Human Powered Submarine: Control Systems
Human Powered Submarine
Controls Systems
Team 2007-2008
Jeff Poirier
Corie Drake
Matt Ledoux
Amy Wing
Dan Getchell
Chris Briggeman
Introduction
Tow Cap
Replacing “Stuff Box”
Fly By Wire
Tow Tank
Bow Planes
Manual Controls
Window Replacement
Conclusion
Introduction
This is a brief summary of the progress made by the Human Powered Submarine: Control Systems group during the Fall 2007 semester. It will cover all of the accomplishments made thus far in the design process. Also mentioned will be goals for the following semester.
Tow Cap
The tow cap was designed for towing the submarine through the water without causing damage to the nose of the sub. The first step in this process was to figure out a way to get the measurements of the inside of the nose of the submarine. The solution was to line the inside of the nose with plastic and fill it with Great Stuff expanding insulation foam. This created a template from which the final tow cap was cast. A rubber mould was created which was then used to form the final cap which was made of plastic. The cap was then sanded and Bondo was applied to smooth out the caps surface. A large Eye-Bolt was used as the tow point. The force in the bolt is distributed to the cone by an aluminum plate beneath a layer of Bondo.
For stress/strain analysis see Appendix A & Appendix B (sidebar)
Replacing “Stuff Box”
One of the first problems addressed was the use of the so called “Stuff Box”. This box initially contained all four actuators as well as the entire control circuitry and battery. It was determined early on that the giant block of aluminum housed in the tail of the sub was excessive. It was decided that the box would be removed. The linear actuator group then took over the bulk of this design problem by making individual boxes for each actuator. In the final design of the sub, there will also be a small box of either Plexiglas or plastic located in the tail to house the electrical components. This will be completed early next semester.
Fly By Wire
One of the biggest problems facing the design team this semester was the Fly-by-Wire control system. This is intended to be the primary mechanism for steering the sub, so it is critical that it works correctly. Thus far, progress has been slower than expected in programming and integrating with the submarine, though it appears that things are beginning to go more quickly.
One of the most important advancements in this section of the design process is the implementation of a joystick. The joystick will be the simple mechanism used to steer the sub. The joystick itself was salvaged from an old PlayStation 2 controller and appears to fit design specifications perfectly. The main requirement of the joystick was that it be analog. Since an analog joystick is essentially two potentiometers mounted perpendicular to one another, it will allow the pilot of the submarine to make turns of varying degrees. That is, the amount of turn of the rudder or elevator is directly related to the amount of movement of the joystick, as opposed to simply have a stick that sends left/right up/down commands that do not vary in magnitude.
Below is a schematic showing how the joystick is wired on the PlayStation controller.
Figure 2 Joystick as wired on PlayStation 2 Controller
Once the joystick was disconnected from the PlayStation controller it was prepared to be hard wired to the control circuitry.
A problem that comes up with the controls of the submarine being analog is that the program Basic Stamp will not be able to read it. It can only take inputs of a certain degree and not varying. To bypass this problem we adapted a capacitor to fill with electrons and discharge through a resistor network that consisted of the joystick and another additional resistor. Since varying the angle on the joystick varies the resistance of the circuit, at different angles of the joystick, the circuit has a different resistance. To measure the difference in resistance (or joystick angle) the program measures the amount of time it takes the capacitor to discharge all it’s built up electricity.
The amount of time it takes for the capacitor to discharge is equal to the total resistance of the circuit times the size of the capacitor.
Where
T= time
R= resistance
C= capacitance
Since any given angle of the joystick gives a certain circuit resistance, the angle of the joystick is now a function of the time it takes the capacitor to discharge. This can then be scaled to the needed output of the servos for any given joystick movement. Below is the schematic of the capacitor resistance circuit of which joystick angle is the only variable.
The figure below shows a basic wiring setup of the fly by wire control chip.
Tow Tank
In order to gain an understanding of how the sub will react in operation, it will need to be tested in the tow tank. To gain access to the tank, confined space entry training as well as lock out/tag out training must be obtained.
Bow Planes
Bow planes are used to help stabilize the trim of the submarine. This is done by forcing the nose down, countering the buoyant force which was a large problem in previous years. The bow planes also remove some of the burden on the rear control surfaces regarding trim, freeing the control planes to focus on maneuvers. The design of the bow planes is such that they are fixed, yet they are allowed to move slightlyto adjust for variations in speed. Though they are not adjustable on the fly, they are easily adjusted between runs.
Manual Controls
The manual control system was also a large problem for the design team. The original controls were thrown together when the electronic controls failed in a previous competition. This system needed to be of the same caliber as the electronic control system so in the event of an electronic failure, the pilot could rely on the manual controls to perform with the same level of certainty.
The manual controls consist of throttle cables which will be controlled by twisting a handle which in turn twists the control surface. Some key modifications made this semester are the moving of the control handles more toward the front of the sub as well as manufacturing new handles which will not get caught on the pilot’s gloves.
To move the handles forward, the mounting brackets were taken off and reattached further toward the front of the sub. This allowed for more room for the driver to control the submarine with more access and maneuverability.
The new attachments for the manual steering controls were made to be much thicker so that the screw attached to throttle cables could be countersunk. Countersinking the screws allows for smooth operation of the handles, without interference with the driver’s hands. The new attachments also were designed with more precision as the old ones were not up to the standards required.
Figure 6 Newly mounted control
mounting brackets
The last step in finishing the manual controls is to attach them to the control surfaces. This is where there is some overlap between the control group and the linear actuator group. Because of this, a joint effort is being made in determining the best way to combine each groups components. It has been decided that the manual controls will not be able to be engaged “on the fly” but instead will need to be engaged in the event of an electronic system failure by manually connecting the controls to the control planes. Further development of this will continue next semester.
Window Replacement
Upon inspection of the submarine at the beginning of the semester it was noted that the lower window was cracked and needed replacement. The cracks and non-unification of the window could result in more drag when propelling through the water which would ultimately slow down our top speed. This window was replaced with plexi-glass and sealed on the inside and out using clear caulking.
Conclusion
There has been a substantial amount of work completed on this project this semester. Although things were slow at the beginning they have picked up as the semester begins to come to a close. There were many bumps in the road as there was already substantial work done on the sub at the start. Repairs and modifications took up a large amount of time as did the planning and group training activities that were required. The design team is now in full swing and it seems the bumps are all out of the road. Next semester the team will be ready to work at a steady pace and begin completing milestones immediately.
1