OpenStar Unveils Levitating Dipole Reactor With Swappable Core

The First New Nuclear Fusion Idea for Decades — Sabine Hossenfelder

Sabine Hossenfelder

The startup bypasses fusion's crippling neutron-damage bottleneck by turning its central magnet into an expendable, robotically replaced module rather than relying on unproven physics.

AI-assisted summary of The First New Nuclear Fusion Idea for Decades — original source. Check the original for full context.

Source published:

Key ideas

  • New Zealand fusion startup OpenStar Technologies published a design for "TMANUI," a levitating dipole commercial reactor projected to generate 208 megawatts of net electric power—roughly matching the capacity of a small natural gas plant. (1:34)
  • Unlike conventional tokamaks and stellarators that surround plasma with external coils, the dipole architecture suspends a single torus-shaped superconducting magnet inside the reaction chamber, naturally suppressing plasma instabilities and reducing hardware costs. (1:34)
  • Because floating a magnet inside the chamber exposes it to destructive, uncharged fast neutrons from deuterium-tritium fusion, the levitating dipole concept had seen minimal progress since the 1980s outside of an MIT test device (2004–2011) and modest work in Japan. (3:17)
  • OpenStar addresses neutron degradation through modular maintenance rather than new physics: the yttrium barium copper oxide magnet is encased in a sacrificial tungsten and boron carbide shield that robots lower out of the reactor annually for a two-week refurbishment, lasting about ten cycles before full replacement. (3:17)

Detailed notes

## Background Published on YouTube by physicist Sabine Hossenfelder on August 23, 2026, this video examines a newly published paper from New Zealand startup OpenStar Technologies. The description highlights their design for a "levitating dipole" fusion reactor—an approach long written off as fundamentally flawed—calling it the first truly new nuclear fusion design idea in decades.

## TL;DR The video breaks down OpenStar Technologies' proposed commercial fusion reactor design, dubbed TMANUI, which revives the historical levitating dipole concept to generate an estimated 208 megawatts of electricity. Instead of housing magnets outside a torus like standard tokamaks, TMANUI places a floating superconducting magnet inside the plasma chamber, stabilizing the plasma naturally with fewer external coils. While high-energy neutrons from deuterium-tritium reactions inevitably bombard this central magnet, OpenStar tackles the issue through pragmatic engineering rather than exotic physics: modular magnet sections protected by tungsten and boron carbide that are robotically replaced during two-week maintenance downtimes every year. Sabine praises the realistic focus on lifecycle maintenance and component replacement, noting it addresses real-world operating realities often ignored by competing fusion projects.

## Atmosphere & style The presentation carries Sabine's signature blend of sharp technical analysis, dry humor, and grounded skepticism toward venture-backed tech claims. Her delivery is measured and direct, cutting through speculative energy hype to focus on nuts-and-bolts engineering challenges like neutron embrittlement and maintenance logistics. Punctured by sardonic jokes—such as comparing the modular system to a radioactive Ship of Theseus—the video feels refreshingly pragmatic and intellectually sober in a field known for overpromising.

## Bulletin - [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) New Zealand fusion startup OpenStar Technologies published a design for "TMANUI," a levitating dipole commercial reactor projected to generate 208 megawatts of net electric power—roughly matching the capacity of a small natural gas plant. - [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) Unlike conventional tokamaks and stellarators that surround plasma with external coils, the dipole architecture suspends a single torus-shaped superconducting magnet inside the reaction chamber, naturally suppressing plasma instabilities and reducing hardware costs. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) Because floating a magnet inside the chamber exposes it to destructive, uncharged fast neutrons from deuterium-tritium fusion, the levitating dipole concept had seen minimal progress since the 1980s outside of an MIT test device (2004–2011) and modest work in Japan. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) OpenStar addresses neutron degradation through modular maintenance rather than new physics: the yttrium barium copper oxide magnet is encased in a sacrificial tungsten and boron carbide shield that robots lower out of the reactor annually for a two-week refurbishment, lasting about ten cycles before full replacement.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=0s) New Zealand startup OpenStar Technologies is reviving the "levitating dipole" fusion reactor concept, an approach mainstream consensus had long written off as fundamentally flawed. - [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) OpenStar's proposed commercial reactor design, designated "TMANUI" (from the Māori word for sun), is projected to generate 208 megawatts of net electric power, roughly matching a small natural gas plant. - [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) Conventional magnetic confinement designs like tokamaks and stellarators position powerful magnetic coils outside a hollow, donut-shaped vacuum vessel to keep scorching plasma away from the containment walls. - [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) The levitating dipole inverts conventional reactor geometry by eliminating the vessel's central hole and floating a large, torus-shaped superconducting magnet unsupported inside the chamber to trap, compress, and heat the surrounding plasma. - [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) Placing the magnet inside the chamber provides two major advantages: the dipole magnetic field geometry naturally suppresses certain disruptive plasma instabilities, and the system requires fewer external magnets, making it substantially cheaper to build. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) Alongside the engineering challenge of magnetically levitating a massive, supercooled magnet, placing hardware inside the reaction chamber exposes it directly to high-energy fast neutrons produced by deuterium-tritium (D-T) fusion. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) Because fast neutrons carry no electric charge, magnetic and electric fields cannot deflect or contain them, allowing them to blast directly into the internal magnet structure and rapidly degrade it. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) While aneutronic fuels like boron avoid heavy neutron production, they require drastically higher ignition temperatures, compounding the difficulty of achieving fusion. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) Due to these physical roadblocks, the levitating dipole concept saw minimal progress after its 1980s inception, aside from an MIT test device operated between 2004 and 2011 and modest experimental work in Japan. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) OpenStar's innovation relies on pragmatic maintenance management rather than unproven physics, utilizing standard high-temperature superconductors (yttrium barium copper oxide class) wrapped in a heavy shield of heat-resistant, neutron-absorbing tungsten and boron carbide. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) Rather than building the magnet as a monolithic block, OpenStar splits the architecture into an inner main section and an expendable outer section that absorbs the brunt of the neutron damage. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) After roughly one year of continuous operation, robotic equipment lowers the magnet assembly out the bottom of the reactor into a radiation-safe bay to replace the sacrificial outer components over a two-week turnaround, repeating the cycle about ten times before retiring the core. - [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) While OpenStar has not solved every challenge of nuclear fusion, its roadmap highlights a critical reality for all fusion ventures: routine maintenance and component replacement will inevitably make up a massive fraction of long-term operating costs.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=0s) Re-evaluating the Levitating Dipole and Sponsor Aside

New Zealand-based startup OpenStar Technologies has brought forward a fresh perspective on nuclear fusion technology, reviving a concept called the "levitating dipole." For decades, mainstream consensus regarded this approach as fundamentally flawed, yet the company aims to turn that very drawback into an operational feature. Before exploring the engineering behind this concept, Sabine showcases a new gadget: the Plaud Note Pro. She describes it as an external memory system for recording conversations so you do not have to endure them twice. It can record continuously for up to 50 hours, transcribe speech with speaker labels, generate various AI-assisted summaries, and keep audio archives searchable without needing multiple separate apps or draining a smartphone battery.

```mermaid flowchart TD accTitle: Conventional Tokamak versus Levitating Dipole accDescr: Diagram contrasting external magnet placement in a tokamak with internal magnet placement in a dipole reactor.

subgraph Tokamak["Conventional Tokamak / Stellarator"] T1["External Toroidal Magnets"] T2["Hollow Torus Vacuum Vessel"] T3["Plasma Confined at Center"] T1 --> T3 T2 --> T3 end

subgraph Dipole["OpenStar Levitating Dipole Concept"] D1["Vessel without Central Hole"] D2["Superconducting Magnet Torus Levitating Inside"] D3["Surrounding Dipole Field Confines Plasma"] D1 --> D2 D2 --> D3 end ```

---

[01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) OpenStar's TMANUI Concept and Dipole Mechanics

OpenStar Technologies published a paper outlining a commercial fusion reactor design designated "TMANUI," a name derived from the Māori word for sun. Their projections suggest the facility could generate 208 megawatts of net electric power, roughly matching the capacity of a small natural gas plant.

Unlike conventional magnetic confinement designs such as tokamaks or stellarators—which position powerful magnetic coils outside a hollow, donut-shaped vacuum vessel to keep scorching fusion plasma away from the containment walls—the levitating dipole inverts the layout. The dipole reactor vessel has no central hole. Instead, a large, torus-shaped superconducting magnet floats unsupported inside the chamber. The resulting magnetic dipole field traps, compresses, and heats the surrounding plasma to initiate fusion. Placing the magnet in the interior delivers two significant advantages: 1. **Geometric Plasma Stability:** The altered magnetic field geometry naturally suppresses certain disruptive plasma instabilities. 2. **Reduced Hardware Costs:** Because the plasma is inherently easier to stabilize and the central magnet does the heavy lifting, the overall system requires fewer external magnets, making it substantially cheaper to build.

---

[03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) Neutron Damage and OpenStar's Modular Solution

Despite its architectural benefits, placing the magnet inside the plasma chamber introduces severe physical drawbacks. In addition to the engineering headache of magnetically levitating a massive, supercooled magnet in place, the hardware is exposed to the core of the reaction. Deuterium-tritium (D-T) fusion produces high-energy fast neutrons that carry no electric charge. Because magnetic and electric fields cannot deflect or contain them, these neutrons blast directly into the magnet structure and rapidly degrade it. While aneutronic fuels like boron avoid heavy neutron production, they require drastically higher ignition temperatures, compounding the difficulty. Because of these roadblocks, the dipole concept saw minimal progress after its 1980s inception, outside of an MIT test device operated from 2004 to 2011 and modest experimental work in Japan.

```mermaid flowchart LR accTitle: TMANUI Maintenance and Modular Replacement Cycle accDescr: Step-by-step cycle of robotically removing and refurbishing the modular dipole magnet.

subgraph Operation["~1 Year Operational Run"] A["Plasma Burns with D-T Fuel"] B["Outer Shield Absorbs Heat & Neutrons"] A --> B end

subgraph Maintenance["Robotic Refurbishment (~2 Weeks)"] C["Magnet Module Lowered by Robots"] D["Moved to Radiation-Safe Room"] E["Outer Damaged Section Replaced"] B --> C C --> D D --> E end

subgraph Lifecycle["Extended Plant Lifecycle"] F["Reinstalled for Next Cycle (Up to ~10 Times)"] G["Full Magnet Assembly Retired/Overhauled"] E --> F F -->|"Repeat x10"| G end ```

OpenStar's innovation avoids unproven fundamental physics in favor of pragmatic maintenance management: - **Material Selection:** The core uses standard high-temperature superconductors (yttrium barium copper oxide class) wrapped in a heavy protective shield of heat-resistant, neutron-absorbing tungsten and boron carbide. - **Segmented Architecture:** Rather than casting the magnet as a monolithic block, the design splits it into an inner main section and an expendable outer section that absorbs the brunt of neutron damage. - **Robotic Swaps:** After roughly one year of continuous operation, robotic equipment lowers the entire magnet assembly out the bottom of the reactor into a shielded, radiation-safe bay. - **Two-Week Turnaround:** Workers and automated tooling replace the sacrificial outer components over approximately two weeks. The core can withstand about ten of these operational cycles before the entire magnet requires replacement.

Sabine jokes that the system is essentially the "fusion reactor of Theseus, except every plank is radioactive and the philosopher has died." OpenStar has not fundamentally solved every challenge of nuclear fusion, but they provide a concrete engineering roadmap to revitalize an overlooked concept. Moreover, she argues that other fusion ventures should take note: routine maintenance and component replacement will inevitably represent a huge fraction of long-term operating costs.

## Notable quotes > [00:00](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=0s) external memory system for recording conversations so you do not have to endure them twice — Sabine Hossenfelder

> [00:00](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=0s) levitating dipole — Sabine Hossenfelder

> [00:00](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=0s) fundamentally flawed — Sabine Hossenfelder

> [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) TMANUI — Sabine Hossenfelder

> [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) fusion reactor of Theseus, except every plank is radioactive and the philosopher has died — Sabine Hossenfelder

> [03:17](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=197s) routine maintenance and component replacement will inevitably represent a huge fraction of long-term operating costs — Sabine Hossenfelder

> [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) Geometric Plasma Stability — Sabine Hossenfelder

> [01:34](https://www.youtube.com/watch?v=PzDe6AP_JfE&t=94s) Reduced Hardware Costs — Sabine Hossenfelder

## Mentioned OpenStar Technologies, Sabine Hossenfelder, Plaud Note Pro, TMANUI, MIT

## Video description Check out Plaud using my link: https://bit.ly/4xaAtrp. Amazon: https://amzn.to/45Ba5La. Use my code SABINE for 10% off!

A nuclear fusion startup in New Zealand just published a design for a "levitating dipole" reactor, an approach almost everyone had written off as fundamentally flawed. It’s the first new idea in nuclear fusion design for decades. Let's take a look.

Paper: https://doi.org/10.1016/j.fusengdes.2026.115864

#physics #science #energy #technology #nuclear #sciencenews

👕T-shirts, mugs, posters and more: ➜ https://sabines-store.dashery.com/ 💌 Support me on Donorbox ➜ https://donorbox.org/swtg 👉 Transcript with links to references on Patreon ➜ https://www.patreon.com/Sabine 📝 Transcripts and written news on Substack ➜ https://sciencewtg.substack.com/ 📩 Free weekly science newsletter ➜ https://sabinehossenfelder.com/newsletter/ 👂 Audio only podcast ➜ https://open.spotify.com/show/0MkNfXlKnMPEUMEeKQYmYC 🔗 Join this channel to get access to perks ➜ https://www.youtube.com/channel/UC1yNl2E66ZzKApQdRuTQ4tw/join 📚 Buy my book ➜ https://amzn.to/3HSAWJW

[Source](https://www.youtube.com/watch?v=PzDe6AP_JfE)

中文简报

OpenStar公布配有可更换堆芯的悬浮偶极聚变反应堆 该初创公司将中央磁体设计为可通过机器人更换的消耗性模块，而非依赖未经证实的物理原理，从而绕过了聚变领域严重的中子辐照损伤瓶颈。 视频详细解析了OpenStar Technologies提出的名为“TMANUI”的商用聚变反应堆设计，该设计重启了历史上的悬浮偶极概念，预计可产生208兆瓦电力。与标准托卡马克将磁体置于环体外部不同，TMANUI将一个悬浮超导磁体直接放置在等离子体室内部，以更少的外部线圈实现等离子体的自然稳定。尽管氘-氚反应产生的高能中子不可避免地会轰击该中央磁体，但OpenStar通过务实的工程手段而非奇异物理机制来应对这一挑战：采用由钨和碳化硼保护的模块化磁体部件，并在每年的两周维护停机期内由机器人进行更换。Sabine对该项目聚焦于全生命周期维护和部件更换的务实态度表示赞赏，指出它解决了竞争对手往往忽视的实际运行现实问题。 新西兰聚变初创公司OpenStar Technologies公布了名为“TMANUI”的悬浮偶极商用反应堆设计，预计净发电功率为208兆瓦——大致相当于一座小型天然气电厂的发电容量。 (1:34) 与用外部线圈包裹等离子体的传统托卡马克和仿星器不同，偶极架构将单个环形超导磁体悬浮在反应室内部，能够自然抑制等离子体不稳定性并降低硬件成本。 (1:34) 由于在反应室内悬浮磁体会使其暴露于来自氘-氚聚变的破坏性、不带电的快中子轰击之下，自20世纪80年代以来，悬浮偶极概念除在MIT的一台测试装置（2004–2011年）以及日本开展的少量研究外，进展一直十分微弱。 (3:17) OpenStar通过模块化维护而非新物理机制来解决中子劣化问题：其钇钡铜氧磁体被包裹在牺牲性的钨和碳化硼屏蔽层内，每年由机器人将其从反应堆中降下进行为期两周的翻新维护，在彻底更换前可维持约十个周期。 (3:17)

Reading summary

New Zealand startup OpenStar Technologies has proposed 'TMANUI,' a commercial fusion reactor design reviving the long-sidelined levitating dipole concept. By placing a superconducting magnet directly inside the plasma vessel rather than outside it, the company aims to improve plasma stability and cut hardware costs while managing severe neutron damage through modular robotic maintenance. Inverted Architecture of the Levitating Dipole (1:34) OpenStar Technologies' proposed reactor design, designated TMANUI, departs from mainstream magnetic confinement architectures such as tokamaks and stellarators. While conventional reactors mount external magnetic coils around a hollow, donut-shaped vacuum vessel, TMANUI utilizes a chamber without a central opening and magnetically levitates an unsupported, torus-shaped superconducting magnet directly inside. The resulting dipole field traps, compresses, and heats the surrounding plasma, with the company projecting an output of 208 megawatts of net electric power, roughly equivalent to a small natural gas plant. Geometric Stability and Reduced Hardware Costs (1:34) Placing the magnet inside the vacuum chamber delivers two key design advantages. First, the altered geometry of the magnetic dipole field naturally suppresses certain disruptive plasma instabilities that routinely affect traditional containment systems. Second, because the plasma is fundamentally easier to stabilize and the central magnet provides the primary confinement, the overall reactor demands fewer external magnetic coils, which OpenStar claims will significantly reduce manufacturing and construction expenses. Mitigating Fast-Neutron Damage via Modular Swaps (3:17) The primary drawback of the internal magnet layout is extreme hardware degradation. In deuterium-tritium (D-T) reactions, uncharged high-energy fast neutrons bypass magnetic fields and directly bombard the internal assembly. Aneutronic fuels avoid severe neutron generation but require substantially higher ignition temperatures, leading the dipole concept to remain largely sidelined after its 1980s introduction, aside from an MIT test device (2004–2011) and modest Japanese experiments. OpenStar's proposed solution relies on modular component replacement rather than unproven materials. The central coil utilizes standard yttrium barium copper oxide (YBCO) high-temperature superconductors encased in tungsten and boron carbide shielding. The magnet is split into an inner core and a sacrificial outer segment. After an operational run of roughly one year, robotic tooling lowers the assembly into a shielded chamber to replace the outer segment in approximately two weeks, with the primary core rated to survive about ten operational cycles before requiring a total overhaul. While OpenStar has not eliminated every barrier to nuclear fusion, its roadmap emphasizes that practical engineering and routine component maintenance may revitalize abandoned reactor concepts. Long-term commercial viability across the fusion sector will depend heavily on managing the operating costs and downtime associated with regular hardware replacement.

Source quotes

[0:00] Sabine Hossenfelder: external memory system for recording conversations so you do not have to endure them twice [0:00] Sabine Hossenfelder: levitating dipole [0:00] Sabine Hossenfelder: fundamentally flawed [1:34] Sabine Hossenfelder: TMANUI [3:17] Sabine Hossenfelder: fusion reactor of Theseus, except every plank is radioactive and the philosopher has died [3:17] Sabine Hossenfelder: routine maintenance and component replacement will inevitably represent a huge fraction of long-term operating costs [1:34] Sabine Hossenfelder: Geometric Plasma Stability [1:34] Sabine Hossenfelder: Reduced Hardware Costs

Original transcript

[0:00] A nuclear fusion startup in New Zealand has  just put forward the first new idea for the [0:06] nuclear fusion technology for decades. They're  called OpenStar Technologies and they're the only [0:14] startup which works on an approach that's called  a "levitating dipole" that everyone thought was [0:20] fundamentally flawed. But they've turned the bug  into a feature. Let's have a look. Oh, I have a [0:27] new gadget. And look at this. Is it a bird? Is it  a pine? No, it's the Plaud Note Pro. A searchable [0:35] external memory for all the conversations I don't  want to sit through twice. It can record speech [0:40] continuously for up to 50 hours. Transcribe it  with speaker labels. Create different summaries [0:46] from the same recording and make everything  AI searchable. But Sabine, just use a phone, [0:52] you might say; drain my phone battery, download  all the recordings, use another app to transcribe [0:57] them, and then another app to search them. I don't  have time for this. Plaud is a one-stop solution. [1:04] Also, with Plaud, you can be sure that your data  is safe. You get 300 minutes of transcription per [1:10] month for free with paid plans available if  you need more. So, if you spend most of your [1:15] time on the beach, Plot probably isn't for you.  But if you handle a lot of verbal information, [1:21] Plot can make your life much easier. And they have  a 30-day no questions asked return policy. Click [1:28] the link in the description and use my code to  get 10% off. And now back to the episode. OpenStar [1:34] Technologies just published a paper with a new  design for a commercial nuclear fusion machine [1:40] that they say is predicted to produce 208 megawatts  of net electric power. To put this into context, [1:47] that's about the power output of a small gas  power plant. They call their design TMANUI, [1:53] a name that's derived from the Māori word for  sun. But what's new about it? Open Star has [1:59] bet on the levitating dipole approach. That's  a variant of nuclear fusion that works with [2:05] magnetic confinement. The usual approach is a  tokamak or stellarator in which one has the fusion [2:11] plasma inside strong magnets that usually form a  sort of torus. The biggest challenge for these [2:17] approaches is to keep the plasma levitating inside  the containment vessel because it's super hot and [2:24] when it touches the reactor walls that will  damage them. In the dipole fusion reactor, the [2:30] plasma vessel does not have a hole in the middle.  Instead, one levitates a big torus shaped magnet [2:36] inside the vessel. And this magnet is what creates  the dipole field. If the magnet is superconducting, [2:42] the field can be strong enough to trap the  plasma and heat and compress it, hopefully [2:47] enough for fusion to set in. The idea is kind of  similar to the Tokamak, but putting the magnet inside [2:54] the reactor vessel has two major advantages. The  first is that it removes some instabilities in the [3:00] plasma just because the geometry of the magnetic  field is different. The second is that because [3:05] it's easier to keep stable and the magnet is in  the middle, you need fewer of the magnets and [3:11] that makes it cheaper. But as usual, there are  trade-offs. The issue is that this idea brings [3:17] in a new instability, which is that now you have  to keep the magnet in place. Worse still, you now [3:23] have the magnet inside the plasma and that makes  it much more likely it'll get damaged. The biggest [3:31] problem is the fast neutrons that are produced in  deuterium-tritium fusion. Because they're neutral, [3:38] they can't be trapped or redirected with the  electric and magnetic fields. They will degrade [3:44] the magnet quickly. Now one could use different  fuels instead like for example Boron but it not [3:51] produce neutrons or not as many neutrons. But  the problem with such other neutron-free fusion [3:57] fuels is that the temperature one must reach is  dramatically higher than for deuterium-tritium fusion [4:04] which is also bad. And this is why in a nutshell  the dipole fusion approach pretty much hasn't [4:09] been pursued though it's been around since the  1980s. MIT then had a small test device which ran [4:16] from 2004 to 2011 and the Japanese have a small  research facility but that's basically it but if [4:22] not the levitation dipole then what's new and the  news is that OpenStar has come up with a way to make [4:30] this idea compatible with deuterium-tritium fuel  their idea is no groundbreaking new physics it's [4:36] rather a very down-to-earth management proposal  they say shield the magnet as well as you can and [4:43] and make sure that you can replace the shield  when needed. So the weakness of the approach [4:48] becomes a maintenance problem. Concretely, they  say they want to use a superconducting magnet of [4:54] barium and copper. That's fairly standard. Around  the magnetic coil, they have a shield of tungsten [5:00] and boron carbide. This is because this material  can withstand a lot of heat and it's also good at [5:06] absorbing neutrons. That too is fairly standard.  What's new is that the magnet is not one block. [5:12] It has a main section and an outer section.  They accept that this outer section will get [5:17] damaged and instead try to find a way to cheaply  replace it in regular intervals. So they plan on [5:24] a mechanism by which after about a year the  entire magnet module is removed from below by [5:30] robots and then put in a radiation safe room  to replace the damaged parts. They estimate [5:36] that this replacement itself would take about 2  weeks and that this can be done about 10 times [5:41] after which the other parts have to be replaced  as well. It's basically the fusion reactor of [5:46] Theseus except every plank is radioactive  and the philosopher has died. In summary, [5:52] OpenStar has not solved nuclear fusion but they have  proposed a serious engineering route for a strange [5:58] old fusion idea that previously had a big problem.  Also, I think there's a bigger lesson here. Other [6:05] startups should spend more time thinking about the  question of regular maintenance because it'll be a [6:10] major contributor to the running cost. And next  time your computer runs into a problem, just [6:17] try turning it off and replacing the radioactive  middle bit. Thanks for watching. See you tomorrow.