Progress Report - Mechanical Design (01/24/2012)

     The development of the vehicle has continued to progress. After several hours of research and speaking with salespeople from a couple suppliers of paintball tanks, regulators, etc., it was determined that compressed air should be used instead of CO2 for several reasons:

(1) The outlet pressure of CO2 is a strong function of the operating temperature, and is thus difficult to predict/regulate. In fact, paintball CO2 tanks don't come with a regulator, while all paintball compressed air tanks come with a regulator already installed as a standard. Apparently the ability to regulate compressed air much better than CO2 is the reason why higher-quality paintball guns and more experienced paintball players use compressed air instead of CO2.

(2) Originally it was thought that CO2 tanks would be far more compact than compressed air tanks, however, this is not necessarily the case. There are compressed air tanks that are just as compact as compressed CO2 tanks (as I have found by shopping online). This discovery eliminates the issue of dimensional constraints.

(3) After a few phone calls, I was able speak to a supplier in northern California who had a very versatile selection of regulators, adapters, etc. In fact, they had a low-pressure regulator that can be directly attached to the tank regulator, and that also has the exact desired threaded outlet (i.e. 1/8" FNPT), thus culminating in a compact, efficient, reliable completion to the gas distribution circuit.

*Note: The compressed air tank (with the tank regulator) and the low pressure regulator (both the tank regulator and the low-pressure regulator (required to drop the pressure down from about 850 psi to 100 psi) have analog gauges installed on them) have been ordered and will arrive shortly.

Furthermore, I developed an efficient design for the compressed air tank mount, which will be machined using the lathe and drill press tomorrow morning (relatively easy to manufacture, and the stock has already been selected).

In addition, a mounting plate for the pressure transducer has been water cut, and I picked up the one-way check valves from Grainger and successfully installed/tested them on the exhaust outlets of the two torpedo launcher air cylinders, as well as on the outlet of the exit solenoid valve for the grasp/release mechanism.

Tomorrow afternoon (when the McMaster-Carr order arrives), I will carry out my improvised modification to the camera enclosures and perform another water tight test (hopefully my strategy will work--I am optimistic as it seems like a good approach). This strategy involves adhering 3/8" thick EPDM rubber (i.e. the same material used for the gaskets) to the bottom surface of the top camera enclosure, and the top surface of the bottom camera enclosure, allowing the neoprene sealing washers, bolts, and threaded rod to pass through clearance holes. The EPDM rubber will be adhered to the aforementioned surfaces using the gasket adhesive (very strong). Thereafter, I will cleanly apply caulking around the four edges of the exposed surface of each of the two rubber pads. Then, I will pour epoxy into each of the six total exposed holes. This should prevent any leakage through the bolt threads, sealing washers, etc., as was the issue earlier. Furthermore, an appropriate tape (specifically for smooth surfaces such as acrylic) was purchased in a recent order. This tape will be placed along the outer edges of the enclosures in a clean fashion such that all the caulked edges will have this extra layer of protection. The final outcome should hopefully be more reliable (i.e. water tight), aesthetically-appealing camera enclosures.

I also made another (hopefully final) McMaster-Carr order this evening, which includes an assortment of items such as another NPT adapter, a liquid rubber wire-protecting agent, heat shrink tubing (with an internal adhesive to assist in preventing water from coming into contact with the soldered wires), a 5/16" thick sheet of acrylic (to be used for the hydrophones mount, as well as to complete the pressure transducer mount), and a few small neodymium magnets which I hope to use on the revised versions of the torpedoes which were 3D printed late this afternoon, and will be retrieved tomorrow morning for surgery (i.e. I need to drill into and install a stainless steel rod at a specific depth--similar to the initial version). The small neodymium magnets were chosen for their compactness due to their high magnetic field/volume ratio, and should provide an even cleaner design than with the bar magnets. Also, this slight revision is being done because I forgot to factor in the addition of the magnet when performing my calculations, and thus the center of buoyancy deviated from the original center of mass, thus yielding torpedoes that tilted back slightly when submerged underwater. This negatively affected their hydrodynamics, and after removing the magnet from one of the torpedoes, the density was perfect, the center of mass and buoyancy were great, and the torpedo moved nicely through the water. The necessary adjustments are being made to the second generation of torpedoes.

I also discovered on Monday that the SEACON connectors will arrive in the first week of February as the order was unfortunately not completely processed until January 6th (despite being submitted prior to winter break). This is okay, though, as there are still other productive tasks that can be completed until then. In addition, the pressure transducer will be ordered on Friday (W-9 forms were not supplied by the company, so the order had to be cancelled, and it will need to instead be ordered out of pocket), and two more SQ26 hydrophones will be purchased by the end of the week (the two from last year were determined to work fine for the frequency range required, and I spoke to branch of the company in Seattle, where I was provided a quote and also made aware of the availability of these hydrophones.

Furthermore, the solenoid valves were successfully tested today, and my proposed circuit will work. I will have Hang help me make a PCB containing four of these identical circuits shortly.

                       -Eric Sloan (ME Project Manager - AUVSI Robosub Competition)

Progress Report - Mechanical Design (01/18/2012)

The vehicle will undergo a watertight test this afternoon. If there is any leakage in the hull or any of the two camera enclosures, more silicone caulking will be added. Hopefully this will not be necessary, however, so that we can hand the enclosures over to Ryan for testing with the web cameras inside. Also, new mounting plates were made for the bottom thruster and the marker dropper mechanism. The heads of the screws for the bottom acrylic supports were intersecting into the locations of the clearance holes in the original thruster mounting plate. The mounting holes were simply migrated inward toward the center in the new mounting plate--a simple adjustment. For the marker dropper mounting bracket, I simply cut an aluminum angle bracket to 1" x 1" x 3.75" dimensions, drilled a few holes, and attached it to the marker dropper and frame. Furthermore, I designed individual identical mounting plates for each of the four solenoid valves. I decided on revised locations to mount the solenoid valves, as well as the four-way split adapter for the gas lines. The new scheme will be very efficient (i.e. more direct), aesthetically appealing, and also distributes the mass of these valves symmetrically throughout the vehicle, thus maintaining proper balance. I also ordered a CO2 tank, a lightweight pressure regulator, and a check valve (if it works properly, I will purchase two more). So, the CO2 distribution system should be ready for testing toward the end of this week. Once the screws and T-slot nuts arrive (this evening), the solenoid valves will be mounted to their corresponding plates, the plates will be attached to the 80/20 frame, and I will start connecting the nylon gas lines (tubing) to the push-to-connect adapters. During this stage, I will also cut the tubing to size so that there won't be excess tubing hanging off the vehicle. When the CO2 tank and pressure regulator arrive, the network can be completed and preliminary tests can ensue. Thereafter, I will make interface circuits (low side drive BJT) to enable easy actuation of the solenoid valves with a micro controller. These tests will likely be done with a Dragon Board due to familiarity. The SEACON connectors should arrive shortly as well. Once they arrive, they will be installed in the inner end caps of both the hull and camera enclosures. The four SBT150 thrusters can then be potted using epoxy, and the leads of all the thrusters can be carefully soldered to the corresponding leads of the male cables. The interface will be protected using shrink wrap at a minimum.

                                -Eric Sloan (ME Project Manager - AUVSI Robosub Competition)

Progress Report - Mechanical Design (01/14/2012)

The aluminum hull end caps were completed with the machining of the slots and handles. I wanted to get them hard anodized and dyed blue (for aesthetic reasons), but the nearest facility, which is located in Valdosta, GA, was under heavy work load, so I decided to go ahead and install both the camera enclosure and hull end caps as they were. The hull end caps were carefully aligned so that the ends of the thermally-conductive aluminum platform which rests on the interior acrylic supports would properly fit into the end caps slots. The torpedo launchers were also installed, including a revised, more efficient mount design, and custom attachments to hold the acrylic torpedo cannons. The machinist, Jeremy, made a couple suggestions to help develop this clean design. Furthermore, I weighed each of the two ABS plastic torpedoes on a highly accurate scale in one of the chemical engineering laboratories. This, in combination with the Pro/Engineer-indicated volume of each of the 3D-printed torpedoes enabled me to determine their density. With this information, I was able to perform a couple detailed calculations in order to determine the proper depth to drill into the bottom of each of the torpedoes so that I could install a dense 3/16"-diameter stainless steel rod into the center of each of the torpedoes without changing the original center of mass location. After reaming out this hole and press fitting the rods into each of the torpedoes, the back end was filled with epoxy. This process allowed me to adjust the density of the torpedoes to the desired density of salt water, while simultaneously adding mass, inertia, and stability. Thereafter, I spray-painted the torpedoes completely white, and then laser-cut magnets to adhere to the bottom surface of each of the torpedoes, as well as to the cylindrical disk attachment on the torpedo launcher CO2 cylinder piston. These magnets will serve to prevent the torpedoes from prematurely dislodging due to disturbance forces, vibrations, etc. during the mission. The camera enclosures were also completed and attached to the frame. I came up with the idea of installing a fully-threaded rod through the center of each of the two external camera enclosures to make use of the tripod mounts on the Logitech C615 web cameras. They can simply screw into the ends of the rod, and then be tilted to orient toward the desired window face. Neoprene sealing washers were used at all locations where a screw was required on the inside of the enclosures (in order to attach the enclosures to the base, and also for the threaded rod). These will serve to maintain a reliable watertight seal. EPDM gaskets were also made at Capital Rubber, and should perform well. The hull, which rests on a bend of acrylic supports, was also installed in the center of the frame. The top acrylic supports were also screwed in to their respective complementary bottom supports in order to clamp the hull into place without introducing any severe stress concentrations. Furthermore, having already made the aluminum mounting plates for the thrusters, my teammates helped me screw the thrusters into these plates and attach the plates to the vehicle. I also ordered larger diameter, stronger, nylon gas lines, as well as compatible adapters and splitters for the CO2 distribution network. The solenoid valves were already purchased months ago, so once the CO2 tank and pressure regulator arrive (I ordered these on Friday), installation and testing of the CO2 distribution system can be performed. I will need to make a mounting plate for the solenoid valves, cut the gas lines to the proper sizes, develop and interface circuits to enable simple actuation of the solenoid valves via a microcontoller, and then write a program to allow me to actuate the solenoid valves individually, and ensure that the pressure-regulated CO2 is distributed as desired (e.g. to the torpedo launchers). The vehicle undergo a watertight test on Tuesday at the FSU Morcom Aquatics Center after an ME presentation. While I work with my teammates on the CO2 distribution system and develop mounts for both the CO2 tank and the solenoid valves, my ECE teammates will work on the computer vision (with the camera enclosures), thruster control, and hydrophones. Once the underwater SEACON connectors and pressure transducer arrive (likely next week), they will be installed on the vehicle. Holes will need to be drilled and threaded into the end caps of the hull in order to allow the female connectors to screw in (Loctite will be used as a thread sealant here). I will also need to solder (and shrink wrap) the electrical leads of the peripheral subsystems to the electrical leads of the corresponding male SEACON cables. I would like all the mechanical aspects of the vehicle complete by spring break so that the focus can be solely on programming, debugging, and developing electrical interface circuits. I also intend to eventually contribute in the development of depth and stability control algorithms, as well and velocity and turn control algorithms based on the integration of pressure sensor and inertial measurement unit. While I help the ECEs with programming, I will have Tra and Kashief work on developing the replicated PVC objects that the AUV will encounter during the actual mission so that we can test the program in a relatively mirrored environment. There is still a lot of work to do, but I will make sure we get it done.

**Also, pictures of the AUV should be up by the end of Monday

              -Eric Sloan (Mechanical Engineering Project Manager)