Fusion: CFS orders 10,000 km of superconducting tape before proving its reactor works
Commonwealth Fusion Systems signed the largest superconducting tape order ever announced with Japan's Fujikura, aimed at its first commercial plant. But the machine that must prove net energy gain has not been switched on yet.
October 04, 2026 · by Equipe Entendendo o Futuro · ⟳ Automatic update
On September 30, 2026, US-based Commonwealth Fusion Systems (CFS) announced that Japan's Fujikura will supply more than 10,000 km of high-temperature superconducting (HTS) tape — according to the company, the largest order of its kind. The material is for ARC, CFS's first grid-connected fusion plant, planned for Chesterfield County, Virginia. It is a big industrial step, taken before the decisive scientific step has happened.
What it is / how it works
Nuclear fusion powers the Sun: light hydrogen nuclei merge and release a lot of energy. To do it on Earth, you must heat a gas (plasma) to over 100 million degrees and keep it away from the reactor walls. In a tokamak — a doughnut-shaped chamber — giant magnets hold the plasma in place.
CFS's bet is on those magnets. Instead of traditional superconductors, it uses a thin tape of "high-temperature" superconducting material (working at about -253 °C rather than near absolute zero). The tape allows much stronger magnetic fields in smaller magnets. A stronger field means a much smaller — and, in theory, cheaper — reactor than giant projects such as ITER in France. Each magnet is wound from kilometres of this tape, which is why a power plant needs so much.
Why it matters
A clean, carbon-free source with no long-lived radioactive waste, available 24 hours a day, would transform the power sector. The order shows CFS preparing to build at scale, not just experiment. It also helps build a supply chain: making HTS tape in large volumes is still a bottleneck for the whole industry.
In CEO Bob Mumgaard's words, the partnership shows the company is "actively placing orders and getting ready to execute our ARC power plant once our SPARC fusion machine achieves Q>1."
Evidence
- Record magnet (2021): a CFS/MIT prototype reached 20 tesla with high-temperature superconductor, a record for this type of magnet.
- SPARC under assembly: the demonstration reactor in Devens, Massachusetts, received the first of its 18 toroidal field magnets in early 2026. The target is first plasma in 2027.
- Customer and capital: in 2025 Google agreed to buy 200 MW from the future ARC plant. CFS says it has raised about $4 billion since 2018.
What could still go wrong
This is the key point: nobody has yet produced electricity from fusion. The "Q>1" the CEO mentions means the plasma producing more energy than is used to heat it — and SPARC still has to be switched on and show that. Even Q>1 in the plasma is not the same as a plant that delivers more energy to the grid than it consumes overall.
Other risks: assembly delays (fusion projects have historically slipped), material wear under neutron bombardment, tritium fuel production, and the final cost of electricity. The tape order is an industrial commitment, not physical proof. CFS itself talks about commercial plants operating "in the 2030s and beyond."
What to watch
- 2027: SPARC first plasma and, soon after, the Q>1 attempt.
- Pace of installation of the 18 toroidal magnets in Devens.
- Permitting and construction of the ARC plant in Virginia.
- Whether Fujikura and other suppliers can deliver HTS tape in volume and on time.
Honest summary: CFS has one of the strongest track records in private fusion, with tested magnets and a major customer. But the proof that matters — net energy — is still ahead.