«Association EURATOM / IPP.CR I N S T I T U T E O F P L A S M A P H Y S I C S, v.v. i. ACADEMY OF SCIENCES OF THE CZECH REPUBLIC ANNUAL REPORT ...»
0,15 0,16 0,18 0,19 0,20 0,23 0,24 Fig. 1. Number of collected photons in each wavelength and spatial channel of the detection camera. Energy of laser is 2.5 J on 532 nm, spectral channel width is 1 nm.
The calculations were performed first and foremost for the plasma edge, since this is the critical part of the measured region. High density in the centre will ensure sufficient number of detected photons. Background radiation by Bremsstrahlung was estimated too.
Part II - PHYSICS Laser system
Following demands on the laser system were set from the analysis of the scattering efficiency:
Nd:YAG laser system running in pulse regime and generating second harmonic, i.e. wavelength 532 nm, energy of 2.5 J on the second harmonic for each pulse, and repetition rate of 30 Hz, which is closed to technological limit for high-power lasers.
Output beam must be polarized, because of polarization dependence of TS. Divergence of the beam should be less than 1 mrad to achieve good focus and beam diameter close to the tokamak vessel. Laser will be placed outside the tokamak hall; the optical path from the laser to the tokamak will be therefore about 15 m. Beam diameter on the laser output should be about 1 cm, thus on the focusing lens close to the tokamak, the diameter will be up to 3 cm. The lens (f = 1 m) will focus the beam into the plasma, having sufficient focus depth to have 1 mm waist and having diameter less than 1 cm all-over the 40 cm long region of the core TS measurements. Beam pointing stability must be good enough to guarantee movements of the beam in the tokamak vessel about 1 mm.
In the future, an upgrade to multi-pass TS system will be possibly done, exploiting the experience of Petra Bilkova and Petr Bohm with the upgrade from mobility-stay on TEXTOR tokamak in the autumn 2007.
Detection system Highly sensitive back thinned intensified CCD or CMOS camera will be used for detection of the scattered light. Required range of detectable electron temperature, number of spatial points and spatial resolution determines requirements for the camera system for fast events and very low light level imaging. Optical design and simulation of the best parameters of imaging optics in the collected light path is being done by the ZEMAX software, as well as the layout of the spectrometer in the Littrow arrangement sense.
In collaboration with:
H. Fernandes, J. Sousa, T. Pereira, I. Carvalho, A. Neto, Association EURATOM-IST, Lisbon, Portugal M. Cavinato, Associazione EURATOM - ENEA sulla Fusione, Padova, Italy
The magnetic diagnostic of COMPASS is a complex set of many magnetic sensors distributed around the plasma column and mounted mostly on the vacuum vessel. The overview of the COMPASS magnetic diagnostic sub-systems including the number of sensors and the target quantity to be measured is given in . The dismantling of the magnetic diagnostics on COMPASS was carried out during two missions of IPP staff in the Culham Science Centre in November 2006 and March 2007. The existing documentation of magnetic diagnostics was checked and completed. The magnetic sensor positions were identified and documented. Cables from magnetic sensors to corresponding cubicles were disconnected. All equipment was prepared and transported to IPP Prague. The magnetic sensors will be checked and calibrated in early 2008.
The most important step in the re-installation of the COMPASS tokamak is to put into operation the feedback system for maintenance of a defined plasma position. The task of the vertical and radial feedback controls is to process the signals from magnetic sensors, which detect the plasma location, and to apply a correction signal to the power supplies to maintain the plasma in the required position.
To build the feedback system for COMPASS, we considered two options: refurbish and use the original COMPASS analogue system, or design and built a new system based on digital approach. Finally, we decided to pursue the later option because it is significantly more transparent, flexible, and up-to-date.
Digital system is developed under contract of collaboration with the IST Lisbon, which is experienced in building similar systems for other Associations .
a) Algorithms. Algorithms for feedback control will be generated by the finite elements MAXFEA code, which is solving the free boundary equilibrium based on the 2D Grad Shafranov equation, developed by P.Barabaschi . Geometry and mesh files were done for COMPASS. They include the exact position of the poloidal field coils, vacuum vessel and limiters. The material properties, resistivity and inductance are described in a material file.
The connections of magnetising, equilibrium, shaping and feedback coils are stated in a circuits file with corresponding number of windings and with the currents to the shaping and equilibrium circuits.
Part II - PHYSICS
b) Hardware and control. The digital system will be based on the ATCA technology standard:
multiple - input - multiple - output (MIMO) controllers, developed in IST, will be used. The output of the modelling will be converted to algorithms, which will be programmed to CODAC nodes. These nodes include ATCA modules, each one with 32 analogue input channels, 4 analogue output channels, and 8 digital input/output channels connected to a processor. Such set-up allows the implementation of the MIMO controllers.
Before the signals from the magnetic sensors (Internal Partial Rogowskis, Flux loops and saddle coils) will go to the digital feedback system, firstly they are integrated by original COMPASS analogue integrators, which were used in Culham. In 2007, we have performed basic test of several original COMPASS integrators connected to CASTOR magnetics. The results of these tests were rather mixed. The integrators showed a significant drift despite their embedded drift correction feature. Some integrators were even found to be not operational. In 2008, these initial tests will be completed systematically for all old COMPASS integrators and the final conclusions about their usability for the new COMPASS magnetic system will be drawn.
c) Power supplies. Three identical fast amplifiers are being built in the IPP for the feedback stabilization of the horizontal (1 piece) and vertical (2 pieces in a bridge) plasma positions.
The amplifiers are based on MOS transistors, each amplifier has maximum current of 5kA, voltage ±50V, frequency range DC-5kHz, output impedance of 5 mΩ, and inductance of 274 µH. A passive cooling of the transistors is adopted through heat absorption in attached mass of aluminium blocks. The circuit safety is designed for overcurrent, overvoltage and optical isolation protection inside the energizer that determines the amplifiers inputs. There were designed three energizers (containing pre-amplifier, optical decoupling and protection circuits for the fast amplifiers) and digital PID controller (for the fast amplifiers control). In addition, amplifier zero-output-voltage pre-shot check and another delayed overcurrent protection is included.
 Application for Preferential Support for enabling a programme of ITER relevant plasma studies by transferring and installing COMPASS-D to the Institute of Plasma Physics AS CR, Association EURATOM-IPP.CR. Phase II (2006)  A.J.N. Batista, Rev. Sci. Instrum. 77 (2006) 10F527  P.Barabaschi, Plasma Control Technical meeting, ITER JHT, Naka, Japan (1993)
COMPASS Control, Data Acquisition, and Communication system M. Hron, J.Písačka, F. Žáček, J. Strnad, J. Vlček, J. Stöckel, R. Pánek
In collaboration with:
H. Fernandes, J. Sousa, B. Carvalho, T. Pereira, I. Carvalho, A. Batista Associação Euratom-IST, Lisbon, Portugal M. Cavinato Associazione Euratom-ENEA sulla Fusione, Consorzio RFX, Padova, Italy An efficient operation of COMPASS in Prague requires a new CODAC (Control, Data Acquisition, and Communication) system, which is being developed jointly by the Associations IPP.CR and IST. The tokamak control involves several areas, each of different levels: 24 hours a day control of building infrastructure, interlock, central experiment control, and real time control during the experiment run. Each of the systems has different purposes and diverse features.
Control of slow processes The slow control of the overall plant during the 24 hours a day 7 days a week cycle includes the building infrastructure (cooling, heating, air conditioning) as well as the machine infrastructure running in the instantaneous service (experimental area access control, deionized water treatment, energetics systems, etc.).
All these systems are, in general, parts of the building, communications installations, and energetics deliveries, respectively. Specifications for these systems were derived by the IPP.CR staff during the design phase, then the construction and delivery was/is a responsibility of the contractors.
The control of these systems will be managed by their own controllers, while the communication links will secure the necessary exchange of information.
Personnel and machine protection Safe operation of COMPASS will be guaranteed by an interlock system. The personnel and machine protection will be separated in independent loops.
States of the experimental area will be based on the operational diagram, shown on right, and they will be set by an Area access control system. The personnel protection will be based on a restriction of the access in certain states of the experimental area, during operation of danger systems.
Next, the status of the machine systems will be checked by a safety loop during the preparation phase before each discharge. This will be done in parallel with the CODAC control checking.
Finally, during the discharge, the operation of the tokamak can be interrupted and/or operation of particular systems can be inhibited, depending on the evolution of the status of the individual systems.
Central control of the experiment Part II - PHYSICS The central controller with an operator interface, called “FireSignal“, will allow us to manage individual subsystems and will be used for the control of the experiment . Data access will be managed by a Java based “SDAS layer” .
During 2007, the CODAC design was evaluated and the HW platform for CODAC nodes was selected. We will use the Advanced Telecommunications Computing Architecture (ATCA) and a real time linux, running on a PC. Such a system has been developed at IST Lisbon for the vertical stabilization for JET  and will be adopted for COMPASS. This set-up builds then a compact CODAC node, where the data acquisition is secured by A/D converters and FPGA arrays on the ATCA boards. The PC running the RealTime Application Interface (RTAI) for Linux gives us the necessary computation power. The available digital and analog outputs of the system are used for communication and control of the subsystems.
We specified the communication link to be used between the CODAC and the power supplies, which are delivered together with the Energetics (toroidal field, equilibrium field, magnetizing field, and shaping field power supplies). Basic features communication protocol were defined.
Next, several particular systems were defined: e.g. a communication module between the CODAC and fast amplifiers for feedback was designed; vacuum system control was designed, incl. a vessel baking module; NBI control requirements (see figure) were specified for the Call for Tender.
Real time control during the experiment run The CODAC nodes described in the previous section will be capable of a real time processing of the acquired signals in a loop well bellow 50 !s. The data acquired by the A/D converters will be pre-processed by the FPGA on board. Then, during the test and start-up phases, the output signals will be calculated in the CPU of the PC. Later, when the algorithms will be fixed, they may be transformed to the FPGA itself to speed-up the response of the feedback loop.
References  A.Neto et al., Fusion Engineering and Design 82 (2007) 1359–1364  A.Neto: Fusion Engineering and Design 82 (2007) 1315–1320]  A.J.N.Batista et al., Rev. Sci. Instrum 77(2006) 10F52 Part II - PHYSICS 3 Development of Concept Improvements and Advances in Fundamental Understanding of Fusion Plasmas
In collaboration with:
Tsv. K. Popov, P. Ivanova, Association EURATOM-INRNE/Faculty of Physics, Sofia, Bulgaria.
F. M. Dias, Association EURATOM-IST/CFP, Lisbon, Portugal.
Langmuir probes (LP) allow local measurements of the plasma potential, the charged particles density and the electron energy distribution functions (EEDF) usually using the second derivative of the I-V characteristic. The recently developed kinetic theory [1, 2] may be used for the calculation of those plasma parameters from the first derivative of the electron saturation current measured by the probe. In this work we report the first results of the EEDFs measurements in the CASTOR tokamak edge plasma using this method.