Generalized Solar/Electric Boat Design Theory

Introduction

The purpose of this paper is to provide a brief insight to the necessary steps involved in designing a solar electric boat for the Solar Splash World Championships held annually in Milwaukee, Wisconsin. This solar / electric regatta provides college and high school students hands on engineering experience. The students are involved in the complete design and construction of a solar / electric powered boat. Designing a prototype system of this nature is obviously a difficult task. A large amount of design and thought is necessary for this type of project.

Unfortunately, because these projects are student run that means team members are around for a relatively brief period of time. This means that with the high turn over rate, the team runs the risk of loosing the information and experience that previous generations accumulated if their work is not completely documented. Technical project team’s are completely about a learning curve. The longer teams are around the more advance they become and the higher their performance is. Unfortunately for team’s entering the competitions for the first time have the same problems that other teams experienced when they started.

The goal of this paper is to provide a generalized background into the workings of a solar electric boat and the considerations in the design of such a vessel. The paper is targeted to the rules and race conditions for the Solar Splash Competition. By no means is this paper suggesting that this is the only way to design a solar electric boat. The very nature of the project dictates that the design processes and construction methods must be re-evaluated constantly in order to ensure peak performance. However, the following method of design has proven successful for multiple teams and provides a good foundation for building on.

It is the author’s hope that the paper provides useful information to future Solar Electric Boat Team members and starting teams in order to promote the mutual benefit and advancement of the competition and performance of the teams involved in it.

Competition Objectives

The Solar Splash regatta started in the 1993-94 school year as the World’s first intercollegiate solar electric boat race. Each team entering the race must design, build, and race a craft that fits into the specified box dimensions of 6 meters in length, 2.4 meters in beam, 1.5 meters in freeboard and 1 meter of draft. Each craft must run on a combination of stored battery energy and solar power to compete in the events. The competition is comprised of five point earning events:

- Technical Report

- Presentation (display of boat and other visual aids for educational purposes)

- Maneuverability Race

- Endurance Race

- Sprint Race

The team earning the most total points is declared the Overall World Champion. This "all around" point system encourages teams to design the best well-rounded vessel. The competition was designed to give hands on engineering experience to colleges and now secondary schools around the world. The University of Michigan Solar Electric Boat Team’s (hereafter referred to as UMSEBT) goal is to design, construct, test, and race an entry in the annual Solar Splash event. Throughout the course of the project, the Students and faculty involved learn valuable skills in design, leadership, teamwork, and engineering.

The two main events of the competition are the sprint and endurance races. The sprint race involves a 300-meter dash. The endurance race involves the craft achieving the maximum number of laps on a one-kilometer course in the allotted two-hour time span. A Solar Electric boat is an extremely complicated vessel to engineer for its relatively small size. Most of the crafts put forth to compete in the Solar Splash World Championships are complete prototypes in their own right. In order to design a vessel that can perform well in two completely different envelopes of operation (i.e. sprint and endurance), great care in design of the vessel must be executed.

The competition is a one event endurance race. It takes place in early May on Lake Burley Griffin in Canberra. Independent entries as well as serious commercial contenders are included in this race, beyond just student teams as found in the Solar Splash competition. With the right funding, the year 2000 will be the first time for the UMSEBT to participate in the Bayer Solar and Advanced Technology Boat Race. This race is fairly new, having just begun in 1997.

The design process for a solar electric boat can best be described as a converging design triangle. Shown below is a pictorial representation of this design process. The design of the craft is encircled by the restrictions in the competition. Within the circle formed by the Solar Splash Rules is the main design triangle which is in turn divided into four triangles, depicting the three main aspects of design in a solar boat (electrical, mechanical, and hydrodynamic), and the pursuit of the optimum design for the competition. The rules of the competition limit the vessels to a certain amount of stored battery energy and solar energy for the duration of the competition. The specific specifications of the energy limitations will be discussed later in the paper. However, since the power source is governed by the race rules it is a logical choice to start there in the design of the vessel. The goal of designing the electrical systems is to construct the most efficient, most powerful electrical configuration under race guidelines. At this point the vessel’s motors should be selected after which the most efficient hull form and propulsion system are determined to ensure maximum performance of the vessel.

The goal of the electrical design process is simple but crucial. The team is trying to provide the most electrical power to the system for the given events. The power source is a combination of battery power and solar energy. According to Solar Splash regulations the team is allowed the following power for the sprint and endurance events. The sprint race dictates 1.5-kilowatt hours of battery energy and the use of the solar array for charging batteries in-between heats. The endurance race allows 1-kilowatt hour of battery energy and 480 watts of solar energy. The primary energy storage devices are lead acid batteries. The competition dictates that the batteries be stock, lead-acid, unaltered, and have a mass of less than 31 kg and 44 kg for the endurance and sprint races respectively. With the power source limitations in mind the electrical analysis can be started. The main topics of interest in the electrical design are battery selection, photovoltaic panel design, power management, the power circuit design and controls, instrumentation, and modularity of design.

Lead acid batteries produce their electrical energy by changing sulfuric acid and lead into water and lead sulfate. As the lead sulfate crystals form on the batteries, the battery depletes its reserve of electro-chemical energy. The faster the battery is discharged (higher amp draw), the larger the crystals that are formed and the more restricted the battery is of its electromotive force. This crystaline-based restriction of electrical flow is the one of the major sources of the internal resistance of the battery. The larger the crystals are that form, the larger the internal resistance and the lower the terminal voltage is. When a lead acid battery is discharged at high amps, it depletes what is called the surface charge. This surface charge is not the entire battery’s reserve. If the battery is left to sit after a high ampere discharge it will actually regain some of its charge. For example, if a battery was run at a several hundred amp discharge until it was dead, after sitting for a half an hour the battery could produce power again (at a much lower draw). Another point of interest in lead acid battery performance is that they perform better when at higher ambient temperatures. In other words the battery can produce more energy at an ambient temperature of 100° F than at 50° F. With this information in mind we’ll shift to the actual battery selection process for sprint and endurance.

For the endurance portion of the competition and for the rules qualification of the battery the important information is the amp hour rating and the operational voltage of the battery. Since the competition dictates the kilowatt-hour ratings on the basis of a two hour discharge, the time dependence amp draw curve must be examined for the battery being considered. Below is an example of such a curve.

Fig 1- Exide H26-60R Time Dependence on Amp Draw

After it has been confirmed that the batteries are within power limitations, the designer should check to see if the battery bank is under weight limitations. If at all possible the designer should consider that the keeping the maximum battery bank voltage (36 V) is ideal because for a given wattage input, an electrical motor will almost always perform better at higher voltages than at higher amps. The iterative process of battery selection should be repeated until the teams battery pack is as close to the kilowatt hour limits as possible.

When choosing sprint batteries, keeping a system voltage of 36 volts is wise because it will reduce the amount of amps necessary to drain the batteries and will allow for better performance of the motors. The governing factor in selecting sprint batteries is choosing a battery that is the least effected in it’s loaded voltage by high amps. As stated at the beginning of this section, the when a battery is loaded heavily the internal resistance lowers the terminal voltage of the battery. Therefore the designer must find the battery that has the best high amp-draw power discharge curve. This curve is the mean voltage vs. amp draw curve and an example of this type of curve is shown below.

Fig 2- Exide H26-60R Terminal Voltage vs. Amp Draw

Unlike the Solar Car competition which limits the surface area of the panel and not the wattage, Solar Splash limitations only limit the wattage. In conjunction with this and the fact that the competition is annual, the majority of the teams have opted to use manufactured panels because of their relatively low cost and high reliability. Of the different types of solar panels, the three main types there are monocrystaline, polycrystaline and thin film. Monocrystaline are the most efficient (17% average) and consequently the most expensive. Polycrystaline are less expensive but are also less efficient (average 14%). Thin film cells are the flexible unlike mono and polycrystaline but have 50% the efficiency of a monocrystaline panel. Another nice feature of manufactured panels is that most of them come with bypass diodes installed. These diodes prevent back flow of current through the panel which can be damaging to the cells. Unfortunately, manufactured panels are not always quite to the power configurations desired by the team. Therefore if the resources and time presents itself, assembling custom panels is recommended.

The only event where power management is a major concern is the endurance race. The sprint is an all out drain of the battery energy and only lasts around 30 seconds so power management would be a mute point. However, in the endurance event the boat must draw maximum power with the goal in mind that the battery bank be completely depleted at the end of the two hours. In order to achieve this some sort of power point tracking system is recommended. Many teams in the competition have encorporated power management systems in various shapes and form. One team in the competition has encorporated a computer controlled charging system that allows the craft to run on one battery while the other is charging. Once the operating battery is depleted to a certain dead voltage, the computer switches the power to the charging battery and switches the charging to the dead battery. Another method is using an actual power point tracker that is a little "black box " that controls the power output of the system to maximize towards a goal (in this case 2 hour discharge). The old fashion method of power management is to adjust the throttle according to the battery’s dicharge curves.

Battery discharge curves

The main concern for the power circuit design is minimization. Minimization of weight, resistance, and complexity while keeping safety in mind is key. The designer should use the least amount of main power cable possible while keeping within amperage limitations of the cable. Heat transfer texts and the National Electrical Code are excellent sources of information on these topics and provide good guidelines for the selection of main power cabling. While most teams have different cable sizes for the endurance race (different power configurations), the sprint race choices for cable are fairly standard. The severely harsh amp loadings restrict most of the teams to using cable between 2/0 and 4/0 AWG (American Wire Guage).

Controls for the main power circuit are fairly simple to design also. There are many methods of power control ranging from thyristors to Pulse Width Modulation Control (PWMC). Sprint configuration controls are very simple because since the events purpose is to deplete the energy of the batteries at such a high rate, throttle control is unnecessary. Most of the teams throttle consists of "dead man’s" on/off switch that controls main connectors, or solenoids. The only concern in the design is the proper selection of the solenoids to ensure that they can handle the loads of the systems high amps.

Any prototype system is going to have the need for excellent instrumentation for the exploration and documentation of the vehicle’s performance. Therefore it is essential that all aspects of the crafts performance be recorded and evaluated. Some of the more important instrumentation necessary are amp and voltage meters for various points in the system, tachometers for monitoring motor rpm’s, and telemetry measurements to give velocity speed. Other useful pieces of information would be torque meters and thrust meters for the propeller. If at all possible use a data logging system that has the serial output capacity for downloading information to a laptop computer for further analysis. One major concern however for the instrumentation readouts is that it is in a logical format and easy to read layout for the driver. Usability for the instrumentation system is of the utmost importance.

Because there are two distinctly different configurations in the competition it is essential that controls, instrumentation and safety features such as fuses be modular to ensure the conversion of the vehicle from endurance to sprint mode and back is as painless and as simple as possible. Last year one team adopted the method of having the sprint and endurance controls and instrumentation plug in through a universal 36-pin connector. This and a similarly modular design for the drive train allowed for a short turn-around time of about a half an hour between sprint and endurance configurations.

Mechanical Design

The goal of the mechanical stage of design is to create the most efficient method for transmitting the available battery to propulsive energy. The only requirements for designing the endurance and sprint drive train dictated by the competition rules are safety ones. The system must have the proper safe guards on dangerous moving parts and the unit must be capable of handling the dynamic loads including a safety factor. The remainder of the design of the mechanical systems relies on the creativity of the team. The design of the mechanical system involves the selection of motors, power transmissions, steering, and drive units.

The differences between the motors mainly differ in their controls and motor curves and will not be discussed in this paper. When looking for different motors search for the one that can give you the highest efficiency and power for the least amount of weight. This type of analysis is performed using the characteristic motor curves. It should be noted that Electric motors are a two independent variable system. Therefore if you know any two characteristics of the motor from its set of curves (any two of volts, amps, torque, rpm, power, efficiency) you can extract the remaining four traits.

Selecting an endurance motor is fairly simple. Most teams use a motor around one horsepower for the endurance race. When considering the use of more than one motor it is important that the power to weight ratios and efficiency to weight ratios be considered because a large portion of most electric motor’s weight is from the casing of the motor. It is recommended that the number of motors being used be kept to a minimum because of these reasons.

Unfortunately selecting a sprint motor(s) is a little more complicated. Unlike the endurance a one motor configuration is not always the most efficient one with respect to horsepower to weight ratio. As mentioned earlier in the report, motors perform better under higher voltage than higher amperage conditions for a given power setting. The battery packs of the sprint configuration put the system at around 22 horsepower for the given terminal voltage and amp draw. This operating condition is particularly hard on the brushes and windings of the motor. It is necessary to consult the amperage time limit curve of the motor being selected to ensure that the motor can handle the abuse that it goes through in a sprint condition. An example of such a curve is shown below.

Fig. 3-Cupex Time-Amperage Limit Curve

These limits cannot be exceeded or the motor will "burn up". Because of the current limitations of the motors dividing up the current load by placing smaller motors in parallel sometimes is beneficial if the horsepower to weight ratio and efficiency is improved.

Another major concern in the selection of motors is the "System Attainable Horsepower" curve. Because of the characteristic of battery terminal voltage decrease due to amp loading, the highest efficiency point of the system is not necessarily at the highest discharge capacity of the battery. In order to construct the attainable horsepower curve, the designer must take the terminal voltage and the amps of the system at a series of points along the curve. Then the designer divides the amps by the number of parallel motor circuits (e.g. a four motor parallel system would yield amps divided by four). The resulting coordinates of (terminal volts, amps) are then applied to the motor curves and the individual attainable horsepowers are extracted. This horsepower is then plotted against the amp draw. The resulting curve should be a concave down, shallow quadratic function. The reason for the plotting of the graph two dimensionally when there are two inputs (terminal voltage and amperage ) is because the terminal voltage is a dependant variable on the amp draw. It should be noted that the ordinate of the graph is at an exaggerated scale of .1 horsepower to emphasize the shape of the curve.

Fig. 4 – Cupex 4 motor system attainable horsepower curve

Even though the scale is exaggerated it can been seen that operating the system at 750 amps as opposed to the maximum battery output for system of 850 amps (based on a 30 second discharge) can save .8 horsepower. This amount may not seem like much but in some craft one more horsepower can mean 4 more miles per hour top end speed.

The goal for the transmission was to be as light weight and efficient as possible and still be able to handle the dynamic loads transmitted by the various motors. Since both endurance and sprint configurations contain the same principles of design, a condensed analysis is possible in this section. In designing the power transmission system there are two major sections to consider. The first is the transmission of the motors power to the drive shaft. The second area is the path of the drive shaft to the exterior of the craft.

Transmission of the motor power to the drive shaft can be achieved in four main methods: chain drive, belt drive, gearing, and direct shaft transmission. Chains are the most forgiving and best well rounded in their traits. The are flexible in their configuration and have relatively medium weight and efficiency. While timing belt drives are generally more efficient than chain drives, they are also heavier. The belts themselves are lighter than chains, but the weight of the hubs are much heavier than their chain counterparts (sprockets).However, if light weight hubs are available the system would be feasible. Also the timing belts of the drives are not variable in length like the chains. Belts are also much more sensitive to alignment than chains. Gearing is more efficient than belts but are fairly inflexible in their design. In order to change gear ratios a physical shift of the motor is necessary in order to align the gears. Direct shafting is the most efficient type of transmission but has no flexibility in gearing ratios, nor is it as vibrationally forgiving as the previous methods. Appendix D contains selection methods and charts for belt and chain drives.

The path of the shaft depends entirely on the type of craft being designed. While most craft can use a straight-shafted system, craft such as hydrofoils sometimes use either z-drives or bend shaft drives in order to get the propeller sufficiently in the water. Straight shafts are much more efficient than z-drives and are much simpler to construct. However in some situations a z-drive is the only option. Whichever system is used it is important that the proper anti-vibrations steps are take so the forces of the propeller are not transmitted to the motor. This is usually achieved by the use of journal and thrust bearings to absorb lateral and longitudinal forces, respectively.

Steering is a very open ended portion of the mechanical design. The only design requirements for the steering is that it give the craft a relatively small turning radius. There are many methods available including, but not excluded to, rudder, daggerboards, and thrust vectoring. The goals of whatever system chosen should be to maximize turning and minimize appendage drag.

Similarly to the modular method of the electrical design, it is recommended that the drive trains be completely self-contained and modular. Designing in this fashion will ensure simple and quick changing between sprint and endurance configurations. Modular design of the drive train units also promotes the reduction of weight in the system because everything is more compact and less material is used.

Hydrodynamics Design

The hydrodynamic design of a solar electric boat is by far the most complicated and most open-ended portion of the design triangle. With a large number of different hull types and configurations, the possibilities of design is only limited by the creativity of the team members. Therefore this portion of the report will only give a brief outline of the basic hull types available. Also the generalized design process of a hull and its propulsion will be discussed. The methods will be described briefly and references will be sighted for further information, as the majority of the math involved in the design of the hull and propulsion are too in-depth for the scope of this paper.

Because of the nature of the competitions set power systems, most teams have roughly the same electrical and mechanical systems. Where the differentiating factor of performance comes in is largely due to the hull and propeller combination. The completely different design goals of the competition provide a difficult challenge for the teams. They have to design a hull that has both low resistance at around one effective horsepower, and also have low resistance at sprint speeds exceeding 30 mph. Of the available classic types of hull forms there are displacement hulls, planing hulls, and assisted lift hulls. Of those hull types they can be broken down into the following sub-groups:

1. Displacement Hull - Monohull, Multihull, SWATH

2. Planing Hull - V-hull, Inverted V-hull, Racing Multihull

3. Assisted Lift Hull - Hydrofoil, Surface Effect Ship, Air Cushion Vehicle

4. Hybrid combinations of the above

All of these hull types have different characteristics when performing in the sprint and endurance configurations and the analysis of them is beyond the scope of this paper.

After the type of the hull being used has been decided, a hydrodynamic and hydrostatic analysis of the hull must be performed in order to predict the performance of the vessel before it is built. Multiple methods for the hydrodynamic analysis including the various dimensionless series for displacement hulls (Series 60, Taylor series, etc.) put forth by the American Towing Tank Conference (ATTC) and the Society of Naval Architects and Marine Engineers (SNAME), various power prediction programs like Power Point Prediction (PPP). Regardless of the prediction method used, it is recommended that a scale model be tested and the results of performance scaled by Froude’s Method. Planing Craft can be approximate by various methods such as Savitsky’s Method and Power Sea, providing the craft is within method limitations. Assisted lift craft can be approximated by the use of the methods outlined in Hydrodynamics of High-Speed craft by Doctors.

At the same time the vessel must be tested for stability. According to the race rules the craft must not list more than 15 degrees with a weight of 10 kg hanging over the shear line (outer edge at the beam). Therefore it is necessary to do hydrostatic calculations for the vessel. Aside from model testing tools such as Hydromax and the Principles of Naval Architecture (PNA) are available to approximate the stability of the craft being designed. In order to calculate the stability a detailed knowledge of the vehicles components, their locations and masses is necessary.

Once the hull design is completed the final step to the design triangle is to design the method for transmitting the mechanical power to the water in order to propel the boat. Many types of propulsion are available ranging from propellers to jet pumps and multiple design methods are available to either select a stock propeller from companies or design one from scratch. Propeller design is a complicated process to optimize and does not fall within the scope of this paper. Two good references for the design and selection of propellers are the PNA series and Marine Propellers and Propulsion by J.S. Carlton.

Conclusion

After the first iteration of design is completed the process should be repeated until the team feels that an optimum design has been reached. The current records for the sprint and endurance are 71 kilometers in 4 hours and 21.86 seconds for the 300-meter dash. These should be the ultimate performance goals of the teams designers. Unfortunately, because of the tight timeline due to an annual competition this is a large amount of technical design work to complete. Because of this, a large amount of the teams have adopted the program where they work with one vessel multiple years optimizing its performance while the design of the next craft is processed. It is the author’s hope that this paper makes the design process a little more straight forward and structured and also provides incoming teams with the necessary information available to get a start on their project. Again the author would like to re-iterate that this paper is based on the findings of the tech teams to date and is constantly changing as the teams all progress on the learning curve.