No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Comparison of Ferrite Materials for Pulse Applications
Abstract
Materials are the limiting factor in many pulse power projects.
The magnetic materials available from several manufacturers were
experimentally compared for their usefulness in high-speed magnetic
field applications. This particular application is a high-speed kicker
magnet for injection of 150 GeV antiprotons into the Tevatron at
Fermilab
... We have used a model of the kicker prototype following the description in [4]. The frequency dependence of the ferrite permeability (CMD5005) has been obtained from the supplier's data sheet (Ceramic Magnetics Inc., Fairfield, NJ), whereas we have taken CMD5005 = 12, see [11]. InFig. 1, top, we compare the numerical results for the horizontal impedance with measurement. ...
Maintaining the impedance budget is an important task in the planning of any new accelerator facility. While estimates from analytical computations and measurements play a central role in doing so, numerical calculations have become an important alternative today. On the basis of the Finite Integration Technique, we have developed a simulation tool for the direct computation of coupling impedances in frequency domain. After discussing the special features of our code as compared to commercial programs, we present results for the transverse and longitudinal coupling impedances of the SNS extraction kicker and the injection/extraction system of the heavy-ion synchrotron SIS-18 at GSI.
... Its properties were tested and compared with other ferrite materials. The results of these tests are described elsewhere [2]. Since we have a tight tolerance on field uniformity across the gap, POISSON was used to optimize the shape of the pole tip. ...
There are six proton and six antiproton bunches used at the
present time for Tevatron Collider operation. As the luminosity is
increased for a fixed number of bunches, the number of interactions per
bunch crossing increases. The quality of the physics data taken by the
CDF and DO detectors is enhanced by reducing the number of interactions
per bunch crossing. To this end it is planned to collide 36 proton
bunches with 36 anti-proton bunches. To do this, it is necessary to
construct and install a new 150 GeV injection kicker system with a
faster rise and fall time than the existing injection kickers. This
paper will describe the design, construction and testing of the new
kicker magnet along with the associated spark gap pulsers, pulse forming
lines and trigger circuits. Difficulties and our solutions are also
presented
... Each cell is comprised of a high frequency Nickel- Zinc ferrite block. The ferrite is CMD 5005, made by Ceramic Magnetics [5]. The ferrites are 2.54 cm thick, and have a 0.635 cm spacing. ...
This kicker functions to inject 6 batches of 8 GeV beam vertically
into the Fermilab Main Injector from the Booster. The trajectory of this
beam must be bent 1.05 mr to place it onto equilibrium orbit. This
kicker system produces a nominal integrated field kick of 0.309 kG-m.
Specifications require a 1.6 microsecond pulse with a 50 ns risetime
(1%-99%) and a 150 ns falltime (99%-5%). The system is comprised of
three magnets and power supplies. Each is composed of a resonant
charger, a 25 Ω PFL, a thyratron-based pulsed power supply, a
ferrite “C” magnet, and a FluorinertTM-cooled
resistive load. Design details and measurements are presented
The study of beam dynamics and the localization of potential sources of instabilities are important tasks in the design of modern, high-intensity particle accelerators. In the case of synchrotrons and storage rings, coupling impedance data are needed to characterize the parasitic interaction of critical components with the beam. In this article we demonstrate the application of numerical field simulations to the computation of transverse kicker coupling impedances. Based on the D simulation results, a parametrized model is developed to incorporate the impedance of an arbitrary pulse-forming network attached to the kicker. Detailed comparisons of numerical results with twin-wire and direct measurements are discussed at the example of the Spallation Neutron Source extraction kicker.
A high precision & non-destructive types of current monitors using Ni-Zn ferrite toroids for the measurements of electron beam currents has been developed. This monitoring system consists of Ni-Zn ferrite toroidal cores, pickup coils, electromagnetic shields, a monitoring housing, current amplifiers & ceramic ducts. The fast current monitors showed fast rise & fall times (< 3 ns), the linearity within 2%, the high sensitivity (0.05 V/mA at 50 Ω load ) & good S/N ratio.
INTRODUCTION can achieve a flux of 136 gauss in that gap. To develop the required kick, a magnetic length of 2.5 meters is required. To achieve a reasonable rise time, three 83.3 cm long kicker magnets are used. The inductance as calculated from aperture dimensions is 598 nH. The field propagation time through the magnet can be calculated from the inductance and the characteristic impedance as: In order to maximize the efficiency of the injection process into the Main Injector Ring, the beam gap required for the extraction and injection kickers can be reduced. A switching magnet which will achieve full field within 1% on the order of 30 to 40 nSec is required to achieve efficient transfers of beam between Booster and the MIR with the removal of only one bunch from the Booster ring. The magnet designed to perform this task is a 25W traveling wave device which reaches its full field of 136 Gauss in 30 nSec. The field is developed across an apertu
To design ferrite core accelerator cells and pulse power sources for use in linear induction accelerators, it is necessary to know how the ferrite behaves when its magnetization reverses in times of the order of 100 ns. To meet this need, a simple model has been developed to describe fast flux reversal in soft ferrite cores. The core material is represented by an effective spherical grain whose size, shape, and orientation of crystal axes are assumed to represent averages over many grains. The magnetization mechanism considered is the collapse of an effective spherical domain within the effective grain. A magnetization equation is derived for the field needed to drive the collapsing domain wall against the viscous force of spin relaxation damping. This equation is compared with experimental data taken on small cores of TDK PE11BL which were reset by a dc field and reversed in about 150 ns. The model fits the data well over a large part of the flux reversal for a wide range of dB/dt. The values of the parameters found from fitting the data are related to the magnetic properties. Alternative domain configurations which have been investigated are also mentioned.
The constitutive law relating the time rate of change of the
magnetic field H to that of the flux density B , via a
differential equation, yields a faithful and yet computationally
tractable representation of magnetic hysteresis. The equation is used to
develop a theory of rate-independent and rate-dependent hysteresis in
ferrites, ferromagnetic materials, magnetic thin films, and permanent
magnetic materials. The theory provides mathematical expressions for the
initial magnetization curve, the anhysteretic curve, the major loop, the
symmetric and asymmetric minor loops, and the energy loss associated
with their traversal. Functional forms for two material functions that
appear in the equation can be scaled to measured values of the closure
point, the remanence, the coercivity and, for rate-dependent
applications, the nonlinear changes in the loop area and energy loss
that accompany increases in dB / dt and
dH / dt . Variations in loop shape and coercive point
with angle observed in uniaxially anisotropic materials are described.
Sample calculations are presented
The phenomena of absorption and dispersion observed in magnetic ferrites at frequencies above one Mc/s are discussed making the assumption that at these frequencies no contribution to the magnetisation is made by the Bloch boundaries.For pure and unstrained polycrystalline aggregates of cubic crystals the following relation between the critical frequency ω0 and the initial susceptibility χ is found: , where and M the magnetic moment per cm3. In deriving this equation the damping is assumed to be small. It is further shown that internal stresses tend to increase the losses at lower frequencies and make the rise in tan δ with frequency less steep. This is actually borne out by experiment.
As part of the Fermilab Tevatron upgrade, a 6.25 Ω ferrite
loaded traveling wave kicker magnet has been designed. The critical
parameters are the field rise time and flatness during and after the
pulse. A picture frame pole piece configuration was chosen which
requires two pulses of equal amplitude but opposite polarity. Low
inductance, high voltage capacitors placed between each of the pole
pieces provide the shunt reactance necessary to achieve the 6.25 Ω
impedance. Cross coupling adjacent cells is used to improve the
transient response of the magnet. The compensated termination resistors
are built into the magnet to minimize reflections. Two spark gap pulsers
provide the two 4800 A fast rise time current pulses necessary to drive
this magnet. The field in this 2.4 m long magnet rises to 1055 G in less
than 400 ns. This paper describes the design choices involved with this
system and preliminary test results
The ISA beam abort (extraction) system must be highly efficient, in the sense of producing minimum beam loss, and reliable to prevent serious damage to accelerator components by the circulating high-energy beams. Since the stored beams will be debunched, the low-loss requirement can be met only with ultra-thin extraction septa and/or fast-acting kickers. This paper examines the design of the ISA extraction kickers subject to a set of extraction channel constraints and a given maximum working voltage. Expressions are derived for determining system parameters for both a lumped parameter magnet and a delay-line magnet. Using these relationships, design parameters are worked out for several possible system configurations. The paper also describes the construction of a full-scale prototype module of the kicker and summarizes the preliminary test results obtained with the module.
The differential equation that underlies the theory of
rate-independent hysteresis for isoperm ferromagnetic materials is
modified and extended to rate-dependent nonisoperm materials. The theory
and its extension exhibit all of the important features of ferromagnetic
hysteresis, including the existence and stability of minor loops. Both
are well suited for use in numerical field-solving codes. Examples in
which the material functions are simple combinations of analytic
functions are presented for MnZn ferrite, NiZn ferrite, NiFe tape, and
CoCr thin film. Also presented is a procedure for constructing a
two-dimensional vector model that yields bell-shaped and M-shaped curves
for graphs of the angular variation of the coercive field
In this paper a method is presented for determining the complex permittivity and permeability of linear materials in the frequency domain by a single time-domain measurement; typically, the frequency band extends from VHF through X band. The technique described involves placing an unknown sample in a microwave TEM-mode fixture and exciting the sample with a subnanosecond baseband pulse. The fixture is used to facilitate the measurement of the forward- and back-scattered energy, s21(t) and s11(t), respectively. It is shown in this paper that the forward- and back-scattered time-domain "signatures" are uniquely related to the intrinsic properties of the materials, namely, e* and ¿*. By appropriately interpreting s21(t) and s11(t), one is able to determine the real and imaginary parts of ¿ and ¿ as a function of frequency. Experimental results are presented describing several familiar materials.
With the advent of the computer and automatic test equipment, new techniques for measuring complex dielectric constant (ε) and permeability (µ) can be considered. Such a technique is described where a system is employed that automatically measures the complex reflection and transmission coefficients that result when a sample of material is inserted in waveguide or a TEM transmission line. Measurement results of ε and µ for two common materials are presented.