LOW POWER STIRLING ENGINE FOR UNDERWATER VEHICLE APPLICATIONS.pdf

LOW POWER STIRLING ENGINE FOR UNDERWATER

VEHICLE APPLICATIONS

Graham T. Reader, Ian J. Potter, Eric J. Clavelle and Owen R.. Fauvel

Department of Mechanical Engineering, The University of Calgary

Calgary, Alberta T2N 1N4, CANADA

readeraenme.ucalgary . ca

Abstract-- The selection and design of a power system for any

form of underwater vehicle is an extremely complex and

difficult task. The system must be capable of providing the

vehicle with the required mission performance in terms of

power and energy and also be volumetrically and

gravimetrically compact.

Navy submarine at the beginning of the 1980s that the

underwater Stirling engine attained technical maturity.

The concept that made the Stirling attractive to navies was the

so-called hybrid or air-independent variant of the conventional

submarine. In the 1970s, and even today, the electro-chemical

batteries used in conventional submarines are the secondary

storage type and invariably employ lead-acid cells. Although

sufficient power can be extracted from such batteries to propel

the submarine at high underwater speeds, their energy storage

capacity is usually too low to enable high speeds to be sustained

for more that a few tens of minutes. Even at low speeds,

normally only tens of hours of submergence can be obtained.

When the vehicle to be used is a newly designed US Navy

Diver Propulsion Vehicle (DPV), other power system

constraints are highlighted. These constraints include limited

vehicle diameter, high performance operation, low power

requirements, safety and a non-magnetic signature. Of the many

power systems available, very few can fulfil the design criteria

for the DPV. One system that can is the hydrocarbon fuelled

Stirling engine - a dynamic heat engine using an external

combustion system.

To improve the “indiscretion rate”, which is a measure of the

time the submarine can remain submerged, of non-nuclear

submarinesmany altemative solutions were investigated and of

these the hybrid concept became the preferred choice. With a

hybrid conventional submarine a secondary power unit is used

to recharge the storage batteries or provide a separate source of

direct power when the vessel is underwater. The former is the

more usual way of using these secondary power systems. Many

Air-Independent Power (AIP) systems have been investigated

and developed [3], but to-date only the Stirling engine system

is in operational submarine service. Countries other than

Sweden have shown an active interest in the use of the Kockum

type Stirling power-pack including, Germany, Australia, Japan,

Norway, Denmark and Malaysia.

This paper describes the application of the Stirling engine for

underwater duties, and in particular the selection, design and

development of a Stirling engine powered DPV. Details are

given of the specialist vehicle requirements, engine selection

and design and the development of a combustion gas

recirculation system to enable pure gaseous oxygen to be used

as the combustion oxidant. In addition, details are given of the

restrictions imposed on component design and manufacture by

the low vehicle power requirements.

I. Introduction

A closed-cycle heat engine which can be made to work

effectively in an airless environment is ideal for use in

underwater power system applications. If such an engine is also

inherently quiet, then it is particularly attractiveto military users

of such systems. The Stirling engine meets all of these

specificationsand it is therefore not surprising that underwater

applications have been proposed for the engine almost from the

time of its re-development in the late 1940’s. In the 1960s and

1970s, significanttechnical progress was made with underwater

Stirling engines for use in submarines and torpedoes during the

Philips-General Motors program [ 11. Other interested parties,

such as the German and Royal Navies also investigated the

concept [1][2]. However, it was not until Kockum (United

Stirling) fitted a Stirlingunit into an operational Royal Swedish

The use of underwater vehicles is not solely restricted to navies.

Over the last thirty years, especially since the discovery of

offshore oil and gas in the European North Sea, such vehicles

have been used, and specifically developed, for industrial,

commercial and scientific activities. Many of these underwater

vessels are tethered, i.e., they receive their power via an

umbilical electric cable attached to a surface or sub-surface ship

or installation. These types of vehicle, which are also used by

the military, can be manned but the majority are operated

without an in-situ crew from a remote location, hence their

description - Remotely Operated Vehicles (ROVs). However,

there are other unmanned underwater vehicles which can

operate autonomously, i.e., they carry their own energy unit.

These autonomous or unmanned underwater vehicles (AUVs

0-7803-4273-9/98/$10.00 @ 1998IEEE

-411-

and UUVs) are often torpedo-like in appearance and size and

can be as large as a small submarine. Most modem AUVs are

alternativepower units are being investigated [4].

affected.

The module would contain the auxiliary power unit(s), control

system and oxygen supply [6]. Depending upon the design

philosophies the pod would also contain the fuel for the engine

and any additional ballasting arrangements generated by the

inclusion of the unit. The type of module used by the Royal

Swedish Navy in their Nacken class submarine of about 1200

tonnes contains two V4-275 Stirling units and all the associated

controls and oxygen supply. The total volume of the module is

about 210 m3 and this allows sufficient energy storage for the

submarineto stay underwater some 2-5 times longer, depending

upon the mission, than with the normal battery pack.

Two additional roles for the Stirling engine that have recently

been identified and are currently being developed is for use in

Diver Propulsion Vehicles (DPVs), Figure 1, and Swimmer

Delivery Vehicles (SDVs).Although these vessels are generally

for military operations, the possible cross-over of the

technology to commercialand possibly sport vehicles must also

be considered.

The Koch engine, a Vee-configuration four cylinder

double-actingengine has been physically described in detail in

the Stirling literature [7]. The combustion system can be

pressurised to about 3 MPa enabling the exhaust to be

discharged without the aid of a compressor down to about a

depth of 300 m. The water vapour in the exhaust is condensed

out and the carbon dioxide removed in a special mixing unit.

The temperature of the heater is kept constant at about 970 K

and the working fluid is helium. The heater head has been

designed for use with liquid diesel fuel. Originally, EXXSOL

D60, a Swedish diesel containing about 1 ppm sulphur was

chosen but now heavy sulphur fuels such as “Citygas” can be

used. Kerosene and other lighter solvent type fuels have also

been tested.

Figure 1: Diver with DPV

The reliability of Stirling’shas, often unfairly, been rancorously

questioned and the underwater Stirling is no exception but in an

operational lifetime the reliability has been between 90% and

100% and the SwedishNavy have been more than satisfied with

this performance [7][8]. A similar system is used in the new

1500 tonnes A-19 GotZand class Swedish submarines, the first

of which was launched in January 1995. Budget restrictions

precluded the installation of a third V4-275 unit which would

have provided sufficient power for higher sustained underwater

speeds without drawing on the battery packs [9]. The next

generation of Swedish boat, Submarine 2000, will also be of a

hybrid configuration and developed in collaboration with the

Norwegian and Danish navies.

11. Manned Submarine Developments

Depending on the hotel load requirement, i.e., all power

demands other than for propulsion, a modem patrol size naval

submarine of between 1200 and 2400 tonnes displacement

would require about 100-250kW of power to travel underwater

a modest cruising speeds of between 3 and 5 knots. For most

submarine surveillance operations these speeds are adequate.

The secondary power unit needed to met these demands is

usually housed in a module within the vessel’s pressure hull.

The module can either be added to an existing submarine (retro-

fitted) by extending its hull length or by designing a new

submarineto accommodate such a module[5]. In both cases, the

maximum extension of the hull length over that normally used

is about 10-15%. Above these figures the vessel’s surface

operating characteristics and stability would be unacceptably

Kawasaki of Japan are currently working with Koskum on the

development of Stirling units for use in future generations of

Japanese submarines [lo]. The Australian Navy also

investigated Stirling technology for use on a modified Collins

class submarine.

The future for the use of Stirling systems in non-nuclear naval

submarines appears bright and although this is a somewhat

limited market it is nevertheless lucrative. The commercial and

tourist submarine market is at the moment too fragile for

Stirling’s or other non-battery power systems to make any

-412-

inroads. The SAGA submarine built for use in the commercial

offshore industry has proved to be a technical success but not

found a market [ 111. Tourist submarines are now in world wide

use but their present operational requirements can be readily

and economically met with battery packs and umbilical feeders.

This situation could change in the future.

and low acoustic signature for use in areas where underwater

ordnance was known or suspected. Mainly due to the low

magnetic signature, the original proposal suggested the use of

a heat engine, such as the Rankine cycle turbine or the Stirling

engine, rather than an electrically powered system, i.e., fuel

cell.

111. Unmanned Underwater Vehicles

The engine was to be capable of repeated starts and stops, and

provide the diver with some means of speed control. For the

combustor, which was to operate in a closed environmentwith

no exposure to air, the suggested system would use a

hydrocarbon fuel and compressed oxygen as the energy source,

although other technologies would be considered.

Unmanned underwater vehicles in the form of self-propelled

torpedoes have been under development for as long as navies

have used the subsea environment. Present heavyweight -

usually submarine launched - torpedoes have speeds between

45 and 60 knots which have to be maintained for about 20

minutes. Although studies have shown that this level of

performance could be achieved -just - by Stirling power units,

until metal combustion systems are perfected, it seems unlikely

that the heavyweight Stirling torpedo will be the focus of much

attention. However, over the last thirty years and particularly

over the last decade there has been a growing interest in and

need for slower longer range untethered and unmanned

underwater vehicles.

A. Initial Design of the Deep Ocean Stirling Engine

In response to the potential sources request, the University of

Calgary was sponsored by Analysis and Technology

Incorporated, Panama City to design and manufacture a Deep

Ocean Stirling Engine (DOSE), comprising a Stirling engine,

combustor and associated control system for the newly

designed DPV. Table 1 lists the initial design requirements for

the non-magnetic DPV which were subsequently used by the

University of Calgary in the initial power system concept

design. In the early stages of the design evaluation, emphasis

was placed on two areas, firstly, establishing a generalised

design and secondly,the use of computer simulationsto provide

concept system and component dimensions. As such, for the

vehicle integration and power system requirements, the

Submersible Power System Selection program (SPSS') was

used to confirm the DPV power requirements and ensure that

the geometric configuration of the vehicle allowed for the

required mission profile. Although these calculations were

based on generic volumetric and gravimetric energy densities

for a Stirling engine system, they provided an initial baseline

for further design analysis and optimisation.

The USA's Defence agency's (DARPA) experimental UUV

vehicle, had an initial requirement for a 10 kW power pack with

an endurance of 336 hours. No suitable power system was

found and the specifications have been radically altered. The

prototype Japanese RI robot had little space left for sensor

payloads, but a second generation design provided better

facilities [ 121. The engine to be used in the R1 was to be a 30

kW(e) four cylinder double acting version using a swash plate

drive. A similar type of vehicle using a Kockum type

Vee-engine of 15 kW was proposed in the late 1980s. Another

Japanese Stirling engine research and development project is

also being undertaken at Mitsubishi Heavy Industries [13].

IV. Diver Propulsion Vehicles

Table 1: DPV Statement of Requirements

A Diver Propulsion Vehicle (DPV) is a craft which propels

divers through the water at speeds the diver could not attain by

swimming. The first DPV was developed, built and used during

World War I by the Italian's Raffaele Rossetti and Raffaele

Paolucci with the intention being to sink the Austrian battleship

Viribus Unitis [14]. This original DPV was powered by

compressed air, however, most subsequent vehicles have been

electrically, or to be more precise, battery powered. These

modern day DPV's are used for sports, professional and

military diving, however, they suffer from limited mission

endurance, low speed and restricted depth capabilities.

Shaft Power (W)

-470

Shall Speed (rpm)

750

Endurance (hours)

Vehicle Size (mm)

$300 x 1830

Weight (kg)

-68

Operating Depth (m)

91.44

Acoustic Signature

Low

Magnetic Signature

Low

Exhaust Gases

Minimal

Refurbishment (hours)

Operating Temperature ("C)

Storage Temperature ("C)

-2 to 32

-40 to 60

Reliability (# of missions)

In an attempt to develop a long endurance deep diving DPV, in

mid 1996 the Coastal Systems Station (CSS) at Panama City

requested potential sources for a kinematic heat engine and

combustor. The vehicle was to have a low magnetic signature

-413-

B. Engine Design

process to be directed onto the engine heater system. Since the

combustionis extemal and separatefrom the working gas in the

Stirling cycle, it is possible to select a suitable over-pressure in

the combustor. If the over-pressure is slightly higher than the

ambient seawater pressure outside the vehicle, the exhaust gas

can be discharged overboard without the use of an exhaust gas

compressor. The exhaust gas from such a system will consist of

s CO,, water vapour and a small quantity of oxygen. In this

project, since the maximum vehicle operating depth is 91.5 m

and the ambient seawater pressure is 10.2 bar, the

recommended combustion system pressure will be 13 bar.

With the wide number and variations of Stirling engines

available, a synthesis approach was required to reduce the

number of options to a single engine configuration. As such, a

hierarchical engineeringdecision tree was developed to specify

the required functions and the successive design selections at

increasing levels of detail. Engine systems and CO i r i

The engine configuration ultimately selected for the DPV was

based on two displacer cylinders sharing a single power

cylinder (double acting Bingham piston), with each displacer

having its own regenerator, cooler and heater bank, Figures 2

and 3. The engine working fluid is helium. Only this

configuration was thought to provide low acoustic signature,

speed control, ease of starting, good power envelope,

portability, reliability and efficiency.

The objectives of the combustion system were identified as,

firstly to design a controllable system that would combust a

variable quantity of hydrocarbon fuel and oxidant, and direct

the exhaust gases over the Stirling Engine heater tubes.

Secondly,the combustor system should provide sufficient heat

to enable the engine to develop a maximum of 470 W shaft

power output.

Theoretical calculationswere carried out to predict the quantity

of kerosene fuel and oxidant required for a given engine power

taking into account combustion stoichiometry and system

efficiencies. In addition, the temperature resulting from the

combustion process was required, together with the effects of

any temperature moderating system that maybe utilised.

With this initial configuration the Stirling engine computer

design simulation, MARWEISS", was used to define and

confm the engine dimensions and parameters. This simulation

program has been used by the authors to give close results to

practical Stirling engine designs, however, as an additional

design confidence procedure, the design criteria were also

integrated into the VARY" simulation program. Although there

were slight differences in the definition of simulation input

parameters, the results from both programs proved very similar.

Regenerator Cooler

@Pressure

Combustion

Figure 3: DOSE Schematic

Figure 2: DOSE Configuration

The fuel supplied to the system was to be in such a condition

that combustion occurs effectively and as efficiently as

possible. The supply and conditioning of the fuel, although

independent processes were viewed interactivelyto provide a

final design solution. To reduce the parasitic power loss on the

engine, a constant pressure, helium filled bladder system was

custom designed to fit within a spherical stainless steel

accumulator (designed to ASME Pressure Vessel Codes,

Section VIII). This svstem was designed to maintain the same

C. Combustion Systems Design

For an underwater Stirling Engine power system using a

hydrocarbon fuel source, the fuel and an oxidant are burnt in an

external combustion chamber mounted on the engine in such a

manner as to allow the heat generated from the combustion

-414-

- -

were assessed based on fundamental engineeringprinciples as

they related to the desired DPV configuration and operational

requirements.

~~~~~~

fuel supply pressure as the fuel quantity in the accumulator

diminished. Although a simple system with no moving parts,

there is space and weight penalty associated with the

accumulator and inert gas pressurising system. The accumulator

was custom design due to unsuccessful attempts at

commercially obtaining an accumulatorthat fitted all the design

criteria, with most failing on magnetics or weight.

was a need to reduce the adiabatic flame temperature from the

high levels associated with the use of pure oxygen by the use of

a thermic ballast for heat absorption. This heat absorption will

initially be provided by the trapped inert gas used in the

combustor purging process. However, the main system for the

heat absorption will be based on a combustion gas recirculation

(CGR) system using the gases trapped within the combustion

chamber. This has a similar effect to that of a separate external

supply of an inert gas. The system is based upon the oxygen

being supplied through an ejector tube driving nozzle housed

within a jet pump, this entrains and mixes with the inert

recirculatedgases in the combustion chamber.An internalback-

flow of combustion gases is therefore created through the

pressure gradient from the oxygen being fed at high pressure

through the nozzle. The resulting combustion atmosphere will

have the same oxygen content as that for air, and hence the

combustiontemperature should be reduced to the normal level.

The extremely low flow rates associated which the fuel system

promoted a variety of alternative design possibilities such as

pulsed jet and bubblejet technology. However, after extensive

evaluation, the simplest, yet most convenient atomiser was

selected - a plain orifice pressure atomizer. However, the low

fuel flow rates necessitated the use of an extremely small sized

laser drilled ruby nozzle. A micro-meteringvalve, then allows

control of the fuel delivery pressure to that required at the

nozzle to ensure atomisation and the correct fuel flow for a

particular engine power condition.

To initially start the combustion, some form of ignition device

is required. The non-magneticrestrictions did not allow the use

of energised coils as would be found in a normal ignition

system, when coupled with the desire for continuous ignition

for a timed period, the only real solution was a specially

designed electric ignitor.

To combust a hydrocarbon based fuel, oxygen is required. In

normal atmospheric operation air supplies this oxygen. The

nitrogen present in the air, although not playing a physical role

in the combustion mechanism, acts as a temperature control

mechanism. This limits the combustion temperature to that

capable of being handled by the engine/combustion system

materials. In the underwater role, other factors will effect the

selection of the oxidant, such as the ability of the

enginelcombustion system to operate with non-air working

fluids and the volumetric/gravimetricstorage densities of the

working fluids and system storage containers.

D. Control System Design

Several control options or techniques were considered for the

power system, including mechanical, electronic and

mechatronic. After extensive evaluation and brainstorming the

mechatronic option was selected as being able to offer the

simplest and safest system. The main element of the control

system was for the diver to be able to vary the engine speed.

This is achieved by a ganged mechanical cable arrangement

linked to the stroke and the fueVoxidant metering valves. In

addition, a computer-basedsystem:

i. interacts with the mechanical system for starting and

If only oxygen were to be used as the working fluid,

excessively high combustiontemperatures are generated. This

is well above the combustion system materials temperature

limitations and a thermic ballast is needed for the additional

heat absorption.

At the pre-contract meeting with CSS, it was clearly stressed

that GOX was the preferred oxidant choice based on logistics

of supply and storage system requirements. In the resulting

oxygen system design, the oxygen is supplied at the correct

flow rate for a particular engine load by the use of a micro-

metering valve and a choked nozzle. The actual oxygen

requirements are based on stoichiometricoxygen conditions. As

a precaution, an 10% oxygen has been added to the

stoichiometricto ensure complete combustion of the kerosine.

ii. provides an emergency shut-down if component failure or

over-temperatureoccurs.

V. Problems Overcome in the Design Process

Whilst several items were commercially procured in accordance

with the commercial-off-the-shelf(COTS) philosophy used at

the University for underwater vehicle design, the custom nature

of the DOSE design has resulted in the main part of the

manufacture has occurred in the University workshops.

The adiabatic flame temperature for the combustion of kerosine

with variations in air at stoichiometriclevels is 2463 K. If pure

oxygen at the stoichiometriclevel was used the adiabatic flame

temperature was calculated to be 6396 K. If 10% excess oxygen

is used, this temperature is reduced to 6086 K.

The main difficultiesfaced during the design and manufacture

process have been focussed in two areas:

i. The extremely low power demands for the system.

ii. The requirement for nonmagnetic materials in the system

construction.

To keep within the metallurgicallimits of the heater tubes there

-415-

stopping.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: