Diesel Electric Systems

Revolutionary propulsion

Like most new technologies, marine diesel electric systems were introduced in the navy long before they were shown to the commercial industry. Their major advantages are low noise and vibration disturbances, lower energy consumption and higher flexibility in the ship design. Numerous diesel engines can be connected in parallel in sound and vibration isolated rooms, almost anywhere on the lower decks of the ship.

These diesel generators are then responsible for the electric ship propulsion, the heating and any other electrical utilities on board. Running diesel-electric engines at a stable load allows smoother transients and constant speeds. While at an efficient load, marine diesel engines also tend towards optimum fuel consumption, thereby reducing emissions and the impact on the environment.

Advantages of Diesel Electric Systems

It should be mentioned that the diesel electric system is extremely valuable for ships with low average speed (10 knots), such as cruise liners. This technique of combined diesel electric systems gains importance when the installed power generating capacity can be used for various ship functions, and different situations such as that needed for passenger services (i.e. electricity, heating).

The safety aspects of diesel electric systems are commonly regarded as being related to redundancy in different ways. The number of electric power-generating units is large enough to ensure propulsion capability and steerage way irrespective of any component failure. In addition the diesel and electric units can be located in different compartments to safeguard against loss of power in case one compartment has been destroyed by fire or flooding. This flexibility also allows the optimisation of cargo space volume and arrangement.

The shipbuilding industry has recently introduced a revolutionary new propulsion system known as PODS in which the shaft has been replaced by a bulb-shaped propulsion unit driven by an electric motor. The only connection between the bulb and the ship is therefore electric cabling. This design significantly reduces wake effects, and propulsion efficiency is improved by as much as 10%. Two PRODIS projects are dedicated to this new system. OPTIPOD is looking at the impact of PODS on ship design and attempting to optimise the shape of the stern. PODS INSERVICE is a support project for OPTIPOD and any other future related research in that it is collecting data from the existing POD fleet. At present Finland has four passenger ships and one ice-breaker/tanker in-service using this new technology. The French company, Chantiers de l'Atlantique has also just delivered a POD vessel called Millennium - using two 20MW PODS, this vessel represents the largest scale use of the technology to date.

Project goal

The final goal of the project is to develop guidelines for the design of podded ships. The intention is to have these guidelines as general as possible, but development work has been focused on four ship types: cruise liner, ro-pax ferry, products tanker, and supply ship. The hydrodynamic development of these ship types is the responsibility of those four consortium partners with model basins: HSVA for the cruise liner, CTO for the ro-pax ferry (as outlined in The Naval Architect's Polish report in July/August this year, page 15), SPA for the products tanker, and VTT for the supply ship. The structure of the project is not, however, based on these ship types but on the following issues, which form eight technical work packages:

• Hydrodynamics• Safety and risk analyses• Structural safety• Impact on environment• Operational aspects• Effects on general arrangement• Cost/benefit evaluation

The last work-package, which deals with the development of guidelines, will mainly be carried out at the end of the project. The other activities run more or less in parallel, withinput/output from one work-package to another.

As can be seen from the list above, the goal of the investigation is to have a very broad base for forming the guidelines. When the project was formed in the spring of 1999 there were many questions regarding podded ships: Could the layout of the ship be more effective, ie, could there be space savings when moving from a conventional propeller-shaftarrangement with rudders to a podded solution?

• Is it possible to improve the overall efficiency of all podded ships compared with conventional?

• How important is the increased manoeuvrability for different ship types using pod propulsion compared with conventional?

• What other benefits or drawbacks can be found using pod propulsion?

The major users of pod units have been cruise liner operators. Several partners have significant experience with using pods for these ships, which is a very good start for the project. The reason why the cruise liner was the first ship type to use pods (although first prototypes were fitted to waterway service craft and an Arctic tanker) is simple; several of the uestions above had a very positive answer for these vessels. However, even for cruise liners there is much development work to be done before finding the 'ultimate design'. As mentioned above, the goal of the project is to develop guidelines for a range of ship types that can benefit from a podded propulsion solution as compared with a conventional one. As can be seen from the work-package list, the investigation will cover the main design tasks that shipyards have to deal with in the initial design of a vessel. The different parts of the work scope are explained below in a little more detail.

Hydrodynamic interaction

Looking back over a number of decades, there have not been so many improvements in hull design in the low- or medium-speed range. The bulbous bow arrived in the 1970s and has been further developed since then. Twin-skeg hull form designs were also developed during that decade. The introduction of pod propulsion, which will allow the propulsion unit to be placed without considering any shaft arrangements or space for machinery will, of course, give the naval architect many new opportunities to design the 'ultimate hullform'. The propeller can be placed more freely in the longitudinal plane and tilted so it will face the up-flow in the stern. To some extent, clearances between the propeller tip and the hull can beoptimised easier than with a conventional arrangement. The traditional balance of propulsive factors such as effective wake and thrust deduction together with naked hull resistance will in some way give a new dimension to design work. The possibility of creating a more homogeneous inflow to the propeller will also improve cavitation properties and allow for higher speeds or reduction of vibration and noise from the propeller. The risk of focusing on efficiency can be

that the sea-keeping properties and maneuverability, including course stability, are put aside.

Safety and risk analyses

From the development work done for current cruise liners, there are no indications that there should be any general problems with course stability, but, in fact, the current general knowledge about manoeuvrability of podded ships is limited. In OPTIPOD, effort will be focused on the development of mathematical models for manoeuvring with pods. Input to these models will, of course, come from model tests with the four ship types mentioned above. One of them, the products tanker, will only have one pod unit. This means that there will be possibilities to compare the behaviour of a single-screw conventional ship with a single-pod ship.Even though manoeuvring simulation models will primarily be based on results obtained from model tests.

Effects on general arrangementWith the freedom to place machinery in the most favourable position onboard, studies on the effect on general arrangements will be carried out for the two most interesting ship types: the ro-pax ferry and the cruise liner. State-ofthe- art design methodology as well as the latest3D virtual-design tools will be used.

Cost/benefit evaluationObviously, there is great interest in podded propulsion from owners of cruise liners, judging by shipyards' orderbooks. Will there be the same interest for other ship types? This is probably totally dependent on the overall economy of podded concepts. It is important to clarify benefits in, eg, enhanced harbour manoeuvring as well possible drawbacks (such as maintenance issues, demands on crew, and sensitivity to rough seas).

PART II

PROPULSION SYSTEMS

In the following chapters we will describe three examples of diesel propulsion, and argue the pro's and con's of the systems and describe a way to solve the previous mentioned light

Conventional diesel propulsion with fixed pitch propellers

Conventional diesel propulsion with controllable pitch propellers Diesel electric propulsion with Pods.

Conventional diesel propulsion with fixed pitch propellers

Pros and Cons

Pros:

it is a usual system and less expensive system.

Cons:

it is necessary to run three engines (two prime movers and one generator);

to promote long engine life the yacht has to sail 13.8 knots or more.

Solution to solve a part of the light load problem:

Install two smaller engines of 660 B.HP each. As can be read in table 3 two engines of 660 B.HP are enough to attain 15 knots. At a nice cruising speed of 13 knots the engines are running at 50 % load.Use of an father/son gear box installation gives better load figures at lower speed, but the cost of such an installation is enormous and therefor left out of this comparison.

Conventional diesel propulsion with controllable pitch propellers

Pros and Cons

Pros:

up to 13.0 knots, one engine can be shut-off and the propeller can be set in feathering position.at 11 knots the engine is running at 50% load;

engine rating curve and propeller rating curve runs equal, by adjusting the pitch;

perfect maneuverability, special at slow speed (no need of trolling valves).

Cons:

low propeller efficiency (ETA-P = 60-61%); more expensive; more maintenance; above 13.0 knots still three engines are running.

 

Summary of the principal disadvantages of conventional diesel propulsion

light load problems; mostly two prime movers and one generator running; engine room lay-out not flexible; no possibility of sound enclosures around prime movers; extensive sound insulation and ventilation system. problems with prime movers can lead to hazardous situations, there is no

back up.

Which options are available to develop a new configuration without the above described disadvantages? The answer should be found in diesel electric propulsion.

Integrated diesel electric propulsion

There are two types of electric power - AC or alternating current and DC or direct current. A basic AC motor operates at a constant speed determined by frequency. A DC motor speed will vary with the applied voltage.A plain rectifier converts AC power to DC power and the DC voltage will be about the same value as the AC. A SCR controls the conversion of AC power to DC power so that a variable DC voltage from the constant AC voltage can be obtained.This provides an easy way to control the speed of a DC motor (analogous to a rheostat for light dimmer). A conventional ship will have four or five diesel engines with one engine dedicated to each propeller and the others dedicated to ship service power.

For a diesel electric ship, the power bus bar arrangement has many inherent advantages, including:

all vessel functions can be supplied by one or more engine/generator sets feeding this common bus bar;

the common bus bar feeds the ship’s service switch panel and both SCRs from where the speed controlled E.motors revolve the propellers by using either conventional shaft installations with fixed pitch propellers or shaft installations with controllable pitch propellers or Z-drive thrusters.

each variable speed function can be precisely controlled from zero to top speed by the individual SCR. The phrase "from zero speed", is important. With a standard mechanical drive the minimum propeller r.p.m. is determined by engine idle speed (only with the aid of a trolling valve, it can be reduced);

diesel generator sets can be put on line or removed as needed by the total system load, without interrupting any operating load. This feature permits fuel conservation when full power is not required;

on-board engine maintenance can be accomplished without laying up the ship or losing any system function;

it is not possible to overload an engine even if the throttles are put full ahead with only one engine running. This feature is provided by a power limit circuit that compares the power available from the generator(s) that is (are) running with the total power being used by all loads when the load reaches 95% of available capacity, the SCR fed loads are restricted to prevent overloading.

In stead of having one engine per load, engine/generator sets now channel their power into a common reservoir, the power bus bar. As power is needed to turn a propeller or to operate other vessels loads, that amount of power is pulled from the bus bar and sent to the bowthruster, galley, air-conditioning motors or any other electrical load.An extensive power management and load sharing system will regulate all of this.With one or more engines shut off, the engine(s) being used will be operating at a higher load.Therefore, they are more efficient and use less fuel per horse-power hour delivered. The result is that the diesel electric boat will use less fuel than a direct drive boat to perform the same mission when less than full horsepower is required. Here is where one of the big advantages of electric drives become evident. Even when only one engine is running, you can still operate both of the propellers in forward or reverse in any combination of speed relationships within the horsepower capability of the one engine and supply ship service power requirements. Again, the pilot cannot overload the engine.

Some other advantages of electric drive: the engine(s) always operate between 50 and 100% load; the engine(s) always operate at constant speed and will, therefore, accept

load changes much more quickly;

the engine(s) produce less black smoke and other pollutants when accelerating the propellers;

more economical use of the fuel being fed to the engine; the speed of the propeller(s) can be changed much faster; reversing will be faster; electric propulsion motor (and bowthruster) can be operated at 1% speed

with no time limit up to and above full rate speed; gear boxes are simple reduction units with no reversing and no clutches; engine room lay-out is more flexible;

Pros and Cons for AC-AC propulsion

Pros:

an AC propulsion motor is of simple construction, no brushes; with an AC system the power factor (cos.phi) is high >0.94 and constant,

also at low speed, therefor less kVA's have to be installed; the efficiency of an AC system at lower speed does not decrease as much

as a DC system; due to a diode rectifier for the AC-DC conversion the disturbance of the

sinusoidal is lower compared to AC; by the application of an AC system in 12-pulse execution a total harmonic

distortion (THD) to less than 5% and any single harmonic to 3% can be reached without the use of expensive filters. This also is partly possible by the AC-DC system;

Cons:

the ambient temperature for the inverter (DC to controllable AC) may not exceed 45° C. This can be solved by application of an air/water cooler;

an AC-AC system needs more volume and weight; the sound level in the high frequency range of the asynchronous motor is

more compared to a DC motor.

General note:

AC-DC with a 6 pulse system may require somewhat larger generators to handle the higher harmonic currents. To develop a 12 pulse system it is necessary to utilise phase shifting transformers and double SCR bridges. In that case the weight and volume of an AC-DC and an AC-AC system are almost the same.The client finally, depending on his sailing area, will make a decision for AC-DC or AC-AC. This decision will often be directed by service possibilities etc.

Schematic drawing of a conventional diesel propulsion and an integrated diesel electric propulsion.

Extra benefits of the diesel electric propulsion in combination with Z-drive thrusters

the engine room lay-out is flexible; the Z-drive thrusters are flexible mounted in a bottom well under a hatch

for easy dismantle and maintenance; no need of difficult engine beds, shaft drives, struts and rudders; each generator set can be installed in an airtight sound enclosure. The

engine air intake will be ducted with a silencer directly from an outside air intake box. The heat rejection of the genset will be absorbed by a water/air fan cooler. The sound enclosure will reduce the noise level with at least 45-50 dB.A. Each generator set will be mounted on air mounts in the sound enclosures. Each assembly is double mounted on the engine bed;

with less sound insulation the sound level in spaces above and in front of the engine room can be reduced to less than 50 dB.A.; a conventional diesel propulsion yacht needs an extensive sound insulation to achieve a sound level of 57-60 dB.A directly outside the engine room.

due to the fact that the combustion engines are placed in airtight and cooled sound enclosures the ambient temperature in the engine room is low and need maybe 10% of the normal air flow, which results in only a small engine room fan, small ducting and less noise at the outside air intake and air exhaust grill

outstanding maneuverability, particularly at slow speeds; Further more a computer linked to a GPS guided dynamic position system

controls the Z-drive thrusters and bow thrusters to enable the yacht to remain stationary at sea;

less man hours for the total installation.

Summary of benefits of diesel electric propulsion.

the possibility to run with one prime mover, which serves both propellers and the ship's service at the same time.

with one or more engines shut off, the engine(s) being used will always operate between 50 and 100% load, which promote long engine life.

the back up of three generators for propulsion. less smoke and other pollutants. the propulsion motors and the bowthruster motor can be precisely

controlled from zero to top speed. it is not possible to overload any engine. the speed of the propellers can be changed much faster. more economical use of the fuel being fed to the engine. engine room lay-out is more flexible. extreme low noise level in- and outside the engine room. low engine room temperature. outstanding maneuverability, particularly at slow speed.

Project Title: ESMA - Efficient Ship Machinery Arrangement Project

Shafts vs. Pods, Comparison between a conventional shaft line and a podded drive

Conclusions

A podded drive has got clear and well proven advantages in manoeuvrability and hydrodynamical efficiency. It also has got advantages in space usage, weight and production efficiency which magnitudes depend on a case.

Compared to a conventional shaft line with a diesel electric power plant a podded drive is today more expensive. A podded drive is a new product and it remains to be seen how the price of the podded drive developes when more units are delivered by several suppliers. Without predicting how splitting the design costs on many units or new suppliers’ desire to enter the market will influence on the price of the podded drives, it should be expected to decrease. Both suppliers and yards can be required to improve their productivity along the experience. Reaching the same price level than conventional shaft line in a few years is a reasonable expectation. This with the advantages of the podded drive will make it very competitive.

Objective

Objective of this study was to summarize actual differences in costs, space utilization, manoeuvring characteristics and performance of a podded drive and a shaft line.

The study was carried out by comparing Fantasy class cruisers with different propulsion systems to each other.

Shafts vs Pods

Building costs

Building costs were divided to material and labour costs. A pod unit has got a significant amount of propulsion system parts preassembled in it. Thereby in a pod unit more labour and design are included in material costs than in case of shaft line. This raises the material costs of the podded drive. On the other hand the more compact design of the podded drive should lower the overall material and installation costs.

Difference in material costs consists of replacing propulsion motors, shafts, bearings, sealings, propellers, castings of the bosses and the shaft supporters, rudders and their machinery and stern thrusters with pod units and their turning, cooling and power supply appliances. The material costs of the podded drive were in case of the ships in this study 19% higher than the costs of the material that they replaced. Between the dates when the corresponding materials were delivered to ships that were compared, a relevant producer price index raised 5,8%. When inflations share is subtracted the difference in material costs decreases to 12.4%.

Labour costs were expected to sink radically because of the podded drive. Savings in labour costs were yet smaller than what was estimated covering 20% of the raising of the price caused by material costs of the podded drive.

A choice of propulsion system has got an impact on other systems also such as cooling, ventilation and lubrication. These differencies don’t seem to cause important costs. However space saving and lack of shaftline eases lay out design which provides potential to weight and costs savings via more compact engine room arrangement and clearer pipe arrangement when a podded drive is chosen.

A podded drive transferres work and tests to be done in workshops instead of to be done on board. This binds more capital earlier but enables to shorten the passing through time. Workshop hours are much more productive than working hours on board.

Since a podded drive has got better overall efficiency, a power-plant with a lower nominal output can be chosen, than in case of shaft line, without raising the utilization. This reflects to almost all of the auxiliary appliances to be rated lighter as well.

To give an image how the building costs of a podded drive and a shaft line relate to each other, some figure has to be presented. Counting in only directly on propulsion system connected material and labour costs results podded drive being 10% more expensive than shaft line.

Operating costs

Producing and transferring electrical power from generator to the shaft of the propulsion motor is approximately 1,5% more efficient in case of shaft line. In order to gain good hydrodynamical efficiency, the diameter of the propulsion motor in pod has been reduced which leads to stated 1,5%. On the other hand, transferring mechanical power from propulsion motor to the propeller shaft is approximately 1,5% more efficient in case of podded drive. Therefore shaft line (power plant) and podded drive can be regarded equal when comparing the power transferring efficiency (PD/PB).

Comparing sea trial measurements and model tests predictions of the Fantasy class cruisers suggests that a podded drive can be expected to have 5-9% better hydrodynamical efficiency than shaft line. Also bigger figures have been presented by some suppliers of podded drives. These figures need to be taken with care until more full scale information is available.

Space requirement

The location of the main engines is defined by the funnel and distribution of the bulkheads. Main engines are positioned at the same time with the propulsion motors whose positions and installation angles are derived from the shaft arrangement. Long under water part of the shaft line is expensive to build and it disturbes the wake field. The attempt to shorten the length of the shaft line increases the angle of the shaft line which

lifts the propulsion motor higher and turns the propeller plane to an worse angle from hydrodynamical point of view.

When choosing a podded drive the problem of positioning the propulsion motors and the shafts does not exist and positioning of the main engines has got more latitude.

When main engines (and propulsion motors) are positioned, the rest of the engine room is built up around them. In case of a podded drive there is considerably more space available (which in addition should be more efficiently arranged by using the advantage of increased latitude in positioning the main engines.)

Cyclo transformers which transform the voltage supplied by the main switch boards suitable for the cyclo converters and the cyclo converters which control the power of propulsion motors are posiotioned in M/S Elation and M/S Paradise at the same spot as in their sister ships , that is, on tank top next to the room where propulsion motors used to be positioned. This was due to M/S Elation and M/S Paradise being conversions in which only the necessary changes were made. Today these appliancies are located outmost aft one deck below the deck of the pod room. The goal is to supply the electricity as far as possible with the higher voltage Besides saved money in cabling ,this means space consumed on more valuable deck.

In lower decks podded drive saves space but on deck above the pod unit it consumes more deck area than it saves. Rudder machinery room is replaced with rooms which contain turning, cooling and power supply appliances. These appliances require more than double the space of the rudder machinery. This deck is considered much more valuable than tank top or the first deck.

In Fantasy class cruisers the spaces that the new arrangement with podded drive released were used to increase waste treating capasity (195m2), for machine stores and spaces (64m2) and for fresh water tanks (150m3). The price was 84m2 lost provision store spaces out of which most was changed to pod room.

Weight

Weight savings were in the ships that were compared in this study in scale of one percent of the light weight of the ships.

Rating the power plant lighter saves weight 9.8-16.5 tons/MW (only diesel engine) supposing that the type of the engine does not change. Changing the type of the engine may have significant influence on weight. The weight of the generator depends on the angular speed. The higher the speed the lighter the generator.

Building procedure

Replacing shaft line with podded propulsion transferres work and testing to workshops. It makes the aft part of the hull more simple and erases the shaft supporters and bosses.

Assembling of the podded drive starts by lifting the pod units to their places and fastening them. Then turning, cooling and power supply accessories are built on top of the pod units and the outfitting of the pod rooms is finnished.

Manoeuvrability

In manoeuvrability a podded drive is without doubt superior to a conventional shaft line. This can be seen by comparing the seatrial measurements of all of the Fantasy class cruisers.

Good controllability of a ship with a podded drive can be seen in results of the zig zag tests of M/S Elation on sea trial.

Noise and Vibrations

Measurements at the sea trials of the M/S Elation and M/S Fantasy show that pressure pulse amplitudes were lower with podded drive, some of the difference resulting from the slightly bigger hull clearance in M/S Elation. In degree of vibration and noise, measurements indicate significant diffrencies for advantage of a podded drive.

Wärtsilä the Ship Power Supplier (Forts.)

Ultra-large container carriers (>10,000 TEU) with single-screw propulsion plant? Engines for ultra-large container vesselsAs mentioned in an article in this issue before, the containerised sea transport market has been growing at a much faster rate than any other major sea transport sectors. With the increasing number of containers transported, the maximum container vessel size has also been growing, and the largest vessel today measures close to 8,000 TEU. The maximum vessel size is expected to keep on getting larger, owing to the benefits from economies of scale. However, the existing machinery configurations are not adequate for the increasing power demand of the super container vessels and these therefore need new machinery products and concepts.Wärtsilä is continuously developing new solutions and products to meet the future demands of the shipbuilding market. As an example of this work, two machinery solutions for future ultra-large container vessels are described in their recent »Marine News No. 1/2002«. Wärtsilä had already presented alternative propulsion concepts ranging from traditional single screw arrangements to more complex hybrid solutions

featuring mechanical propulsion combined with electric pod drives.The maximum power of a RTA 96 C two-stroke diesel engine is 5,720 kW/cyl. For the time being cylinder in-line engines were fitted with up to 12 cylinders generating 68,640 kW. Plans exist to build them also with up to 14 cylinders producing 80,080 kW. But such an engine reaches a weight of 2,300 t being close to 27 m and 11 m high. This might be too much in dimensions and also in weight. There are various disadvantages of this single-engine arrangement which can be compensated by a double-plant configuration. But always when comes to weight problems the hybrid solutions might offer changing advantages. One big problem for the huge two-stroke engines is part-load running of e.g. only 20% when manoeuvring at low speeds in port entrances.To overcome the manoeuvring problem at part-loads and low speeds best, a hybrid solution of a single-screw driven by a huge two-stroke diesel engine in combination with a diesel-electric podded drive behind the propeller instead of a rudder gives a lot of all advantages together. The maximum power from a single diesel engine could create around 70 MW. If 100 MW is needed to propel a – lets us say – 10,000 TEU container vessel some additional 30 MW might come the diesel-electric pod. With these 30 MW alone extreme good manoeuvring conditions can be reached as the pod can be turned to all wanted directions.

Steerable Propulsion Units: Hydrodynamic issues and DesignConsequences

by Tom van Terwisga, Frans Quadvlieg and Henk Valkhof (MARIN, Wageningen, TheNetherlands)Paper written on the occasion of the 80th anniversary of Schottel GmbH & Co;Presented on 11 August 2001

Introduction

The last three decades have shown a strong development in the market for steerablepropulsion units. This paper addresses several main developments and places them in ahistoric perspective. The major objective of the paper is to present a review of issues relevantto steerable propulsor units. These issues are essentially of a hydrodynamic nature. Althoughit is thought that hydrodynamic issues often have a heavy impact on the design, theprofessional background of the authors rather than anything else prompts the choice for an

emphasis on hydrodynamic aspects. Starting from the hydrodynamic aspects, we drawseveral conclusions towards the design and operations of vessels equipped with steerablepropulsion units.Steerable propulsion units refer here to those units that are able to actively deliver a steeringmoment by rotating the thrust vector through the rotation of the thruster. Such propulsion unitsmay occur in different concepts. The most renowned example and one of the oldest productsin this range is the steerable thruster unit (Figure 1).Recently, since the early nineties, a distinct concept has made its way into the marine world.This new concept is referred to as podded propulsion (or in short: pods) and is distinguishedfrom the original thruster in that its prime mover is an electric motor, situated in the hubunderneath the strut, directly driving the propeller (Figure 2).Figure 1 Steerable thruster unit

Figure 2 Podded propulsorApart from the steerable thruster and the pod, a number of other steerable propulsor unitsexist. One of the oldest is the Voith Schneider Cycloidal propeller (see Figure 3). Thispropeller is characterised by a number of foils rotating about a vertical axis, with a blade anglethat depends on the blade position. The blade angle is controlled by a mechanical actuatormechanism, which essentially determines the thrust/torque ratio in every position.A special type of waterjet that is worth mentioning is the Schottel Pumpjet, whichdistinguishes itself by the combination of intake, pump and nozzle in one rotatable unit (seeFigure 4). The gain in space and the consequent flexibility in the ship design are obvious.Page 2 of 15Figure 3 Voith Schneider PropellerFigure 4 Schottel Pump JetThe paper first aims at providing some historic background to the development of steerablepropulsion units. This is followed by a discussion on hydrodynamic issues and designconsequences for perhaps the two most popular steerable propulsors: the steerable thruster

and the podded propulsor.Historic developmentAn early example of a propulsor applying the principle of the vectored thrust is the GermanVoith Schneider Propeller. The development of the Voith Schneider Propeller started in 1926and the first application powered an inland waterway vessel in 1929. The first tug with theVSP Cycloidal propeller installed (Figure 5) was launched in 1950.Figure 5 First VSP Tractor Tug in 1950Schottel has played an essential role in the history of azimuthing thrusters. Some 50 yearsago, Schottel introduced the Schottel Rudder Propeller SRP. This rudder propeller could berotated over 360 deg (vertical axis), where the full propulsive power could be used for anyangle (Figure 6). These days, azimuthing thrusters are available up to some 6 MW, allowingfor a wide range of applicability (Figure 7).Page 3 of 15Figure 6 First Schottel Rudder Propellerlaunched in 1950Figure 7 Largest Schottel RudderPropellerThe popularity of steerable thruster units can be explained by the various applications ofDynamic Positioning (DP) or Dynamic Tracking (DT), both in the offshore industry, as well asin other areas of seagoing activities. The conventional thruster unit which is widely applied inDP/DT applications, makes use of a mechanical power transmission, where the prime mover(mostly a diesel engine or an electric motor) is connected with the propeller through one ortwo right angle gears (designated respectively L or Z drive).The increase in popularity of the conventional thruster in the early seventies was caused byseveral factors, according to Nienhuis [8]. "In the offshore industry activities were shiftingtowards increased water depths which in some cases prohibit the use of conventional passivemooring systems. The flexibility and mobility of DP systems led to its application for theexploitation of marginal oil fields, with the added advantage that assistance of anchor

handling vessels is no longer necessary. This latter advantage is also beneficial for cable orpipe laying vessels, which nowadays may be fitted with dynamic tracking (DT) systems.Indeed there seems to be a trend for oil companies to require the use of actively controlledships in the vicinity of subsea pipe lines to avoid the risk that these may be damaged by theuse of anchors." Other applications of DP or DT systems can be found in dredging vessels(e.g. trenching, stone dumping, beach replenishment) and naval ships (mine hunters inhunting or hovering mode, frigates in mine sweeped areas, replenishment at sea operations).The number of applications of the rotatable thruster for other ship types also grew. This wasa.o. promoted to a large extent by Bussemaker [1], who proposed tractor tugs with azimuthingpropellers. In the mean time the application of rotatable thrusters has grown to many othership types, such as double-ended ferries, stern drive tugs, inland passenger ships, minehunters and offshore workships.The traditional stronghold of the azimuthing thruster is the application where goodmanoeuvrability at low speeds is essential, such as e.g. for DP and DT. With the maturing ofthe concept of the azimuthing thruster and the availability of electric motors with a high powerdensity, the thruster with an electric motor in the pod came within reach. The first so-calledpodded propulsors, using this design principle came into service in the early nineties. One ofthe main assets of this podded propulsor is probably that it has important consequences forthe general arrangement of the ship as well, because of the different layout of the propellershafting-engine chain. Other important aspects refer to the overall propulsive efficiency andthe manoeuvrability.The idea of placing the electric propulsion motor inside a submerged azimuthing propulsorarose in the late 1980s by Kvaerner Masa-Yards, together with ABB Industry. A 1.5 MW unitwas first installed in 1990 on the Finnish waterway service vessel Seili [7].

Over the last five to six years, podded propulsors have become more and more important.Particularly on cruise liners, the units have proven to be of major importance as a means toreduce cavitation and vibration and hence have lead to a new standard for a high comfortclass of cruise ships. At MARIN, it all started with the request of Kvaerner Masa-Yards andPage 4 of 15Carnival Cruise Lines to compare the results of the twin screw open shaft Fantasy class ofships with similar ships provided with pods.Based on encouraging results with pods as main propulsor, Carnival Cruise Lines decided toselect ABB Azipod® propulsion on the last two passenger cruise ships of the Fantasy class."Elation", delivered in early 1998 from Kvaerner Masa Yards' Helsinki Yard was thus the firstcruise ship fitted with electric azimuthing propulsion units. Two units were installed withpulling propellers in the front end of the pods. The electric motors feature a power output of14 MW each and a rotation rate range from 0-146 rpm. At present, the largest podded drivesthat are offered by the industry go up to powers of about 30 MW.The podded propulsor (with the electric motors placed in the pod) have proven to offer anumber of benefits, "such as a remarkably increased manoeuvrability. The crash stop forinstance was half of the original, and the vessel remains manoeuvrable during a crash stop.Other benefits are less fuel consumption, reduced engine room size and flexible machineryarrangement, as well as low noise and vibrations. The need for long shaftlines, conventionalrudders, CP-propellers and reduction gears are eliminated, resulting in space and weightsavings and reduced need for maintenance." [7].In the meantime, all major propulsor manufacturers have developed their own poddedpropulsor (Figure 8, Figure 9 and Figure 10). A noteworthy deviation from the mainstream poddesign is the Siemens - Schottel Propulsor (SSP, see Figure 11).Figure 8 Azipod from ABBFigure 9 Dolphin from John Crane-LipsFigure 10 Mermaid from Rolls-Royce

Kamewa Figure 11 SSP from Siemens-SchottelThe hydrodynamic design of the SSP is characterised by two propellers, rotating in the samedirection. By dividing the total thrust over two propellers, a number of potential advantagesoccur:The mass flow through the propeller disk is increased when compared to only onepropeller. This is caused by the contraction of the streamtube due to the acting propellerwhen going downstream. The wake of the first propeller at the downstream propeller hasPage 5 of 15therefore a diameter that is smaller than the propeller diameter. Consequently additionalmassflow is ingested, leading to a higher efficiency.The loading over the blades is lower when the thrust is divided over two propellers,causing improved cavitation characteristics. Alternatively, at comparable cavitationbehaviour, the blade area ratio can be decreased which decreases the frictional dragcontribution to the torque.At a lower loading per blade, there is room to decrease the propeller rotation rate, alsoresulting in smaller frictional drag contributions to the torque.Decreasing the propeller rotation rate leads to larger rotational losses in the wake. Theselosses can largely be recovered when a proper stator (such as the stator fins and the struton the SSP) is placed, downstream or upstream of the propeller.The above tendencies may lead to improved powering performance, which is almost always atrade off between efficiency and the risk of vibration hindrance and erosion. These potentialadvantages do however not automatically lead to an improved overall performance of theship. Much will depend for example on the constraints with regard to propeller diameter.A derivative of the hydrodynamic considerations is that it will be important to have a highpower-density electric motor. This motor should be able to operate at low rotation rates, or atthe same rotation rate at a reduced pod diameter. The SSP was the first podded propulsorfitted with a Permanent Magnet Motor, allowing for a high power density

Thrusters: Hydrodynamic issues and design consequencesThis section touches upon some of the more dominant hydrodynamic issues in the designand operation of thruster units: thrust effectiveness, maximum thrust density andmanoeuvrability.

Thrust effectivenessPerhaps the most important issue in DP, DT or low speed manoeuvring is knowledge on theeffective forces that the thrusters exert on the ship in the encountered conditions. Theseforces determine the thruster effectiveness for a given input power and consequently affectthe selection of the type of thruster, its size and the overall thruster layout.Thruster effectiveness does not follow simply from a consideration of a thruster in open waterconditions. The thruster always operates in the vicinity of the hull and of other propulsors inan environment determined by waves and the motions of the ship.Nienhuis [8] acknowledged the following disturbing factors: "In the first place, unsteadyconditions are inherent due to the low-frequency motions, the variable thrust vectors as wellas the first-order ship motions. Secondly, the low speeds encountered may lead to inflowdirections, which deviate significantly from the alongship direction. Further, it may beexpected that other propellers operating in the vicinity of the considered thruster or propellerwill not only alter its effective inflow velocity, and hence its thrust, but will also affect the netforce which this thruster exerts on the ship. Next, the effect of wind and waves, which moreoften than not dominates the current, leads to thrust levels of the propeller which are not inbalance with the current forces. This is similar to a tug in towing condition. Finally, restrictedwater (shallow water or the presence of quays) is often encountered, changing theperformance of the propulsion devices.”These phenomena all combine to the fact that for a proper design and operation of a vesseloperating at low speeds, it is not sufficient to know the bollard pull of each of the propellers.Still relatively little knowledge is available for conditions inherent to DP, tracking or low speed

manoeuvring. These conditions being:low propeller inflow speedsdrift angle varying from 0 to 360 degreesthrust vectors largely uncorrelated with the current force vectorwidely varying propulsion arrangementsrestricted waterPage 6 of 15unsteady dynamic behaviour as a function of waves and low and high frequent shipmotions."The effectiveness of a whole DP/DT system is often presented in a so-called DP capabilityprediction. Such a prediction aims at providing the sustainable conditions for a ship with agiven thruster configuration. The sustainability is then defined in a simple way by determiningthe static balance between excitation forces imposed by the environment and reaction forcesby the thrusters (Figure 12). As explained above, the conditions for a DP operated ship arehighly dynamic and due to amongst others the effect of large inertia and damping forces andsecond order wave drift forces, this static approach suffers from severe limitations. Wichers etal. [11] conclude that static analysis (for a monohull) is inadequate in determining the DPcapability of a 3 axis weathervaning vessel.Figure 12 Example of a DP Capability plot showing the reduction in capability bythruster - hull and thruster – thruster interaction. In this case, 6 azimuthing thrusterswere applied, each of 200 kN bollard pull.Another complication in the use of DP capability predictions is the large variety ofcomputational models that are used, each with their own simplifications and neglects. In manycases for example, no interaction effects with other thrusters or with the hull are used. In evenmore cases the effects of e.g. bow tunnel thruster degradation in waves are neglected. Theseare mostly outside the scope of the DP Capability programs, whereas these effects can havean important bearing on the capability.To fully incorporate the above effects and other non-linear effects such as the second orderwave drift forces on the hull, one should apply a simulation model that solves the equations of

motion in the time domain (see e.g. Wichers et al. [11]).

Maximum thrust densityBecause of structural considerations, the size of a thruster is usually heavily constrained. Thisconstraint, together with the desire to keep the number of thrusters as low as possible, hasposed the issue of the maximum thrust density (thrust per unit propeller disk area). Although itis recognised that the thruster efficiency decreases with increasing thrust density in general,there is nevertheless a drive toward higher thrust densities for DP as a result of the overalldesign problem.The minimum dimensions of thrusters are however limited by cavitation induced thrustbreakdown, cavitation-induced vibrations, erosion and possibly mechanical constraintsimposed by the construction. Simple rules of thumb are mainly used in practice by engineersand propeller manufacturers to determine the minimal propeller size in an early design stage.Page 7 of 15These rules mostly use a propeller tip speed criterion or a power density criterion. Morerefined criteria, deduced from model experiments, such as proposed by Auf'm Keller [3] andHoltrop [2], show that parameters as blade area ratio and number of blades should also betaken into account. Current computational tools such as lifting surface or panel codes are ableto also show the effect of blade geometry on the maximum thrust density.Van Rijsbergen and Van Terwisga [10] review methods to determine the minimal propellerdiameter originating from full-scale experience, model-scale experiments and theoretical andcomputational considerations. Their paper focuses on thrust breakdown due to the presenceof a certain amount of sheet cavitation on the propeller blade. Other types of cavitation, suchas Propeller Hull Vortex (PHV) cavitation and erosive bubble cavitation can also impose alimit on the thrust density, but are not yet amenable to computational analysis.It was concluded from this study that the minimum propeller is determined by two criteria: A

non-dimensional thrust density criterion KT/n, and a non-dimensional tip speed criterion n.Dimensional equivalents of these criteria are less reliable because they show too large adependency on shaft immersion and efficiency. Furthermore, the thrust capability of apropulsor was pointed out to be dependent on wake field, propulsor type (open propeller,ducted propeller or waterjet) and propeller design. These parameters should preferably beincorporated in the criteria.

ManoeuvrabilityOne of the most important goals of the azimuthing thrusters is to have the ability to direct thethrust in all directions. This allocation offers an excellent freedom in manoeuvrability and is ofgreat use for the offshore industry, especially for the purpose of dynamic positioning. Also forother ships, it turned out to be a good solution.The large amount of tugs that are presently equipped with azimuthing propellers is a goodexample of ships that are combining on-the-spot manoeuvrability with the required vectoredthrust ability at low and high speeds. The nozzle on the steerable propeller combines thisgood manoeuvrability with a good bollard pull. A good example is the ship type AzimuthingStern Drive (ASD) tug. The number of azimuthing stern drive tugs that are delivered in therecent years is enormous. The Azimuthing Stern Drive tug is hereby developed as thestandard tug type, taking over from the tractor tugs and the conventional tugs. Even a newtype of tug is developed and equipped with Schottel thrusters. This is the Rotor® tug [5], ofwhich an impression is given in Figure 13.

Page 8 of 15The concept of three thrusters under the vessel without skegs yielded an enormous freedomin manoeuvrability (Figure 14), allowing even pure sideways movements of up to 6 knots(Figure 15).

Figure 15 Sidestepping at 6 knots

For ships equipped with thrusters and their manoeuvrability, it becomes an issue whether thehuman helmsman is able to control the ship. This manoeuvring problem is in a way related tothe control of the jet fighter F16. The F16 system in itself is course unstable and somanoeuvrable that one human cannot handle it. Placing a computer between the controls ofthe pilot and the actual steered flaps on the F16 formed a good solution. For the Rotor Tug,Schottel developed also such a device, called the Master-Pilot. Also for the ships equippedwith DP capabilities, such computer systems are required to allocate the thrusts of thepropellers in such a way that the environmental loads can be withstood, not only effective, butalso efficient. This means that the DP job has to be done with as little power use as possible.During all these manoeuvres, it is important that the thrusters will have as little mutualinteraction as possible. One thruster, blowing in the direction of a second thruster, reducesthe effectivity of the leeward thruster to a large extent, see Figure 16 from Nienhuis [8].Figure 16 Mutual interference between thrustersFor manoeuvring and course keeping purposes, one is interested in the characteristics of thepropellers in oblique flow at relative high speeds. The side force and the longitudinal force asfunction of larger forward speeds and oblique inflow angles are discussed in [6]. Withdecreasing skeg sizes and in some cases no skeg at all, the aspect of the course stabilitybecomes more critical. For the above mentioned Rotor-tug, it was found that for small anglesof attack, the side force generated by an operating thruster with nozzle is of the same order ofmagnitude as a typical skeg. This is illustrated in Figure 17 from [5]. These smaller angels ofattack (say up to 15°) are important for course keeping. For an important part, the coursestabilising effect is due to the nozzle. The following example of a double-ended ferryillustrates that for thrusters without nozzle, the situation is different.Page 9 of 15

Comparison of lateral force0.E+001.E+052.E+053.E+054.E+055.E+056.E+050 10 20 30 40 50Angle of attack (deg)Transverse force (N)SkegThrusterFigure 17 Comparing the lateral forces working on the ship due to thrusters or skegSufficient course keeping ability with rotatable propulsors is not trivial. Double-ended ferriesare sometimes equipped with thrusters without nozzles. In Figure 18, reported in [4] anillustration is given of two hull forms. While the upper hull form (initial design) suffered from anunacceptable course instability, the lower hull form appeared to show an acceptablebehaviour. In this case, stability obviously has to come from both the hull form and thepropulsors.Figure 18 Hull form design consequences for sufficient course keeping abilityPods: Hydrodynamic issues and design consequencesWith every new development, new uncertainties occur that need to be controlled.Hydrodynamic issues that arose during the development of pods were uncertainty about scaleeffects in the power-speed prediction based on model tests, and the loads and stresses thatoccur on the pod during its operational life. MARIN has recognised these problems in an earlystage and has invested in developing and validating an extrapolation method to scale thepower-speed relation from model to full scale. This is reflected in the pod models used forhydrodynamic testing and in a Joint Industry Project on Pods in Service. The objectives and adescription of the monitoring campaign are given later.A number of design questions arose with the advent of the pod:Will pods save money and will they show lower fuel consumption?

Will they lead to higher passenger and crew comfort, achieved by lower propeller inducedhull pressures and excitation forces through better cavitation properties?What about the manoeuvrability and the course keeping ability?What about the safety and reliability of the new systems?What are the hydrostructural loads under operational and extreme conditions?Will they cavitate due to steering angles during course keeping?Page 10 of 15Can they replace stern thrusters?Some general trends referring to these questions are given below. These trends were foundfrom some 50 commercial and research projects on podded ships that were carried out atMARIN during the last 5-6 years. Due to an early investigation of the hydrodynamic issues,We are proud to say that some 80% of the commercial researches towards pods are carriedout at MARIN. This includes the very prestigious projects towards the Eagle class of cruisevessels (Figure 19) and the Queen Mary II (Figure 20).Figure 19 Eagle class cruise vesselFigure 20 Queen Mary II

Propulsive EfficiencyBefore establishing the power speed relation, one should make sure that the pods are ideallypositioned in the flow, respecting possible design constraints. It has become clear that theoptimisation of the so called tilt and rudder angles and transverse and longitudinal position incombination with the best rotational direction can lead to power savings of about 3-5%.Although some trends between optimal position and hull form can be distinguished, theoptimum position strongly depends on the shape of the hull, the aft body fullness and the L/Bratio of the ship.Assuming that the pod configuration has been optimised, predicted power improvementsrelative to conventional propulsion configurations in the range of 7-12% are not unusual. Upto now, MARIN was able to validate her power predictions with the trial results of some 7ships. These results showed that the predicted power is close to the full-scale measuredpower, with a slight tendency to be somewhat conservative.

ComfortAddressing the comfort issue, it can be stated that the minimisation of propeller inducedpressure fluctuations is of utmost importance. Especially for cruise liners and ferries this is animportant issue. The increase in cruising and crossing speeds over the last decade and thegrowing importance of passenger comfort has led to a decreasing feasible design space forpropellers in a conventional shaft arrangement. Large propeller-hull clearances and highlyskewed, tip unloaded propellers were the result. The deterioration of propulsive efficiency wasthereby accepted.With pods, excellent inflow characteristics and small cavitation extents on the propeller bladeshave been observed. Even the complete absence of cavitation has been observed. Aconsequent reduction of propeller induced hull pressure fluctuations and excitation forces wasmeasured, even under steering angles of about plus or minus 7 degrees. It is thereforeexpected that in the near future, more sophisticated wake adapted propellers on pods cangain a few percent in efficiency without sacrificing the excellent vibration levels of the ship.Page 11 of 15Manoeuvrability and course keepingThe introduction of podded propulsors with electric motors in the hub introduced the vectoredthrust in a new market segment: the very large powers. This allowed for example cruise shipsto be equipped with pods. The need for this was also obvious. Besides the already presenttrend to go for an All Electric Ship, there was a need for better manoeuvrability with cruiseships. Cruise ships are becoming larger and larger while ports stay at similar sizes andmarine traffic becomes denser. A further improvement in controllability of cruise ships shouldtherefore be pursued. The application of the podded propulsors stimulated this enormously.Besides the almost standard application of pods for cruise vessels, nowadays pods are alsoapplied in other ships. The first application of the SSP was on a chemical tanker, but there are

other applications possible such as heavy load ships.Equipping ships with podded instead of conventional propulsion can improve themanoeuvring characteristics of a ship considerably. However, the use of the word can shouldbe emphasised here: worsening is also possible. Several manoeuvring aspects are dealt within the following.Low speed manoeuvringFor manoeuvrability at low speeds, the pod developments are the necessary leap forward.Due to the use of new materials and client requirements towards all balcony ships, thesuperstructure of modern cruise vessels and ferries is becoming very high. The resulting windloads are enormous. Therefore, more powerful bow and stern thrusters are required.Especially stern thrusters have insufficient power. Now, the all-turnable podded propulsorsare overcoming this in the aft ship. The most recently observed trend is that the amount ofbow thruster power is the limiting factor in reaching the vessels’ low speed manoeuvringtargets.Course keeping abilityA design consequence of the application of pods is that freedom is obtained to design a veryflat aft ship. This is often favourable from a resistance point of view, and creates a veryhomogeneous flow towards the pod, which is good to avoid cavitation and vibrations.Especially when three or four pods are used, this freedom is also needed from a design andconstruction point of view. References are the Queen Mary II and the Eagle class of cruisevessels. The open aft ship does not have much lateral resistance and hence the coursekeeping ability will be small. The podded propulsors are furthermore in general withoutnozzle. It was already stipulated in this paper that non-ducted propellers have inherentlymuch lower course keeping stability than ducted propellers. Together, this makes that poddedships are in general more course unstable than conventional ships (see Figure 21). Therecent trends of applying pods to full ships (such as tankers and LNG carriers) can become a

real challenge from the course keeping point of view.The possible operational consequences of an insufficient course stability is serious:Not fulfilling the IMO resolution A751(18) towards course keeping ability,Excessive steering actions imposed by the autopilot, causing wear and tear of thebearings and steering engine, increased resistance, loss of propulsive efficiency andpossible cavitation.An increased risk of broaching due to the loss of directional and transverse stability instern quartering waves.Increased risk of collisions due to the inability of the ship to counteract turns adequately.Increased required power because of the additional hull resistance resulting from the nonzerodrift angles.Excessive steering in calm water or waves should be avoided at all times from a cavitationpoint of view. The consequences are a constantly varying loading of the propeller, resulting inmany peak loadings. There is an increased risk for adverse effects by cavitation on thepropeller (when the propeller is in oblique flow, the cavitation inception speed is lower). Shipsat higher speed may additionally suffer from cavitation on the struts of the pods.Page 12 of 15Figure 21 Aft ship equipped with conventional propulsion arrangement and a podarrangementBased on the above, it is the firm belief of MARIN that the course keeping ability anddirectional stability should be investigated thoroughly before building the ship. More importantthan ever seems here that the behaviour and performance of the vessel is the result of amarriage of the hull form with the propulsor.Heel anglesA third important aspect of steering with pods is the occurrence of large heel angles. Thepods are very powerful steering tools. The side force that can be generated is so large, thatthe steered vessel can suffer from very large heel angles. At MARIN, heel angles of up to 25°have been measured with ship models due to regular steering. Knowing that the panic limit for

passengers is at some 7°, it is obvious that this is undesired. The design consequences arethat the hull form will have to be modified to assure that the heel angles will stay withinacceptable values. It is important to check this with model tests before the ship is build.Practical operation of pods and thrustersExperience from past projects learns that crew training is becoming very important whenships are equipped with podded propulsors, which is true for steerable thruster units as well.Operating pods is a different way of sailing. The manoeuvring capabilities of vessels equippedwith pods are potentially high, but full use of these capabilities requires crew training,preferably on a manoeuvring simulator in order to cover also propulsion emergencies.Examples of such projects are e.g. the cruise vessels built at MeyerWerft in Papenburg, whohad to sail through the Ems to reach open sea. Very accurate steering is necessary and theslightest mistake will cause a risk on the loss of the ship. Other examples are the training oftugmasters, the handling of double-ended ferries such as for the PSD ferries and the TESOferries. Figure 22 gives an illustration of a training for tugmasters on the handling of a tugwhile escorting large vessels.Page 13 of 15Figure 22 Master training in tug handling at MARIN's simulator centreSafety and structural loadsTo get an appreciation of the structural loads that are met during operations with podpropelled ships, a large European project was initiated by MARIN. The reliability and safety ofpods under operational conditions had to be monitored on full scale. This Joint IndustryProject was designated "Pods In Service" and has the following objectives:1. Assess the reliability and safety of pods under operational conditions2. Evaluate the operational performance and benefits for the ship owners3. Develop design, construction and classification methods.In this Joint Industries Project, 25 parties are collaborating world-wide. Besides MARIN, theseare the cruise line operators, navies, the pod manufacturers ABB Azipod, KaMeWa, Siemens-Schottel, shipyards and classification societies and VTT Finland (see Figure 23).

Figure 23 Participants in the Pods in Service Project (Kvaerner Masa-Yards meanwhilealso joined the project)Page 14 of 15During this project, four vessels will be monitored on full-scale during a period between 6 to12 months of their operational life. The measured ships are the Summit (Millenium-class), theTT-line’s Nils Holgerson, RCCL’s Radiance of the Seas and the Finnish Botnica. During themonitoring campaign, there is a focus on structural excitation and response. To this end, thefollowing signals are measured continuously: strains in shaft, gear and pod housing, hullpressure fluctuations, hull accelerations and vibrations and propeller blade strains.Simultaneously, the conditions are monitored continuously by registration of azimuth shafttorque and angle, input power and propeller rpm, ship draft, motions, speed and track andwind, waves and current. For one of the vessels, the underwater-radiated noise will bemeasured.From the measured quantities, important feed back is obtained. This is not only hydrodynamicfeed back with respect to the efficiencies and vibrations. Much structural feed back ispresented and classification societies are using this to upgrade or determine the rules for theclassification of podded vessels. A special work group consisting of all classification societiesis developing and verifying computational and design methods for pod and hull strength.The first ship, the Botnica, owned by the Finnish Maritime Administration, experienced aextreme severe storm situation (15 m significant waves) during the monitoring campaign. Theresults are being analysed during the first months of 2001. Then it also will become clearerwhat happened during that extreme event. But of course also the other information will be ofimportance to increase the knowledge related to the behaviour of the ship and its PODsystem during a longer period of time.Final remarks

This paper gives a review of current issues in the design and application of steerablethrusters and podded propulsors. One can conclude from this review that the concept ofsteerable thrusters and its design space is relatively well known territory, yet leaving anumber of pitfalls for the designer. The concept of the podded propulsor is relatively new, andrelatively little empirical knowledge has yet been accumulated. Hence, designers andoperators have to rely on model tests, supplemented with CFD calculations that requirerelatively little empiricism. For pods, one can state that the necessary empirical knowledge isgenerated more quickly than was the case with the steerable thruster some 50 years ago.This is achieved through sophisticated model tests supplemented with CFD computations andcomprehensive full-scale measurement campaigns.On podded propulsors, different applications and more sophisticated designs can beexpected. An extension of the pod applications can be expected toward full block vessels andcontainer ships. Research programs are already initiated for this. A higher degree ofsophistication of the design seems especially possible in an optimisation of the combined hullform – pod system design (e.g. adaptation of hull lines) and in further reductions of the poddiameter and the optimisation of the stay (strut arm of the pod). In addition, the propelleroptimisation will lead to a further improvement in efficiency and in cavitation and vibrationreduction. It is expected that the range of applications will also grow with increasing insight incourse keeping properties in calm water and waves.Although this paper has dealt especially with hydrodynamic issues, we cannot evade theever-important issue of economics. Even hydrodynamicists can see that a reduction of theprice of the pods will definitely be beneficial toward extension of its use.

References:[1] Bussemaker, O. and Corlett, E.C.B.; Tractor tug family fitted with rudder propeller.Proceedings of 2nd International Tug Convention. London, 1972

[2] Holtrop, J.; A statistical re-analysis of resistance and propulsion data, InternationalShipbuilding Progress, Vol 31 - No. 363 (1984)[3] Keller, auf'm J., "Enige aspecten bij het ontwerpen van scheepsschroeven (in dutch)",Schip en Werf, Dec. 1966.Page 15 of 15[4] Kristensen, H.O.H.; The manoeuvrability of double ended ferries. Design considerations,construction and service experience. Int. Conference on Ship Motions andManoeuvrability. RINA, February 1998, London.[5] Kooren, T., Aalbers A. and Quadvlieg, F.; Rotor Tugnology; ITS2000, The 16thInternational Tug and Salvage Convention, Jersey, Channel Islands, May 2000.[6] Oosterveld, M.W.C. and van Oortmerssen, G.; Thruster systems for improving themanoeuvrability and position keeping capability of floating objects. OTC paper 1625, May1972[7] Marine Propulsion International; '2000 Years of Propulsion History', Sept. 2000[8] Nienhuis, U.; 'Analysis of Thruster Effectivity', PhD Thesis, Oct. 1992[9] Valkhof, Henk H.; "Podded propulsors, it has all just started", Marine Propulsion 2001Conf., The Motorship, March 22-23, , London[10] Van Rijsbergen, M.X. and Van Terwisga, T.J.C.; 'On the maximum thrust density ofpropellers', NCT'50 International Conference on Propeller Cavitation, Newcastle, 3-5 April2000[11] Wichers, J, Bultema, S. and Matten, R.; 'Hydrodynamic research on and optimizingdynamic positioning system of a deep water drilling vessel', Offshore TechnologyConference OTC, Vol 4, no. 8854, May 1998

Diesel Electric Propulsion Systemfor P&O Cruise Liner “Aurora”Electric Propulsion System for P&O "Aurora" -a Propulsion System with a high Level of RedundancySTN ATLAS Marine Electronics´ system responsibility for:Diesel-electric propulsion system, electrical power generation and power distribution system;Integrated navigation systemMain data:Length overall approx. 270 mBreadth 32.20 mDesign draught approx. 7.9 mGross measurement est. 76000 tPassenger cabins 934Service Speed 24 knotsPower plant 56.0 MWPropulsion drives 2 x 20 MWThruster drives 4 x 1.5 MWChiller drives 3 x 1.35 MWPropulsion and power components supplied by STN ATLAS Marine Electronics2 Diesel-electric main propulsion drives with:- 2 Double-winding synchronous motors 20 MW, 140 rpm- 2 Synchro-Converters 12/12-pulse, each with two pure water coolers and tworedundant excitation and control systems in master/slave-configuration- 8 Cast resin propulsion transformers 8.4 MVA6 Diesel generators4 x 17.5 MVA, 6.6 kV, 60 Hz, 514 rpm2 x 1.25 MVA, 690 V, 60 Hz, 1800 rpmSwitchboards 6.6 kV with 11 and 12 panels for medium voltage distribution4 Harmonic filter banks reducing the total harmonic distortion of the mainsvoltage to max. 5 %8 Ring main substation units 4 x 1.5 MVA and 4 x 1.1 MVA incl. medium and

low voltage switchboards and cast resin transformers for low voltage supply5 Cast resin transformers for engine room and emergency switchboard supply7 AC motors 4 x 1.5 MW, 6.6 kV, 1200 rpm for thrusters and 3 x 1.35 MW,6.6 kV, 3600 rpm for air-condition compressors3GeneralFor the new flagship of P&O, built atMeyer Shipyard, STN ATLAS MarineElectronics was selected as maincontractor for the complete electricpower generation, main power distributionand the electric propulsion system.Based on the existing experienceacquired from the delivery of more than140 electrical propulsion drives, STNATLAS Marine Electronics has takenover also the system responsibility forthe power generation in combinationwith the propulsion system. Mathematicalmodels as well as own softwaresimulation programs allow precisepredictions concerning the behaviourand the quality of the power network.The "Aurora" is designed as a passengervessel with a very high standard for allaccommodation areas. This high standardis consequently achieved also for themachinery, especially for the propulsionsystem. Intensive safety requirementsof P&O lead to a design with a high levelof redundancy. This guarantees a highavailability under all sailing conditions.System DescriptionThe "Aurora" is equipped with four dieselgenerators providing all electric energyfor the propulsion, all machinery and

hotel consumers. The medium voltageswitchboard is split into two sections,which feed all essential consumers. Aring bus system supplies all substationsvia two short circuit current limitationreactors. Two fixed pitch propellers, eachdriven by a synchroconverter system,provide the required propulsion power.Chiller and thruster motors with directon line start are connected to the mediumvoltage switchboard. Four filter circuitsensure a low harmonic distortion factor.Synchro-Converter Drive DesignEach propulsion motor is designed assix phase salient pole synchronousmachine with air cooling. Both of thewinding systems are fed by separateconverters. This results for the motor ina 12-pulse torque ripple and a smoothoperation. Each of this two converters,which are feeding one motor, is suppliedby 2 parallel transformers with a 30°shift for a 12-pulse line reaction. Thisdesign provides an unchanged harmonicsignature with a 12 pulse line reactionalso in case of operation with only oneof the converters. The two convertersare consequently realised with theirown control, excitation and coolingcircuits.The converter power section isdesigned as direct water cooled systemto reduce the dimensions.Switchboard DesignThe 6,6 kV voltage switchboard isdesigned in two separated parts connectedby SF 6 tie breakers. For themain consumers like generators SF 6breakers are used, for the smalleroutgoing panels with a higher switching

sequence contactors are used. A verycompact arrangement was achievedwith double stock compartments.The substations operate with separateSF 6 insulated switchboards feedingthree winding transformers.Propulsion Redundancy ConceptThe "Aurora" is equipped with an exceptionalhigh level of redundancy. Besidethe two propulsion motors there are intotal four independent converters. Oneconverter is connected to one windingsystem of the motor. It is controlled andmonitored by it´s own control system.The same applies for the water coolingcircuit, where the power part of eachconverter is separately cooled. Eachconverter is even equipped with anexcitation system, which will be automaticallyswitched over in case of a failure.Each converter uses separate sensors,so the monitoring is independent fromeach other. As special design aspecteach converter is designed for 140% ofit´s nominal power. This built-in sparepower will be activated automatically, ifonly one converter is in operation. Atotal converter power of 28MW is4therefore installed per shaftline. Besidethe lower temperature level in normaloperation mode this power provides incase of operation with one converter avery low reduction of speed. Testsshow a reduction of only one knot. Thisallows the vessel to maintain it´s timeschedule.Consequently all power and auxiliaryfeedings for one shaftline are supplied

from different distribution switchboards.Also the filter circuits follow the principleof redundancy. In case of a problem withone of the filters, an unlimited operationis possible without any restrictions.The benefit of this concept is anavailability of a partial higher propulsionpower in case of a converter failure.Telediagnostic SystemThe propulsion system is equipped withintensive diagnostic and failure identificationsystems. A transient recorder forstoring data in a memory is provided asstandard since many years for all STNATLAS Marine Electronics synchroconverters.Now a telediagnostic systemis realised, which allows a serviceengineer to perform an instantaneousevaluation of an event from his homeoffice via satcom.The service engineer can perform thesame analysis as being locally at theconverter board. In many cases thefailure can be corrected by the shipsstaff engineers by data transmissionand phone/ fax support via Satcom.Filter circuits/ Quality of the mainsFour filter circuits in total are provided,each with an incorporated filter for the5./7.th. and the 11./13.th harmonic frequency.This configuration allows STNATLAS Marine Electronics to guaranteea THD factor of <5% under all operatingconditions. This prediction was verifiedby extensive measurements. The resultlies well under any classification requirementand ships specification. The filtercircuits combine the reduction of theTHD factor to specified values with an

improvement of the power factor. Thisreduces the reactive power andincreases the efficiency. The filters areswitched depending on load conditions inorder to avoid a capacitive power factoron the mains.Monitoring conceptDistributed I/O´s for motor and transformerscollect all data from analogue anddigital sensors. The collected data aresent via a bus system to the converter.This concept reduces the cabling effortto only a bus system and the DC supplycable between the main components.Beside this are all data available formonitoring purposes by the converterCPU. A bus transfer of all data to theautomation system is provided.System AdvantagesThis advanced propulsion concept ofthe P&O AURORA guarantees a veryhigh level of redundancy. This ensures,that even with a fault in one of theconverters, the vessels time schedulewill be fulfilled. A high level of availabilityis achieved by installing four independentconverters. In case of converterproblems a telediagnostic systemallows world wide an immediatesupport by specialists of the supplier.5The single line shows the arrangement of the main supply with the propulsion system. The consequentseparation between the propulsion ensures the high availability.

Second tanker to have SSP podded propulsionSwedish shipowners order chemical products tanker with Siemens SCHOTTEL Propulsor

The Siemens SCHOTTEL Propulsor Consortium (SSP) has received an order from Swedish shipowners AB Donsötank of Donsö to supply a Siemens SCHOTTEL Propulsor 'podded' propulsion system for another chemical products tanker. The order is worth around 5.5 million euros and the new vessel is scheduled for delivery in early 2003. The new vessel's sister ship the "Prospero", which also has SSP propulsion, entered service in November 2000.

Owners "Rederi AB Donsötank" of Donsö near Gothenburg in Sweden have recently placed an order with Chinese builders Shanghai Edwards Shipyard for a new chemical products tanker of approximately 12,000GRT and 145 metres overall length. The vessel will be propelled by a Siemens SCHOTTEL SSP 7 Propulsor podded propulsion system with an output of 5.1MW.

At the heart of the SSP propulsion system is a permanent-field electric motor that needs no costly and complex cooling system. Compact, hydrodynamically-efficient construction together with a twin-propeller concept produces a highly effective propulsion system which, compared to conventional systems, offers better efficiency and needs substantially less space. These features help save energy and cut costs and so bring about a very useful improvement in the amount of revenue-earning cargo capacity available. The fact that the pod swivels also gives a useful boost to the vessel's manœuvrability. Thanks to the system's modular construction installing it on board the vessel is an extremely simple procedure. At the same time the system offers other key features such as excellent reliability and low

maintenance costs.

One important reason for the new order from AB Donsötank has been the highly successful performance of the M/T "Prospero" - the new vessel's sister ship which entered service at the end of last year - over the past few months.For the press only: Data of illustrations on request: kscholz@SCHOTTEL.de

Getting twin propeller efficiency from a podAn innovative podded drive system is designed to give significant improvements in economy and maneuverability for a broad range of vessels, including cruise ships, chemical tankers, icegoing vessels, offshore units and naval vessels.

Available in power outputs from 5 MW to 30 MW per unit, the SSP (Siemens Schottel Propulsor) promises energy savings of better than 10%, thanks to a combination of the benefits of the Schottel Twin Propeller and a new permanently-excited synchronous motor, developed by Siemens, that allows the maximum efficiency in the transmission of electrical energy within a minimum installation space.The unit builds on Schottel's experience in the development of steerable right-angle drives (its Rudderpropellers). The benefits of these units in applications up to 6 MW are well known and there are more than 23,000 Schottel units of this type in service.To improve the efficiency of the Rudderpropeller, Schottel developed the Schottel Twin Propeller (STP), where the propeller load is distributed 50/50 to two propellers, one forward and one aft of a lower housing.

This housing features two airplane type fins that recover rotational energy from the forward propeller.

The STP achieves efficiencies up to 20% higher than standard Rudderpropellers. However, mechanical right angle drives, including the STP, are limited to around 7 MW per shaft.

The SSP (Siemens Schottel Propulsor) was developed to make the advantages of the STP available for higher powers. This was only possible by incorporating powerful electric motors in the lower housing of the azimuthing drive.

Conventional high power, low speed synchronous motors are so large and heavy that they must be housed within underwater housings with a diameter of as much as 60% that of the propeller diameter, with a dramatic negative influence on the unit's overall efficiency. Siemens has for some time been developing permanently-excited synchronous motors with a longitudinal electrical flow design. A 1,000 kW propulsion test unit has been in operation on a naval vessel for several years. This type of unit, available in a power range of about 5 MW-30 MW at low speeds, allows a significant reduction in the diameter of the motor and, in turn, in the diameter of the housing of a podded drive. This allows optimum hub/propeller diameter ratios to be achieved.

The SSP is a marriage of the Schottel Twin Propeller and this new type of electric motor.A standard Rudderpropeller: replaces the steering and propulsion systems of vessels; gives optimum maneuverability without additional stern thrusters; lowers noise and vibration thanks to special supports; occupies less space and requires a smaller engine room than a conventional system; can be installed later in the construction stage than the conventional system, saving on construction time and costs.

Further advantages claimed for the SSP are: no risk of vibration excitation by gear sets and cooling fans; simple surface-cooled motor; mounting of the lower housing is possible without drydocking.

COMPARISON

A propulsion analysis was performed comparing the SSP with the propulsion system of the 70,000 gt cruise vessel Century built by Meyer Werft in 1995. Tank tests were performed by SVA, Potsdam, Germany, taking into consideration the original tank tests of the vessel by SSPA, Gothenburg, Sweden.

The 248.5 m LOA Century displaces 35,200 tons on a design draft of 7.5 m. Design speed is 22 knots. Its diesel mechanical propulsion system includes two shaft lines, each with a 5.8 m diameter propeller absorbing 14 MW at 120 rpm.

SVA Potsdam carried out tank test and cavitation tunnel analyses of both:

a standard podded drive system using two units each with a 5.2 m propeller diameter and propeller speeds of 160 rpm (hub diameter 60% of propeller diameter);

an SSP system with two units each with propeller diameters of 5.4 m and propeller speeds of 150 rpm.

The study indicated that the SSP system would reduce propulsion plant power consumption by 10%, translating into either an 0.5 knot speed increase or a 10% fuel savings. Though the study considered the advantages of the standard podded drive's lower resistance (due to the absence of stern thrusters and shaft brackets), its speed was found to be no better than with the conventional drive.The SSP gives considerable space savings in applications such as cruise ships. In addition, a weight comparison of the 14 MW SSP installation and a conventional diesel electric installation indicates a total weight of 510 tons for the SSP units, and associated cabling and structures. This compares with a total weight of 760 tons for the conventional system.

ELECTRICAL SYSTEM

The hydrodynamic requirement that the lower housing diameter not exceed 30-40% that of the propeller, ruled out conventional synchronous motors. The concept selected was, instead, the permanently-excited synchronous motor. Siemens has many years' experience in designing this type of motor for naval submarines.

Magnetic flux is generated by high performance magnetic elements. Generally arranged on the motor's rotor, these substitute for the conventional excitation winding and such auxiliaries as slip-rings, rectifier, cooling air ducts and cooling fans. Besides significant volume and weight savings, this gives a considerable gain in efficiency by avoiding core losses and heat losses due to the excitation current.

The flux distribution selected for the SSP application was longitudinal. This avoided the need for a disc-type rotor, giving flexibility in selecting the relationship between axial and radial dimensions of the motor's active components. The resulting design is very similar to that of a conventional synchronous motor and has similar electrical characteristics, avoiding problems with electrical supply via conventional converters.To optimize the drive configuration, a self-commutated converter is required.

Depending on load requirements, the SSP will be offered with a cyclo or PWM converter.

Depending on propulsion system demand, the motor will be designed with either one winding system or two independent winding systems (in the latter case, emergency operation with half the motor is possible).

A motor designed for a 14 MW drive will have the following characteristics:

 Output power  14 MW Supply voltage  3.3 kV Rated current  2.9 kA Efficiency  0.98 Power factor cos phi  0.85 Max speed  150 rpm Number of poles  18

The motor will be supplied by a directly water-cooled, fuseless and short circuit proofed cycloconverter. To reduce total harmonic distortion in the ship's

network, each individual drive system is designed with 12 pulse configuration. The converter will be selected to guarantee a near sinusoidal shape, resulting in a low level of structure-borne noise.SAVINGSWhile power savings of 10% have been indicated by studies of a 70,000 gt cruise ship, Schottel and Siemens believe that for other types of ships, power savings may be even higher. ML

PODS FOR ROPAXSiemens-Schottel propulsor systems will be powering two new RoPax vessels ordered by the TT-Line shipping company and to be built by SSW Fähr- und Spezialschiffbau GmbH of Bremerhaven. The ships will be operating in the ferry service between Travemünde and Trelleborg. The Siemens Marine Engineering Subdivision in Hamburg will be supplying and installing all the electrical machinery and systems as a turn-key contract worth around $40 million.

Each of the 190 m ships will be propelled by two Siemens-Schottel SSP 10 Propulsors. The power output of each SSP 10 will be 11 MW. The turn-key project also includes all the automation equipment employing proven “SIMOS IMAC 55” equipment and the communications systems.

The SSP utilizes a compact permanent magnet electric motor that allows the pod to have a lower profile, permitting more efficient water flow into the propellers.

The SSP is particularly suitable for passenger vessels such as TT-Line’s two new RoPax ferries because its twin screws produce much lower noise and vibration levels so passengers can enjoy higher standards of comfort.

Since the Propulsor comes in sizes from 5,000 kW to 30,000 kW it is suitable for the whole range of outputs needed by seagoing vessels. Further development of electric propulsion systems will undoubtedly be expanding the range of application of pod-type propulsion systems in the future .In the permanent-magnet electric motor and the twin propeller concept, the Siemens-Schottel Propulsor is employing two basic technologies that, since they are both innovative and trend-setting, provide a solid foundation for successful penetration of the market. ML

POD PROPULSION

by F. Mewis

Interest in pod propulsion has been stimulated by the successful sea trials of several ice breakers and cruise liners during the last few years. At present electrically driven azimuth drives of up to 20 MW per unit are available.

From the hydrodynamic standpoint the primary advantages of this type of drive unit are

lower power requirement lower level of propeller-induced pressure pulses improvement of the manoeuvrability

Powering Performance

Much research and development work has been done by the various pod producers for the purpose of improving the pod drive performance.

By optimizing the pod propellers as well as the configuration of the pod housings, the efficiency of the units can be improved dramatically. The use of modern CFD-methods

for optimization has led to a better understanding of the flow around the pod housings, also in connection with the working propellers. The figures below show the pressure distribution on a pod housing. The distribution is different on the starboard and port side due to the influence of the working propeller on the flow.

Starboard Side Port Side

(Abb. part1.ps und part2.ps)

Model tests in the towing tanks with pod units include both propulsion tests and open water tests.

HSVA has developed new model pod testing equipment which enables accurate measurements of the propeller torque and thrust at both ends of the drive. The pod unit thrust is also measured. The photograph shows a Siemens-Schottel Propulser (SSP) with two 4-bladed propellers installed on HSVA's pod open water device.

SSP model installed on HSVA pod open water device

Pod units are also well suited as booster drives for increasing the speed of existing vessels. HSVA has carried out extensive model tests for a commercial

conversion project: a cruise vessel with an additional azimuthing pod on the centerline. Within the scope of the work, the powering performance of four different pod systems was investigated.

"Costa Classica" conversion with additional SSP

Diesel – Electric Power PlantDiesel – electric power plants and propulsion solutions have clear merits which may vary from one ship type to another. The safety aspects of diesel – electric propulsion are commonly regarded as being related to redundancy in different ways. The number of power-generating units is large enough to ensure propulsion capability and steerage way irrespective of any component failure. In addition the units can be located in different compartments to safeguard against loss of power in case one compartment has been destroyed by fire or flooding.

A summary of advantages:

Flexibility, the installed power generating capacity can be used for various ship functions and different situations;

Propeller torque capability, full torque at any propeller speeds;

Permits running diesel engines at a stable load with smoother transients;

Permits running diesel engines at a constant speed;

Permits running diesels engines at a more efficient load at optimum specific fuel consumption, hereby reducing emissions and impact on the environment;

Uniform machinery; simple spare parts logistics, maintenance, crew training, etc;

Flexibility in location of main engines allows to optimise cargo space volume and arrangement;

Redundancy, both in the sense of safety and freedom in maintenance routines.

 

Propulsion for Performance

Commercial marine propulsion has been seeing a good deal of innovation over the past few years, most of it in the cruise ship sector where podded propulsion drive and gas turbines have been brought into play. Very large diesel engines, the largest ever built for commercial application, are being developed for mega-container vessels capable of transporting nearly 10,000 TEUs.     In the workboat sector, there has been a near full-scale shift to azimuthing drive for ship-assist tugs, and tug owners have been choosing from a wide variety of drive combinations. Fast ferry operators have been demanding cutting-edge propulsion technology, and new hull forms are being developed to use advanced medium-speed diesels as well as compact gas turbines.     Over-shadowing all these developments are environmental regulations, many emanating from California, that will influence marine propulsion design over the next decade.     While propulsion equipment manufactures have come up with a number of new concepts to reduce pollution and enhance ship operation, not all installations have gone completely as planned.     In the cruise sector, Carnival Corporation has settled out of court with

Finland's ABB Industry Oy for problems the cruise line has experienced with ABB-developed Azipod podded propulsion drives fitted on two Carnival cruise liners, the Elation and Paradise. Repairs to the Paradise, said to have involved pod bearings and seals, were carried out by Virginia's Newport News Shipbuilding while the Elation was handled at California's San Francisco Drydock. Carnival demanded compensation from ABB because of canceled cruises and lost income due to the two drydockings. Newport News has also handled Celebrity Cruises, new cruise ship Millennium, which experienced problems with its Mermaid podded drives, resulting in a forced speed reduction of about four knots. Most recently, the new European car ferry Nils Holgersson experienced bearing failure in a Siemens-Schottel podded propulsion unit during sea trails in the Baltic Sea, which required drydocking of the vessel for complete removal of the pod. Nevertheless, the benefits of podded propulsion are many and manufacturers feel that "once the bugs are worked out" they will become a standard propulsion device. Podded drive, in fact, has already been specified for the Coast Guard's new icebreaker being planned as a replacement for the 1944-built USCGC Mackinaw on the Great Lakes.     Although podded propulsion has yet to be fitted to a large containership, it is already finding application in the tanker field. Swedish operator Rederi Donsotank has taken delivery of a 16,800 dwt product tanker, the Prospero, from Chinese shipbuilders that makes use of a single Siemens Schottel propulsion pod. The unit has two propellers, one at each end, and is powered by four 9L20 Wartsila diesel generating plants with a combined output of 6,480 kW. This gives the 478-foot vessel a loaded speed of 14.5 knots and the capacity to accommodate about 1,000 cubic meters more cargo than it would if conventional diesel-screw propulsion was used. Podded propulsion has also been chosen for a series of larger 106,000 dwt tankers being built for Finland's Fortum Oil & Gas by Japan's Sumitomo Heavy Industries. Fortum is a pioneer Azipod user, having had pods fitted to two of its smaller ships, the Uikku and Lunni, almost a decade ago. The new Sumitomo-built vessels will be "Double Acting Tankers" in that they will move forward in open water but will use the Azipods to move stern-first in ice conditions.     Podded propulsion is also making its way into the ferry sector and ABB has developed a compact version of the Azipod that is being installed on a double-ended ferry being completed in Europe for service on the Baltic Sea. The smaller version of podded propulsion unit uses a permanent magnet motor with direct cooling by the surrounding sea water. This in turn allows a smaller diameter propeller to be used which, according to ABB, gives more dynamic efficiency.     A zipod has been designed to have a reversible propeller for working

astern, which means that the units do not have to be completely azimuthing. This, in turn removes the need for slip ring connections, allowing cables to be used to transfer the required electrical power between generating plant and drive.

Prop Shaft Bearing Failures AgainSideline New Cruise Ships

Celebrity Cruises revealed two of its new pod-drive cruise ships are being sidelined due to multiple propeller shaft bearing failures, one for the second time in under a year. Even with insurance, parent Royal Caribbean expects the cost and lost profits to total $11.5 million.

The Millennium-class Infinity and Summit are each equipped with two Rolls-Royce Mermaid pod propulsion systems. Two large bearings, approximately 2-1/2 feet in diameter, carry the propeller shaft in each pod. The bearings are apparently failing under thrust loading.

The early-stage failures are showing up as spalling, evidenced by metal in the pressurized lubrication system. Oil filters prevent further contamination damage to the bearings, but to reduce thrust loading and prolong bearing life, the drives are being operated at significantly restricted power. Under full power, the motors are capable of using 19.8 megawatts to swing the 18-foot fixed-pitch propellers.

Infinity and Summit are scheduled to drydock for two weeks each at the Grand Bahama Shipyard in Freeport. Repairs had to wait until both drydock space and new bearings were available. Summit will have all four of its drive bearings replaced beginning March 29, 2002, with Infinity to arrive on April 13, 2002 for the same work.

Celebrity President Jack Williams said, "The ships are operating safely, but unless these repairs are made, we cannot guarantee the integrity of our advertised itineraries."

Last summer, Infinity experienced the first prop shaft bearing failure, shortly after being put into service. It was emergency drydocked in Vancouver for two weeks to replace one bearing in the port side pod.

• click here to read the Infinity article, which also contains information about pod drives and teething problems with the ships

Rolls-Royce analyzed the Infinity's 2001 bearing failure and a few months ago released a reengineered bearing. Because Infinity's starboard-side pod is now showing the same failure, all of the prop shaft bearings in both Infinity and Summit are being replaced with the new bearing - eight in all. While the new Summit has not shown signs of failure, they are being replaced as a precaution. Similarly, Constellation, the final Millennium-class ship still under construction, is being retrofitted before it hits the water.

Millennium Class Specifications

The Millennium class represents the latest in advanced cruise ship technology and drive systems. Millennium class are 91,000 tons, 965 feet long and can cruise at up to 24 knots. They carry 1,950 passengers and 1,000 crew. The Infinity has the world's first conservatory at sea (including six Magnolia trees), a cyber cafe, 25,000 square foot spa, Internet access in every stateroom and numerous other features. They are built to Panamax standards and so are certified to pass through the Panama and Suez canals. All Millennium class will be built by the Chantiers de L'Atlantique shipyard at St. Nazaire, France. Each ship costs over USD $350 million.

New Drive and Propulsion Technologies Introduced

The Millennium class employs two new drive system technologies.

First, they are the world's first gas turbine powered cruise ships. Power is generated by two GE LM2500+ aeroderivative gas turbine engines from GE Marine Engines division, GE Aircraft Engines. The LM2500+ is a combined gas turbine and steam turbine integrated electric drive systems (COGES). Each 22-foot, 11,000-pound engine produces 40,500 horsepower at 3600 RPM. The exhaust gas temperature is 965 degrees Fahrenheit. The gas turbines are the cleanest burning powerplants for any cruise ship in operation today.

Second, the ship is powered through the water by two Kamewa (Rolls-Royce AB) / Alstom pod propulsion systems called Mermaid™. Each Mermaid pod propulsion system consists of a 19.5 MW electric motor turning an 18-foot fixed pitch propeller. The electric motor is contained within the pod, completely submersed, and has infinitely variable speed control. Most importantly, the two pods can be rotated through 360 degrees, providing thrust in any direction. The propellers normally point forward, but their infinite speed adjustment and infinite directional adjustment allow the ship to be steered in any direction at any speed up to 24 knots. The propulsion pods not only allow the rudder to be eliminated, but putting the power unit in the pod frees up substantial space onboard.

Other advantages to the pod propulsion system are that the ship can easily dock anywhere without tugboat assistance, and that by pointing the propellers into the oncoming water, pressure pulses are reduced or eliminated. A propeller's pressure pulses create intrusive vibrations within the ship; reduction of propulsion system noise and vibration has long been a key design criteria for cruise ships.

Production

Royal Caribbean ordered four Millennium class ships to be delivered by the end of 2002. The first, Millennium, was delivered in mid-2000. The Infinity was delivered on February 26, 2001, over a month late, and was not inaugurated until April 29, 2001.

Chantiers de L'Atlantique is France's biggest shipyard and has built most of the world's largest and most advanced oil tankers, over 120 advanced technology warships and produces 40% of the world's cruise ships. The shipyard employs over 8,500 people.

Teething Problems Plague Both Ships

Even though the shipyard's advanced technology capabilities are well established, this has not prevented the Millennium class from experiencing a series of expensive and debilitating teething problems.

The first ship, the Millennium, had to be taken out of service only a few months after inauguration due to unacceptably high vibration levels amidships, traced to the gas turbines. The ship was put in drydock at Newport News where it was fitted with a ducktail and additional buffer section in the stern. Several cruises had to be cancelled in that case.

The Infinity was in its in final stages of completion just as the Millennium's unwanted vibration problems surfaced. Infinity's launch was delayed by over a month as a solution was engineered and incorporated into the ship. Several of Infinity's early cruises were cancelled.

In January 2001, the Millennium was once again out of service for two weeks due to an "under-performing" electric motor in one of the Mermaid propulsion units. The weak motor limited the ship's top speed to 20.5 knots instead of 24 knots, making it impossible to stay on schedule. Two cruises were cancelled during the repair.

The Infinity's port-side pod drive bearing failure sidelined it for two weeks and forced the cancellation of two more cruises. However, it was not immediately clear why the Infinity had to return to drydock for repairs. According to Rolls-Royce, the entire Mermaid propulsion system can be serviced or replaced in the water.

Pod Drive Technology Questions Persist

Bearing failures and other problems are not unique to Celebrity, lending credence to several leading experts' opinions that pod drive systems are not yet a mature product or technology.

In December 2000, ABB Industry reached a financial settlement with Carnival Cruise Lines over a propeller bearing failure in one of the 14 MW Azipods which power the company's ship Paradise. The Paradise's Azipod bearing failure was blamed on lubrication problems, although an analysis pointed to a "series" of unspecified problems.

The latest generation of Pod Propulsion Systems, the Mermaid, are developed by Rolls-Royce (former Kamewa) with Alstom responsible for the electric drive.

Rolls-Royce's extensive experience of propellers and thrusters, and Alstom's position as a leading supplier of electric propulsion systems makes this team the world's leading independant partner for shipowners and shipyards.

Both companies are long established suppliers to the demanding cruise and offshore sectors, and the Mermaid is expected to have significant influence on future propulsion technology for a wide range of applications suitable for electric propulsion. These include offshore units, naval ships, cruise vessels and ferries.

108/22/02

The Commercial Rim-Driven Permanent MagnetMotor Propulsor PodBill Van Blarcom, Juha Hanhinen, and Friedrich Mewis 1Patent PendingABSTRACTPodded propulsion is gaining more widespread use in the marine industry and is prevalent innewer cruise ships in particular. This propulsion system can provide many advantages to the shipowner, including improved propulsion efficiency, arrangement flexibility, payload and harbormaneuverability. A new unique podded propulsor concept is being developed that allowsoptimization of each element of the system. The concept integrates a ducted, multiple blade rowpropulsor with a permanent magnet, radial flux motor rotor mounted on the tips of the propulsorrotor blades and the motor stator mounted within the duct of the propulsor. This concept,designated a Commercial Rim-Drive Propulsor Pod (CRDP), when compared to a conventionalhub-drive pod, offers improved performance and attributes in a number of areas, including:smaller weight and size, and equal or improved efficiency and efficiency bandwidth, cavitationand hull unsteady pressures. The combination of these CRDP attributes and performanceparameters could allow the ship designer greater flexibility to provide improved ship performance

at reduced cost, as compared to that of a hub-drive pod. The advantages extend across the entireoperating range, from sea trial to off design conditions. The advantages when compared to a hubdrivepod could allow a CRDP to achieve higher ship speeds, or to be applied to a wider range ofplatforms, or to extend the operating envelope of those platforms. The present paper discusses theCRDP’s advantages for both the ship designer and operator, compared to currently availablehub-drive pods.Bill Van Blarcom is with General Dynamics Electric Boat (EB), Groton, ConnecticutJuha Hanhinen is with Deltamarin Ltd, Helsinki, FinlandFriedrich Mewis, is with the Hamburg Ship Model Basin (HSVA), Hamburg, Germany208/22/02INTRODUCTIONGeneral Dynamics Electric Boat (EB) hasdeveloped a commercial rim-driven propulsor pod(CRDP, patents pending) and recently completedhydrodynamic model testing of an 18MW CRDP at 1/25model scale to demonstrate performance potential [1].Testing included powering (open water and selfpropulsion) and measurement of cavitation and hullpressure fluctuations (at 0 o and 8o angle of incidence).The purpose of this paper is to expand on those testresults and provide an assessment of the benefits of theCRDP for a variety of platforms.Principles Of CRDP DesignThe CRDP design balances the hydrodynamicperformance and structural integrity of the propulsorwhile integrating the motor. The key hydrodynamicperformance parameters for the CRDP are highefficiency, good cavitation performance and off-designperformance while maintaining a compact overall size(length and diameter), light weight and structuralintegrity. One of the main advantages of a rim-drivedesign is the mounting of the motor rotor on the rimattached to the propulsor rotor. This allows the motor toproduce a higher torque, thus enabling operation at alow RPM. The low RPM results in low relative velocityover the rotor blades, which contributes to goodefficiency and cavitation performance. An additionaladvantage is reduced flow distortion due to the strutbeing located outside the propeller flow stream. Theseadvantages are enabled by radial field PM motors.The strut and duct of the CRDP are designed withinthe constraints imposed by the motor. The motorrequires provisions for both cooling and electricalconnections and cabling, which affect the strut chordlength and duct geometry. Motor cooling is providedvia seawater flow through the gap between the rotor andstator and seawater flow over the outside surface of theduct immediately behind the motor stator. Formaximum efficiency the strut span can be minimized,but must be sufficient to provide a hull to propulsorstandoff to achieve acceptable hull unsteady pressuresas well as clearance for pod azimuthing. The ductdiameter and thickness should also be minimizedresulting in a short duct length to minimize drag,maximize efficiency, and reduce maneuver resistance of

the pod. The duct diameter is driven by the rotor andstator blade design, while the duct thickness is driven bythe motor design.Scale Model Hydrodynamic Test Results SummaryAnd ConclusionsA complete series of model scale hydrodynamicperformance tests were conducted on a small-scalemodel of the CRDP. These tests includedmeasurements of the open water and behind hullpowering performance in the Hamburg Ship ModelBasin (HSVA) tow tank and cavitation inception, torquebreakdown due to cavitation and hull pressurefluctuation measurements in the HSVA large cavitationtunnel. The CRDP tested was a 1/25.11 scale model ofa unit that was designed to operate on a typical twinscrew panamax cruise ship at a power level of ~18 MWper pod. The CRDP was designed to provide improvedpowering (efficiency), cavitation inception and hullpressure fluctuation performance compared to that of acomparable power and size hub-drive pod with an openpropeller. In addition, the CRDP was designed to haveacceptable cavitation breakdown performance andexperience no cavitation erosion during operation.Conclusions of the test program conducted atHSVA are summarized [1] as:(1) At 1/25th model scale the open water efficiency atthe design advance coefficient of the CRDP is =67.2% and of the comparative hub-drive pod is =64.3% [2], representing a relative improvement of4.5% for the CRDP ({67.2/64.3}-1).(2) Scaling model results to full-scale, the open waterefficiency at the CRDP design advance ratio is =71.7%, and the peak open water efficiency is =72.1%. Applying the same scaling methods to aparticular hub-drive pod yielded consistent resultswith previous full-scale efficiencies, and showedthe CRDP relative improvement at the peakefficiency point to increase further, to about 6%.(3) The efficiency versus advance coefficientdependence of the CRDP (Figure 6 of [1]) showsmuch less sensitivity to off design operation(variation in blade loading) at model scale than withthe comparative hub-drive pod; i.e., even largerimprovements in the CRDP efficiency at off-designconditions. Efficiency curves at full-scale cannotbe shown due to business sensitivities, but thesensitivity difference between the CRDP and hubdrivepod are even more pronounced at full-scale.(4) The improved behind hull efficiency (D) of theCRDP results in the use of less power for givenship speed or increased speed for given power.(5) The CRDP as designed exhibited cavitation-freeoperation at full-scale up to a a cavitation index ()of 2.55 at a 0o angle of incidence and 2.95 at an 8o

angle of incidence. The small amount and types ofcavitation, exhibited above incidence speed by the

CRDP and their stable nature led HSVA to statethat no cavitation erosion would occur on theCRDP at or below the maximum speed, angle ofincidence and blade loading tested. The maximumspeed tested exceeded 26 knots, maximum angle ofincidence tested was 8o at 26 knots, and a test at 24knots straight ahead with a 15% blade overload alsomet the cavitation erosion free criteria. The tests308/22/02did not go to high enough speed or loading topredict when erosion would occur, so the limits ofthe specific design tested are unknown.(6) The hull pressure fluctuations induced by the CRDPhave extremely low amplitudes, in fact the lowestever seen by HSVA or Deltamarin at the clearancestested. CRDP maximum level at blade rate was0.25 kPa at 26 knots and 24 knots with a 15% bladeoverload, even at 8o angle of incidence, comparedto ~1.4 kPa at blade rate at 24 knots at 0o angle ofincidence for a good comparative hub-drive pod.And those CRDP levels showed only a gradualincrease throughout the speed range tested, thussupporting that continuous operation at higherspeeds is viable for the specific pod designed (i.e.,that specific CRDP could potentially be applied to alower resistance, higher speed platform).(7) The hull pressure fluctuations induced by the CRDPat higher harmonics, i.e., 3 times, 4 times, and 5times blade rate, have extremely low amplitudes;less than or equal to 0.09 kPa at 24 knots for 0o and8o angle of incidence.(8) Torque (or thrust) breakdown due to cavitationoccurred at about 26 knots with the hull tested, wellabove the maximum operating speed for the shiphull tested of 24.5 knots. But the falloff rate, afterbreakdown is very gradual, thus supporting thatcontinuous operation above the breakdown point isviable, but with a slight impact on efficiency. Bycontrast, open propellers, such as those of hub-drivepods, typically experience a very rapid fall-off tonear zero from the breakdown point.Benefits Assessment BasisThe following assessment identifies key aspects ofthe CRDP that are considered potentially the mostattractive to the commercial market. Most of thesebenefits are a direct outcome of and supported by theCRDP hydrodynamic demonstration effort performed atHSVA. This assessment combines that knowledge withinsight gained by market research and with additionalCRDP design knowledge (e.g., motor efficiency,cooling).This effort involved market research includingdiscussions with knowledgeable commercial ship andpropulsion machinery designers, builders, testers andowners. Other efforts included gathering reports and

papers, technical research, attending presentations andsymposia, etc., covering numerous types of platformsand propulsion systems. From that research and anexpanding knowledge base of the physical andperformance features of EB’s CRDP in a number ofconfigurations, assessments of the potential benefits anddetractions of the CRDP vs. currently available hubdrivepods in various platforms was accomplished.Since new commercial propulsion system advances arecontinuing, this effort also continues.SIZE AND RATINGSThe initial assessment point is a comparison of theCRDP’s size and rating compared to hub-drive podscapable or near-capable of delivering the same thrust;Table 1 provides that comparison. Table 1 was compiledfrom open literature for the hub-drive pods, with the podratings also based on open literature which identifiesthose ratings to panamax cruise ships.EFFICIENCY AND ASSOCIATED SAVINGSHydrodynamic EfficiencyThe maximum speeds for panamax cruise shipshave also appeared in open literature, with themaximum of any today being ~25.5 kts. Based on thatmaximum speed and the 20MW HDP ratings above, amaximum powering point can be compared to thepredicted maximum powering point for the CRDP on apanamax cruise ship derived from self propulsiontesting [1]. That powering point shows the CRDP tohave a ~7% efficiency advantage, or ~2.6MW lowerpower comsumption. A powering curve based on that7% advantage at all speeds is considered reasonablyrepresentative of the sea trial powering performance ofTable 1- Comparison of CRDP to Commercial Pods @ Approx. Comparable Behind-Hull Unit ThrustABB Azipod [2] Mermaid [2] Dolphin [4] SSP [5] EB’s CRDPNominal Power Rating ~20MW ~20MW ~19MW 20MW 18.5MW (nominal)Continuous Torque Rating ~1340kN-m(est) ~1250kN-m(est) ~1396kN-m 1470kN-m(est) 1918kN-m*Length 11.40 m 11.15 m 13.05 m 11 m 3.90 mHub diameter 2.85 m 2.90 m ~2.8m.2 (est) ~2.9m.2 (est) 1.46 mPropeller diameter 5.80 m 5.75 m 6.0m 6.25 m 4.9m (propeller)5.85 m (duct)*The CRDP continuous torque rating includes a minimum of 30% margin on pullout; that margin is thereforeavailable for temporary maneuvering loads.408/22/02today’s best at-sea hub-drive pods on a panamax cruiseship hull form. That powering curve is shown in Figure1 below alongside the CRDP powering curve fromtesting at HSVA [1]. Alternatively, at the same powerlevel the ship can be propelled at higher speeds with theCRDP than with the comparative hub-drive pod, an ~.4knots higher speed than the hub-drive pod at the CRDPcontinuous torque rating.Motor EfficiencyBesides the hydrodynamic efficiency advantage

demonstrated by model testing, the CRDP could alsoprovide a significant motor efficiency advantage.Wound field synchronous (WFS) motors power mostcurrent hub-drive pods. The CRDP is powered by apermanent magnet (PM) radial field motor, which has aclear efficiency advantage over WFS motors sincepower does not have to be applied to the field (rotor),and field power losses are thus eliminated. Figure 2below shows the efficiency of the PM motor (blue) forthe 18MW CRDP vs. a typical WFS motor (red), plottedvs. ship speed. At full power the PM motor is about 2%more efficient (PM = 98.8%), but the PM motorefficiency advantages are most dramatic at lower powerlevels, approaching a 50% improvement at 4 knots.Figure 2 also shows the combined efficiency advantagefrom the 18MW PM motor and CRDP hydrodynamicperformance, and the resulting power savings vs. shipspeed, with the power savings scale to the right of thisplot. Taking this projection one step further, Figure 3shows revised ship powering curves which now includethe effect of the PM vs. WFS motor efficiencyadvantage. It is noted that these efficiency comparisonsdo not take into account novel drive schemes that mayallow these motors to be driven more efficiently atlower power. For instance, drive technology exists thatcan power part of the windings (e.g., reduced phases),thus reducing winding losses at part loads, but these canbe applied to both motor types. The efficiencydifference might therefore be reduced at loads lowerthan 50%, but the CRDP motor would still be moreefficient than the WFS with the same type of drive.However, there are other factors that bias thiscomparison in favor of the hub-drive pod as discussedbelow, so the overall comparison is consideredreasonable.Figure 1 Behind Hull Powering Results, CRDP vs. Good Representative Hub-Drive Pod508/22/02Figure 2 CRDP w/PM Motor Efficiency vs. Conventional Hub Drive and WFS MotorFigure 3 Behind Hull Powering Curve, CRDP vs. Comparative Hub-Drive Pod with Both Hydrodynamic andPredicted PM vs. Expected WFS Motor Efficiency Accounted forPotential Annual SavingsFigure 4 below takes this efficiency comparisoneven further, by showing a potential annual power/costsavings for the CRDP in a panamax ship. Thisprojection was developed from an average annualoperating profile of several panamax ships from dataprovided by Deltamarin and projecting the poweringdifference of the CRDP hydrodynamic and motorperformance vs. a conventional hub-drive pod withWFS motor from Figure 3. Note that for ~40% of theannual hours the ship is at standstill. The powergenerating cost and efficiency, $0.10/kW-hr and 95%608/22/02

drive efficiency, have been used in other commercialmarine papers and appear reasonable by comparison toother more complex estimating methods evaluated.Projected Efficiency Gains Are ConservativeSome factors not accounted for in this savingsprojection that bias the results in favor of the hub-drivepod, and thus bring some degree of conservatism to thisprojection are:Projections Are Based on Sea Trial Conditions,Resulting in Lower Than Average Power RequirementThe powering projections at all speeds, as shown inFigures 2, 3, and 4 are for straight ahead, sea trialconditions (clean smooth hull, deep calm water, nocurrent, 2.365m/sec headwind (Beaufort 2, ~5kts)).These are not representative of even average conditionsover the life of these ships, which include operating atthe following conditions:Deepwater conditions: Trim, wind, current, wavesand hull fouling are factors having significantimpact on ship resistance. A +15% loading factor isconsidered a normal adjustment from sea trial toaverage deep-water conditions. In heavy weatherthe overload condition can easily be 50%.Shallow water conditions: Water depth also hasextremely strong influence on resistance. In onereport it was noted that that for panamax size cruiseships (~8m draft) strong depth impact starts around30m water depth and in 15m deep water thesevessels can typically only reach 50% of top speed.Low speed operation: At lower speeds in particular,sea trial conditions are the most unrepresentative,since lower speed ranges are likely in shallowdepth, high harbor maneuvering conditions wherepropeller loading would be considerably increased.And in those maneuvering conditions the pods areusually turned into a “crabbing” orientation, inwhich they are typically oriented between 30 to 90degrees to each other to allow rapid thrust vectoring(e.g., see Figure 5). The 4 to 12 knot poweringportion of the Figure 4 powering comparison isbased on both pods powering from the 0o angle ofincidence position; the crabbing position changesthis. Thus higher blade loads will be experiencedduring low speed operation than has been analyzed,and those operations will be at inflow angles ofincidence to the pod, both factors increasing theadvantage of the CRDP.CRDP Operation at These More Severe OperatingConditions Will Be Even More EfficientThe CRDP’s higher efficiency and flatter efficiencyvs. speed of advance curve as shown in Figure 6 below(Figure 6 of [1]) demonstrates that the CRDP willperform even more efficiently at higher, more normalloading conditions (lower advance coefficient, J) andresult in additional savings. This figure shows modelFigure 4 Potential CRDP Annual Fuel Savings for Representative Panamax Cruise Ship

708/22/02scale open water efficiency (o) versus advancecoefficient for the CRDP and a good comparative hubdrivepod. As shown, at the peak efficiency advancecoefficient (J/JPeak = 1), the CRDP shows ~4.5% higherefficiency at model scale compared to therepresentative hub-drive pod. But in addition, at offpeak advance coefficients the CRDP efficiency isshown to be much less sensitive than that of thecomparative hub-drive pod. As an example (from [1]),if the off design operation is limited to a 3% drop inefficiency from the HDP peak (to about 61.3%), the offdesign operation range of the CRDP is almost twicethat of the hub- drive pod range (.54/.28 = 1.93). Thisinsensitivity of the CRDP to off design operation canenable lower ship operating costs and higher operatingspeeds in heavier sea states with the CRDP. Inaddition, it allows the design and use of fewer CRDPunits for operation over a range of power levels thanpossible with hub-drive pods. And while impressive atmodel scale the difference is even more at full-scale,although it cannot be presented due to businesssensitivities.Thus using sea trial conditions is conservative inthe annual powering projections of Figures 3 and 4.Figure 5 Typical Pod Crabbing Orientation for HarborManeuveringAdditional CRDP System Efficiencies Are NotAccounted ForAdditional electrical system power savings. TheCRDP’s PM motor operates at a higher powerfactor (~.94) than does a WFS motor (~.72 to .82);that difference can amount to a ~ 1% higherefficiency of the generator and distribution system.Less secondary system power consumption.The CRDP does not require a dedicatedcooling system, and therefore the energy to runsuch a system is saved (See “Secondary ShipDesign Impacts/Opportunities“ discussion onpage 10).The CRDP also uses seawater lubricatedbearings instead of lubricated oil bearings.The ship’s lubrication systems energyconsumption is thus also reduced.These differences are also not accounted for in Figures3 or 4, thus further adding to their conservatism.oFigure 6 Open Water Efficiency at Model Scale of aCRDP Developed for a Panamax Cruise Ship vs. aRepresentative, Good Hub-Drive PodSHIP DESIGN OPPORTUNITIESThe CRDP could offer more freedom to the shipdesigner; in certain circumstances this may be asignificant advantage.

Narrower Ship Beams, And Unique Configurations:The pod size itself offers an obvious benefit fornarrow beam ships, evident by the length comparison ofTable 1 and also depicted in Figures 7 and 8. Butbeyond the conventional twin screw ships as tested theCRDP size can also support more unique configurationsthat conventional hub-drive pods cannot, or can supportthem in more flexible arrangements.808/22/02Figure 7 Typical ~20MW Commercial Pod vs.~18.5MWCRDP Shown in Relative Size(CRDP ~1/3 length)Figure 8 Typical Twin-Screw Panamax Cruise Ship -Hub-Drive Pod vs. CRDP Arrangement ComparisonConsider, for instance, a three-pod arrangementsimilar to the “Voyager of the Seas” Class (Figure9). Those ships have two azimuthing “pulling”pods (facing forwards) and one fixed pushing pod,with the pod size (power) and spacing betweenpods dictated by the azimuthing pods turningcircle. The CRDP, being shorter in length, couldsupport a two-pod arrangement deliveringcomparable thrust, or a three across, all podsazimuthing configuration, if desirable to the shipdesigner or owner (Figure 10). It could alsosupport more pod arrangements than the hub-drivepods, which might be of advantage for locatingpods in more ideal wake locations and thus furtherimprove hull efficiency (H) or cavitationperformance as desired.Figure 9 Three 14MW Pod Arrangement on Voyagerof the SeasFigure 10 HDP vs. CRDP in Narrow Ship BeamArrangementsIn the case of a four pod ship, such as the QueenMary 2 (as noted in reference [3]), the CRDP couldobviously support more pod arrangementopportunities than a conventional hub-drive pod.The 4 hub-drive pod arrangement for the QueenMary 2 will include two azimuthing and two fixedpods. A more flexible arrangement, if desired,with all azimuthing pods can easily be imaginedwith the CRDP, given the significantly smaller pod908/22/02size. But also of note in reference [3] is astatement that the four-pod configuration wasselected after a three-pod configuration wasevaluated (26.5MW each); the three-podconfiguration was abandoned due to excess per podweight, in excess of 300 tonnes. Since the CRDPalso offers a weight advantage it carries a lowerprobability of creating a trim problem for the ship,allowing for a more rationale distribution of themachinery and load items, and it might thus

support a three-pod configuration where the hubdrivecould not.Also consider a booster pod arrangement, such asthe Costa Classica extension project. The smallerCRDP length can allow the pod to be locatedfurther behind the center skeg, due to the smallerpod turning circle. The ship designer would haveto determine whether there was any advantage on aparticular ship, but with that freedom it mightallow internal arrangement improvements or resultin higher efficiency and ship speed because of amore favorable wake in that position. Anotheropportunity for the designer would be putting in amore powerful azimuthing pod in the same space.More Freedom In Stern Configuration:The CRDP also offers opportunities to reconfigurethe stern lines, which could enable increased payload,increased hull efficiency (and further reduce operatingcosts), or podded propulsion of Ro-Ro platforms whereconventional hub-drive pods aren’t feasible.Two CRDP Features Enable These DesignOpportunities:Lower Vibration Levels.The CRDP pressure fluctuation amplitudes aresmaller than hub-drive pods, which may allowreduced clearance between the pod and hull.Hull clearance (clearance between propellerblade tips to nearest point of hull) is typically setby propeller cavitation effects, and in particularhull vibration associated with cavitation. Thepropeller operates in a flow field affected by thehull, which is decelerated and non-uniform into thepropeller, and has negative effects on propelleroperation. The propeller induces an unsteadypressure field that affects the submerged part of thestern, mainly caused by cavitation. This unsteadycavitation is often the main cause of ship vibrationproblems [2]. The vibration tolerance level variesdependent on the type of ship. As a rule of thumb,cruise liners and cruise ferries are typicallydesigned to achieve pressure amplitudes on the hullof less than 2kPa at blade rate. By comparison fastferries would typically allow 3 to 3.5kPa, containerships with installed propulsion power in the rangeof 20 to 30MW would typically allow 3 to 5kPaand tankers would typically allow 5 to 6kPa.Clearance for a cruise ship has typically been 25%to 35% of the propeller tip diameter to achieve itsvibration tolerance level; other ship types would ofcourse have different typical clearances. The tip (orhull) clearance both provides distance to dissipateenergy from the source (propeller cavitation) aswell as placing the propeller in a more benign wakeand thus limiting cavitation.In the CRDP configuration tested, withroughly comparable hull clearance (CRDP duct

outside diameter and clearance hub-drive podblade tip outside diameter and clearance), pressurefluctuation levels are so dramatically lower thanconventional pods that the CRDP clearance to theship could be reduced. CRDP levels are less than20% of the hub-drive pod levels as noted inreference [1] and page 3, paragraph (7) herein.Note also that the CRDP levels are ~10% of the“normal tolerance level” of 2kPa for a cruise ship,even at an 8 o angle of incidence. Besides showingdramatically lower levels these results also showless sensitivity to wake than do hub-drive pods,further supporting the tolerance to smaller hullclearances.Pod ConfigurationThe CRDP 18MW design for thisdemonstration was developed with a self imposedspecification on duct diameter and clearance propeller diameter and blade tip clearance of thecomparable hub-drive pod. But the CRDP couldbe redesigned with a smaller duct diameter ifdesired. Although the impact would be a reductionin efficiency and lengthening of the CRDP boththose features have significant margin to trade offvs. the comparable hub-drive pod. And efficiency,although affected by a reduced diameter would notbe impacted as much as would a conventional hubdrivepod if reduced by the same amount.-- Resultant Stern Configuration OpportunitiesThe stern could be lowered thus allowing moregentle ship lines.Thus providing a more gentle distribution ofdisplacement, reducing the boundary layerthickness and steady pressure gradient.Consequently this could reduce ship drag andthereby improve behind hull efficiency (D).This gain might be great enough to offsetreduced pod efficiency in the case of applyinga smaller diameter CRDP in order to lower thestern.1008/22/02More gentle ship lines could create more aftpayload space.Ro-Ro ships must work the requirement to rollcargo on and off the ship around the draft andinternal height constraints for the propulsion andazimuthing system. In the case of an azimuthingpodded propulsion system the azimuthing systemimposes a height requirement directly above thepropeller that non-podded ships do not have. Thishas been a reason for disqualifying poddedpropulsion in some Ro-Ro designs. The reducedhull clearances and/or smaller propeller/ductdiameters enabled by the CRDP might thus enable

more extensive podded propulsion on Ro-Ro ships,by allowing deeper stern lines and thus loweringthe internal height constraint.More PayloadAs noted above, the CRDP allows sternconfiguration opportunities that may open up additionalspace for payload. But by virtue of the CRDP’simproved efficiency it can also offer additional payloadopportunities by allowing fuel bunker reductions whilestill supporting the same service/refueling range.Secondary Ship Design Impacts/OpportunitiesThe CRDP motor does not require a dedicatedcooling system, therefore there is less coolingsystem demand resulting in secondary efficiencysavings and cooling system equipment reduction.The CRDP features allowing the elimination of thecooling system are:The CRDP rotor, being a properly designedPM machine, generates little losses.The CRDP stator, being located in the duct, iscooled by the seawater passing by the hull,both internal and external of the duct.- The surface area of the duct immediatelysurrounding the motor, being muchgreater and more uniformly exposed to thepassing flow than the comparable surfaceof a hub-drive pod motor, enables thiscooling method.- Also, the CRDP motor stator core lengthis considerably shorter than a comparablehub-drive pod, since the stator corediameter is considerably greater than thehub-drive pod’s. The shorter core lengthshortens the conduction path from thecenter of the core to the end-turns,enabling more uniform temperaturesthroughout the CRDP stator.Lower power demands due to higher CRDPefficiency can support lowering power plant space,weight and cost.Passenger comfort at higher speeds/higher seas.The low pressure fluctuation levels demonstratedby the CRDP both straight ahead at high coursekeeping angles of attack could allow vessels tooperate at higher speeds or fill more spaces aft withpassengers without a reduction in their comfortlevel.SHIP OPERATION OPPORTUNITIESHigher performance in all operating conditionsAs previously noted, the CRDP’s higherefficiency and flatter efficiency vs. speed of advancecurve, as shown in Figure 5, demonstrates that theCRDP will perform more effectively and efficiently atall loading conditions. A maximum continuous thrustvs. ship speed analysis for the tested panamax hull wasperformed. The analysis assumed the CRDP and the

hub-drive sea trial powering curves of Figure 1 as astarting point. Based on that starting point themaximum continuous powering/thrust capability ateach ship speed is based on the motor continuoustorque ratings (Table 1) and open water performancecurves for each pod (e.g., Figure 5 of [1] and Figure (6)of [2]). The advance coefficient (J), was varied untilthe torque coefficienct (KQ) of each pod produced themaximum continuous torque; the thrust coefficient (KT)was then determined and used to calculate the thrust.Figure 11 below shows some of the sea trial andmaximum thrust operating points on the CRDP andhub-drive pod open water efficiency curves. At the seatrial operating points for each pod they are deliveringequivalent thrust, at the maximum continuous thrustpoints they are both operating at the motor continuoustorque point. In addition, operating points are shownfor the CRDP which match the hub-drive pod’smaximum continuous thrust point, thereby enablingcomparison of efficiency at the same thrust.By comparing the sea trial operating points itshould be noted that the hub-drive pod was given aslight additional advantage in this analysis since its seatrial starting point is at a more favorable point than theCRDP’s for all ship speeds. The hub-drive pod’s seatrial point is just past its peak efficiency point whereasthe CRDP‘s is just before that point. For thrust andblade loading to increase, propeller speed must alsoincrease, and J therefore decreases. From Figure 11 itcan be seen that efficiency will therefore initiallyincrease from the sea trial condition as J decreases forthe hub-drive pod whereas efficiency will only decreaseas J decreases for the CRDP. Thus the relativeefficiency and maximum thrust benefit predictions forthe CRDP thus computed should be conservative at allthrust conditions and all ship speeds.1108/22/02The analysis showed the CRDP relativehydrodynamic efficiency advantage growing to over10% at 13 knots and over 13% at 10 knots whenmatching maximum thrust capability of therepresentative hub-drive pod at those speeds. Or, theCRDP can produce steadily greater thrust than the hubdrivepod, 2.5% additional at 24 knots but increasing to20% additional thrust at 10 knots, thus enabling highership speeds in heavier seastates, casualty conditions,etc. In addition, cavitation and pressure fluctuationtesting has demonstrated that the higher CRDPefficiency performance also comes with improvedcavitation performance over a broader range of bladeloading as well.Figure 11 CRDP vs. Hub-Drive Pod, Operating Points onOpen Water Performance Curve Predicted for Sea Trial andMaximum Thrust ConditionsLess Maintenance

The CRDP is expected to require significantly lessmaintenance than other pods for a number of reasons aslisted below:Less cavitation erosion can be expected, thereforeresulting in lower maintenance costs. Besides thecavitation testing that supports this conclusion isone design difference of note: the strut positionrelative to the propeller’s swirling discharge. Inthe case of hub-drive pods, usually 1/3 to 1/2 thestrut is exposed to the strong propeller discharge,and there have been reports that this exposed areais highly susceptible to cavitation erosion as wellas vibration excitation. The CRDP strut on theother hand is entirely outside and protected fromthe propeller discharge and should not experiencecomparable erosion or vibration. CRDP statorvane erosion (stationary blades) might be ofconcern since they are exposed to propellerdischarge, but these vanes are designed as amatched set with the rotor blades to minimizecavitation erosion amongst other factors.Bearings. The CRDP uses seawater lubricatedjournal and thrust bearings, thus avoiding thenecessity for seals to protect oil filled bearingcavities. This type of bearings has been used formany years and while successful they have evolvedfrom yesteryear’s brass backed rubber stavebearings to special polymer materials today thatimprove bearing life and reduce friction. TheCRDP journal bearing is designed within industrystandard design guidelines for projected areapressure loading. The thrust bearing is designed tooperate at higher pressure than that calculated forthe journal bearing projected area pressure. Thethrust bearing is designed to operate atapproximately the peak calculated journal bearingpressure, which is approximately six times theprojected area pressure. The thrust bearing designwas tested and verified using a scale model bearingapproximately 1/3rd the diameter of the CRDPthrust bearing at maximum CRDP surface speedand pressure. The thrust bearing designdemonstrated little or no wear while operating atmaximum CRDP conditions with a frictioncoefficient of 0.005. The expected maintenanceinterval of both the CRDP journal and thrustbearings is at least 12 years.Current commercial propulsion pods bycontrast do not appear to either incorporate a robustbearing service life or separate bearing cavity sealsto prevent seawater contamination of thelubrication system. The typical commercial poduses oil-lubricated roller bearings for both the shaftradial and thrust bearings, which have become amaintenance problem for many ship operators,requiring expensive dry-dock periods to

disassemble the pod(s) and replace roller bearings.The current bearing problems appear to beaggravated by early seal failures, in some reportedinstances at least, that introduce seawater into thebearing cavity and lead to rapid bearing failure.Also, the failed seals and flooded bearing cavitiescan allow oil to escape the pods and becomepenalizing environmental spills.Cooling System: Less cooling system equipment isrequired since the CRDP is totally cooled bynaturally passing seawater past the duct and by thepressure developed by the propeller rotor causingseawater flow through the gap between the motorstator and rotor. This feature also lowers overallsystem noise levels by eliminating disturbingnoises emitting from cooling fans, etc., which arerequired for most other pods.Lubrication System: Separate bearing lubricationsystem is avoided, since the bearings are seawaterlubricated.1208/22/02Higher Attainable Ship SpeedsFrom information gathered, the greatest apparentobstacle to achieving higher ship speeds with currentcommercial hub-drive pods is primarily due to highcavitation and pressure fluctuation levels, aggravatedby pod dithering in high speed course keeping and bypotential for strong propeller/strut vibration and erosioneffects (as also noted in the “less maintenance”discussion above). The CRDP's demonstratedperformance on this initial pod demonstrator, which bythe way was not designed to achieve higher speed thanthe maximum speed identified for the particularpanamax cruise ship, was dramatically better than thecomparative hub-drive pod.Higher performance in single pod operation. Manypod propelled cruise ships have had publicizedpropulsion system problems. Other podded ships,particularly other ship types, may have also experiencedproblems but have not been as widely publicized.Some of these casualties were known to involve at leastone pod, and others, while not necessarily caused by thepod, may have still disabled powering one pod. Someof these resulted in cruise cancellation in mid-cruisewith considerable revenue impact. It is reasonable toassume several of these resulted in single pod operationto either complete the cruise or get back to port.Infinity and Summit in fact were noted to have operatedin single pod mode at lower ship speeds and modifieditineraries in order to support scheduled cruises whileawaiting a time window for repairs. It is thereforerealistic to consider this operating mode as being ofsome interest to a commercial ship owner/operator. Asalready noted the CRDP can deliver more thrust atsomewhat higher blade loading than the hub-drive pods

and therefore will support higher speeds in thesecasualty conditions.CONCLUSIONSIn comparison to currently available commercialhub-drive pods:CRDP delivers required net thrust with a smallerpod (~1/3 as long).CRDP is more efficient in all operating conditions(e.g., sea trial condition, fouled hull, maneuvering,and varying sea states) than current hub-drive pods.Overall efficiency advantage is a result ofhydrodynamic, motor and secondary systemefficiency advantages.CRDP is less prone than hub-drive pods toperformance degradation, both straight ahead andat steering angles.CRDP provides better passenger comfort; ~5 timeslower hull unsteady pressure levels for given hullclearance will result in reduced hull vibrations.CRDP can achieve higher ship speeds withacceptable cavitation and no risk of cavitationerosion.CRDP allows more hull design flexibility(narrower ship beams, reduced clearance to hull,more gentle and fuller stern lines, etc.).CRDP can support a wider range of operatingmodes, such as single pod propulsion on multiscrewships, than conventional hub-drive pods.CRDP allows use of an alternate water lubricatedbearing system that does not require lubricating oilor seals and is expected to be more reliable than thecurrent practice of oil lubricated roller bearings.CRDP is expected to require less maintenance.The maintenance advantage results from reducedcavitation erosion, reduced support systemequipment (e.g., reduced cooling and lubricatingsystem requirements) and the alternate bearingsystem noted above.The CRDP, therefore, should be a moreeconomical choice for a wider range of ship types thanother propulsion alternatives.REFERENCES[1] Lea, M.; Thompson, D.; Van Blarcom, B.; Eaton,J.; Richards, J.; & Friesch, J; “Scale ModelTesting of a Commercial Rim-Driven PropulsorPod,” September 2002[2] Mewis, F., “HSVA Seminar for Ship Ownersand Operators,” 10May01[3] “Naval Architect” (publication of the RoyalInstitute of Naval Architects, UK), January 2001[4] DOLPHIN – A John Crane –Lips/STN ATLASMarine Electronics Podded Propulsion System,DS 1.036.01/2000[5] “The SSP Propulsor, An Ingenious Podded DriveSystem,” 159U538 02981NOMENCLATURE

ARL The Pennsylvania State University AppliedResearch LaboratoryCRDP commercial rim–drive propulsor pod, patentspendingDR rotor diameterEB Electric Boat Corporation, a GeneralDynamics CompanyHSVA Hamburg Ship Model Basin (HamburgischeSchiffbau-Versuchsanstalt GmbH)ITTC International Towing Tank ConferenceJ advance ratio or coefficient = V/nDR

KT thrust coefficient = T/n2DR

KQ torque coefficient = Q/n2DR51308/22/02L lengthMW megawattn rotor rotational speed, in revolutions persecondQ steady torquePD Power delivered to propeller = 2nQPE Effective Power delivered by propeller pod =RT VRT Total Ship ResistanceRPM revolutions per minuteT steady thrustt thrust-deduction fraction = (T-RT)/TV ship speedVA speed of advance of propeller = V(1-w)VRel relative velocityw Taylor wake fraction = (V-VA)/VO complete pod efficiency in open water= (JKT)/(2KQ)D propulsive efficiency (a.k.a., quasi-propulsivecoefficient)= PE/PD = R O H

H hull efficiency = (1-t)/(1-w)R relative rotative efficiency= KQ (open water test) /KQ (propulsion test)

nwater kinematic viscositywater mass densitycavitation index (related to ship speed)= (po-pv)/ (0.5(VA)2)multiplication signAUTHORSBill Van Blarcom is a Principal Engineer atGeneral Dynamics Electric Boat, Groton, CT. He holdsa Bachelor of Science in Mechanical Engineering fromRensselaer Polytechnic Institute, Rensselaer, NY, and aMaster of Science in Mechanical Engineering from theUniversity of Connecticut, Storrs, CT. He has workedin various engineering positions at Electric BoatCorporation over the last 28 years including analysis,design development, program management andsupervision. His current efforts at Electric Boat have

been in advanced concept development and evaluationof integrated electric power/propulsion systemcomponents.Electric BoatEastern Point Rd.Groton, CT 06355(860) 437-5631; (860) 437-5428 (fax)bvanblar@ebmail.gdeb.comJuha Hanhinen is a Naval Architect fromDeltamarin Helsinki Office. He did his masters thesisfor Azipod about azimuthing podded propulsors andreceived his degree from the Technical University ofHelsinki in –94. After short periods in HelsinkiShipyard and Finnish propeller manufacturerFinnScrew, he joined the Deltamarin team in summer –95. Since then he’s been heading the company’shydrodynamic group and closely followed thedevelopment of commercial pod propulsors and theirintroduction in cruise ship applications.Deltamarin LtdSarkiniementie 7FIN-00210 Helsinki, Finland358 2 4377 311; 358 2 4380 378 (fax)juha.hanhinen@deltamarin.comFriedrich Mewis is a Director of the Hamburg ShipModel Basin (HSVA) where he is also head of theResistance and Propulsion department. He holds a“Diplomingenieur für Schiffbau” from the University ofRostock, Germany. For the previous 27 years he wasemployed at the Potsdam Ship Model Basin (SVA),where he specialized in the areas of resistance andpropulsion as well as hull form and propulsionoptimization. He has continued his work in this field atHSVA, where he has been for 6 years. He has published15 major papers, among them “The Efficiency of PodPropulsion”, HADMAR’2001, Varna, Bulgaria. He issecretary of the Propulsion Committee of the 23rd

ITTC.Hamburgische Schiffbau-Versuchsanstalt (HSVA)GmbHBramfelder Strasse 164D-22305 Hamburg, Germany49 (40) 69 203-224; 49 (40) 69 203-345 (fax)Mewis@hsva.deACKNOWLEDGEMENTSThe authors recognize that numerous, diverseteams of people on both sides of the Atlantic worked tosupport this demonstration effort, and it is with regretwe cannot name them all, but all are thanked herewith.The following limited list of individuals wereconsidered the greatest supporters, and most gratefullyacknowledged:From Electric Boat: Al Franco, Scott Forney, Chas.St. Germain, Dan Kane, Charles Knight, Michelle Lea,Donald Thompson, Spyro Pappas, Stu Peil, MarkWarburton. From HSVA: John Richards, Jürgen

Friesch, Karl-Heinz Koop, Eckhard Praefke, DietrichWittekind and Gerhard Jensen. From Penn State ARL,Jon Eaton.

Manufacture of a Prototype Advanced PermanentMagnet Motor PodPiet Van DinePrincipal EngineerElectric Boat CorporationSeptember 2002AUTHORPiet Van Dine is a Principal Engineer at Electric Boat Corporation, Groton, CT.He holds a Bachelor of Science in Naval Architecture and Marine Engineeringfrom the University of Michigan, Ann Arbor, MI. He has worked in variouspositions in the marine industry over the last 21 years including field engineeringfor General Electric, engineering assistant for Norfolk Naval Shipyard and 19years in engineering positions at Electric Boat. His efforts at Electric Boat havebeen in advanced concept and component development. Mr. Van Dine has beenactively working with composite material over the past twelve years and isresponsible for advanced concepts composites efforts. He has written severalpapers and submitted several patents relating to composites manufacturingprocesses and applications, including eleven that have been awarded.40 Overlook Ave.Mystic CT 06355(860) 437-5220 (phone)(860) 437-5428 (fax)Manufacture of a Prototype Advanced PermanentMagnet Motor PodPiet Van DinePrincipal EngineerElectric Boat CorporationABSTRACTPodded propulsion is prevalent in the marine industry. Poddedpropulsion systems provide many advantages to the ship owner,including increased propulsion efficiency and reduced constructioncost. To evaluate the potential of a new pod configuration, aprototype machine was constructed and tested. This prototypemachine was mainly constructed of composite parts. The propeller,housings, structural blading, motor canning and fairings wereconstructed of composite materials. Composite materials werechosen as a cost saving, schedule reduction, performance

enhancement and as a technology demonstration. This paper willreview the unit construction, and test results, focusing on thelessons learned for the composite part manufacture.KEY WORDS: Pods, CompositesINTRODUCTIONPodded propulsion has become a shipbuilding standard for commercial ships.This propulsion alternative offers the ship owner many advantages. Theadvantages include reduced ship acquisition costs, improved propulsionefficiency and improved ship arrangements. Podded propulsion has taken theform of a driver contained in a hull, turning the propeller by a shaft. The driverhull (or pod) includes a shaft sealing system to prevent water from entering. Thepod is connected to the ship hull by a strut. The strut can be connected directlyto ship structure or to an azmuthing system. The azimuthing system offers theship owner improved maneuvering over standard rudder systems. Figure 1 is anartist’s depiction of one of the most popular podded propulsion systems,produced by Asea Brown Boveri (ABB).Figure 1: ABB Azipod Podded Propulsion {1}An alternate podded propulsion system has been invented, incorporating thedriver or motor on the rim of the propeller. The motor rotor is attached to the rimand driven by the motor stator, which is located outside the rotor. Thispropulsion system has been shown analytically and empirically to offer improvedefficiency and reduced weight when compared to current podded propulsion.The configuration of a rim driven pod (RDP) is represented by Figure 2. The podconfiguration and many details are Patent Pending.Figure 2: Rim Driven Pod ConceptA prototype RDP was designed and constructed to provide an empiricaldatabase to validate the analysis. This hardware was designed to be testedstatically, and so reproduced the flow path and driver hardware properly. Theexternal shape was not faired due to the static test conditions. The bearingsused for the initial demonstration were angular contact ball bearings. Thissystem offered a low risk, low cost alternative to hydrodynamic bearings as areexpected to be used in many future applications. The unit was manufactured ofcomposite materials to minimize costs and schedule. Cost and scheduleadvantages stemmed from the non-production nature of the part and novelcomposite manufacture techniques employed. The composite material alsooffered some definable performance advantages. Figure 3 is a picture of theassembled prototype RDP.Figure 3: Prototype Rim Driven Propulsion PodDESIGNThe prototype RDP was designed for static operation, the thrust was absorbed ina moored barge. The integration of the various components to create the RDPassembly was fully evaluated in the design and manufacturing stages to precludeproblems in future units. Other features of the prototype RDP included:Rolling Element (ball) bearings enclosed in a pressure compensated housing.Static lip seals to prevent bearing oil from leaking into the water.A permanent magnet (PM) motor design.

A two stage propeller blade set, the first rotating, the second stationary,canceling swirl from the first to maximize efficiency.Two composite stator can concepts, a solid can and a hollow pressurecompensated can.A solid composite canning for the motor rotor.Composite blading manufactured with a patent pending concept, whichmanufactures each blade independently and then joins them into a monolithicstructure.End bells and cones manufactured of composite materials.Figure 4 is a cross sectional drawing of the prototype RDP with a pressurecompensated hollow stator can. This drawing represents the option that wasmanufactured, assembled and tested.Figure 4: Prototype RDP DrawingThe design attributes of the prototype RDP are:

• 120 Horsepower

• 500 RPM

• 24 Poles

• 144 Stator Slots

• 6 Phase Motor

• 0.45 Inch Electrical GapMANUFACTUREThis section will review the manufacture of the key parts of the prototype RDP,the methods used and the lessons learned. These key parts include the rotor,the stator can, and the stationary blading. The manufacturing methods chosenreflected the required part configuration. The two primary manufacturing428 Volts75Hz98.188% Motor Efficiencymethods used were a vacuum assisted resin transfer molding (VARTM) andfilament winding. The VARTM methods used both soft and hard molds. Theresin system for the parts was a DOW DEREKANE Vinyl Ester with E – Glassreinforcement. A metallic structure was included as a permanent part of someparts as reinforcement, load carrying, wear surfaces or thread pads. Partshrinkage rates and cure were calculated based on empirical VARTM data. A2% average shrink rate was used for the thick parts. This rate was proven to beappropriate for the reinforcement, resin and fillers used.RotorThe rotor was manufactured (using a patent pending process) by first moldingimpeller sections that include the entire blade span and portions of the hub andshroud, see Figure 5-1. These sections were then assembled with adhesive toform a complete set of blades. The set of blades was then placed in a mold with

a glass wrap on the outside of the blade assembly, see Figure 5-2. Thisassembly was then injected to form the shroud. The metallic hub was insertedwith glass fiber and injected with resin, see Figure 5-3. The part was thendemolded and the rotor finish-machined to accept the motor rotor. The motorrotor is then installed, see Figure 5-4. Glass was packed around the rotor, andthe rotor VARTM injected to form a solidly canned part, see Figure 5-5.Figure 5-1. Composite Rotor SegmentsFigure 5-3. Rotor with Shroud and HubFigure 5-2. Rotor Segments in ToolingFigure 5-4. Rotor with Motor RotorFigure 5-5: Completed Rotor with Motor Rotor CannedStatorThe stator vanes were manufactured (using a patent pending process) in thesame manner as the rotor, by first molding sections that include the entire bladespan and portions of the hub and shroud, see Figure 6-1. These sections werethen assembled with adhesive to form a complete set of blades, see Figure 6-2.The set of blades was then placed in a mold with a glass wrap on the outside ofthe blade assembly. This assembly was then injected to form the shroud. Theassembly was then installed in a VARTM cylinder on which aluminum stiffenersand thread plates were attached with resin, see Figure 6-3. The part was thendemolded, see Figure 6-4. The part finish machining completed the effort, seeFigure 6-5Figure 6-1: Stator Segments Figure 6-2: Assembled Stator SegmentsFigure 6-4: VARTM Stator Assembly Figure 6-3: Stator AssemblyFigure 6-5: Finished StatorStator CanThe stator can was manufactured using two different methods. The first methodused was a solid encapsulation. A hollow, pressure compensated canningmethod was also used. Figure 7-1 is the motor stator that was canned.Figure 7-1: Motor StatorThe solid canning was manufactured by inserting the motor stator in a hard toolwith locating features and dry glass cloth. The part was then VARTM processed,resulting in a single monolithic structure, see Figure 7-2. This method resulted insome shifting of the motor stator during the VARTM process. The reason for theshifting was the curing pattern of the part and the inadequate locating features.Based on these problems a process revision was developed for future use. Thesolid canning process is protected under US Patents #06069421, #6150747 andpending patents.Figure 7-2: Solid Canned StatorThe other configuration used to manufacture the stator canning was a pressurecompensated hollow can. This configuration is also protected by pendingpatents. The manufacturing process entailed filament winding of inner and outercomposite cylinders, see Figures 7-3 and 7-4. These cylinders are assembledwith metal end plates and O – Rings to form a sealed container, see Figures 7-5,7-6 and 7-7. The assembly was tested by immersion in water and pressurizingwith air. The air pressure was held for 30 minutes and the water watched for

bubbles. The container was then disassembled for motor stator installation. Themotor stator was installed and positively located by mechanical connection to anend plate. The container assembly was then completed and installed on the unit,see Figure 7-8. This container was filled with dielectric fluid and was externallypressure compensated. This assembly resulted in a reliable test configurationwith positive locating features. The unit testing proceeded based on this motorstator can configuration.Figure 7-3: Outer CanFigure 7-8: Canned Stator on Unit Figure 7-7: Outer CanFigure 7-6: Partially Assembled Can Figure 7-5: Can End PlatesFigure 7-4: Inner CanOther PartsThe other major parts were manufactured in two methods; the end bells weremanufactured using soft molds and the VARTM process. The cones weremanufactured by forming a metallic core and filament winding over the core. Theparts were then finish machined. Figure 8 is a picture of the parts ready forassembly.Figure 8: Unit PartsProcess ReviewThe processes used to manufacture the prototype RDP were successful. Theprocess could be improved for the end cones by reshaping the metallic core tobe more conducive to the filament winding process. See Figure 9 for the conewinding form that was used. Various details and permanent tooling wouldimprove efficiency.Figure 9: End Cone Winding FormASSEMBLYThe parts were assembled to complete the prototype RDP. The processrequired a careful attention to details to ensure that parts were not damaged.The fact that powerful permanent magnets were a part of the unit caused strongforces during assembly. The use of composite materials reduced this attractiveforce and the parts were positively controlled during the entire assembly. Figure10 is a photograph of the completed unit.Figure 10: Completed prototype RDPTESTTesting was conducted by mounting a pipe to the top mounting plate. Fourtethers were used to attach to the corners of the test barge. Figure 11 is apicture of the prototype RDP entering the water for testing. This configurationformed a solid truss support that held the prototype RDP steady. The testingincluded operation across the range of speeds and at varying depths to stressthe design. The test results proved the predicted performance results, increasedefficiency of the motor design was in the 6 efficiency point range over metalliccanning (predicted 98.188% for composite, 92.285% for metallic). The metalliccanning losses would have been directly related to eddy currents in the statorside gap. The hydraulic design resulted in better efficiency than predicted. Thehydraulic efficiency differences were due to blade design improvements betweenpredictions and manufacture as well as better than expected inflows. No

integration issues or design flaws were found as a result of testing. Thecomposite parts in particular performed well. The stator can resulted in no eddycurrent losses as would result from a metallic can.Figure 11: Prototype RDP in Test ConfigurationCONCLUSIONSAn alternate pod configuration was manufactured and tested. The prototypeRDP met the analysis predictions. The cost to produce this unit in its entirety was35% lower than was quoted for a metallic unit. The time to reach the test for theprototype (including unscheduled problems) was the same as quoted for themetallic manufacture. First of a kind units usually require some scheduleslippage. This indicates that even with problems, the composite parts were lessexpensive and time consuming than metallic parts. The composite partmanufacture was successful.REFERENCES{1} ABB Azipod Project Guide, dated 3 December 1998