of blades are to be avoided because when one blade passes through the disturbed flow-field, the “opposite” blade is also passing through a disturbed area of flow because of the cruciform arrangement of the control-surfaces. This coincidence produces an enhanced vibration. Four-bladed propellers are the “worst” offenders because of the coincidence of each blade simultaneously passing through the disturbed flow-field aft of the four control-surfaces. A variation of the four-bladed propellers, the tandem propellers (two four-bladed props mounted on the same shaft but separated by less than half the diameter of each propeller) produce mechanical blade-rates for both a four-bladed propeller and an eight-bladed propeller.

The mid-line of each propeller blade is a straight line so that when that blade passes through disturbed flow-fields aft of a control-surface, the entire length of the blade experiences the effects simultaneously thus producing an enhanced vibration or pulse transmitted to the hull through the propeller shaft.

When discussing the PAPA Class Soviet SSGN, MORSKOI SBORNIK, a Soviet ship-building journal, described the PAPA as having “special eight-bladed (tandem) propellers.” (Later changed to single, five-bladed propellers on each shaft.) Numerous open-source documents described the installation of tandem propellers on VICTOR and CHARLIE Class Soviet nuclear submarines. The linked site (see below) shows images of these tandem propellers (not counter-rotating) including one view from directly astern of a VICTOR Class Soviet SSN. (Ninth low of images) This is a remarkable site which also shows images of Soviet, Chinese and even North Korean skewed propellers. The North Korean unit also has a five-bladed vortex dissipator (More on skewed propellers below).

The design of Soviet submarine propellers evolved to settle on seven-bladed configurations now known to be used on some DELTA, OSCAR, TYPHOON, AKULA, KILO, and the new YASEN Class (Project 885) Russian SSN.

The new single-shaft BOREY Class Russian SSBN appears to use a pump-jet propulsion system with rotors (moving blades) and stators (stationary structural support members for the shroud). If, for example, the BOREY has a seven-bladed rotor (call it the propeller) and 11 stators, the system may produce a narrowband source at eleven times the mechanical blade-rate or 77 times the propeller shaft-rate.

Several internet sites show images of the pump jet installed on a single KILO Class Russian SS – the ALROSA – now in the Black Sea Fleet, apparently as a test platform. The KILO pump-jet appears to have 11 rotors (blades) and perhaps 7 stators.

Getting back to the statement by ADM Rickover, it should now be obvious why, if feasible, it would be useful to locate the propeller as far aft of the control-surfaces as possible to avoid excitation of the submarine's hull by varying thrust forces created by the propeller turning in a disturbed flow-field aft of those appendages. Since the location of the propeller extremely far aft is not practical, skewed propellers have been be employed to reduce the magnitude of the vibration (pulse) induced in the hull – which produces mechanical blade-rate – by curving (skewing) the blades both port to starboard (outboard for two screw platforms) and fore to aft as shown by images previously discussed – especially the Russian DELTA III SSBN.

When viewed from the side, skewed propellers look like the “ribs” of a raised umbrella twisted with reference to the umbrella shaft. The result of this design is that, unlike the straight blades of non-skewed propellers, area of the blades of a skewed propeller pass through area of disturbed flow “gradually” which reduces the magnitude of the resulting vibration pulse that is transmitted to the submarine hull through the propeller shaft thus reducing the level of the mechanical blade-rate radiated by the hull. You might say that the vibration is “smeared” over time by the curved shape of the blade and reduced by the displacement of the blade tips further aft than would be the case with an unskewed propeller for which the blades are perpendicular to the shaft for their entire length..

Note: as shown by, the pump-jet propulsor of the British MK50 torpedo has 11 rotors (blades) and 13 stators.

Is there a down-side to the use of skewed propellers? Possibly. Limited data suggests they are not as efficient as unskewed propellers from a propulsion standpoint and may result in the loss of 1-2 knots at about 30 knots. That does not seem like much until one calculates the horsepower required to recover that loss of speed at the top end of the range. Remember, the PAPA Class Soviet SSGN could achieve 42-knots on 80-percent power but only 44.7-knots on 97-percent power and 44.97 knots on 100-percent power. Those additional increments in speed are very "expensive" at high-speed.