Thea Energy has announced a collaboration with MIT’s Plasma Science and Fusion Center (PSFC) to test a planar winding coil at full magnetic field and full strength. The coil is intended for Eos, the company’s large-scale integrated stellarator, being developed by the US firm based in Kearny, New Jersey.
The tests will validate the coil under real operating conditions and verify its durability over time, a key factor in ensuring the reliability and high availability of a future power plant. According to reports, the Eos design uses 12 winding coils and around 300 programmable shaping magnets. Thea Energy says the shaping coils have already been built and validated at full scale. The winding coil is the next component to be put to the test.
The background is the history of the stellarator. Earlier generations relied on complex three-dimensional coils, which are expensive to produce because of the extreme tolerances required. Thea Energy has chosen a different geometry: planar coils that are simpler to manufacture, install and maintain, with the ability to control individual magnets through software. In March 2025 the company presented results from Canis, a 3×3 array of planar coils made with high-temperature superconducting material. That test, conducted at 20 kelvin, showed field control to within 1% of the predicted value.
Zachary Hartwig, an MIT professor and head of the PSFC’s engineering group, said the center’s test infrastructure is well suited to evaluating Thea Energy’s coils. The PSFC has already worked with Commonwealth Fusion Systems on large-scale magnets for the SPARC tokamak, and so brings direct experience in this type of test campaign.
The same planar-coil architecture will be used in Helios, the fusion power plant that follows Eos, for both winding and shaping coils. A single magnet technology across two machines reduces manufacturing complexity and makes industrial-scale deployment more realistic. Thea Energy was founded in 2022 as a spin-out of Princeton University and the Princeton Plasma Physics Laboratory. In 2024 it was selected as one of the eight initial awardees of the US Department of Energy’s Milestone-Based Fusion Development Program. It has also launched a collaboration with NVIDIA, Synopsys, Argonne and Princeton to build a digital twin of Helios.
If the coil passes the full-field test, the roadmap to commercial fusion by 2035 loses one of its most uncertain steps. For Europe, where the Wendelstein 7-X stellarator in Germany remains the experimental benchmark, an alternative based on magnets that are easier to produce would be a development worth watching closely.
