Design and implementation of front-end circuit of virtual ultrasonic nondestructive testing system

O Introduction
With the development of modern industry and science and technology, nondestructive testing technology plays an increasingly important role in the operation of equipment and equipment, product quality assurance, productivity improvement, cost reduction and other fields. Nondestructive testing has also developed into an independent comprehensive discipline, and ultrasonic flaw detection technology occupies an extremely important position in the field of nondestructive testing and has been widely used in many fields.
Since ultrasonic nondestructive flaw detection equipment has different requirements for probes and different processing algorithms for received echo signals in different application occasions, a certain type of nondestructive flaw detection equipment can usually only be adapted to one or several application occasions. In order to make ultrasonic flaw detection equipment have a better application range, higher processing algorithms and faster update cycles, virtual ultrasonic nondestructive flaw detection equipment can be used. The virtual ultrasonic flaw detection system uses the function of the computer display to simulate the control panel of the traditional flaw detector, outputs the test results in various forms, and uses software functions to realize the calculation, analysis and processing of digital signals. The input and output (I/O) interface device is used to complete the signal acquisition, measurement and debugging, thereby completing the ultrasonic nondestructive flaw detection system with various test functions. This system is a product of the combination of virtual instrument technology, digital technology and traditional ultrasonic flaw detection system. When designing a virtual digital ultrasonic system, it is necessary to combine the design of the analog channel in the traditional ultrasonic flaw detection system, because any ultrasonic flaw detection system must include ultrasonic transmission circuit, receiving circuit and signal conditioning circuit to further carry out subsequent processing work. The design of these circuits will directly affect the reliability and test accuracy of the entire ultrasonic flaw detection system. The circuit designed here is the front-end circuit before A/D conversion.

1 System composition of virtual ultrasonic nondestructive testing equipment
Ultrasonic testing generally refers to the technology of making ultrasonic waves interact with test pieces, studying the reflected, transmitted and scattered waves, detecting macroscopic defects, measuring geometric characteristics, detecting and characterizing changes in organizational structure and mechanical properties, and then evaluating their specific applicability. The devices used for ultrasonic testing of solid materials can be divided into three categories according to the parameters they indicate: the first category indicates the penetration energy of sound, which is often used in the penetration method; the second category indicates the formation of standing waves in the test piece by ultrasonic continuous waves with variable frequencies, which can be used for resonance thickness measurement, but is rarely used at present; the third category indicates the amplitude and running time of reflected sound waves, which is often used in the pulse reflection method. In view of the widespread use of the pulse reflection method at present, the virtual instrument is designed using the reflection detection method.
The basic principle of the pulse reflection method is shown in Figure 1, and generally only one probe is used for both transmission and reception. When the workpiece is intact, only the initial wave T and the bottom echo B are displayed on the display, as shown in Figure 1 (a). When there is a small defect in the workpiece that is smaller than the cross section of the sound beam, there is also a defect echo F between the initial wave T and the bottom wave B, as shown in Figure 1(b). The size of the defect can be evaluated based on the height of the defect wave F, and the buried depth of the defect can be obtained based on the time difference between the defect echo F and the initial wave T. When there is a defect echo, the bottom echo height decreases. When there is a large defect in the workpiece that is larger than the cross section of the sound beam, the ultrasonic beam is completely reflected by the defect, and only the initial wave T and the defect echo F are displayed on the display, and the bottom wave B disappears, as shown in Figure 1(c).

The overall structure of the virtual ultrasonic flaw detection system is shown in Figure 2. The system uses the AT89C52 single-chip microcomputer as the core control device. Data acquisition is completed by a computer combined with a dedicated data acquisition card. Then, computer software is used to calculate, analyze and process the data. The panel design of the flaw detection system and the design and construction of some functions are combined with LabVIEW application software to display the test results. The single-chip microcomputer controls the transmitting circuit to transmit a 400 V high-voltage narrow pulse to stimulate the probe to generate ultrasonic waves, and then switches the switch to the signal receiving circuit. At the same time, the data acquisition card is started to collect data. The communication between the single-chip microcomputer control system and the computer (such as pulse repetition frequency, pulse duty cycle and gain adjustment, etc.) is carried out using a USB interface.

[page]

2 Transmitting circuit and receiving circuit
2.1 Ultrasonic transmitting circuit
The ultrasonic transmitting circuit is the key part of the pulse echo ultrasonic flaw detector and has a great influence on the performance of the ultrasonic flaw detection system. Transmitting circuits are usually of two types: tuned and untuned. The tuned circuit has a tuned coil, and the resonant frequency is determined by the inductance and capacitance of the tuned circuit. The ultrasonic pulse frequency band is narrow. The untuned circuit transmits a peak pulse with a wide pulse frequency band, which can adapt to probes with different frequency bands. At this time, the ultrasonic frequency emitted is mainly determined by the inherent parameters of the piezoelectric chip. The system uses the B5S single crystal longitudinal wave straight probe of German K.K Company (Krautkramer). The probe has a nominal frequency of 5 MHz, a working range of 15 to 6 000 mm and a near field length of 110 mm. In order to facilitate the flexible debugging of the system, an untuned transmitting circuit is used, and its pulse control parameters can be easily modified and set through the core controller AT89C52 single chip microcomputer. The transmitting circuit is shown in Figure 3.

In the transmitting circuit, the high voltage DC passes through the current limiting resistor R1, and the DC blocking capacitor C is charged to VH. In the commonly used ultrasonic testing system, the VH voltage ranges from tens of volts to hundreds of volts. In order to fully stimulate the piezoelectric performance of the probe, a 400 V high voltage DC power supply module is used. Thyristor Q is a high-speed switching device controlled by the pulse signal generated by the single chip microcomputer. At a low level, the thyristor Q is in the cut-off state, and the capacitor C is charged to 400 V. When the high level signal arrives, the thyristor Q is in the on state, causing the capacitor C to discharge through the thyristor Q and the adjustable resistor R2, and generating a high voltage pulse on R2 to excite the probe. The variable resistor R2 determines the damping of the circuit, and the intensity of the emission can be changed by changing the resistance value of R2. When the resistance is large, the damping is small, the emission intensity is large, and the resolution of the instrument is low. It is suitable for detecting specimens with large thickness and low resolution requirements. When the resistance is small, the damping is large and the resolution is high. It is used when detecting near-surface defects or when the resolution requirements are high. The design uses a bidirectional thyristor BTl36-600, which has a reverse peak voltage of 600 V and a rated average current of 4 A. In order to fully drive the thyristor, a dual-channel power MOSFET driver ICL7667 is specially selected to design the drive circuit. The high-voltage pulse and ultrasonic signal generated by the transmitting circuit are shown in Figure 4.

2.2 Limiting and receiving amplifier circuit
When the detection range is large, the reflected wave signal of deep defects or bottom waves is very weak, so high-gain amplification processing is required before processing. Since the probe is a transmitter-receiver integrated, the transmitting signal is very strong, and it acts on the receiving circuit at the same time. In addition, strong interference may be added during the test process. Therefore, in order to protect the amplifier circuit from damage and keep the amplifier circuit in a linear dynamic range, it is necessary to limit the receiving signal before amplification. The limiting circuit is shown in Figure 5. The resistor R3 in the figure should be large enough relative to the adjustable resistor R2 in the transmitting circuit to eliminate the load effect of the receiving circuit on the transmitting circuit. Diodes with large forward current (such as 2K61701) D2 and D3 are selected to form a bidirectional limiting circuit to prevent the high-voltage pulse in the transmitting circuit from entering the back-end receiving circuit. In this way, the output of the limiting circuit is about ±0.7 V, achieving the expected effect of the circuit. After the amplitude limiter, the amplifier circuit is next. In order to measure the amplitude change value, the echo signal passes through a calibrated attenuator before entering the amplifier, so as to quantitatively adjust the signal amplitude to adapt to different signal ranges. This design uses the voltage-controlled gain amplifier AD603 launched by AD (ANALOG DEVICES) to design the programmable gain amplifier circuit module. AD603 has the characteristics of linear decibel, low noise, wide bandwidth, high gain accuracy and flexible gain control. Its impedance of up to 50 MΩ can ensure that the signal is fully loaded into the subsequent circuit. The AD603 programmable gain schematic is shown in Figure 6, and its pin description is shown in Table 1.

[page]

AD603 provides accurate, pin-selectable gain, and its gain is linearly variable, and has high stability when temperature and power supply voltage change. The gain change range is 40 dB, the gain control conversion ratio is 25 mV/dB, the response speed is 40 dB, and the change range takes less than 1μs. As shown in Figure 6, AD603 contains a 0-42 dB variable attenuator composed of a seven-level R-2R ladder network and a fixed gain amplifier. The gain of this fixed gain amplifier can be changed by connecting different feedback networks to select different gain change ranges of AD603. This variable gain function of AD603 is unmatched by other operational amplifiers.

After the ultrasonic echo signal enters the attenuator from VINP and is attenuated, it is amplified by the fixed gain amplifier. The gain control of the attenuator is completed by the control voltage VG. VG is the gain control voltage of the differential input, that is, the difference between GPOS and GNEG, and the range is -0.5 to +0.5 V. The gain of the fixed gain amplifier can be changed by connecting different feedback networks to select different gain change ranges of AD603.
(1) When the output terminal VOUT of AD603 is short-circuited with the feedback terminal FDBK, Gain (dB) = 40VG + 10; at this time, the gain range is -10 to +30 dB, and the bandwidth is 90 MHz.
(2) When the output terminal VOUT of AD603 and the feedback terminal FDBK are connected to the feedback resistor, Gain (dB) = 40VG + 20; at this time, the gain range is 0 to +40 dB, and the bandwidth is 30 MHz.
(3) When the feedback terminal FDBK is grounded, Gain (dB) = 40VG + 30; at this time, the gain range is 10 to 50 dB, and the bandwidth is 9 MHz.
It can be seen that the gain control of AD603 is quite flexible. In actual use, multiple AD603 chips can be connected in series to achieve greater amplification and dynamic range control. This design first used a single AD603 as a preamplifier, and then used two AD603 chips in series as an AGC automatic gain amplifier. The preamplifier is shown in Figure 7.

As shown in Figure 7, in the design, the output terminal VOUT and the feedback terminal FDBK are connected with a potentiometer R3, which can flexibly select the gain range. By adjusting the potentiometer R2, the voltage between GPOS and GNEG can be adjusted between 0 and 0.5 V. If the resistance value of the potentiometer R3 is adjusted to 0, the gain range of the amplifier is 10 to 30 dB. The design principle of the AGC automatic gain amplifier circuit is similar. Due to space limitations, it will not be repeated here. The output waveform of the circuit is shown in Figure 8.

3 Bandpass filter circuit
The high-frequency amplifier circuit will introduce noise in the process of amplifying the echo signal. In order to control the introduced noise and improve the overall signal-to-noise ratio of the system, a bandpass filter needs to be designed to filter out the noise. This design uses the MAX4104 produced by MAXIM to design a bandpass filter circuit with a center frequency of 5 MHz, a gain of K=4, a quality factor of Q=5, and a bandwidth of B=1 MHz. Figure 9 shows the bandpass filter circuit and the echo signal after filtering.

4 Conclusion
The design and implementation of the front-end circuit of virtual ultrasonic nondestructive testing equipment are introduced. In order to verify the function of the designed circuit, the CSK-IA test block specified in the JB/T4730.3-2005 standard is used for experiments. The results show that these circuits can well complete the signal conditioning task before A/D conversion, and the circuit performance is stable and reliable.