Wednesday What We’re Reading (Dec. 18, 2019)

It’s an even more eclectic week here at Soapbox World HQ than usual.

Lead Story: Magnetized Target Fusion is awesome

  • Fusion with pistons: General Fusion gets Bezos backing – If I were a billionaire, I’d definitely fund cool Future Technology projects, too.
  • Here’s the 2007 paper describing General Fusion’s concept – Start with a steel sphere two meters across, studded with steam-powered pistons. Fill it with a liquid lead-lithium alloy. Use equatorial pumps to turn it into a vortex, and polar pumps to pull it out the top and bottom, so you get a vertical cavity. Inject plasma into that cavity. Fire the pistons, which all impact the steel sphere at the same time and make a compression wave in the lead-lithium. That compression wave ignites the plasma, heating the lead-lithium. Run the hot lead-lithium from the polar pumps through a heat exchanger, which generates steam for turbines and the pistons. Repeat once per second. Neutron activation turns the lithium into tritium, and the use of liquid lead as a working fluid means you don’t need to worry about neutron bombardment turning your steel sphere into Swiss cheese.
  • Here’s a video, if my description isn’t clear – Obviously, General Fusion is invested in saying it’ll work, but the math seems to check out, and this particular approach to fusion seems to have a number of advantages over pure magnetic confinement and pure inertial confinement. The biggest one, as I see it, is that it operates on principles broadly familiar to today’s industrial equipment: it has some pumps, and it has some steam-powered pistons (with some electronics to control impact timing), neither of which is all that complicated. The devices to generate the plasma are a bit more esoteric, but well-understood.

All told, a nifty system, and one with a number of seeming practical advantages. Also, I love the idea of a fusion power plant being bulky, spiky, and loud. (All those pistons hitting a steel sphere…)


Science and Technology

Grab Bag

  1. I can’t get over how stupid an architectural choice they made here. Let’s say the average request to the Gameanalytics service in the previous bullet point takes 0.1 second to handle each request, which is likely faster than the real figure. Even then, it’s handling around 5800 requests simultaneously, more than Plaid’s 

Leave a Reply