Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Mechanical measurement system
Reexamination Certificate
2001-06-19
2003-02-25
Nghiem, Michael (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Mechanical measurement system
Reexamination Certificate
active
06526356
ABSTRACT:
BACKGROUND OF THE INVENTION
The monitoring of rotating machinery and gears using vibration measurements for determining the operating performance and condition is an established process in many industries. Vibration measurements are commonly acquired for test data processing using accelerometers placed on the rotating machinery to evaluate the condition of rotating equipment. Monitoring methods have been applied to the preflight evaluations of rocket engines. Effective monitoring tools are needed for preflight testing of rocket engine turbopumps. With the current trend toward reusable launch vehicles that will require the turbomachinery to operate for extended periods of time and on multiple missions, active monitoring of the condition of the internal components of the rocket engine becomes more important. In an effort to increase performance, expendable launch vehicle rocket engine turbopumps are sometimes operated at speeds and loads for which the engines were not initially designed. The consequence of this increased loading is that structural margins are decreased and the potential for hardware damage or catastrophic failure increases. For a particular expendable launch vehicle engine, two cases of liquid oxygen gear damage, including a catastrophic failure, have been observed during acceptance and development ground testing of the hardware. In order to mitigate the risk associated with the decreased structural margins, a drive train diagnostic procedure is needed.
In order to gain insight into the behavior of a rocket engine prior to flight, vibration response data is acquired during acceptance tests known as static firings or hot runs. During hot runs, the engine is fixed in a test stand and ignited. The steady-state data acquired is then analyzed to determine quantitative parameters that are used to assess the vibration signature of an engine. During rotating machinery analysis, particular vibration signatures are related to specific types of component defects. For example, discrete gear tooth defects are often characterized in the frequency domain by the appearance of spectral components at higher order harmonics of the speed of the shaft upon which the faulty gear is located.
The simplest fault detection techniques use the change in statistical properties of the vibration signal as a measure of engine health. Relevant vibration parameters that have been used include both the root mean square value and the kurtosis. While these vibration parameters provide a single number that can potentially indicate a defect in the system, the vibration parameters can not identify the source leading to a change in vibration level. Some gear fault detection methods use an analytic envelope signal to provide information on the modulation of the gear mesh frequency. However, due to the extremely high operating speeds of rocket engine turbopumps, measurements of the vibration responses up to the gear mesh frequency are often beyond the capability of the data acquisition instrumentation. More recently, wavelet transforms have been used for gear fault detection.
There are several problems associated with current methods of testing rocket engine turbomachinery. Current methods of testing rocket engine turbomachinery are often used to monitor bearing conditions and typically use trend analysis with a one-sided cepstrum analysis. These methods do not provide quantitative parameters for monitoring rocket engine gears developed from the one-sided cepstrum analysis. For turbomachinery with operating ranges typical of those experienced in rocket engines, the resolution of the one-sided cepstrum analysis approach yields results that are not as easily interpreted.
In conjunction with vibration measurements acquired during tests or operation of a rotating machine, digital signal processing techniques are used in condition assessment procedures. Analysts have used a one-sided autospectral density and the one-sided cepstrum for computation purposes to indicate gear performance. The one-sided cepstrum method in particular, has been used to detect damage in both rolling element bearings and gears. Partly due to the susceptibility of engine and transmission components to fatigue failures, there has been research directed at effective detection of gear tooth damage.
The typical practice in most machinery analysis is to establish a baseline for a specific machine and subsequently implement a regular monitoring schedule. Typical rotating machinery can be compared to a baseline over an expected operating life measured over many years. In a trend analysis process, changes in relevant parameters are then tracked over the life of the machine. However, the life of an expendable launch vehicle rocket engine turbopump, including acceptance testing and operational missions, is measured in minutes. The turbopump is usually test fired at least twice prior to delivery to the customer in order to show that acceptable performance limits are met. These test firings generally last for several minutes each. Due to limited available engine life, it is desirable to perform as few tests as possible to assure nominal performance. From a diagnostics perspective, a consequence of only two hot fire tests is a limited amount of operating time of the engine during which to assess small changes in the vibration signature rendering trend analysis ineffectual.
The inherent difficulty of comparing different pieces of hardware are mitigated by maintaining a database of previous comparisons between known hardware health and associated vibration characteristics. For example, the variability in vibration characteristics for engines that perform in a nominal fashion can be established with some simple statistics. Additionally, when a correlation has previously been established between documented hardware damage and a unique vibration signature, correlation can be useful in providing an early differentiation between a nominally operating engine and one that contains a defect.
In many methods, the comparison of measurements between different machines is not recommended because of variability in transmission path effects due either to manufacturing tolerances or differences in the instrumentation setup. Also, infant operational characteristics of a machine may not apply across an entire class of machines. The constraints imposed by the methods applied to rocket engine turbomachinery make a comparison between engines impracticable. Rotating machinery can be compared to a baseline operation over an expected operating life measured in years, but life time baseline comparisons are unsuitable for an expendable launch vehicle rocket engine turbopump life, including acceptance testing, that is measured in minutes. While trend analysis is performed from hot fire to hot fire on a typical engine, trend analysis is not applied to compare specific engine parameters to other engines parameters of known operating conditions. While traditional trend analysis may be performed on a single engine, parametric cepstrum analysis has not been accurately used to compare vibration signatures across a class of engines. While cepstrum analysis has been used to perform trend analysis for a particular engine, an easily interpreted parametric database containing cepstrum parameters for both healthy and faulty engines has not been used to provide accurate indication of engine health. These and other disadvantages are solved or reduced using the invention.
SUMMARY OF THE INVENTION
An object of the invention is to provide a test method for rotating machinery.
Another object of the invention is to provide a method for detecting anomalous gear performance in rocket engine turbomachinery during acceptance hot firing.
Yet another object of the invention is to provide a method for detecting anomalous gear performance in rocket engine turbomachinery during acceptance hot firing using two-sided cepstrum analysis.
Still another object of the invention is to provide a method for monitoring rocket engine turbomachinery using two-sided cepstrum analysis for generating a quantitative pa
DiMaggio Samuel J.
Sako Brian H.
Nghiem Michael
Reid Derrick Michael
The Aerospace Corporation
LandOfFree
Rocket engine gear defect monitoring method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Rocket engine gear defect monitoring method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Rocket engine gear defect monitoring method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3144654