Zephyr Fusion: In-orbit fusion power

[rklaehn]

https://www.ycombinator.com/launches/Oox-zephyr-fusion-in-orbit-fusion-power

This looks very interesting. The basic idea is that while high quality vacuum is extremely expensive on Earth, it is free in space. So instead of buying an extremely expensive vacuum chamber and dealing with first wall issues, they just create a strong magnetic dipole in space, fill it with fusion plasma (presumably with something like neutral particle injectors and RF heating) and then attempt to get fusion in space.

I am not sure how they intend to convert the fusion energy into usable energy. Also there is no shielding, so you don't want to run this anywhere near anything you care about. But still, extremely interesting concept.

We will see more of these in the future. Startups that take cheap launch for granted and use the specific conditions in space for some processes that are impossible or very expensive on Earth.

Edit:

Maybe a short argument why this might be a good idea. A well known way to make fusion work is to just make the plasma bigger. Take existing tech and scale it up to absurd sizes, and you will have a practical fusion plant on paper. That is why ITER is so big.

The problem is that on Earth this is prohibitively expensive.

τₑ ~ Rē

The energy confinement time is roughly proportional to the plasma radius squared!

[Asteroza]

It's Buck Naked Fusion, a very redneck engineering application of the two greatest merits of commercial LEO activities, microgravity and vacuum.

[edzieba]

1) Is the cost-driver for fusion reactors really the vacuum chamber, or is it all the other parts of the fusion reactor (superconducting electromagnets, electromagnet driving power electronics, plasma sensing and stabilising systems, gas injectors, energy recover, etc)?

2) The majority of fusion reactor designs are thermal recovery breeders: A molten Lithium blanket is used to heat water for steam generators and to generate Tritium for fuel. Removing the walls means you have removed this system, so need to replace it with a new system.

Basically, is making the heat recovery jacket not be vacuum-tight (or not have a vacuum-tight outer wall protected by the heat recovery blanket) worth the headaches of adding a bunch of radiators and having to launch the thing into orbit and then assemble it?

[rklaehn]

It's Buck Naked Fusion, a very redneck engineering application of the two greatest merits of commercial LEO activities, microgravity and vacuum.

I am not sure you would want this in LEO. Unless they do aneutronic fusion, which seems unlikely with thermal approaches, this will be an intense neutron source. You will certainly be able to measure it going overhead.

[rklaehn]

1) Is the cost-driver for fusion reactors really the vacuum chamber, or is it all the other parts of the fusion reactor (superconducting electromagnets, electromagnet driving power electronics, plasma sensing and stabilising systems, gas injectors, energy recover, etc)?

Giant vacuum chambers certainly are not cheap. Regarding energy conversion, we will have to see what their plans are. But I would say that giant plasma radii are a very underexplored area of the design space for obvious reasons, and it would be very interesting to see experiments even if they don't lead to a commercially competitive reactor.

This is very similar technology to magnetic solar wind sails and artificial magnetospheres for shielding. So if the only outcome is to turn investor money into research into advanced space applications without achieving breakeven I would still be OK with it... 

2) The majority of fusion reactor designs are thermal recovery breeders: A molten Lithium blanket is used to heat water for steam generators and to generate Tritium for fuel. Removing the walls means you have removed this system, so need to replace it with a new system.

Basically, is making the heat recovery jacket not be vacuum-tight (or not have a vacuum-tight outer wall protected by the heat recovery blanket) worth the headaches of adding a bunch of radiators and having to launch the thing into orbit and then assemble it?

From the illustrations it looks like they don't want to do a traditional recovery jacket at all. We don't really know much about their approach, not even if it is D-T or something more exotic.

[baking]

It's not DT fusion, it's boron-sulfur fusion.

[Robotbeat]

Why? Do you have something useful to say other than just insults?

[laszlo]

...But I would say that giant plasma radii are a very underexplored area of the design space for obvious reasons, ...

Respectfully disagree. It's the medium plasma radii that are underexplored. The giant plasmas (the ones big enough to have sufficient mass to generate their own gravitational compression and confinement fields) have been well-studied for over a century 

[baking]

Why? Do you have something useful to say other than just insults?

No blanket, no heat output, no power conversion, no tritium breeding.  It is a literal joke, but they got money from Y Combinator.  Very low effort scam.

[Robotbeat]

I'm sorry, but how exactly is it a scam? I'm calling "boron sulfur" on your "boron sulfur." I'm fine with companies taking alternative paths to a goal, and there are applications of fusion that don't require integrated tritium breeding.

[DanClemmensen]

Why? Do you have something useful to say other than just insults?

No blanket, no heat output, no power conversion, no tritium breeding.  It is a literal joke, but they got money from Y Combinator.  Very low effort scam.

If your goal is a space drive, none of those things may matter. Fusion reactors work by containing a plasma, and they fail when the plasma leaks. But if you can direct that leak, it becomes a space drive.

[Robotbeat]

It's also pretty useful if your goal is to demonstrate ignition. You can add the other stuff later.

Not sure if it overcomes the greater cost due to launch, but it really is a kind of interesting approach, i.e. taking advantage of the space environment to make fusion easier to develop.

[baking]

Why? Do you have something useful to say other than just insults?

No blanket, no heat output, no power conversion, no tritium breeding.  It is a literal joke, but they got money from Y Combinator.  Very low effort scam.

If your goal is a space drive, none of those things may matter. Fusion reactors work by containing a plasma, and they fail when the plasma leaks. But if you can direct that leak, it becomes a space drive.

The fusion propulsion concept it was based on was impractical, but at least it addressed some basic questions.  A proposal to generate power from fusion without mentioning how you are going to generate power from fusion is a joke.  I'm not saying it can't be done, but pick something, anything, so at least we can critique it.  As it stands, all you can do is laugh or shake your head in amazement.

[baking]

It's also pretty useful if your goal is to demonstrate ignition. You can add the other stuff later.

Not sure if it overcomes the greater cost due to launch, but it really is a kind of interesting approach, i.e. taking advantage of the space environment to make fusion easier to develop.

Yes, it's an interesting science experiment, but is that what Y Combinator is doing now?

[Robotbeat]

It's also pretty useful if your goal is to demonstrate ignition. You can add the other stuff later.

Not sure if it overcomes the greater cost due to launch, but it really is a kind of interesting approach, i.e. taking advantage of the space environment to make fusion easier to develop.

Yes, it's an interesting science experiment, but is that what Y Combinator is doing now?

Who in their right mind would be mad about Y Combinator not investing in yet another B2B SaaS?

[Twark_Main]

1) Is the cost-driver for fusion reactors really the vacuum chamber, or is it all the other parts of the fusion reactor (superconducting electromagnets, electromagnet driving power electronics, plasma sensing and stabilising systems, gas injectors, energy recover, etc)?

2) The majority of fusion reactor designs are thermal recovery breeders: A molten Lithium blanket is used to heat water for steam generators and to generate Tritium for fuel. Removing the walls means you have removed this system, so need to replace it with a new system.

Basically, is making the heat recovery jacket not be vacuum-tight (or not have a vacuum-tight outer wall protected by the heat recovery blanket) worth the headaches of adding a bunch of radiators and having to launch the thing into orbit and then assemble it?

The issue isn't being "vacuum tight" (leakproofing). The issue is that it needs to be a pressure vessel withstanding 1 atmosphere of negative pressure. For reference this is roughly 10x more than the rating for the floor of a parking garage.   

Pressure vessel mass scales as the volume. If you want a 2x length scale-up, the mass of your pressure vessel increases by 8x. Ditto for the pump-down time, which becomes a significant rate limiter.

In space you could (if necessary) add a thin MLI-insulated blanket with tubes for breeding. Building this system in space (with free vacuum and cheap insulation via MLI) should be simpler and lighter compared to a huge pressure vessel.

[laszlo]

... But if you can direct that leak, it becomes a space drive.

Or a weapon.

[DanClemmensen]

... But if you can direct that leak, it becomes a space drive.

Or a weapon.

No more or less a weapon than any ion drive.

[InterestedEngineer]

I predict that thermal sourced power generation from fusion will never really catch on.

Direct to electricity is far more promising.  Especially for space-based electric power.  (e.g. Helion)