Analysis of the switching process of crystal diodes

0 Introduction

Crystal diode switching circuits are widely used in digital systems and automation systems. In the crystal diode switching characteristic experiment, there is a time delay or lag between the output and input during the switching conversion process. The study of the crystal diode switching characteristics is mainly to study the length of time required for the switching state conversion process. The DQ series diodes developed by Microsemi have the advantages of ultra-fast soft recovery, which greatly improves the switching speed of the crystal diode. With the development of technology, the new SiC Schottky barrier diodes can greatly reduce switching losses and increase switching frequency compared with traditional power diodes using Si or GaAS technology. In AM-LCD, the barrier diodes made of C60 are used as switches of the active matrix, and their working speed is also very fast. When used as a switching device, the shorter the time required from on to off or from off to on, the better. Therefore, the reasons for the speed of the crystal diode switching need to be carefully analyzed and discussed. On this basis, through a simple experimental circuit, the appropriate pulse signal and load are selected according to the parameters of the crystal diode, and the delay of the diode switching conversion process can be clearly observed.

1 Diode switching characteristics

In the gate circuit of digital electronic technology, under the action of pulse signals, the diode is sometimes turned on and sometimes turned off, which is equivalent to the "on" and "off" of the switch. The time taken by the diode from cutoff to opening is called the turn-on time, and the time taken from opening to cutoff is called the turn-off time. To study its switching characteristics is to analyze the speed of the on and off conversion. When the frequency of the pulse signal is very high, the rate of change of the switch state is high. As a switching device, the faster the switching speed, the better, but the diode is an electronic device made of semiconductor materials such as silicon or germanium through special processes. It has a maximum limit working speed. When the switching speed is greater than the limit working speed, the diode cannot work normally. In order for the diode to work safely, reliably and quickly, the conversion frequency of the high and low levels of the external pulse signal must be less than the frequency of the diode switch.

As shown in Figure 1, a pulse signal Vi is applied to the input end, and its amplitude is +V1 and -V2. When the voltage applied to the two ends of the diode is +V1, the diode is turned on; when the voltage applied to the two ends of the diode is -V2, the diode is cut off, and the input and output waveforms are shown in Figure 2. When the voltage across the diode changes from forward bias +V1 to reverse bias -V2, the diode is not cut off instantaneously, but it is maintained for a period of time ts before the current begins to decrease. After tf, the reverse current is equal to the reverse drift current I0 in the static characteristic, and its value is very small. ts is called the storage time, tf is called the fall time, and ts+tf=trr is called the turn-off time. When the voltage across the diode changes from reverse bias -V2 to forward bias +V1, the diode is not turned on instantaneously, but it is stably turned on after the conduction delay time and rise time. This period of time is called the turn-on time. Obviously, the turn-on and turn-off moments of the diode always lag behind the moments when the high and low levels are applied to its two ends. The turn-on time of the diode from cutoff to forward conduction is very small compared to the turn-off time from conduction to cutoff, and its effect on the switching speed is very small. In the analysis and discussion, the effect of the turn-off time is mainly considered.

2 Analysis of the causes of diode switching time delay

There are two kinds of current in semiconductors. The current formed by different carrier concentrations is diffusion current, and the current formed by the electric field is drift current. When a P-type semiconductor and an N-type semiconductor are brought close together, a diffusion motion that decays exponentially will occur at the contact point of the two semiconductors due to the difference in carrier concentration. During the diffusion process, electrons and holes will recombine when they meet, and an internal electric field will be generated at the junction. The internal electric field will prevent the diffusion motion and promote the drift motion. Finally, the diffusion motion and the drift motion reach a dynamic balance. When the voltage applied to both ends of the diode changes, on the one hand, the width of the PN junction changes, and the number of donor anions and acceptor cations in the barrier region will change; on the other hand, the number of diffused majority carriers and drifted minority carriers will also change due to voltage changes. This situation is similar to the effect of a capacitor, which is represented by barrier capacitance and diffusion capacitance respectively. When a forward voltage is applied to both ends of the diode, it weakens the internal electric field of the PN junction, the diffusion motion is strengthened, the drift motion is weakened, the dynamic balance of diffusion and drift is destroyed, and the diffusion motion is greater than the drift motion. As a result, the majority holes in the P region flow to the N region, and the majority electrons in the N region flow to the P region. The electrons entering the P region and the holes entering the N region become minority carriers in the region respectively. Therefore, there are more minority carriers in the P region and the N region than when there is no external voltage. These extra minority carriers are called unbalanced minority carriers. Under the action of the forward voltage, the holes in the P region cross the PN junction and accumulate on the boundary of the N region. The electrons in the N region cross the PN junction and accumulate on the boundary of the P region. These unbalanced minority carriers diffuse in the N region based on the concentration difference during accumulation, forming a certain concentration gradient release, with high concentration near the boundary and low concentration far from the boundary. In the process of holes diffusing to the N region, some of them meet and recombine with the majority electrons in the N region. The farther away from the PN junction boundary, the more holes recombine. Vice versa, when electrons diffuse to the P region, some of them meet and recombine with majority holes in the P region. The farther away from the PN junction boundary, the more electrons recombine. When the diode is forward-conducting, unbalanced minority carriers will accumulate near the boundary, resulting in a charge storage effect.

When the input voltage suddenly changes from a high level to a low level, the electrons stored in the P region and the holes stored in the N region will not disappear instantly, but will gradually decrease through two ways. First, under the action of the reverse electric field, the electrons in the P region are pulled back to the N region, and the holes in the N region are pulled back to the P region, forming a reverse drift current I0. Secondly, they recombine with the majority carriers and disappear. Before these stored charges suddenly disappear, the width of the PN junction barrier region remains unchanged and is still very narrow, so the reverse current is large and basically remains unchanged at this time. After a period of time, the charges stored in the P region and the N region have been significantly reduced, and the barrier region gradually becomes wider. After a period of decline, the reverse current gradually decreases to the value of the normal reverse saturation current I0, and the diode is cut off. Therefore, the diode turn-off time is also called the reverse recovery time. When the input voltage suddenly changes from a low level to a high level, the PN junction will become narrower from wide, and the diode will only turn on after the barrier capacitor is discharged. The turn-on time is very short than the turn-off time and can be ignored. The current flowing through the diode increases with the increase of the diffused stored charge and gradually reaches a stable value.

The switching time delay of the diode during the switching process is actually caused by the capacitance effect of the PN junction. The transient switching process of the diode is the charging and discharging process of the PN junction capacitance. The transition of the diode from cut-off to conduction is equivalent to the charging of the capacitor. The transition of the diode from conduction to cut-off is equivalent to the discharge of the capacitor. The diode junction capacitance is small, the charging and discharging time is short, and the transition process is short, then the transient switching characteristics of the diode are good, and the switching speed is fast. The delay time is the time required for the capacitor to charge and discharge. The length of the delay time is determined by the structure of the diode itself and is related to the external circuit. The larger the PN junction area of ​​the diode, the more charge is stored in the tube, and the longer the delay time. In addition, the larger the forward current determined by the external circuit, the more charge will be stored, and the longer the turn-off time; the larger the reverse current, the faster the stored charge disappears, and the shorter the turn-off time. In order to increase the switching speed and reduce the delay time, the junction area of ​​the general switch tube is made smaller to store less charge. At the same time, through the "gold doping" inside the diode, the stored charge can be quickly recombined and disappear, reducing the delay time.

3 Experimental observation of the switching conversion process of the crystal diode

In order to observe the switching characteristics of the diode, the experiment can be carried out according to the circuit shown in Figure 1. First, determine the pulse signal applied to both ends of the diode, and its amplitude and period must be appropriate. Otherwise, it may take a long time to debug before observing the delay in the switching process of the diode, and it may also cause damage to the diode. The selection of the pulse signal should be based on the main working parameters of the diode, such as the maximum forward working current of the diode, the maximum reverse working voltage of the diode, and the reverse recovery time. Based on these parameters, determine the amplitude of the pulse signal. The signal period must be greater than the reverse recovery time trr. Select a certain load connection circuit, and observe the delay of the diode switch conversion time through a dual-trace oscilloscope. Change the signal period and load respectively, record multiple experimental results, and further analyze the relationship between the delay time of the diode switch conversion process and the change of the pulse signal period and external load. The delay time has a minimum value for both the diode junction area and the load resistance. When designing a switch circuit, the diode junction area and the load resistance should be selected from the optimal values ​​corresponding to the extreme point. The length of the N region also has an optimal value. In theory, it should be the length value when the device is loaded at the required critical breakdown voltage and is just in the punch-through state. The lengths of the P region and the N region do not have much effect, but should be slightly larger than their respective punch-through lengths. The concentration should be as high as possible, and the lower the doping concentration of the N region, the better.

4 Conclusion

The structure of the crystal diode determines its characteristics when used as a switch. It takes a while for the gate to open and close in the gate circuit of digital electronic technology. The length of time varies for tubes with different structures. With the rapid development of modern electronic technology, the switching speed of crystal diodes is required to be faster and faster. Therefore, the design requirements for device structure and working circuit are also getting higher and higher. In the experiment of studying the delay process of crystal diode switching time, the cycle, amplitude and circuit load of the input signal have a great influence on the observation of delay time. Only by multiple experiments of a certain switching circuit can the delay of the switching conversion process of the diode be clearly observed.