Tuesday, May 8, 2018

If This Thing Was 20kW, I'd Have to Start Looking for Financing

Machine Design relays a report from NASA and the National Nuclear Security Administration (NNSA) that a small, fission-powered, nuclear power plant has undergone some robust initial testing and is working perfectly.
A small, lightweight fission reactor built at NASA’s Glenn Research Center in Cleveland recently passed extensive testing in the Nevada desert “with flying floors,” putting it one step closer to powering human outposts on the moon or Mars.
In what may be a desired name in search of an acronym (and a nod towards “The Simpsons”), the reactor is called KRUSTY - Kilopower Reactor Using Stirling Technology.  KRUSTY went through testing from November 2017 through March.  NASA report pdf here.
The experiment culminated with a 28-hour, full-power test that simulated a mission, including reactor startup, ramp to full power, steady operation, and shutdown. Throughout the experiment, the team simulated power reduction, failed engines, and failed heat pipes, showing that the system could continue to operate and successfully handle multiple failures.
KRUSTY is said to be able to deliver 10 kW for up to 10 years.  I know I've talked about this before, but while your house may be rated for 30kW (120V with 250 Amp service), you typically don't use a third of that.  The way power is distributed to progressively smaller groups of houses in a neighborhood, the power companies budget around 6kW per house.  The higher voltage going into the neighborhood is dropped to the 120V that goes to your house (in the US, of course) by distribution transformers.  There is usually one of these to every five houses, and it's rated for 25 kW.  They budget 6kW for the average house and you can quickly see that 5 houses makes 30 kW not 25, but they rationalize that by saying it's not likely everyone will be using the full 6 kW at the same time.   

That says KRUSTY could power a typical house easily; maybe even two houses.  Cool.  What's inside?  And exactly what do they mean by "Stirling Technology" in the name? 
The reactor core is in a solid, cast-uranium-235 core about the size of a roll of paper towels. The prototype power plant uses passive sodium heat pipes instead of cumbersome pipes, pumps, plumbing, and a liquid working fluid used by traditional reactors to carry heat from the reactor.
The heat powers two Sterling heat engines which convert heat energy to mechanical work and mechanical work into electricity.  This means it could power a human outpost, which is ideal for the Moon, where generating power from sunlight is difficult due to the 14-day lunar nights. The  reactor is also fault tolerant, so any loss of cooling leads to an automatic reduction in fission power with no possibility of uncontrolled  reactions.  Ultimately, though, the target is to power more exploration, including Mars.


Artist's sketch of an array of four KRUSTY power plants on Mars.  From NASA's Kilopower project page.

Back in the first year of this blog, 2010, I offered a strange thought.  The US Navy has been powering ships with nuclear reactors for around 50 years.  These reactors have become a standard federal part number.  At least in concept, they could say "we need a new A4W" or "we need a new A1B" and order one from the approved suppliers.  One A4W delivers 104 Megawatts, or more than 10,000 times the power KRUSTY can deliver.  If KRUSTY can power a home, an A4W can power 10,000 homes.  One A1B delivers 700 MW.  You can do the math. 
What if we covered the country with these reactors?  Instead of giant, centralized power plants, what if we had a distributed network of thousands of these reactors?
Now maybe this isn't the exact ticket.  Maybe the Navy's reactors require too much human attention and maintenance, and maybe the design is based on having as much cooling water as they could ever want - the entire ocean as a coolant source.  I just still think the answer lies in distributed power systems and fewer centralized, single-point failure scenarios.  I'd love to have a KRUSTY in my backyard.  Yeah, considering it's NASA hardware I don't think there's a chance I could ever afford it.

Edit 0710 EDT 5/9/2018 - correct typo.


16 comments:

  1. Every 4 years or so a news article comes along dealing with small, portable, safe nuclear power plants that are actually quite affordable (compared to, say, a full sized powerplant that needs a metric ton load of auxiliary equipment and huge amounts of space.

    Like something that would fit on a tractor trailer and weigh within tractor trailer's weight limits.

    But... It's nuclear, it's bad, we'll all die from radiation poisoning and will irradiate the whole world. Green bastards!

    Please, quicker, faster, under this administration.

    ReplyDelete
  2. An A1B delivers just 700m(illi)Watts?

    ReplyDelete
  3. I saw that typo, too, and decided to cut SiG some slack. :-)

    ReplyDelete
    Replies
    1. The most common mistake I make is to leave out words, especially articles. Lately, I find myself typing a space between syllables in places it doesn't belong, like "in stead" rather than "instead", or something like that.

      Delete
  4. Stirling engines won't work in space. Not even on Mars. Not enough ambient air pressure, and would have real problems with heat build up. To even begin to work, they'd have to have excellent heat transfer properties, and enormous radiators. I suppose you could build one with a closed fluid loop. At least it could be relatively low pressure.

    They appear to work really well on submarines, though.

    ReplyDelete
    Replies
    1. Stirling engines do not require external atmosphere. But true that maintaining a suitable heat sink is a design concern.

      Delete
  5. How do the numbers look in terms of costs and return on investment for domestic/industrial use here on the Third Rock?

    ReplyDelete
    Replies
    1. Hard to calculate. We don't know what cost savings would come from mass production. I agree with SiG though that distributed power production is the answer to our fragile grid.

      Of course, you know as well as I do that they would be stolen by terrorists, beaten apart, and the radioactive material assembled into dirty bombs (or just spread around by hand, allahu ackbar style).

      Delete
    2. Malatrope says it well. I can't get numbers on what they cost or any of those important details. It's a prototype of something that will become certified for manned spaceflight (where the paperwork has to outweigh the launch vehicle). I'm sure they could be made cheaper than what this one costs.

      Delete
  6. The reactor part should be amazingly simple and relatively safe. But to turn that heat into moving electrons requires mechanical motion. That is the weakest part of such a power plant. Anything that moves eventually wears out and has a failure. A design that transforms heat into electrical power directly without steam orsome other mechanical generator would be almost foolproof.....and unbreakable is what is needed when your tens of millions of miles away from the nearest hardware store.

    ReplyDelete
  7. On reflection, to expand on my previous reply, Stirlings can certainly be built to use ambient atmosphere, but it's not an inherent requirement. Indeed, commercial and industrial Stirlings are frequently sealed and pressurized with various gasses to increase power. Pressurized air in hot oil mist is not a good idea-people have been killed by exploding engines. I have been wondering about sulfur hexafloride as a working gas.

    ReplyDelete
    Replies
    1. I just want to point out that this system is being developed by NASA for use in a hard vacuum and on Mars, which is pretty close to a hard vacuum.

      I'd be inclined to believe that they understand the environment as well as anyone would, and probably have tested the system in a vacuum chamber as part of their "November to March" testing period.

      The heat pipes run on sodium vapor and vapor condensing back to liquid sodium in the Stirling power generators.

      It's all described on page 4 of the NASA pdf I link to in my third paragraph.

      Delete
  8. One of the NASA Stirling designs I like supports the moving part on flexible metal blade-shaped springs rather than rolling or sliding bearings. The metal flexes within its elastic range. There is no rubbing seal between the moving part and the container; the gap between the moving part and the wall is small enough to reduce the gas leakage enough. Result is a 30 year lifetime.

    Is there any conservation-of-energy type reason to believe it would be net-energy-negative to run a machine to turn matter into antimatter? Would that machine fit in a garage?

    ReplyDelete
    Replies
    1. There aren't any shortcuts. To make antimatter, you have to first shatter an atom. That takes LOTS of energy. The shattering then releases a particle and its opposite, in addition to lots of other byproducts. Then the particle-antiparticle pair recombine,and release more energy, in all sorts of forms. Mostly by creating other weird particles.

      The real trick would then be to capture energy in usable form. You're not going to get more out than you put in, unfortunately. (You need a cyclotron just to get things started.) And the vessel will end up being radioactive as all get-out.

      Delete
  9. Except that U235 is mind-bogglingly expensive and the US barely has enough spare for these few NASA generators. It would be very expensive electrictiy.

    ReplyDelete