Analytic Essay

Our quest to build a completely original submarine began when we decided to become partners. As soon as we decided to join together as DEEP cohorts, we knew that we would be building a submarine that would be completely different and original from any sub built from an instruction book. We would use our minds and our knowledge to try and build something that had never before been designed. We ended up with a sub that looked a little like the Starship Enterprise from Star Trek. The first time we sat down together in class, we came up with our original designs, which included a main pod with two engine pods on either side. We decided upon an on board power system and, unlike the subs built using the instruction book, an on board ballast system. Our plan for the ballast system was to have two water chambers on either side of the battery that would automatically fill when the sub submerged. We then had an air hose, controlled presumably by a blowing into it, running out the back and connecting to the two water chambers. We then made our revised designs the next time we sat down for class. Though our engine pod, along with our electrical system, design remained nearly the same, the ballast system changed dramatically. We first decided to put it in the front of the sub and the battery, in an air-tight compartment, in the back. We also decided to use a balloon that we would inflate to push out the water instead of pumping the air directly into the water chamber. Finally, we decided to use a CO2 bicycle pump, which could operate underwater, to inflate the balloon instead of an air hose controlled by blowing air in.
The sub, which I have just explained to you, was all built thanks to hours of diligent work by the two of us and ready to experiment on as we took our trip to Long Beach. We placed the sub on a bleacher when we got there. We left it there until it was our turn to go in the water. However, we did not notice that underneath our sub was a backpack strap that led to a backpack. Someone picked up that backpack. The sub was flipped over and tumbled to the ground, a good 4 feet below. All my partner and I heard was a thunderous crash, and then everything went quiet. We rushed over and looked on in shock as both of our engine pods lay separated from the main body. This meant that our damaged sub was inoperable, for as soon as we would put it in the water, the motor compartments would have flooded, ruining them and our sub. But there was hope. For just one day later, we took the sub home and did the impossible. Using brilliant engineering techniques and sophisticated technology, we fixed the submarine by shortening the connectors between the engine pods and the main body, and then resealing the sub with hours of painstaking work. The sub was now operable and we tested it in the St. Matthew's pool, running a speed test, turning test, and buoyancy test.
The speed results showed our sub's motors and battery were quite powerful and were able to move the vehicle rapidly through the water. We were able to predict in our hypothesis surface speed fairly accurately and our submerged speed not quite as accurately. The vehicle moved nearly as fast submerged as it did on the surface despite underwater drag. The error analysis for the speed test showed that its speed was fairly consistent, as we only had a 2.83% error above water and a 4.94% error below water. This test, however, was not a completely accurate representation of the sub's speed because it consistently turned right during the tests. We believe this is because of the vehicle's starboard tilt.
The turning radius results showed that the sub is not the most agile vehicle. The sub would have probably been able to turn better before it was disabled, as the engine struts were longer. This is because the engine pods would have been further away from the body, causing more torque. Our hypothesis for the turning radius was well off because we had never actually tested this before the actual experiment. The sub, as you can see turned much better to the right than to the left. We believe, this is once again, due to the starboard tilt of the vehicle. The error for this test, 12.46% to the right and 5.26% to the left, was not so bad considering the crude measuring techniques. We measured this by running the submarine in a circle and having one person mark a point on the circle with a float, while another marked the opposite spot. We then measured the distance between the two points, which was the diameter, and cut it in half.
The buoyancy test showed that the sub is much faster at surfacing than submerging. We used compressed C02 to inflate the bladder and relied on the water pressure on the surface to deflate it. Therefore, the compressed gas was able to force itself into the bladder with much greater force than the water could force out. This manifested itself in the fact that the sub could surface from a depth of five feet much faster than it could sink to the same depth. Our large margins of error on the surfacing portion were largely due to the gas canisters. When the canister was new, the pressure it exerted was much greater than the pressure exerted by an already spent one. This affected the speed at which the bladder inflated. The huge margin of error for the submerging portion was due to the fact the sub's tank had to be filled with water before the bladder could deflate. The vehicle was not held at constant depths. The deeper it was held the faster the submerging speed due to the increased water pressure. The shallower it was held, the slower the submerging speed due to the decreased water pressure.
Scuba divers are, by nature explorers. They can be aided by automate underwater vehicles (AUV's) such as ours. However, one major difference between our sub and those used for scientific exploration is that our sub had a much, much more limited depth because of the silicon and epoxy seals on the engine pod and battery compartment. However, the vehicle was able to make it to 8.5 feet deep without rupturing any seals. Another influencing factor was the fact that our engines run hot. If they had run for too long a time, the air inside the pod would have heated to dangerous levels. When air molecules heat up, they expand. The expanding hot air could blow holes in the seal in the nose cone of the engine pod. In our preliminary construction tests, we learned this the hard way. We managed to overheat the pods but luckily surfaced the vehicle in time to salvage the motors.
The sub was primarily composed of an acrylic shell. We used acrylic instead of PVC because we wanted it to be clear. The advantages of having the sub be clear were we could see inside and ascertain any leaks or construction flaws. One problem with acrylic is that when holes are drilled in it, it weakens the surrounding plastic causing cracks that in time could leak. When disaster struck in Long Beach, we broke the struts that connected the pods to the main body. The fractures were behind the epoxy seals that would have allowed water to enter the engine compartment. If you look at our diagrams, you will see that the aft portion of the engine pods were supposed to be flooded, however, the engine compartment could not get any water in it or else the motors would short. To repair the extensive damage, we had to chop the struts shorter and remove the debris in the mounting holes. The end result was the pods were closer to the body, reducing the amount of available torque to turn the sub. We also had to add a pair of reinforcing girders across both pods. The repairs functioned optimally and the components managed to operate at nearly maximum efficiency. We would not have really modified the hull, but given time, we would have exchanged the acrylic struts for aluminum ones so that they would not break as easily.
The buoyancy system worked fairly well. When we first designed it, the buoyancy bladder was full length. This was a problem because the air would move around throughout it depending upon which end of the sub was downward and therefore which end of the bladder had more water pressure on it. Because of this, either one side of the sub or the other was higher most of the time, causing the sub to never be completely level underwater unless it were completely flooded. We then shortened the bladder to try and solve this problem by sticking a rod through the middle of the sub. This caused the bladder to take up less volume, therefore limiting the amount of air that could move around. This only marginally improved performance and we still had an almost impossible time trying to get the sub level and neutrally buoyant. In the future, if we had more time and resources, we would have had the buoyancy system consist of several compartments, all of which we could have controlled separately. This would have let us control the buoyancy of each section of the sub and let us get it pretty level underwater while it was neutrally buoyant. Also, instead of controlling the amount of water in each compartment with an air hose, which caused drag, we probably should have used two-way electronic valves and pumps, which could have forced water in and out of the ballast compartments. With these valves and pumps, we could have more easily controlled the weight and buoyancy of each ballast compartment and therefore the weight and buoyancy of the submarine.
Another problem we found at the beginning of the testing was that the back, which was not filled with water, was much lighter than the front. This caused the sub to always be tilted forward. We fixed this by adding weight to the back. We also added a little weight to the front of the sub to counter the effects of the air movement in the ballast bladder. Ultimately, to fix the problem of the uneven weight between the battery compartment and the front of the sub, we should have put the battery compartment, which was filled with air, not water, in the middle of the sub, with ballast compartments on either side.
The engine system could have been improved but, without a machine shop, there really isn't much we could do. The primary flaw was the fact the seals could be ruptured when the engines heated up. With a machine shop, we could have built the pods out of aluminum and make them in one piece. Other than that, the system was pretty efficient.
Under the electrical/control system, the improvement that we would make given time, would be to make the control remote. The current design includes wires that run to the control box. We would have like to eliminate those wires and install a transmitter and receiver in the control box and sub. In addition, we would have liked to make the diving planes powered instead of simply friction fit. Otherwise, electrical ad control systems were pretty effective.
One problem that doesn't really fit under any category was our starboard tilt. We are not sure exactly what caused this tilt, but it affected our turning and our ability to run the sub straight.
The scientific principles involved in our project basically had to do with buoyancy. We should have done some calculations on buoyancy comparing our sub's weight with the amount of water it displaced if the balloon was either fully inflated or deflated. We could possibly have made our sub more neutrally buoyant if we had found the amount of air we needed in the buoyancy bladder to have our sub's weight equal the weight of the water it displaced. However, we really did not have time to make these calculations.
Overall, our submarine worked fairly well. Despite setbacks which included heavy damage to our vehicle, we were able to fix the sub and run the tests we needed to run. I would qualify our project as a success.

Background Essay

 Problem Statement

 Method & Materials



Analytic Essay 









 Special Thanks