Saturday, May 30, 2009, 09:49 PM
After a long day of plumbing and computer troubles, we ended the day with a 2 second test firing of the rocket motor - reaching 1700 lbs of thrust!
We began the day with several water flow tests to measure the alcohol system pressure drop with the new high pressure helium valve. The new data matched the predictions well and provided a reasonable estimate for the LOX side pressure drop with the new helium valve without performing a water test on that system.
After the water flow test, we performed a cold flow test of the new LOX plumbing and level monitoring system: Using liquid nitrogen to simulate liquid oxygen we loaded the LOX tank half full (identified using the new liquid level monitoring system) and let the system sit for ~30 minutes to allow ice to form. Good results all around no ice on the foam and no ice downstream of the main LOX valve. At this point, we performed a complete hot fire simulation using the liquid nitrogen in the LOX system and water in the fuel system, demonstrating the correct operation of the rocket motor and computer control system. During this test, we observed nitrogen leaking past the main LOX valve. This leak was unacceptable and we spent the next two hours diagnosing and repairing the valve (needed a new teflon shim seal).
After a few more hours of working through issues like the leaky valve, we filled the alcohol tank to the top and the LOX tank to the 25% fill mark (using the new level monitoring system). At last we were ready to test fire the motor. The first countdown occurred around 8:30 pm but the countdown aborted due to lack of prime ignition! We spent the next ~ 30 minutes pouring through the test data but could not determine why ignition did not occur. Hence, we pressurized the system and tried the countdown again.
This time ignition occurred at the correct time during the countdown. The motor ran for two seconds before the liquid oxygen ran out, reaching ~1700 pounds of thrust:
| permalink
Saturday, May 2, 2009, 08:06 PM
The propulsion team started the day by epoxying the thermocouple patch heaters around the six LOX tank level sensing thermocouples. After a couple of rounds of testing, urethane foam was applied to the tank over the sensors.
Later in the day, the team installed the hopeful fix for our ignitor problems a pre-mixer for the propane and compressed air. This pre-mixer is located just upstream of the existing ignitor in the port previously occupied by a chamber pressure sensor and thermocouple. The benefit of the pre-mixer was immediately obvious smooth runs of the ignitor over a much wider range of mixture ratios. With this success, the chamber pressure and ignitor temperature sensors were relocated and the ignition system work was postponed until further combustion testing could be performed in a better location.
In addition to this fix to the ignition system, two backup ignition methods have been identified and will be ready on test day in case they are needed: paper soaked in alcohol placed inside the combustion chamber to burn with the ignition system, and a remotely-ignited road flare to replace the ignition system.
The computer team performed several complete checks of the new software and rewired sensors today, a great milestone for their hard work.
We ended the day with several countdowns and aborts of the integrated propulsion/computer control system. Were in good shape for the upcoming test firings!
| permalink
Saturday, April 25, 2009, 06:22 PM
The goal behind sensing the fluid level in the liquid oxygen tank is to reduce the time required to fill the liquid oxygen tank and to reduce the quantity of LOX wasted for a given test (i.e. just load the quantity of LOX needed for that test firing).
Today the computer and propulsion teams started the day by performing some fluid level experiments with external temperature sensors. A simulated tank was created using 2 cast iron piping, sealed at one end. The tank was insulated using several layers of pipe wrap, and a few thermocouples were mounted to the exterior of the tank, under the insulation. The test fluid was alcohol, chilled to -100 deg F using dry ice. When the tank was filled with cryogenic alcohol to various levels, the thermocouples quickly responded those next to fluid indicated very cold temperatures, whereas those next to an empty part of the tank indicated a much warmer temperature. Using this temperature difference, the level of alcohol in the tank could be readily determined, even if the tank was fully chilled and then part of the fluid was poured out (though not as fast, in this case).
The team was concerned, though, that the significant thermal sink of the liquid oxygen might make the results more difficult to determine in the actual application. So, small patch heaters were made using the nichrome wire from a home heating pad. The heaters ringed each thermocouple and when activated, the sensors next to fluid responded much differently than those next to an empty tank: the temperature indication of the sensors next to fluid did not rise as high as those next to gas, and when the heater power was turned off, the temperature of the sensors next to fluid decayed much faster than those next to gas.
Based on the success of this test and the simplicity of the sensors and heaters, this method was chosen for the LOX tank.
By the end of the day, a strip of foam insulation on the LOX tank was removed and the surface was prepped for the lox level sensing system.
After the students left for the day, the LOX tank was taken to a safe area and six thermocouples were welded to the body of the LOX tank at the 5, 20, 25, 30, 50, and 75% full locations.
| permalink
Next
