166 Matching Results

Search Results

Advanced search parameters have been applied.

D7H-test results

Description: Data were reduced from the voltage-time relations stored in files D7H001 to D7H090 on HP1000. The I-B calibration curve is included. The data base is shown and can be used by the 9845B. The data include the quench location, Q/sub 2/ layer 1 top, Q/sub 3/ layer 1 bottom and the quench current and its normalized value with respect to short sample, I/sub c/ = 4920A at 4.4 K, I/sub c/ = 6710 A at 1.8 K. The resistance (..cap omega../cm) was calculated using the propagation time according to the voltage change across the measured sections. The conductor potential length are L/sub 5/ /sub 9/ = 48.6 cm, L/sub 6/ /sub 10/ = 17.9 cm, L/sub 7/ /sub 11/ = 40.6 cm. The turn to turn velocity V/sub t/ was calculated dividing the nominal turn to turn distance (58 mil) by the propagation time (Trans. Time). The quench time T/sub q/ was measured from the time the resistive rise starts until the energy extraction system fires.
Date: July 30, 1982
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Cooldown of an infinitely long hollow cylinder and application to the SSC cooldown

Description: This writeup is an attempt to calculate some of the cooldown parameters of a long string of magnets such as the case might be for the SSC. Besides the cooldown time, temperature gradients along a magnet, and in the transverse direction, are influenced by the mass flow of the refrigerant. A number of assumptions and simplifications have been made so that an analytical solution can be obtained. Part I of this report assumes a one dimensional model with a finite axial conductivity and infinite transverse conductivity. Part II, considers the cooldown in the transverse direction only. A common example for both parts points out the limitation of the assumptions made in Part I and suggests the need for a two dimensional time dependent model T = T(r,z,t).
Date: February 1, 1985
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Calorimetry in superfluid He II to measure losses in superconducting magnets

Description: A method using calorimetry to measure magnet losses in pressurized Helium II is described. The isothermal nature of He II is used in measuring the overall heat capacity of the system and the net refrigeration power. During the measurements, the refrigeration power is held fixed, and the system (400 liters) temperature is near 1.92 K. The calorimetric measurement was calibrated against known power inputs between 1 and 20 W. This technique can even measure heat loads higher than the available refrigeration. Results of loss measurement on two dipole magnets are reported.
Date: April 1, 1982
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Gravitational convection of subcooled He I and the transition into superfluid He II at atmospheric pressure

Description: The propagation velocity of the He II-He I interface at 1 atm was measured in a large experimental cryostat. During cooldown, He II at the lambda temperature propagates up and down in a gravitational field with an interface velocity V/sub b/ which depends on the He I temperature T/sub infinity/. For T/sub infinity/ = 4 K, V/sub b/ = 0.1 mm/sec and for T/sub infinity/ = 2.2 K, V/sub b/ = 2 mm/sec. During warm up, He I propagated with an interface velocity V/sub b/ > 30 mm/sec that was measured as low bound only. A mathematical model gives a temperature dependent interface thickness in the range 2 x 10/sup -3/ approx. = delta < 4 x 10/sup -1/ mm.
Date: June 1, 1980
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Magnetic conditioning in superfluid

Description: Improvements in superconducting magnet technology have reduced to a handful the number of training quenches typical of dipole magnets. The number of training quenches in long (17 m) and short (1--2 m) SSC magnets are now about the same (operating at 6.6 tesla and 4.4 K). Yet the steps necessary to totally eliminate training are in the future RandD plans for magnet construction and conductor motion prevention. The accepted hypothesis is that Lorentz forces and poor mechanical properties of superconducting cables are the cause of conductor motion. Conductor motion reduces the stored energy in the cable by converting it into heat. The small amount of heat generated (millijoules) during motion is usually enough to quench the magnet when it is close to short sample. During training, the magnet performance normally improves with the number of quenches. It is not the quench itself that improves magnet performance but rather the fact that once conductor motion has occurred it will probably not repeat itself unless subjected to higher forces. Conditioning is a process that enables the magnet to reduce its stored energy without causing a premature quench. During the conditioning process the magnet is further cooled from its operating temperature of 4.4 K to 1.8 K by converting He I into He II. As a result the magnet is placed in a state where it has excess stability as well as excellent heat transfer capabilities. Although this does not eliminate motion, if the magnet is now cycled to /approximately/10% above its operating field at 4.4 K (which is above short sample) the excess stability should be enough to prevent quenching and reduce the probability of conductor motion and training once the magnet has been warmed back up to its operating temperature of 4.4 K. 3 refs., 5 figs.
Date: August 1, 1988
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Measurements of heat transfer to helium II at atmospheric pressure in a confined geometry

Description: Recently the enhanced heat removal capability of unsaturated superfluid helium II has been exploited in fusion and accelerator dipole magnets. In superfluid the internal convection mechanism dominates the heat removal process and orientation with respect to gravity becomes of secondary importance. Heat transfer, however, can be influenced by the thermodynamic state of the liquid, especially with regard to possible phase transformations. The transformation from non-saturated He II must involve an He I state before the film boiling transition is experienced. Some steady state measurements of heat transfer to non-saturated He II have been previously reported. In typical magnet designs, cooling passages between turns result from gaps between the electrical insulation, and are typically on the order of a fraction of a millimeter. The purpose of the work reported here is to measure the attenuation of the heat transfer within such a restrictive geometry.
Date: August 1, 1981
Creator: Warren, R.P. & Caspi, S.
Partner: UNT Libraries Government Documents Department

Design of a 9 T 5 cm bore dipole

Description: A conceptual design is presented for an accelerator dipole magnet intended for a collider application with central field of 9T and a coil bore diameter of 5 cm. Emphasis is on identifying a superconducting cable and a coil design that requires little development and which will result in a reasonable operating ''margin.'' Conventional NbTi superconducting cable is used, operating at 2 K to increase the critical current density. The cable and strand proposed are similar to designs now being manufactured for the SSC and the Tevatron. 4 refs., 9 figs., 1 tab.
Date: July 1, 1988
Creator: Taylor, C.E. & Caspi, S.
Partner: UNT Libraries Government Documents Department

Heat transfer through He II in a 9. 6 m long 35 mm ID tube

Description: The limiting heat flux at the onset of He I was measured in a 9.6 m long tube of 35 mm ID at a bath temperature between 1.8 K and T/sub lambda/ and a pressure of 1 atm. The measured limiting heat flux during axial heating is 50% more than end heating at the same bath temperature. Both cases agrees with the Gorter-Mellink mutual friction theory.
Date: July 1, 1984
Creator: Caspi, S. & Schafer, R.V.
Partner: UNT Libraries Government Documents Department

Source, origin, and propagation of quenches measured in superconducting dipole magnets

Description: Transitions from the superconducting to normal state at 4.4 and 1.8 K in several model accelerator dipoles were recorded by a fast data-acquisition system. The resistive voltage rise in the conductor during the transitions is used to determine accurately the location of the quench source in the magnets and to estimate the axial and turn-to-turn quench velocities. The quench velocity, temperature evolution and energy deposition in the coil were calculated using the program QUENCH, and are in reasonable agreement with the data. In the two dipole magnets studied, the transitions almost always occurred in the regions of highest field. In one coil the high field region is in the straight section because the field in this region is enhanced by iron support rings. In the other magnet the high field region is at the end, in the innermost turn of the first layer. Some quenches were preceded by large voltage spikes that can be ascribed to conductor motion. Other quenches do not appear to be associated with any large energy release. Acoustic emission (AE) was monitored during the tests and AE bursts were observed simultaneous with the initial voltage spike. An increased AE signal continued as the quench progressed.
Date: November 1, 1982
Creator: Caspi, S. & Hassenzahl, W.V.
Partner: UNT Libraries Government Documents Department

Forces in a thin cosine(n{theta}) helical wiggler

Description: We commence with the derivation of the Lorentz force density on a surface of discontinuity based on the expressions of fields and currents previously derived (Appendix A). Applying such Lorentz body forces to the equilibrium condition of an infinitesimal surface area yields a set of differential equations for the local total force. In attempting to solve such differential equations it may prove to be useful and prudent to reduce their complexity by first transforming all fields, current densities and Lorentz forces to a coordinate system that is aligned with the direction of the current flow. A Frenet--Serret rotating unit vector coordinate system may serve such a purpose and will reduce the 3 components of the Lorentz force to 2. We proceed with obtaining such a conversion through the use of differential geometry, although a more straight forward approach may exist through the use of surface developability and coordinate transformation. Following a solution to the force equations we continue with and example of a nested set of a combined function dipole and quadrupole that employ an identical periodicity {omega}. The expressions for the self force and the mutual force on each magnet element are obtained. Finally, by reducing the periodicity {omega} to zero we obtain the force expressions for long (2D) multipole magnets including both the self and interactive forces.
Date: March 1, 1995
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

The vector potential and stored energy of thin cosine (n{theta}) helical wiggler magnet

Description: Expressions for pure multipole field components that are present in helical devices have been derived from a current distribution on the surface of an infinitely thin cylinder of radius R. The strength of such magnetic fields varies purely as a Fourier sinusoidal series of the longitudinal coordinate Z in proportion to cos(n{theta}- {omega}{sub m}z), where {omega}{sub m} = (2m-1){pi}/L, L denotes the half-period and m = 1, 2, 3 etc. As an alternative to describing such field components as given by the negative gradient of a scalar potential function (Appendix A), one of course can derive these same fields as the curle of a vector potential function {rvec A}--specifically one for which {nabla} {times} {nabla} {times} {rvec A} = 0 and {nabla}{center_dot}{rvec A} = 0. It is noted that we seek a divergence-free vector that exhibits continuity in any of its components across the interface r = R, a feature that is free of possible concern when applying Stokes` theorem in connection with this form of vector potential. Alternative simpler forms of vector potential, that individually are divergence-free in their respective regions (r < R and r > R), do not exhibit full continuity on r = R and whose curl evaluations provide in these respective regions the correct components of magnetic field are not considered here. Such alternative forms must differ merely by the gradient of scalar functions that with the divergence-free property are required to be ``harmonic`` ({nabla}{sup 2}{Psi} = 0).
Date: December 1, 1995
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Application of program POISSON to axially-symmetric problems - magnetostatic and electrostatic - with use of a prolate spheroidal boundary

Description: A version of the relaxation program POISSON has been produced that, for magnetostatic problems, can apply a boundary condition consistent with no external sources being present. This capability includes the treatment of axially-symmetric cases (using A* = rhoA as the working variable) with a boundary whose form is that of a prolate spheroid (and hence tends toward spherical in the limit eta = a/..sqrt..a/sup 2/ - b/sup 2/ ..-->.. infinity). (LBL-18798/UC-28 (December 1984)). The treatment of electrostatic problems (to obtain solutions for the scalar potential V) necessarily must differ in detail from the treatment of magnetostatic problems in cases of axial symmetry. It seems desirable, therefore, first to review the magnetostatic treatment that has been adopted for such axially-symmetric magnetostatic problems and then to suggest an analogous treatment that might similarly be introduced into the program to permit solution of similar electrostatic problems (again through the introduction of a prolate spheroidal boundary).
Date: January 1, 1986
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

Incorporation of a circular boundary condition into the program POISSON

Description: Two-dimensional problems in electrostatics or magnetostatics frequently are solved numerically by means of relaxation techniques. In many such problems the ''sources'' (charges or currents, and regions of permeable material) lie exclusively within a finite closed boundary curve and the relaxation process in principle then could be confined to the region interior to such a boundary - provided a suitable boundary condition is imposed onto the solution at that boundary. The present notes discuss and illustrate the use of a boundary condition of such a nature as to imply the absence of external sources, in order thereby to avoid the inaccuracies and more extensive meshes present when alternatively a simple Dirichlet or Neumann boundary condition is specified on a somewhat more remote outer boundary.
Date: March 2, 1984
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

Incorporation of toroidal boundary conditions into program POISSON

Description: A technique is developed for introduction of a boundary condition applicable to relaxation computations for magnetic problems with axial symmetry and with no sources (currents, or magnetized material) external to the boundary. The procedure as described in this note is restricted to cases in which the (toroidal) boundary will surround completely the region of physical interest but will not encompass the axis of rotational symmetry. The technique accordingly provides the opportunity of economically excluding from the relaxation process regions of no direct concern in the immediate neighborhood of the symmetry axis and hence can have useful application to annular magnetic devices with axial symmetry. The procedure adopted makes use internally of the characteristic form of the vector-potential function, in a source-free region, when expressed in toroidal coordinates. The relevant properties of associated Legendre functions of half-integral degree are summarized in this connection and their introduction into the program POISSON is outlined. Results of some test cases are included, to illustrate the application of this technique for configurations with median-plane symmetry. 8 refs., 9 figs.
Date: July 17, 1987
Creator: Laslett, L.J.; Caspi, S. & Helm, M.
Partner: UNT Libraries Government Documents Department

Incorporation of a boundary condition to numerical solution of POISSON's equation

Description: Two-dimensional and axially-symmetric problems in electrostatics, magnetostatics or potential fluid flow frequently are solved numerically by means of relaxation techniques -- employing, for example, the finite-difference program POISSON. In many such problems, the ''sources'' (charges or currents, vorticity, and regions of permeable material) lie exclusively within a finite closed boundary curve and the relaxation process, in principle, then can be confined to the region interior to such a boundary -- provided that a suitable boundary condition is imposed on the solution at the boundary. This paper is a review and illustration of a computational method that uses a boundary condition of such a nature as to avoid the inaccuracies and more extensive meshes present when, alternatively, a simple Dirichlet or Neumann boundary condition is specified on a somewhat more remote outer boundary. 2 refs., 5 figs., 1 tab.
Date: October 1, 1988
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

Incorporation of boundary condition into the program POISSON

Description: Two dimensional Cartesian and axially-symmetric problems in electrostatics or magnetostatics frequently are solved numerically by means of relaxation techniques - employing, for example, the program POISSON. In many such problems the ''sources'' (charges or currents, and regions of permeable material) lie exclusively within a finite closed boundary curve and the relaxation process in principle then could be confined to the region interior to such a boundary - provided a suitable boundary condition is imposed onto the solution at the boundary. This paper discussed and illustrates the use of a boundary condition of such a nature, in order thereby to avoid the inaccuracies and more extensive meshes present when alternatively a simple Dirichlet or Neumann boundary condition is specified on a somewhat more remote outer boundary.
Date: August 1, 1985
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

Incorporation of an elliptical boundary condition into the program POISSON

Description: This report is the third in a series which takes into account the boundary condition in electromagnetic problems such as used by the program POISSON. Here we extend the analysis to permit the use of an elliptical boundary both for two-dimensional and ax
Date: December 1, 1984
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

Numerical solution of boundary condition to POISSON's equation and its incorporation into the program POISSON

Description: Two dimensional cartesian and axially-symmetric problems in electrostatics or magnetostatics frequently are solved numerically by means of relaxation techniques - employing, for example, the program POISSON. In many such problems the ''sources'' (charges or currents, and regions of permeable material) lie exclusively within a finite closed boundary curve and the relaxation process in principle then could be confined to the region interior to such a boundary - provided a suitable boundary condition is imposed onto the solution at the boundary. This paper discusses and illustrates the use of a boundary condition of such a nature in order thereby to avoid the inaccuracies and more extensive meshes present when alternatively a simple Dirichlet or Neumann boundary condition is specified on a somewhat more remote outer boundary.
Date: May 1, 1985
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

The use of toroidal boundary conditions in the program POISSON

Description: In circular particle accelerators of moderate size, one cannot entirely neglect the curvature of the structure and of the guide field. In practice, one may wish to restrict the region of analysis to that near the working aperture, while excluding a very substantial area closer to (and including) the axis of rotational symmetry. In this way, a more efficient mesh can be generated for a program such as POISSON. In restricting the solution to the region of interest, there must be concern regarding a suitable termination of the problem at the boundary of the mesh. For these reasons, we have employed toroidal coordinates in constructing the boundary to a relaxation mesh, and in formulating the boundary conditions that then would be imposed at such boundaries. 11 refs., 6 figs.
Date: September 1, 1987
Creator: Caspi, S.; Helm, M. & Laslett, L.J.
Partner: UNT Libraries Government Documents Department

Vector Potential and Stored Energy of a Quadrupole Magnet Array

Description: The vector potential, magnetic field and stored energy of a quadrupole magnet array are derived. Each magnet within the array is a current sheet with a current density proportional to the azimuthal angle 2{theta} and the longitudinal periodicity (2m-1){pi}/L. Individual quadrupoles within the array are oriented in a way that maximizes the field gradient The array does not have to be of equal spacing and can be of a finite size, however when the array is equally spaced and is of infinite size the solution can be simplified. We note that whereas, in a single quadrupole magnet with a current density proportional to cos2{theta} the gradient is pure, such purity is not preserved in a quadrupole array.
Date: March 15, 1999
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Forces and Stored Energy in Thin Cosine (n0) Accelerator Magnets

Description: We wish to compute Lorentz forces, equilibrium stress and stored energy in thin multipole magnets (Fig.1), that are proportional to cos(n{theta}) and whose strength varies purely as a Fourier sinusoidal series of the longitudinal coordinate z (say proportional to cos (2m-1){pi}z/L where L denotes the half-period and m = 1,2,3...). We shall demonstrate that in cases where the current is situated on such a surface of discontinuity at r = R (i.e. J = f({theta},z)), by computing the Lorentz force and solving the state of equilibrium on that surface, a closed form solution can be obtained for single function magnets as well as for any combination of interacting nested multi function magnets. The results that have been obtained, indicate that the total axial force on the end of a single multipole magnet n is independent (orthogonal) to any other multipole magnet i as long as n {ne} i. The same is true for the stored energy, the total energy of a nested set of multipole magnets is equal to the some of the energy of the individual magnets (of the same period length 2L). Finally we demonstrate our results on a nested set of magnets a dipole (n = 1) and a quadmpole (n=2) that have an identical single periodicity {omega}{sub 1}. We show that in the limiting 2D case (period 2L tends to infinity), the force reduces to the commonly known 2D case.
Date: March 18, 1996
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

Forces in a Thin Cosine (nTheta) Helical Wiggler

Description: We wish to calculate the Lorentz body force associated with pure multipole helical magnetic fields (i.e, proportional to cos(n{theta})) whose strength varies purely as a Fourier sinusoidal series of the longitudinal coordinate z (say proportional to cos(2m-1){pi}z)/L, where L denotes the half-period of the wiggler field and m= 1,2,3... We also wish to apply such forces to the current sheet, and solve for the stress distribution required to maintain such a coil in equilibrium. In the calculations of Lorentz forces we include the self field contribution as well as possible contributions arising from additional nested helical windings. We shall demonstrate that in cases where the current is situated on a surface of discontinuity at r=R (i.e. J=f({theta},z)) and the Lorentz body force is integrated on that surface, a closed form solution for the stress distribution can be obtained and such a solution includes contributions from possible nested multi pole magnets. Finally we demonstrate that in the limiting 2D case where the field strength does not vary with z ( period 2L tends to infinity) the stress reduces to known 2D expressions.
Date: May 2, 1997
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

The 3D Vector Potential, Magnetic Field and Stored Energy in a Thin cos2 theta Coil Array

Description: The vector potential and the magnetic field have been derived for an arrays of quadrupole magnets with thin Cos(2{theta}) current sheet placed at r = R.{sup bc}. The field strength of each coil within the array, varies purely as a Fourier sinusoidal series of the longidutinal coordinate z in proportion to {omega}{sub m}z, where {omega}{sub m} = (2m-1){pi}/L, L denotes the half-period, and m = 1,2,3 etc. The analysis is based on the expansion of the vector potential in the region external to the windings of a linear 3D quad, and a revision of that expansion by the application of the 'Addition Theorem' from that around the coil center to that around any arbitrary point in space.
Date: July 9, 1997
Creator: Caspi, S.
Partner: UNT Libraries Government Documents Department

High-field strong-focusing undulator designs for X-ray Linac Coherent Light Source (LCLS) applications

Description: Linac-driven X-Ray Free Electron Lasers (e.g., Linac Coherent Light Sources (LCLSs)), operating on the principle of single-pass saturation in the Self-Amplified Spontaneous Emission (SASE) regime typically require multi-GeV beam energies and undulator lengths in excess of tens of meters to attain sufficient gain in the 1{angstrom}--0.1{angstrom} range. In this parameter regime, the undulator structure must provide: (1) field amplitudes B{sub 0} in excess of 1T within periods of 4cm or less, (2) peak on-axis focusing gradients on the order of 30T/m, and (3) field quality in the 0.1%--0.3% range. In this paper the authors report on designs under consideration for a 4.5--1.5 {angstrom} LCLS based on superconducting (SC), hybrid/PM, and pulsed-Cu technologies.
Date: May 1, 1995
Creator: Caspi, S.; Schlueter, R. & Tatchyn, R.
Partner: UNT Libraries Government Documents Department