Design of ultrasonic emission and control circuit based on S3C2440A

　　With the development of science and technology, high temperature, high pressure, high speed and high load have become important symbols of modern industry, but its realization is based on the high quality of materials. To ensure this excellent quality, various methods must be used. A testing method that destroys the original shape of the product without changing its performance and performs 100% testing on the product to ensure its reliability and safety. This technology is non-destructive testing technology.

　　Ultrasonic testing occupies a major position in non-destructive testing and is widely used in fields such as metals, non-metallic materials, and medical instruments. In recent years, information technology based on microelectronics and computer technology has developed rapidly, and ultrasonic non-destructive testing instruments have also received unprecedented development momentum. In order to improve the reliability of detection and improve detection efficiency, digital, intelligent, automated, and image-based instruments have been developed. Ultrasonic instrument is an important development trend in the field of non-destructive testing today. However, traditional ultrasonic detectors have poor accuracy, low precision, large size, high power consumption, and human-machine interface problems.

　　Unfriendly and other issues. The ultrasonic emission and control circuit is based on an ARM-based ultrasonic detection system, with an ARM microcontroller as the core and using C language programming to easily realize the adjustment of the emission frequency and excitation voltage pulse amplitude.

　　1 Overall design structure of ultrasonic detection system

　　The overall structural block diagram of the ARM ultrasonic detection system is shown in Figure 1. The system mainly consists of 3 parts: ultrasonic front-end transmitting and receiving circuit, DSP and ARM processor.

　　The ultrasonic front-end transmitting circuit is responsible for generating ultrasonic waves with adjustable excitation pulse voltage and repetition frequency. The receiving circuit first processes the reflected weak signal through amplification, filtering and other circuits, then collects the signal through the A/D conversion circuit and sends the collected signal to the DSP through the data buffer FIF0.

　　DSP receives the data buffered by FIF0 from the A/D converter, and mainly completes signal processing algorithms with complex calculation structures to improve the accuracy and data processing capabilities of ultrasonic flaw detection instruments.

　　The ARM processor mainly completes two functions: one is the control function, which adjusts the width and repetition frequency of the excitation pulse and the amplification factor of the amplification circuit; the other is the real-time display, storage and external communication of signals. The ARM microprocessor uses the 16/32-bit RISC microprocessor S3C2440A based on ARM920T. Its core frequency is up to 400 MHz, it has low power consumption, small size, many integrated peripherals, and good data processing capabilities, so it can be widely used in handheld devices.

　　2 Ultrasonic transmitter circuit

　　According to different conditions such as the material and thickness of the tested part, the frequency and emission voltage of the corresponding ultrasonic probe required are also different. The frequency of the emitted ultrasonic wave is generally several MHz, the high-voltage excitation pulse is generally tens to hundreds of volts, and the rise time of the pulse does not exceed 100 ns. According to spectrum analysis, there is an optimal relationship between the frequencies of the excitation pulse width probes. When the pulse width meets this relationship, the receiving probe has the best signal quality. The relationship is:

　　In the formula, f0 is the probe frequency and 2a is the pulse width. The probe frequency selected for this design is 2.5 MHz, and the pulse width determined by equation (1) is 600 ns, so the discharge time should be controlled at 600 ns as much as possible.

　　There are many types of ultrasonic flaw detection methods. In practical applications, most of them use the pulse reflection method, and their transmitting circuits are mostly non-tuned. The ultrasonic transmitting circuit is shown in Figure 2. The circuit consists of an adjustable high-voltage power supply, resistors R1 and R2, energy storage capacitor C, insulated gate bipolar transistor (IGBT) VQ, fast recovery diodes VD1, VD2 and probe. The equivalent resistance of the diode is R3, and the equivalent resistance of the switch is The resistor is R4. The PWM module of the ARM microprocessor generates pulses with adjustable frequency and duty cycle, which are sent to the gate of the switching tube VQ through the IGBT drive and protection circuit to form the control pulse V1. When V1 has a negative pulse, the IGBT is turned off, and the high-voltage power supply charges the capacitor C through R1 and VD2. The charging time constant is τ1=C(R1+R3). When t>5τ1, the capacitor C is considered to be full. When V1 is a positive pulse, the IGBT is turned on, and the capacitor C discharges the probe through the switching tube VQ, R2 and diode VD1. The discharge time constant is τl=C(R2+R3+R4). The ultrasonic probe generates ultrasonic waves of a certain frequency after receiving the excitation of high-voltage negative pulses.

　　Function of components in the circuit:

　　1) Resistor R1 is used to limit the charging current of capacitor C from the high-voltage power supply during charging, that is, it plays a current limiting role and reduces the impact on the power supply when the transmitting unit is working. From this point of view, the larger the resistance of resistor R1, the better. . On the other hand, the repetition frequency f of the circuit is relatively high. In order for the capacitor C to be fully charged before triggering, CR1<1/5f must be satisfied. So choose the appropriate resistance value of resistor R1.

　　2) Resistor R2 has two functions: one is to adjust the discharge time and transmission power, and the other is to act as a damping resistor to adjust the ultrasonic pulse width. The smaller the resistance of R2, the smaller the transmit power and the narrower the transmit pulse; the greater the resistance of R2, the greater the transmit power and the wider the transmit pulse.

　　3) The fast recovery diodes Vd1 and Vd2 filter out the charging pulse, so that point A only has the negative voltage excitation pulse during discharge.

　　During charging, the relationship between current i and voltage UR is as shown in equations (2) to (3).

　　The circuit board developed can excite the probe to generate ultrasonic waves of 0.5 to 10 MHz. The excitation pulse voltage can reach up to 830 V, and the rise time of the pulse is less than 50 ns.

　　3 Generation of PWM pulses based on ARM

　　ARM embedded processor is a high-performance processor with extremely low power consumption and extremely low cost. It has fast computing speed and high precision. It is also easy to transplant the real-time operating system and truly becomes a real-time multi-tasking system. The S3C2440A built-in PWM pulse module contains 4-channel 16-bit timers with programmable duty cycle, frequency, and polarity, and has automatic reloading and double buffering functions. The main frequency FCLK is up to 400M-Hz, and the PCLK used by APB bus equipment is up to 68 MHz. The specific process is: first, turn on the auto-reload function, and set various parameters of the PWM pulse through the PWM register, such as the timer configuration register (TCFGn), timer control register (TCON), timer count buffer register (TCNTBn), Settings for timer comparison buffer register (TCMPBn), timer count observation counter (TCNTOn), etc. Secondly, set the manual update bit of the corresponding timer, and then set the start bit. After the waiting time, the timer starts counting down. When the values ​​of TCNTn and TCMPn are the same, the logic level of TOUTn changes from low to high. When TCNTn is 0, TCNTn is automatically reloaded with the value of TCNTBn. If you want to reset the initial value of TCNTn, you need to perform manual update.

　　By using TCMPBn to perform the PWM function, the frequency of PWM is determined by TCNTBn. The double buffering function allows TCMPBn to be rewritten by the ISR or other programs at any point in the current PWM cycle for the next PWM cycle.

　　4 High-voltage power supply and its control

　　The ultrasonic transmitting circuit has high requirements for the excitation voltage pulse, which requires a certain amplitude. The smaller the pulse width, the better, and it must have a certain transmit power. This determines the sensitivity of ultrasonic flaw detection and is also related to the depth of flaw detection of the workpiece. If you want to penetrate thicker workpieces, you need to convert greater electrical power into acoustic power. The transmitting power is:

　　In the formula, uA0 is the instantaneous voltage when the capacitor is discharged, C is the capacitor capacity, t is the discharge time, and is the effective power.

　　When the discharge time constant is determined, the discharge time and C are determined. Therefore, increasing the transmit voltage is the main way to increase the transmit power. According to the discharge voltage formula, in addition to the influence of various resistors in the circuit, the voltage of the high-voltage power supply is a major factor. But the voltage cannot be too high, otherwise the piezoelectric chip will accelerate aging. Generally, the emission voltage does not exceed 1800 V.

　　The high-voltage power module from Ultravoh Company of the United States is used here. Among them, the model of the "V" series is 1V12-P0.4 power module, which can fully meet the needs of this design. Its input voltage is 12 V, the output voltage is 0~1000 V, the control voltage is 0~5 V, and the power is 0.4W. Low power consumption, small size, light weight, and equipped with output voltage monitoring and self-protection circuit. The high voltage power supply control circuit is shown in Figure 3.