Hacker News Reader

Comment by ohiovr | Hacker News Reader

A tokamak can only constrain a plasma in 2 dimensions. To get confinement to last longer the magnetic field concentration needs to increase. An entirely new way to constrain plasma to 3 dimensions is needed for practical fusion applications. Otherwise accidents will be disastrous for plant operators.

View on HN

> To get confinement to last longer the magnetic field concentration needs to increase.

Which is what the new project will do - using a classic tokamak with the much higher field magnets that can be built with high-field superconductors.

> Otherwise accidents will be disastrous for plant operators

What's going to go wrong? I've seen talks that breaches will be rapidly cooling and will be contained by a modest amount of concrete.

Plasma disruptions

https://www.jp-petit.org/NUCLEAIRE/ITER/ITER_fusion_non_cont...

Field quenching of the super conducting coils. The stronger the field the worse it can be when it collapses.

Quenching is less of a concern with HTS coils. They are operated with liquid helium for higher critical current. The HTS is embedded in copper.

One argument going on right now is insulated vs uninsulated coils. Insulated are the norm, but uninsulated are virtually impossible to damage from quenching. The copper is a virtual open when the core has zero resistance. When the core quenches then suddenly the windings are all parallel and the coil turns into a single turn copper coil that immediately dumps the current.

Tokamaks are in the limelight, but the older brother hanging out in the shadow will take the stage in our lifetime :)

Keep an ear out for new stellarator projects (assuming an increase in research funding).

What's your opinion on stellerators? From what I heard about Wendelstein 7-X they were very happy with the results achieved in the experimental phases so far and are currently upgrading it to achieve longer runs in the future.

Disclaimer: I work on a stellarator. Specifically, the first MHD optimized stellarator and the first with 3D modular coils. W7-X is the only other one in the world. I feel that the current state of stellarator theory is encouraging. Coil optimization has integrated lots of important factors into their cost functions, including tolerance. Flux surface geometry optimization has integrated theoretical turbulence models into the cost functions. A pure MHD optimized machine will not be a fusion reactor, but we don't know enough about turbulent transport to build a working machine today. We need mid-scale machines to develop the theory. Tokamaks get the vast majority of the funding because their performance metrics are higher, but imo stellarators are capable of matching tokamaks in performance with much less control issues.

Reply on news.ycombinator.com