Yeah, the aqueous reprocessing methods are messy, and produces lots of (mildly) irradiated nitric acid. Pyroprocessing improves matters, but you still have to mechanically strip and break down the clad. Metal fuel really helps... too bad it's not that great neutronically! It's nice that edge heating isn't a big deal on this, and I doubt that anything else other than CERMET would be good enough to get away with that.
In principle, water heated to 800C has a comparable ISP to hydrazine, so the heat could be used directly.Thrust is small, though perhaps a hundred times the ion engine.
For launch safety I can't imagine the long-pole being the hydrazine though.
An operating Kilopower reactor is not safe to approach, but on the pad I'd be more worried about the hypergols.
Quote from: john smith 19 on 05/07/2018 07:38 amAn operating Kilopower reactor is not safe to approach, but on the pad I'd be more worried about the hypergols.Would this be a hazard to the electronics on board a deep space probe? These things are usually not very big, maybe 1m on a side, and it is hard to get very far away.
{snip}An operating Kilopower reactor is not safe to approach, but on the pad I'd be more worried about the hypergols.
Quote from: ThereIWas3 on 05/07/2018 02:15 pmWould this be a hazard to the electronics on board a deep space probe? These things are usually not very big, maybe 1m on a side, and it is hard to get very far away. 'yes' - but no.To make the weight low, the shielding is only in one direction. You put the probe in the shadow of this block of shielding, usually on a bit of an arm.It is a concern, but less dynamically than - for example 20m*20m of solar panel.
Would this be a hazard to the electronics on board a deep space probe? These things are usually not very big, maybe 1m on a side, and it is hard to get very far away.
Given the placing of the heat pipes for the 1Kw version it's fair to say that edge heating is a feature of the design. ...BTW Balance of plant costs were recognized as an issue as far back as the 1960's, hence the various "Advanced" gas cooled designs to more closely match SOP of conventional plants. An obvious question would be if you could go LEU while staying fast spectrum you're effectively looking at a breeder design (of sorts), turning U238 to Pu. However one of the features of space nuclear is its avoidance of fuel swelling issues by a very low burnup level. If this rises swelling might be more of a problem.An under recognized aspect of plant costs is that control rod drives for commercial reactors are roughly $1m each, because of their critical role in plant safety. OTOH Kilopower "load follows" automatically.In fact connecting the top end of those heat pipes to a water cooled heat spreader would make quite a good boiler, and companies like Siemens can supply turbines down to this power.https://www.siemens.com/global/en/home/products/energy/power-generation/steam-turbines/d-r-steam-turbines.html Which suggests (in principle) you could transport a complete power package in a 20' container, but you'd probably need to put the reactor inside a shield and connects some well insulated (and reasonably long) insulated pipes between the reactor and the turbogenerator package. I think the whole "Cast-in-place" nature of a one piece reactor core is quite attractive, especially insuring intimated contact with the heat pipes, or at least a well fitting sleeve they can be slid down.
A spacecraft propulsion bus that combined nuclear powered Hall thrusters and chemical propulsion (likely hypergolic) would form a formidable basis for outer solar system exploration. Imagine a 'Block' series of standardized 'Buses' that can be attached to differing spec spacecraft... That would be very cool. And scaling them up for crewed spacecraft use would give terrific capability.
Indeed. There are various tweaks in a reactor powered design. A key one is that the reactor could be in front of the experimental payload with just the power and monitoring cables coming through (around?) the shield. Obviously the further forward the reactor is in front of the shield the smaller a sector of the sphere the reactor is emitting over is an absorption target for the electronics (True for gamma rays. Works for neutrons as well?). Given the fairly low thrusts of even large ion thrusters the truss separating the reactor from the shield and payload can still be pretty light.
Wow, there's a lot to unpack here... and I'd love to pick your brain about a few things in the future. Feel free to get in contact with me through my blog!The edge heating is a bonus only in the 1 kWe version, since none of the others have peripheral heat pipes. According to a couple far more knowlegeable friends than I, the metal fuel kinda oblivates the edge heating advantage in the balance of systems, but... you gotta play the margins in astronautical engineering, so it's worth looking at moving the outer ring of heat pipes on the larger models closer to the edge. I've got a file with KRUSTY's MCNP model, if you're interested, but I'm pretty sure you have it already!http://permalink.lanl.gov/object/tr?what=info:lanl-repo/lareport/LA-UR-16-28377
There are two companies, to ny knowledge, that are exploiting heat pipe cooling already: Oklo Power is the more attractive one, IMO, for a number of reasons, but the biggest one is dirt simple: they're using a vertical core/PCS architecture, which allows heat pipes to also act as thermosiphons, which is a balance-of-energy-budget bonus to a large degree. Westinghouse also has a design - closer in line to the Megapower project through DNSA and DOD (possibly, considering the squirrely reporting the get away with) which I suspect are the primary funders of the higher powered (and UO2 fueled,prismatic FE variants), but it's horizontally oriented in operation. There's something going on with that PCS that I'd love to find out about, but... welcome to nuclear. Sometimes there's a door in you face, and it can't be breached with money.
As a non-Americocentric note, the UK pretty much only has high-temp gas cooled reactors. I didn't realize this until recently. I heartily endorse the quotes, and think that this is another example of British excptionalism being demonstrated by dropping the ball, as is their wont on advanced topics. IMO, between HTCGR and MSR (let the amazing peoples exult!), with a fission fragment reactor in the basement of every hospital, we've got EVERYTHING that fission has to offer - and let the civilians be as epically amazed as they should be!
Cast-in-place is epic... except as done at Y12. They're a weapons manufacturing facility that occasionally gets to pla with civilian work, so the Cold War is still very real. Just having the core manufactured there adds exceptional expense due to the fissile material reporting requirements... I think that you, John, understand the antipathy in the civ/mil nuc spectrum). So, in any other facility, it would potentially be cheaper... BWXT, are you listening?Regulatory burden of casting/manufacturing aside, just casting the KRUSTY core seems to be an incredible achievement... U7 is the ideal, and they managed in on a corewith THAT cross section? I'd love to knkow about their thermal management system...assuming I don't have the DNA/NNSS/DOE tapping politely at my door at 3AM!
Ed Pheil seems skeptical of breeding in this design... and in these matters I tend to trust him! In the coming months I hope to rewrite my blogs into a proper system description (and webpage) of Kilopower, since my posts have focused on KRUSTY up until now... hopefully I'll be able to play with the fast spectrum LEU implications, and again, I'd love to pick your brain, John!
Another option is to have your power and propulsion in one leading module, and have your payload on the end of a cable being pulled behind... it adds GNC complexity, but saves a lot of mass compared to a load bearing truss.
All PCS equipment is on the payload side of the shield, so the only thing that you really need is a control cable for the stepping motor used for the control rod. One benefit to this design is that by monitoring the hot end temp of the Stirlings, you are also monitoring the fission power level to a reasonably high degree of confidence.Yes, the inverse cube law also applies... but unfortunately, as the shield is slowing the neutrons you get additional gamma production within the shield, so it's not perfectly "spherical." Also, if anything is outside the shield then you will get backscatter from it, so you add a sphere (for a point), cylinder, etc of radiological source for each unshielded component.
IIRC one of the papers mentioned that the radiation flux recieved by the payload would be lower than that from an MMRTG, but I can't find the reference anywhere, and that seems kinda wonky to me. However, this is a matter of optimizing the shield thickness for the needs of the sensors being used, and is a fairly minor issue.
One thing to keep in mind is that most of the attention is given to the 10 kWe version right now, since it's the "big boy" in the design family, but the 1 kWe version is the most advanced - and that's more than enough for a lot of space probe missions. The flux off this reactor will be significantly lower than the 10 kWe version (although the shielding is also less, so the overall flux exposure is likely similar).
A "Fission fragment" reactor? That's normally associated with spaceflight (the actual fragment speed is about 5% of the speed of light. The sort of thing you want if you want to look at interstellar, rather than interplanetary distances).
Sorry it took me so long to respond... 1000+ things going on right now.Yes, it's talked about in terms of spaceflight, but... it's got a non-Carnot-based PCS by default as its' primary PCS, and as a side effect, you get isotopic enrichment of EVERY fission fragment FOR FREE! Really, that seems like the holy grail (and a containment challenge, but an easily solvable one) for a hospital: you can get plenty of gamma sources of ANY intensity for gamma knife or gamma-isotopic therapy, separate out the pure alpha and beta isotopes for therapies that can be delivered by pneumatic tube, so if you've got an hour's half-life that isn't a deal-killer (because the sucker is in the basement), and any beamline you WANT, including neutron, is readily available!For those not familiar:http://www.rbsp.info/rbs/PDF/aiaa05.pdfhttps://www.nasa.gov/pdf/718391main_Werka_2011_PhI_FFRE.pdfhttp://www.rbsp.info/rbs/PDF/nets16b-ppt.pdf
It's been very interesting to watch the way the ability to tailor the emitance and absorbance spectrums of radiation to and from surfaces has improved. This will always be a trade off between extracting the heat (from the core) and simple reflecting it back (mirror surfaces tuned to one wavelength can be >99% reflecting at that frequency).
On a side topic. I was thinking about how fast the human race gets around it's own back yard IE the Solar System. It took New Horizons 9.5years to reach Pluto.The whole Solar System is roughly 11 light hours across. Not even half a light day. A drive capable of reaching 1/1000 the speed of light would be an enormous improvement. It would put Pluto on an 8 month flight (point to point, ignoring orbits). I think the bottom line is that there is a lot of room for improvement before we get into the realms of "breakthrough physics," "warp drives" etc.
I wonder... is there anything you can think of along these lines to replace the quartz in a nuclear lightbulb? The silicon mirrors are obviously too thermally limited, but... it's always been a favorite design of mine, despite the incredible mass of the system.
I couldn't agree more. Unfortunately, the vast majority of systems with even 1/10th of that capability are so far down the TRL spectrum that it's rather depressing, and often only due to readioillogical fears preventing testing. I've been lucky enough to make friends with a couple people who were students looking at vapor core and open cycle gas core NTRs, and they came away from their programs tearing their hair out at how many problems would be solved if only there were a way to do extensive criticality and containment experiments - something well within the capabilities of a modestly expanded NCERC. Both of us know that's not going to happen anytime soon, though, and unfortunately if it was propulsion systems should probably be well down on the list - Gen IV reactors need a TON of work to be ready for deployment, and they probably should have priority.My favorite near-term design is the LARS out of Brookhaven, but you want to talk about a testing nightmare to qualify the fuel elements: https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19910012832.pdf Establishing criticality thresholds, bootstrapping methods, and the like would be unlike most things we've ever looked at in any depth, but it seems like such an elegant system (and at 1900 s isp, a game changer), with very little outside the fuel elements themselves that would need extensive testing.Either that or Borowski's LANTR: https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20170009140.pdfTRITON is also attractive for near-term, but balance-of-plant issues seem like they would be a major challenge: http://alternatewars.com/BBOW/Space_Engines/AIAA-2004-3863_TRITON.pdfAny of these designs would open up the inner Solar System out to the Main Belt for... well, whatever we want to do. Getting past the Belt gets harder in any near-term system, though... at least in any timeframe that humans would want to endure without artificial gravity and a quite extensive ship.
{snip}The problem with all nuclear systems is the attitude that failure is not an option. The nuclear equivalent of SpaceX's first 3 F1 launches exploding (reactor designs melting down?) and the CEO saying "Well, we learned a lot more this time round and our next reactor won't explode" is (literally) unimaginable IRL. Throw in the desperately poor T/W of current NTR designs to date and it look like you're in a very nasty crevice in the optimization landscape, difficult to improve without taking risks, difficult to take risks (even by other industries standards, quite small ones) in the first place. This is what makes the Kilopower's team of getting to a live reactor test for 10s of $m such an extraordinary feat of both management and engineering skills, as they negotiated their way through the massive H&S issues around modern nuclear testing.
Quote from: BeyondNERVA on 05/13/2018 04:30 amSorry it took me so long to respond... 1000+ things going on right now.Not a problem. Quote from: BeyondNERVA linkYes, it's talked about in terms of spaceflight, but... it's got a non-Carnot-based PCS by default as its' primary PCS, and as a side effect, you get isotopic enrichment of EVERY fission fragment FOR FREE! Really, that seems like the holy grail (and a containment challenge, but an easily solvable one) for a hospital: you can get plenty of gamma sources of ANY intensity for gamma knife or gamma-isotopic therapy, separate out the pure alpha and beta isotopes for therapies that can be delivered by pneumatic tube, so if you've got an hour's half-life that isn't a deal-killer (because the sucker is in the basement), and any beamline you WANT, including neutron, is readily available!
Sorry it took me so long to respond... 1000+ things going on right now.
Yes, it's talked about in terms of spaceflight, but... it's got a non-Carnot-based PCS by default as its' primary PCS, and as a side effect, you get isotopic enrichment of EVERY fission fragment FOR FREE! Really, that seems like the holy grail (and a containment challenge, but an easily solvable one) for a hospital: you can get plenty of gamma sources of ANY intensity for gamma knife or gamma-isotopic therapy, separate out the pure alpha and beta isotopes for therapies that can be delivered by pneumatic tube, so if you've got an hour's half-life that isn't a deal-killer (because the sucker is in the basement), and any beamline you WANT, including neutron, is readily available!
The Kilopower series appears the be designed to produce electricity such as nuclear electric propulsion. Can it be used in a nuclear thermal system?
Quote from: ThomasGadd on 05/15/2018 03:49 pmThe Kilopower series appears the be designed to produce electricity such as nuclear electric propulsion. Can it be used in a nuclear thermal system?Yes, but the temperatures are so low, and the ISP is so low due to that, and the power output is so low that it is somewhat better than an ion engine in terms of thrust, and at best on a par with hypergolics.If you can come up with a scenario where for example you have free water, can't electrolyse it for some reason, and need a few hundred m/s delta-v over a week or so, it may have a point.Outer planet comet hopper?Otherwise you need _lots_ more power for meaningful thermal rockets, as well as for them to get a whole lot hotter.