Sub-Pages

Pressurized Air Leak Testing

        The primary goal of testing the air delivery system was to find any leaks between the fittings and to verify that the solenoid valves could handle the CO2 pressure.  The pictures above are of the complete air system.  The components consist of a 12oz CO2 canister, four solenoid valves, an assortment of brass flare fittings, and copper tubing.  The test was run by pressing the on/off valve on the CO2 canister to the on position.  Pushing in this valve allows the CO2 to exit the canister and flow through the system.  The team noticed leaks at the flared fittings connecting the components to one another.  After locating the leaks the team shut off and removed the CO2 canister from the air system.

    The team came up with two approaches to the problem, one being less complicated.  The first possible solution to the problem was to add a pressure regulator but the problem with this approach was the requirement to minimize weight and also the fact that there was no guarantee that the system was leaking because of excessive pressure.  The system was not only leaking at the flared fitting where it connected to the adjoining piece but there were also leaks coming from the backside of the flared fittings where they slide along the piping.  So adding a pressure regulator would not control the leaking at the fitting to tubing connection.  So the second option available to the team was to JB weld the connection between each fitting and also the connections to the piping.  The team chose this option because it was cheap, quick and added little weight, and provided a realistic solution to the problem.  This approach led the team to the realization that once the air system was JB welded together it would no longer be able to be disassembled.  Due to this complication the team discussed a plan of attack for assembling the submersible robot while still JB welding the fittings together.  The first step the team took was to run the tubing that runs between the two housings, through the holes drilled into the threaded end caps.  After completing this step the team inserted and glued in the barb fittings into the end caps.  Next the team ran the plastic hose from barb fitting to barb fitting which prevents water from entering into the housings but still allows for the tubing and wiring, to pass from housing to housing.  While holding the air system in the dry housing with one hand and threading the PVC housing into the end cap with the other hand the team had one side of the robot together.  The next step was to JB weld all the fittings in the wet portion of the wet/dry housing except for those fittings which run through the divider between the two sections.  The reason for this is because if the fittings connected to the divider are JB welded then the PVC housing cannot be threaded into the end cap because the divider has a larger diameter than the PVC housing.  So the plan of attack was to JB weld those fittings which were not connected to the divider, then the team threaded the second PVC housing into the threaded end cap in a similar many as was done for the other housing.  Next the team JB welded the fittings into the divider and inserted the divider into the coupler which joins the wet/dry sections together in the wet/dry PVC housing pipe.  After that it was a matter of gluing PVC pieces together and then connecting the tubing to the CO2 canister.  In conducting the initial test on the air system the team was able to decide what the best approach was to ensure the CO2 did not leak.  This decision lead the team in how the robot had to be assembled and required the air system to be permanently assembled but still allowed for access into the housing to remove other components as needed.         

          After assembly of the robot was completed and the air system was allowed to dry properly the team then retested the air supply system and found that the system withstood the pressure and that the leaks were fixed.  The team also found that the pressure from the carbon dioxide canister exceeded the solenoid valves maximum operating pressure.  While this was disappointing to the team it was an important step in making a successful submersible robot.  At the conclusion of the air system testing the team decided that a pressure regulator would be required and this is a recommendation to any future teams who may try to improve on the design that the team has come up with.