Computing for Magnet Design
Solving the coupled electromagnetic, mechanical, thermal, and superconductor physics that decide whether a magnet can be built.
Why magnets are a computing problem
A high-field superconducting magnet is a multi-physics object. Current in the winding creates a field; the field acting on the current creates Lorentz forces of enormous magnitude; those forces create stress and strain; strain and temperature change the superconductor's ability to carry current. All four physics couple, and the design closes only when every one of them stays inside limits simultaneously.
What gets computed
- Field maps: the current distribution needed for the target field shape
- Forces and stress: structural analysis of coils, cases, and supports
- Current-sharing margin: how close the conductor is to losing superconductivity
- Cryogenic heat loads: what the cooling system must remove
The high-field regime
REBCO high-temperature superconductor allows fields well beyond older niobium-based conductors, which is what makes a compact, high-field design feasible. The Hyperion breeder targets a peak field of 16.84 T with roughly 8 T on axis; the burner uses far higher local fields, about 26.49 T at the mirror plug and 17 T at the throat. At these fields the stress is severe and the mechanical design is as demanding as the electromagnetic one.
def hoop_stress(B, r, j):
# order-of-magnitude Lorentz hoop load on a winding
return B * j * r # field * current density * radius
# stress scales with the product; high B forces careful structureVerification and margin
Because a magnet failure is expensive and slow to diagnose, models are cross-checked against measured conductor data and small-coil tests, and the design carries margin against the current-sharing temperature. The related discipline of quench analysis (see Quench Analysis) covers what happens if that margin is lost.
All Hyperion and burner magnet work is design and simulation; construction of the breeder begins in the second quarter of 2027.