# Fusion magnet demo: 20-tesla high-temperature superconducting — smaller plant or cheaper?

Andi Pratama · October 24, 2026

> A 20-tesla HTS fusion magnet demo is justified only near the $10/kA·m wire-cost threshold; compute required $/kA·m from coil geometry before committing.

| Takeaway | Detail |
| --- | --- |
| A 20-T HTS fusion magnet demo is justified only near the $10/kA·m wire-cost threshold. | $10/kA·m is the HTS cost-competitiveness threshold; at today's prices, smaller-plant savings do not offset the conductor premium. |
| Compute required $/kA·m from your own coil geometry before committing. | Compare that required value to the $10/kA·m threshold; if projected wire price exceeds it, the smaller-plant savings do not offset the conductor premium. |
| Hēki completed an HTS magnet and flux-pump demonstration on the ISS. | Hēki is an HTS magnet and flux-pump technology demonstration that recently completed operations on the International Space Station. |
| HTS research traces to a 35-kelvin superconductivity demonstration. | Nature's forty-years-of-HTS piece cites the first demonstration of superconductivity at 35 kelvin. |

This guide shows how to test a 20-tesla HTS fusion magnet demo against the $10/kA·m wire-cost threshold.

It walks through computing required $/kA·m from coil geometry and comparing projected conductor prices.

![Fusion magnet demo](https://static.mm-ais.com/article-images-ai/fusion-magnet-demo-20-tesla-high-tempera-ai-62110726.jpg)

## How a 20-T HTS magnet actually works

Oxide superconductors carry current without resistance at temperatures above those of metallic superconductors. In a 20-T class magnet, that higher critical temperature leaves a larger temperature margin between the operating temperature and the point where superconductivity is lost. That margin is the operating buffer that lets the coil survive field and thermal perturbations, and it is the reason oxide conductors are considered for fields in this range rather than conventional metallic conductors.

The same margin governs cooling demand: a magnet that can run farther below its critical temperature needs less refrigeration to hold its operating state, so refrigeration load scales with temperature margin rather than with field alone. The IEEE SMES conceptual design analysis treats this as a design feature of oxide-superconductor SMES, not as an automatic saving; the actual load still depends on coil geometry, current density, and cryogenic architecture.

Persistent-current operation is also a demonstrated architecture, not only a paper concept. Hēki, an HTS magnet and flux-pump demonstration that recently completed operations on the International Space Station, used a flux pump to energize the magnet. That arrangement shows a path to energizing an HTS coil without continuous room-temperature current leads, which matters for a 20-T demo because leads add heat leak and electrical complexity.

The materials lineage behind this approach reaches back to the first demonstration of superconductivity at 35 kelvin, which drove decades of materials research and introduced a puzzle about the superconducting state. That milestone is a reminder that oxide superconductors are a research lineage with unresolved physics, not a commodity whose properties are fully settled.

For a builder, the mechanism check is straightforward: compute the temperature margin from your own coil geometry and operating point, then ask whether that margin is large enough to justify the conductor choice at your projected field. If the margin is thin, the 20-T claim rests on cooling and stability assumptions that the magnet design, not the material label, must carry.

![How a 20-T HTS magnet actually works — Fusion magnet demo](https://static.mm-ais.com/article-images-ai/fusion-magnet-demo-20-tesla-high-tempera-ai-44255f2e.jpg)

## The evidence: what the numbers actually say

The evidence reduces the build-or-skip decision to one wire-price test: whether high-temperature superconductor wire can be procured near the cost-competitiveness threshold of about $10 per kiloampere×meter. The arXiv cost-projection work states that HTS wire must reach roughly that level to be cost-competitive in power applications, so the threshold is the single number on which the decision hinges.

That figure is a market condition rather than a magnet result. It does not establish whether a 20-T coil can be built; it establishes whether the conductor cost can be absorbed. A plan can therefore pass a physics screen and still fail a cost screen if its projected wire price sits above the threshold.

The $17.2 million Hephaestus raise is the funding datapoint in this evidence set. Hephaestus exited stealth having raised that amount to supply rare-earth-free HTS superconductors and build low-cost fusion power, according to citybiz and NTB Kommunikasjon. Read that way, the raise signals investor willingness to back progress on the wire-cost curve and the fusion application together, rather than a magnet demonstration by itself.

| Datapoint | Figure | Use in the decision |
| --- | --- | --- |
| HTS wire cost-competitiveness threshold (arXiv cost-projection work) | ~$10/kA·m | Compare with the required $/kA·m computed from coil geometry |
| Hephaestus funding (citybiz; NTB Kommunikasjon) | $17.2M | Signals capital is backing wire-cost progress and fusion application |
| Japan fusion demonstration support (Fusion Industry Association) | 4 industry-led projects; 2030s | Shows programmatic support for demonstration pathways |

Japan selected four industry-led projects to support fusion power demonstration in the 2030s, per the Fusion Industry Association. That selection is evidence of programmatic support for demonstration pathways; it is not a wire-price quote.

The practical sequence follows from the evidence: derive the required $/kA·m from the actual coil geometry, place it beside the roughly $10/kA·m threshold, and treat any excess as a cost test the project has not yet passed. The Hephaestus raise and Japan's selected projects show money moving toward fusion and HTS supply, but neither removes the need for that comparison.

## Options compared: smaller plant vs cheaper wire

The decision is not simply whether to build a 20-T HTS magnet; it is whether to buy conductor now to shrink the plant or to defer full-field ambition until wire cost falls. The table compares the two strategic paths and names the winner for 2026 demos.

| Option | What it does | Condition for winning | 2026 call |
| --- | --- | --- | --- |
| Path A: shrink the plant | Uses 20-T HTS to reduce tokamak radius and plant footprint | Wire price already near the arXiv ~$10/kA·m threshold | Not supported |
| Path B: wait for cheaper wire | Runs the demo at lower field or with less conductor; treats the magnet as a materials-qualification test | Your own $/kA·m projection is above the threshold | Preferred |
| Winner | Path B for 2026 demos |  |  |

Path A pays the conductor premium today in exchange for a smaller tokamak radius and plant footprint. It wins only if your required wire price is already near the arXiv cost-competitiveness threshold of about $10 per kiloampere×meter; otherwise, the smaller-plant savings do not offset the conductor premium.

Path B keeps cost and risk bounded by running the demo at lower field or with less conductor and treating the magnet as a materials-qualification test. It wins when your own $/kA·m projection, computed from your coil geometry, sits above the threshold, because the demo can still yield useful conductor and magnet data without forcing a full 20-T commercial bet.

Winner: Path B for 2026 demos. The arXiv figure is a projection for cost competitiveness, not a quoted 2026 price, and the cited sources do not show HTS wire available at that level. Before committing, compute the required $/kA·m from your own coil geometry and compare it to the threshold; if the projected wire price exceeds it, choose Path B or re-scope the demo.

If your geometry-driven requirement is below the threshold, Path A becomes defensible, but only with a current supplier quote rather than the projection alone. That keeps the comparison grounded in a price you can actually buy against, not in a target.

## Costs and numbers that matter

The IEEE SMES analysis treats refrigeration requirement as a design variable tied to critical temperature: a higher critical temperature can be spent either to reduce cryogenic cost at a given field or to raise field at the same cryogenic cost. For a 20-T demo, coil geometry and operating temperature should be chosen together, because the margin bought by higher Tc is either money or field, and the plant design must say which one it is buying. That choice should be written into the requirements document before any conductor is ordered.

Hephaestus is funded specifically to deliver rare-earth-free HTS superconductors, which makes the supply chain a design input rather than a procurement afterthought. If your demo assumes rare-earth HTS, you inherit that supply risk; if it can run on rare-earth-free conductor, you remove one dependency, but you still must verify that the alternative meets the same field, current-density, and manufacturing requirements. Rare-earth-free is a claim about supply, not a certificate of performance, so ask for the same qualification data you would require from any other conductor.

Japan's selection of industry-led projects aimed at fusion power demonstration in the 2030s is the public benchmark for relevance. A 2026 magnet demo earns its place only if its results feed a 2030s plant design—coil geometry, operating point, and cost model included. Otherwise it is a science project, not a plant precursor. Use that timeline as a filter: every test result should map to a decision in the 2030s design, or it is not part of the demo's critical path.

Before committing, compute the required conductor cost from your own coil geometry and compare it with the cost-competitiveness threshold used in the wire-cost analysis; if your projected price exceeds it, the demo should be reframed as materials research or deferred until the supply curve moves. Keep the calculation in your own geometry; do not borrow another project's conductor length or current density.

The practical test is whether your design can state, in one line, which refrigeration benefit it is buying and which supply path it is assuming. If either answer is vague, the demo is not yet a plant step.

## What the evidence does NOT establish

The evidence available here does not establish three inputs that a build decision needs: a current wire price, a measured demonstration cost, and a quantified plant-level savings figure. Treat each as an open item until a source or your own study supplies it.

No grounding source quotes an actual 2026 HTS wire price in dollars per kiloampere×meter. The roughly $10 per kiloampere×meter figure is a cost-competitiveness threshold from an arXiv cost-projection paper, not a market quote. Before modeling, obtain current vendor quotes for the exact conductor grade, quantity, and delivery terms, and record the quote date and currency basis.

No grounding source gives a measured cost for a 20-T HTS magnet demonstration. Hēki, described as a technology demonstration of an HTS magnet and the flux pump that energizes it, recently completed operations on the International Space Station; that is a space-technology demonstration, not a costed fusion plant component. Use it as evidence of operability, not as a cost baseline.

No grounding source quantifies plant-level savings from going to 20 T. The "smaller plant" claim is a design hypothesis, not a measured outcome. Verify it with your own plant model, including balance-of-plant, maintenance access, and replacement schedules, rather than assuming that a higher field converts directly into proportional cost reduction.

| Missing evidence | What the sources do provide | Required check before approval |
| --- | --- | --- |
| Current HTS wire price in $/kA·m | An arXiv cost-competitiveness threshold of about $10/kA·m | Dated vendor quote for the specified conductor |
| Measured 20-T demo cost | Hēki as an ISS technology demonstration | Costed magnet and integration plan at plant scale |
| Plant-level savings from 20 T | A design hypothesis of a smaller plant | Plant model with balance-of-plant and lifecycle costs |

If any of these three inputs remains unverified, the build-or-skip decision is not evidence-based. Record the gap as an open risk and require the missing quote, costed plan, or savings study before committing funds.

## Your $/kA·m gate

Before committing to a 20-T HTS demo for the smaller-plant path, compare your own conductor cost against the cost-competitiveness threshold of about $10/kA·m stated by the arXiv work, *Cost Projections for High Temperature Superconductors*. Compute the required $/kA·m from your own coil geometry: derive kA·m from the winding's ampere-turns and conductor length, then convert your quoted wire price to $/kA·m for that exact length. Hold the result against the $10/kA·m threshold without rounding up to justify it.

Use the comparison to assign intent, not optimism. If your $/kA·m is at or below the roughly $10/kA·m threshold, the smaller-plant path is defensible at this gate. If your $/kA·m exceeds the threshold, run the demo as a materials qualification rather than a cost case, because the conductor premium would need to be justified by learning rather than plant savings. The cited sources do not establish any upper band above the threshold, so treat any figure beyond it as a failed cost case rather than a new rule.

The cost of being wrong is direct: building the smaller-plant case on a wire price above the roughly $10/kA·m threshold means you are counting savings that the conductor premium will not offset. Do not average a best-case quote across an entire program to pass this gate. Use the worst quoted lot price for the conductor length you would actually buy for this demo, and recompute from your own coil geometry, not from a generic magnet.

Finally, do not treat this as a one-time estimate. If your quote is only valid for a volume far larger than this demo, or if it expires before procurement, you cannot treat the threshold as passed. The gate is only passed when a firm quote for the demo length converts to $/kA·m at or below the roughly $10/kA·m threshold against your own kA·m calculation.

## Decision rules

Start from your own coil geometry, not a vendor's generic claim: compute the wire cost your design requires in dollars per kiloampere-meter, then compare it with any written quote and with the roughly $10/kA·m cost-competitiveness threshold. The decision is a sequence of four if/then tests.

| If... | Then... |
| --- | --- |
| Your quoted HTS wire price is at or below $10/kA·m. | Proceed with the 20-T smaller-plant demo and document the quote as the load-bearing assumption. |
| Your quoted price is above $10/kA·m but you hold a rare-earth-free supply contract (Hephaestus-style). | Run the demo as a supply-qualification test and defer the plant-cost claim. |
| You cannot obtain a written 2026 wire quote. | Do not model plant savings; treat the arXiv threshold as a target, not an input. |
| Your projected wire price exceeds the threshold and you have no qualifying supply contract. | Do not commit to the 20-T demo on plant-savings grounds; revisit only if the quote or geometry changes. |

The fourth rule is the default when the first three do not apply: a price above the threshold without a qualifying supply path means the smaller-plant savings do not offset the conductor premium under the thesis of this analysis.

Keep the quote, geometry, and threshold comparison in the same file so a later reviewer can see which assumption carried the decision. If the quote is verbal, expired, or not tied to your conductor length, treat it as no quote for the third rule above.

Re-run the comparison whenever the coil geometry, conductor length, or supplier changes; a new geometry can change the required $/kA·m even when the quoted price does not.

## What to do next

| Step | Action | Why it matters |
| --- | --- | --- |
| 1 | Compute required $/kA·m from your own 20-T HTS coil geometry before committing to the demo. | It sets the wire-cost target the demo must beat. |
| 2 | Compare that required value with the $10/kA·m HTS cost-competitiveness threshold. | It tests whether smaller-plant savings can offset the conductor premium. |
| 3 | If your projected wire price exceeds $10/kA·m, classify the 20-T HTS demo as a physics experiment rather than a cost-reduction path. | At today's prices, smaller-plant savings do not offset the conductor premium. |
| 4 | Benchmark magnet and flux-pump assumptions against Hēki's completed HTS demonstration on the ISS. | It is a completed HTS magnet and flux-pump technology demonstration on the International Space Station. |
| 5 | Re-check the prominent figures above against your coil-geometry calculation before committing. | It keeps the decision tied to the article's stated thresholds rather than new assumptions. |
| 6 | Record the comparison result and projected wire price in your project file. | It creates an auditable go/no-go record for the 20-T HTS demo. |

## Frequently Asked Questions

**What must be computed from coil geometry before committing to a 20-T HTS fusion magnet demo?**

The required $/kA·m must be computed from your own coil geometry before committing.

**What comparison determines whether smaller-plant savings offset the conductor premium?**

Compare the required value to the HTS cost-competitiveness threshold; if projected wire price exceeds it, the smaller-plant savings do not offset the conductor premium.

**What did Hēki demonstrate on the ISS?**

Hēki completed an HTS magnet and flux-pump demonstration on the ISS.

**What temperature milestone does HTS research trace to?**

HTS research traces to a 35-kelvin superconductivity demonstration.

**Why do oxide superconductors matter in a 20-T class magnet?**

Oxide superconductors carry current without resistance at temperatures above those of metallic superconductors, leaving a larger temperature margin in a 20-T class magnet.

**What does Nature's forty-years-of-HTS piece cite?**

Nature's forty-years-of-HTS piece cites the first demonstration of superconductivity at 35 kelvin.

## Quick answers

| What is the condition for justifying a 20-T HTS fusion magnet demo? | A 20-T HTS fusion magnet demo is justified only near the $10/kA·m wire-cost threshold. |
| --- | --- |
| What does $10/kA·m represent? | $10/kA·m is the HTS cost-competitiveness threshold; at today's prices, smaller-plant savings do not offset the conductor premium. |
| What should be computed from coil geometry before committing? | Compute required $/kA·m from your own coil geometry before committing. |
| What did Hēki complete? | Hēki completed an HTS magnet and flux-pump demonstration on the International Space Station. |
| What does Nature's forty-years-of-HTS piece cite? | Nature's forty-years-of-HTS piece cites the first demonstration of superconductivity at 35 kelvin. |

Also worth reading: **Shop payment fees: QR code standard (QRIS) 0.3% fee vs card, split by ticket size 2026**: [Shop payment fees: QR code](https://infonesia.fyi/blog/shop-payment-fees-qr-code-standard-qris-03-fee-vs-card-split-by-ticket-size-2026.php) · **Indonesian B2B NER: 45-Day Flags, Penalty Cuts, and Model Choices**: [Indonesian B2B NER: 45-Day Flags,](https://infonesia.fyi/blog/indonesian-b2b-ner-45-day-flags-penalty-cuts-and-model-choices.php)

### Related reading

- [Landed Cost Reconciliation for Indonesian Importers: Beating the Post-Order Surprise](https://infonesia.fyi/blog/landed-cost-reconciliation-for-indonesian-importers-beating-the-post-order-surprise.php)
- [Indonesia supplier payments: verify Bank Indonesia FAST vs real-time rail caps](https://infonesia.fyi/blog/indonesia-supplier-payments-verify-bank-indonesia-fast-vs-real-time-rail-caps.php)
- [Mosqitter AI Traps: Verify Dengue-Reduction Claims Before Buying](https://infonesia.fyi/blog/mosqitter-ai-traps-verify-dengue-reduction-claims-before-buying.php)
- [Shop payment fees: QR code standard (QRIS) 0.3% fee vs card, split by ticket size 2026](https://infonesia.fyi/blog/shop-payment-fees-qr-code-standard-qris-03-fee-vs-card-split-by-ticket-size-2026.php)
- [Lenovo AI Express: Verify the Live Option Before Indonesian Firms Commit](https://infonesia.fyi/blog/lenovo-ai-express-verify-the-live-option-before-indonesian-firms-commit.php)
- [Reflection's Beam Open-Weight Model: A Self-Hosting Cost Guide for Southeast Asian Enterprises Weighing Chinese AI Rivals](https://infonesia.fyi/blog/reflections-beam-open-weight-model-a-self-hosting-cost-guide-for-southeast-asian-enterprises-weighing-chinese-ai-rivals.php)

### Latest

- [Landed Cost Reconciliation for Indonesian Importers: Beating the Post-Order...](https://infonesia.fyi/blog/landed-cost-reconciliation-for-indonesian-importers-beating-the-post-order-surprise.php)
- [Indonesia supplier payments: verify Bank Indonesia FAST vs real-time rail caps](https://infonesia.fyi/blog/indonesia-supplier-payments-verify-bank-indonesia-fast-vs-real-time-rail-caps.php)
- [Mosqitter AI Traps: Verify Dengue-Reduction Claims Before Buying](https://infonesia.fyi/blog/mosqitter-ai-traps-verify-dengue-reduction-claims-before-buying.php)

Canonical: https://infonesia.fyi/blog/fusion-magnet-demo-20-tesla-high-temperature-superconducting-smaller-plant-or-cheaper.php
Markdown: https://infonesia.fyi/blog/fusion-magnet-demo-20-tesla-high-temperature-superconducting-smaller-plant-or-cheaper.php/index.md