One existing method to quantify the gas loss from a closed system is the mass point leak rate method. This traditional empirical method is capable of quantifying the loss of a known type of gas from a volume of known size. Using this method, measurement devices quantify the gas pressure and temperature within a closed system throughout the duration of the test. At the onset of the test, the operator establishes boundary conditions to create a pressure differential across the test article that is higher than the pressure differential of interest. During the test, the pressure differential decreases due to leakage. When the operator subjectively determines that the desired pressure differential has been achieved and sufficient data has been collected, the test is stopped. Subsequently, the data analyst identifies a subset of the collected data to be used for mass loss computations. A typical computation utilizes a linear fit of the mass-time data set, wherein the slope of the line is the mass loss rate. It is common to use the largest data subset to minimize the measurement uncertainty; however, the data set must not be so large that the curve fit is nonlinear.
A newly developed process is similar to the traditional mass point leak rate technique, but solves a number of deficiencies. The primary modification is the addition of a gas pressure control system to the volume into which the closed volume gas leaks. The following benefits are realized as a consequence of this addition: 1) The process quantifies leak rates in real time. The result is that tests run until a desired quality or measurement uncertainty is achieved; the need for data post-processing is eliminated. 2) Tests have the proper conditions at both the onset and conclusion of the test. The desired test parameters are maintained throughout the test, eliminating the need to discard portions of collected data sets. 3) Test duration is optimized. In the traditional method, tests could be terminated prematurely, resulting in lower than expected quality, or run longer than necessary, returning results that have better quality than required. The new process prevents this from happening.
Compared to the traditional method setup, as shown in figure (a), the new method requires the addition of a control system consisting of a differential pressure measurement device, controller, and pressure regulator, as shown in figure (b). The control system compares the test pressure differential to the desired target and adjusts the downstream (i.e., low-pressure side) pressure such that the target pressure differential is met at the onset of the test and maintained throughout its duration. Based on feedback from the differential pressure measurement device, the controller commands a pressure regulator to increase or decrease the downstream gas pressure, as necessary. The adjustment is facilitated through connections to two gas pressure sources, one containing gas at a pressure higher than the downstream pressure region (e.g., vent to atmosphere), and the other source containing gas at a pressure lower than the downstream pressure region (e.g., vacuum pump).
In application of the new process, the test pressure is set by the operator to a pressure greater than the target pressure differential, similar to the traditional method. The control system then changes the downstream pressure to achieve the target pressure differential. The test begins and data are recorded. As leakage decreases the test pressure, the controller decreases the downstream pressure, and the target pressure differential is maintained throughout the experiment. The resulting mass-time data set is linear; therefore, the entire data set is used to compute the leak rate. Because real-time calculations of mass loss rate and its uncertainty are known during the test, the software can stop collecting data when the desired criteria are met.
The reduction in test duration and elimination of data post-processing results in the additional benefits of reduced manpower costs and reduced schedule. Additionally, manpower is reduced since tests require less oversight as they are known to end as specified.
This work was done by Christopher Daniels, Minel “Jack” Braun, Heather Oravec, and Janice Mather of the University of Akron; and Shawn Taylor of the University of Toledo for Glenn Research Center. NASA is seeking partners to further develop this technology through joint cooperative research and development. For more information about this technology and to explore opportunities, please contact http://technology.grc.nasa.gov . LEW-19292-1