Design of vibration standard device based on random comparison method

2.1. Hardware architecture

DH-11 vibration standard device was composed of PXI cage, PXI controllers, PXI-4461 data acquisition card, standard vibrating transducer, vibrostand, power amplifier, charge amplifier and measuring instruments, which is shown in Fig. 1.

Fig. 1. Hardware architecture of vibration standard device

The dynamic data acquisition card PXI-4461 is a dedicated module for sound and vibration analysis. Its main functions are: 2 synchronous analog inputs, measuring range ±316 mV-42.4 V; 2 synchronous analog outputs; 24-bit resolution; 204.8 kS/s maximum sampling rate. The two synchronous analog inputs are used for the acquisition of the vibration signals of the standard vibration transducer and the calibrated transducer respectively; one synchronous analog output is used as the signal output of the power amplifier to control the vibration of the vibration table. Standard shakers (including signal generators and power amplifiers) produces motion with adjustable frequency and amplitude as the mechanical input to the transducer being calibrated. The vibration parameter measurement section and other supporting instrument combinations (including standard transducers, charge voltage converters, data acquisition units, and computers) are used to accurately measure vibration parameters such as amplitude peak, frequency, and period. The system software made PXI 4461exporting vibration analog signal as input signal of power amplifier to control vibrostand. The output signals of standard vibrating transducer and be calibrated vibrating transducer are amplified by charge amplifier, then entered into PXI 4461 and measuring instruments.

The standard transducer uses the 8305 vibration transducer of Danish B&K company. The standard vibrating transducer 8305 was made from ceramic material with high stability. Design with innovation can reduce environmental impact and have long-term stability. The calibrated transducer and the standard transducer “back to back” are mounted in the center of the vibrating table. The controller and the data acquisition card are integrated through the PXI chassis. The function is to acquire the signals of the standard transducer and the calibrated transducer. According to the input command, the control module outputs various signals required for transmission to the power amplifier. The power amplifier is used to amplify the signal generated by the controller to drive the shaker.

2.2. Software architecture

The system software includes OS and an application. OS adopts WindowsXP. The application was designed using LabVIEW 2013. The application includes self-checking module, management module and calibration module, which is shown in Fig. 2.

The calibration of the vibration transducer is mainly to calibrate the transducer's normalized sensitivity value, frequency response and nonlinearity.

The automatic measuring module of the vibrometer and vibrostand mainly includes the selection of the verification procedure, the parameter configuration of the working vibration meter, the test procedure, the configuration of the test process parameters, and the automatic verification of the working vibration meter.

In the software menu, select the “Data Management” menu to query the previous verification results of the transducer and the working vibration meter.

Fig. 2Software architecture of vibration standard device

3.1. Work process

Comparative method calibration as shown in Fig. 1. This method is also suitable for calibrating speedometers or displacement meters.

The workflow is shown in Fig. 3. First, select the verification mode (manual calibration, automatic transducer calibration, automatic vibration meter calibration and vibration table automatic calibration) through the software user interface. The system sets different vibration parameters according to different verification methods. The software system reads the amplitude and frequency of the vibration generated by the standard vibration table, sends it to the data control module to generate an exciting signal, and then sends the exciting signal to the power amplifier. After the signal is amplified, the standard vibration table vibration is driven. At the same time, the standard transducer and the calibrated transducer mounted on the standard vibration table transmit the signal of the test to the data acquisition module, and compare the signal collected by the standard transducer with the expected ideal signal. If the error requirement cannot be met, the signal output should be readjusted within the allowable range of the system according to the error level until the error requirement is met. If the error requirements are met, the "comparative method" calibration process can be entered. The standard transducer and the calibrated transducer transfer the test data to the data acquisition module. The software system reads the test data of the standard transducer and the calibrated transducer, and displays the calibration result in the form of curve and tabular data respectively. The curve includes the time domain curve, the amplitude frequency characteristic curve, the phase frequency characteristic curve and the amplitude nonlinear curve of the actual acquisition of the transducer. After the calibration is completed, the system prompts whether to save the result to the database. After the save operation is completed, the system continues to prompt whether to generate the calibration certificate. The system can automatically generate the calibration certificate according to the set parameters and test data. The calibration certificate format is the standard Microsoft Word format.

3.2. Random comparison calibration

In the actual engineering, the pure sinusoidal vibration is almost non-existent, and often it is the shock transient and random vibration which are large and small, fast and slow, and the vibration changes frequently. When these vibrations are measured with absolute vibration transducers, their response does not follow the amplitude-frequency characteristic law derived from steady-state sinusoidal vibration. That is to say, the transient response of the transducer is different from the steady state response. If a transducer calibrated in a sinusoidal steady state is used to measure transient and random vibrations, and the steady state calibration data is used to calculate transient values and random vibration values, a certain amount of error is inevitable.

Fig. 3Flow chart of the calibration

Because the transducer is used in a random vibration environment, it is more practical to use a random vibration method to calibrate the vibration transducer.

When the accelerometer is calibrated using the FFT random calibration method, both charge amplifiers have errors due to the dual-channel FFT dynamic signal analyzer. Therefore, using only one comparison method will result in a large calibration error. To reduce the effects of these systematic errors, a switching comparison method is used to eliminate these errors. The method is shown in Fig. 4. Where a is vibration acceleration, $S_u\left(f\right)$ is charge sensitivity of the calibrated transducer, $S_r\left(f\right)$ is charge sensitivity of standard transducer, $H_A\left(f\right)$ is frequency response function of charge amplifier A, $H_B\left(f\right)$ is frequency response function of charge amplifier B, $U_{A1}\left(f\right)$ is the output voltage of charge amplifier A at the time of the first measurement, $U_{B1}\left(f\right)$ is the output voltage of charge amplifier B at the time of the first measurement, $U_{A2}\left(f\right)$ is the output voltage of charge amplifier A at the time of the first measurement, $U_{B2}\left(f\right)$ is the output voltage of charge amplifier B at the time of the first measurement.

At the first measurement, the calibrated transducer is mounted on the upper end of the bracket, the standard transducer is installed at the lower end, and the frequency response function between the two transducers is measured; after the first measurement, the cable connector of the transducer is exchanged, and then the second measurement is performed again as the first method.

Fig. 4Schematic diagram of FFT random switching method

It can be known from the principle of FFT random switching method:

From Eqs. (1) and (2), the following equation can be obtained:

After the second measurement, the following equation can be obtained:

From Eqs. (4) and (5), the following equation can be obtained:

By dividing the Eq. (3) by Eq. (6), the sensitivity of the calibrated transducer can be obtained:

The sensitivity of standard transducer $S_r\left(f\right)$ can be obtained by point-by-point calibration on low frequency, intermediate frequency and high frequency vibration standards. The sensitivity of the calibrated transducer $S_u\left(f\right)$ can be obtained by Eq. (7).