MIT claim they will build the SPARC

[JoeStrout]

A new story (not much new detail though) about this project:
https://news.mit.edu/2024/commonwealth- ... plant-1217

[crowberry]

It is interesting to note that CFS will finance the ARC site by itself. For this they will need to raise a lot of new capital of course. Here are the links to the CFS press release and the blog post about the ARC site announcement:
https://www.cfs.energy/news-and-media/c ... n-virginia
https://blog.cfs.energy/cfs-will-build- ... -virginia/

[jrvz]

I decided quite some time ago that Commonwealth Fusion Systems would be successful with ARC. That was mainly because the they would use high temperature superconductors that would give them a stronger magnetic field (suppressing instabilities and allowing the device to be much smaller, saving cost and construction time), but also because they would use demountable joints, allowing them to split the toroidal field coils and install or remove the vacuum vessel in one piece (see "Demountable Superconducting Magnet Coils" by Sorbom et al. at https://www.burningplasma.org/activitie ... lDraft.pdf). Their ideas to use stainless steel as the "insulator" for their superconductive magnet coils and a molten salt blanket are also important.

However, I was surprised to see in their paper "Design, Fabrication, and Assembly of the SPARC Toroidal Field Model Coil" by Vieira et al. (https://ieeexplore.ieee.org/document/10411128) from a year ago, their toroidal field model coil for SPARC uses a few demountable joints, but not the two joints for each turn of each coil required for vertical disassembly. Also, the SPARC render in "SPARC as a platform to advance tokamak science" by Creely et al. (at https://pubs.aip.org/aip/pop/article-pd ... 162457.pdf) doesn't appear to show split toroidal field coils. So they'll wait to introduce that in ARC? Or are they not yet convinced that they can use vertical disassembly?

[Skipjack]

I believe the disassembly was always planned for ARC and not SPARC. SPARC won't need it because it won't be operating for long enough to need that sort of maintenance. It is also small enough to make taking apart easier than for ARC.

[baking]

They are testing low-temperature solders for demountable magnets for ARC. The idea is to melt the solder in the joints without melting the solder in the rest of the magnet. https://iopscience.iop.org/article/10.1 ... d0b2b/meta

I don't think ARC works without demountable magnets, but the exact design hasn't been published.

We will see SPARC assembled this year and it will be very different. The vacuum vessel will be in two halves and each half will be threaded through 9 magnets. The assembly tooling is currently in the tokamak hall. The halves will be welded together and the cryostat will be built around it. The cryostat base arrived from Italy a few weeks ago and will fill the hole in the floor of the tokamak hall. The magnet and VV assembly will be placed (or built) on the base and the walls of the cryostat will be assembled around it.

[jrvz]

Yes, the article about the soldered joints was very interesting. Apparently they plan to use compressed air to blow solder wherever they want it. I wondered how they'd avoid the solder short-circuiting the various joints - but then realized that would be no worse than the stainless steel that already connects the joints. By comparison with the superconductor, the solder qualifies as a slightly lossy insulator.

Dividing the vacuum vessel into only two halves sounds much better than the segmented architecture shown in their earlier paper - requiring only two welds instead of 18. I'd like to see what their tooling looks like. I'm envisioning a semicircular cradle for each half of the vacuum vessel, with a semicircular support rail, and rollers interspersed between nine stationary toroidal field coils. The rest of the coils would be mounted to a flat support. After threading the other half of the vacuum vessel into those coils with a similar cradle and rollers, the flat support would be slid over so the two halves of the vacuum vessel can be welded.

Something like that?

[baking]

Dividing the vacuum vessel into only two halves sounds much better than the segmented architecture shown in their earlier paper - requiring only two welds instead of 18. I'd like to see what their tooling looks like. I'm envisioning a semicircular cradle for each half of the vacuum vessel, with a semicircular support rail, and rollers interspersed between nine stationary toroidal field coils. The rest of the coils would be mounted to a flat support. After threading the other half of the vacuum vessel into those coils with a similar cradle and rollers, the flat support would be slid over so the two halves of the vacuum vessel can be welded.

the two stands can be seen at 1:25 in this video: https://www.youtube.com/watch?v=4WCLcmjFYiw&t=85s

Each will hold 9 TF magnets. How the vacuum vessel will be inserted, how the two halves will be joined, and how the assembly will be placed in the cryostat is unknown.