Experimental Results from Early Non-Planar NI-HTS Magnet Prototypes for the Columbia Stellarator eXperiment (CSX)

This paper reports on the successful staged development of three ReBCO high-temperature superconducting magnet prototypes (P1–P3) utilizing 3D-printed aluminum frames, specialized winding mechanics, and solder potting to validate manufacturing, thermal management, and quench mitigation strategies for the upcoming non-planar Columbia Stellarator eXperiment (CSX).

The Gist

Imagine trying to build a miniature, self-sustaining sun inside a machine to generate clean energy. This is the goal of nuclear fusion. To keep this super-hot "sun" from melting the machine, scientists use incredibly powerful magnets to hold the plasma (the hot gas) in place.

For a long time, these magnets were like flat, rigid pancakes. But to make the machine smaller and more efficient, scientists want to twist these magnets into complex, 3D shapes (like a twisted pretzel or a Mobius strip). This is called a stellarator.

The problem? The super-material used to make these magnets (called HTS tape) is like a very thin, brittle ribbon of glass. It works great when bent flat, but if you try to twist it into a 3D shape, it can snap or lose its superpowers.

This paper describes how a team at Columbia University built a series of "training wheels" (prototypes) to figure out how to twist this fragile ribbon into the complex shapes needed for their new machine, the Columbia Stellarator eXperiment (CSX).

Here is the story of their journey, broken down into simple steps:

1. The Challenge: The "Brittle Ribbon"

Think of the superconducting tape as a high-tech, super-strong shoelace.

• The Goal: We want to weave this shoelace into a complex, twisted 3D knot to create a magnetic cage.
• The Problem: If you pull the shoelace too tight or bend it the wrong way, it snaps. Also, if it gets too hot, it stops working instantly (this is called a "quench").
• The Solution: The team had to invent new ways to hold the ribbon, bend it gently, and keep it cool.

2. The Three-Step Training Program (P1, P2, P3)

Instead of trying to build the final, perfect machine immediately, they built three smaller versions to learn from their mistakes.

• Prototype 1 (P1): The "Flat Practice"

  • What it was: A simple, flat, oval-shaped coil.
  • The Lesson: They tested if they could 3D-print a frame to hold the ribbon and if they could glue the layers together with special solder.
  • Result: Success! They proved the basic idea worked in a bath of liquid nitrogen (very cold, like -196°C).

• Prototype 2 (P2): The "Twist Test"

  • What it was: A twisted, non-flat shape. This is where the ribbon gets stressed.
  • The Innovation: They built a special robotic arm with a gimbal (like a camera stabilizer). As they wound the ribbon, this arm rotated the coil so the ribbon was always pulled straight into the channel, never forced to bend awkwardly.
  • The Glue: They used a "solder potting" technique. Imagine pouring molten metal over the ribbon layers to glue them together. This acts like a safety net: if one spot gets hot, the current can flow sideways through the glue to cool it down, preventing a disaster.
  • Result: They successfully wound 42 turns and cooled it down to 20 Kelvin (colder than outer space!). It worked, but they noticed the metal wires connecting to the coil got too hot.

• Prototype 3 (P3): The "Final Boss"

  • What it is: The most complex version yet. It has "concave" parts (dents) and is designed to hold even more ribbon (200 turns) to create a stronger magnetic field.
  • The Goal: To prove they can handle the most difficult shapes and generate the target magnetic field of 0.5 Tesla (strong enough to hold the fusion plasma).
  • Status: It has been built and is currently being tested.

3. The "Super-Connector" Problem

To make a full-sized machine, they need to join thousands of feet of this ribbon together.

• The Analogy: Imagine trying to connect two pieces of a very delicate, high-speed fiber-optic cable. If the connection is even slightly loose, the signal (or electricity) gets lost as heat.
• The Fix: They developed a special way to overlap and solder the ribbons. They tested these "joints" and found they are incredibly efficient, losing almost no energy (resistance is lower than a micro-ohm). This means they can build a machine that is miles long without it falling apart electrically.

4. The Results: What Did They Learn?

• It Works: They successfully created twisted magnets that generate the expected magnetic fields.
• Cooling is Key: They figured out how to transfer heat away from the magnets using a "cold head" (a mechanical refrigerator) and special sapphire interfaces, keeping the magnets at a frosty 20 Kelvin.
• Safety First: The "solder potting" worked as a safety valve. When they pushed the current too high, the magnet didn't explode; it just gently warmed up and stopped, proving their safety design works.
• The Hiccup: The metal wires connecting the magnet got too hot and melted a small part of the ribbon. This is a "bug" they are fixing by improving how they clamp the wires together.

Why Does This Matter?

This paper is a roadmap. It proves that we can take fragile, high-tech materials and twist them into the complex shapes needed for fusion energy.

Think of it like learning to ride a bike:

1. P1 was learning to balance on a tricycle.
2. P2 was learning to ride a regular bike on a flat road.
3. P3 is learning to ride up a steep, winding hill.

Once they master P3, they will be ready to build the full-sized Columbia Stellarator, a machine that could one day provide limitless, clean energy for our cities. They have successfully "de-risked" the hardest parts of the engineering, paving the way for the future of fusion power.

Technical Summary

1. Problem Statement

The Columbia Stellarator eXperiment (CSX) aims to upgrade the existing Columbia Non-neutral Torus (CNT) into a university-scale, quasi-axisymmetric stellarator using High-Temperature Superconductor (HTS) technology. The target is an on-axis magnetic field of 0.5 T and a peak on-coil field of 5.3 T.

Key Challenges:

• Strain Sensitivity: ReBCO (Rare-earth Barium Copper Oxide) tapes are highly sensitive to strain, particularly torsion and "hard-way" bending. This makes winding them into the complex, non-planar geometries required for stellarators difficult.
• Manufacturing Complexity: Traditional methods for non-planar magnets (e.g., twisted CICC) are often too bulky for university-scale devices.
• Quench Management: HTS magnets are prone to rapid heating and damage if they transition out of the superconducting state (quench) due to local defects or exceeding critical current.
• Joint Resistance: Final magnets require kilometers of tape, necessitating low-resistance lap joints that have not been fully optimized for these specific non-planar, high-strain conditions.

2. Methodology

The authors employed a staged prototype program (P1, P2, P3) to de-risk manufacturing, winding, structural joining, cooling, and energizing strategies.

Key Technologies & Design Features:

• No-Insulation No-Twist (NINT) Architecture: HTS tape is wound directly into 3D-printed aluminum (AlSi10Mg) channels without insulation between turns, followed by solder potting.
• 3D-Printed Modular Frames: Due to printer size limits, frames are printed in sections and joined via dovetail joints (P1/P2) or bolted interfaces with indium for thermal contact (P3).
• Gimballed Winding Mechanism: A specialized rig allows the coil to rotate and slide, ensuring the tape remains perpendicular to the channel during winding to minimize torsional strain.
• Passive Quench Mitigation: The use of solder potting (ChipQuik® Sn63Pb37) allows current to redistribute radially through the tape stack and into the bobbin during a hot-spot event, preventing catastrophic damage.
• Parallel-Wound Non-Insulated (PWNI) Architecture (P3): Used in the final prototype to improve current sharing between tapes in the same channel.
• Cryogenic Test-stand: A modular 20 K system using a Sumitomo 408S cold head. It features conductive cooling via sapphire interfaces, multi-layer insulation (MLI), and an actively cooled aluminum heat shield. Diagnostics include silicon diodes, voltage taps, and a Hall probe.

3. Key Contributions

• Demonstration of NINT Non-Planar Winding: Successfully validated a winding strategy that maintains constant tension and minimizes strain on ReBCO tape in complex 3D geometries.
• Modular Additive Manufacturing: Proved that 3D-printed aluminum frames with mechanical (dovetail/bolted) joining can achieve the geometric accuracy required for quasi-symmetry while withstanding thermal cycling.
• Low-Resistance Joint Development: Developed and characterized HTS lap joints using SnPb solder, achieving sub-micro-ohm resistance levels essential for multi-kilometer tape lengths.
• Thermal Management Strategy: Established a conductive cooling path from 50 K to 20 K using sapphire interfaces, successfully managing thermal gradients in a non-planar coil.

4. Experimental Results

Prototype P1 (Planar Elliptical):

• Configuration: 22 turns, double-pancake.
• Test: Immersed in liquid nitrogen (77 K).
• Result: Achieved predicted magnetic fields (0.55 mT at 10 A) and total resistance of 30.1 µΩ. Validated additive manufacturing and baseline winding.

Prototype P2 (Non-Planar, High Strain):

• Configuration: 42 turns (limited by high-torsion regions), double-pancake.
• Test: Cryogenic test-stand at 30–42 K.
• Results:
  • Resistance: Measured $R = (1.67 \pm 0.02) \, \mu\Omega$ (linear up to 70 A).
  • Field: Produced 4.5 mT at 110 A (4.5 kAt), matching Biot-Savart predictions.
  • Thermal: Temperatures reached 32–48 K. Quench occurred near 120 A due to resistive heating at the copper-HTS interface and soldered joints.
  • Time Constant: L/R time constant measured as $\tau \approx 160$ s (consistent between voltage decay and field decay).
  • Inductance: Measured 1.04 mH (vs. predicted 1.15 mH).

Prototype P3 (Concave, High-Field):

• Configuration: 200 turns, dual double-pancakes, concave geometry, PWNI architecture.
• Status: Commissioned at 20 K; high-field characterization is ongoing. Designed to test the viability of winding concave sections and managing high-torsion zones.

Joint Testing:

• 30 mm lap joints tested at 77 K yielded a resistance of 380 nΩ, confirming the feasibility of low-resistance connections for the final multi-kilometer coils.

5. Significance

This work represents a critical step toward realizing compact, university-scale stellarators using HTS technology.

• De-risking: The staged approach successfully isolated and solved key engineering challenges: strain management during winding, structural integrity of 3D-printed frames, and passive quench mitigation.
• Scalability: The results validate the NINT approach for non-planar geometries, offering a path to higher current densities and smaller bend radii than twisted cable approaches.
• Future Outlook: The successful testing of P2 and the commissioning of P3 pave the way for the full-scale CSX coils. Future work will focus on optimizing the copper-HTS thermal interface, expanding diagnostics (embedded sensors), and testing P3 at the target 0.5 T field.

In summary, the paper demonstrates that non-planar HTS stellarator magnets are manufacturable and operable using 3D-printed frames, gimballed winding, and solder potting, providing a robust blueprint for next-generation fusion devices.