Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Portable nuclear plants

  1. Nov 9, 2008 #1

    jal

    User Avatar

    I found the following
    note: Moderators
    I would like to know the opinions of experts from here before you move this thread to another forum.

    Is this all hyp?
    http://www.hyperionpowergeneration.com/
    Home page of Hyperion power generation
    I have just read the thread Small nuclear power plants. that showed up after posting this thread.
    jal
     
    Last edited: Nov 9, 2008
  2. jcsd
  3. Nov 9, 2008 #2

    Astronuc

    User Avatar

    Staff: Mentor

    It's not hype. Hyperion and others are serious about small, compact and self-contained nuclear systems. This is a commercial approach to what the Navy does for nuclear reactors in submarines and surface ships - more or less - in the sense that the core is not refueled periodically, e.g. annual, sesquiannual, or biannual basis. Instead the core operates for a life-time without refueling, after which it is removed and replaced. Traditionally NPP's have been designed for 40 years, with the initial capital costs retired by that time. Now plants are having their lives extended to 60 yrs and there is a new effort to push 80 years, although O&M costs will likely increase, or new capital costs incurred as some large components are replaced.
     
  4. Jul 15, 2009 #3

    mheslep

    User Avatar
    Gold Member

    A search turned up this thread, and I see recently Hyperion identified a buyer for unit, so I'll append here, as I have a feasibility interest for portable reactors.

    Apparently the Hyperion unit design is 25-30 MW(e), 70MW(t). My question: how is it possible to effectively reject the 40-45 MW(t) of heat with out a large water source, or absent that an extremely large heat exchanger. A portable reactor can have neither.
    Edit: This design uses a Brayton cycle.

    http://technorati.com/posts/HK3w8tp1oLKudb99ak4pTORJxp%2Ba_rsUr_0TPMtmQQs%3D
     
    Last edited: Jul 15, 2009
  5. Jul 15, 2009 #4

    Astronuc

    User Avatar

    Staff: Mentor

    They could possibly do a Brayton cycle, or a combined Brayton-Rankine cycle.

    And perhaps they plant to sell the rejected heat as 'process heat' or district heating.


    Some nuclear and conventional plants use dry forced-convection cooling systems.

    Using geothermal heat sink may be an alternative.
     
  6. Jul 15, 2009 #5

    mheslep

    User Avatar
    Gold Member

    Yes Wiki says Hyperion is using a Brayton cycle w/ heat pipes and high temperature UH3 gas. Wiki doesn't specify the reference for that fact; I gather its all in the patent.

    Edit: From World Nuclear
     
    Last edited: Jul 15, 2009
  7. Jul 16, 2009 #6

    Astronuc

    User Avatar

    Staff: Mentor

    The Wikipedia article and the press surely provide lofty praise and expectations for the Hyerpion modular power unit, but I'd imagine that it's going to take a couple of years to get through the NRC process.

    I'd be interested in testing of the UH3 concept in their core configuration. It sounds like there is a hydrogen feed and bleed system (H2, D2), which also must deal with tritium. I am also curious how they handle the noble radioisotopes of Xe and Kr, and the volatiles such as Cs, I and Br. And I have to wonder about Xe-transients/swings.
     
  8. Jul 16, 2009 #7

    mheslep

    User Avatar
    Gold Member

    Yes here's NRC's web status on them
    http://www.nrc.gov/reactors/advanced/hyperion.html
    I saw elsewhere a public NRC comment that was disappointing. It read, paraphrasing, that, well, the Hyperion design was unusual, different, so they were never going to touch it. So I expect that their sales will all be international, or to the military.
     
  9. Jul 16, 2009 #8

    Astronuc

    User Avatar

    Staff: Mentor

    To pass muster with the NRC, I would expect that Hyperion would have to build a demonstrable module, and put in a reactor like ATR to demonstrate the concept. It is radially different than current reactor designs.

    The NRC has difficulty with just the conventional Gen 3+ LWRs, so I imagine they'd be reluctant to take on the Hyperion UH3 system at the moment.
     
  10. Jul 16, 2009 #9

    mheslep

    User Avatar
    Gold Member

    Sorry I don't follow the tritium source. Aside from the standard in 0.01% tritium production in general fission, how does the introduction of H2 or D2 increase that in any way? Same goes for the other radionulclides. Why would you expect they would be handled in a different waste process than is used with removal in PWR?
     
  11. Jul 16, 2009 #10

    Astronuc

    User Avatar

    Staff: Mentor

    d + n => t.

    In conventional LWR fuel, the Xe, Kr, Cs, I mostly accumulate in the UO2 matrix. Some of the gases and volatiles (~1% - ~20% under normal operation) find their way to the void volume. The void volume and fuel are surrounded by cladding, which ideally retains its hermiticity.

    The UH3 can volatize, and the H can react with other metals like Zr (also a fission and decay product). Part of the NRC's requirement would be a demonstration of the fuel behavior to maximum design burnup, and a demonstration that Hyperion scientists/engineers can predict the thermo-mechanical and neutronic/reactivity behavior of the UH3 fuel system.
     
  12. Jul 16, 2009 #11

    mheslep

    User Avatar
    Gold Member

    Arg, yes of course, missed that, though the D cross section is small.

    Well I believe the idea is that since they have need no control rod mechanicals, their containment mechanicals are much simplified.
     
  13. Jul 16, 2009 #12

    Astronuc

    User Avatar

    Staff: Mentor

    I was just talking with a colleague about the NRC's thoughts on the Hyperion system. Apparently, Hyperion has talked with the NRC, but has only provide 'design on paper' details. The NRC is not going to review it unless the folks at Hyperion provide experimental data, and demonstrate that they can simulate the physics of the reactor and fuel.

    Hyperion, like the Toshiba4S and BWXT's design, are 'low capacity' systems. Given that the NRC's experience is LWR technology, they are not about to seriously considered alternative systems without a lot of experimental and analytical work. The NRC would also need to hire staff members with experience in such technology (that may be a kind of Catch 22).

    Apparently there is also an ANL/LLNL design.
     
  14. Jul 16, 2009 #13

    mheslep

    User Avatar
    Gold Member

    Yes the NRC states as much online.

    That's the trap: The NRC doesn't want to be forced to look at this technology. The best face on this is that it puts NRC out there where they have to do more work to cover themselves politically, and can't simply point to years or PWR data to show due diligence. <speculation>The nastier side is that they draw a lot of staff from Big Power (Westinghouse and the operators), have grown comfortable with the multimillion dollar fees they can charge Big Power for approval, and are just overly cosy with them. So to cover for all this, they demand expensive documentation with the sub-text that they won't act on it even if submitted, i.e., go away, and Big Power happens to be just fine with that too</speculation>.
     
  15. Jul 17, 2009 #14
    Well what else would we expect? That NRC provide the experimental data and NRC develop calculational methodologies? Remember, everyone got their wish back in the 1970's and the NRC was split from AEC - to ensure that the regulators were not synonymous with the promoters. It is not the NRC's job/function to develop new technology.

    And as far as 'big power = W' , I suspect Westinghouse isn't involved because they don't develop products for markets that don't exist yet. hmmm that's not very clear, is it? WEC's customers (the power companies) are not interested in anything that isn't licensed. And if the customers aren't interested, why would WEC be interested? Another Catch-22.

    Also, the power companies are looking at 1100 MWe plants as "too small" so I don't know who is going to buy one of these little things. Finally, W and Areva make their money by selling the reactor fuel, so a device that requires refueling every 10 years isn't high on their list of things to look into. The naval reactors are a different story, 'big power' was interested in them to (1) get into the technology, and (2) bragging rights/national pride in the 1950s - 60s.

    Sorry if this has a ranting tone, it wasn't meant to be
     
  16. Jul 17, 2009 #15

    Astronuc

    User Avatar

    Staff: Mentor

    Um no. I was responding the hype by Hyperion. Yes - they've talked to the NRC. But they only have a design on paper. The NRC will not evaluate it until there is experimental evidence that the concept works - i.e. it exhibits predictable behavior, particular with respect to reactivity control. It is ludicrous for Hyperion to expect otherwise.

    The AEC did developmental research, and DOE still does, e.g. Gen IV. The Hyperion concept comes out of LANL. Does LANL have experimental data? If so, then Hyperion would have to submit those to the NRC before being seriously considered.

    As for 1100 MWe being too small, various US utilities have selected the AP1000 (~3415 MWt/1200 MWe (gross)/1115 MWe (net)), as opposed to the EPR (4590 MWt / 1700 MWe) or APWR (4450 MWt / 1700 MWe). Some utilities are concerned about putting all their 'eggs in one basket' at a given station, although most Gen3+ plants are likely to be built at existing sites.

    The ABWR is rated 3926 MWt / 1350 MWe (gross) and the ESBWR is projected to provide 4500 MWt / 1590 MWe (gross).

    So far US utilities seem to favor PWR technology.
     
  17. Jul 17, 2009 #16

    mheslep

    User Avatar
    Gold Member

    Nor is it their job to impede new technology .

    Who says they're 'too small'? I think you have that exactly backwards. What they are is too expensive at $4 to $8 per W(e) installed. Hyperion is quoting their 25MW(e) reactor at $1/W(e). The question is more likely who would bye an AP1000? So far no one in the US.

    Well then thankfully we have the like of Hyperion, Babcock and Wilcox, etc, to address this market.
     
  18. Jul 17, 2009 #17
    In this context, their job is to review and approve (or not) license applications. The job of the applicant is to provide a complete application. That seems pretty clear to me; I don't think you can blame the NRC for lack of progress by Hyperion, or LANL, or DOE, or anyone else.

    That's a good point.

    Can you help me understand this? I can't figure out what is included in that cost. What's confusing me is the combination of "MWe," "brayton cycle," and "no moving parts." How do they propose to convert the heat into electric power w/o moving parts? Or, are the moving parts an additional cost (for those customers who want MWe no MWth)? I haven't seen this explained in the Hyperion website.
     
  19. Jul 18, 2009 #18

    Astronuc

    User Avatar

    Staff: Mentor

    This is a key point. Does the cost projected by Hyperion cover just the modular unit? If so then what does the rest of the plant cost? It will have to be secure!

    A 25MW(e) reactor will cost $25M at $1/W(e), but will that require a $25 million facility or a $100 M facility or more in which to locate it?

    And what happens with the back end? Will Hyperion take back the unit? Where will it be stored? Will it be reprocessed? If so, where? What about ultimate disposal? What about liability?

    And Hyperion still has to demonstrate that the technology will perform as claimed, and reliably for the design lifetime of 5 years or so.
     
  20. Jul 18, 2009 #19

    mheslep

    User Avatar
    Gold Member

    I had not seen the no moving parts description, but if true that must apply only to the reactor chamber, emphasizing that the design uses no mechanical control rods.
     
  21. Jul 18, 2009 #20

    mheslep

    User Avatar
    Gold Member

    Almost the cost is just the modular unit, as site costs will be variable and Hyperion could not quote them.

    The simple graphics shown so far indicate they believe a simple hole in the ground with some concrete lining would suffice. No doubt thats true for nominal operation, but then what about seismic events, the protection against which drive costs so high elsewhere? I venture they would say there reactor is inherently fail safe - if the reactor is damaged the hydrogen escapes and all reaction except the normal decay just stops.

    This they do address to some degree: refueling is accomplished by shipping back some kind of core to the factory.

    As this technology came out of Los Alamos, there must have been some prototyping done there.

    It occurs to me a small portable low cost portable reactor such as this with siting complications would be idea for one customer in particular: the Army/Airforce. They looked at small reactors decades ago (Army Nuclear Power Program) and decided the state of the art at the time was too much trouble. But now this, a $1/W(e) reactor that could be flown in on one C-130 and installed at an ~ airbase in a hole dug in a couple days, and continue to run without refueling for five years? And, the main hurdle for which, the NRC, does not apply to them.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Portable nuclear plants
Loading...