1.0 About the Solar Splash Race/Competition
The American Society of Mechanical Engineers (ASME) sponsored Solar Splash '94,
which was held August 18-21, 1994 on Pewaukee Lake just
west of Milwaukee, Wisconsin. The event was hosted by Marquette
University and the
Pewaukee Lake Yacht Club. Teams arrived during the day on August 18 to
register and prepare for the weekend of racing. Qualifying and
technical inspections
were held on August 19. Qualifying consisted of each entry completing one
lap of a course designed to test maneuverability as well as passing
several safety tests,
including static stability and driver egression. Times from the
qualifying laps were then used to seed the entries for the sprint races
held on August 20. Each
sprint race consisted of running down a two hundred meter straight-away
course starting from a stationary position. There were four rounds in
all. The first round
was used to seed the entries for the second round and to confirm the
qualifying results. The second and third rounds were elimination rounds
setting up a four
boat final heat. The endurance races were held August 21 on a closed loop
course around which participants piloted their craft for two hours.
Participants were
judged on the number of laps completed within the alloted time. There
was a preliminary round held in the morning and a final round held in the
afternoon.
Solar Splash '94 gained international recognition and attracted
participants from all over the world. The participants were University of
Michigan, Kanazawa Institute of Technology (Japan), Kobe University of Mercantile
Marine(Japan), University of Puerto Rico, Marquette University, McNeese
State University,
University of South Carolina, Western New England College, University of
Arkansas, and Grand Valley State University. The Solar Splash staff has already
reported inquiries from over sixty potential entries for the 1995
Regatta.
2.0 Solar Splash '94 Results
The results from the 1994 Solar Splash Regatta were as follows:
| Result | Sprint Race | Endurance Race |
| 1st Place | Kanazawa Institute of Technology | Kanazawa Institute of Technology |
| 2nd Place | University of Arkansas | University of Puerto Rico |
| 3rd Place | University of Michigan |
Kobe University of Mercantile Marine |
University of Michigan highlights: 3rd Place Sprint; Mechanical Technical
Achievement Award; DNF Endurance. The University of Michigan was unable to
compete in the endurance race due to structural failure of a universal
joint.
3.0 Team Development
The University of Michigan Solar/Electric Boat Team began to take shape
in January 1994. The team is a derivative of another design team that had
begun working on an entry for the Human Powered Submarine races that are
held annually off of the coast of Florida. The team leaders of the original
Human Powered Submarine Team received a fax from Marquette University
inviting them to participate in Solar Splash '94. The Human Powered Submarine
Team, while still in its infancy and short on personnel, was forced to
make a decision. The decision was whether to proceed with the Human Powered
Submarine, or to withdraw from that event and to enter Solar Splash '94.
After lengthy discussions and a team vote, the decision was made to
proceed with
the solar/electric boat pro ject for several reasons. The most attractive
aspect of the solar/electric boat project was that it was an engineering
contest that eliminated
the dependence on athletic prowess. Another attractive feature of the
solar/electric boat project was that the technology developed would be
more applicable to
real world problems. Team members were also excited by the prospect that
virtually anyone could pilot the boat. This was not true for a human powered
submarine because only certified SCUBA divers are allowed t o pilot such
a vehicle. These three factors proved to be enough to sway the team to
enter the 1994 Solar Splash Regatta.
4.0 Design Direction
The team was late in entering the event and with only six months to
design and construct a solar/electric boat, we decided to focus our
efforts on the sprint race.
Producing a boat that could win both the sprint and the endurance races
would require more time and funds than were available. In addition, the
University of
Michigan has expertise in the area of planning hull design which we felt
was unavailable to other teams. Concentrating on a planning sprint boat
would also allow
the team to test innovative ideas as in planning hull technology.
Trading viscous drag for pressure drag (i.e. reducing viscous drag and
increasing pressure drag) is in general a positive trade with regard to
increasing speed.
Thus, increasing the ratio of pressure drag to viscous drag can produce
an increase in speed for a given power requirement (i.e. net
drag).Therefore, in order to get the highest speed out of a given power supplied,
the objective becomes to reduce the wetted surface of the hull.
This reduction in wetted surface can be accomplished by decreasing the
draft of the boat via dynamic lift. There are several types of watercraft
that produce dynamic lift which include planing hulls, hydrofoil assisted
hulls, and surface effect hulls. The use of a surface effect hull was
ruled out early in
the design process because it either requires a large amount of power at
zero forward speed (i.e. air cushion vehicles), or it requires very high
forward speeds
(i.e. ramjet vehicles) in order to produce dynamic lift. Over the course
of two hundred meters one can neither afford to expend power that does
not result in
forward speed nor expect to attain speeds high enough to achieve ramjet
effects. Hydrofoil assisted hulls are often difficult to design and
build. This is due to the
difficulties associated with structurally supporting the hydrofoils, and
the need for complex control surfaces. These factors, combined with the
availability of
planing hull technology at the University of Michigan led the team to
design and construct a planing hull. In hindsight,this was an excellent
decision because
none of the three teams who designed their craft to use hydrofoils
actually used them at Solar Splash '94. One team encountered structural
problems with
supporting their foils, and the underwater lake vegetation inhibited any
team using underwater appendages.
The planing hull surface and propeller were designed using computer
modeling tools developed at the University of Michigan in conjunction
with Michigan
Sea Grant and Vorus & Associates Incorporated. In order to investigate
unproven concepts in planing hull technology, the team decided to attempt
the design and
construction of an inverted vee planing hull under the
advisement of our faculty advisor Dr. William S. Vorus. It was theorized
that the inverted vee
would not only provide lateral stability in shallow draft conditions,
but also provide added lift from the interaction of the internal jet
heads and reduce the wetted
surface by causing flow separation at the wing tips. In the
process of making the decision to design and construct this planing hull,
the team set two
goals. The first goal was to get such an innovative concept to
successfully plane given the maximum electrical energy available. The
second goal was to attain a
top speed of twenty miles per hour.
The decision to use a surface piercing propeller was driven by the shape
of the transom and the desire to minimize appendage drag. With an inverted
vee shape at the transom, it is difficult to support a fully submerged propeller
without encountering increased appendage drag from subsurface structural
members.
In addition, a properly designed fully cavitating surface piercing
propeller can take advantage of the aforementioned trade-off between
viscous and pressure drag.
Appendage drag could be further reduced by not having submerged
steering devices. The team therefore decided that maneuvering would be
accomplished
by shifting the pilot's lateral position. This would create a net turning
moment due to a lateral displacement of both the center of gravity and
the net drag vector.
5.0 Hull Design & Construction
5.1 Design
Computer simulation was used to determine an internal deadrise
distribution over the running wetted length that would bring the hull lift
to drag ratio to a
predetermined level. Once the deadrise distribution was determined,
the internal running wetted surface was modeled using the software
package FastShip4.13.
The rest of the hull surfaces were then designed to meet both hydrostatic
displacement and functional requirements. With all of the
surfaces modeled,
offsets were taken from the hull sections and exported to AutoCAD to
produce design drawings for lofting.
5.2 Construction
With funding from Trinity Marine Group the UMSEBT participated in
and supervised construction of the hull at United States Marine,
Incorporated. The male
mold was fabricated with 3/4" plywood sections covered with 1/8"
Lewan plywood. The composite hull was constructed by bonding a single skin of
carbon fiber to each side of 1/2" Divinycell H60 foam core using Derakane
8084 vinylester resin as the bonding agent. Because of the inverted vee
design, it was impossible to pull the hull in a single part from the male mold.
Therefore,the decision was made to lay the foam core over the mold and
then bond the external layer of carbon fiber to the core. The mold then had to
be disassembled and removed from within the partially completed hull. With
the internal surface of the foam core exposed, the internal layer of carbon fiber could be
bonded in place. With the basic hull form complete, the hull was shipped
back to the University of Michigan and structural members and motor supports were
installed. The external surface was then finish-sanded, primed and painted.
6.0 Propeller Design & Fabrication
The propeller was custom designed using computer modeling software at
the University of Michigan. The final design was for a
five-bladed,twelve
inch diameter, fully cavitating, surface piercing propeller that would be
fifty percent submerged in the running condition. NuCon Marine
Systems Corporation
provided funding for and fabrication of the propeller from
T6061-T6 aluminum using five-axis milling technology.
7.0 Propulsion System and Machinery
7.1 Motor Selection
Preliminary power requirements guided the search for four to ten
horsepower motors. Initial race specifications required that the motors
operate on a system
voltage not to exceed twenty-four volts. Cost and high
amperage requirements eliminated the use of brushless DC motors and
narrowed the search to permanent
magnet DC motors. Under advisement from Gabriel Marine, the team selected
two Advanced DC A-89 series wound traction motors for their capability of
handling the high current loads expected in the sprint race.
7.2 Thrust Vectoring
A surfacing propeller produces significant lateral forces because only
half of the blades are submerged, and thus experience side forces in the
running condition.
This produces a turning moment acting on the hull. In order to eliminate
this turning moment the net lateral force on the system must be zeroed.
This can be
achieved by laterally canting the shaft relative to the centerline of the
hull to produce a zero net turning moment.
7.3 Adjustable External Shaft Support
Uncertainties exist in the prediction of propeller placement necessary
for both zeroing the turning moment and achieving fifty percent propeller
submergence
in the running condition. Therefore both the horizontal and vertical
position of the propeller had to be adjustable. These requirements lead
to the design
and fabrication of a propeller positioning system consisting of a
universal joint and an external thrust bearing constrained by five
adjustable tie-rods. The universal
joint allowed the shaft to be canted without altering the position of the
internal shaft and motors. The thrust bearing consisted of two
Rulon flange bearings press
fit into a stainless steel collar. The collar was fabricated at Brighton
NC Machining and was held in place by the five adjustable tie-rods which
transmitted the
thrust from the thrust bearing to the hull. This in turn kept the thrust
loads from reaching the universal joint.
7.4 Gearing
The two Advanced DC A-89 motors were connected to the shaft via two
rollerchains and sprockets. The propeller was designed to operate at 1800
RPM in
the running condition. The roller chain/sprocket combination was used to
allow forgearing changes based on propeller RPM during trial runs. The
design gear ratio was 3:1 (motor/shaft).
8.0 Electrical System
All teams were required to remove their solar arrays during the sprint
event due to the short duration of the races. However, the Solar Splash
'94 rules stipulated
that all battery recharging must be accomplished with solar panels. Since
we decided to focus the sprint race and had minimal resources, the
team decided not to
invest in the design and fabrication of a costly, high quality solar
array. In lieu of such an extravagant system, we decided to purchase
two off-the-shelf solar
panels that would produce enough power to top off the batteries after
each sprint race.
The batteries for all of the teams competing in Solar Splash '94 were
donated by Johnson Controls. Two types of 12 VDC batteries were made
available to
the teams, one with an 81 amp-hour rating and the other with a 51 amp-hour
rating. Both types of batterie s had a maximum discharge capacity of one
thousand amperes.
Due to the high current loads in the high power system, 2/0 cable was
used to minimize the power loss through the cables. The rules called for a
high
power battery disconnect in line with a separate high power motor
disconnect. The motor disconnect consist ed of four high power contactors
connected in parallel. To maximize power output for the sprint race, the control system
simply consisted of a spring loaded on/off switch which engaged the
contactors. However, for testing purposes and the endurance race a variable
speed control was desired. This was accomplished by installing two
DC motor
controllers which were both operated from a single spring loaded
potentiometer. Ford Motor Company financed the purchase of the motors,
motor controllers,
and associated electrical equipment from Electric Vehicles of America.
9.0 Operational Phases
9.1 Testing
In order to establish the boat as a safe and stable platform, it was
towed behind a ski-boat to test planing stability and maneuvering prior to
the installation of the
motors and electrical system. Once the boat was deemed seaworthy, the
motors were installed and fully operational sea trials commenced. After
making several
adjustments to the propeller position and removing part of the deck to
reduce weight, the boat was not performing up to expectations. The team
was unable to get
the boat planing prior to departing for Solar Splash '94 and the top
speed attained was under ten miles per hour.
9.2 Racing
It was felt that the propeller was not developing enough thrust at low
speed and that lowering the propeller should develop enough increased
thrust to get the boat
up on plane. Despite further modifications to the propeller position,the
boat's performance was not improved for race qualifying. However, it
was established
during qualifying that the boat was more than maneuverable enough under
sufficient forward speed to complete the qualifying course and meet
the minimum
turning radius requirement of forty meters.
The team did not feel that the Vee-N-Verse's performance in the
qualifying round would be competitive in the sprint event. The boat was
not performing upto
our design goal, which was to plane. For this reason the team decided
to make two major changes in the propulsion system with hopes of increasing
the performance. However, there would be no time to test the new
configuration prior to the first heat in the sprint event, which made the
team's decision a risky
one. The first major alteration in the propulsion system was to
replace our two 12 VDC 81 amp-hour batteries with three 12 VDC 51 amp-hour
batteries. This alteration was approved by the race committee after another team
was allowed to boost their system voltage to 96 volts by means of a
DC-DC converter. The
second major change to the propulsion system was to modify the sprockets
on both the shaft and the motors to give the system a 1:1 gear ratio. This
allowed the
motors to operate at the design RPM on the new system voltage of 36 VDC
with an increase in power. Despite these drastic adjustments
the Vee-N-Verse was
still not able to plane in the first round of the sprint competition.
Before the second round, which was to be the first elimination round, the
team made one final adjustment to the propeller location by lowering it about
two inches in an attempt to improve performance. The increased thrust at
low speed due to this adjustment was enough to vault Vee-N-Verse up onto
plane for
the first time. This cut our elapsed time from seventy-four seconds in the
first round to thirty-nine seconds in the second round. This improved time was
good enough to place us in the semifinals where we ran the course in
thirty-three seconds and advanced to the final round. In the final round the
Vee-N-Verse repeated her semifinal time of thirty-three seconds which
secured third place overall in the sprint competition for the UMSEBT.
Unfortunately, the Vee-N-Verse suffered mechanical failure of a universal
joint during the sprint race finals. For fear of suffering irreparable
damage to theboat
and endangering the pilot's safety, the UMSEBT was forced to withdraw
the Vee-N-Verse from the endurance competition. After a successful weekend
of racing
the UMSEBT returned to Ann Arbor, Michigan having earned the
Mechanical Technical Achievement Award and Third Place overall in the
Sprint Event ofSolar
Splash '94.
9.3 Post Race Improvements
Although the Vee-N-Verse performed exceptionally well under the
harsh conditions encountered at Solar Splash '94, it was felt that the
boat's performance could
still be improved. The team decided to attempt to reach the design speed
of twenty miles per hour . The boat was refitted with a needle-bearing
universal joint that
would provide smoother operation, and taken back to North Lake for final
testing. During this testing the Vee-N-Verse exceeded the design goal by
attaining a
top speed of twenty-one miles per hour and handling exceptionally well
during rigorous maneuvering tests. The testing also allowed the three
members of the
team who designed and constructed the boat, James Criner, Greg Beers, and
Bryan Johnson to each experience the thrill of piloting the Vee-N-Verse.
10.0 Future Improvements
The UMSEBT has compiled a list of potential improvements to the
existing Vee-N-Verse. One of the simplest yet critical improvements is to
realign the motors
and internal shaft to reduce the angle on the universal joint. This will
increase the torque bearing capacity of the universal joint and
will reduce mechanical losses.
Another important modification is to strip and refinish the exterior
surface of the hull. Doing this under strict quality control will improve
both the boat's
appearance and reduce the weight associated with excess paint. Further
reduction of the overall boat weight will allow the Vee-N-Verse to attain
higher speeds
and quicker acceleration. To achieve this the boat will undergo a weight
reduction program which ranges from cutting limber holes in many of the
outfitting
components to removing any unnecessary structure. Reducing the overall
weight by as little as ten pounds has been shown to have a significant
impact on the
boat's performance. A substantial improvement to the drive system will be
to use timing belts to transmit power in lieu of the existing roller
chains and sprockets.
This will reduce mechanical losses associated with the gearing and lessen
the vibrations in the system.
The electrical system needs to be completely revised to make better use
of the energy available in the batteries. This should also reduce the
amount of copper cable
used in the existing electrical system which will in turn save weight. The
team also plans to design, construct, and employ a high quality solar
array for use in the
1995 Solar Splash. This is an expensive and time consuming activity that
will require significant sponsorship. The goal is to develop an array
which could
provide enough power to al low the Vee-N-Verse to be competitive in the
endurance event without drastic hull modifications. Also, the team will
need to design
and fabricate a second propeller for use in the endurance event. The
existing surface piercing propeller, while providing excellent
performance in the design
condition, is not intended to operate at low speeds.
An eventual goal for Vee-N-Verse is to have a computer controlled
propellerpositioning system. This system will move the propeller from an
optimum position
for accelerations from a standing start to an optimum position for
the running condition. If this goal is not attainable for Solar Splash
'95, the existing propeller
positioning system still must be improved because it can be cumbersome to
work with.
11.0 Conclusion
Throughout this discussion we have touched on the rationale behind most
of the decisions and actions made by the UMSEBT during the course of the
design
and construction of Vee-N-Verse. While a complete discussion of the
rational engineering used to produce Vee-N-Verse is beyond the scope of
this discussion,
the validation of such an approach is not. In section 4.0 Design
Direction, it was stated that the team's goal for Vee-N-Verse was to
successfully achieve a
top planing speed of twenty miles per hour. After six months of hard work
and generous sponsorship, the UMSEBT exceeded that goal by attaining a top
planing speed of twenty-one miles per hour. With only six months to
complete the project, the UMSEBT was not afforded the luxury of design
iteration
and testing. It is our contention that the success of this project was a
direct result of the persistent use of rational engineering.
12.0 Acknowledgments
The University of Michigan Solar/Electric Boat Team would like to take
this opportunity to recognize the following companies and individuals for
their generous
sponsorship. Without your support, our dream could never have become a
reality.