Florida Institute of Technology


Spring 2004

Preliminary Design Review


By: Justin Enjo, Lyndsay Freeman, Arial Nulph,

Daniel Olday, and Rafael Oller


The AUV Kamikaze project is aiming to design a fully automated submarine capable of performing 2 to 3 hour preprogrammed missions in extreme ocean depths and circumstances. More specifically, the three major tasks of our assignment are to create an AUV that: 1) is capable of characterizing the data field around the Vailulu’u crater and seamount using a wide array of traditional oceanographic tools; 2) is able to use this data to make necessary mission corrections in order to optimize data collection; and 3) can return home safely. Issue three may sound like a given condition, but in the case of our research and development, it is of up most importance. All in all, our project is helping to push the realm of human exploration into harmful areas where humans cannot possible go and the use of automated robots is of vital importance.

The most fundamental theory behind AUV design is tailoring the AUV to be mission specific while trying to maintain a high level of modularity. The AUV must be able to complete its mission. However, there are many outside factors that can affect the integrity and motivation at any time. Our AUV must be able to be easily adjusted to meet changing mission demands at any given moment, whether these factors are environmental or from the governing mission authority. Throughout the research and development of the AUV, this modularity capability must be kept in mind. The major sectors and components of the AUV that must remain high modular are the payload systems, the communications links, and the powering capabilities.

We begin our development by breaking our research into a several categories. This allows us to create a division of powers among the development phase of the AUV, and to help in our deadline predictions for development. The major sections of our AUV development process are Hull Design, Power Design, Propulsion Design, and Communication Design. Each of these sections takes into account a new aspect of the overall system design. We are then ranking the categories in developmental order. For example, the Hull Design precedes everything because it starts our project with the original dimensional guidelines, sort of like building from the inside out. Also, the complexity increases throughout the developmental stages with each stage becoming somewhat more convoluted in theory. For example, software implementations of the neural networks within the AUV require a much more advanced level of technology than the PVC hull casing.

Once we have a fully powered and moving AUV, only then can we begin sensor design and payload development. Our AUV is going to incorporate cutting edge technology available to us by our sponsors and our budget. All of our payload systems are integrated independent systems that require no software development on our parts. The harmony of our system depends on our ability to give these sensor systems an environment suitable for there functionality.

Currently, our major focus is on the major hardware development of our AUV, namely Hull, Propulsion, and Powering Design. We also have been focusing on funding and donations from companies throughout the country in order to make the implementation of our timeline smoother.


Powering Design:

Powering design is second on the order of precedence for a number of reasons. Once the hull is designed, we will know the dimensions of the batteries that we are going to use. Today’s market has a wide variety of batteries, so obtaining certain dimensions will not be a major problem. Powering design is the next section to be developed because it is the basis for the rest of our project development. We need power. No power no nothing.

When designing the power system we must incorporate all attributes that contribute to power consumption. A list of the primary components in the overall powering calculation is the motors (main propulsion and control), payload systems such as sensors, mainframe computer including data back-up, and power losses through connections and wiring.

The overall design of the power system must take into account several factors including supplying adequate power, being reliable and consistent with power supply, and being safe so that shorts and faults can be prevented. The most popular power distribution is a battery bank that consists of a series of two or more batteries connected in parallel and diode protected. This configuration ensures an even distribution of power draw amongst all the batteries in the system and the diode regulates the direction of current flow. Each payload system should contain its own fuse and be powered independently from the battery bank source. In case of a system malfunction, only the faulty system will shut down and the rest of the AUV will remain operational.

The power system must be incorporated into the hull so that the system is externally accessible in two ways. First the modularity of the AUV hull should allow easy access to the batteries in order to change them or add more power. The batteries should also be connected to an umbilical connection on the exterior of the pressure hull. This will allow for "shore power" capabilities and allow the system to be recharged with out dismantling the hull. Oh yah, we are going to use rechargeable batteries.

Ideally we want to use deep cycle marine batteries that are sealed lead acid batteries. Deep cycle marine batteries usually have tremendous amp-hour ratings, and unlike conventional car batteries, they can be run down to a very low charge without affecting the overall current draw of the battery. This capability is ideal for our AUV design. Lead acid batteries are low cost, reliable in both time performance and spillage avoidance, and generally produce a reasonable cycle life.

Our power system design begins with the motor selection. Right now we are focusing on 12-volt Minnkota trolling motors. Most of Minnkota’s 12-volt motors have a max map draw of 45 amps. This value is consistent with all the 12-volt motors ranging from 35 to 55 pounds of thrust. The length of running time operating the trolling motor at full speed is a function of the motor’s amp draw and the battery amp hour rating. The governing equation published by Minnkota for estimating the running time for their batteries is:

0.85 (Battery Amp Hour Rating) = (Hours of Running Time) x (Motor Amp Rating)

Of course the amp draw of the motor is decreased as the speed setting is reduced, but the maximum case is used in the overall system design.

During our design of the payload systems and the mainframe computer, we will probably design the overall system outside the machine before implementing it into the hull itself. This predicament leaves this area of research a somewhat of a stand still. We need to estimate the amount of power required by our payload systems before we choose the right battery configuration.

Two procedures can be taken to prevent this situation. First, we can try to estimate the required amperage draw of our payload system. Later in our research we will have a comprehensive knowledge of the desired components and there published amp ratings in order to design a harmonious system. The alternative to this is too designing the AUV with ample power, meaning more power than is necessary for our 2-hour mission. This idea might seem uneconomical at first, but the extension of the hull by 8 inches and the vehicle weight increase of 20 pounds might be worth given the system more than enough power to complete the missions assigned to the AUV. Also, greatly increasing the power supply adds to the modularity of the AUV. Even though we are destined to perform a 2-hour mission, the capability of running four-hour missions may open up new avenues for the little AUV Kamikaze.

Propulsion Design:

Originally the design for our planes was having them attached to the AUV by a rectangular brace that was connected to the rear of the AUV. The design seemed to be very unorthodox. Just from looking at the design, it would seem to generate a lot of unnecessary turbulent flow; this is why we have decided to taper off the end of the AUV so that we may obtain more of a laminar flow.

The design that we want to implement is to get rid of the rectangular design and employ a circular ring design. The basis for this design is a circular bracket that is basically made of a bunch of rings, each one getting smaller and smaller as it nears the end, so that the end will be tapered. Attached to the rings will be magnetic couplers, and our propulsion system, a trolling motor. The magnetic couplers will be used to attach the planes to the hull without a direct connection from the inside of the hull to the outside. The main propulsion system will be mostly enclosed within the circular bracket with the exception of the propellers. After attaching all the planes and trolling motor, we will be able to fiberglass off the end, to hopefully achieve laminar flow at the rear.

For our propulsion system, we have decided to go with the single engine design, so that we may save on power, and reduce turbulent flow. We were originally talking about implementing a twin-engine design, for safety reasons, and running them at half power, but decided to scrap that idea, for the time being.

We are primarily focusing on the trolling motors developed by the Minnkota Corporation. Minnkota, by our research standards, sets the industry standard on trolling motors. The plastic housings can be pressure compensated by boring into the housing and fixing a flexible tubing. The pressure will be able to compress this tubing causing equilibrium within the motors. We estimate that these motors will be fully operational to 1000 meters.

The Minnkota trolling motors are factory efficient meaning that we do not need to do any modifications to the gear ratios, transmissions, or duty cycles of the engines. Our only experimentation will centered on the power consumption economies of the various engines. This data can be determined experimental or estimated using published equations from the Minnkota Corporation. We have established a correspondence with Minnkota requesting educational samples of the new saltwater series of motors, however the only reply has been a product catalog that only gives us a few pictures to gloat at.

Hull Design:

This design project was setup to use an existing sectioned torpedo hull created by a senior team from 1997. Therefore, the initial planning stages were centered on improving the existing systems. The key features or components of the existing vehicle that had to be improved included: fastening system of each section, fin and rudder control system and motors, trolling motor bracket, battery pressure vessel, computer and instrument housings, and communications tower. The following subsections will cover some of the problems, ideas, and progress that have been encountered recently. Keep in mind that many of the decisions required to design each component rely heavily on the decisions taken for other components, such as any pressure vessel will depend on the size of the hull or instruments required.

Hull Design (Initial)

This sub-section is designated as the initial hull design because it is the basic structure that was devised to visualize how every component could fit together comfortably, while keeping in mind the ease of assembly, ease of manufacturing, space maximization, and most importantly structural strength and integrity. Figure-1, shown below, demonstrates and labels the ideas and components that were devised.



This could reduce both the time required to assemble and the number of bolts at risk of coming loose if not properly tightened. The long rods could be inserted through every section without requiring threading and provide transverse strength along the hull; additional strength could come from the outer sections supporting each other while pressed together. The figure also shows a neat design to carry different payloads such as batteries, instruments, or sensors by just sliding in a prefabricated Payload Frame between the long rods. This system would also reduce assembly time when adding or removing components. One drawback to the system is that the rods take up around 1" of usable diameter from the inner space that could be used in bigger size spherical instrument housings.


Instrument Housings

Pressure housings are required to protect the sensitive PCU, PC-104 boards, and other electrical components that navigate and control the vehicle. The extreme pressures of depths of 1000m coupled with the extremely volatile saltwater environment make the decision for these pressure vessels very important. This design goal has proven to be the defining characteristic around which the entire submarine would be developed. The problem is that unlike pressure compensating, the differences in pressures require a precisely manufactured vessel and seal to withstand the extreme forces. This amount of precision can only be obtained if the spheres were purchased from or donated by the few manufacturers who make them. Because the initial design confined the sphere to an 8.7" diameter within the long rods, it was hoped that a two spherical vessels of 8.5"inches could be purchased and used; moreover, if only a 9" inch diameter sphere where available, the rod system could be replaced by a different fastener system.

After researching glass sphere pressure vessels manufacturers it became increasingly clear that the housings were not manufactured in sizes under 10" diameters. This past week, around March 10th, Rafael contacted one of the few main manufacturers of these types of pressure spheres in the world, McLane Laboratories. Although the contact in the company, Wilson Peters, stated interest in sponsorship of the program, he confirmed that the spheres from their company are no smaller than 12" diameter. This fact proved to be a crossroad in the design process since an entire new hull and system would have to be developed to incorporate such large balls. Mr. Peters provided an additional option in that his company also manufactures cylindrical aluminum vessels rated to 6000-meter depths. Although this vessel shape fits nicely into the existing design, it lacks the buoyancy force generated by the glass sphere and it increases the cost of a unit due to the additional machining and work. It seemed that Mr. Peters would be willing to donate the glass spheres but not the aluminum vessels.

Battery Pressure Vessel

Although the design of the battery housing will ultimately depend on the battery type and ratings chosen, guidelines exist that limit the battery size to the hull of the vessel and how it is attached. The basic design of the pressure vessel will require that the space be pressure compensated with mineral oil to give a higher factor of safety when constructing a pressure rated, watertight compartment. With pressure compensation, the stress differences at the seals are kept to a minimum since theoretically no difference exists between either of the sides. Incorporating the existing sectional type rings, seen in Figure-2, limit the diameter of a cylindrical shape to an 8" diameter. Using this diameter allows the length of the cylinder to be expanded because the vessel could in turn fit within several 15" PVC sections. This additional length can be used for larger batteries, battery chargers, or other components that requires pressure protection. Every component that goes inside the battery pressure vessel has to be adjusted to allow safe function while immersed in the mineral oil.


Design Strategy

Because the development with the instrument housings represents a big obstacle in the ongoing design process, a new strategy was implemented that opens up the way for three hull options:

Keep the initial hull design and implement the more expensive aluminum pressure vessels –positive: keeps the initial design advantages, saves some manufacturing time, adds more space for instruments due to the longer vessel space –negative: requires additional components to reach neutral buoyancy and increases the overall cost.

Redesign the entire hull from scratch using larger diameter pipes, or the existing human powered sub –positive: increases the payload size for the spheres and other components, allows the use of the donated spheres –negative: would require a lot more work to build a new sub hull or work up from the human powered sub, increases the cost of the project for new hull materials, larger diameter stock could be much harder to find.

Design a system that could attach the glass sphere balls into the existing hull components –positive: incorporates the donated spheres & existing components, stabilizes the budget for the project with a slight increase for materials, saves time on overall assembly, increases size inside the existing PVC sections for additional payloads –negative: complicates the structural design by adding different size sections and components, could decrease the structural strength if not properly designed, increases assembly time and the number of fasteners, changes the sleek look of a torpedo

Instead of choosing one option and risking a bad choice, the logical procedure to follow in a decision is to begin thinking of ways and designs to implement option three, to closely examine the work required to build a new hull to decide, and to hold the first option on as a last resort, since it would be easier to fork over extra cash and incorporate the metal vessel.

Motor Bracket

A simple motor bracket was devised where a square aluminum part cradles the trolling motor rod and sandwiches it with the AUV end cap. As the motor is turned on, it pushed against the end cap itself, which allows the square cradle to be reduced in size and strength. Four rods bolt the cradle to the end cap and can be unscrewed from either side by cutting a slit in the ends of the rod to stop it from spinning when applying the torque. A set screw in the bottom of the cradle secures the trolling motor from moving around while not in operation. Although this system is extremely simple, it is very easy to manufacture and provides an easy means to attach a single motor. See Figure-2 below:

Hull Design (Option-3)

A preliminary design was drawn to attach the larger 12" sphere within the 10.5" outer diameter sections by using a set of brackets that cradle the sphere in place and provide structural stability along the hull. The basic system is shown in Figure-3 below:


As seen above, six arm brackets hold the sphere in place and act as support for the structure of the hull sections. Six arms were chosen at an angle of 60˚ between them because this is consistent with the previously described initial design. The arrangement of the arms also allows the sub to rest on the bottom two brackets and could be raised by the top two. On the left side of the figure, the arms are bolted into the Sphere Ring as shown; however, on the right side a different system was devised as discussed next.

Since the control systems require two sealed spheres they would have to be placed in two of this type cages. The main obstacle in this proposal lies in the attachment system between the two cages, highlighted in the theoretical assembly below.










Figure-4 above shows how the assembly of the submarine would look with the sphere "cage" or bracket system. The disadvantage in this design is than the points highlighted as Problem Area lie in an inaccessible location when assembling and disassembling the sections. One option to get around this problem is to change the arrangement of the bracket arms in either one of the rings that faces the other. In other words, the first cage (on the left) could be bolted to the left side ring using the six rod pattern In turn, the bracket arms from the 2nd cage and be bolted into that left ring using different screws than those connected to the 1st cage. This means that the battery compartment is sandwiched in between the two cages but can be reached by simply unbolting the left ring bolt pattern connected to the 1st cage. As seen in this assembly area, one of the PVC sections is cut in half to allow an even distribution of the battery pressure vessel. The sections on the front of the 1st cage and the back of the 2nd can be used for instrumentation and rudder control motors respectively.

Although this design seems workable when visualizing it, it would be much more complicated to manufacture with a suitable structural strength for the entire body. Adding these components complicates the assembly of the vehicle and the amount of screws that have to be properly tightened. Without fiberglass housings to cover up the brackets, the entire system would look complicated and unreliable, which is also measure of the work put into the vehicle. A nice appearance improves exposition of the AUV in different markets and to different companies.


Funding and Donations:

Thus far, the team has arranged for several components to be donated. Two of these products have already been delivered. Denise Shirey of Sparton Electronics has sent an SP3000D Digital Compass which will help to guide our vehicle and save our team nearly $800. Some Company in Melbourne has delivered a $150 cylindrical piece of aluminum that will be machined and used to attach the dome onto the front of the vehicle. Vinnie Herbert of Measurement Computing will soon be sending a $500 data acquisition board that will be instrumental in our sensor activity for the mission. Quinn Jones of MaxStream will be donating two refurbished radio modems so that our vehicle can receive new commands and missions wirelessly. The contact information and product numbers can be found in the appendix of this report.



The major concern of the AUV Kamikaze project continues to be financial dilemmas. We have been provided with a hull, bindings, and a few pressure vessels that we are "determined" to use in our design. The overall optimal design of the AUV would be to start from scratch and purchase a new hull that is more compatible to our mission standards, however, more research and brainstorming is showing an increasing potential for the existing hull.

Even though we have engineered a thorough schedule for the AUV production is seems that the harmony of the system requires the hull design, propulsion design, pressure vessels design, and control housing design to all be implemented at once. We are currently modifying our design protocol to try and implement all of these considerations into the design at one time, which is becoming time consuming.

The single largest break through we have accessed is to modify the existing hull structure that we have. We will primarily accomplish this by extending the length of the AUV and by reducing the amount of cross sections within the AUV hull. Current research is underway in trying to find replica PVC housing for the AUV. The integration of longer hull sections will greatly increase our working room and the internal holding capacity of the AUV as well as magnify the modularity that we original expected. We might be sending this AUV to the moon after the volcano trip!


Home | Team Members | Schedule | Archives | Search | Discussions | Contact Us

 Copyright 2004 by Florida Institute of Technology
and The Department of Marine and Environmental Systems
For problems or questions regarding this web contact [jenjo@fit.edu].
Last updated: 04/24/04.