Reseal pressure tests

The same IBP test configuration used to measure the bubble point is the ideal test configuration for measuring the reseal pressure for a given screen. Each controlled bubble breakthrough test allows for a controlled reseal pressure test. Details of the inverted reseal pressure test are reserved for Chapter 4 along with the IBP test configuration. 

The purpose of this chapter is to present the LAD performance experiments carried out in room temperature liquids. Bubble point and reseal pressure tests for a 325x2300, 450 X 2750, and 510 x 3600 Dutch Twill screen are conducted in storable liquids, methanol, acetone, IPA, water, and binary methanol/water mixtures of various methanol concentrations. First screen pore diameters are estimated based on analysis from scanning electron microscopy and historical data. Experimental results are used to compare methods for determining effective pore diameter. Next, contact angles are measured for both pure and binary mixture fluids using a modified version of the Sessile Drop technique. Then, the equation of state analysis from Neumann and Good (1979) is used to determine the critical Zisman surface tension for stainless steel LAD screens, which... 

Initially, the as-received pump could not pass a 13.8-bar (200 psig) static pressure test at ambient temperature. Consequently, the pump was returned to the manufacturer for extensive seal modifications. Upon receiving the modified pump, the unit was placed into service for a hot shakedown test with C30 oil at 175 psig pressure under the present experimental setup. The modified seal continued to leak at a small but acceptable rate on the order of 1 g/min. However, leak problems were encountered in the threaded seal of the stator and the pump housing. The stator was resealed with a thick Teflon tape and a high-temperature pipe sealant that initially slowed the total system leaks to less than 50 g/h. 

Bubble point tests were then repeated numerous times at identical conditions to ensure quality results, repeatability, and consistency and to null out the possibility of defects by ensuring bubbles would break through various locations on the screen. A sample test run is depicted in Figure 4.4 where the pressure differential at break through and at reseal is superimposed on the raw DPT signal. Note that the reseal pressure is always lower than the breakthrough pressure. 

Room temperature reseal pressure data was collected simultaneously with bubble point data. The same inverted bubble point test configuration used to measure the bubble point was used to measure reseal pressures. A bubble point test ceases once pressurant gas breaks through the wetted pore. To commence a reseal point test, the pressurant gas flow rate beneath the screen is slowly reduced in fixed quasi-static increments to slowly encroach upon the differential pressure across the screen at which the screen reseals. Eventually the screen rewets/reseals itself as evident in visualization of no more bubbles... 

Figure 4.4 illustrates this effect for a room temperature bubble point and reseal test where the corrected reseal pressure is superimposed on the raw DPT signal. As shown, the reseal pressure is always lower than the bubble point pressure. Unlike bubble point data reduction, the DPT across the screen carmot alone be used to determine screen reseal. Rather, time synchronization with the visualization system is required to determine the exact differential pressure across the screen at reseal. Sole reliance on visualization makes reseal point data inherently noisier than bubble point data. 

Trends in room temperature reseal pressure data mirrors trends in bubble point pressure data. All reseal pressures collected here are about 90% of the corresponding bubble point values. Operationally, this implies that only a ->10% reduction in differential pressure across the screen is required to reseal the screen and prolong the point of total LAD failure to yield a higher overall expulsion efficiency. Wicking rate test results performed in IPA align nicely with historical trends as coarser meshes outperform finer meshes. 

The simplified reseal pressure model presented in Chapter 3 predicts reseal pressures well when reseal diameters are based on reference fluid tests (Method 1) or through summation of historical data (Method 2) ... 

As mentioned in Section 3.9.2, there are only three previous studies where reseal data was reported. Reseal pressure data was collected alongside all bubble point test data from both room temperature as well as cryogenic bubble point tests for a 200 x 1400, 325 x 2300, 450x2750, and 510x3600 Dutch Twill LAD screen in IPA, methanol, acetone, water, LH2, LN2, LQX, and LCH4. A total of 4836 reseal pressure data points, of which 4815 were new points, were collected, processed, and analyzed to develop this model. 

The functionality of relief lutes should be tested by raising the system pressure and checking the actual relief pressure against calibrated instrumentation. The ability to reseal should also be checked. 

Additional outflow tests should be conducted in a room temperature fluid, such as water, in order to quantify frictional and dynamic pressure losses that were difficult to measure in cryogenic liquids. These tests could also be used to validate the analytical LAD channel solver and sizing/design methodology. With carefully designed LAD channels, this rig could also be used to easily quantify pressure transients. The rig could also be modified to further investigate the phenomena of screen reseal after initial breakdown. 

---