Talk:Pebble-bed reactor

"Written like an advert"

The company in South Africa that is developing the PBR always appear to gloss over the safety concerns in an attempt to minimize them in the eyes of the public. I personally think this technology is viable, but when playing with nuclear power, one shouldn't ever downplay the risks.

There is a strong likelihood that this Wikipedia article is being modified by a public relations company, or an expert in PR. —Preceding unsigned comment added by ObseloV (talk • contribs) 04:57, 21 January 2009 (UTC)

Chicken McNuggets

For anyone who's wondering about the change in emphasis between the old reactors and the PBR designs, the PBR is using the "Chicken McNugget" business model, you put all your high technology investment up-front into the material manufacturing (the "pebble factory"), and then you can set up cheap franchised outlets wherever you want, taking delivery of the high-tech pebbles, bunging them into cheap reactors, and serving up hot and tasty electricity using staff who don't have to be geniuses, and who don't get the chance to do anything potentially dangerous.

Someone who used to design structural steelwork for UK reactors once said that they were designed like cathedrals: every one was different. It's like the restaurant business: you can have a single restaurant with specialist equipment and staff, all working together to create a great product and avoid food-poisoning disasters, but when it comes to franchising it, you have problems: You can't easily duplicate key staff, monitoring quality control through the chain is difficult (so you get stupid accidents) and you don't get bulk deals on the key equipment. You may get an economy of scale regarding know-how, but not for staff or hardware, because you aren't intending to build that many restaurants, and all your equipment is still very specialised.

The alternative is the McDonalds approach: to redesign the entire business model around mass production and franchising. Design once, build many (but spend a lot of money on the design). MacDonalds is the home of the Chicken McNugget: It takes a product that is variable and messy, and needs to be handled with care (chicken) and it turns it into a centrally-produced, encapsulated, idiot-proof format. You move all your high-tech staff and plant and specialist equipment into a central location where you use computers and robots and liquid nitrogen and god knows what else to churn out "nuggets", then you ship the nuggets to your local franchises, where the cheap staff dunk them in hot oil and serve them. The local staff never get to touch a knife or a piece of raw chicken themselves.

With cheap local staff and simple local plant, and there's a limit to how badly they can mess things up, because their job has already been made as simple as possible. You try to think of everything beforehand and streamline the local processes to be as close to idiot-proof as possible. The guys with the tough job are the ones back at the nugget factory (or the pebble factory) -- the cost and consistency of what happens there dictates the viability of the entire chain. If the mass-produced designs are right, and the product is designed well, the system runs smoothly and efficiently. If there's a problem at the nugget plant, it affects the whole chain. So it's critically important to get everything right (or nearly right) before you take the thing into mass-production.

This is why we haven't already had a big PBR roll-out: they want to hold off and make sure that they get it right, and then if a few pilot plants show that the model works, and there aren't any further cost or reliability issues with the pebbles ... poom ... the things will be everywhere. But the integrity, durability, reliability, manufacturing safety and cost of the pebbles themselves is critical. ErkDemon 16:22, 22 May 2007 (UTC)

A good essay. It leaves a few things out:

• The standardisation of reactor designs isn't just a PBMR phenomenon, all reactor designers are doing it. It's part of the maturing of the industry, which may be about to put the "green" hiccup in the past and enter the boom that everyone in the 1950s thought was only 20 years away. I said may.

• The pebble technology has applications in other reactor designs. If I were directing research, I'd be very interested in investigating a (slightly) higher-temperature Magnox-style plant using the cladding techniques that are involved in the pebbles... and looking at that sort of thing is very much part of the generation IV reactor project. The AGR would have been a world-beater if their original intention of using beryllium-based cladding had worked... but it didn't.

• There are a number of sleeping problems political still to be encountered... the traditional ones of course are bombs, wastes and accidents. All three have the potential to scuttle the PBMR in favour of other designs (remember we are talking here about politics, not engineering):
  • Bombs - the pebble-bed concept lends itself to online refuelling, and even if the fuel pellets are reprocessing-proof, what's to stop you slipping in some fertile pellets that aren't? Made for it.

Because of online refuelling, you can keep the excess reactivity low. So, if someone is adding extra fertile pellets in, they will soak up noticeable amounts of neutrons, which will increase the amount of fuel used, so producing plutonium will either be slow or will be noticed. Jrincayc 01:52, 29 May 2007 (UTC)

That's the argument that is generally used. There are three problems with it. Firstly, it's a technical argument. Even if valid, frankly these subtleties have generally been lost on the antis. Secondly, it ignores the possibility that the operators will not care that they're being noticed, as happened with India. Thirdly and most important IMO, the point is not so much that the proliferation risk of a PBMR is all that high, it's rather that the risk however small is far greater than that posed by (say) a PWR or BWR - types which have both been resisted on the grounds of proliferation risk (however foolishly) - which gets us back to the politics. Our article on Nuclear fuel currently reads Reprocessing of spent commercial-reactor nuclear fuel has not been permitted in the United States due to nonproliferation considerations. That is high-burnup spent PWR and BWR fuel we are talking about there! Andrewa Andrewa 01:07, 31 May 2007 (UTC)

• • Wastes - the PBMR concept not only abandons recycling of uranium and transuranics, it also throws away the moderator. First proposed nuclear fuel cycle in history to produce more nuclear waste per megawatt-hour than a once-through Westinghouse PWR.
  • Accidents - Graphite-moderated reactors quite simply have the worst accident record of any moderator type, so far. Windscale, Chernobyl. TMI was a hiccup by comparison.

But I like your analysis. Interesting times. Andrewa 01:58, 26 May 2007 (UTC)

Though interesting, this is supposed to be a talk page about the article, not a discussion forum about the topic. There are lots of forums out there, only one Wikipedia though. Pro crast in a tor 10:06, 29 May 2007 (UTC)

Sorry, Wikipedia is not that special. theanphibian 19:16, 29 May 2007 (UTC)

I think that's a valid point... but one of the special things about Wikipedia is that we do have articles on current events and controversial topics, and we try to be NPOV even there. And it's a challenge at times! Where to draw the line? I think this essay is helpful, and suitable for a talk page, because I think it will help us to improve the article. But, if that's not the case, then you're quite right, it doesn't belong here. There are lots of interesting things that don't belong here. The acid test is simple: Will it help us to improve the articles? Andrewa 00:35, 31 May 2007 (UTC)

The topic was meant to prompt a few thoughts or discussions that might be useful to the development of the main page. But it ended up longer than intended, and if it isn't considered to be sufficiently useful I've no objection to it being deleted. All is happy! ErkDemon 00:33, 12 June 2007 (UTC)

As I said above, it will probably be archived eventually. Even if not, it will probably remain in the history.

Another interesting question that it raises... is the McNugget approach successful? In the 1970s I shared a house with other undergraduate university students, and one was working as a cook in a fast food chain. He would never eat there, he looked rather green when any of the rest of us did, and occasionally told stories which don't bear repeating to try to convince us not to either. The problem seeemd to be that the job was so simple, there was no sense of responsibility for the product. Some chains now appear to be quite intentionally working to correct this, to introduce some responsibility and accountability and even pride into jobs at all levels.

On the other extreme, one of the common strands to several nuclear reactor accidents (in particular SL-1 and Chernobyl) is that the operators had been told and believed that they couldn't break the reactor. The Soviets had been saying since the mid-1950s that their control systems were foolproof; This was their justification for building reactors of a type that the rest of the world thought dangerously unstable, and they were very proud of the achievement. And they'd evidently communicated this to the management and staff of Chernobyl NPP to the point that nobody was worried about conducting an experiment in the absence of anyone trained in nuclear engineering. The collegues of the three victims of the SL-1 accident have replied to suggestions that one of the victims might have deliberately lifted the control rod assembly in order to commit suicide by pointing out that in their training they were repeatedly told how safe the whole rig was, and that they doubted that any of them knew of the danger that this action posed. Presumably, the same helio belief in the inherent safety of SL-1 was behind the (in hindsight appalling) decision to have these people unbolting and manually operating the control rod without having a nuclear engineer present to supervise.

It's impossible to make anything foolproof, because fools are so ingenious. Andrewa 02:54, 3 July 2007 (UTC)

Firstly, the safety of HTRs is impeccable. Their safety has even been tested several times with accident conditions were ANY other nuclear reactor would have had a meltdown. Full control rod pull-out coupled with a complete shutdown of all cooling has been demonstrated several times in the AVR and the HT-10. HTRs are safe because no operator action is required to prevent the release of radioactive fission products.

There are two hypothetical accident cases situated around the inability to switch off the blowers. In these cases the staff has several hours time to switch off the blowers. That's why the ONLY safety grade equipment within the entire reactor are the switch-off controls for the blowers. And even then, there are multiple ways to switch something off, you know. Just get a fire axe and whack the power cable. You've got several hours time to do that...

Got a leak? No problem. Get a mechanic and weld a piece of metal above the hole once the core is depressurized to minimize future damage from air ingress. BTW, constructing a dangerous case wrt. to air ingress is more difficult than you think, it doesn't just happen with a hole...

The safety of the HTR is that if even only a fraction of the equipment works, and a severe accident happens, the staff can first have a lunch, then go down to the pub for a drink and then afterwards go on weekend trip, and NOTHING would happen. Sure the reactor would be scrap, but no release of fission products would occur. The only way that the core can overheat, and cause about one in a million fuel particles to break, resolves around the inability to switch the blower off during extremely severe accidents. And I am talking here about a terrorist-style-bombing-of-selected-components-extremely-severe accident...

Secondly about fuel waste. HTRs need much, much less fuel due to their insane burn-ups AND the uniformity of that burn-up within the spent fuel. Theoretically, one could mill the spent fuel and then do the usual reprocessing afterwards, but due to the extremely high burn-ups of HTRs this is highly uneconomical. Burn-ups around 200GWd/tHM in HTRs btw. translates to about 70-80% of all energy being created by fissioning breeded Plutonium. There is a reason why the US chose this reactor design to destroy Russian Plutonium: its simply insanely good in utilizing nuclear fuel. --Dio1982 (talk) 19:42, 31 January 2008 (UTC)

I just want to say, this is by far, one of the most interesting "article" I had read on wikipedia. It is quite POV, I agree, but man does it provide insight. I guess that's the problem with POV, what some consider to be insightful would be considered to be total BS by others. 129.173.242.237 19:49, 5 November 2007 (UTC)

Is Doppler Broadening a Unique Safety Feature?

At the top of the safety feature list is the doppler broadening effect. They way this effect is explained, it should happen anytime you have hot U238 in the fuel. This would seem to apply to almost any other reactor design. Why is it touted as particularly important in PBR's

Sign yourself, and it is an unique feature. Of course, all reactors observe the effect, but the relative size of their's inner core and density make the effect irrelevent. However since radioactive material is in such smalle quantity per pebble, the effect become relevent. The natural geometry and design of the Pebble Bed Reactor makes it an unique feature. 129.173.242.237 20:32, 5 November 2007 (UTC)

This has a lot to do with the core physics.

The doppler broadening effect is significant in any reactor. It is the prime means, besides of negative feedback from steam voids, to prevent an overheating of the core if excess reactivity is accidentally inserted into the core by a loss of boron or accidential control rod removal. Doppler broadening is also the prime reason why the core will shut down during severe accidents where the shit is truly hitting the fan (think Chernobyl bad). BUT, the effect is relatively small since any water cooled reactor will not deviate a lot from its steady state temperature, hence the doppler broadening effect is very small. "Active" measures of shut down control rods and/or Boron control is necessary to kill the fission reaction during more "normal" accidents.

By definition HTRs may not rely on any active measures to remain safe. If you look at the available physical processes that would "eat up" neutrons if the reactor becomes too hot, there are not many options. To be honest there are only two big ones: moderator steam voids and doppler broadening. Since HTRs are gas cooled, that only leaves you with doppler broadening.

And in order to get any meaningfull doppler broadening effects, you need huge temperature variations here. We are talking about the +600°C range here. What happens at these high temperatures, is that some slow thermal neutrons get bumped by 1200°C carbon moderator atoms into a much higher energy state (backscattering). It just so happens that this higher energy state is situated just around the first absorption harmonic of U238. This actually adds a tremendous 100pcc of negative reactivity to the core.

The neat thing about HTRs relying completely on doppler broadening to shut down is that you can do instantaneous load following without any external inputs. The core is held at its equilibrium temperature no matter what. If the coolant gases enter the core too hot, the core will produce less power, if they enter too cold, it'll produce more power make up the difference.

The downside is that you can't really completely switch the reactor off. Even at shutdown, the reactor will produce some minor amount of fission energy (~1-2MW of thermal power) to keep the core at a lower equilibrium temperature. This is a huge problem for people designing these reactors to get license approval.

Safety wise this is btw, not a problem. The reactor vessel walls will be just too hot to touch, nothing else. But the reactor would need months to cool down to room temperature, unless the core is evacuated which takes about a month on its own. --Dio1982 (talk) 17:34, 11 March 2008 (UTC)

Reference 9 is missing

I just tried to access reference 9 but it appears to have 404d... 87.112.74.244 (talk) 11:46, 3 January 2008 (UTC)

Citation needed on public radiation release

In section Pebble_bed_reactor#Safety_features, it says:

Later problems with the AVR reactor resulted in a small release of radiation to the public.

I really think we need a citation for this event to clarify just how safe or unsafe these reactors are. I would also like to see information on exactly how "small" it was, if that information can be found.

The best reference I found in Googling is at the end of a page on how pebbles circulate within the reactor, but it is still very vague, and not a very authoritative source. Another potential reference is Reactor Watchdog Project's info page on PBMR, which has more detail, but also comes from a site with an obvious anti-nuclear spin.

It is also mentioned in THTR-300#Decommissioning, but I don't see a citation there either.

Different sources identify the event as happening in 1986, 1988, or May 4, 1985.

       Correct is May 4, 1986  —Preceding unsigned comment added by 80.137.39.188 (talk) 10:00, 14 October 2010 (UTC) 

LeBleu (talk) 00:44, 4 February 2008 (UTC)

The NIRS is not a credible source. I have removed the link. —Preceding unsigned comment added by 70.162.2.55 (talk) 00:09, 17 November 2008 (UTC)

Also, it is very difficult to get exact information on these accidents, since most of them have to do with national security interests. 128.196.210.75 (talk) 22:35, 17 November 2008 (UTC)

Check Dr. Moormann's paper http://hdl.handle.net/2128/3136 Jrincayc (talk) 13:16, 18 November 2008 (UTC)

For the AVR, the best resource is probably "Decommissioning of the AVR Experimental Nuclear Power Plant, Disassembly of Plant Components and Entombment" by Arbeitsgemeinschaft Versuchsreaktor AVR in Nov 1991, available on the NEA cdrom http://www.nea.fr/abs/html/nea-1739.html. In it it states that the releases off site were less than 1 μSv/a from the exhaust air, 3.5 mSv/a at the edge of the fence when the plant was operating, and 0.42 mSv/a at the edge of the fence at worst when the plant was shutdown (due to some stored radioactive waste). Jrincayc (talk) 04:51, 27 November 2008 (UTC)

The initial quotation was incorrect. The AVR had no problems. It was the THTR that had the pebble problem. I would imagine there's a decommissioning white paper on the THTR out there that discusses the incident, but I cannot find anything in English. 70.162.2.55 (talk) 06:36, 3 December 2008 (UTC)

Neolithic nuclear reactors?

Articles like this give me the impression that potentially, a shaman would need merely mix a powdered mineral containing unenriched uranium with a pure lampblack in order to make an ever-burning fire. (Though some ill might have befallen some villagers burned with it, he can scarcely be blamed for the judgments of the gods) Is it plausible for people to have made reactors this way in Neolithic times? Wnt (talk) 22:06, 7 August 2008 (UTC)

Rather implausible. You might be able to do it with natural uranium, but you would need pure uranium, not uranium ore. See Chicago Pile-1 and Natural nuclear fission reactor Jrincayc (talk) 04:02, 17 November 2008 (UTC)

burial at sea to disrupt thermoclines

A pile of "spent" pebbles will still get hot, right? It seems like if the issues with cesium and strontium leaking out could be resolved, one could dump them in a sea with circulation issues to circulate the water by heating the depths. 65.46.169.246 (talk) 15:07, 16 March 2011 (UTC)

Dimensions of the pebbles

What are the typical dimensions of the pebbles used in such reactors? This should be included in the text of the Design section, as now it seems one can only roughly deduce their size by looking at one of the pictures. —Centrx→talk • 05:21, 18 November 2008 (UTC)

6 cm in diameter is what AVR, HTR-10 and THTR used. Other sizes have been discussed (Someone from MIT put out a paper on this, but I don't remember the exact reference Kadak and others 2004ish maybe?) The pdf you can get from http://hdl.handle.net/2128/3136 has the pebble dimensions on page 2 for AVR. Jrincayc (talk) 03:39, 20 November 2008 (UTC)

The pebble diameter for the AVR, THTR, HTR-10 as well as for the planned PBMR in South Africa and the planned HTR-PM in China are all 6cm. There is a lot of experience in manufacturing and handling as well as experimental data about the 6cm diameter pebbles due to the AVR and THTR program. There isn't really any benefit to changing this diameter in terms of reactor physics and thermohydraulics, but it would cause a significant amount of work in getting new reactors designed and licensed. --129.69.25.217 (talk) 12:36, 20 January 2009 (UTC)

Textual error?

Section "Safety Features", final paragraph contains the text "The graphite is the moderator for the reactor, and are strong containment vessels", which defies grammar and seems meaningless. —Preceding unsigned comment added by 212.159.59.5 (talk) 18:33, 27 February 2009 (UTC)

Cleaning up citations

There's a scattering of [citation needed] markers -- any chemists out there who can assist? E.g.:

• in 'Containment', this might be appropriate for "Pyrolytic carbon can burn in air when the reaction is catalyzed by a hydroxyl radical (e.g. from water)". Is there a Chemistry Wikipedia group that could help? ThomasNichols (talk) 11:46, 6 October 2009 (UTC)

pebble leakage

This paragraph ...

"A pebble-bed reactor thus can have all of its supporting machinery fail, and the reactor will not crack, melt, explode or spew hazardous wastes. It simply goes up to a designed "idle" temperature, and stays there. In that state, the reactor vessel radiates heat, but the vessel and fuel spheres remain intact and undamaged. The machinery can be repaired or the fuel can be removed. These safety features were tested (and filmed) with the German AVR reactor.[10]. All the control rods were removed, and the coolant flow was halted. Afterward, the fuel balls were sampled and examined for damage and there was none."

... is in contrast to ...

"Darüber hinaus wurde im Jahre 2008 ein Bericht von Rainer Moormann, Mitarbeiter des FZJ, veröffentlicht, in dem die übermäßig starke radioaktive Kontamination des Reaktors auf eine unzureichende Überwachung des Reaktorkerns sowie einen länger andauernden Betrieb bei unzulässig hohen Temperaturen zurückzuführen ist. Dies habe u. a. dazu geführt, dass Spaltprodukte aus den Graphitkugeln austreten konnten."

... from http://de.wikipedia.org/wiki/AVR_%28J%C3%BClich%29 , which translates to ...

Furthermore in 2008 Rainer Moormann, who works at the FZJ, published a report where the overly high contamination of the reactor was caused by inadequate monitoring of the reactor core as well as by lengthy operations at inadmissively high temperatures. This has among others led to the release of fission products from the graphite balls.

Is Rainer Moormann's report correct? Did the pebbles leak due to excess temperature? What is excess temperature? I thought the reactor can go up to idle temperature which is not excessive and safe. Or was the leakage caused by additionally leakage of water into the reactor? —Preceding unsigned comment added by Darsie42 (talk • contribs) 15:31, 18 July 2010 (UTC)

Reference 6 missing

Reference 6 returns a 404 error mostly in German. Some sleuthing may be required to find the correct new link. —Preceding unsigned comment added by 65.160.229.33 (talk) 17:01, 14 March 2011 (UTC)