Bipolar high-voltage narrow pulse generation system based on time-sharing trigger Marx circuit

Abstract

The invention provides a bipolar pulse generation system based on a time-sharing trigger Marx circuit. The bipolar pulse generation system aims at solving the problems that in the prior art, a bipolar pulse generation system is high in circuit complexity, the output capacity is limited, and the output amplitude and the pulse width cannot be adjusted. According to the system, firstly, an FPGA generates two paths of driving signals, one path of driving signals is a reference driving signal, and the other path of driving signals is an adjustable delay driving signal; the two paths of signals are driven by two avalanche triode Marx circuits, and two positive and negative unipolar pulse signals are output; and then the signals are superposed through a combiner, and bipolar pulse signals are output. According to the scheme, one path of nanosecond-level unipolar negative pulse and the other path of nanosecond-level unipolar positive pulse are synthesized, and finally nanosecond-level high-amplitude bipolar pulse of which the output amplitude is adjustable in the range of 600V to 1000V and the peak-to-peak pulse width is adjustable in the range of 0.6 ns to 1.4 ns is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of pulse power and radar detection, and in particular relates to a bipolar high-voltage narrow pulse generating system based on a time-sharing triggered Marx circuit. Background Art

[0002] With the widespread application of underground detection technology in resource exploration, urban infrastructure, and geological surveys, ground-penetrating radar (GPR) systems have become one of the core detection tools due to their non-destructive, high-resolution, and real-time imaging capabilities. GPR transmits high-frequency electromagnetic pulse signals underground and inverts the underground medium structure based on the time delay and amplitude changes of the echo signal. The quality of the transmitted pulse (such as amplitude, pulse width, and spectral characteristics) directly determines the system's detection resolution, signal-to-noise ratio, and penetration depth. However, existing pulse generation technologies still have significant bottlenecks in high-amplitude output, waveform fidelity, and system integration, which restricts the performance improvement of GPR in complex scenarios.

[0003] Traditional ground-penetrating radars (GPRs) mostly use a unipolar pulse transmission scheme, relying on avalanche transistors (AVTs) and Marx circuits to generate high-voltage, narrow pulses. Due to structural limitations of the Marx circuit, the output amplitude growth rate decreases significantly when the number of stages exceeds 10. Furthermore, the Marx circuit structure significantly increases the difficulty of synchronously triggering the AVTs at multiple stages, resulting in severe pulse jitter in the output signal, significantly reducing system stability. Furthermore, the spectral energy of the unipolar pulse is concentrated in the low-frequency band, and the electromagnetic waves generated by the antenna also primarily concentrate in the low-frequency band. The low attenuation of low-frequency components in the underground medium results in insufficient resolution for shallow targets. Furthermore, the Marx circuit suffers from delays in the turn-off of switching devices and the discharge of energy storage capacitors, leading to signal overcharging and severe signal smearing.

[0004] To overcome the shortcomings of unipolar pulses, bipolar pulse technology, with its advantages of high frequency and ultra-wideband, has gradually become a research hotspot. The H-bridge topology utilizes MOSFETs or IGBTs to switch current direction, generating bipolar pulses. However, the high switching speeds and low voltage withstand of these devices make it difficult to generate sub-nanosecond high-amplitude pulses. Transformer coupling schemes can convert unipolar signals into bipolar outputs using pulse transformers. However, high-frequency transformers suffer from high core losses and low bandwidth, which can lead to output waveform distortion and low amplitude. Multi-path pulse synthesis to achieve bipolar output typically involves increasing transmission delays by varying line lengths. However, extended lines introduce signal coupling, which also results in waveform distortion. Furthermore, in the field of pulse power technology, traditional bipolar pulse generation schemes suffer from parameter rigidity. The core issue lies in the fact that the physical characteristics of critical paths within the circuit architecture (such as transmission lines, delay modules, and switching devices) dictate fixed signal transmission delays. This hardware-level delay rigidity directly locks the pulse waveform amplitude and width within preset parameters. Specifically, pulse amplitude is limited by the fixed capacity configuration of energy storage elements (such as capacitors) and the voltage threshold of switching devices, while pulse width is completely determined by the characteristic impedance of the transmission line and the length of the delay unit. This invention uses FPGAs to implement two signals with adjustable delays, thereby achieving flexible control of bipolar pulse parameters, allowing pulse amplitude and pulse width to be dynamically adjusted according to actual needs. Summary of the Invention

[0005] The purpose of the present invention is to provide a bipolar, high-voltage, narrow pulse generation system based on a time-sharing, triggered Marx circuit. This system generates unipolar positive and negative pulses through delay compensation using an FPGA and a combination of two avalanche transistors in a ten-stage Marx circuit. This system achieves nanosecond-level, high-amplitude bipolar pulses with an adjustable output amplitude between 600V and 1000V and a peak-to-peak pulse width between 0.6ns and 1.4ns. The system can adjust the peak-to-peak voltage and peak-to-peak pulse width of the bipolar pulses by adjusting the delay of two FPGA signals. The bipolar pulse signals generated by this system have a high bandwidth and a high center frequency, meeting the ultra-wideband, high-frequency output requirements of ground-penetrating radar (GPR), effectively improving the detection resolution of GPR. This reduces the design difficulty of the bipolar pulse signal and increases the peak-to-peak voltage of the bipolar pulses.

[0006] The hardware components of the present invention include: an FPGA chip, a unipolar positive pulse circuit, a unipolar negative pulse circuit, a power module, and a combiner. First, the FPGA chip generates two drive signals: one is a reference drive signal, and the other is an adjustable delay drive signal. These two signals are driven by two avalanche transistor Marx circuits, which output two unipolar pulse signals, one positive and one negative. Second, the signals are superimposed by the combiner to output a bipolar pulse signal. The specific implementation steps include:

[0007] Step 1: Generate two drive signals

[0008] First, the clock generation module divides the external input clock through the PLL circuit inside the FPGA chip to generate two clock signals. One signal is used as the reference signal and is directly input into the unipolar positive pulse circuit to drive the generation of a positive pulse signal. The other signal has an adjustable delay and maintains a certain phase difference with the reference signal. It is input into the unipolar negative pulse circuit to drive the generation of a negative pulse signal. The frequency of the two output signals is no higher than 50kHz.

[0009] Sub-step 1: Determine the delay time

[0010] Taking the reference signal as a reference, the delay time of the delay signal is usually set to the sum of the half-peak pulse width of the positive pulse and the delay of the two transmission lines. The calculation formula of the delay t is:

[0011] t=T0+Δt1

[0012] Where T0 is the half-peak pulse width of the positive pulse, and Δt1 is the delay of the two transmission lines.

[0013] Δt1 is mainly the transmission delay on the coaxial line. The calculation method of the coaxial line transmission delay is as follows:

[0014]

[0015] Where Δτ represents the delay, v p Indicates the speed of electrons moving in the transmission line, ε r represents the relative dielectric constant of the transmission medium, c is the speed of light in a vacuum, and ΔL represents the difference in length between the two transmission lines, which usually does not exceed 10 cm.

[0016] Step 2: Implement a unipolar pulse circuit based on Marx cascade

[0017] First, the power module outputs a high-amplitude voltage to power the circuit, causing the avalanche transistor to enter the avalanche state. The input drive signal outputs a trigger signal through the RC differentiator circuit. The first-stage avalanche transistor is triggered to turn on by the trigger signal. The ten-stage Marx cascade method is used to achieve step-by-step triggering and conduction. The output signal is connected to a 50Ω SMA connector as the signal output end to realize a unipolar negative pulse generation circuit.

[0018] Sub-step 1: RC Differentiator Circuit

[0019] The capacitor in the C0603 package is used as the RC differential capacitor, and the resistor in the R0603 package is used as the RC differential resistor.

[0020] Sub-step 2: Ten-level Marx pulse generation circuit

[0021] The first few stages use avalanche transistors with lower trigger conditions and more stable output (such as FMM415TD), while the latter stages use avalanche transistors with higher trigger conditions and higher output performance (such as FMM417TD). Through the combined design, the complementary advantages of the two are integrated to achieve higher output amplitude under lower trigger conditions. The intermediate energy storage capacitor uses a chip capacitor packaged as C1206, and the current limiting resistor uses a chip resistor packaged as R1210. The local signal line routing adopts a microstrip line design. In a double-layer PCB circuit board, the formula for the characteristic impedance Z0 of the microstrip line is

[0022]

[0023] Where, ε eff W is the trace width of the microstrip line, H is the thickness of the circuit board substrate, and h is the trace thickness.

[0024] Sub-step 3: Implement a unipolar positive pulse source

[0025] Reverse the polarity by connecting the signal pin and the ground pin of the SMA connector at the output end of a unipolar negative pulse generating circuit to realize polarity reversal and realize a unipolar positive pulse generating circuit.

[0026] Step 3: Combiner signal superposition method

[0027] The combiner is a Wilkinson power combiner, with both the input and output ports equipped with 50Ω SMA connectors. One input port is connected to the output of a unipolar positive pulse source, and the other is connected to the output of a unipolar negative pulse source. Due to the triggering time difference, the negative pulse arrives at the end of the positive pulse, forming a bipolar pulse. A 50-ohm coaxial line is used for connection and signal transmission. The output port is connected to an antenna feed or connected to an attenuator through a 50-ohm coaxial line for testing with an oscilloscope. The insertion loss of the combiner affects the peak voltage of the input and output signals. The numerical relationship is as follows:

[0028]

[0029] Where L represents insertion loss, V in Indicates the peak voltage of the input combiner, V out Indicates the peak value of the combiner output voltage.

[0030] The maximum peak-to-peak voltage of the bipolar pulse signal generated by the superposition of unipolar positive and negative pulse signals after passing through the combiner is calculated as follows:

[0031]

[0032] Where L represents insertion loss, V in1 、V in2 Respectively represent the peak voltage of the unipolar positive and negative pulse signals of the input combiner, Vmax Indicates the maximum peak-to-peak value of the bipolar pulse signal that the combiner can generate.

[0033] Beneficial effects

[0034] 1. The present invention uses an FPGA driver module and two pulse generators, and a combiner to synthesize a bipolar pulse signal, achieving nanosecond-level high-amplitude bipolar pulses with an output voltage adjustable in the range of 600V to 1000V and a peak-to-peak pulse width adjustable in the range of 0.6ns to 1.4ns.

[0035] 2. The present invention uses a ten-stage Marx pulse generating circuit designed by combining avalanche transistors FMMT415TD and FMMT417TD, which increases the pulse amplitude and reduces the half-peak pulse width without changing the manufacturing cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a structural diagram of a specific implementation of the bipolar pulse generating system based on the time-sharing triggered Marx circuit of the present invention.

[0037] Figure 2 The present invention is based on the circuit principle diagram of a unipolar negative pulse source.

[0038] Figure 3 The present invention is based on the circuit principle diagram of a unipolar positive pulse source. DETAILED DESCRIPTION

[0039] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. The description is provided to help those skilled in the art understand the present invention for the purpose of explanation rather than limitation.

[0040] To address the limited output amplitude of unipolar pulse circuits, this invention proposes a method for combining avalanche transistors (ABs), significantly improving the output peak voltage. Conventional approaches to bipolar pulse circuit design increase delay differences by changing wiring harness length, often resulting in lengthy wiring harnesses and distorted output waveforms. To address this issue, this invention introduces a bipolar, high-voltage, narrow pulse generation system based on a time-sharing triggered Marx circuit. This effectively addresses this issue and optimizes system performance, enabling flexible control of both output amplitude (adjustable from 600V to 1000V) and peak-to-peak pulse width (adjustable from 0.6ns to 1.4ns). This system utilizes an FPGA for delay compensation and a ten-stage Marx circuit design combining two avalanche transistors to generate unipolar positive and negative pulses. This results in nanosecond-level high-amplitude bipolar pulses with an adjustable output amplitude between 600V and 1000V and a peak-to-peak pulse width between 0.6ns and 1.4ns. The system can adjust the peak-to-peak voltage and peak-to-peak pulse width of the bipolar pulses by adjusting the delays of two FPGA signals. The bipolar pulse signal generated by this system has a high bandwidth and a high center frequency, meeting the ultra-wideband and high-frequency output requirements of ground-penetrating radar, effectively improving the detection resolution of ground-penetrating radar. This reduces the design difficulty of the bipolar pulse signal and increases the peak-to-peak voltage of the bipolar pulse.

[0041] The present invention is based on the bipolar pulse generation system of the time-sharing trigger Marx circuit. Figure 1 As shown, the specific hardware parts include: FPGA driver module, unipolar positive pulse circuit, unipolar negative pulse circuit, power supply module, and combiner.

[0042] Step 1: Generate two drive signals

[0043] First, the clock generation module divides the external input clock through the PLL circuit inside the FPGA chip to generate two clock signals. One signal is used as the reference signal and is directly input into the unipolar positive pulse circuit to drive the generation of a positive pulse signal. The other signal has an adjustable delay and maintains a certain phase difference with the reference signal. It is input into the unipolar negative pulse circuit to drive the generation of a negative pulse signal. The frequency of the two output signals is no higher than 50kHz.

[0044] Sub-step 1: Determine the delay time

[0045] Taking the reference signal as a reference, the delay time of the delay signal is usually set to the sum of the half-peak pulse width of the positive pulse and the delay of the two transmission lines. The calculation formula of the delay t is:

[0046] t=T0+Δt1

[0047] Where T0 is the half-peak pulse width of the positive pulse, and Δt1 is the delay of the two transmission lines.

[0048] Δt1 is mainly the transmission delay on the coaxial line. The coaxial line transmission delay is calculated as follows:

[0049]

[0050] Where Δτ represents the delay, v p Indicates the speed of electrons moving in the transmission line, ε r represents the relative dielectric constant of the transmission medium, c is the speed of light in a vacuum, and ΔL represents the difference in length between the two transmission lines, which usually does not exceed 10 cm.

[0051] Step 2: Implement a unipolar pulse circuit based on Marx cascade

[0052] The present invention is based on the negative pulse source of the ten-level Marx pulse generating circuit. Figure 2 First, the power module outputs a high-amplitude voltage to power the circuit, causing the avalanche transistor to enter the avalanche state. The input drive signal outputs a trigger signal through the RC differentiator circuit. The first-stage avalanche transistor is triggered to turn on by the trigger signal. The ten-stage Marx cascade method is used to achieve step-by-step triggering and conduction. The output signal is connected to a 50Ω SMA connector as the signal output end to realize a unipolar negative pulse generation circuit.

[0053] Sub-step 1: RC Differentiator Circuit

[0054] The capacitor in the C0603 package is used as the RC differential capacitor, and the resistor in the R0603 package is used as the RC differential resistor.

[0055] Sub-step 2: Ten-level Marx pulse generation circuit

[0056] The first few stages use avalanche transistors with lower trigger conditions and more stable output (such as FMM415TD), while the latter stages use avalanche transistors with higher trigger conditions and higher output performance (such as FMM417TD). Through the combined design, the complementary advantages of the two are integrated to achieve higher output amplitude under lower trigger conditions. The intermediate energy storage capacitor uses a chip capacitor packaged as C1206, and the current limiting resistor uses a chip resistor packaged as R1210. The local signal line routing adopts a microstrip line design. In a double-layer PCB circuit board, the formula for the characteristic impedance Z0 of the microstrip line is

[0057]

[0058] Where, ε eff W is the trace width of the microstrip line, H is the thickness of the circuit board substrate, and h is the trace thickness.

[0059] Sub-step 3: Implement a unipolar positive pulse source

[0060] The present invention is based on the positive pulse source of the ten-level Marx pulse generating circuit. Figure 3 , reverse the signal pin and ground pin of the SMA connector at the output end of a unipolar negative pulse generating circuit to achieve polarity reversal and realize a unipolar positive pulse generating circuit.

[0061] Step 3: Combiner signal superposition method

[0062] The combiner is a Wilkinson power combiner, with both the input and output ports equipped with 50Ω SMA connectors. One input port is connected to the output of a unipolar positive pulse source, and the other is connected to the output of a unipolar negative pulse source. Due to the triggering time difference, the negative pulse arrives at the end of the positive pulse, forming a bipolar pulse. A 50-ohm coaxial line is used for connection and signal transmission. The output port is connected to an antenna feed or connected to an attenuator through a 50-ohm coaxial line for testing with an oscilloscope. The insertion loss of the combiner affects the peak voltage of the input and output signals. The numerical relationship is as follows:

[0063]

[0064] Where L represents insertion loss, V in Indicates the peak voltage of the input combiner, V out Indicates the peak value of the combiner output voltage.

[0065] The maximum peak-to-peak voltage of the bipolar pulse signal generated by the superposition of unipolar positive and negative pulse signals after passing through the combiner is calculated as follows:

[0066]

[0067] Where L represents insertion loss, V in1 、V in2 Respectively represent the peak voltage of the unipolar positive and negative pulse signals of the input combiner, V max Indicates the maximum peak-to-peak value of the bipolar pulse signal that the combiner can generate.

Claims

1. A bipolar pulse generation system based on a time-sharing triggered Marx circuit, characterized in that: include: FPGA driver module, unipolar positive pulse circuit, unipolar negative pulse circuit, power supply module and combiner. The FPGA chip is divided into a clock generation module and a trigger signal generation module. First, the FPGA chip generates two drive signals, one for the reference drive signal and the other for the adjustable delay drive signal. The two signals are driven by two avalanche transistor Marx circuits, outputting two unipolar pulse signals, one positive and one negative. Second, the signals are superimposed through the combiner to output a bipolar pulse signal. The specific implementation steps include: Step 1: Generate two drive signals First, the clock generation module divides the external input clock through the PLL circuit inside the FPGA chip to generate two clock signals. One signal is used as the reference signal and is directly input into the unipolar positive pulse circuit to drive the generation of a positive pulse signal. The other signal has an adjustable delay and maintains a certain phase difference with the reference signal. It is input into the unipolar negative pulse circuit to drive the generation of a negative pulse signal. The frequency of the two output signals is no higher than 50kHz. Sub-step 1: Determine the delay time Taking the reference signal as reference, the delay time of the delay signal is set to the sum of the half-peak pulse width of the positive pulse and the delay of the two transmission lines; The calculation formula of delay t is: t=T0+Δt1 Where T0 is the half-peak pulse width of the positive pulse, and Δt1 is the delay of the two transmission lines. Δt1 is the transmission delay on the coaxial line. The coaxial line transmission delay is calculated as follows: Where Δτ represents the delay, v p Indicates the speed of electrons moving in the transmission line, ε r represents the relative dielectric constant of the transmission medium, c is the speed of light in a vacuum, and ΔL represents the difference in length between the two transmission lines; Step 2: Implement a unipolar pulse circuit based on Marx cascade First, the power module outputs a high-amplitude voltage to power the circuit, causing the avalanche transistor to enter the avalanche state. The input drive signal outputs a trigger signal through the RC differentiator circuit. The first-stage avalanche transistor is triggered to conduct by the trigger signal. The ten-stage Marx cascade method is used to achieve step-by-step triggering and conduction. The output signal is connected to a 50Ω SMA connector as the signal output terminal to realize a unipolar negative pulse generation circuit. Sub-step 1: RC Differentiator Circuit The capacitor in C0603 package is used as the RC differential capacitor, and the resistor in R0603 package is used as the RC differential resistor; Sub-step 2: Ten-level Marx pulse generation circuit The first few stages use avalanche transistors with lower trigger conditions and more stable output, while the latter stages use avalanche transistors with higher trigger conditions and higher output performance. The combined design integrates the complementary advantages of the two to achieve higher output amplitude under lower trigger conditions; the intermediate energy storage capacitor uses a chip capacitor packaged as C1206, and the current limiting resistor uses a chip resistor packaged as R1210; the local signal line routing adopts a microstrip line design; in a double-layer PCB circuit board, the formula for the characteristic impedance Z0 of the microstrip line is Where, ε eff W is the width of the microstrip line, H is the thickness of the circuit board substrate, and h is the thickness of the trace; Sub-step 3: Implement a unipolar positive pulse source Reverse the polarity of the SMA connector at the output of a unipolar negative pulse generator circuit to connect the signal pin and the ground pin to realize polarity reversal and realize a unipolar positive pulse generator circuit. Step 3: Combiner signal superposition method The combiner is a Wilkinson power combiner, with both the input and output ports having 50Ω SMA interfaces. One input port is connected to the output of a unipolar positive pulse source, and the other input port is connected to the output of a unipolar negative pulse source. Due to the triggering time difference, the negative pulse arrives at the end of the positive pulse, forming a bipolar pulse. A 50-ohm coaxial line is used for connection and signal transmission. The output port is connected to the antenna feed or connected to an attenuator through a 50-ohm coaxial line for testing with an oscilloscope. The insertion loss of the combiner affects the peak voltage of the input and output signals. The numerical relationship is as follows: Where L represents insertion loss, V in Indicates the peak voltage of the input combiner, V out Indicates the peak value of the combiner output voltage; The maximum peak-to-peak voltage of the bipolar pulse signal generated by the superposition of unipolar positive and negative pulse signals after passing through the combiner is calculated as follows: Where L represents insertion loss, V in1 、V in2 Respectively represent the peak voltage of the unipolar positive and negative pulse signals of the input combiner, V max Indicates the maximum peak-to-peak value of the bipolar pulse signal that the combiner can generate.

Images