CN1158173A - Method and apparatus for automatically balancing rotating machinery - Google Patents

Abstract

A method of and apparatus for balancing rotating machinery involves the transient acquisition of transient data (100) over the operational speed range of the engine being analyzed. A spectrum of the transient data is acquired using a real time spectrum analyzer (7) capturing an RPM triggered waterfall (34, 38). Further, a computer calculates a one shot balance solution. The balance solution (44) is shown on a video display screen (850) and includes the suggested solution to the balance problem based on the acquired data and generic sensitivity coefficients (46) and the predicted vibrational behavior of the proposed solution. By acquiring a second set of data, it can be verified whether the one shot balance solution remedied the balance problem. If a problem remains, the operator may calculate engine specific sensitivity coefficients based on at least two previous data acquisition runs. A new trim balance solution is then calculated based on the immediately previous run and the calculated engine specific sensitivity coefficients.

Description

Method and apparatus for automatically balancing rotating machinery

Field of Invention

The present invention relates generally to rotating machinery, and more particularly to methods and apparatus for automatically obtaining a balancing solution for balancing rotating machinery.

Background of the invention

Vibration forces can cause mechanical equipment like turbine engines to fail or not work efficiently. In a turbine engine environment, many high-precision components rotate at high speeds at high temperatures. Consequently, significant vibration from the imbalance can severely damage engine components.

To avoid this drawback, techniques have been developed to monitor vibration and indicate when rotating mechanical equipment is unbalanced. Reliable engine vibration monitoring usually starts with an accelerometer. Special cables must be used to carry the electrical output signals of the accelerometers, usually located deep in the engine, to the signal processing electronics. Electronics to condition accelerometer signals for mechanical storage or instrumentation display.

Once a balance problem is identified, an engine vibration analysis is required to diagnose and correct the problem. Specifically, the engine had to be taken out of service and analyzed, which typically required mounting the engine on a test stand, which was time consuming and expensive.

A common method of performing this analysis begins with vibration measurements. Accelerometers and tachometers read out vibration data at various engine speeds to determine if vibration limits have been exceeded. If the limit is not exceeded during the shock measurement, the engine has been properly balanced and no further steps need to be taken. Once the shock limit is exceeded, a balance problem is usually assumed first and a balancing procedure is performed. However, if the vibration is not mitigated by balancing, use spectral analysis to pinpoint the cause of the vibration problem.

Data are necessary to quantify the amplitude of the shock. In the past, to obtain the amplitude and phase of the vibration as a function of velocity, steady state dwell has been performed at about ten discrete acquisition points over the useful range of engine operation. Then install the test carrier, which will be a first guess or try at how to balance the system. A second steady-state stop is then performed at the same discrete acquisition point to measure changes in vibration due to changes in the balance load. This change determines the imbalance sensitivity value (coefficient) used to calculate the new equilibrium solution using the least squares method. A new equilibrium solution is usually calculated assuming the engine is still unbalanced. This equilibrium solution is theoretically the best solution for the entire operating range of the engine.

In another common approach, the data for each steady-state standstill is manually recorded one by one, and the recorded data is entered into a computer that calculates the equilibrium solution.

According to another balancing system of the prior art, Mechanical Technology Inc.'s MTI ⁽ R) PBS-4100( ^TM ), engine specifications can be stored in a computer so that the system knows where to make a steady state standstill. The data obtained from these pauses are processed by a computer using the method of least squares to obtain an equilibrium solution. After the first pause, perform a one-time balance using the PBS-4100 ^TM system with the common sensitivity or influence coefficient of the engine model under test stored in the computer, then perform a second steady state pause to confirm that the system has been balanced. After the PBS-4100 ^™ system has calculated the balance solution, the system provides a display showing the proposed solution and the magnitude of the balance load that should be installed on the engine.

A problem with this approach is that it is difficult to guarantee that the first and second steady state stops are performed at the same speed. There can be a 10-15 PRM (revolutions per minute) difference between the first and second steady state pause, which can adversely affect data quality. This is especially true when data are obtained within a tolerance range of engine speed. For example, data may be obtained in the range of 2950-3050 RPM for a target speed of 3000 RPM. Speed errors cause errors in calculating engine-specific sensitivity coefficients and errors in balance calculations. Another problem with the above method is that it takes a considerable amount of time to balance an unbalanced engine.

Invention Summary

The present invention provides methods and apparatus that overcome the deficiencies associated with conventional rotating machinery balancing systems.

For example, the present invention provides methods and apparatus for balancing rotating components in a reduced time over conventional methods. Data acquisition of the drive provides fast and accurate balance solutions in minutes and also eliminates standstill or constant speed operation, saving fuel.

In addition, the present invention enables more accurate estimation of balance solutions for unbalanced rotating components. One way to achieve a more accurate estimate is to have automatic or instantaneous data collection during slow engine acceleration or deceleration so that no stalls in engine speed are required. In this way, the data collection time can be shortened and data can be obtained at more collection points. 50 to 100 acquisition points can be automatically obtained using instantaneous data acquisition with real-time processing capabilities, while only about 10 acquisition points are obtained according to the common steady-state pause method. The accuracy of the equilibrium solution is improved by using more collection points. Therefore, the use of instantaneous data acquisition achieves a more accurate balance calculation than the ordinary steady-state standstill method.

Specifically, instantaneous data acquisition is performed based on the measured speed of the rotating part. Thus, if in the first data acquisition operation, data is acquired in increments of 50 RPM up to 4000 RPM starting at 1000 RPM, the data acquisition will occur precisely at these speeds. For subsequent data acquisition operations, data will also be acquired at these precise speeds compared to normal steady-state pause methods. Therefore, the engine-specific sensitivity and equilibrium solutions calculated from the obtained data are more accurate than those obtained by the ordinary steady-state standstill method.

Also, the present invention provides an estimated response to the equilibrium solution after it has been calculated and before its implementation, to allow the user to determine whether the solution is desired.

In addition, the present invention is also capable of proposing a balanced solution for each collection point within the range of collection points from which data has been obtained, so that the user can tailor the balanced solution according to the desired characteristics of the system. For example, users can calculate equilibrium solutions for specific speed ranges or specific sensors.

The balancing system of the present invention does not require disassembly of the rotating member from the structure in which it is installed, such as an aircraft, boat, etc., and therefore does not require the engine to be shut down. The system therefore minimizes costly engine disassembly and reduces unnecessary maintenance by allowing the user to test and balance the engine in situ. In other words, the engine does not have to be disassembled and attached on the test bench.

A method of balancing a rotating component in an interactive computer system in which component data including characteristics of the component type is pre-stored in memory involves several steps. These steps include slowly changing the rotation speed of the component within a predetermined speed range according to the component type, detecting instantaneous vibration data at multiple collection points within the speed range while changing the rotation speed of the component, and judging whether the component is is in an unbalanced state, and computes a balance solution from the part data and the instantaneous data when the part is in an unbalanced state. The step of detecting instantaneous data may include the steps of generating a waterfall file and displaying a three-dimensional representation of the waterfall file data known as a waterfall graph. A waterfall chart is a graphical representation of data on three axes. For example, one axis is RPM, the second axis is frequency, and the third axis is the amplitude of the vibration.

Specifically, the step of detecting transient signals acquires and digitizes vibration signals and tachometer signals using a real-time spectrum analyzer. Data is collected in increments of RPM as the engine accelerates or decelerates. For each instant of the data, the time-domain digitized waveform is transformed into the frequency domain using the "Fast Fourier Transform (FFT) method". The resulting FFT representation of the vibration signal is a representation of the characteristics of the frequency spectrum of the vibration component measured by the system.

The step of detecting the instantaneous data includes the step of processing the waterfall data to obtain synchronized shock amplitude and phase data at various acquisition points for one or more positions of the component within the velocity range. The location where the measurements are taken can be selected by the system operator. The component data used to calculate the equilibrium solution includes sensitivity (ie, influence) coefficients that depend on the component type. The operator can enter new part data into memory. The step of judging includes the step of comparing the instantaneous data with predetermined shock limits.

The method may comprise the step of installing at least one load at a determined location according to an equilibrium solution. The installation of the load can be identified after removal of the previous balance load, or the installation of the balance load can be identified without removal of the previous balance load. At least one location on the component of the equilibrium solution and a corrective load to be installed at the at least one location can be identified and displayed. In addition, the graph of the first vibration characteristic waveform of the component based on the instantaneous data can be displayed, and the graph of the second vibration characteristic waveform of the component based on the equilibrium solution can be estimated and displayed. A vibration signature is a graph of synchronous vibration amplitude as a function of rotational speed of a component, indicative of the state of equilibrium of the rotating component.

The speed range includes component speeds from idle state to maximum power state. The speed range may be selected and determined with two speeds between an idle state speed and a maximum power state speed depending on the component type.

The device for balancing a rotating part of the present invention includes a sensing device connected to the part to detect instantaneous data at multiple collection points within the speed range of the part when the rotating speed of the part is slowly changed, and the sensing device connected to the sensing device , performing spectral analysis to generate waterfall data for detailed vibration analysis, a data processor for generating balance data from the waterfall data and comparing the balance data with vibration limits, and connected to the data processor, when the balance data exceeds the vibration limit means for calculating an equilibrium solution based on the first sensitivity coefficient and the equilibrium data.

       Overview of the attached drawings

The invention will now be described in more detail with reference to preferred embodiments of the invention shown in the accompanying drawings, given by way of example only, in which:

Figure 1 shows an exemplary turbine engine that may be balanced in accordance with the present invention;

Figures 2(a)-2(c) represent an exemplary structure for connecting an aircraft engine to the engine vibration and balance analysis system of the present invention;

Fig. 3 (a) is an exemplary engine vibration and balance analysis system structure of an embodiment of the present invention;

Fig. 3 (b) is the data flow diagram of Fig. 3 (a) system;

Figure 4 is an exemplary waterfall file display of the present invention;

Fig. 5 is a flowchart of an exemplary single-stage (single plane) balancing method of the present invention;

Figures 6(a)-6(m) represent exemplary screens of the intuitive interactive environment of the present invention;

7(a) and 7(b) show a flowchart of an exemplary two-stage balancing method of the present invention.

A detailed description

Portions of the disclosure of this patent document contain material that is protected by copyright. The copyright owner has no objection to the facsimile reproduction by anyone of any of the patent disclosures contained in the patent files or records of the US Patent and Trademark Office, but reserves all other copyrights.

While the following description is within the scope of aircraft engine systems, those of ordinary skill in the art will appreciate that the present invention may be applied to other types of rotating machinery including gas turbines, compressors, generators, pumps, electric motors, and steam turbines.

Figure 1 shows an exemplary engine to which the engine vibration and balance analysis system of the present invention may be applied. The engine has a sector section including blades 1 for low pressure propulsion and a supercharger 2 , a main section 4 including a high pressure gas generating device or compressor 3 and a high pressure turbine 4 , and a low pressure turbine section (LPT) 5 . The engine balancing system uses vibration data and engine speed data. In order to obtain vibration data, it is necessary to connect sensing components, such as accelerometers, to the turbine engine to measure suitable data. In the exemplary engine of FIG. 1 , on the sector bearing (not shown) connected to the blade 1 of the sector, on the sector cover (not shown) connected to the blade 1 of the sector, among other places One or more accelerometers for measuring vibrations are fixed on the tailstock (not shown) of the turbine 3 of the LPT section 5 and/or on the tailstock of the compressor 3 of the main section. The accelerometers at various locations act as a balance and vibration path for measuring the vibrations of the engine.

Many types of accelerometers can be used to measure engine vibrations or to detect and measure vibrations, such as compression or shear accelerometers. In a typical compression-type accelerometer, a mass applies pressure to a piezoelectric element. In a typical shear accelerometer, the housing applies a shear force to the piezoelectric element. Although the invention is not limited by the type of accelerometer, general-purpose accelerometers with a sensitivity of 1 to 100 PC/g, a weight between 10 and 50 grams, and a frequency range of 0 to 12 KHz are suitable for the invention. In addition, miniature accelerometers with a sensitivity between 0.05 and 0.3 PC/g, a weight of 0.5 to 2 grams, and a frequency range of 1 to 25 Khz are also satisfactory. Other types of acceptable accelerometers include the following: 1) for three-dimensional measurements; 2) for permanent monitoring measurements on industrial machines; 3) for use at very high temperatures; 4) for calibration and other benchmarking purposes; 5) for vibration measurements on buildings and other structures; and 6) for very large shock measurements. Accelerometers that measure vibration are already installed on some aircraft. In the event that the aircraft does not have an accelerometer installed, the preferred accelerometer type is a DYTRAN(R) Model 3174C high temperature accelerometer.

In a practical accelerometer design, the piezoelectric element is arranged so that when the assembly vibrates, the component housing exerts a force from the piezoelectric element proportional to the acceleration of the shock. It should be noted that direct loading of the piezoelectric output at the accelerometer output can significantly degrade the accelerometer's sensitivity and limit its frequency response. To minimize these effects, a preamplifier is used to feed the output signal of the accelerometer to an impedance well below the accelerometer impedance, which is suitable for interfacing with the relatively low input impedance of typical measurement and analysis equipment . Also, vibrations can be detected and measured using velocity transducers or displacement probes. The shock sensor can be an existing engine transducer or a ground test transducer.

The rotational speed of the engine may be determined with a tachometer such as a magnetic tachometer or an optical sensor tachometer. In the present invention, an eddy current probe type magnetic tachometer that calculates the rotational speed using a magnetic field can be used. On aircraft equipped with an Aircraft Vibration Monitoring (AVM) system, a typical magnetic tachometer measures engine RPM by sensing a purposefully deep tooth in the tail. A typical light sensor tachometer measures speed using a beam of light bouncing off a marking point. For example, the marking point is a reflective tape affixed to the first blade of the fan blade.

Aircraft engines can be classified into one of three configurations: 1) No AVM; 2) Old AVM; and 3) New AVM. In an AVM-less configuration, in order to measure the shock, a shock sensor must be placed by a technician. The AVM structure includes shock sensors. In older AVM architectures, the sensor output was obtained using a direct connection, while in newer AVM systems, the sensor output is obtained using a front panel connector inside the cockpit.

Aircraft with no AVM platforms include Boeing 727, 737-200 and 757, DC10 and MD80. The equipment required for an AVM-free system is engine and aircraft dependent and can include: two accelerometers, approximately 250 feet of AVM-free cable assembly (including two charge amplifiers (connected) and a light sensor tachometer, engine-specific Rack tool, and vibration analyzer with BNC plug and cable.

Figure 2(a) shows a typical non-AVM cable connection between an aircraft and the vibration analysis and balancing system of the present invention. In an AVM-less system, the engine vibration and balance analyzer 7 is directly connected to the turbine engine 6 . According to this exemplary connection, three or more data paths are connected to three or more sensors on the turbine. Channel 1 is connected with the tachometer, channel 2 is connected with the front vibration sensor on the blade bearing, and channel 3 is connected with the rear vibration sensor on the turbine tailstock. In addition, the tachometer and accelerometer are powered using the power line connected to the analyzer 7 to actuate the system for vibration and balance analysis.

Aircraft using the Tabular AVM platform include the Boeing 737-300. A typical older AVM cable connection is shown in Figure 2(b). Typically required in-flight equipment includes: a gauge AVM cable on one side of the aircraft 8, a tachometer regulator box 9, and a vibration analyzer 7 with BNC plug and cable. In this old style AVM connection, passage 1 is connected to the tachometer regulator 9 which is connected to the AVM along with passage 2 for the front shock (eg sector shift bearing) and passage 3 for the rear shock (eg turbine tailstock) The housing 11 is connected to the AVM passenger seat 12 to obtain vibration data from the side 8 of the aircraft via the AVM. The power line is connected with the tachometer regulator 9 .

The new AVM platform is suitable for all modern aircraft. Figure 2(c) shows a typical modern AVM cable connection. Typically required equipment includes: a new AVM cable connecting the analyzer 7 to one side of the aircraft 8, a tachometer regulator box 9, and a vibration analyzer 7 with a BNC plug and cable. In this new AVM connection, channel 1 is connected to tachometer adjuster 9 which, along with front shock channel 2 and rear shock channel 3, is connected to AVM housing 13 on side 8 of the aircraft for data acquisition. The power line is connected with the tachometer regulator 9 .

In the AVM connection of Figures 2(b) and 2(c), the tachometer regulator box 9 receives a series of pulses of engine rotation and outputs one pulse for each revolution of engine speed. The trigger input of the vibration and balance analyzer 7 receives pulses from the tachometer, measures the time between pulses and calculates the engine speed in RPM. Shock sensors measure vibrations of all frequencies. For engine balancing, the vibration components required for balance calculations are the fundamental frequency or synchronous vibration components. The frequency of this component is the frequency of the engine speed. The RPM value calculated from the external trigger signal is used to determine the frequency components in the amplitude and phase waterfall in order to obtain balanced vibration amplitude and phase.

During engine operation, aircraft vibration monitors will indicate or sensors will detect vibrations. To analyze the vibrations, airline mechanics run outgoing cables to the AVM or install sensors (such as accelerometers) and tachometers.

Several components make up the illustrative engine vibration and balance analyzer shown in Figure 3(a). Shock data and engine speed are received from sensors and provided to the instantaneous data acquisition circuit 100 including an analog-to-digital (A/D) converter 150 (e.g., 16-bit A/D converter) through an external trigger input, so that the acquired data into digital form. Each shock sensor is equivalent to a balancing channel. The engine can be accelerated or decelerated 1) following the normal steady-state velocity method of collecting vibration amplitude, phase, and velocity (RPM) information at selected constant speed operating conditions or 2) during slow transient engine acceleration or deceleration of on-line vibration amplitude, phase, and RPM data data during the period. The pre-processed digital shock data is transmitted to a digital signal processor (DSP) memory 250 within the signal processing circuit 200 . According to an exemplary embodiment, DSP memory 250 is 8 megabytes of RAM, although the memory can be larger depending on the number of lanes (sensor locations) on the engine.

Within the signal processing circuit 200, the DSP processing section 350 processes the vibration and tachometer signals (ie, the acquired instantaneous data) using the function of a real-time spectrum analyzer. In a typical implementation, signal processing circuit 200 includes a 32-bit DSP microprocessor operating at 50 MHz. The vibration and tachometer data are averaged and processed at predetermined incremental RPM changes (eg, 50 RPM) and stored in the waterfall element of DSP memory 250 . Incremental changes in engine speed are selected for signal processing with reference to the analyzer's ability to acquire data. Therefore, to obtain the best resolution, it is best to choose increments as small as the analyzer will allow.

At the conclusion of the data collection process, waterfall file data is written from memory 250 to disk. The waterfall data is then post-processed to extract balanced amplitude and phase signatures.

The signal processing section 200 is controlled by a user interface 400 including a host interface 450 that downloads DSP software to the DSP processing section 350 . DSP memory 250 transmits displayable records, such as graphical displayable information, including time and waterfall files, as well as amplitude, phase and velocity information, to host interface 450 . The display record may represent, in addition to the data, the shock limits plotted against data entered by the user or pre-stored in the hard drive 500 .

In an exemplary embodiment, the user interface includes a MICROSOFT ^(R) compatible graphical user interface that guides the user through the balancing process. A typical host interface would be a standard 80386 or an optional 80486 with a companion coprocessor running at 33MHz.

The host interface 450 is connected to a digital input/output (I/O) card 550 and downloads the spectrum analysis software (ie, the FFT algorithm 32 ) for digital signal processing operations to the digital I/O card 550 . Digital I/O card 550 can interface with function keys on a keyboard, mouse, printer, and the like. User interface 400 is connected to disk drive controller 600 and stores and receives data from internal hard drive 500 (i.e. 50 megabyte hard drive, 120 megabyte optional) and floppy disk 700 (e.g. 1.44 megabyte 3.5 inch floppy disk) )The data. Balancing software 650 that guides the user step-by-step in balancing the engine is stored within the hard disk drive 500 . Other information typically stored in hard drive 500 includes spectrum analyzer software, data collection programs, data for various engine types, engine operating data and waterfall data after data collection is complete, and equilibrium solution data. Data for each engine type includes preparation information on how to collect the data and balance characteristics of a particular engine. Information output from hard drive 500 to floppy disk 700 typically includes engine running and waterfall data and balance files. A balance file contains vibration data collected in a previous run, which when combined at any time with the engine sensitivity file for the engine on which the balance file was applied, can be used to calculate a balance solution. The user interface 400 is also connected to a video graphics adapter (MT) 750 capable of transmitting video data to a gas plasma display 800 and/or an external MT display 850 for displaying equilibrium solutions, waterfall files, etc. An exemplary waterfall file display is shown in FIG. 4 .

The data flow diagram of the system described in Fig. 3(a) is shown in Fig. 3(b). The vibration data and the engine speed data of the rotating member 20 are detected by a plurality of vibration sensors 22 and a speed sensor 25, respectively. Sensors are equivalent to balanced pathways. The data information from the vibration sensor 22 and the speed sensor 25 is input to the spectrum analysis part of the balance analyzer. In particular, the shock data from the shock sensor 22 is transformed from the time domain to the frequency domain using FFT 32 . The parameters required to perform this transformation can be modified by the user with typical values set as follows: Fmax = 500MHz, N = 200 lines, overlap = max. The result of FFT 32 is transformed into an amplitude spectrum in amplitude waterfall section 34 . The cross-correlation spectrum section 36 also calculates a cross-correlation spectrum which is input to a phase waterfall section 38 which determines the phase response of the shock sensor 22 to the speed sensor 25 (ie, the tachometer). Acquisition of each amplitude and phase spectrum is triggered by a tachometer. The balanced amplitude of the amplitude cascade section 34 can be used for spectrum analysis diagnostics in the diagnostic section 40 . Balanced amplitude and phase signature data from amplitude waterfall section 34 and phase waterfall section 38 , respectively, are saved in balanced shock data file 42 . The data within the balance data file 42 is used to calculate a balance solution 44 along with a sensitivity file 46 which includes sensitivity coefficients associated with the engine being balanced.

A high-performance real-time spectrum analyzer designed for diagnosing vibration problems including at least balance-related problems may be used as a digital signal processing device. A typical analyzer for practicing the present invention is the "Dynamic Signal Analyzer" model SA390 manufactured by Scientific-Atlanta, Inc. An analyzer capable of distinguishing between balanced and unbalanced problems, including acoustic problems, may also be employed. The spectrum analyzer must be capable of engine vibration measurements over the entire engine speed range, ideally including a versatile time domain (oscilloscope) and spectrum display for use in identifying vibration sources. This device eliminates unnecessary engine disassembly and solves vibration problems at a significantly reduced time and cost.

By using a spectrum analyzer capable of processing data spectrum or waterfall data in real time, the engine vibration and balance analysis system can perform instantaneous data acquisition for multi-speed multi-level balancing, during which the engine is accelerated from idle to full power. The spectrum analyzer collects multiple vibration data including amplitude, phase and frequency over the power range of the engine. More data improves accuracy and yields better predictive equilibrium solutions. Real-time instantaneous data collection provides more data collection points while avoiding the pause that common balancing systems have. In addition, the engine only needs to run for about 2 to 5 minutes to obtain the data, which is not only shorter than the ordinary method, but also saves fuel and shortens the time to obtain the equilibrium solution.

The engine vibration and balance analysis system of the present invention is also adapted to receive vibration data from an aircraft engine vibration monitoring system, such as ^ENDEVCO® MICROTRAC II ^™ . Therefore, flight data can be input as a data run to the engine vibration and balance analyzer and a balance solution can be calculated.

The steps included in the balancing operation of the typical single-stage balancing solution performed by the above-mentioned engine vibration and balancing analysis system are shown in the flow chart of FIG. 5 .

In step ST1, equipment for connecting the engine and the analyzer is installed. The controls reach to collect engine operating information on engine speed and vibration levels. After the data is collected, it is judged in step ST3 whether the engine vibration exceeds a predetermined vibration limit. If the shock limit is not exceeded, there is no balance problem to be found in the ST3. Control then goes to step ST4 where the device is removed and the operation ends at ST5. Otherwise, if the shock limit is exceeded, a balance problem has been found and control passes to step ST6, where a first-attempt or one-shot balance solution to the problem is calculated using the general sensitivity coefficients. Control then passes to step ST7 where the correction load is installed based on the one-shot equilibrium solution calculated in step ST6.

After the calibration load is installed in step ST7, the engine is run again to collect vibration data in step ST8. Then it is judged in step ST9 whether there is still a vibration problem. That is to say, in step ST9, the collected data is compared with the vibration limit to determine whether there is still a balance problem. If no balance problem is found in step ST9, control passes to step ST10 where the device is removed and the operation ends in ST11. If there is still a problem in step ST9, a new engine-specific sensitivity is calculated in step ST12 based on the previous two operations. Control passes to step ST13 where a new equilibrium solution is calculated based on the new engine specific sensitivity and data collected during the previous engine run. Execute steps ST7, ST8, ST9 again. If there is still a balance problem, steps ST12, ST13, ST7, ST8 and ST9 are performed until the problem is solved. Once the required solution is obtained, various cables and devices connected to the engine are removed in step ST10, and the operation is ended in step ST11, and the engine can run normally.

Figures 7(a) and 7(b) show the flow of balancing operations required to obtain a two-stage balancing solution. The same steps as those shown in Fig. 5 are denoted by the same reference numerals in Figs. 7(a) and 7(b). If the user wishes to compute a two-level solution, a second attempted solution is computed at step ST14 using the common sensitivity after finding a balance problem at step ST9. Then install the balance load and run the engine again in steps ST15 and ST16, respectively. The obtained data is compared with the shock limit in step ST17 to judge whether there is a balance problem. If not, steps ST10 and ST11 are executed. Otherwise control passes to step ST18, where a new sensitivity is calculated from data collected during the previous three engine runs (ST2, ST8, ST16). An equilibrium solution is calculated at step ST19 using this new sensitivity and the data from the last run. Then install the load and run the engine again in steps ST15 and ST16, respectively. In step ST17, a comparison is made to determine whether balancing is still required. If there is still a problem, execute steps ST18, ST19, ST15, ST16 and ST17 until the problem is solved. Once it is confirmed that there is no more balance problem, steps ST10 and ST11 are executed to end the operation.

It will be appreciated that the system is capable of more than two stages of balancing by modifying the above process, depending on the number of stages required.

Interactive user/system communication can be performed using the MICROSOFT(R ⁾ WINDOWS( ^TM) operating system, an open architecture with full internal spectrum analysis capabilities. Design the system to be PC compatible. Other operating systems known in the art, such as DOS, OS/2 ^™ , etc. are also suitable for use with the present invention, but WINDOWS ^™ is preferred.

In an exemplary implementation of the present invention, various modes of operation are accessible by the user within the WINDOWS ^™ environment where the user enters data and selects the operations to be performed.

The system user enters the WINDOWS ^™ environment after properly wiring and connecting the necessary equipment as described with reference to Figures 2(a)-2(c), as shown in Figure 6(a) in the Engine Vibration and Balance Analyzer The main menu of options appears on the screen. If the user selects the SETUP button by, for example, placing the pointer on the SETUP button with a trackball and pressing the trackball, a Select Engine Type screen appears as shown in FIG. 6(b). The Select Engine Type screen includes a list field 300 listing various pre-stored data files containing specifications related to a number of engine types. These files are preferably stored on hard drive 500 of FIG. 3(a). The user then selects the appropriate engine type corresponding to the engine to be balanced. Or the user may determine an engine type that is not stored in memory, in which case the engine vibration and balance analyzer will prompt the user to enter data pertaining to the specifications of the engine to be balanced. The Select Engine Type screen also includes a table field 310 and drop down fields 320 and 330 to change the drive and directory respectively and to change the list of files appearing in table field 300 according to methods well known in the art.

After selecting the file corresponding to the desired engine, the MainMenu screen of Figure 6(c) appears. This screen provides an indication of the engine type, comments about the engine, and an indication of the connection of the shock sensor to the analyzer. Other engine information may be displayed, as the invention is not limited in this respect. After confirming that the correct engine type has been selected, the user presses the ACQUIRE DATA button and the Engine Run Setup screen of Figure 6(d) appears. At this time, the user can enter the serial number of a specific engine in the engine serial number field. If the user enters a serial number, the serial number is saved so that the instantaneous data collected related to this engine is automatically saved for future reference at the end of the run. The Weight (load) and Angle (angle) fields on the Engine Run Setup screen allow the user to enter load information for existing loads on the front plane or on the rear plane. The ACQUISITION (acquisition button allows the user to select the desired type of operation, i.e. manual or automatic. In the manual mode of operation, data is collected for a specific rotational speed, as is done with the normal steady-state dwell method. In the automatic mode of operation, the data Acquired automatically when the engine is accelerated gradually.Data will be collected according to the programmed rotational speed interval.For example, an exemplary data collection cycle of the present invention can be about 2 to 5 minutes, within 50 to 5 minutes of this period of time Collect data at 100 collection points.

Selecting the CANCEL button will return the system to the screen of Figure 6(c) without initializing the selected item. Selecting the OK button will set the system to collect data. If the tachometer signal is not present at this time, a screen appears telling the user that an improper RPM has been detected and prompting the user to check the tachometer connected to the balance system. The user is also asked whether to continue data collection. The user can abort the test in response to this query. If a tachometer signal is detected, the user can respond positively to the query and proceed with the test.

If the test continues, a screen like Figure 6(d) will appear. The screen of Figure 6(e) appears while collecting engine operating information related to engine speed and vibration levels. In one embodiment, a spectrum analyzer that usually displays data information in a waterfall format (eg, FIG. 4 ) after data collection can be used. Of course, waterfall data can also be displayed while data collection is in progress. When the engine is running below the minimum engine RPM, the user can select the START button. Once the START button is pressed, the engine speed increases gradually to maximum speed. Once the engine's maximum speed has been reached, the engine has completed its stroke. At this point should press the STOP (stop) button to stop data acquisition. Once the STOP button is pressed, the Main Menu (main menu) of Figure 6(c) appears, and the user can now calculate the equilibrium solution.

When the BALANCE SOLUTION (balance solution) button among Fig. 6 (c) is pressed, just appear and show and run the relevant information and indicate whether the engine vibration exceeds the preset limit stored in the hard disk drive 500 of Fig. 3 (a) Balance Menu screen. The collected data is then compared to the shock limits stored in the memory 250 of the user interface 400 to determine if a balance problem exists. An exemplary Balance Menu screen is shown in Figure 6(f). Once a balance problem is found, the CALCULATEBALANCE SOLUTION button can be pressed to calculate a first-attempt balanced solution or a one-time balanced solution to the problem.

The complex values called universal sensitivities or influence coefficients are part of the data pre-stored in the engine vibration and balance analyzer memory 250 for each model of engine. Each stored general sensitivity coefficient represents a value related to how engine vibration typically varies for a model of engine due to changes in balance load. By combining the acquired vibration data with the general sensitivity data, the analyzer is able to calculate a one-shot equilibrium solution.

Once the user presses the CALCULATE BALANCESOLUTION button on the Balance Menu screen, the BALANCE SOLUTION screen shown in Figure 6(g) appears. When the one-shot balance solution is calculated, a graphic representing the engine balance flange is displayed showing the suggested weight, part number and hole location for the load to be mounted. The balance solution is preferably computed using the least squares method for all valid vibration and balance paths over the entire engine speed range. Thomas P. Goodman described in the paper "A Least Squares Method for Computing Balance Corrections" published in the "Journal of Engineering for Industry" in August 1965, pp. 273-279 This algorithm.

Calculates the predicted shock data for the specified one-shot equilibrium solution and displays it along with the shock data from the last run. After each run, without first computing an equilibrium solution, the user can observe the data and can determine that the vibration problem is not an equilibrium problem by analyzing the condensate vibration signature. Thus, for each balance solution calculated, after the balance load corresponding to the calculated balance solution has been installed, the graph of the current (i.e., associated with the previous run) vibration signature can be compared with the estimated resulting vibration signature The waveform is displayed together. Ordinary systems do not have the ability to find faults because the amount of data is almost, if not always, inappropriate for analyzing the shock data signature in the manner required to draw such conclusions.

To analyze the vibration problem before calculating the balance solution, the user activates Vibration Data in the menu bar of the Balance Menu screen of Figure 6(f). There is no submenu in VibrationData, which includes the options of Select new Data File (select a new data file) and Plot Vibration Data (draw a graph of vibration data). The SelectNew Data File option should be selected first, this allows the user to identify the desired file to be analyzed. Thereafter, the user activates the Plot Vibration Data option, and the system displays the graph of the vibration characteristic waveform for the selected file.

The installation of the balance load can be identified after removing the previous balance load, or the installation of the balance load can be identified without removing the previous balance load. After the corrective loads are installed, the user can run the engine again and collect data as described above to determine if a vibration problem exists. Then compare the collected data with the vibration limit to determine whether there is a balance problem. If there is a problem or to ensure the correct balance is achieved, the user should calculate new engine specific sensitivities after the load has been changed. Then, after collecting new operating data based on load changes, new engine-specific sensitivities should be calculated, as opposed to new equilibrium solutions. Engine-specific sensitivities differ from generic sensitivities in that, unlike estimates for the type of engine operation being determined, they are specific to the engine being tested. To calculate the sensitivity again, the user starts Sensitivity Data on the menu bar of the Balance Menu screen (sensitivity data, the screen shown in Figure 6(h) appears).

To calculate new sensitivities, the user selects the Calculate New Sensitivities option and the Sensitivity Menu screen of Figure 6(i) appears. The user is now able to select the type of balancing required, ie single stage or multi-stage (balance position on the engine). For example, as shown in Figure 6(i), you can choose Front plane (pre-stage), rear plane (rear stage) or dual plane (two-stage). If you are performing pre- or post-stage balancing, only two data acquisition files are required to calculate the new engine-specific sensitivities. However, if two-stage balancing is to be performed, three data files for the load variation of the two-stage are required.

If the user presses the CALCULATE FRONT PLANE button, the first data file screen (not shown) will appear, and the collected data files stored in the system memory will be listed on the screen, and the user can select the first data file accordingly. document. Then, a second data file screen (not shown) with the same form appears for the user to select a second data file. After selecting these two data files, the Sensitivity Menu screen shown in Figure 6(j) appears, showing a summary of the selections. At this point the user should select the SaveSensitivity option on the menu bar to save the information just selected.

After the sensitivities are saved, they are automatically loaded for use in the next equilibrium solution. If the user selects any sensitivity for calculation, the user can access the file using the Select New Sensitivity File option on the screen shown in Figure 6(h). A series of files appears on the screen from which the user selects the new sensitivity file (not shown) required for the solution. The user should then select the Sensitivity Data (sensitivity data) option of the menu bar (that is, a screen similar to the screen in Fig. Graphics of level data. A Plot Sensitivity screen like Figure 6(k) appears showing the current sensitivity of the selected file.

The user should then request the engine vibration and balance analysis system to compute a new adjusted balance solution. The calculated new tuned balance solution will be based on newly collected data and calculated engine specific sensitivities rather than generic sensitivities. The user then carefully tests the chosen balancing solution to ensure that the vibration problem has been eliminated. If this problem persists, the user should command the engine vibration and balance analysis system to calculate new engine specific sensitivities based on all data collected while the engine is running, continuing as described above until an acceptable tuned balance solution is obtained . Each adjustment balance also allows the user to fine-tune the balance solution.

After calculating the one-time or adjusted balance solution, a screen (not shown) with the balance menu bar of FIG. 6(g) appears, providing the user with the Several options for the Select Plane (selection level) option. The Return option takes the user back to the previous screen or window. Assuming the Engine Vibration and Balance Analyzer is connected to a compatible printer, the Print option allows the user to send a copy of the data to the printer. In addition, the Save option saves the equilibrium solution and proposal in a file on floppy disk. All balance vibration data are automatically saved after each engine run.

Users can also subtract, add or combine weights with the Adjust Weights option. Also, the user can set the maximum number of loads allowed by the balanced solution. A new run can be performed with or without removing the load from the last run, according to the user's wishes. Also, if there were loads before the carrier was installed, the system can compensate the loads determined by calculating a new solution based on the removal of the existing loads.

The Select Plane option allows the user to select the desired plane for the equilibrium solution. The submenu under Select Plane provides level options. The user can then, for example, ask the analyzer to display the pre- or post-balanced solution, or to recalculate the pre- or post-stage solution when acquiring data in pre-post mode.

The Custom option allows the user to specify a certain balance and vibration path to be considered when determining the balance solution. For example, users can customize balance loads for specific transducers or RPM ranges. If the desired solution is modified when the Customige option is selected, the system will recalculate the load proposal (ie balanced solution) for the custom data.

Data compression, payload versatility, and the ability to store data externally can all be very important. The Data Management button in the Main Menu (Figure 6(c)) provides these capabilities. Once the Data Management button is pressed. The screen of Fig. 6(e) appears. Pushing the Sum Weights button enables the user to manually enter the load type and location and calculate the net weight and vector sum of the entered loads. Pressing the Plot Data button enables the user to compare two vibration signatures from different runs or vibration between two different serial number engines (eg runs). Once the Plot Data button is pressed, the user is prompted to select the first and second serial numbered engines for comparison. The data for the individual engine runs can then be plotted together as shown in Figure 6(m). Pressing the Export data to Floppy button enables the user to externally store all information related to the current serial number engine selected on a floppy disk.

Waterfall data is used to determine vibration amplitude and phase for balance calculations, and to find the cause and diagnose vibration problems associated with unbalance. For example, waterfall data can be analyzed using the capabilities of a spectrum analyzer to find the cause of imbalance problems. Individual records within a waterfall file can be post-processed to determine vibration frequency and trend. This file also indicates the correctness of the vibration signal to help correctly point out the problem of the equipment.

A common approach to balancing data utilizes an analog tracking filter to obtain synchronous (1/rev) vibration levels at steady state velocity. Sometimes for diagnostic purposes, instantaneous data is collected during engine acceleration or deceleration, but the prior art makes no attempt to balance this data due to acquisition limitations associated with non-real-time spectrum analyzers.

While specific embodiments of the invention have been described, it will be appreciated that the invention is not limited thereto, since modifications may be made by those skilled in the art. This application contemplates any and all modifications within the spirit and scope of the basic invention disclosed and claimed herein.

Claims (37)

1. A method of balancing a rotating component in an interactive computer system, wherein component data including characteristics of the component type are pre-stored in memory, said method comprising the steps of: varying the rotational speed of the component within a predetermined speed range according to the component type; Detect instantaneous data at multiple acquisition points across the speed range while varying the rotational speed of the part; Judging whether the component is in an unbalanced state according to the instantaneous data; When the part is in an unbalanced state, the equilibrium solution is calculated from the part data and the instantaneous data. 2. The method of claim 1, wherein the detecting step further comprises the step of generating waterfall data from the instantaneous data. 3. The method of claim 2, wherein the detecting step further comprises the step of processing the waterfall data to obtain amplitude and phase data of vibrations at acquisition points of the component within the velocity range. 4. The method of claim 3, further comprising the steps of: display processed waterfall data; and When a part is out of balance, the processed waterfall data is analyzed to determine if there is a balance problem. 5. The method of claim 1, wherein said step of determining includes the step of comparing the instantaneous data to predetermined shock limits. 6. The method of claim 1, wherein said detecting data step includes the step of measuring, at one or more locations on the component, the amplitude and phase of vibrations of the component at acquisition points within the velocity range. 7. The method of claim 6, wherein said measuring step includes the step of selecting one or more locations on the part. 8. The method of claim 6, further comprising the steps of: generating waterfall data from instantaneous data; and The waterfall data is processed to obtain vibration and phase data of vibrations of the component at acquisition points over a range of velocities. 9. The method of claim 1, wherein the component data used in said calculating step includes sensitivity coefficients associated with component types. 10. The method of claim 1, further comprising the step of entering new part data into memory. 11. The method of claim 1, further comprising the step of installing at least one load at the determined location based on the equilibrium solution. 12. The method of claim 1, further comprising the step of identifying a load of the equilibrium solution to be installed at a location of the component. 13. The method of claim 12, further comprising the step of: Display the first vibration characteristic waveform of the component according to the instantaneous data; estimating a second vibration signature of the component based on the equilibrium solution; and Displays the second vibration signature of the component. 14. The method of claim 1, wherein the speed range includes component speeds from an idle state to a maximum power state. 15. The method of claim 1, further comprising the step of selecting a speed range to be determined by two speeds between the idle state speed and the maximum power state speed based on the component type. 16. The method of claim 1, wherein the part data includes characteristics of a plurality of part types, the method further comprising the initial step of selecting the part type from among the plurality of memory types. 17. A method of balancing a rotating part comprising the steps of: Vary the rotational speed of the part within the part speed range according to the part type; Detecting instantaneous data at multiple acquisition points across the speed range while varying the rotational speed of the part during the first run; Judging whether a component is out of balance based on instantaneous data detected during the first run; When the component is in an unbalanced state, calculating a first balance solution according to the first sensitivity coefficient of the component type and the instantaneous data; changing the first load amount of the component according to the first balance solution or according to the load change input by the user; Varying the rotational speed of a component including a changed load over the speed range; Detecting instantaneous data at multiple acquisition points across the speed range while varying the rotational speed of the part during the second run; Based on the instantaneous data during the second run, it is determined whether the component is out of balance. 18. The method of claim 17, further comprising the step of: calculating a second sensitivity coefficient based on the instantaneous data detected during the first and second runs; and A second equilibrium solution is calculated from the instantaneous data detected during the second run and a second sensitivity coefficient. 19. The method of claim 17, further comprising the step of selecting a speed range to be determined by two speeds between the idle state level and the maximum power state speed based on the component type. 20. The method of claim 17, wherein said step of varying the first amount of load further comprises the step of installing the load according to a first equilibrium solution. 21. The method of claim 17, wherein said step of changing the first load amount further comprises the steps of inputting a user's desired load change and installing and/or removing the load according to the user's desired load change. 22. The method of claim 17, wherein said step of detecting instantaneous data includes the step of measuring, at one or more locations on the component, the amplitude and phase of vibrations of the component at acquisition points within the velocity range. 23. The method of claim 22, wherein said measuring step includes the step of selecting one or more locations on the part. 24. The method of claim 17, further comprising the steps of selecting the airfoil required for the first equilibrium solution and displaying the selected level of the first equilibrium solution. 25. The method of claim 24, wherein the selected level includes one of a single level and multiple levels. 26. The method of claim 17, further comprising the step of selecting a part type associated with the part from a plurality of part types. 27. The method of claim 17, further comprising the step of: calculating a second balance solution based on a second sensitivity coefficient for the component type and instantaneous data detected during the second run when the component is in an unbalanced state; changing the second load amount of the component according to the second balance solution or according to the load change input by the user; continuously varying the rotational speed of the component including the changed second load amount within the speed range; detecting instantaneous data at multiple acquisition points across the speed range while continuously varying the rotational speed of the part during the second run; Judging whether the component is in an unbalanced state based on the instantaneous data during the third run; calculating a third sensitivity coefficient based on the instantaneous data detected during the first, second and third runs; and A third equilibrium solution is calculated from the instantaneous data detected during the third run and a third sensitivity coefficient. 28. Devices for balancing rotating parts, comprising: a measurement device, coupled to the component, for detecting instantaneous data at a plurality of acquisition points within the velocity range of the component while continuously varying the rotational speed of the component; a digital processor, coupled to the measurement device, performing spectral analysis on the instantaneous data to generate waterfall data, generating balance data from the waterfall data, and comparing the balance data to shock limits; and means connected to said digital processor for calculating a first balance solution based on the first sensitivity coefficient and the balance data when the balance data exceeds the vibration limit; wherein the component load distribution is adjusted according to the first balance balance solution. 29. The apparatus of claim 28, further comprising means for displaying waterfall data. 30. The apparatus of claim 28, further comprising means for displaying the equilibrium solution. 31. The apparatus of claim 28, wherein said measuring means includes means for measuring, at one or more locations on the component, the amplitude and phase of vibrations of the component at collection points within the velocity range. 32. The apparatus of claim 31, further comprising input means coupled to said measuring means for inputting one or more positions on the member. 33. The apparatus of claim 28, wherein said measuring means detects instantaneous data during two independent operations, said computing means calculates a second sensitivity coefficient based on at least two instantaneous data detection operations, said computing means also based on the most recent The instantaneous data detected in one instantaneous data detection operation and the second sensitivity coefficient are used to calculate the second equilibrium solution. 34. The apparatus of claim 28, wherein said rotating member is an engine fixed to an aircraft. 35. The apparatus of claim 28, further comprising: storing part data including sensitivity coefficients associated with each of the plurality of part types and memory storing transient data; Input means, connected to said memory, for receiving a selection signal for selecting a component type, the component type defining a first sensitivity coefficient of the component to be balanced. 36. The apparatus of claim 35, wherein said input means further comprises means for inputting data detection operating data comprising a prescribed speed or range of speeds for which an equilibrium solution will be calculated, said measuring means based on said Operational data detects transient data. 37. The apparatus of claim 35, wherein said input means further includes means for inputting a prescribed speed or speed range for which the equilibrium solution is to be calculated.

Images