New testing validates Thea Energy's thesis

Last week, Thea Energy released a preprint detailing a major milestone in the company’s effort to build a simpler and more practical stellarator. The research centers on test results from Canis, their prototype magnet system. 

Traditional stellarators require highly customized, contoured coils to shape the magnetic fields responsible for confinement. Thea’s approach eliminates these complex coils in favor of an array of simple, mass-manufacturable planar coils, each contributing a portion of the total magnetic field. By shifting complexity from mechanical design to precise software-based current control, Thea can focus on developing a robust software control platform and then manufacture relatively inexpensive magnet hardware at scale.

The Canis array consisted of nine planar HTS coils arranged in a 3×3 grid, cooled to cryogenic temperatures (~20 K), and powered individually. In testing, the system produced compelling results, including:

• Field strengths capable of supporting stellarator confinement: fields up to 47.2 millitesla at 25 cm from the coils, with strengths at the coil surfaces calculated at over 3 Tesla​.

• Precise field shaping: Canis was able to reproduce target magnetic field shapes based on simulations from Thea’s planned reactor design, matching predicted field contours within a 1% margin of error.

• Consistent performance under tight operating parameters: Each coil operated with currents up to ±140 amps without quenching, and the array demonstrated stable performance across multiple test runs.

One of the biggest technical hurdles the team addressed was magnetic coupling between the coils. When coils are packed closely together, their magnetic fields naturally interact, making it difficult to adjust one coil without affecting others. To compensate for this, Thea developed a control system that adjusts each coil’s current in real-time.

In testing, the system performed as designed: each coil could be tuned independently while maintaining the desired overall field configuration. This paves the way for closed-loop control in a full-scale device, where plasma diagnostics could drive dynamic adjustments across the magnet array.

Perhaps the most substantial proof of Thea’s thesis wasn’t in the technical performance: it’s that the entire Canis platform—including all nine coils, the cryogenic systems, and the power and control infrastructure—was built in-house in a matter of months. This rapid development cycle underscores a core benefit of Thea’s approach: with identical, modular components, the engineering process becomes far more efficient and cost-effective. 

If Thea can scale from 9 coils to the target 450-coil array with the same speed and efficiency, it could be positioned to deploy a pilot reactor well ahead of other fusion startups. And as a participant in the U.S. Department of Energy’s Milestone-Based Fusion Development Program, hitting development targets on time is also key to securing cost-share funding. As part of this announcement, Thea confirmed that the DOE certified the completion of this test campaign’s performance milestones.

Meeting government-defined milestones also signals concrete progress to private investors. With this demonstration, Thea can make a compelling case for additional investment by showing that key risks, including magnet design, cryogenics, and field control, have been successfully addressed. Thea’s leadership has indicated these results will support raising a Series B funding round in the coming years—progress towards which we’ll continue to follow closely.