2. OUTLINE OF VIBRATION MONITORING
This monitoring system is mainly aimed at the measurement of high frequency vibrations which are induced by rubbing of liner/rings. In preliminary test, however, cylinder pressure and lift of F. O. needle valve were additionally measured to clarify the features of vibrations in combustion process. Fig. 1 shows the outline of the system including all measured items. Vibration signals were measured by accelerometers and charge amplifiers containing low-pass filter, and output signals from the charge amplifiers were acquired by PC incorporating A/D conversion board via isolation amplifiers.
Representative locations of vibration measurements are shown in Fig.2. In preliminary test, measurements of vibration were made at several locations on one cylinder unit. This is because the vibrations are induced by the exciting force such as combustion and exhaust valve closing as well as the friction of liner/rings, and measurements at other location than cylinder liner are necessary to specify the vibration sources. While most of the accelerometers were mounted on magnet base attached to cylinder, for the measurement of cylinder liner vibration, additional accelerometers were directly fitted to the exposed liner surface using adhesive cement to observe the difference in responses due to contact resonance.
In terms of the condition of A/D conversion, the system is so designed that vibrations have high responses at the frequency range from 5 kHz to 30 kHz taking advantage of the resonance characteristics of accelerometer (shown in Fig. 3). The sampling frequency was adjusted to 100 kHz per channel, which is higher than usual, considering the setting of cut-off frequency of low-pass filter. Table 1 shows the specifications of vibration measuring system.
3. Results of the measurement
3.1 Features of waveform
As a procedure for the test running of engine, in a way of stepwise increase of load, the running for proper fit of liner/rings was initiated and subsequently various kinds of performance test were carried out. The results of measurement at 75% load in the period of running-in are demonstrated in Fig. 4.
Accelerometer
-B&K type 4371
-Natural frequency: 48 kHz
Charge amplifier
-B&K NEXUS type 2692A
-L. P. F. cut-off frequency: 30 kHz
- L. P. F. attenuation slope: 40 dB/dec
-H. P. F. cut-off frequency: 1 Hz
A/D conversion board
-National Instruments type PCI-MIO- 16E-1
-Channels: 16 SE
-Max. sampling rate: 1.25 MS/s
-Resolution: 12 bits
The signals shown in this figure, in order from top to bottom, represent rotation mark (T. D. C. ), lift of F. O. needle valve, cylinder pressure and accelerations of exhaust valve, cylinder cover, cylinder liner and cylinder block. At cylinder cover the acceleration was measured in the vertical direction and at other locations measured in the horizontal direction. And all acceleration signals shown here are the results of measurements in use of magnet base. From Fig. 4, vibration sources are summarized as follows.
Fig. 2 Arrangement of accelerometers
Fig. 3 Frequency characteristics of accelerometer
