Design and Evaluation of Type C Pulse Forming Network for Different Component Values

Transcription

1 Volume 2, Issue 1, January-March, 2014, pp , IASTER Online: , Print: Design and Evaluation of Type C Pulse Forming Network for Different Component Values Ramnik Kour 1, Nishant 2 1 M. Tech Student, Dept of EEE, Arni University, Kathgarh, Himachal Pardesh, India 2 Assistant Professor, Dept of EEE, Arni University, Kathgarh, Himachal Pardesh, India ABSTRACT The line-type pulser does not need complex circuits as those required by the hard-tube pulser. The relatively large "on" resistance of the vacuum-tube and the power dissipated make the hard-tube pulser much less efficient than the line-type pulser. Line-type pulsers are often lumped-parameter realizations of transmission lines. Realizations for line-type pulsers based on transmission line theory generally require a large number of elements to generate an accurate representation of a rectangular pulse. Consequently, techniques for designing pulse forming networks which require a relatively small number of elements have been derived. In this work, a 3-stage type-c pulse forming network have been studied and designed to evaluate the effect of output parameters. The results are evaluated for different load resistance value while keeping the other components value as static. The reference impedance was set to 50Ω. The input applied voltage of 10V is considered for the design and result evaluation purposes. The results are evaluated to see the effect of load current (I Z ), capacitor current (I C ) and inductor current (I L ) for different load resistances. Keywords: Network Components, Pulse Forming Network, Type C Ladder Network, Transmission Line, 3-Stage Network. I. INTRODUCTION Pulse forming networks (PFN) has several advantages over the hard-tube pulser in terms of simplicity, size, weight, and efficiency. Generally, the line-type pulser does not need as complex drive circuits for grid control as those required by the hard-tube pulser. Pulse forming networks are often lumpedparameter realizations of transmission lines and are also known as line-type pulsers [1-3]. Realizations for line-type pulsers based on transmission line theory generally require a large number of elements to generate an accurate representation of a rectangular pulse [4]. Consequently, techniques for designing pulse forming networks which require a relatively small number of elements have been derived. These techniques are based on the synthesis of a pulse having finite rise and fall times as opposed to the zero rise and fall times for a rectangular pulse. Line-type pulsers do have disadvantages and limitations [5]. The pulse generated by the line-type pulser is generally not as rectangular as the pulse generated by the hard-tube pulser. The line-type generated pulse usually has a longer rise and fall time and has oscillation on the top of the pulse. Overshoot on the leading and trailing edges is coramon [6]. These distortions of the pulse shape may be reduced by special network design. Important disadvantages of the line-type pulser are the difficulty of changing the impedance or the pulse duration [7] [8]. These factors are determined by the element values. The element values are not easily varied by conventional methods. The pulse duration can be reduced by decreasing the value of the network inductance. The transmission line, composed of distributed inductances and capacitances, has several interesting characteristics. Ideally, the transmission line can transmit signals with no distortion, and theoretically it can produce perfectly rectangular pulses into a matched load [9]. In a near ideal transmission line, such as coaxial cable, there are distinct limitations. The major limitation is the excessive length of cable required to generate a pulse of even a few microseconds duration [10]. Several hundred feet of cable, depending on the cable's propagation time, are required to generate a one microsecond pulse. Size 9

2 limitations may be overcome by using a lumped parameter transmission line. A lumped parameter line consists of discrete inductors and capacitors arranged such that the characteristics of a transmission line are approached [11]. The design of a lumped parameter line permits considerable flexibility in the choice of the pulse duration time and the characteristic impedance. A major disadvantage of the lumped parameter line used for pulse generation is its pulse shape. Rise and fall times are limited by parameter choice [12]. Distortion due to oscillation during the pulse and overshoot on the leading and trailing edges of the pulse are common. However, deviations from the ideal pulse may be reduced by using special inductor-capacitor configurations [13]. In this work, Type C pulse forming network has been studied, analysed and modified to generate pulse waveforms. Further, the value of each component is to be optimized for their applications in high voltage pulse forming networks. II. METHODOLOGY A type-c network using lumped parameter line consists of discrete inductors and capacitors arranged such that the characteristics of a transmission line are approached and is shown in fig. 1. The design of a lumped parameter line permits considerable flexibility in the choice of the pulse duration time and the characteristic impedance. Fig. 1: Type-C PFN Network. The design of the type-c PFN network can be explained using the flow diagram as shown in fig. 2. The network is initiated by applying ac voltage of 10V. The input is then filtered and amplified and is then applied to 3-stage PFN network. The remaining dc component, if any, is again filtered out and the output is generated for the load. Stop Start Input Initialization Output Generated Filtering Varying different component values Altering component values Type-C 3 rd order Network Selection Fig. 2: Flow Process for Evaluating Type-C PFN Network In this work, the type-c ladder has been studied and simulated for different ladder networks and component parameters. Only 3-stage networks have been simulated for this experiment to know their effect on output characteristics. The pulse duration can be reduced by decreasing the value of the network inductance. This method for reducing the pulse duration has effects on other PFN parameters and these effects are studied. For this type of network, equicells are used. The value of resistances R were selected as zero while for inductors L and capacitors C, the selected values are 1μH and 1μF respectively. Different load resistances RL were used to evaluate the parametric changes at output. Different loads were taken as 1Ω, 10Ω, 100Ω and 1KΩ. The input voltage applied to the network is 10

3 10V ac voltage with time interval Tint of 50 μs. Distortion due to oscillation during the pulse and overshoot on the leading and trailing edges of the pulse are common. However, deviations from the ideal pulse may be reduced by using special inductor-capacitor configurations. III. RESULTS AND DISCUSSIONS The experiment was performed in FEM based software MATLAB. Fig. 3 shows the output values for 1Ω load resistance. Fig. 3(a) Capacitor current for R L =1Ω Fig. 3(c) Inductor current for R L =1Ω Fig. 3(b) Capacitor voltage current for R L =1Ω Fig. 3(d) Load current for RL=1Ω Similarly the results were noted while taken the value of load resistance as 10Ω and is shown in fig. 4. We kept on increasing the load resistance to see its effect on the component parameters. In this regard, the graphs have been noted on the value of load resistance 100Ω as shown in fig. 5. Fig. 4(a) Capacitor current for R L =10Ω Fig. 4(b) Capacitor voltage for R L =10Ω Fig. 4(c) Inductor current for R L =10Ω Fig. 4(d) Load current for R L =10Ω 11

4 Fig. 5(a) Capacitor current for R L =100Ω Fig. 5(b) Capacitor voltage for R L =100Ω Fig. 5(c) Inductor current for R L =100Ω Fig. 5(d) Load current for R L =100Ω Similarly the results were noted while taking the value of load resistance as 10Ω and are shown in fig. 6. Fig. 6(a) Capacitor current for R L =1000Ω Fig. 6(b) Capacitor voltage for R L =1000Ω Fig. 6(c) Inductor current for R L =1000Ω Fig. 6(d) Load current for R L =1000Ω From the plots, it is clear that the frequency response of the network is limited due to inductance in the connections and the energy storage capacitors. Moreover, the response of the circuit would be limited to such an extent that there would be no peak at all. However, the accuracy depends on values of the equivalent circuits with respect to the various components. Knowing these values, it is a simple circuit calculation to determine the approximate frequency response or the load current of such a circuit built by these passive components, to get the desired waveform in a specific network. The lesser is the value of load, higher is the rise and fall time for the circuit. The slow rise and fall time for across the circuit for higher load value is due to the increased leakage inductance of the circuit. The charging circuit serves to isolate the power supply from the PFN during operation and yet allows the PFN to be properly charged by 12

5 the power supply. The steps for generating a pulse with the PFN are as follows: (a) the PFN capacitors are charged to an initial voltage; (b) the trigger circuit turns the switch "on" and the switch becomes a shortcircuit; (c) the PFN discharges through the load resistance generating a voltage pulse across the load; (d) after the PFN completely discharges, the switch deionizes and becomes an open-circuit; (e) the PFN capacitors begin to recharge. The switch is normally a thyratron. The pulse's magnitude varies linearly with the initial voltage across the capacitors so the pulse shape is always the same. IV. CONCLUSION The design is formed and studied for a 3-stage Type-C PFN network to produce any desired pulse shape. The experiment is done by changing the values of the network components as well as the load resistance. The plots described in the figures show a parabolic rise and fall and ultimately become zero. Moreover, the simulation results show no significant differences in the rise and fall time for different cases, with and without consideration of losses. PFNs have significant advantages over other methods of pulse generation in terms of circuit simplicity and higher efficiency. If simple, easily-applied methods for changing PFN pulse durations were available, applications for PFNs and PFN usage would greatly increase. The pulse duration can be reduced by decreasing the value of the total network inductance, the total network capacitance, or both. REFERENCES [1] P.C. Magnusson, Transmission Lines and Wave Propagation (2nd ed. Boston: Allyn and Bacon, 1970). [2] W. C. Johnson, Transmission Lines and Networks (New York: McGraw-Hill Book Company, Inc, 1950). [3] G. N. Glasoe and J. F. Lebacqz, eds. Pulse Generators (2nd ed. New York: Dover Publications, Inc, 1965). [4] E. A. Guillemin, Synthesis of Passive Networks (New York: John Wiley and Sons, Inc, 1957). [5] Deepak Gupta, and P I John, Design and construction of double-blumlein HV pulse power Supply, Sadhana, 26(5), 2001, [6] H. Li, H. J. Ryoo, J. S. Kim, G. H. Rim, Y. B. Kim, and J. Deng, Development of Rectangle- Pulse Marx Generator Based on PFN, IEEE Trans. Plasma Sci., 37, 2009, [7] D. Wang, T. Namihira, K. Fujiya, S. Katsuki, and H. Akiyama, The reactor design for diesel exhaust control using a magnetic pulse compressor, IEEE Trans. Plasma Sci., 32, 2004, [8] Ramnik Kour, and Randeep Singh Chib, Effect of Different Load Resistances on 3- Stage Type-A Pulse Forming Network, International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, 2(11), 2013, [9] M. Bollen, and R.A.A. degraaf, Behavior of AC and DC drives during voltage sags with phaseangle jump and three-phase unbalance, Power Engineering Society 1999 Winter Meeting, 2, 1999, [10] B.N. Singh, B. Singh, A. Chandra, and K. Al-Haddad, Design and digital implementation of active filter with power balance theory, IEE Proc.-Electric Power Applications, 152(5), 2005, [11] H. Akagi, E.H. Watanable, and M. Aredes, Instantaneous power theory and application to power conditioning (Jhon Wiley & sons, New Jersey, USA, 2007). [12] N. Hingorani, G. Laszlo, Understanding FACTS: Concepts and Technology of Flexible AC Transmission Systems (IEEE Press, Inc, 2000). [13] G. F. Reed, M. Takeda, and I. Iyoda, Improved power quality solutions using advanced solidstate switching and static compensation technologies, IEEE Power Engineering Society 1999 Winter Meeting, IEEE. 2, 1999,