Welcome to the ITER! The US Congress adopted a budget for fiscal year 2020 with sharply increased funding for the US part of ITER – $ 257 million. After several years of under-funding, this is great news! In general, by the end of 2019, we can ascertain the implementation of more than 65% of the work plan before the first plasma. But ahead is the most difficult third.
So, to launch ITER, we need:
- Specialized buildings of the ITER complex
- Electricity, water, air and other infrastructure stuff
- Heat removal system
- Cryogenic Fluid Supply System
- Superconducting Magnet Power Subsystem, Switching Matrix and Emergency Resistors for Magnetic Energy Reset
- Tokamak Vacuum and Fuel System
- Cryostat and thermal cryoscreens
- Finished superconducting magnets – total 43 pieces
- Vacuum chamber in which the plasma will burn
- A system for measuring plasma parameters, operating parameters of equipment, control and visualization – thousands of sensors and actuators and hundreds of racks throughout the complex
- And the most important thing is to put it all together, install, set up and run. We have exactly 6 years for this.
Now let’s look at these points in detail
The most significant event of 2019 is the top-out of the tokamak building. Already in March 2020, we are promised the completion of construction and the beginning of the movement of cranes from the preliminary assembly building (opened back in 2017) to the tokamak building, and as a result, the beginning of the assembly of the reactor in the reactor shaft.
Yes, the project went on for a long time at this moment – the excavation of the pit began in 2010, the filling of the seismic insulating foundation in 2011, and the construction of “working floors” began at the end of 2015 (a rather long pause was associated with the redesign of the building after the accident at the Fukushima nuclear power plant). And now – the design height has been reached! Interestingly, according to the plans of 2014, this was supposed to happen in July 2019, in general, we can say that the task was completed almost without delay.
Of the approximately 40 buildings and structures necessary for the first plasma, almost everything is already ready or is at the final stage of construction. Of the unprepared ones, it is worth noting the control building, the building of magnet energy dump resistors (these resistors are produced in Russia), and the tritium building, built about half. However, for the 6 remaining years, they can be completely completed and saturated with equipment.
Render of a completed site. Gray indicates the already built and saturated equipment, purple – tokamak building still under construction, blue – future buildings. All this blue building around the tokamak building itself is not needed for the first plasma and will be built later.
In addition, in 2019, builders handed over finished buildings for power converters of the magnetic system, a building for reactive power compensation equipment, and at the end of 2018, also the construction of a heat rejection system.
The ITER complex in full-scale launches will be the largest consumer of electricity – about 110 megawatts for supporting systems and up to 250 megawatts for heating and power supply systems for magnets. All this will be distributed among the systems with a complex multi-level system consisting of 7 transformers and two ASU stations connected to a 400 kilovolt switchgear. The first part, namely the general distribution medium-voltage device providing 110 megawatts of loads, was commissioned in January 2019 and took over the supply of very few consumers (builders and installers). This input will make it possible to test all the main infrastructure facilities of the tokamak – a cryocombine, a heat discharge system (these two consumers are responsible for the lion’s share of the load – almost 100 MW for two), however, work on the construction of local transformer substations and distribution networks is still underway.
Also in 2019, the first 400 MVAR transformer (out of three) from the power subsystem of variable loads (magnets, heating systems) was commissioned. It will be used to test the power converters of the magnetic system, which, however, will take place not earlier than in two years.
In 2019, the equipment of a heat release system was actively installed – and this is no less than 5 autonomous water circulation systems with different water chemistry and reliability level, 10 fan cooling towers with a total capacity of about 300 megawatts and two hot and cold water buffer pools, as well as 4 a dozen pumps, heat exchange equipment, etc. The whole system should receive up to 1150 megawatts of heat from the tokamak and its auxiliary systems at the time of launch, and buffering this heat will gradually dump it in pauses. For the first plasma, however, it is clear that the power of this system will be used in a small fraction of the possibilities.
The cryocombine is one of the largest liquid helium production facilities in the world in 2019 … was actively redesigned. In principle, this is the scourge of any complex “first of its kind” project – a huge number of relationships lead to the fact that the neglect of some little things translates into big alterations. In particular, as ITER engineers explained to me, the load review led to the need for a slight increase in equipment and the addition of ventilation and air conditioning systems, and its total volume turned out to be higher than the roof’s capabilities, and some of the air conditioners had to be moved to the annex, and all ventilation routes redesign. So, a small change led to the suspension of equipment installation here for a year.
However, this moment is behind, and in 2020 the start of autonomous testing of units is expected. After the construction of the building of emergency resistors (in the year 2022), an overpass with pipelines of cryogenic liquids from the cryocombinade of the tokamak building will be installed and, apparently, somewhere after 2023 the phased introduction of the cryosystem will go in the tokamak building, it should be quite interesting.
Another of the most important events of 2019 – the installation of cryolines began on the lower floor of the tokamak building, from which cryofeeders of superconducting magnets and various other things like cryosorption vacuum pumps will be powered.
This point is important in that the installation of the first (of very many) communications in the tokamak building finally began. This process will be complex and long, which means that it is important to start it as soon as possible.
ITER superconducting magnets in operation will store up to 46 gigajoules and operate on current up to 68 kiloamperes. Moreover, the tokamak in operation requires a fairly rapid change in the current in the magnets, which means powerful current sources “pumping” and “pumping” magnets with current. Two buildings will have about 40 separate converters, which are controlled multiphase rectifiers of grandiose sizes (the largest converters will be up to 90 megawatts of power, and the total power of all converters is 2.1 GW). Since power is needed specifically for changing the current, the magnetic system will be paired with a reactive power compensation system – roughly speaking, a set of capacitors and inductors switched to an alternating current network. This will allow storing part of the extracted magnetic energy and returning it back in the next cycle without “pulling” the high-voltage power line.
In 2019, in both buildings of converters, the installation of busbars (Russian-made) began, which will connect the converters and magnetic feeders, and the preparation of the bases for installation of the inverter blocks themselves began. There is also the installation of transformers (each inverter relies on an input trans), clearly visible on the general plans (they are outside the building).
In 2020, inverters will be installed and all components will be combined, but the electrical tests themselves are still far away.
An extremely important system as part of ITER, including as many as 400 vacuum pumps and 10 kilometers of vacuum pipes. It seems that in 2018-2019 she was struck by the redesign virus, in any case, the construction of that part of the tritium building, where, so to speak, the vacuum workshop with several dozen main pumping pumps was to be located, has been standing since mid-2018. However, on the floor above this room, the whole floor of the tritium building is reserved for another actively changing system – water cooling of the tokamak, the tasks for which were transferred from the United States to the EU in 2018. Some new elements of the vacuum system, however, lit up.
A cryostat is a vacuum vessel in which a tokamak will be located together with a magnetic system. The vacuum here is mainly for thermal insulation of very cold magnets from a rather hot vacuum chamber and the surrounding building. In 2019, the production of the “lower cylinder of the cryostat” was completed – this is the second part of the cryostat from the bottom (out of 4), and the work with the base of the cryostat is almost finished – this is the lowest part. From the installation of the base on the bearings and then the lower cylinder on the base in March 2020, the ITER assembly should begin (more on that below). In fact, hundreds of supporting elements for thermal cryo screens, sensors and their cable lines have yet to be welded on both pieces of the cryostat, but this work can be done both in the remaining months and even after installation of the reactor in the mine.
Cryo screens, by the way, are also on their way to the site. They are cunning 10-20 mm thick stainless steel sheets with welded cooling tubes through which helium will flow at a temperature of 80-100 K and silver plated to improve the reflection of infrared radiation. Some of the cryoscreens are included in the very first assemblies that need to be installed in the mine, so we are pleased to note that their production was completed on time (South Korea is engaged in it)
If you read my articles on ITER before, then you know that I do not get tired of admiring the grandeur of the main superconducting magnets of the international thermonuclear reactor. Actually, all 25 large ITER magnets will become the 25 largest superconducting magnets in the world. For the first plasma, it is necessary to collect all of them – however, the assembly order determines which of the magnets is the most priority. Actually, already this year at least the first 2 magnets should be installed in the mine – these are the lower poloidal PF6 and PF5, which will be located under the tokamak chamber. The first of which is made in China and moves in the direction of Kadarash, and the second is now undergoing final production operations right at the ITER site. Both magnets will have cryotesting (on site) and additional equipment with sensors, but we can expect that no later than the end of summer they will be lowered to the design position. Given the weight (~ 400 tons each) and dimensions (10 and 18 meters in diameter), installation operations should be quite epic.
The readiness of the TF toroidal field magnets is no less important – about a year after the start of installation, the assembly of the vacuum chamber in the mine should begin according to plans (this operation will take 2-2.5 years), and for it, the preliminary assembly of 2 TF coils and one sector ( as well as related cryoscreens) at the sector assembly stand in the preliminary assembly building (have you already written about the assembly?). Those. Sometime in the summer of 2020, ideally, the first two TF magnets and the first sector of the vacuum chamber should arrive at the ITER site and continue to do this in a regular manner.
Toroidal magnets are assembled in Europe and Japan. In particular, a year ago, “under the Christmas tree” in Europe, they carried out the operation of sliding the halves of the case onto the winding package (in Japan they did this in March 2019) and throughout 2019 continued to bring the first “combat” magnet to readiness. To do this, it was necessary to precisely expose the winding bag inside the case, weld the halves of the case, weld the lids through which the bag was inserted, fill the interior with epoxy. All this was successfully completed, and the last operation remained – machining of the body according to the allowances left for inaccuracy of assembly. This technological complexity is due to the fact that it is necessary to obtain a coincidence of real and theoretical magnetic axes within 1 mm in three axes with a product dimension of 16x10x3 meters.
While we know a lot about the successes of Europe, Japan (alas, traditionally) for the year did not publish anything on the progress of the TF assembly. A year ago, the lag was literally a couple of months, so maybe in 2020 the Japanese TF will arrive in Kadarash, which would be very useful – with European magnets alone, there’s no way to keep up with the assembly dates.
In addition to the above, there is also a central solenoid manufactured in the USA (according to which there has been no news since May), magnets PF4,3,2,1 (of which only 3,4 have not been started) – but all this will be needed for installation in 2-4 years, so today we won’t touch them.
Even at the dawn of the formation of tokamaks, as the most promising type of controlled fusion reactors, engineers noted that the toroidal shape of the chamber is a technological nightmare for industry. However, the reality turned out to be much worse: you need not just a toroidal chamber, but a two-walled chamber with high rigidity (and therefore with thick and numerous ribs), extreme requirements for welds and even exorbitant requirements for geometry accuracy (this is on surfaces of double curvature – where the ruler and even the template is so easy not to measure the accuracy).
The ITER vacuum chamber will be assembled from 9 sectors, 4 of which are manufactured in South Korea (Hyundai Heavy Industry), and 5 – in Europe (Walter Tosto / Ansaldo / ENSA). The production cycle includes hot stamping of workpieces, their machining, welding in 4 stages of enlargement with intermediate machining – and all this requires a lot of equipment of an uneven shape, and its own optical metrology. Things are moving very slowly – Koreans began cutting metal for the first sector in 2012 and only in the fall of 2019 did they get to welding 4 segments in the finished sector. Europe lags behind by about 2 years, and, in my opinion, it will not be able to issue 1 sector before the end of 2021, which with a high probability means a slide in plans for the first ITER plasma for 1 year.
The project of the vacuum chamber also has the contribution of India (makes blocks of neutron absorbers from boron steel) and Russia (pays for the manufacture of 9 top pipes in Germany, each weighing 18 tons) – but everything is fine here, no drama and emotions.
Finally, there are intra-chamber devices – a divertor, a first wall, a blanket, “warm” magnets to suppress ELF instabilities, numerous sensors and water lines. However, this is not necessary for the first plasma, so today we will skip this topic, although there is progress and achievements in this (very high-tech) area of ITER work.
In large industrial projects, the adjustment of the process control system often determines the delay in launching the entire project. Firstly, it logically goes to the last stage (it is impossible to configure the control system of non-installed equipment) and secondly it collects all installation, manufacturing, etc. errors that pop up on debugging. Here, ITER risks assembling full bingo: not only is the project extremely complicated, many components are also the first of their kind, they come from different countries and include their own local control systems. Despite the preventive measures taken in the form of open source software as a control stack (RHEL + CODAC + EPICS) and distribution of software and hardware sets to all interested, the launch of the system with tens of thousands of sensors, thousands of actuators (many with their own software and hardware inside) some of which must also meet the reliability criteria for nuclear facilities will be a very difficult task.
It is sad that all this is slightly postponed until later – although the management stack is already working not only purely in the laboratory, but also “steers”, for example, with the standard ITER power supply system (which took 6 months to start), the ITER control building and data center have not even began to build, so that in less than 3 years we won’t see the process of putting the ITERS ITER into operation.
In 2020, the assembly of the reactor in the mine should begin, which is what the project was going for about 12 years of real work and 35 years from the idea. The assembly promises to be an extremely difficult event if only because a lot of contractors will work in different systems and areas: from heavy rigging to optical alignment, from thousands of low-voltage cables to 300×200 mm busbars for current up to 70 kiloamperes, vacuum, cryogenic, water, gas lines – All this converges in a mine 30 meters in diameter and 30 depths.
A bit outdated, but generally true video, telling about the process of assembling a tokamak in a mine.
The installation of the reactor, in fact, has already begun – in the concrete base – the “crown” 18 bearings were installed under the movable support, and alignment pads for the future base of the cryostat are installed around. After installing the two lower sections of the cryostat, they will need to be welded and in parallel all “sub-clamp” products should be laid in the base – PF6.5 magnets, six correcting coils, a large collector for distributing cryogenics and currents to correcting magnets, cryovacuum screens (partially), then install the mounting column around which sectors of the vacuum chamber will be hung. (the convoy is undergoing acceptance in South Korea and will soon leave for France)
By the way, interesting articles will be put on the column – 6 compression rings – these are four-meter fiberglass parts that will hold the lower TF segments together. These products are currently manufactured in France.
In parallel with the “sub-chamber space”, 18 supports for toroidal magnets (made in China) will be installed and the lower cryofeeders of superconducting magnets will be installed.
Shot on the L1 floor of the tokamak building is just to feel the atmosphere.
This entire period should take about a year, after which the assembly of the toroidal vacuum chamber should begin. However, we need to live up to this stage – let’s talk about it in a year.
The ITER project progressed well in 2019, and even received an unexpected resolution to the funding problem from the United States. Nevertheless, implementation problems continue to creep out here and there – for example, the installation of systems in the tokamak building began with an annual lag, there is a strong delay in the production of the vacuum chamber. But I am glad that ITER is already on the verge of assembling the reactor itself in the mine – in a few months we will see this grand event with our own eyes.
P.S. If you read it to the end – a small bonus, a link to a wonderful 3D tour of the site, shot in October 2019