Effect of Water and Humidity on Hypergolic Propellant Ignition Delay
Download 0.9 Mb. View original pdf
|
- Navigate this page:
- Table 2. Hypertester test matrix
| B. Hypertester Results Each condition was repeated three times with all instrumentation recording properly. Due to operational issues, certain conditions were tested more than three times, but not all instrumentation successfully recorded for the additional tests. Case #3 test did not record properly on the data acquisition systems, but was not noticed until much later so only two tests at this condition recorded on all systems. Fifty-seven drop-on-pool tests were completed and the test matrix is shown in Table 2. Table 2. Hypertester test matrix Case Fuel Water Concentration in WFNA [%] Relative Humidity [%] Fuel Aging Conditions Total Tests Completed 1 TEAB 0 low not aged 11 2 TEAB 10 low not aged 4 3 TEAB 0 high not aged 3 4 TEAB 10 high not aged 11 5 MMH 0 low not aged 3 6 MMH 10 low not aged 4 7 MMH 0 high not aged 4 8 MMH 10 high not aged 3 9 TEAB 0 low 2 days in humidity air 3 10 TEAB 0 low 4 days in humidity air 3 11 TEAB 0 low 7 days in dry air 5 12 TEAB 0 low 14 days in dry nitrogen 3 Each testis analyzed below for significant variables and a statistical analysis is used to determine any strong trend. Trends are only discussed if the p-value for the linear regression coefficient is less than 0.20 or greater than 0.75; otherwise results are considered inconclusive. 1. Single Test Analysis To determine the moment of drop impact, three different measurements are used. The moment of contact on video is selected as the first frame where the reflected light of the drop changes due to the surface interaction with the pool. For the piezo sensor, contact is the first data point after the trigger which is 50% higher than the noise level. For the microphone the same technique is used, but instead the signal threshold is 75% over the noise level. The noise level is the maximum value minus the mean value for the first few seconds of data recording. The values of 50% and 75% are chosen to give consistent response times. The video contact frame is used primarily for the exact contact time. For tests where the video camera did not record, the impact sensor data is preferred. In the eight tests where both the video and impact sensor recorded, the video is, on average, slower by 48 s with a standard deviation of 114 s. The burning propellant frequently burnt and corroded the piezo sensor leads so, given its limited usefulness, the sensor was removed for the last 27 tests. The microphone also records a signal when impact occurs, but it lags behind the video by 866 son average with a standard deviation of 654 s. Video is also used to verify whether ignition occurs in each test. To determine the moment of ignition on the video, relevant frames are inspected for noticeable chemiluminescence in the visible spectrum. The frames are then backtracked to the first indication of chemiluminescence and that frame is recorded as the IDT. The video frames from a case #1 test are shown in Figure 5. Downloaded by PURDUE UNIVERSITY on July 21, 2017 | http://arc.aiaa.org | DOI: 10.2514/6.2015-3867 |