Post by Roger Clifton on Oct 19, 2016 16:47:15 GMT 9.5
Rolls-Royce, following 50 years of experience designing and building submarine PWR's, has announced its intention to set up factory production of SMRs in the UK. The link shows an image of a SMR being delivered as a single truckhaul.
They see a base market for SMRs in the UK of 7 GW, amounting to 17 to 35 modules. They foresee a further 9 GW elsewhere, presumably in Europe. Just the base market falls into the category of "factory production", and imminently "mass production" as the market heats up.
Post by Roger Clifton on Aug 17, 2017 15:56:30 GMT 9.5
Rolls-Royce (UK) has elaborated on its SMR plans with a variation on the convention. The reactor is still small in that the reactor pressure vessel is small enough to be built in a factory, however there are four steam generators, not one, and they are coupled to the outside of the pressure vessel instead of being integrated inside it. The uncrowded core is thus able to achieve a much higher thermal power, so that the four turbine trains can generate a total of 450 MWe.
Presumably the steam generators and the conventional generation train on each of them can be added one by one, allowing the plant operator to generate cash flow before completing the capital commitment of the power station.
Rolls-Royce SMR ... can generate a total of 450 MWe.
Britain is decommissioning 26 Magnox reactors that each delivered typically 300 MWe to the grid. The sites are appropriate for replacement reactors of the RR SMR size, ie 220-440 MWe. They still have grid connections and water supply - perhaps even the same cooling towers - of that capacity. Further, the neighbors are not scared of the so-uneventful nuclear electricity, if they ever were.
"Study Finds Advanced Reactors Will Have Competitive Costs"
"A new study of contemporary nuclear industry cost projections, previously unavailable to the public, provides new insight into a potential path breaking cost trend for the next generation of advanced nuclear plants."
Post by Roger Clifton on Oct 16, 2017 10:07:16 GMT 9.5
Being "advanced" means that they are not fully developed, an essential requirement for any mass rollout of nuclear energy. For non-nuclear countries like Australia, introduction to non-carbon baseload requires us to dip our toes in the water tentatively. Reassurance will come when we find that the newly installed (and more fully developed) SMRs are cheap, plug-and-play, and boringly reliable in performance. Thus silencing the naysayers.
However the advanced reactors are of real interest for the subsequent generation of nuclear energy, as countries move to supply the entirety of their baseload with non-carbon power. For that rollout, we would be especially interested in how well these proposed designs can be mass produced, in factories.
China is way ahead in planning their transition. China's plans for decarbonisation between now and 2050 involve the steady expansion of the slow neutron reactor fleet to 200 GW, with concurrent development of fast neutron (as in "advanced") reactors. At the same time, the reprocessing of used fuel accumulates fissile material. By 2050, the fast neutron reactors are planned to take over the lead. They would initially be fuelled from this stockpile.
Because most SMRs have passive safety features, cost may be the main factor in determining how quickly these reactors are accepted. Figure 3 (Levelized Cost of Electricity for All Participating Companies) of Yurman's article provides some cause for optimism in this regard.
But, of course. a company must get its design certified and build a full-scale working module before making any sales.
Yes--apparently NuScale's has pre-sold its first plant. So my last comment was not quite accurate. I did some digging and found this recent quotation on World Nuclear News:
"The DOE [Department of Energy] in 2015 awarded a three-year cooperative agreement to NuScale and UAMPS [Utah Associated Municipal Power Systems] to conduct site characterisation activities and to prepare documentation that will lead to a COL [construction and operating licence] application for a commercial 12-module NuScale power plant. This is planned to be built on the site of the Idaho National Laboratory. It will be owned by UAMPS and operated by Energy Northwest [which also operates the Columbia Generating Station, the Northwest's only current nuclear plant]."
A quick aside: In the comment above I happened to mention the Columbia Generating Station, the only nuclear power plant in the Northwest (USA). Just west of this plant is Washington state's gravitational wave observatory (LIGO), which a few days ago observed a neutron star merger. So while the power plant consumes uranium, a few miles away LIGO can observe events which create it.
Post by Roger Clifton on Dec 18, 2017 8:31:19 GMT 9.5
The ARC reactor is a plutonium burner/breeder, a scaled-down version of the S-PRISM. Both have a well tested prototype (well, precedent anyway) in the EBR-2 that ran successfully for decades. Both ARC and S-PRISM could be factory-made and potentially mass-produced, however the current supply of plutonium for a mass startup would be limited. Instead, the incoming trend for nuclear reactors is to burn low enriched uranium in the PWR tradition in SMRs. Enrichment facilities can be expanded faster than plutonium breeding for any rollout.
The Chinese plan (and face it, these are the people who are actually getting things done) is for PWRs up to 2050 and fast neutron reactors after that. It will take until 2050 for the PWRs to accumulate enough plutonium to start up the fleet of fast neutron reactors. By the year 2100 their plan would provide ~1 kW of nuclear electricity per person in the country.
I hope for a mass rollout of SMRs to occur worldwide, to give us a means of arresting the worsening climate change. However, stasis would require more than 1 kW/pax as it must also include power for transport, smelting and cement.
Post by Roger Clifton on Dec 21, 2017 8:36:48 GMT 9.5
Construction has begun on China's prototype fast neutron reactor, CFR-600. The documentation describes its fuel as UO2 and MOX, that is, oxide fuel rather than metal fuel. Currently their capacity to separate plutonium is limited, so they would start up on UO2, possibly HEU. However the documentation also describes the reactor as developing a breeding program. High breeding ratios are best achieved with metal fuel. We can expect to hear news of them experimenting with U-Zr-Pu alloys, where the American IFR program met success.
At 600 MWe, CFR-600 does not quite fit into the usual definition of SMRs. However, as it is sodium cooled rather than water cooled, it does not need the large pressure vessel forgings that SMRs promise to avoid. Factory production is clearly intended for its subsequent commercial version, and potentially mass production.
NASA has just tested its prototype Kilopower nuclear reactor.
"The National Aeronautics and Space Administration (NASA) yesterday announced the successful testing of a uranium-fuelled Stirling engine for use in possible future missions to Mars. Testing of the Kilopower reactor - which could be used to provide power to missions to the Moon, Mars and beyond - began at NASA's Nevada National Security Site in November 2017 and was completed in March." www.world-nuclear-news.org/ON-NASA-successfully-tests-kilopower-reactor-0305185.html
One NASA official said that, "Safe, efficient and plentiful energy will be the key to future robotic and human exploration [of space]." And safe, cost-efficient and plentiful energy, much of it also supplied by nuclear, will help shape the destiny of humans on Earth.
Post by Roger Clifton on Jun 5, 2018 10:12:57 GMT 9.5
According to WNN, DoE reports on the vulnerability of all energy infrastructure to cyber attack.
Small modular reactors (SMRs) need to be robust against hacking. Although the passive designs lend themselves to autonomous operation without any connection with the Internet, there are clear advantages in giving remote access to operators and inspectors.
The prospect of mass production of SMRs implies a shortage of trained nuclear operators, as their training cannot be completed at the same speed as an emergency rollout of hardware. The distribution of small reactors around the periphery of the grid implies that an operator watching over several reactors would have to be remote from most or all of them. During production, an autonomous power plant would be sending a steady stream of status reports requiring a human to monitor them. Analysing deviations requires a trained operator, and taking action requires a certain level of remote control.
Conceivably, an antinuclear hacker could trigger a flow of false alarms, forcing the remote operator to make a precautionary shutdown. The system would have to be designed to protect against such interference.
Post by Roger Clifton on Jun 9, 2018 11:07:43 GMT 9.5
I guess we are in times when the nuclear regulator of every country would require a human on site at every nuclear plant, no matter how autonomously it runs. Unfortunately nuclear reactors will eventually be needed to be produced faster than nuclear operators can be trained to staff them. How long will it be before this obstructive requirement is lifted?
Similarly obstructive regulation, the Red flag traffic laws required that every horseless carriage be preceded by a man waving a red flag at walking pace, some 60 yards ahead. Originally intended to warn and advise drivers of horses what to do, they soon became redundant as know-how spread, but distrust lingered and the obstruction continued.
It took 30 years (1896) for this requirement to be repealed. Presumably over that time, the public ceased to believe that the newfangled devices would despatch random passers-by to horrible deaths, explode violently without warning, or that the unnatural forces that propelled the horseless carriages would escape and spread lurking death across the region.
It is now more than thirty years since the Three Mile Island accident (1979) raised fearful public reaction, despite killing nobody at the time. Fearmongers warned that fatal diseases were only yet to emerge, but no one has died so in the 40 years since. Neither have evil-doers snatched the magic fire from the heart of a working reactor. Still reacting, regulators continue require a small army of watchmen on every nuclear site. You would think people had learned by now!
Post by Roger Clifton on Jun 11, 2018 11:34:44 GMT 9.5
With few moving parts, an SMR's nuclear island (the reactor and its primary cooling system) is more likely to be designed to be autonomous. The much busier conventional island (with the secondary steam, turbine, generator and transformer yard) would more likely require manpower. It is tempting to imagine the nuclear technicians watching the reactor remotely, while occasionally directing an on-site fitter to take an instrument or video camera up to this or that location on the nuclear island.
Most of the running cost to the community of Galena, Alaska, for their planned 10 MW "4S" reactor, was for a watchman. The reactor and its steam train was to do almost everything by itself.
Also on the NuScale front, as davidbbenson reports above, the company has found a way to get 20% more power from its modules. What is especially interesting is that this additional power comes at almost no extra cost, so the cost of electricity can fall up to 18%.
Post by Roger Clifton on Jun 24, 2018 19:20:55 GMT 9.5
Can we train operators as fast as we can produce SMRs?
At some point in the future, we can expect decarbonisation to accelerate, with a rollout of mass-produced SMRs implied. Installation times can be expected to shorten with experience. Already, NuScale predicts that they can get a first module producing electricity within 24 months of the completion of earthworks. However such short scales are much shorter than the timescales normally required to train a matching surge of operators.
It is instructive to compare manpower requirements for a classic ~1 GW power plant, of ~6 years construction. In a study by Verma (2012), routine operation of a PWR requires 867 staff, of whom ~400 are technicians. Technicians require three years training, of which one year is on site. Engineers and scientists (amounting to quarter of the staff initially!) require five years training, including one year on site. However, the author implies that the nuclear component of the academic training can be made postgraduate, so that existing engineers and physicists could be upskilled in a year or so. The author also indicates that these numbers could be trimmed back by redesigning the skill structure of the staff community.
An isolated 50 MW SMR would provide one twentieth of the cashflow of a 1000 MW PWR power plant. Proportionate staff numbers would be 20 technicians and half a dozen engineers/physicists. It is hard to see communities of such people being trained and assembled at sites where an expected surge of demand indicated installing a single SMR. So unless remote operation becomes permissible, invoking the size flexibility implied by small units may not be feasible. Instead, clustering in energy parks may be necessitated by a requirement for on-site staffing. NuScale, at least, is designed to be planted in clusters of 600 MW, increasing the possibility of streamlining a more quickly-trained staff structure.
A fairly recent article about NuScale had this to say about plant staffing:
"Design efficiencies have allowed NuScale to reduce the estimated operational staff count for a 600 MW plant to around 0.7 staff per MW, lower than the nuclear industry average, Snuggerud told Nuclear Energy Insider.
"'We've put a lot of effort into developing high levels of automation and leveraging the simplicity of the design,' he said."
But since this article appeared, NuScale has announced a potential upgrade of their module from 5O to 60 MW at essentially no extra cost (comments, Jun 8,11 above). So the staffing ratio for a 12 module, 720 MW plant might be closer to 0.6.
"ARC Chairman and CEO Don Wolf said the company was excited by the new collaboration and applauded New Brunswick for its strategic decision. 'We intend to demonstrate that the inherent safety features of our reactor enable a simple and cost-effective design which will be competitive with all other forms of electricity generation, all while protecting our environment and complying with the export control rules of Canada and the United States,' he said."
"Understanding the super safety of the Moltex nuclear plant which will be built in Canada" (link)
This article maintains that because Moltex reactors are inherently safe, they can forgo many of the expensive control features required by traditional nuclear plants. And if SMRs are cheap enough, they will become the dominant energy source in the coming decades.
The article is based on, and embeds, an audio presentation by the co-founder of Moltex.