A tokamak plasma was achieved in JT-60SA for the first time on 23 October, making it the world’s largest operational superconducting tokamak to date.
The result will be carefully examined and the team will continue to perform more tests during the next weeks.
A ceremony to officially inaugurate JT-60SA will be held in Naka on 01 December, with representatives and guests from both Japan and Europe.
JT-60SA has 6 ring-shaped Equilibrium Field (EF) coils which will confine and shape the plasma. They are wound from niobium titanium cable-in-conduit superconductor.
Energization tests on the EF coils started on 19 September, beginning with the largest coil EF 1. EF 1 has a diameter of 12 m and weighs 30 tonnes, making it one of the largest pulsed magnets in the world.
The voltages and currents applied to the EF coils will be limited at first to avoid unnecessary risk. The first target is to commission the EF coils to ± 3 kA together with their base power supplies and Quench Protection Circuits (QPC).
The first tests for each EF coil are to confirm that an emergency stop can be reliably operated from both plus and minus 1 kA. That is, the QPC intervenes correctly to discharge the magnet and the power supply shuts down safely. This check is repeated whenever the current is increased, e.g. to plus and minus 3 kA. Then the control of the magnet is confirmed by applying current and voltage waveforms.
The toroidal field (TF) magnet provides the primary confinement for the JT-60SA plasma. The 18 D-shaped coils are designed to operate at 25.7kA and produce together a field of 2.25 Tesla at the toroidal axis running around the tokamak. The corresponding energy stored in the toroidal magnetic field reaches 1.06 GJ.
After completing the vacuum vessel bake at about 200°C and returning the vacuum vessel to its operating temperature of 50°C on 02 August (and confirming it is still leak-free!), and after performing high voltage tests on the magnets under their operating vacuum condition, TF energization began on 07 August.
The first step was to “balance”, or adjust, the quench detectors during repeated ramps up to 3kA. The voltage across each of the 18 individual TF coils is continually compared to that across its neighbour in order to detect any loss of superconductivity (known as a quench). Each measurement is made using a sensitive bridge circuit which should detect any change in resistance but should not react to the inductive voltages arising when the current in the coils is changed.
Next a comprehensive functional test of the quench detectors was made. This was achieved by deliberately warming up some of the coils just enough for them to lose superconductivity and become resistive. Then the current applied to the magnet was gradually increased. As soon as the threshold for quench detection was reached, the quench protection circuits (QPC) were activated. This shuts down the magnet power supply and opens bypass switches forcing the TF current into large resistors to rapidly release the huge magnetic energy. This test was repeated until each of the 9 pairs of quench detectors had triggered correctly (each pair of coils has 2 detectors for redundancy).
Then the current in the magnet could be progressively increased, with test operation of the QPC at each step. The energy in the magnet increases with the square of the current, and it is important to confirm that the energy released during QPC operation can be safely handled. The QPC is so important to protect the magnet that each bypass switch has its own backup pyrobreaker: an explosive charge that can intervene to interrupt the current in case the bypass switch fails. Operation of a pyrobreaker was deliberately tested during QPC operation at 15 kA on 21 August.
QPC operation (or to a greater extent an actual magnet quench) is also disruptive for the cryogenic system due to the sudden additional heat load generated. Hence in preparation for QPC operation at higher currents tests have also been made to trial temporarily disconnecting the TF magnet cooling loop from the cryoplant, and to check that when the pressure in the loop increases the automatic valves relieve the pressure correctly by directing helium to the quench tank.
Confirming the correct operation of all the processes designed to protect the tokamak is a key part of its commissioning.
The central solenoid is the heart of the JT-60SA tokamak and is critical to initiate the plasma and then to drive current in it. Plasma current is a defining feature of a tokamak which substantially enhances the confinement of the superheated particles.
The central solenoid is wound from niobium tin (Nb3Sn) cable-in-conduit conductor (CICC), which in the absence of any magnetic field becomes superconducting at about 18 K (-255°C). Early on 22 July the four modules of the central solenoid reached this temperature and the resistance of the magnets dropped to almost zero, demonstrating that the central solenoid is superconducting.
The 18 toroidal field coils and 6 equilibrium field coils are made using niobium titanium (NbTi) CICC. Their superconducting transition occurs at about 9 K (-264°C), meaning that it took a couple more days to reach and could be observed on 24 July.
By allowing current to flow with no resistance (unlike a conventional electromagnet), the superconducting magnets of JT-60SA will allow plasmas to be produced for long durations without excessive power consumption. JT-60SA is the largest superconducting tokamak built so far. The largest magnet has a diameter of 12m.
Meanwhile the vacuum vessel that will contain the JT-60SA plasma continues to be ‘baked’ at an elevated temperature of 200°C in order to remove impurities attached to its surface.
On 10 July the refrigerator turbines were started in order to cool the magnets down from 80 K to their operating temperature of 4.5 K. Following this the temperature of the vacuum vessel which will contain the JT-60SA plasma has been increased from 50°C to 200°C.
It is necessary to ‘bake’ the vacuum vessel like this in order to achieve sufficiently clean conditions for the plasma. The high temperature helps to drive out water and other impurities from the surfaces inside the vessel so that they don’t end up polluting the plasma. The quality of the vacuum will be a key factor in achieving a tokamak plasma. The vacuum pressure increases temporarily during the bake, but should ultimately be lower and the plasma-facing surfaces should be cleaner.
Meanwhile the magnets continue to get colder. They are protected from the increased thermal radiation from the vacuum vessel by the 80 K (-193°C) double-walled helium-cooled stainless steel thermal shield surrounding them. The vacuum vessel bake causes the highest thermal loads on the cryogenic system, which now consumes 8 truck loads of liquid nitrogen each day.
The vacuum vessel that will contain the JT-60SA plasma is designed to operate at 50°C. It was warmed up to this temperature over 12-13 June by circulating hot nitrogen gas through its double wall structure.
The helium coolant in the superconducting magnets and associated plant was purified by circulating it at room temperature starting on 03 June. By 13 June the dewpoint was below -70°C and less than 7 ppm nitrogen could be detected at the magnet outlet.
The cooling of the JT-60SA magnets started on 14 June and is proceeding steadily. The distribution of helium is carefully optimised to avoid excessive differences in temperature for the components while maximising heat extraction. Nevertheless, the combined mass of the toroidal field, equilibrium field and central solenoid magnets is about 640 tonnes, so it takes a long time to get down to 4 K! Today the magnets have reached about 173 K (-100°C).
Above 80 K (-193°C) refrigeration is provided using liquid nitrogen. Consumption is now around 1400 litres per hour, which means 5 deliveries by truck every day.
Following the successful leak testing of the cryostat and the helium pipes, it was confirmed that also the main vacuum vessel is leak free.
This week a number of experiments have been made in the cryostat in order to test new vacuum monitoring sensors that have been installed.
Now helium gas is being circulated at room temperature through the magnet cooling pipes and the cryoplant purification system (80 K adsorber). This will flush out any impurities such as air from the system ready for cooling down. In parallel several dry nitrogen purges have been performed in the JT-60SA cryostat to help remove water vapour adsorbed on its multi-layer insulation.
JT-60SA operations restarted on 30 May when the flanges providing access to the cryostat were sealed closed again and vacuum pumping started. This was a big moment for us after the long suspension for improvement works!
Evacuation of both the vacuum vessel, which will eventually contain the tokamak plasma, and the cryostat surrounding the machine core proceeded smoothly. Now both the cryostat and the numerous helium pipes inside it, which will supply coolant to the superconducting magnets, have been successfully leak tested.