Field singularities and coil fields
Transcription
Field singularities and coil fields
Field Computation and Magnetic Measurements for Accelerator Magnets Superconducting Accelerator Magnets (Field singularities and coil fields) Stephan Russenschuck, 2015 Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Where we are ➔ We have studied the mathematical foundations of magnetic fields – – – – Vectorfields Lumped circuit techniques for normal conducting magnets The Laplace equation Harmonic fields ➔ So far we assumed that the fields on the domain boundary where known (from measurements or calculations) ➔ Now we need to address how to calculate these fields – The field of line currents and the Biot Savart law – Development of these field solutions into the Eigenfunctions Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 2 Rutherford (Roebel) Kabel, Strand, Nb-‐Ti Filament Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 3 Cabling Machine for Rutherford Cables Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Turk’s Head Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Field of Line Currents Why bother? Reciprocity; except for sign it does not matter if we exchange the source and field points Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Greens Functions of Free Space Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Greens Functions of Free Space Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Greens Functions of Free Space Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Green’s Functions of Free Space Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Green’s Functions of Free Space But what if boundaries are present? Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot-‐Savart’s Law This works only in Cartesian Coordinates Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot-‐Savart’s Law This works only in Cartesian Coordinates Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot-‐Savart’s Law This works only in Cartesian Coordinates Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot-‐Savart’s Law This works only in Cartesian Coordinates Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot Savart’s Law But wait a minute: Are we finished? Are we sure that the divergence of the vector potential is zero as it was required for the Laplace equation? Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot Savart’s Law But wait a minute: Are we finished? Are we sure that the divergence of the vector potential is zero as it was required for the Laplace equation? Current loops must always be closed and must not leave the problem domain Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Biot-‐Savart’s Law for Line Currents Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 11 Vector Potential of a Line Current Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Field of a Line Current Segment Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Field of a Line Current (Infinitely Long) Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Field of a Line Current (Infinitely Long) Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Field of a Line Current (Infinitely Long) Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Field of a Line Current Appearance of elliptic integrals: To be solved numerically. On axis: In the center: Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Magnetic Dipole Moment Far field approximation Definition Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Solid Angle and Magnetic Scalar Potential Solid angle (easy to compute) gives the magnetic scalar potential of a current loop Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Imaging Current Method Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Imaging Current Method Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Notes on the Imaging Method ➔ Domain 1: Domain with current sources ➔ Domain 2: Highly permeable material – All imaging currents must be in domain 2 – The sources and the images must create a field that satisfies the continuity conditions at the interface between domains 1 and 2 – The image of the image must be the original source – The field generated in domain 1 is identical to the source field plus the field from the (iron) magnetization. – The field generated in domain 2 has no physical significance Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Imaging Current Method n Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Imaging Current Method n Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Imaging Current Method Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Imaging Current Method Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The Field of a Line Current (2D) Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Imaging Method Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Ideal Current Distributions Exercise: Do this for the quadrupole and compare the results in terms of bore radius Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Ideal Current Distributions Exercise: Do this for the quadrupole and compare the results in terms of bore radius Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Nested Helices Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Nested Helices Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Coil-‐Block Approximations Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Generation of Multipole Field Errors Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The LHC Magnet Zoo Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Finite-‐Element / Boundary-‐Element Coupling Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 29 29 Corrector Magnets Octupole Sextupole-spool pieces Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 Magnet Extremities Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 The CERN Field Computation Program ROXIE Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015 32 Numerical Field Computation ➔ Principles of numerical field computation – Formulation of the Problem – Weighted residual – Weak form – Discretization – Higher order elements – Numerical example ➔ Weak forms in 3-‐D ➔ Element shape functions – Global shape functions – Barycentric coordinates ➔ Mesh generation Stephan Russenschuck, CERN TE-MSC-MM, 1211 Geneva 23 FII-2015