SoylentNews is people
SoylentNews
NASA's Kilopower nuclear reactor with Stirling converters (not an RTG) has passed key tests:
The Kilopower team conducted the experiment in four phases. The first two phases, conducted without power, confirmed that each component of the system behaved as expected. During the third phase, the team increased power to heat the core incrementally before moving on to the final phase. 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.
"We put the system through its paces," said Gibson. "We understand the reactor very well, and this test proved that the system works the way we designed it to work. No matter what environment we expose it to, the reactor performs very well."
The Kilopower project is developing mission concepts and performing additional risk reduction activities to prepare for a possible future flight demonstration. The project will remain a part of the STMD's Game Changing Development program with the goal of transitioning to the Technology Demonstration Mission program in Fiscal Year 2020.
The full system will generate 10 kW of power, but the prototype tested from November to March was designed to produce just 1 kW. The solid uranium-235 core is safe to handle.
The Kilopower Reactor Using Stirling Technology (KRUSTY) prototype exceeded almost all performance metrics.
Multiple units could power missions on the Moon, Mars, or other destinations:
"Kilopower's compact size and robustness allows us to deliver multiple units on a single lander to the surface that provides tens of kilowatts of power," NASA Associate Administrator Steve Jurczyk said in January.
Also at Beyond Nerva. 3m8s video.
Previously: NASA's Kilopower Project Testing a Nuclear Stirling Engine
Initial Tests of NASA's Kilopower Nuclear System Successful
(Score: 2, Interesting) by suburbanitemediocrity on Thursday May 03 2018, @01:21PM (13 children)
"doesn't sound like anything I'd want to use on a long-term mission."
It's all about mtbf and reliability. With proper engineering you can make anything as safe as you want or need. There is nothing intrinsically wrong with mechanical systems.
(Score: 5, Informative) by VLM on Thursday May 03 2018, @01:59PM (12 children)
Well, I'll try to explain it without any bad SN automobile analogies.
With a RTG, heat flows across a bulk solid material just like a boring heatsink. The heat-resistance crosses sort of a back-EMF like, um, force, such that its a slightly worse conductor in exchange for a like amount of electrical power generated. It turns out to be pretty easy to size the parts such that short circuiting the RTG won't melt the reactor (as semiconductors are usually not-terrible heat conductors) and open circuiting the RTG obviously means no "force" holding back the heat so if a fuse blows the reactor runs colder than usual, which is usually really good news. Like dumping the load off a series wound motor, nothing too scary happens (usually). Maybe this is a really bad non-technical analogy if I'm talking about series-wound electric motors and stuff.
With a Stirling, heat flows thru a gas and no engineer has ever made anything that can contain a gas forever, at least relative to nuclear waste length lifetimes, so someday the core will not be cooled at all by the Stirling. Also its like a shunt-wound motor in that the reactor will pump out constant watts of heat (more or less) while shutting off the generator means no gas movement means no cooling, so life gets exciting even if the gas doesn't leak out, if a fuse blows.
There are engineering solutions to prevent stirling engine meltdown. On the earth we use two, four, six, maybe more coolant loops so none ever fail simultaneously and then put in triple redundant backup emergency coolant systems. And we have smart humans on the ground. Of course you kill lots of the population and destroy all the infrastructure in an earthquake and tsunami like happened in Japan, you're still screwed. Or if the sorcerers apprentice starts pulling levers in a fun experiment like Chernobyl, it goes poorly. Or if something really freaking weird happens and the ops misinterpret the wrong gauges thru a long chain of bad luck, you get three mile island which none the less was pretty safe outcome compared to Chernobyl or Japan's events, so in some ways TMI wasn't a total failure. In space you could hook the reactor up to enough solid metal boring heatsinks such that if the electric stuff totally failed it would still have maybe 25% cooling maybe enough to prevent a meltdown.
Oh and this is before other fun... The worst thing solids do when you abuse them is crack or melt so RTGs can have the crap beaten out of them, even re-enter the atmosphere, and it'll be "OK" for some value of "OK". However gas/liquid stuff explodes nicely even for strictly speaking non-nuclear explosion reasons. See Japan, Chernobyl, etc. You could build hardware so incredibly strong that the buildings in Japan wouldn't pop or the pipes at Chernobyl wouldn't pop and that would ... more or less work until the fuel melted. But a solid block of a RTG is much more inherently safe.
Might look like I'm anti-reactor pro-RTG but there are various engineering tradeoffs where one is better than the other despite comments above. A RTG might be a thousand times less likely to create a nuclear superfund site, but it might take ten thousand times the mass of radioactive junk. So seemingly paradoxically a reactor is a thousand times more likely to fail the mission while on average being ten times easier to clean up which superficially appears impossible but its actually consistent. A less reliable reactor sounds like a bad idea for a mars mission unless you have like 10-way redundancy in which case losing one isn't a biggie and we don't have the lift capacity to launch a RTG large enough. Or another fascinating engineering tradeoff, I suspect the total predicted environment damage over time of launching small reliable tested complete reactors into space would be much lower than launching individual components of RTGs and assembling a freakin' huge RTG in orbit before a mission, not to mention a reactor built on the surface MIGHT be more reliable than a RTG built in space, at least until we have substantial space infrastructure.
Another odd reactor vs RTG issue is total activity at launch time (in case of a launcher accident). Activity of a reactor thats shutdown and hasn't been run at full power for a long time is actually ridiculously low and real reactors on earth need high-activity sticks to safely and predictably boot them up from cold start because there's not that many thermal neutrons floating around randomly. Initiators of implosion a-bombs are non-trivial design problem for same reason, how to make the thing go nuke-mode at maximum density and not too early or late is actually quite a classified puzzle. RTGs of course have both way more material and its hot hot hot no off switch active as hell at launch time. So in terms of "disintegrations per second" I suspect a brand new reactor is vastly staggeringly cleaner than a RTG. AND after long years at full power, its the other way around, RTGs have worn out and a couple half lives have gone by and they're slightly less than at launch time, but reactors are a smoking stinking ruin of almost random extremely radioactive isotopes from U-235 split in half (and some neutron activation but lets keep this simple, etc).
(Score: 2) by HiThere on Thursday May 03 2018, @05:40PM (5 children)
I *think* the slug of uranium in the reactor is small enough that even without active cooling it wouldn't generate enough heat to melt. Or at least that's the way I read the summary.
That said, a closed mechanical system makes me nervous too, but for the use case of this thing, I think that failure would just translate into "mission failure". They don't seem to be planning to use it on man-rated missions. And even if it were to melt, that would just cause it to change into a less heat-retaining shape if there were any gravity present. It's quite unlikely there would be a solvent (like water) to spread it around, and even if there were, it's only a small amount, and so heavy it wouldn't easily spread far.
So my nervousness has more to do with mission failure than anything else. But out past the orbit of Mars there really isn't much other choice. Mirrors + solar cells could work in free-fall, but that would be more complex.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by dry on Friday May 04 2018, @05:15AM (4 children)
Eventually we will need something like this for manned missions. a Mars mission for example can't rely on solar as there are dust storms that can last months.
(Score: 2) by HiThere on Friday May 04 2018, @04:44PM (3 children)
Yes, but eventually isn't now, and a manned mission had better have a repair tech available.
That said, Stirling engines are among the most reliable mechanical engines. For one thing, they can be sealed against contamination. So for the manned mission you'll need multiple generators, the ability to make parts, and the tools to open and repair individual failures. And those are going to be rare, so that can't be anybody's main job.
This thing, though, reminds me of vacuum tube computers. Just barely good enough. What the equivalent of transistors is, though, I have no idea. As you say, solar power looks dubious. MHD hasn't ever really panned out. Etc. Of course, I could be excessively pessimistic, but moving parts wear.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by dry on Friday May 04 2018, @10:59PM (2 children)
I'd guess by the time something like this is needed for manned missions, it'll be the mark 3 or so version. As well a Mars mission is going to use a mixture of solar and nuclear and may not usually depend on the nuclear.
(Score: 2) by HiThere on Saturday May 05 2018, @05:57PM (1 child)
Well, you need to remember that solar power falls off as the square of the distance from the sun. It's true that Mars' atmosphere is thin, and usually rather transparent, but solar power will be weaker than at earth's orbit even so.
earth orbit radius = 93 million miles (approx.)
mars orbit radius = 142 million miles (approx.)
93^2/142^2 = 0.43 (approx.)
So figure you'll need twice as many solar cells to generate a given amount of power on Mars than on Earth. This is *quite* rough, of course, but it should be in the ballpark.
IOW, it's quite likely that given that you're going to need nuclear power, adding the solar power wouldn't be worth it. Better to have redundant nuclear power, so if one goes down you can fix it. I just wish there were something better than a steam engine to extract the power with. OTOH, the lower average temperature on Mars should mean the steam engine is more efficient, and at least with a Stirling engine you don't need to keep replacing the working fluid.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 3, Interesting) by dry on Saturday May 05 2018, @06:26PM
Have to consider things like power distribution. Probably going to be a shortage of wire at first and a real shortage of poles. Then there's how long it'll take the nuclear plant to come on line. Then any power needs far from camp.
Needing twice the solar panels as Earth isn't that bad. With lack of cloud cover and a higher flux of UV should boost that 43% up a bit.
(Score: 2) by choose another one on Thursday May 03 2018, @06:13PM (5 children)
There are other non-technical tradeoffs too:
1. RTGs are no longer a viable option because NASA doesn't _have_ much more Plutonium (3 missions left I think), no one is making large quantities of it for weapons any more and restarting production for space exploration alone is challenging (and requires a reactor with all the disadvantages listed, albeit not on top of a rocket) - see e.g. http://spacenews.com/plutonium-supply-for-nasa-missions-faces-long-term-challenges/ [spacenews.com]
2. At-launch activity/toxicity level of RTGs has already become a fairly large political / environmental issue
3. For reactors, at-launch is much cleaner, and failure in deep space or on another planet is a _mission_ problem, but not a political / environmental issue - yes, the reactor is a smoking stinking ruin of almost random extremely radioactive isotopes, but by that time it is in deep space or at worst it is on the moon/mars, and hardly anyone gives a shit, really. The _only_ issue would be if there is a live reactor on a sample-return mission or an earth-slingshot, which isn't proposed and probably wouldn't fly for exactly that reason...
So, yeah the tradeoff is exactly as you describe, IF we (well NASA) had plenty more Pu, AND was intending to stick reactors in earth-return missions.
(Score: 2) by JoeMerchant on Thursday May 03 2018, @09:33PM (4 children)
The French could fix that for you, they're not afraid of breeders.
🌻🌻 [google.com]
(Score: 2) by dry on Friday May 04 2018, @05:23AM (1 child)
The problem is that Pu-238 is harder to make/separate and if IRC needs more then a simple breeder reactor. Basically neutron irradiation of neptunium-237 or Americium.
(Score: 2) by JoeMerchant on Friday May 04 2018, @03:04PM
We, the French, the Canadians (probably not the French Canadians), and many other countries have had the technology for over 50 years. What we lack is the will to use it.
🌻🌻 [google.com]
(Score: 2) by VLM on Friday May 04 2018, @03:13PM (1 child)
One of the plutonium isotopes is literally thermally hot in bulk which makes for crappy bombs because of fizzle danger (self triggered at the wrong instant resulting in bad yields) and great RTGs, the other isotope is the opposite. So you need a military bomb focused breeder program along with a bomb-focused processing plant.
You can make a-bombs that are not isotopically clean, they just work better if cleaned up (much hand waving about engineering effort vs manufacturing effort removed). Likewise already big RTGs would be immense if they used mixed or a-bomb grade Pu.
AFAIK the French are not into RTG generation but thats political and has varied from decade to decade.
(Score: 2) by JoeMerchant on Friday May 04 2018, @04:02PM
It is all VERY political, and has been from the beginning.
I'd like to see the US take a Space 1999-esque style leadership role on the moon - build breeder reactors there, preferably not on the "dark side", and keep them open from day 1 for true international inspection and verification of activities. Load the "bad" isotopes on rockets to the sun (WCGW?), and distribute/sell the good ones for the betterment of mankind. I know, the 1999 plot was just a waste dump, and this one would require full-court-press to be running by 2099, but it seems like a good use for the real-estate, and I'd rather put the resources into something like that than bashing each other for control of the remaining oil reserves.
🌻🌻 [google.com]