Lettres d'information

Choisissez ce que vous souhaitez recevoir :

Merci de renseigner votre adresse de messagerie électronique :


Actu & Médias


Of Interest

See archived articles


Byung Su Lim, leader of ITER's Poloidal Field Coil Section, posing in front of the completed winding facility.
Over the course of six years, the Poloidal Field Coils Winding Facility will house the successive winding and assembly of ITER's poloidal field coils, the huge, circular coils that will be positioned horizontally around the toroidal field magnet system.

All but one* that is. Of ITER's six poloidal field coils the smallest—the eight-metre PF1—will be procured by Russia and delivered to the ITER site. The five others are too large to be transported in their finished state: with diameters up to 24 metres, PF2, PF3, PF4, PF5, and PF6* will be manufactured under European procurement in the 257-metre-long winding facility.
Manufacture will take place in three sequential phases. From the reception of spools of conductor at the south end of the building to the exit of completed coils from the opposite end, the fabrication of each poloidal field coil will require at least 24 months.

Phase One: Winding - The raw material for the poloidal field coils is delivered on 20-ton spools from factories in China, Europe and Russia. Some 1,174 tons of niobium-titanium (NbTi) conductor will arrive at the Poloidal Field Coils Winding Facility in staggered deliveries.

Adjacent to the unloading area a climate-controlled "clean" enclosure is the theatre for winding operations. Lengths of NbTi conductor fed from two spools simultaneously ("two-in-hand" winding) are insulated and wound into a flat, spiralled coil called a double pancake.

"Starting from the conductor spools, two-in-hand double pancakes are wound, and insulated with glass-fibre tape", Byung Su Lim, Poloidal Field Coil Section Leader at ITER, explains. "The winding speed must be synchronized with the speed of insulation wrapping and the tension of insulation tape controlled. The shaping of each termination (for connection one to the other) is performed after the completion of winding."

The total conductor lengths required for the winding stage of manufacture varies from 6 km for PF2 to 14 km for PF3. "The size of each double pancake depends on the dimensions of the final coil," stresses Lim. "Depending on the number of turns in each double pancake and the number of pancakes stacked to form the final coil, each poloidal field coil is unique." Thirty-nine double pancakes will leave the winding facility's assembly line over a six-year period.

Phase Two: Impregnation - For the second phase of operations, the double pancake windings are transferred by overhead bridge crane to the impregnation area in the centre of the Winding Facility. "The double pancakes at this stage are still considered 'light components,' weighing a maximum of 37 tonnes," says Lim. The cranes lower the double pancakes into moulds for vacuum pressure impregnation (VPI) with epoxy resin. The resin—acting inside of a sealed mould and under the effect of heat—hardens the glass tape to bond each double pancake into a rigid assembly.

Current passed through the conductor heats the double pancake to approximately 80 degrees Celsius—once hot it is put under vacuum. At the same time, the resin is put under vacuum, degassed, and heated to 80 degrees.

"The challenge during this phase is to distribute the resin in a uniform manner," explains Lim. "Pressure is our main ally: propelled by pressure from the lowest point of the double pancake, the resin reaches all areas before flowing out at the highest point. We'll leave plenty of time for the resin to wet the insulation completely and to avoid creating trapped bubbles in the winding."

Once impregnated, the double pancake is put first at atmospheric pressure and then at overpressure, and "cured" for more than 24 hours above 100 degrees Celsius.

Phase Three: Assembly: The resulting solid double pancake winding is transferred to the assembly area of the building where it is stacked and joined, a second vacuum impregnation is performed to harden the stacked assembly, and additional components added such as clamps, protection covers, and pipes.

The floor in this area of the winding facility has been reinforced for the exceptional weight of the final assemblies. "Eight double pancakes are stacked to form the winding packs for PF coils 1,3,4 and 5, an respectively six and nine double pancakes for the winding packs of PF 2 and PF6. That puts the completed coils—winding pack plus additional components—at between 200 tonnes and 400 tonnes," says Lim. "These massive components will exit the winding facility on self-propelled transporters."

The size and weight of the poloidal field coils are a particular challenge for manufacturing operations, according to Lim. "Manufacturing tolerance targets have been defined at 3-4 mm—a particularly challenging target for components that measure up to 24 metres in diameter."

Avoiding deformation during handling is also an issue. "At frequent intervals throughout the manufacturing process first the double pancakes and then the coils will have to be lifted and manoeuvred as smoothly and evenly as possible, with a maximum authorized tilt of 10 mm," says Lim.

A heavyweight crane located at the far end of the Winding Facility will have the capacity to transport loads of up to 100 tonnes (or 50 tonnes with its circular spreader beam attached). The final assemblies—which will surpass these limits—will be mounted on retractable pneumatic supports.

"At ITER, we are building some of the largest coils in the world," concludes Lim, who joined ITER after ten years at the KSTAR tokamak in Korea where he oversaw the manufacture, assembly, installation and commissioning of the toroidal field coils. "It's especially exciting to be heading into the construction phase of poloidal field coils in earnest. All of our careful planning and teamwork is about to be put to the test. All of us—at ITER and at the Domestic Agencies—are looking forward to getting started."

Click here to watch a video produced by F4E on the poloidal field coil manufacturing.

* Editor's Note: The European Domestic Agency concluded an agreement with China in October 2013 for the fabrication of PF6. As a result, only four poloidal field coils will be produced in the winding facility on the ITER site.

It was one of the coldest days of the year when the ITER management, including Director-General Osamu Motojima and delegations from the seven ITER Members, met in Cadarache to discuss outstanding issues for the Test Blanket Module Program.
ITER will provide a unique opportunity to test mock-ups of breeding blankets, called Test Blanket Modules (TBM), and associated ancillary systems in a real fusion environment. Within these Test Blanket Systems (TBS), viable techniques for ensuring tritium breeding self-sufficiency will be explored in the framework of the TBM Program.

The TBM Program has a special standing within the ITER research program in that the Test Blanket Systems are developed by the Members and remain the property of the Members ... even though they will be tested at ITER. It is furthermore an essential element of the common purpose of the ITER Members to demonstrate the scientific and technological feasibility of fusion power for peaceful purposes. 

That is why members of ITER Organization's management and delegations from the seven ITER Members came together in Cadarache this week, to discuss the generic "Test Blanket Module Arrangement." This Arrangement will be used as a template for the individual TBM Arrangements, which will govern the relationship between the ITER Organization and each Member during the development and construction of the Test Blanket Systems.

"The aim of the meeting was to converge on some outstanding issues in the definition of the template for the TBM Arrangements, so that the Members will have a common legal framework for working with the ITER Organization during the development and construction of the Test Blanket Systems," commented Luciano Giancarli, ITER responsible officer for the TBM Program.

Guy Bonnaud, in charge of the French Fusion Master class, welcoming the students.
Each year in February, students from French graduate universities within a federation called "Education for Fusion Sciences" regroup at Cadarache to follow advanced courses on frontline science and technology related to magnetic fusion. The master course aims to provide interdisciplinary knowledge and skills to scientists and engineers from France and foreign countries that are keen on studying in energy and fusion research programs, specifically within the framework of large projects, both in national or private laboratories.

This week, 25 students from the University of Nancy, universities from the Île-de-France area around Paris, and Marseille gathered in the amphitheatre of the IRFM (l'Institut de Recherche sur la Fusion Magnétique), the fusion branch of the CEA Cadarache that operates the Tore Supra tokamak, for the launching of this year's Master course. For the next four weeks the students will get the chance to gather knowledge on the physics and the technology of fusion—both in theory and by hands-on exercises. Scientists from both the IRFM and ITER will be engaged in teaching the next generation of fusion scientists and engineers.

Click here for further information.

The Fusion Expo on show at the University of Nancy.
The show may just have ended in France, but don't you worry! The Fusion Expo will be back shortly—this time in Aix-en-Provence. After ten days on stage at the University of Nancy, France, the Fusion Expo is currently taking a break to get freshened up. From 13-30 November the travelling exhibition will celebrate its comeback in Aix-en-Provence, where it will be on show for three weeks in the heart of the city at the Tourist Office, just 30 kilometres away from the ITER site.  

David Beltran watches as flames devour a set of cables in the SATURNE facility (IRSN) at CEA-Cadarache. Note the small blue lamp in the lower left corner of the image.
Safety is about anticipating aggressions, whatever their nature and their probability. An earthquake, a flood, an airplane crash, a fire breaking out inside the installation—all these are among the events that must be identified, analyzed and simulated prior to the designing of an installation.

Let's take a look at fire, for instance. It is of vital importance to know what material will burn and how fast; how the flames would propagate within an installation; what chemical elements would be released through the smoke; and how these releases would affect the filtration systems, etc.

In ITER, which is mostly concrete and steel, a "fuel of choice" was identified: the large amount of cables that wind and wiggle for dozens of kilometres in the innards of the installation, forming what Electrical Engineer David Beltran calls "the largest mass of combustible material available in ITER."

Not only do cables burn, they may propagate flames and induce other fires. "Cables are long; they run close to each other; they get into every corner of the installation ... In the absence of strong safety measures, they could spread a fire all over the facility," says David.

In order to take these strong measures, one has to know how different cables would behave in the different circumstances a fire would create. "We needed data, as precise as possible, to feed our simulation codes," explains Pierre Cortes, ITER Safety, Analysis and Assessment section leader.

All the cables in ITER will be of a particular type called "halogen-free" (or LSZH for Low Smoke Zero Halogen). In case of a fire and subsequent outgassing, they do not produce halogen elements like fluorine, chlorine or bromine that might harm some of the installation's systems.

"Unfortunately, R&D on halogen-free cables is scarce," says Cortes. "The solution was to perform the tests ourselves." And so they did, burning all sorts of cables: thick and thin, long and short, in an upright or lying position, with or without ventilation ...

The data acquired covered issues such as ignition temperature, heat release rate, smoke production, the nature of outgassed elements and the speed of propagation. Tests were conducted throughout the year 2011 in the world-class SATURNE and CARINEA installations (IRSN, Institut de Radioprotection et de Sûreté Nucléaire) where all kinds of different fires can be recreated and their dynamics closely and precisely monitored. Both installations, quite conveniently, are located at CEA-Cadarache.

The tests that were performed also enabled the safety team to test the integrity and functionality of some of the safety-relevant cables under conditions of fire. The experience was both simple and spectacular: an electrical cable, feeding power to a lamp, was slowly devoured by the flames. How long would it hold? How long would the lamp keep shining?

The tests demonstrated that specific cables could retain their functionality—whether they carried power or signal—for quite a long time. Seeing the little light continue to shine as the cable sheath carbonized was an impressive sight indeed ...

Cortes and his team have now accumulated a large amount of data on cable behaviour, type by type, brand by brand, and in all imaginable situations. This is not, however, the end of the story. "The question now is to analyze this data in order to assess whether we should test specific firewall systems or other configurations."

One thing is certain: cables will continue to burn in the dedicated fire-test facilities in Cadarache in order to mitigate the consequences of their potential burning in ITER.

The CEA-F4E CSC team standing between a section of the Helios supercomputer, from left to right: Jacques David, François Robin, Jacques Noé (CEA) and Susana Clement Lorenzo (F4E).
The Helios supercomputer is operational according to schedule at the International Fusion Energy Research Centre (IFERC) hosted by the Japanese Atomic Energy Authority (JAEA) in Rokkasho. The machine, whose mission it is to perform complex calculations for plasma physics and fusion technology, has passed its acceptance tests achieving 1,132 Petaflop LINPACK [1] performance.

The Computer Simulation Centre (CSC), where Helios operates, is an important component of Europe's contribution to the Broader Approach, an agreement signed between Europe and Japan to complement the ITER project through various R&D activities in the field of nuclear fusion. The European participation to the Broader Approach is coordinated by Fusion for Energy. The supercomputer was provided by France as a part of its voluntary contribution to the Broader Approach, through a contract between the Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA) and manufacturer Bull.

The acceptance tests of the supercomputer were carried out between 13-22 December 2011 in Rokkasho, Japan. The tight construction schedule was successfully met offsetting any disruptions caused by the great East-Japan earthquake in March 2011. The installation of the equipment was completed in early December and by the end of the month a 1,132 Petaflops LINPACK performance was achieved, ranking Helios fifth in the TOP-500 November 2011 list.

The operation of the supercomputer will kick off with four high visibility runs ("light-house projects") which are expected to shed light on plasma calculations. From January to March 2012, the four selected codes will run one at a time to test drive the capacities of the supercomputer and achieve maximum performance. The first call for proposals has attracted high numbers of submissions from both European and Japanese researchers that are currently under review. It is expected that routine operation will start in April 2012.

Based on the number of proposals submitted to the first call, there has been an oversubscription by a factor of three of the computer's time, demonstrating the great interest from the European and Japanese fusion communities in the supercomputer facility. The majority of proposals address issues related to plasma physics (turbulence, MHD, edge physics and integrated modelling) together with an important number of proposals addressing technology issues.

[1] The LINPACK benchmark is a measure of a computer's floating point rate of execution. It is the performance parameter used to classify the TOP 500 list of supercomputers.

Since "acrobats" began bolting together the pre-assembled arms of the pylons, work on the 400kV power line has provided some spectacular sights. © CEA/Corinne Guis
Installing a 400kV power line requires, in the following order: a crane, a team of acrobats, a truck-mounted winch and, finally, a helicopter.

In difficult areas—such as the ridge that separates the ITER site from CEA-Cadarache to the east of the present Headquarters Building—helicopters are used to position temporary "pulling cables" that will later be attached to the actual power cables and pulled by a ground-based winch.

Such an operation was organized last Wednesday 1 February after the sky had cleared following the previous day's heavy snowfall. "Pulling cables" were installed between pylon 18, located at CEA-Cadarache's highest point, and pylon 15, located close to canteen n°2.

The same operation, this time between pylons 15 and 10, is scheduled for 28 February.

Work on the 400keV power line, which began in September, continues to provide spectacular entertainment.