SEARCH · Engineering Papers

Results for “inductive charging”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3

All sources NASA NTRS DOE OSTI

NASA NTRS2011

Performance Effects of Adding a Parallel Capacitor to a Pulse Inductive Plasma Accelerator Powertrain

Polzin, Kurt A.↗

NASA NTRS2012

Comparison of Computed and Measured Performance of a Pulsed Inductive Thruster Operating on Argon Propellant

Pulsed inductive plasma accelerators are electrodeless space propulsion devices where a capacitor is charged to an initial voltage and then discharged through a coil as a high-current pulse that inductively couples energy into the propellant. The field produced by this pulse ionizes the propellant, producing a plasma near the face of the coil. Once a plasma is formed if can be accelerated and expelled at a high exhaust velocity by the Lorentz force arising from the interaction of an induced plasma current and the magnetic field. A recent review of the developmental history of planar-geometry pulsed inductive thrusters, where the coil take the shape of a flat spiral, can be found in Ref. [1]. Two concepts that have employed this geometry are the Pulsed Inductive Thruster (PIT)[2, 3] and the Faraday Accelerator with Radio-frequency Assisted Discharge (FARAD)[4]. There exists a 1-D pulsed inductive acceleration model that employs a set of circuit equations coupled to a one-dimensional momentum equation. The model was originally developed and used by Lovberg and Dailey[2, 3] and has since been nondimensionalized and used by Polzin et al.[5, 6] to define a set of scaling parameters and gain general insight into their effect on thruster performance. The circuit presented in Fig. 1 provides a description of the electrical coupling between the current flowing in the thruster I1 and the plasma current I2. Recently, the model was upgraded to include an equation governing the deposition of energy into various modes present in a pulsed inductive thruster system (acceleration, magnetic flux generation, resistive heating, etc.)[7]. An MHD description of the plasma energy density evolution was tailored to the thruster geometry by assuming only one-dimensional motion and averaging the plasma properties over the spatial dimensions of the current sheet to obtain an equation for the time-evolution of the total energy. The equation set governing the dynamics of the coupled electrodynamic-current sheet system is composed of first-order, coupled ordinary differential equations that can be easily solved numerically without having to resort to much more complex 2-D finite element plasma simulations.

Polzin, Kurt A.↗

NASA NTRS1992

Development of a High Resolution Liquid Xenon Imaging Telescope for Medium Energy Gamma Ray Astrophysics

In the third year of the research project, we have (1) tested a 3.5 liter prototype of the Liquid Xenon Time Projection Chamber, (2) used a prototype having a 4.4 cm drift gap to study the charge and energy resolution response of the 3.5 liter chamber, (3) obtained an energy resolution as good as that previously measured by us using chambers with drift gaps of the order of millimeters, (4) observed the induction signals produced by MeV gamma rays, (4) used the 20 hybrid charge sensitive preamplifiers for a nondestructive readout of the electron image on the induction wires, (5) performed extensive Monte Carlo simulations to obtain results on efficiency, background rejection capability, and source flux sensitivity, and (6) developed a reconstruction algorithm for events with multiple interaction points.

Aprile, Elena↗

NASA NTRS1991

The physics of the oscillating lightning

The lightning is proposed to start in a limited region of electric breakdown field. A growing charge pool in a cumulonimbus cloud is likely to produce such a field in its outer layers. When the field is not aligned with the potential surfaces, the electric force pushes positive charge in one direction and negative in the opposite. Self inductance develops and an embryo open electric circuit is born. Oscillations swiftly grow in strength when potential energy of the cloud is released. The channel extends in steps during those cycle stages where charge is crammed into the channel terminals. Electrostatic consequences of a narrow conductor among the cloud charges are computed. The conductor is charged by influence to such a degree that the charge is bound to leak out to the ambient air and appear as a charge sheath some distance off the channel. Ample energy is released to sustain the oscillations. The period increases with channel length and agrees well with observed time between steps. A ground strike leads to a shock pulse travelling up the channel.

Storebo, Per B.↗

NASA NTRS1994

Charged-particle mutagenesis II. Mutagenic effects of high energy charged particles in normal human fibroblasts

The biological effects of high LET charged particles are a subject of great concern with regard to the prediction of radiation risk in space. In this report, mutagenic effects of high LET charged particles are quantitatively measured using primary cultures of human skin fibroblasts, and the spectrum of induced mutations are analyzed. The LET of the charged particles ranged from 25 KeV/micrometer to 975 KeV/micrometer with particle energy (on the cells) between 94-603 MeV/u. The X-chromosome linked hypoxanthine guanine phosphoribosyl transferase (hprt) locus was used as the target gene. Exposure to these high LET charged particles resulted in exponential survival curves; whereas, mutation induction was fitted by a linear model. The Relative Biological Effect (RBE) for cell-killing ranged from 3.73 to 1.25, while that for mutant induction ranged from 5.74 to 0.48. Maximum RBE values were obtained at the LET of 150 keV/micrometer. The inactivation cross-section (alpha i) and the action cross-section for mutant induction (alpha m) ranged from 2.2 to 92.0 micrometer2 and 0.09 to 5.56 x 10(-3) micrometer2, respectively. The maximum values were obtained by 56Fe with an LET of 200 keV/micrometer. The mutagenicity (alpha m/alpha i) ranged from 2.05 to 7.99 x 10(-5) with the maximum value at 150 keV/micrometer. Furthermore, molecular analysis of mutants induced by charged particles indicates that higher LET beams are more likely to cause larger deletions in the hprt locus.

NASA Center JSC↗

NASA NTRS1994

Charged-particle mutagenesis 2. Mutagenic effects of high energy charged particles in normal human fibroblasts

The biological effects of high Linear Energy Transfer (LET) charged particles are a subject of great concern with regard to the prediction of radiation risk in space. In this report, mutagenic effects of high LET charged particles are quantitatively measured using primary cultures of human skin fibroblasts, and the spectrum of induced mutations are analyzed. The LET of the charged particles ranged from 25 KeV/micrometer to 975 KeV/micrometer with particle energy (on the cells) between 94-603 MeV/u. The X-chromosome linked hypoxanthine guanine phosphoribosyl transferase (hprt) locus was used as the target gene. Exposure to these high LET charged particles resulted in exponential survival curves; whereas, mutation induction was fitted by a linear model. The Relative Biological Effect (RBE) for cell-killing ranged from 3.73 to 1.25, while that for mutant induction ranged from 5.74 to 0.48. Maximum RBE values were obtained at the LET of 150 keV/micrometer. The inactivation cross-section (alpha i) and the action cross-section for mutant induction (alpha m) ranged from 2.2 to 92.0 sq micrometer and 0.09 to 5.56 x 10(exp -3) sq micrometer respectively. The maximum values were obtained by Fe-56 with an LET of 200 keV/micrometer. The mutagenicity (alpha m/alpha i) ranged from 2.05 to 7.99 x 10(exp -5) with the maximum value at 150 keV/micrometer. Furthermore, molecular analysis of mutants induced by charged particles indicates that higher LET beams are more likely to cause larger deletions in the hprt locus.

Chen, D. J.↗

NASA NTRS1991

Electrostatic fields in a dusty Martian environment

While there have been several studies suggesting the possibility of electrical activity on Mars, to date there have been no measurements to search for evidence of such activity. In the absence of widespread water clouds and convective storm systems similar to those on the Earth and Jupiter, the most likely candidate for the creation of electrostatic charges and fields is triboelectric charging of dust, i.e., the friction between blown dust and the ground, and of dust particles with each other. Terrestrial experience demonstrates that electric fields 5 to 15 kV-m(exp -1) are not uncommon in dust storms and dust devils in desert regions, where the polarity varies according to the chemical composition and grain size. Simple laboratory experiments have demonstrated that modest electrostatic fields of roughly 5,000 V-m(exp -1) may be produced, along with electrical spark discharges and glow discharges, in a simulation of a dusty, turbulent Martian surface environment. While the Viking landers operated for several years with no apparent deleterious effects from electrostatic charging, this may have been at least partly due to good engineering design utilizing pre-1976 electronic circuitry to minimize the possibility of differential charging among the various system components. However, free roaming rovers, astronauts, and airborne probes may conceivably encounter an environment where electrostatic charging is a frequent occurrence, either by way of induction from a static electric field or friction with the dusty surface and atmosphere. This raises the possibility of spark discharges or current surges when subsequent contact is made with other pieces of electrical equipment, and the possibility of damage to modern microelectronic circuitry. Measurements of electrostatic fields on the surface of Mars could therefore be valuable for assessing this danger. Electric field measurements could also be useful for detecting natural discharges that originate in dust storms. This detection could be performed at distances ranging from 10s of km in the case of J-charge-like discharge signatures, to planetary distances if there exists a global electrical circuit or Schumann resonance spectrum.

Sentman, D. D.↗

NASA NTRS2011

Track Structure and the Biological Effectiveness of Accelerated Particles for the Induction of Chromosome Damage

Track structure models predict that at a fixed value of LET, particles with lower charge number, Z will have a higher biological effectiveness compared to particles with a higher Z. In this report we investigated how track structure effects induction of chromosomal aberration in human cells. Human lymphocytes were irradiated in vitro with various energies of accelerated iron, silicon, neon, or titanium ions and chromosome damage was assessed in using three color FISH chromosome painting in chemically induced PCC samples collected a first cell division post irradiation. The LET values for these ions ranged from 30 to195 keV/micron. Of the particles studied, Neon ions have the highest biological effectiveness for induction of total chromosome damage, which is consistent with track structure model predictions. For complex-type exchanges 64 MeV/ u Neon and 450 MeV/u Iron were equally effective and induced the most complex damage. In addition we present data on chromosomes exchanges induced by six different energies of protons (5 MeV/u to 2.5 GeV/u). The linear dose response term was similar for all energies of protons suggesting that the effect of the higher LET at low proton energies is balanced by the production of nuclear secondaries from the high energy protons.

George, K.↗

NASA NTRS2013

Inductive Pulsed Plasma Thruster Development and Testing at NASA-MSFC

THE inductive pulsed plasma thruster (IPPT) is an electrodeless space propulsion device where a capacitor is charged to an initial voltage and then discharged producing a high current pulse through a coil. The field produced by this pulse ionizes propellant, inductively driving current in a plasma located near the face of the coil. Once the plasma is formed it can be accelerated and expelled at a high exhaust velocity by the electromagnetic Lorentz body force arising from the interaction of the induced plasma current and the magnetic field produced by the current in the coil. In the present work, we present a summary of the IPPT research and development conducted at NASA's Marshall Space Flight Center (MSFC). As a higher-power, still relatively low readiness level system, there are many issues associated with the eventual deployment and use of the IPPT as a primary propulsion system on spacecraft that remain to be addressed. The present program aimed to fabricate and test hardware to explore how these issues could be addressed. The following specific areas were addressed within the program and will be discussed within this paper. a) Conical theta-pinch IPPT geometry thruster configuration. b) Repetition-rate multi-kW thruster pulsing. c) Long-lifetime pulsed gas valve. d) Fast pulsed gas valve driver and controller. e) High-voltage, repetitive capacitor charging power processing unit. During the course of testing, a number of specific tests were conducted, including several that, to our knowledge, have either never been previously conducted (such as multi-KW repetition-rate operation) or have not been performed since the early 1990s (direct IPPT thrust measurements).2 Conical theta-pinch IPPT thrust stand measurements are presented in Fig. 1 while various time-integrated and time

Polzin, Kurt A.↗

NASA NTRS1979

Inductive energy storage for MPD thrusters

The high thrust density of the self-field magnetoplasmadynamic (MPD) thruster makes it a promising candidate for many advanced space missions. The high power requirements of this thruster lead to its operation in a pulsed mode from an energy storage device. The system characteristics of an inductive energy storage circuit with a solar array power from 25 kwe to 400 kwe are considered, by solving the circuit equations for the inductor charge and discharge phases. Using simple analytic models of the circuit components, the total system efficiency and inductance are determined as functions of the array output power and circuit resistance. The total system efficiency increases with array power and is acceptable only for low values of circuit resistance, indicating that superconducting circuitry may be desirable. The optimum charge-discharge cycle changes fundamentally as the circuit resistance is decreased through a critical value dependent on the thruster operating characteristics.

Rudolph, L. K.↗

NASA NTRS1999

Determination of Thunderstorm Anvil Ice Contents and Other Cloud Properties from Satellite Observations of Lightning

The lightning frequency model developed by Baker, Christian and Latham (1995) has been refined and extended, in an effort to provide a more realistic framework from which to examine computationally the relationships that might exist between lightning frequency f (which is now being routinely measured from satellites, using NASA/MSFC devices) and a variety of cloud physical parameters, including precipitation rate, updraught speed and non-precipitating ice content. The charging mechanism incorporated into the model is the non-inductive ice-ice mechanism - involving rebounding collisions between ice crystals and growing graupel pellets - which has achieved significant field, laboratory and modelling support in recent years. Model results indicate the existence of a simple relationship between lightning frequency f and the upward flux of ice crystals into the thunderstorm anvil. It follows that, for a particular situation, one can assign a specific mass of non-precipitating ice to an individual lightning stroke. Therefore it is possible - using satellite measurements of global lightning - to estimate the atmospheric loading of ice crystals in thunderstorm anvils: a parameter of climatological importance. One can also estimate the ice crystal loading associated with more localized storm systems. Early results from this work will be presented, together with further work on the relationships between f and other thundercloud parameters.

Blyth, Alan M.↗

NASA NTRS2014

RBE of Energetic Iron Ions for the Induction of Early and Late Chromosome Aberrations in Different Cell Types

Numerous published studies have reported the RBE values for chromosome chromosomes induced by charged particles of different LET. The RBE for chromosome aberrations in human lymphocytes exposed ex vivo showed a similar relationship as the quality factor for cancer induction. Consequently, increased chromosome aberrations in the astronauts' white blood cells post long-duration missions are used to determine the biological doses from exposures to space radiation. The RBE value is known to be very different for different types of cancer. Previously, we reported that the RBE for initial chromosome damages was high in human lymphocytes exposed to Fe ions. After multiple cell divisions post irradiation, the RBE was significantly smaller. To test the hypothesis that the RBE values for chromosome aberrations are different between early and late damages and also different between different cell types, we exposed human lymphocytes ex vivo, and human fibroblast cells and human mammary epithelial cells in vitro to 600 MeV/u Fe ions. Post irradiation, the cells were collected at first mitosis, or cultured for multiple generations for collections of remaining or late arising chromosome aberrations. The chromosome aberrations were quantified using fluorescent in situ hybridization (FISH) with whole chromosome specific probes. This study attempts to offer an explanation for the varying RBE values for different cancer types.

Zhang, Ye↗

NASA NTRS1994

The influence of dose, dose-rate and particle fragmentation on cataract induction by energetic iron ions

Because activities in space necessarily involve chronic exposure to a heterogeneous charged particle radiation field it is important to assess the influence of dose-rate and the possible modulating role of heavy particle fragmentation on biological systems. Using the well-studied cataract model, mice were exposed to plateau 600 MeV/amu Fe-56 ions either as acute or fractionated exposures at total doses of 5-504 cGy. Additional groups of mice received 20, 360 and 504 cGy behind 50 mm of polyethylene, which simulates body shielding. The reference radiation consisted of Co-60 gamma radiation. The animals were examined by slit lamp biomicroscopy over their three year life spans. In accordance with our previous observations with heavy particles, the cataractogenic potential of the 600 MeV/amu Fe-56 ions was greater than for low-Linear Energy Transfer (LET) radiation and increased with decreasing dose relative to gamma rays. Fractionation of a given dose of Fe-56 ions did not reduce the cataractogenicity of the radiation compared to the acute regimen. Fragmentation of the beam in the polyethylene did not alter the cataractotoxicity of the ions, either when administered singly or in fractions.

Medvedovsky, C.↗

NASA NTRS2012

The Biological Effectiveness of Accelerated Particles for the Induction of Chromosome Damage: Track Structure Effects and Cytogenetic Signatures of High-LET Exposure

Track structure models predict that at a fixed value of LET, particles with lower charge number, Z will have a higher biological effectiveness compared to particles with a higher Z. In this report we investigated how track structure effects induction of chromosomal aberration in human cells. Human lymphocytes were irradiated in vitro with various energies of accelerated iron, silicon, neon, or titanium ions and chromosome damage was assessed in using three color FISH chromosome painting in chemically induced PCC samples collected a first cell division post irradiation. The LET values for these ions ranged from 30 to 195 keV/micrometers. Of the particles studied, Neon ions have the highest biological effectiveness for induction of total chromosome damage, which is consistent with track structure model predictions. For complex-type exchanges 64 MeV/ u Neon and 450 MeV/u Iron were equally effective and induced the most complex damage. In addition we present data on chromosomes exchanges induced by six different energies of protons (5 MeV/u to 2.5 GeV/u). The linear dose response term was similar for all energies of protons suggesting that the effect of the higher LET at low proton energies is balanced by the production of nuclear secondaries from the high energy protons. All energies of protons have a much higher percentage of complex-type chromosome exchanges than gamma rays, signifying a cytogenetic signature for proton exposures.

George, K.↗

NASA NTRS2015

RBE of Energetic Iron Ions for the Induction of Early and Late Chromosome Aberrations in Different Cell Types

Numerous published studies have reported the Relative Biological Effectiveness (RBE) values for chromosome aberrations induced by charged particles of different LET. The RBE for chromosome aberrations in human lymphocytes exposed ex vivo has been suggested to show a similar relationship as the quality factor for cancer induction. Therefore, increased chromosome aberrations in the astronauts' white blood cells post long-duration missions are used to determine the biological doses from exposures to space radiation. However, the RBE value is known to be very different for different types of cancer. Previously, we reported that, even though the RBE for initial chromosome damages was high in human lymphocytes exposed to Fe ions, the RBE was significantly reduced after multiple cell divisions post irradiation. To test the hypothesis that RBE values for chromosome aberrations are cell type dependent, and different between early and late damages, we exposed human lymphocytes ex vivo, and human mammary epithelial cells in vitro to various charged particles. Chromosome aberrations were quantified using the samples collected at first mitosis post irradiation for initial damages, and the samples collected after multiple generations for the remaining or late arising aberrations. Results of the study suggested that the effectiveness of high-LET charged particles for late chromosome aberrations may be cell type dependent, even though the RBE values are similar for early damages.

Zhang, Ye↗

NASA NTRS2011

SA13B-1900 Auroral Charging of the International Space Station

Electrostatic potential variations of the International Space Station (ISS) relative to the space plasma environment are dominated by interaction of the negatively grounded 160 volt US photovoltaic power system with the plasma environment in sunlight and inductive potential variations across the ISS structure generated by motion of the vehicle across the Earth's magnetic field. Auroral charging is also a source of potential variations because the 51.6? orbital inclination of ISS takes the vehicle to sufficiently high magnetic latitudes to encounter precipitating electrons during geomagnetic storms. Analysis of auroral charging for small spacecraft or isolated insulating regions on ISS predict rapid charging to high potentials of hundreds of volts but it has been thought that the large capacitance of the entire ISS structure on the order of 0.01 F will limit frame potentials to less than a volt when exposed to auroral conditions. We present three candidate auroral charging events characterized by transient ISS structure potentials varying from approximately 2 to 17 volts. The events occur primarily at night when the solar arrays are unbiased and cannot therefore be due to solar array current collection. ISS potential decreases to more negative values during the events indicating electron current collection and the events are always observed at the highest latitudes along the ISS trajectory. Comparison of the events with integral >30 keV electron flux measurements from NOAA TIROS spacecraft demonstrate they occur within regions of precipitating electron flux at levels consistent with the energetic electron thresholds reported for onset of auroral charging of the DMSP and Freja satellites. In contrast to the DMSP and Freja events, one of the ISS charging events occur in sunlight.

Minow, Joseph I.↗

NASA NTRS2015

Development of Long-Lifetime Pulsed Gas Valves for Pulsed Electric Thrusters

It is advantageous for gas-fed pulsed electric thrusters to employ pulsed valves so propellant is only flowing to the device during operation. The propellant utilization of the thruster will be maximized when all the gas injected into the thruster is acted upon by the fields produced by the electrical pulse. Gas that is injected too early will diffuse away from the thruster before the electrical pulse can act to accelerate the propellant. Gas that is injected too late will miss being accelerated by the already-completed electrical pulse. As a consequence, the valve must open quickly and close equally quickly, only remaining open for a short duration. In addition, the valve must have only a small amount of volume between the sealing body and the thruster so the front and back ends of the pulse are as coincident as possible with the valve cycling, with very little latent propellant remaining in the feed lines after the valve is closed. For a real mission of interest, a pulsed thruster can be expected to pulse at least 10(exp 10) - 10(exp 11) times, setting the range for the number of times a valve must open and close. The valves described in this paper have been fabricated and tested for operation in an inductive pulsed plasma thruster (IPPT) for in-space propulsion. In general, an IPPT is an electrodeless space propulsion device where a capacitor is charged to an initial voltage and then discharged, producing a high-current pulse through a coil. The field produced by this pulse ionizes propellant, inductively driving current in a plasma located near the face of the coil. Once the plasma is formed, it can be accelerated and expelled at a high exhaust velocity by the electromagnetic Lorentz body force arising from the interaction of the induced plasma current and the magnetic field produced by the current in the coil. The valve characteristics needed for the IPPT application require a fast-acting valve capable of a minimum of 10(exp 10) valve actuation cycles. Since even 10(exp 9) cycles is well above anything demonstrated, this lower value was selected as the design point for the present work. The valve seal must remain leak-tight throughout operation, and the body must maintain a low internal leakage at relatively high operating temperatures. The full set of design requirements used for this program are summarized in Table 1. In this work, we describe two pulsed gas valves that have been fabricated to have long lifetime and demonstrate the characteristics listed above. The first is a miniaturized, conventional electromagnet-based valve while the second is a piezoelectric-based valve design. The conventional valve, shown in Fig. 1, is opened by use of a solenoid electromagnetic actuator. When current is applied to the solenoid coil, magnetic forces pull the plunger away from the valve seat, allowing fluid to flow through the valve. Removal of electrical current permits the spring and fluid pressure to seat the plunger, halting the flow of fluid. The valve body is fabricated from 304L corrosion resistant steel (CRES) and while the parts that form the magnetic circuit are fabricated from 430 CRES. This material does not have optimum magnetic properties, but its corrosion resistance permits incorporation into a design without requiring an additional plating process. A viton O-ring compound (Parker V0884-75), selected for its mechanical strength at elevated temperatures, was used for the valve seat seal. The design was based solely on the use of analytical sizing calculations, as opposed to a more rigorous finite element analysis. While this valve is small and relatively lightweight, it does not represent a design that is optimized for mass and/or a given volume envelope. The piezoelectric valve is a "puller" valve design. Applying a voltage to the piezo crystal causes it to elongate and pull a pintle off the seat, opening the valve. The valve seal consists of the pintle with an external, spherically-formed tip fabricated from Inconel 625 and the seat fabricated from Inconel 718 and possessing a spherical surface that permits the pintle to align with the valve body centerline after each cycle, maintaining the circumferential metal-on-metal seal. The contact stress on the valve pintle is expected in the worst case to be 96.5 MPa (14 ksi), which is well below the work hardening value of 413.7 MPa (60 ksi) for Inconel 615. Results presented in this work will compare the performance of each valve, both to each other and to the valve design requirements given in Table 1. In general, both valves meet all the performance envelope requirements, but the piezoelectric valve consumes considerably less power and has a faster response time over the conventional solenoid-driven valves.

Burkhardt, Wendel M.↗

NASA NTRS2007

Observed Enhancement of Reflectivity and Electric Field in Long-Lived Florida Anvils

A study of two long-lived Florida anvils showed that reflectivity >20 dBZ increased in area, thickness and sometimes magnitude at mid-level well downstream of the convective cores. In these same regions electric fields maintained strengths >10 kV m1 for many tens of minutes and became quite uniform over tens of kilometers. Millimetric aggregates persisted at 9 to 10 km for extended times and distances. Aggregation of ice particles enhanced by strong electric fields might have contributed to reflectivity growth in the early anvil, but is unlikely to explain observations further out in the anvil. The enhanced reflectivity and existence of small, medium and large ice particles far out into the anvil suggest that an updraft was acting, perhaps in weak convective cells formed by instability generated from the evaporation and melting of falling ice particles. We conclude that charge separation must have occurred in these anvils, perhaps at the melting level but also at higher altitudes, in order to maintain fields >10 kV m 1 at 9 to 10 km for extended periods of time over large distances. We speculate that charge separation occurred as a result of ice-ice particle collisions (without supercooled water being present) via either a non-inductive or perhaps even an inductive mechanism, given the observed broad ice particle spectra, the strong pre-existing electric fields and the many tens of minutes available for particle interactions. The observations, particularly in the early anvil, show that the charge structure in these anvils was quite complex.

Dye, James E.↗