Electronics Guide

Energy Storage and Compression

The core principle of pulsed power is the temporal compression of energy. A conventional supply charges a storage element slowly, over seconds or minutes, and a switch then releases that energy in microseconds or nanoseconds. The average charging power may be only a few kilowatts while the delivered peak power reaches gigawatts, because the two differ by the ratio of charge time to discharge time.

Capacitors dominate practical energy storage. They discharge quickly, tolerate repeated cycling, and reach energy densities on the order of a joule per cubic centimeter in film dielectrics designed for pulse duty. Inductive storage offers a higher theoretical density, since the magnetic energy density rises with the square of flux density, but it is far harder to use: releasing energy from an inductor requires an opening switch that interrupts a large current rather than a closing switch that starts one. Exploding-wire and exploding-foil fuses, plasma opening switches, and superconducting coils have all been applied to this problem, with mixed practical success. Rotating machines, including flywheel generators, homopolar generators, and compulsators, store the largest energies of all and serve where a pulse of milliseconds rather than nanoseconds is acceptable.

Most large systems compress energy in more than one stage. A slow primary store hands energy to an intermediate store at higher voltage and shorter duration, which in turn drives the final pulse forming stage. Each stage trades voltage for time, and each introduces losses, so the overall efficiency of a multistage chain is the product of the stage efficiencies.

High-Voltage Switching

Switching is the critical enabling technology for pulsed power. A closing switch must hold off extreme voltage, then transition to a conducting state in nanoseconds while carrying currents that can reach hundreds of kiloamperes, and it must do so with timing jitter small compared with the pulse it initiates.

Gas- and vacuum-filled switches remain important at the highest power levels because they tolerate enormous transient stress. Spark gaps, whether pressurized with air, nitrogen, or sulfur hexafluoride, handle the largest currents and the fastest closures but erode their electrodes and require maintenance. Hydrogen thyratrons switch tens of kilovolts and kiloamperes with subnanosecond jitter and are the traditional choice for radar and accelerator modulators. Ignitrons, which use a mercury pool cathode, excel at transferring very large charge in a single shot, and their toxicity has driven replacement programs wherever alternatives exist.

Solid-state switches built from series-connected insulated-gate bipolar transistors, thyristors, or gate turn-off devices offer far longer life and precise controllability at moderate power. Magnetic switches, which exploit the saturation of inductor cores, contain no electrodes at all and provide effectively maintenance-free operation for repetitive systems. Selecting among these families means weighing peak current, hold-off voltage, rate of current rise, repetition rate, jitter, and the shot count expected between overhauls.

Pulse Shaping and Forming

Many applications require precisely shaped pulses rather than the decaying waveform of a simple capacitor discharge. Pulse forming networks combine inductors and capacitors to synthesize flat-topped or otherwise tailored pulses. A charged transmission line, discharged into a matched load, produces a rectangular pulse whose duration is twice the one-way transit time of the line and whose amplitude is half the charging voltage. The Blumlein configuration folds two such lines together so that the full charging voltage appears across a matched load, doubling the output for the same charge level. Magnetic pulse compression stages sharpen rise times toward the nanosecond scale. Designing these systems demands a working command of transmission-line theory, impedance matching, and network synthesis.

High-Voltage Engineering

Pulsed power systems routinely operate from tens of kilovolts to several megavolts. At these levels, insulation design becomes paramount, and phenomena such as corona discharge, surface flashover, and breakdown in gases, liquids, and solids must be managed carefully. Transformer oil, sulfur hexafluoride, and deionized water serve as insulating and energy-storage media in different designs, the last valued for its high permittivity, which yields compact low-impedance pulse forming lines when the charge is applied for only a microsecond or two.

Breakdown under pulsed stress is time dependent: a dielectric withstands a considerably higher field for a microsecond than it does under direct voltage, because the streamers that initiate breakdown need time to cross the gap. Empirical scaling relations developed by J. C. Martin at Aldermaston in the 1960s express this dependence on pulse duration and stressed area, and they remain a standard first-cut design tool. Grading rings, corona shields, and field-shaping electrodes smooth electric-field distributions to prevent premature breakdown, and generous creepage paths along insulator surfaces guard against flashover, which usually occurs at fields well below the bulk breakdown strength of either material.

Repetition Rate and Average Power

Single-shot systems and repetitive systems face different constraints. A single-shot machine may fire a few times a day, so component erosion and thermal load matter little and the designer optimizes for peak performance. A repetitive system firing hundreds or thousands of times per second must dissipate the resistive losses of every stage continuously, and its switches must recover their hold-off strength between pulses. Average power, not peak power, then sets the size of the cooling system and the choice of switch. This distinction explains why magnetic and solid-state switching dominate industrial and medical equipment while spark gaps persist in the largest research machines.

Marx Generators

A Marx generator multiplies voltage by charging capacitors in parallel through resistors or inductors and then discharging them in series through a chain of switches. With n stages each charged to voltage V, the ideal output approaches nV, the practical figure falling short because of stray capacitance to ground, incomplete switch closure, and losses in the charging network. The first gap is normally triggered deliberately, and the overvoltage it produces cascades through the remaining gaps in a process called erection.

Marx generators are workhorses of the field, producing outputs from hundreds of kilovolts to tens of megavolts. Sandia National Laboratories' Z machine illustrates the scale the topology reaches: thirty-six Marx generators charge in parallel and store roughly twenty-two megajoules, which is then compressed through intermediate storage and pulse forming lines into a current pulse rising in about one hundred nanoseconds to a peak of some twenty-six megaamperes. At the opposite extreme, compact Marx generators built on printed circuit boards with semiconductor switches make portable high-voltage sources practical for field instruments and pulsed-electric-field equipment.

Pulse Forming Networks

A pulse forming network shapes the energy released from storage into a pulse of specific duration and waveshape, approximating the behavior of an ideal charged transmission line with a finite ladder of inductors and capacitors. The Guillemin networks, designated types A through E, are derived from a Fourier expansion of the transmission-line response and are the classic choice for radar line-type modulators. The type E form is the most widely used: it employs equal-value capacitors and a single continuously wound, tapped coil whose geometry sets the mutual coupling between adjacent sections, with roughly fifteen percent coupling giving the flattest top for a given number of meshes.

A true pulse forming network reproduces the desired waveform with far fewer sections than the older Rayleigh ladder, which approximated the line with many identical LC segments. Pulse duration follows from the total inductance and capacitance, while the rise time is governed chiefly by the mesh nearest the load, and the flatness of the top improves as sections are added. The chosen topology depends on pulse duration, rise-time requirements, load impedance, and the acceptable droop or ripple on the flat top.

Magnetic Pulse Compression

Magnetic switches use saturable inductors that fall from high to low impedance as their cores saturate. By cascading stages whose saturation times shorten progressively, a magnetic pulse compression chain can sharpen a pulse from microseconds toward nanoseconds, each stage gaining perhaps a factor of three to ten in compression. Because these stages contain no contacts or electrodes to erode, they offer long life and high reliability for repetitive operation, an advantage that suits industrial and scientific systems delivering millions of pulses. Effective design hinges on the selection of magnetic materials, typically amorphous or nanocrystalline alloys or ferrites chosen for a square hysteresis loop and low loss at high frequency, together with careful optimization of core geometry, reset biasing, and winding configuration.

Solid-State Pulsed Power

Advances in power semiconductors have produced solid-state alternatives to gas and vacuum switches for many applications. Series-connected insulated-gate bipolar transistors, arranged as stacks or as solid-state Marx generators, reach hundreds of kilovolts: press-pack devices have been used in klystron modulators delivering pulses near 450 kilovolts with rise times of a few hundred nanoseconds at repetition rates above one hundred hertz. Building such a modulator from many identical low-voltage cells, each holding only a few kilovolts, simplifies insulation and allows the output waveform to be shaped by switching cells in sequence.

Wide-bandgap devices in silicon carbide and gallium nitride extend switching speed and voltage further, supporting pulse generators with rise times near one hundred nanoseconds at repetition rates of tens of kilohertz. Solid-state systems provide precise triggering, long lifetime, high repetition rates, and graceful fault behavior, though they still do not match the peak current and standoff voltage of the best gas-filled switches in single-shot machines.

Energy Storage Selection

Capacitor selection balances energy density, voltage rating, equivalent series resistance, equivalent series inductance, lifetime, and cost. Metallized-film and film-foil capacitors offer the low inductance needed for fast discharge and dominate pulse-duty service; metallized construction adds a measure of self-healing, while film-foil construction carries higher peak current. Ceramic capacitors suit compact, lower-energy systems, while electrolytic types maximize energy density at the cost of higher losses and limited pulse current. Pulse capacitors are rated not only in volts and microfarads but in shots to failure at a given voltage reversal, since the reverse swing that follows an underdamped discharge is a leading cause of degradation. For the largest energies, banks may contain thousands of individual units, with careful attention to current sharing, low-inductance interconnection, and fault isolation so that a single failure does not cascade through the bank.

Circuit Inductance and Layout

Rate of current rise is limited by the inductance of the discharge loop, so physical layout is as much a part of the design as component selection. Coaxial and parallel-plate geometries, wide flat conductors, and short return paths all reduce loop inductance; a bank that is electrically excellent but mechanically sprawling will not deliver a fast pulse. Large discharges also produce mechanical forces, because parallel conductors carrying megaamperes repel one another with substantial pressure, and busbars, capacitor terminals, and load hardware must be braced accordingly.

Triggering and Synchronization

Precise timing is critical, particularly in multi-module systems and in applications that must coordinate with external events. Trigger generators must deliver enough voltage and energy to fire switches reliably while holding timing jitter to nanoseconds. Fiber-optic triggering provides galvanic isolation and immunity to the electromagnetic interference these systems generate. Master oscillators and programmable delay generators sequence complex installations in which many modules fire simultaneously or in a deliberate cascade, and in the largest machines the summed jitter of hundreds of switches determines how sharply the individual contributions add at the load.

Electromagnetic Compatibility

The rapid current changes in pulsed power systems generate intense electromagnetic fields that can disrupt nearby equipment and instrumentation. Shielding, filtering, and disciplined grounding are essential. Faraday enclosures protect sensitive electronics, while coaxial and balanced transmission-line geometries limit radiated emission. Transient-voltage suppressors and protective spark gaps guard against conducted interference. Sound electromagnetic-compatibility design must account not only for the primary pulse but also for the ringing and reflections that follow it, and it must anticipate the ground potential rise that accompanies any large current returning through the structure.

Diagnostics and Measurement

Measuring a pulse that lasts nanoseconds and carries kiloamperes calls for instruments the general laboratory rarely holds. Current is sensed with Rogowski coils, low-inductance coaxial shunts known as current-viewing resistors, and inductive B-dot probes; voltage is sensed with compensated resistive and capacitive dividers, and, where galvanic isolation is essential, with optical sensors that exploit the Pockels or Faraday effect. Every probe must be rated for the full transient, and signal cables must be routed and shielded so that the measurement records the event rather than the interference it produces. Bandwidth, response time, and calibration against a known waveform matter as much as accuracy at direct voltage.

Safety Systems

The lethal hazards of stored high-voltage energy demand comprehensive safety measures. Interlocked enclosures prevent access to energized components, and grounding hooks and shorting bars give visible confirmation that storage elements are discharged. Dump resistors and crowbar circuits divert stored energy quickly under fault conditions. Clear warning indicators signal charging status and hazardous states, and designers must allow for dielectric absorption, which can restore a dangerous voltage on a capacitor that was discharged only briefly. Personnel training, documented procedures, and formal safety reviews matter as much as the hardware in preventing accidents.

Scientific Research

Pulsed power enables fundamental physics from particle acceleration to fusion energy. Accelerators rely on pulsed modulators to drive klystrons and other radio-frequency sources, and on fast kicker magnets whose currents rise within the gap between beam bunches. Inertial confinement fusion experiments deliver megajoules in nanoseconds. At the National Ignition Facility, a power conditioning system storing roughly four hundred megajoules dumps its energy into flash lamps over a few hundred microseconds, and the laser then places about two megajoules of ultraviolet light on the target in a few nanoseconds; the December 2022 shot that produced more fusion energy than the laser energy delivered to the target was the first ignition result of its kind. Pulsed-power-driven z-pinches, as at Sandia's Z machine, take a different route, compressing plasmas magnetically to extreme density and temperature for high-energy-density physics. Flash radiography systems such as the dual-axis facility at Los Alamos generate intense X-ray bursts to image fast hydrodynamic events. Such work continually pushes the limits of the technology and drives further innovation.

Industrial Processing

Electromagnetic forming uses pulsed magnetic fields to shape metal without physical contact, reaching strain rates that conventional presses cannot and forming alloys that crack under quasi-static dies. Pulsed-electric-field processing applies fields of tens of kilovolts per centimeter in pulses lasting microseconds to inactivate microorganisms in pumpable foods while largely preserving flavor, color, and heat-sensitive nutrients; the same technique softens plant tissue before juicing or drying. Electrohydraulic forming harnesses underwater electrical discharges to shape metal, and magnetic pulse welding joins dissimilar metals such as aluminum to steel or copper, which resist conventional fusion welding because they form brittle intermetallic compounds when melted. Each application exploits the capacity of pulsed power to concentrate energy in both space and time.

Medical Applications

Pulsed electromagnetic field therapy is used to support the healing of bone fractures. Electrochemotherapy and irreversible electroporation apply controlled trains of short, high-field pulses to enhance the uptake of cytotoxic drugs or to ablate tumor tissue without the thermal damage that radiofrequency or microwave ablation causes, which allows treatment near vessels and ducts. Defibrillators deliver shaped biphasic pulses, typically of one hundred to two hundred joules in an external unit, to restore an organized heart rhythm. Pulsed sources also drive surgical and therapeutic lasers and the flash lamps of dermatological systems. All of these uses demand tight control of pulse amplitude, duration, and timing to achieve a therapeutic effect while safeguarding the patient, and all fall under medical device regulation that governs their electrical safety and reliability.

Defense and Directed Energy

High-power radar transmitters were among the first large-scale pulsed power applications and remain among the most numerous, with line-type modulators built around a thyratron, a pulse forming network, and a pulse transformer driving magnetrons and klystrons. High-power microwave sources, including virtual cathode oscillators, magnetrons, and backward-wave oscillators, convert pulsed electron beams into intense radiation for electronic warfare and counter-electronics research. Electromagnetic launchers accelerate projectiles with currents of megaamperes, although rail erosion, barrel life, and the mass of the pulsed power supply have kept the technology from fielded service; the United States Navy ended funding for its railgun program in 2021 after roughly two decades of development. Electromagnetic aircraft launch systems on modern carriers apply the same principles at longer pulse durations, drawing on stored rotational energy rather than capacitors.