Research and implementation of 100 A pulsed current source pulse edge compression

Transcription

1 April 016, 3(: The Journal of China Universities of Posts and Telecommunications Research and implementation of 100 A pulsed current source pulse edge compression Tian Xiaojian (, Zhang Dapeng, Mudare Joseph, Gao Fubin College of Electronic Science and Engineering, Jilin University, Changchun 13001, China Abstract In order to increase the usefulness of pulsed current source in engineering practice, research and study was carried out on how to increase the pulse current amplitude, reduce the rise /fall time of output pulse and MOSFET switching losses, etc. Through the analysis of the pulsed current source works theory and the mathematical derivation of the circuit model, the deduction and calculation of the pulse edge compression control methods, and improve the overall circuit structure and optimize the manufacturing process according to the theory. The following indicators was realized: the output pulse current amplitude can be up to 100 A, the shortest pulse rise / fall time was 18.8 ns and 16.1 ns respectively when the maximum amplitude output, the pulse width could be narrowest to 40 ns, repetition frequency could achieve 10 Hz to 10 khz, MOSFET switching losses decreased by 30.9 %. This pulsed current source can be used, not only as the power supply for the ordinary high speed narrow pulse width laser diode, but also as an ideal drive power for the high energy, narrow width pulse laser diode. Keywords optoelectronics, high-power narrow-width pulse driver, 100 A pulse current, short rise/fall time 1 Introduction As the development of laser technology and applications continue to expand, the scope of application of pulsed semiconductor laser has been extended to many other areas of optoelectronics, becoming the core technology of today s photonics. It has been widely used in optical communications, optical fiber sensing, pump light source, laser ranging, laser-guidance, laser aiming and laser radar and some other fields [1 3]. High peak power, short pulse width and steep rising pulse driving can increase the laser application distance and improve resolution of the relevant sensor [4]. For pulsed laser ranging, shortening the rise time of the laser pulse can improve the measurement accuracy [5]. Different peak power output pulsed semiconductor laser have different areas of application. The light pulse emitted from a pulsed semiconductor laser are modulated by electrical pulse generated by the driving power source. In addition to the characteristics of the Received date: Corresponding author: Tian Xiaojian, tianxj@jlu.edu.cn DOI: /S ( semiconductor laser itself, the driving current s parameters generated by pulsed current source have a significant impact on the laser output [6 7]. Domestic research also has made some remarkable progress on pulsed current source, some research institutions have achieved control the pulse current width less than 50 ns, but the pulse peak can only reach tens amperes [7 10]. Seldom can achieve 100 A and nanosecond short pulse current simultaneously, and the research of pulse edge compression technology mainly focused on the small amplitude pulse current, but the compression results were undesirability [11 1]. Based on the analysis of the pulsed current source works theory and improving the overall circuit structure, redesigning the pulsed current source s charge and discharge circuit, and improving the MOSFET high speed switching circuits, the pulsed current source output pulse current with narrowest width of 40 ns, peak current up to 105 A, and continuously adjustable output pulse current amplitude of 0 A~ 100 A, thereby increased the scope of application of pulsed current source. Meanwhile, in this paper, we study the compression method of pulsed current source output pulse

2 74 The Journal of China Universities of Posts and Telecommunications 016 rise / fall time, using the maximum amplitude of the pulsed current source output, pulse rise time can be controlled at 16.1 ns, pulse fall time can be controlled at 18.8 ns, thereby reduced the MOSFET switching losses, and increased the life of the pulsed current source. Discussion and analysis of the pulsed current source applications Pulsed current source is to study the electro-physical science and technology that using some kind of pulse high power devices store up the small instantaneous energy firstly, then carry out a morphological transformation shift or compression adjustment, and finally quickly release to the load [13]. The discharge time of the pulsed current source is very short, but the discharge energy is large, and currently we are unable to use the average power with an equivalent direct power supply system [14]. The diagram of the pulsed current source is shown in Fig. 1. Firstly, the storage network consisting of a group of parallel capacitors is used to store the small instantaneous energy for a short period of time, until the high speed switching circuit triggered to release the energy. The high voltage charging circuit is responsible for supplying the charging voltage for the storage network, the charging circuit consists of the limiting resistor R 1, the storage capacitor C, resistor R and diode D 3. When the high speed switching circuit's MOSFET Q 1 is off, the high voltage charging circuit output current follows the charging circuit to charge the energy storage network. When the Q 1 is turned on, the storage network through the Q 1, ground, the sampling resistor R4, Laser diode (LD and resistor R constitute a discharge circuit for fast discharge. Wherein the pulsed current source output pulse current's width and the rising/falling edge steepness was direct related to the drive current pulse s width and the rising/falling edge steepness provided by the drive circuit, and the output pulse amplitude is direct related to the charging voltage. The output pulse current from the pulsed current source is sampled in real time by the sampling resistor R 4 and external processing circuit by judging the sampled voltage value to decide whether or not switch on the appropriate protection mechanisms A pulse current key technology research Since the output pulse current s repetition frequency is 10 Hz~10 khz, the fundamental signal energy of the pulse current is very strong, and its higher harmonic component energy is very weak. This article ignores the impact of higher harmonics, the pulsed current source s equivalent circuit is shown in Fig.. By analyzing the equivalent circuit model, when pulsed current source is discharging, the discharge circuit can be seen as a zero-input response of RLC circuit [11]. The discharge equation as a second order linear differential equation with constant coefficients Fig. 1 Diagram of the pulsed current source is expressed as di di 1 L + R + i = 0 (1 dt dt C where R is the total equivalent resistance, L is the total equivalent inductance, C is the total equivalent capacitance of the discharge circuit. The two characteristic roots for solving the above differential equation are: r = 1 ( R ( L + ( R ( L ( 1 ( LC and r = ( R ( L ( R ( L ( 1 ( LC, the relationship between R and L/ C determines the pulsed current source output

3 Issue Tian Xiaojian, et al. / Research and implementation of 100 A pulsed current source pulse edge compression 75 current. i When R > LC U 0 rt 1 rt = (e e Lr ( r 1 When R LC, Fig. Equivalent circuit R U t 0 1 L R i = e sin t L 1 R LC L LC L ( (3 The distribution parameters in the circuit will affect the output waveform [9], and the circuit distribution parameters can be affected by a lot of external factors. In this paper, through software simulation, the maximum distributed resistance capacitance and inductance of the circuit were estimated as, R = R P + R 0 + R G Ω, C = C P + C C + C iss + C rss pf, L = L P + L G + L S + L D 8 nh. Because the distributed capacitance and inductance in the circuit cannot be changed after the circuit is designed, Through the model analysis, changing the resistance R can control and adjusted the circuit distribution parameters and can control the effect for waveform effectively. From the parameter calculations it shows that R < L/ C, output currents are obtained using Eq. (3, the oscillation phenomenon will occur in the discharge circuit. Eq. (3 shows that, reducing the parasitic parameters, and increasing the charging voltage U 0, is one way of increasing the output pulse amplitude. In order to increase the output pulse current amplitude, in this paper we proposed an improved charging module for the storage network. Fig. 3 shows the schematic diagram of the improved storage network charging module. Fig. 3 Schematic of storage network charging module The high voltage adjustable regulator TL783 is used as the core of the storage network rectifier circuit, by adding a peripheral voltage compensation circuit and soft-start circuitry, the pulsed current source storage network achieved charging voltage up to 15 V and with changing soft start. TL783 output voltage is R + R 5 R + R 5 U = Uref 1+ = (4 R3 R3 R 3 and R 5 are constant resistors, by adjusting the resistance of R, different output voltages U can be obtained. The minimum value of U is about 1.5 V. In order to broaden the range of output pulse current amplitude, a voltage compensation circuit constituted by the resistor R 1 and Zener diode D 1 was added, to offset the 1.5 V minimum output voltage of the TL783, which is U 0 =U U D. Pulsed current source output pulse current amplitude could be up to 100 A. Because the high power laser production process is very complicated and the lasers distribution parameters are relatively large. Therefore, different types of lasers have different impedances and Lasers are not good pure resistive load. Their inductance, capacitance characteristics have an impact on the current source. In order to measure the maximum short-circuit current of the pulsed current source, a 1 Ω resistor and a high-power high-speed switching diode connected in series was used as the equivalent load. Fig. 4 shows the curve of

4 76 The Journal of China Universities of Posts and Telecommunications 016 the charging voltage and the output pulse current amplitude measured across the 1 Ω resistor. Fig. 5 shows the waveform when the pulsed current source output amplitude was at maximum 100 A, that was measured by a 10 attenuation oscilloscope probe. Table 1 The measurement data of peak current stability Time/h Peak current/a Time/h Peak current/a Study of the pulse edge compression control method Fig. 4 Graph of charge voltage and the output amplitude Fig. 5 Waveform of the maximum amplitude of the pulsed current source output R = Ω, L = 8 nh, C = pf, U = 15 V were substituted into the Eq. (3, and the theoretical maximum value of the pulsed current source discharge current I 118 A was basically the same with the actual max measured value. After preheating for a half-hour the pulsed current source is activated, every half hour the pulsed current source s peak output current was measured and recorded. The testing was done continuously for 8 h, the measurements were divided into three groups and the averaged measured data of the three groups are shown in Table 1. According to the stability formula γ = ( I I ( 100% = Imax Imin I 100%, pulsed current source s peak output current for 8 h stability is.37 %. From the model analysis of the MOSFET as a high-speed switch in Refs. [9 10,15], we know that the pulse width and the steepness of the rising edge of the output pulse current of the pulsed current source is directly related to the pulse width and the steepness of the rising edge of the drive current provided by the MOSFET circuit s drive circuit. Since the turn on and off of the MOSFET is not instantaneous, when the current flowing through MOSFET is rising, the MOSFET tube has a very large switching loss. And the faster the switching frequency, the greater the losses. In this paper, when the output pulse current width is significantly large, the product of the MOSFET turn-on voltage and current is big, and the losses are also big. In order to reduce the switching losses, the pulse rise/fall time must be shortened. Considering the Miller effect [15], when the gate of the MOSFET is turned on, the required drive current I g, is CissUGS + Crss( UGS + UDS Ig = (5 t r Rise time of the discharge circuit is π LC tr = + g(off ( RC (6 Fall time of the discharge circuit is CrssUDS tf = (7 I Switching losses of the MOSFET circuit are 1 PS = UDSIDfsw ( tr + tf (8 C iss, C rss, U GS, U DS, L, C, R, I g(off, I D, f SW representing the input capacitance, feedback capacitance, the gate-source voltage, the drain-source voltage, the total equivalent inductance of the discharge circuit, the total equivalent capacitance of the discharge circuit, total equivalent resistance of the discharge circuit, gate turn-off current

5 Issue Tian Xiaojian, et al. / Research and implementation of 100 A pulsed current source pulse edge compression 77 drain pole output current and output pulse frequency of the MOSFET respectively. Eqs. (5 (7 shows that, as the drive current provided by drive circuit increases, t r and t f decreases. At the same time, reducing the circuit s parasitic parameters, also can reduces t r. In this paper, By using symmetrical components layout and traces to reduce the circuit parasitic parameters, and optimizing the structure of the circuit, the discharge circuit s parasitic parameters were controlled. In the circuit, the total equivalent resistance, R = Ω, L = 8 nh, C = pf, the parameters C iss, C rss, U GS, U DS of the selected MOSFET switching devices are 500 pf, 4 pf, 15 V and 15 V respectively, from Eq. (6 we can calculate π LC tr = + ( RC = 9 1 π ( ns (9 Substituting t r into Eq. (5 we can calculate the MOSFET turn on drive current I g CissUGS + Crss( UGS + UDS Ig = = ( A (10 In order to ensure the circuit is stable and reliable, the instantaneous drive current should reach to 3 times I g [1], the pre-driver chip IXDD404 of the MOSFET can provide 8 A maximum drive current which can meet the requirements. As t r decreases, I g increases, the gate-off voltage of the MOSFET will increase, and according to Eq. (7 t f also decreases correspondingly. In the experiment, a 1 Ω resistor and a high-power high-speed switching diode connected in series was used as the equivalent load, and observing and recording the waveform across the 1 Ω resistor, when the output amplitude of the pulsed current source is maximum, we measured it by 10 attenuation oscilloscope probe. Fig. 6 shows the waveform information before the pulse edge compression control process, the rising time is ns and the falling time is ns. Fig. 7 shows the waveform information after the pulse edge compression control process, the rising time is ns and the falling time is ns. The rise time shortened ns and the falling time shortened ns after the pulse edge compression control process. Under the same output pulse amplitude, frequency and power conditions, the reduction of the MOSFET switching circuit switching losses can be calculated by Eq. (8. 1 UDSIDfsw ( tr + t P f S ( tr + tf = = = P 1 S U ( ( tr tf DSIDfsw tr + t + f % = 30.9% ( Fig. 6 Waveform information of before the pulse edge compression control process Fig. 7 Waveform information of after the pulse edge compression control process Through the equivalent circuit model analysis of the pulsed current source, we know that it is only when the total equivalent resistance, capacitance, and inductance satisfy the condition of R > LC, the pulsed current source has an ideal non-oscillating discharge process. In this paper, by adjusting the coupling resistor can increase the total equivalent resistor value and can effectively suppress the output pulse current oscillations. But it will also affect the pulse rise/fall time and will reduce the output pulse amplitude. How to combine the three to achieve the optimal output pulse is part of the future work in this research.

6 78 The Journal of China Universities of Posts and Telecommunications Conclusions In this paper, we analyzed and discussed the working principle of the pulsed current source. Increasing the pulsed current source output pulse current amplitude and the theoretical analysis of the pulse edge compression control method are basically consistent with the experimental results. Pulsed current source output pulse current in the range of 0 A to 100 A was achieved, thereby increasing the scope of its application. When the output pulse current amplitude is maximum, the pulse rising can be controlled in 18.8 ns and falling can be controlled in 16.1 ns. And the MOSFET switching losses were decreased by 30.9%, thereby improving the stability and extending the service life of the pulsed current source, making it one of the most ideal choices for the next pulsed semiconductor laser drive source. But high power laser production process is very complicated, the distribution parameters of high power laser are much larger than the estimated maximum value of this article. When the pulsed current source is actually driving the laser, the actual laser driving data and the data given in this paper have a difference. The next research will focus on the study of the compression method for the output pulse current width and suppression of the output pulse current overshoot, designing a narrower pulse width, higher amplitude, shorter pulse rise/fall time and smaller overshoot of the pulsed current source. Acknowledgements This work was supported by the Changchun Science and Technology Project (13KG8, the Jilin Province Science and Technology Development Plan ( References 1. Ma T X, Tian X J. Laser diode driver circuit design and improvement based on the MOSFET. Chinese Journal of Lasers, 01, 39(B06: (in Chinese. Yao S, Tao G T, LU G G, et al nm wavelength high power diode array module. Optics & Precision Enginering, 006, 14(1: 8 11 (in Chinese 3. Wang J H, Yao H B, Xue Z J, et al. Design of sending circuit for laser short range dynamic detection system. Infrared & Laser Engineering, 006, 35(6: (in Chinese 4. Dai Q, Song W W, Wang X J. Design and stability analysis of high frequency LD s driving circuit. Optics & Precision Enginering, 006, 14(5: (in Chinese 5. Bu H Y, Lin Y K, Meng Z H, et al. Research of centimeter-level ranging technology based on ultra-narrow pulsed laser. Laser & Infrared, 009, 39(8: (in Chinese 6. Dang J M, Zhai B, Gao Z L, et al. Nanosecond driver for multiple pulse-modulated infrared quantum cascade lasers. Optics & Precision Enginering, 013, 1(9: (in Chinese 7. Li X, Gu G H. Narrow pulse laser driver design based on power transistors. Laser & Infrared, 013, 43(5: (in Chinese 8. Wang W, Gao X, Zhou Z P, et al. Steady-state thermal analysis of hundred-watt semiconductor laser with multichip-packaging. Infrared & Laser Enginering, 014, 43(5: (in Chinese 9. Tian X J, Wu Y M, Wu G, et al. Study on design and electro magnetic compatibility of high-power pulse driver. The Journal of China Universities of Posts and Telecommunications, 014, 1(S1: Chen Y C, Fen Y G, Zhang X B. Large current nanosecond pulse generating circuit for driving semiconductor laser. Optics & Precision Enginering, 014, (11: (in Chinese 11. Yan B. Analysis of pulse laser diode driver power supply. Optics & Optoelectronic Technolgy, 008, 6(4: 1 3 (in Chinese 1. Cheng W, Miao Q M, Sun F, et al. Design of high current narrow width pulsed power supply of laser diode. High Power Laser and Particle Beams, 010, (6: (in Chinese 13. Wang Y. Pulsed power science and technology. Beijing, China: Beijing University of Aeronautics and Astronautics Press, 010 (in Chinese 14. Li S M. Principle and design of the laser device. Beijing, China: National Defense Industry Press, 005 (in Chinese 15. Hu C S, Qin S Q, Wang X S. An extremely fast and high-power laser diode driver module. Proceedings of the SPIE Conference on Semiconductor Lasers and Applications II, Nov 8, 004, Beijing, China. Vol 568. Bellingham, WA, USA: SPIE, 005: 1 17 (Editor: Wang Xuying