Skip to main content
European Commission logo print header

High efficiency hybrid drive trains for small and medium sized marine craft

Final Report Summary - HYMAR (High efficiency hybrid drive trains for small and medium sized marine craft)

Executive summary:

the HYMAR project had a number of numerical objectives to meet as part of the project outcomes. These included zero emissions to air and zero external noise and vibration in port and a reduction of overall fuel consumption by 30%, tending to greater than90% on applications such as long distance sailing boats using regenerative techniques. A new ISO standard for high voltage DC propulsion systems and comprehensive energy management were also targets.

for a number of technical reasons outlined in the original proposal, it was anticipated that these objectives would be met through an optimized serial hybrid platform (i.e. the internal combustion engine in the system solely powers a generator and is not mechanically connected to the propeller shaft). In the event, the more research that was done, and the more data that was collected, the greater the challenges to the underlying premises.

analysis based on the extensive data collected demonstrated that other than in niche applications the project objectives would be best met through an optimized parallel hybrid platform (i.e. the engine is mechanically connected as is an electric 'machine' that can operate as both a propulsion motor and a generator). A project extension was applied for and granted to explore the parallel option.

the data from testing conventional, serial, and parallel systems demonstrated that an optimized parallel system could deliver the core project objectives of:

- Zero emissions to air and zero external noise and vibration in port
- Radical reductions in fuel consumption for the generation of 'house' power with substantial reductions in propulsion fuel consumption below a defined 'cross-over' speed and modest reductions above this speed for a net gain of better than 30% in many applications (depending on the duty cycle) tending to greater than90% on applications such as long distance sailing boats using regenerative techniques
- CO² reduction of greater than30% in all "off design point" applications (e.g. fishing boats when trolling, canal boats, and small commercial ferries operating at low average loads)

extensive generator, electric motor and propeller work resulted in generators and electric motors with high levels of efficiency at the loads experienced in a hybrid system. Mechanisms for dynamic control of generators and electric motors (in both propulsion and regeneration modes), and for matching electric motors to propellers over a wide range of operating conditions, have been refined. A new self-pitching propeller has been designed and built with significantly modified thrust characteristics that facilitate the matching of electric motors and propellers for optimized propulsion and regeneration performance.

A comprehensive energy management module and graphical user interface has been built. This controls whole-boat energy flows. Modified lead-acid battery chemistries have been developed with enhanced performance in hybrid applications. Battery management strategies have been devised that optimize cycle life without necessitating inefficient engine use in the challenging conditions of partial state of charge operation.

HYMAR members have helped to expand the message base of the NMEA 2000 networking protocols to include messages necessary for hybrid systems. The HYMAR project has taken a lead role in developing a new ISO safety standard that addresses DC propulsion systems above 50 volts. This is now at the Draft International Standard (DIS) level.

project Context and Objectives:
the European recreational marine industry is large and brings significant economic and social benefit. Key statistics are:
272,000 people are employed by 37,200 companies
23.4 EUROS Bn of revenue, equating to 0.14% of the EU's GDP
5.99 million boats are owned within the EU and some 32 million people per annum participate in leisure boating

in addition there is significant induced economic activity from other retail service providers such as shops, hotels and restaurants as a result of people using the water. In the UK alone, provisional figures indicate that day visits by people involved in leisure activities on the water generate a further 2.0 Bn pounds of retail spend.

environmentally, leisure boats are frequently operated in areas of high environmental sensitivity and are designed and regulated accordingly. Recreational craft engines are subject to stringent emissions regulations for gases, particulates and sound (Directive 2003/44/EC) and further legislation is planned.

notwithstanding legislative pressures, the European Union recreational marine industry is acutely aware of its responsibility to lead the drive to create an industry which is sustainable in the long term and which constantly drives down its environmental impact and its carbon emissions. As an active member of the Waterborne Technology Platform, ICOMIA has developed a comprehensive technology route map which is used as a tool to position its long term research within the broader maritime community.

the route map covers all of the R&D activities required to ensure the long term future of the industry and covers the key customer and regulatory ambitions for the next 10 years:
Low through life costs
Greater comfort and convenience
Fast attractive and versatile

the most extensive area of work required is in the "greener" category and it is here that the issue of electric hybrid propulsion appears.

these requirements are also reflected in the Waterborne Implementation Route Map (see online) where development of the sustainable recreational craft is identified as one of the primary objectives . In order to achieve this long term objective a number of intermediate steps are required, one of which is development of the next generation of power and propulsion concepts .

recreational and light commercial vessels universally employ internal combustion engines for propulsion purposes, and to generate most of the non-propulsion 'house' energy that is used on board. Almost all vessels are driven by propellers (the exception being a small number of boats driven by jet drives).

for any given engine, a fuel map defines the quantity of fuel it takes to create a unit of output energy at any particular engine speed up to the engine's full rated speed, and at any particular power level up to the engine's rated power at any given speed. The fuel consumption is typically expressed in terms of grams of fuel burned per kilowatt-hour (g/kWh) of output energy (measured at the engine's flywheel or output shaft). This is known as specific fuel consumption (SFC). Note that the full power curve that is displayed on a fuel map is almost always convex.

the manner in which a propeller absorbs energy and transmits it to the water can be described by a propeller curve which plots the energy absorbed against the propeller's speed of rotation. The resulting curve is almost always concave. If plotted on an engine fuel map, the propeller curve can never be made to follow the engine's full power curve. As a result, propellers are typically sized such that the propeller curve crosses the engine's full power curve at, or close to, the engine's full rated speed.

given a propeller matched to an engine in the conventional fashion just described, and plotted on that engine's fuel map, the propeller curve rarely, if ever, passes through the region of peak efficiency on the fuel map. At lower engine and boat speeds, the deviation from peak efficiency is dramatic. In general, the propeller curve approaches closest to peak efficiency at higher boat speeds and loads.

at best, marine engines are operating sub-optimally. In many duty cycles, especially those that involve extended dockside running time, harbour manoeuvring, and low speed propulsion (such as operating in canals and other restricted waterways or fishing boats in trolling mode) they spend much of their time at far below optimal efficiency levels. This creates the opportunity for significant reductions in carbon emissions, other emissions and engine run times. It defines the core context in which the HYMAR project was developed.

the relationship between a propeller curve and peak engine efficiency can be expressed in terms of a graph which displays SFC consumption on one axis and boat speed on the other. It will be found that at slow boat speeds the SFC is high (i.e. engine operation is extremely inefficient) with the SFC trending downwards (i.e. improving efficiency) as boat speeds (and propeller loads) increase, and then trending upwards again above a certain boat speed.

hybrid propulsion systems can be defined as serial – in which an engine drives a generator which powers a propeller shaft via an electric propulsion motor, with no mechanical connection between the engine and the propeller shaft – and parallel – in which the engine is mechanically connected to the propeller shaft as is also an electric motor such that both can drive the shaft together or separately, and the electric motor can also be driven by the engine as a generator.

in the serial case, the output of the generator can be fed directly to the electric motor ('diesel-electric' mode) or can be stored in batteries and subsequently used to power the electric motor. In the parallel case, the electric motor can only run off stored energy. In either case the system can, in theory, be designed such that when the engine in the system is creating electrical energy it is only run at peak efficiency, and if it cannot be run at peak efficiency it is not run at all. If the engine in the parallel system is being used for propulsion purposes, it can simultaneously be used to power the electric motor as a generator, with the generator load manipulated to maintain the engine at peak efficiency for whatever speed is demanded for propulsion purposes. In other words, the engine always runs at, or close to, peak efficiency.

if the peak engine efficiency is plotted on the graph of SFC versus boat speed for a given propeller curve, from a theoretical perspective the hybrid systems capture all the efficiency gains in the area of the graph beneath the propeller curve down to the line representing peak engine efficiency.

these theoretical gains have to be qualified by (i) the electrical losses in the generator that arise when converting mechanical energy into electrical energy, (ii) the losses in the electric motor that arise when converting that electrical energy back into mechanical energy, and (iii) the losses incurred in charging and discharging batteries whenever the electrical energy is stored in a battery before being used. If these losses are factored into the equation, typically a hybrid system is found to be more efficient than a conventional system at low boat speeds, with the conventional system more efficient at high boat speeds. The HYMAR project defined the transition point from superior hybrid efficiency to superior conventional efficiency as the 'cross-over' point or speed.

set against this, the batteries provide a mechanism for storing electrical energy from other sources ('energy displacement') such as shore power, solar, wind, fuel cells, and regeneration from a freewheeling propeller on a sailboat under sail. All such energy can reduce overall fossil fuel consumption (depending on the source of the energy) and average SFC levels. The greater the non-engine contribution to propulsion needs, the higher the average 'cross-over' speed.

the batteries are a critical component in any hybrid system, perhaps the most critical component. In both serial and parallel systems, they determine range and performance under electric power when no engine is running; they provide an essential energy 'buffer' at various points in the duty cycle; and their charge and discharge efficiencies play a significant role in the overall efficiency equation. With respect to batteries, primarily for reasons of cost, the HYMAR project focused on EnerSys Thin Plate Pure Lead (TPPL) batteries.

the HYMAR project was designed to measure the variables in both conventional and serial and parallel systems, to quantify the increased levels of efficiency that could be achieved in the hybrid systems, and to develop an optimized hybrid system that would be close to being market ready, including developing a new generation of TPPL batteries. Additional work was done on exploring mechanisms to change the shape of propeller curves to better match propellers to engine and electric motor efficiency maps, and on developing certain power electronics components needed by a hybrid system (battery chargers, inverters, DC-to-DC converters, monitoring devices, etc.).

based on preliminary studies, the principal HYMAR objectives were defined as:
- Fully integrated marine hybrid drive system for commercial and recreational craft up to 24m, but scalable to larger vessels
- Zero emissions to air and zero discernible noise and vibration in harbour
- An increase of 50% in the lifetime kilowatt-hour performance of lead-acid batteries in marine hybrid applications and an investigation into other appropriate energy storage technologies including lithium-ion
- Full compliance with European Commission Directive 2003/10/EC (Merchant Shipping and Fishing Vessels Control of Noise at Work Regulations 2007) which came into force on 23 February 2008
- Propeller efficiency increased by 5% at full load and greater than15% at "off design point" operation
- Reduction of overall fuel consumption by 20%, tending to greater than90% on long distance sailing boats
- CO2 reduction of 20% in all "off design point" applications such as fishing boats, pilot boats and small commercial ferries"

in order to achieve these objectives, certain other sub-objectives were defined, notably:
- Development of generators with an electrical efficiency of 90% or better across the operating range for the HYMAR hybrid system
- Development of electric motors and controllers with a combined efficiency of 90% or better across the operating range for the HYMAR hybrid system
- Development of dynamic control mechanisms for the generators and electric motors
- Design and build various high efficiency power electronic components (inverters, chargers, DC-to-DC converters)
- Create an integrated management and control system (the HYMAR system Energy Management Module – EMM)
- Integration of the control system into the NMEA 2000 communications protocol, with the development of any new PGNs that this necessitated
- Create a new international (ISO) safety standard for DC systems above 50 volts
- Develop a load-following, self-pitching propeller

it was theorized that these objectives could best be achieved in the context of a serial hybrid architecture. In the event, the more data that was collected and evaluated, the clearer it became that serial systems are suitable for niche applications whereas parallel better fit the overall goals of the HYMAR project. The project was modified to focus on parallel systems and an extension applied for and granted to enable the work on this new focus to be completed.

the detailed results are described in the main scientific and technical section of this report, and in the attached 'Final Report' PDF.

project Results:
in-the-water data collection
in order to assess the efficiency benefits of various hybrid options, and to determine how best to optimize these options, the HYMAR project required detailed knowledge of the efficiency of conventional propulsion systems. A Malo 46 sailboat was outfitted with extensive instrumentation and a custom-built datalogger (from TML) that was continuously refined throughout the project. Fuel maps for the engine in the test boat were developed by Steyr for different test conditions. The instrumentation was extensively checked for accuracy and calibration. Bruntons Propellers developed theoretical propeller efficiency and performance data for various propellers and reduction gears.
A far broader range of baseline trials was conducted in the trials boat than was originally intended, using a number of different propellers in widely varying wind and sea states, resulting in what has to be one of the world's best databases on the performance of conventional propulsion systems in recreational and light commercial displacement vessels.

based on the baseline data, and on a review of existing brushless permanent magnet DC (PMDC) generators and electric motors (together with their motor controllers), the HYMAR project defined electrical efficiency targets of 90% over the necessary operating range for generators and electric motors in serial hybrid systems, and mechanical efficiency targets for the engines driving generators of peak fuel efficiency +5%, as measured by specific fuel consumption (SFC). Based on data from the manufacturer, the efficiency of the TPPL batteries used in the project was set at 87%.

these efficiency factors were used to define the theoretical window of opportunity of an optimised serial hybrid system and the 'cross-over' speed below which a hybrid is more efficient and above which a conventional system is more efficient. A range of generators and electric motors were tested in a serial configuration, generating a mass of highly detailed data from which propulsion efficiencies could be determined for any given vessel operating profile. So far as is known, this is the only database of this kind in the world. This testing demonstrated the difficulty inherent in achieving the target hybrid efficiencies.
an ancillary result of the baseline data collection was to gather information on the extreme inefficiencies of the three principal conventional methods of generating 'house' power on boats while at sea, which are using an engine-driven alternator either underway or at anchor, and running an AC generator. These inefficiencies derive from the low efficiency of traditional alternators (typically, the peak electrical efficiency is not much above 50% and much of the time it is well below this) and AC generators (typically, the peak electrical efficiency is between 70% and 80%, while most of the time it is well below this), coupled to the fact that normal operation is at well off the point of peak fuel efficiency for the engine driving the alternator or generator.
generator development

the project developed specifications for an optimized brushless PMDC generator and a dynamic controller, designed to meet the targets defined above over the full power and voltage ranges specified for a nominal 144 volt serial hybrid system. A prototype generator was built by Homewood Products Corporation and a controller by ESP. Both were tested by HYMAR partner, Steyr Motors. The generator achieved the design target of 90% or greater electrical efficiency over a broad operating range, and also met the efficiency target of operating within 5% of peak fuel efficiency for the Steyr Motors engine driving it. However, the steps necessary to make this possible more-or-less made the generator unmarketable (see below).

ESP conducted a considerable amount of additional PMDC generator evaluation in the test boat with different stator windings and at different voltages and engine rpm, none of which achieved the overall fuel efficiency target. It is clear that this target can be met, but only by running an engine at a relatively slow speed and a relatively high load for this speed. The net result is to reduce the peak power available from the engine, driving up the weight and purchase price per kilowatt (kW) of rated output, while at the same time holding the engine at high average torque levels, which can reduce engine operating life (primarily due to bearing stress and high exhaust temperatures).

marketing (i.e. cost, size and weight per kW of rated output) and other considerations make it unlikely that the HYMAR target will be met with the relatively small diesel engines used in recreational and light commercial serial hybrid systems, and as such the target SFC should be revised upwards. It should be noted that the target is easier to achieve in larger systems (and in fact there are a number of claims of better than 90% generator electrical efficiency with engine SFC numbers at, or below, the HYMAR target).

fuel efficiency versus cost of power
the baseline and generator data enabled another concept developed during the early days of the HYMAR project to be quantified in detail. This is the 'real' cost of power. This is a calculation that includes the amortized cost of operating an engine plus the maintenance cost, as well as the fuel cost, from which a total kilowatt-hour (kWh) cost is derived. It should be noted that at low speeds and power levels, the fuel cost is typically just a fraction of the total kWh cost.

one of the key benefits of an optimized hybrid system is to eliminate all low speed/low load engine operation. The net result is to consolidate several hours of conventional engine run time into an hour of hybrid generator run time. On the HYMAR test boat, for example, it was possible to consolidate up to six hours of harbour manoeuvring time into an hour of generator run time. The average load on the engine is several times higher than in the conventional system, and as a result the amortization cost per kWh is several times lower (the amortization cost is directly proportional to the average engine load). Given such a consolidation of engine run hours, even if there is no net gain in terms of fuel efficiency, there is a major gain in terms of reduced cost of power. There are additional substantial benefits in terms of the potential for pollution-free operation in harbours and other sensitive areas, and the lifestyle benefits of substantially reduced overall engine run time.

note that in order to derive kWh cost numbers, the project made generalized assumptions about the lifetime operating hours of engines and used this as the basis for calculating amortized kWh costs. In practice, lifetime operating hours will be a function of load. In order to develop accurate amortized kWh cost numbers a map is needed, analogous to a fuel map, which gives the lifetime operating hours over the full range of engine rpm and loads. From this, a lifetime kWh map can be derived, giving the total kWh an engine will deliver over its lifetime at any speed and load. This, in turn, would enable accurate kWh amortization costs to be calculated at any speed and power level.

so far as ESP knows, the data for a lifetime kWh map has never been developed. It would be extremely expensive to generate with conventional (mechanical fuel injection) diesel engines, but with the increasingly widespread adoption of electronically-controlled fuel injection, which, on many engines, includes extensive data logging capabilities, the necessary data should become available over time. Collecting this data would be an extremely useful future research project. Note that it is unlikely that peak kWh delivery will coincide with peak fuel efficiency. If the two differ, a further calculation that includes fuel costs will enable the engine speed and power levels to be calculated for the minimum cost of power at any particular fuel price.

'Electric machines'
the HYMAR project began with a number of hypotheses - primarily based on David Tether's prior research, experimentation and significant experience - concerning PMDC 'electric machines' (devices which can be used for both propulsion and generation) used for marine propulsion (i.e. to turn propellers in a variety of conditions) and generators (including 'regeneration' under sail). Tether advocated the development of high torque, low speed 'pancake' style (large diameter, short rotor length) motors driven by relatively unsophisticated motor controllers. It was hypothesized that given the high torque of PMDC motors it would be possible to drive an 'oversized' propeller, achieving significant efficiency gains at the propeller (in displacement craft, larger propellers are inherently more efficient than smaller propellers).

the project carried out extensive testing in both propulsion and regeneration modes of three different PMDC motors and their associated motor controllers: a pancake style, slow-speed, high-torque, motor built specifically to Tether's design; a 'long shaft', smaller diameter, slow-speed, high-torque motor built in collaboration with Tether; and an extremely compact high-speed, low-torque motor sourced from Electric Yacht, another marine hybrid company.

aside from evaluating the different design and construction approaches, the objectives were to:
1. Achieve 90% or better total electrical efficiency (electric machine plus motor controller) over the necessary operating range for a serial hybrid system;
2. Enable and optimize regeneration under sail;
3. Produce a dynamic means of controlling the motor in order to optimize the interaction with the propeller attached to it.

the high-speed, low-torque electric machine was found to be just as effective and efficient (via a reduction gear) in moving the boat in a wide range of conditions as the low-speed, high-torque pancake-style machine. Given the huge price differential between the large, low-speed machine and the compact high-speed machine, it was concluded that there is no justification for the added expense of the low-speed machine. What is more, there is a wide range of extremely powerful and cost-effective high-speed machines coming out of the automotive industry which can be adapted for propulsion purposes. This is almost certainly the way forward.

the project tested the hypothesis concerning the efficiency benefits of an 'oversized' propeller. It was found that the problem with such an approach is it drives the electric machine to full rated torque before the machine develops full rated power. Although electric machines (unlike combustion engines) can be driven beyond full rated torque (sometimes up to twice rated torque) for limited periods of time (the limiting factor is the temperature in the stator windings and the motor controller) a system cannot be designed on the assumption that the extra torque and power will be available for extended periods of time. As such, the design point in a serial hybrid system still has to be full rated power, which results in 'normal' propeller sizing procedures.

however, the high torque of electric machines does create opportunities to use higher-than-normal reduction gears, with slower shaft speeds. If this is done, 'normal' propeller sizing procedures will result in larger diameter (and more efficient) propellers than can typically be used with an internal combustion engine, delivering significant efficiency benefits.

in terms of electric machine efficiency, during testing it became clear that while 90% electric machine + motor controller efficiency is achievable, it cannot be sustained to relatively low motor and propulsion speeds, and in fact below a certain threshold electric machine efficiency falls off quite rapidly. This presents a substantial problem for serial hybrid systems in as much as large and powerful electric machines are needed to handle peak propulsion loads, but the more powerful the machine the higher the speed and power threshold below which efficiency begins to suffer and, as a result, the higher the boat speed below which efficiency begins to suffer. Given that much of the logic supporting hybrid systems is based on the substantial loss of efficiency of a conventional system at low speeds and loads, this similar electric machine loss of efficiency at low speeds and loads cuts directly into the rationale for hybrid systems.

the project conducted an experiment in which two smaller electric motors were connected to the propeller shaft with the ability to run either one independent of the other, or as a synchronized pair. At low propulsion loads, only one motor was run, effectively doubling the motor load over that when operated as a synchronized pair. This lowered the boat speed threshold below which the motors became inefficient, improving overall efficiency. Such an approach (parallel electric motors) has the benefit of increasing system redundancy, in as much as each motor installation is separate, at the expense of increased cost and complexity.

dynamic motor controllers
ESP conducted extensive experiments to determine mechanisms for interacting with motors and controllers in order to dynamically manipulate electric machine performance. The goal was to optimize the interaction with propellers in order to achieve maximum propulsion and regeneration system efficiency.

in general, given a constant power supply (which will not be the case in practice) the PMDC machines under investigation have a constant torque up to a certain speed, which defines their rated power, at which point the torque declines proportionately to increases in rpm for a more-or-less constant power, ultimately reaching an rpm limit.

the torque of a PMDC motor is proportional to the current (amps) passing through the motor (this defines the 'torque constant'), and the speed is proportional to voltage (this defines the 'voltage constant'). These relationships are fixed in the construction of the motor. However, to some extent they can be manipulated by changing the 'torque constant' within the motor controller. Lowering the torque constant lowers the torque per amp and raises the rpm per volt, enabling control to be exercised in the region of constant power.

in marine hybrid applications, if the propeller driven by an electric motor is oversized, the motor will hit its torque rating before it reaches full power. Lowering the torque constant will simply further reduce the torque and power. On the other hand, if the propeller is undersized, the motor will hit its rpm limit before it reaches full power. Lowering the torque constant will raise the rpm limit, enabling more power to be pulled from the motor. At some point, if the system is properly sized, the torque, rpm and propeller curves will all intersect in the electric motor's region of constant power. In order to exploit the full power of the electric motor, the system must be sized to achieve this intersection. It is a unique point for any given system.

unfortunately, a propeller curve varies according to such things as vessel loading, wind and sea states, and fouling of the vessel and propeller, and the available power of a PMDC motor will vary with changing bus voltages. In other words, the intersection point of the propeller curve and region of constant power is dynamically changing. ESP conducted multiple experiments to determine the relationship between torque, voltage, motor power and propeller curves. A unique database has been developed. ESP successfully devised a strategy to match electric motors and propellers over a wide range of conditions through manipulation of the torque constant.

in order to use the torque constant for control purposes under the dyunamic conditions found at sea, the overall system controller (HYMAR's Energy Management Module, or EMM) needs the appropriate access to the motor controller. ESP worked with TML to develop the necessary interface with the EMM. It was found that manipulation of the interface between the electric machine and the propeller via the torque constant is most effective if the propeller is nominally 'undersized', as opposed to the original HYMAR assumption that it would be possible to achieve higher propeller efficiencies by 'oversizing' the propeller (see above). 'Undersizing' has the effect of allowing for higher than normal propeller loads and/or lower than normal bus voltages while still enabling a motor and propeller curve intersection to occur in the region of constant power.

in this respect, there are significant differences between serial and parallel systems. The electric motor in a serial system is directly matched to the propeller as described. In a parallel system the propeller will be matched to the maximum power of the engine in the conventional manner and not to the electric machine. Electric propulsion will only be used at relatively low boat speeds, which require little power. The electric motor will almost always have more torque than is needed at these speeds, and as such could drive a larger (and more efficient) propeller than can be driven by the conventional system. In the event a controllable pitch propeller is installed, the pitch can be cranked up for more efficient low speed operation.

without the controllable pitch propeller, the electric motor will typically reach its rpm limit before it reaches its torque limit, in which case its full power will not be available to the system (and propulsion speeds under electric power will be less than might be expected). In such a situation, manipulation of the torque constant will lower the motor's torque limit and proportionately raise the motor's rpm limit until the propeller load reaches the motor's rated power, at which point any further reduction in the torque limit will cause a loss of speed and power.

propeller research
in conventional systems the propeller curve of a 'matched' propeller results in a concave propeller curve with low propeller loads at low shaft rpm. These loads are well below the available power from the engine and poorly matched to engine fuel maps.

regeneration under sail
the project explored the potential for regeneration under sail, and management strategies to control and optimize regeneration, using all three electric machines and several different propellers (including the INSEAN propellers). Contrary to expectations, it was found that regeneration had little effect on boat speed. This is explained by the fact that regeneration rates are a function of boat speed. Given a displacement hull, as boat speed increases so too does the boat's hull resistance (drag).

battery testing
one of the key technological developments that make marine hybrid systems potentially viable is the development of batteries with high charge acceptance rates (CAR). A high CAR allows a battery pack to be used as an energy sink.L
battery management
it was determined through the EnerSys testing that to maintain battery cycle life and health, the state of charge (SoC) of the batteries should not be allowed to fall below 60% DoD. If the SoC nears this level, and if no other charging sources are available, the generator must be started. Initially, battery charge acceptance rates (CAR) are high, enabling a generator to be run at peak efficiency, but once the batteries reach their 'absorption' voltage the battery CAR progressively declines.

ISO standard
as noted above, a serial hybrid system has to have sufficient electric propulsion power to deal with the most adverse conditions anticipated. This requires powerful electric motors. In order to keep cable and other component sizes down to acceptable levels, the voltage on the central DC bus must be raised above voltages typically seen on boats.

communications protocols
A goal of the HYMAR system was to create a generic communications system and set of messages based on the NMEA 2000 protocol, which is becoming a marine industry standard replacing the long-standing NMEA 0183 protocol. The HYMAR concept was that communications standardization based on NMEA 2000 would enable hybrid system integrators to select equipment - such as generators, electric machines, battery chargers, DC-to-AC inverters, and DC-to-DC converters – from any NMEA 2000 compliant supplier, maximizing equipment supply choices.

power electronics
first Victron, and then Mastervolt, were tasked with developing various power electronics equipment for the HYMAR project, integrated via the NMEA 2000 network.

block level sensing devices
as the HYMAR project progressed it became increasingly clear that managing the battery charge and discharge cycles was a major task that would be critical to the end goals of the project. It was found that under the high discharge and recharge currents of the HYMAR system, major differences in individual block temperatures and voltages could occur, with potentially unsafe results.

energy Management Module (EMM)
at the beginning of the HYMAR project, ESP, along with the Bosch Engineering Group (BEG), was given responsibility for developing the HYMAR system Energy Management Module (EMM). ESP developed a higher level design document for the EMM and a definition of the user interface and functional algorithms as the starting point for detailed development of the algorithms in collaboration with BEG, at which point BEG withdrew from the project and EMM development was re-assigned to TML.

starting from the early definitions, TML implemented the user interface using its 'Jason' EMM hardware as the data source. The control system hardware does not aim to replace the control systems integrated into each device – the engine has its own ECU controller, as does the electrical machine and each major generator or consumer of energy. The EMM hardware sits in between every device and translates each unique communications language into a common format, processes this data to allow decisions to be made, then acts accordingly by informing the user or sending the correct control information to the correct device.

the first test boat had a very basic control strategy, mostly enforced by the fact that the data collection and systems monitoring at that point was basic. Battery bank levels and basic engine parameters were gathered, and this allowed the generator to be started and stopped to keep the batteries within certain ranges to maximize performance and cycle life.

this was then expanded to include a degree of user input through the use of profiles. For example, the 'silent' profile would use electric propulsion unless the battery levels dropped dangerously low, whereas the 'speed' profile would keep the generator running most of the time to give greater available electrical power at the expense of fuel consumption, engine run hours and noise. A series of more balanced profiles were also designed for more general use, maximizing such values as fuel efficiency or battery cycle life.

an initial set of control algorithms was produced for a serial hybrid system and early trials carried out in September 2011 in Sweden. More work was needed in order to refine the logic, including an accurate definition of the battery state of charge, battery capacity and optimum charge cycle, and to define more accurately the trigger levels for certain actions. This was continued over the winter of 2011/12 and on the test boat in the UK in 2012.

version 1 of the user interface took the data over a wireless link into a tablet type computer running the 'Android' operating system developed by Google. This was selected because it is now common enough that the hardware is reliable, inexpensive and readily available, providing a full colour touch-screen display and relatively high powered processing capability. An internal battery means the interface can be used anywhere on the boat as required.

version 2: A revised version of the software was developed and demonstrated at the METS trade show in November 2011 using a tablet PC as the main source of data processing and once again using the Android operating system as the user interface. This version was designed to have a more 'commercial' appearance, as well as greatly increased speed of screen redrawing, adding a more fluid feel to the software. The main data points were broken down into pages, to cut down on the information displayed at one time. A commercial embedded touchscreen computer was modified to take the data input and a panel-mount for it designed. The intention was to run the decisions and interface on this module, and simply build basic interfaces to each device.

version 2 included a GPS loopback module, allowing real-time GPS position and speed to be fed into any other software on the tablet. This, for example, gave a real time boat position on the Google Maps application, or on any installed charting software. This new code gave any application on the Android direct access to the boat navigational information.

as the project continued, limitations in this plan started to become apparent. The interface had to be powered up continuously, and boat management was reliant on a hardware device that was bought in and not customisable for HYMAR purposes.

version 3: For the next stage of the interface development it was decided to move the critical management decisions to a custom piece of hardware that could remain powered up but consume minimal power, and which would be independent enough and robust enough to keep the major systems running without the full user interface. This is the device that became 'Jason'. It acts as the gateway between the various devices on the boat, converting everything into one coherent data stream, acting on the major data points according to the selected algorithms, and then passing the data on to the full user interface for display.

A shift to parallel systems
at the completion of the third year in-the-water trials, the project engaged in a major re-evaluation of all the data collected to that point, especially including the 2011 electric machine data. The conclusion of this was that the initial HYMAR focus on serial systems was misplaced, and that parallel systems would better serve the efficiency goals of the project and much of the marketplace.

key factors leading to this conclusion were:
1. The inefficiency of electric machines at low loads and speeds, which has a far more pronounced effect in serial systems than in parallel systems because of the need for much more powerful propulsion motors in the serial system.
2. The failure of generators to meet the SFC goals of the project. Once again, this cuts into the presumed efficiency gains of hybrid systems, in particular undercutting the rationale for serial systems which rely on the generator for energy production at all speeds.
3. The test program demonstrated that the three key benefits of a hybrid system are the ability to create house energy far more efficiently than a conventional approach; the reduction in engine run hours, and cost of power, through consolidation of conventional light-load engine run hours into briefer periods of higher load generator engine run time, making possible pollution-free harbours and other sensitive areas; and the platform that is created for integrating other energy sources, notably shore power, solar, wind and regenerative energy. All three benefits can be as readily achieved in a parallel system as in a serial system.
4. The serial system requires enormous reserves of battery power for even a relatively modest electric propulsion capability.
5. If the torque constant is to be used to manipulate electric machine performance, the propeller in a serial system must be matched to the electric machine's full power rating, or else be undersized. In a parallel system, the electric machine will almost always have more torque than necessary for the low propulsion speeds at which it is used, and as such could drive a larger, more efficient propeller.
6. A parallel electric machine with a power rating as low as 5 kW will still be able to recapture all available regenerative energy up to sailing speeds of 15 knots and higher.
7. The reduced electric propulsion and generator needs of the parallel system can be met with lower DC system voltages. Systems can be engineered for the vast majority of recreational and light commercial applications that fall within the confines of the existing ISO standards, and which are far less demanding in terms of safety requirements than higher voltage systems. The lower voltage substantially reduces the number of battery cells in series, which significantly reduces battery management issues.
8. The reduced electric propulsion, generator and battery demands of a parallel system substantially reduce the cost as compared to a serial system.
9. Parallel systems are more likely to win widespread market acceptance than serial because of a perceived greater reliability in as much as the trusted diesel engine is still connected to the propeller shaft with the electric propulsion adding a redundant system.

optimized parallel systems
following the shift in focus to parallel systems, the existing extensive test data was used to develop an operating profile for an optimized parallel system. This has the following core components:
1. When in generator-only mode, the electric machine must be powerful enough to load the engine to the point of peak fuel efficiency (lowest SFC). The combined system load and battery charge acceptance rate (CAR) must be high enough to keep the generator engine operating at this load. The CAR will define the minimum battery pack size (which will vary according to battery technologies, and also the CAR at different states of charge).
2. Any electric machine rated as in (1) above will be more than powerful enough to have the necessary propulsion capabilities. The electric machine's efficiency curve needs to be such that the machine is out of the low-speed, low-efficiency part of its operating range by the time the boat is up to harbour manoeuvring speeds.
3. If the engine is being used for low speed propulsion (for example, when the batteries are too discharged to use electric propulsion) the EMM needs to manipulate the generator load such that the combined propulsion and generator load holds the engine at its peak efficiency for the speed at which the engine is operating. Both propulsion and house loads will then be met at peak available efficiency, with a dramatic improvement in the efficiency of both as compared to a conventional system.
4. If the engine is being used for higher-speed propulsion, once again the EMM needs to manipulate the generator load such that the combined propulsion and generator load holds the engine at its peak efficiency for the speed at which it is operating. In practice, in most systems this will require the generator to be operated at, or close to, peak output at the cross-over point from electric propulsion to diesel propulsion, with the load being tapered off as the propulsion load increases until a second cross-over point is reached, after which peak efficiency will be maintained by boosting the engine-driven propulsion with electric propulsion.
5. If (3) and (4) are implemented, the overall system efficiency (propulsion power + house power) will be higher at almost all speeds than a serial system, and at all speeds than a conventional system. The principal limitation will be the inability to maintain generator output at desired levels during extended periods of engine run time as battery packs approach a fully-charged state and the CAR begins to decline.
6. Such an operating profile requires the electric machine to perform over a relatively narrow voltage range but a wide speed range. Given that voltage and rpm are directly connected in a brushless PMDC machine, this creates a considerable challenge.

parallel testing
one of the principal goals of the 2012 in-the-water program was to test the core hypotheses concerning parallel hybrid optimization set out above. To this end, multiple tests were conducted in electric propulsion mode, measuring power into the electric motor controller and power out at the propeller shaft. From this, the efficiency of the combined motor controller and electric propulsion motor was calculated.

referring back to the original HYMAR proposal, the principal objective was defined as follows:

"To research, develop and validate the design specifications, tools and components necessary to build a fully integrated and optimised marine electric hybrid drive train for small to medium sized commercial and recreational vessels."

the original proposal stated that "the major outcomes will be:

- Fully integrated marine hybrid drive system for commercial and recreational craft up to 24m, but scalable to larger vessels
- Zero emissions to air and zero discernible noise and vibration in harbour
- An increase of 50% in the lifetime kilowatt-hour performance of lead-acid batteries in marine hybrid applications and an investigation into other appropriate energy storage technologies including lithium ion
- Full compliance with European Commission Directive 2003/10/EC (Merchant Shipping and Fishing Vessels Control of Noise at Work Regulations 2007) which came into force on 23 February 2008
- Propeller efficiency increased by 5% at full load and greater than15% at "off design point" operation
- Reduction of overall fuel consumption by 20%, tending to greater than90% on long distance sailing boats
- CO2 reduction of 20% in all "off design point" applications such as fishing boats, pilot boats and small commercial ferries"

looking at these outcomes one at a time, we find the following:

fully integrated marine hybrid drive system

as noted, a fully integrated parallel marine hybrid system has been developed, installed, tested and validated. Some relatively minor work remains to be done to make it a market-ready solution.

zero emissions to air and zero discernible noise and vibration in harbour

both the serial and parallel platforms provide a mechanism to conduct all dockside and harbour manoeuvring under electric propulsion and thus with zero emissions and zero discernible noise. The limit here is the capacity of the battery pack, which will typically be higher in a serial system than in a parallel system. Nevertheless, even a modest parallel pack will provide hours of manoeuvring capability at typical harbour speeds and loads.

propeller efficiency increased by 5% at full load and greater than15% at "off design point" operation

the results with respect to this objective are nuanced. The proposal had assumed that the consistently high torque of electric motors as compared to diesel engines would enable larger, inherently more efficient propellers to be used for hybrid systems in displacement hulls. It was anticipated that the gains in efficiency would be significantly greater than 5%. In the event, the torque constant control strategy developed for electric propulsion and regeneration is most easily implemented if a propeller is nominally 'undersized' for the system driving it. On the surface of things, this would seem to imply a lowered efficiency at the propeller.

reduction of overall fuel consumption by 20%, tending to greater than90% on long distance sailing boats

fuel reduction benefits are entirely dependent on the operating profile for a boat and the extent to which a hybrid system is fully optimized. If, for example, a boat operates more-or-less full time below the 'cross-over' speed, reduction of fuel consumption by substantially more than 20% is readily achievable in both parallel and serial applications.

another aspect of the HYMAR proposal was stated as follows:

"The hybrid platform will integrate bio fuels and fuel cells as and when they become available. It represents a necessary first step towards the complete replacement of the internal combustion engine. It will be compatible with, and provide a framework for developing, emerging International Standards Organisation (ISO) and American Boat and Yacht Council (ABYC) standards."

finally, the proposal stated:

"The major new technologies resulting from the project will be:

- An Energy Management Module monitoring all elements of the system and providing efficient and holistic control
- A generic design tool to specify the components required to produce an optimum hybrid drive system for a specific vessel
- A torque adapted, self-pitching propeller
- A design tool for self-pitching propellers
- A new generation of Thin Plate Pure Lead (TPPL) batteries with optimised geometry and electro-chemistry for hybrid applications
- A framework for the integration into the hybrid platform of large-scale, cell-balanced lithium-ion energy storage packs, using 'intrinsically safe' chemistries
- A dynamic permanent magnet DC (PMDC) motor controller
- A dynamic PMDC generator controller
- A study into a rim drive propeller based on a PMDC motor and self-pitching propeller
- A design specification for large PMDC drive motors
- A design specification for PMDC generators
- An outline design for a keel mounted hybrid drive unit
- Contributions to NMEA 2000 standards
- Contributions to the development of a harmonised ISO/ABYC safety standard for electric propulsion systems above 50 volts DC
- Universal AC input to 144 volt DC battery charger
- Efficient 144 volt DC-to-AC inverter and DC-to-DC converter from 144 volts to 12 volts and 24 volts"

potential Impact:
use and dissemination of foreground
section A – dissemination measures
extensive dissemination activity was undertaken making full use of the wide communications networks available to ICOMIA and ready access to the press through Nigel Calder. Some 67 separate activities have been recorded over the three year period of the project equating to one every 2 to 3 weeks.

activities range from presentation of academic papers by INSEAN, through multiple public seminars at boat and trade shows, trade magazine articles and exhibition appearances by the final test boat. The HYMAR team (under the ICOMIA banner) also organised a conference dedicated to the project outcomes at the Marine Equipment Trade Show (METS) in Amsterdam

academic papers
- "A wireless underwater torque measurement system for self pitching propellers" – presented by INSEAN at Nantes
- "Theoretical and numerical hydrodynamic analysis of self pitching propellers" – presented by INSEAN at Hamburg
magazine articles
A total of 22 magazine articles have been published in the following magazines:
- International Boat Builder
- Professional Boat Builder
- Yachting Monthly
- Sail Magazine
- Ocean Navigator
- ABYC Reference Point
- International Boat Industry

the HR42 trials vessel was invited to three separate exhibitions as a feature boat where it was open to the public. Exact numbers of visitors are hard to judge but, at the Southampton boat show for example, as many as 12 people were on board at any one time and an appointments system had to be run. The boat led the parade of sail for three days and produced the largest number of enquiries of any vessel, according to the show organisers.

- National Maritime Museum Cornwall – 3 days alongside and open to visitors
- Mylor Yacht Harbour Cornwall – 3 days alongside and open to visitors
- Lymington – 1 day alongside and open to visitors
- Southampton Boat Show – 3 day alongside and open to visitors. Led parade of sail twice a day for three days.

national Maritime Museum Cornwall - June 2012
southampton Boat Show - September 2012

mylor Yacht Harbour open day – June 2012
industrial conferences
A total of 12 industrial conference papers were presented at the following events:
- British Marine Equipment Association annual conference – Southampton, October 2009
- Rhode Island Marine Trade Association conference – Rhode Island USA, November 2010
- 31st Symposium Yachtentwurf und Yachtbau – Hamburg, November 2010
- HYMAR conference METS – Amsterdam, November 2011
A further five HYMAR papers were presented at ICOMIA's 2nd hybrid marine propulsion conference held in November 2012, just after the end of the project.

the project was fortunate to have ready access to the major EU boat shows through ICOMIA and a very wide range of other public, technical events. Every opportunity was taken to disseminate the project results and to interact with potential users of the technologies being developed. 23 significant workshops were held and a number of lesser public presentations at yacht clubs and other marine venues.

major events took place at:
- International Boat builder exposition, USA
- Bilbao University, Spain
- International Yacht Restoration School, USA
- Seattle Technology Day, USA
- Open Yards, Kungsviken, Sweden
- Port Townsend, USA
- Annapolis, USA
- Gothenburg, Sweden
- London Boat Show, UK
- Dusseldorf Boat Show, Germany
- Southampton Boat Show, UK
- British Marine Federation, UK
- Marine Equipment Trade Show, Amsterdam, Netherlands
- TBAT workshop, Derby, UK
- ICOMIA, London, UK

section B – exploitable foreground
HYMAR was an industrially focussed project and the main industrial partners sought to achieve exploitable outcomes. Results were as follows:
1. ICOMIA – draft international standard 16315 on the installation of marine electrical propulsion systems
2. ICOMIA – widespread industrial dissemination of data on hybrid propulsion systems
3. ESP – large database of hybrid systems performance results and new knowledge related to installation and operation
4. Bruntons – self pitching propeller optimised for hybrid propulsion
5. Mastervolt – DC-to-DC converter suitable for high power operations across a wide range of DC voltages
6. Triskel Marine Ltd – precision, networked voltage and temperature sensor for individual battery blocks
7. Triskel Marine Ltd – precision, networked current sensor and battery monitoring system for high power DC electrical system
8. Triskel Marine Ltd – whole boat energy monitoring and management system
9. Triskel Marine Ltd – whole boat remote monitoring system

1. Draft International Standard 16315
it has been recognised for some time that hybrid propulsion systems will not be widely adopted by the marine industry unless their installation is covered by a recognised standard. Boat builders need to ensure that their construction methods are demonstrably safe and compliance with an ISO standard is an accepted method of achieving this.

3. Database of hybrid systems' performance
one outcome that was not anticipated was the development of an unparalleled database covering the performance of conventional inboard propulsion systems, and of serial and parallel hybrid propulsion systems, including various core components within these systems (notably generators and electric machines). This database has no equal in the world. It has already been used in a limited fashion as the basis for numerous technical presentations and in various trade and consumer publications. Nigel Calder and Ken Wittamore, the two primary authors, are becoming increasingly widely known within the industry as the source for reliable, objective and unbiased information concerning the relative performance of marine hybrid propulsion systems given different operating environments and duty cycles.

4. Self pitching propeller optimised for hybrid propulsion
A hybrid sailing vessel has 3 different operational modes and needs a propeller which is effective in each mode. Matching the propeller to the prime mover is an essential part of delivering whole vessel energy efficiency but the electric motor and the diesel engine have different characteristics and would normally require different propellers. A sailing vessel when sailing is capable of regenerating power and this requires a propeller which is effective as a water turbine.

5. DC-to-DC converter suitable for high power operations across a wide range of DC voltages
boat systems are predominantly DC with propulsion systems operating at a relatively high voltage (144v and 48v on the HYMAR boats) and domestic systems at lower voltages (12v and 24v). An important aspect of energy management is the ability to transfer energy around the vessel to where it is most needed and this requires the ability to interconnect the high and low voltage sides of the DC power distribution busses. This is the role of the prescaler.

6. Precision, networked voltage and temperature sensor
the power electrical side of hybrid propulsion systems are electrically noisy. Although the power source is DC, the motor/generator itself is AC and the power electronics which converts from one to the other produces electrical noise. In addition, rapidly slowing a large inductive load such as a motor generator can inject large electrical spikes into the electrical system. The power source for the motor is a large bank of batteries and these need accurate management at the individual block level in order to maintain a reasonable operational life. Shunt based voltage measurement is neither accurate enough nor responsive enough to handle this particular task and TML, as a HYMAR partner, developed a new set of voltage measurement devices for this purpose.

7. Precision, networked current sensor and battery monitoring system
as with the voltage sensors, precision measurement of current throughout the vessel is an essential part of hybrid management and control. Shunt based measurement is nether sufficiently accurate nor responsive enough to follow power flows within the system so TML, as part of the HYMAR project, developed precision current measurement sensors. In addition to measuring current and voltage, the sensors also carry out amp hour and watt hour counting to calculate battery state of charge. Opportunistic correction of state of charge is undertaken using battery specific look up tables.

8. Whole boat energy monitoring and management system
key to the management of energy on the vessel is accurate knowledge of the energy resources available, their cost and how they are being used. By combining data from all of the voltage and current sensors on the vessel, a complete energy flow system has been produced which shows in diagrammatic form all of the energy flows. These flows are displayed at the chart table and are also available to view on board on any device with a web browser.

cost can be assigned to each source of energy allowing sources to be prioritised based on the cost of supply. At sea for example, bulk charging of the batteries will be achieved using the generator but as soon as the generator moves outside of its optimum operational range it will be shut down. Charging could then be transferred to a fuel cell (if fitted) with final conditioning charging coming from solar. Alternatively, under sail, regeneration could be used and alongside, priority can be given to shore power.

being able to monitor, manage and visualise the energy flow proved to be a key element of the overall hybrid system as this is one area which is very difficult to model mathematically.

further work remains to be done before this system is ready for the open market but the concept has been proved, its value demonstrated and there has been an encouraging level of interest from the commercial shipping industry where whole boat energy management is becoming an important issue.

live energy flow diagram displayed on an iPad

9. Whole boat remote monitoring system
once all of the boat data has been collected together at one central point it is a relatively small step to make it available on the internet. TML have produced a remote monitoring module which allows all of the boat data to be viewed on the internet in real time. The main purpose of this is to provide remote diagnostics of instrumentation, electrical and engine problems, as well as allowing the vessel owner to produce long term efficiency and duty cycle statistics.

list of Websites: