All opinions expressed herein are those of the authors and should not be reproduced, quoted in publications,
or used as a reference without the author’s consent.
Oak Ridge National Laboratory is managed by UT-Battelle, LLC, for the U.S. Department of Energy.
Published by Oak Ridge National Laboratory
Building CR 5600 P.O. Box 2008 Oak Ridge, TN 37831-6169, USA
Editor: James A. Rome Issue 148 June 2015
E-Mail: jamesrome@gmail.com Phone (865) 482-5643
On the Web at http://www.ornl.gov/sci/fed/stelnews
Europe and W7-X:
An example of European
collaboration
Collaboration in European fusion research has a longstanding
history: in the 1970s, the European fusion laboratories
joined to build and operate the Joint European Torus
(JET). To coordinate research activities beyond JET, the
European Fusion Development Agreement (EFDA) was
created in 1999. In 2012, EFDA issued “Fusion Electricity:
A roadmap to the realisation of fusion energy,” and in
2014 the European Consortium for the Development of
Fusion Energy (EUROfusion), representing 29 national
European fusion laboratories, was formed to further
develop the collaboration and efficiently implement the
Roadmap. Research under EUROfusion is organized into
work packages that are aligned with the eight missions
outlined in the Roadmap.
One of the Roadmap missions is to develop the stellarator
line to maturity as an alternative way to fusion power.
Wendelstein 7-X (W7-X) is the cornerstone experiment of
Mission 8. The strategic benefit for Europe is to take leadership
in three-dimensional (3D) magnetic confinement
physics and technology and to mitigate scientific risks in
the challenges on the road to fusion electricity. In addition,
W7-X is a broadly collaborative scientific project for
European fusion research.
The stellarator work package comprises theory and modeling,
with the preparation of experimental schemes benefiting
from the smaller, but easy-to-access device TJ-II in
Spain and from collaborations abroad. A backbone of
European participation in W7-X, however, is the delivery,
operation, and exploitation of components and diagnostics.
To this end, European expertise in leading-edge systems
is being developed and applied to the scientific
program of W7-X. More than 12 European Research Units
will contribute to the work program of W7-X in 2015.
The Hungarian video camera system is a prominent example
of how W7-X benefits from European high-tech. For
more than 20 years, collaborations between the Mak
Planck Institut für Plasmaphysik and the Hungarian partners
at the Plasma Physics Department of the Wigner
Research Centre for Physics (the former KFKI-RMKI)
have existed. Led by Gábor Kocsis, the Hungarian team
(Gábor Bodnár, Gábor Cseh, Tamás Ilkei, Tamás Szabolics,
Tamás Szepesi and Sándor Zoletnik) is delivering an
intelligent fast camera system for the protection of the stellarator
wall. Details are presented in the next article.
In this issue . . .
Europe and W7-X: An example of European collaboration
EUROfusion was formed in 2014 as a consortium of
29 national European fusion laboratories. It is now
implementing a fusion Roadmap that breaks the overall
task into eight missions. As the cornerstone experiment
of Mission 8, Wendelstein 7-X provides a
platform for collaborative activity. ........................... 1
Event Detection Intelligent Camera for real-time
plasma diagnostics and control at W7-X
An innovative concept has been developed to deal
with the massive amount of data needed to monitor
the entire W7-X interior with sub-ms resolution during
long pulses and with multiple cameras. ................. 2
Stellarator News -2- June 2015
Event Detection Intelligent
Camera for real-time plasma
diagnostics and
control at W7-X
Safety first
Wendelstein 7-X will allow quasi-steady-state operation,
delivering very long discharges that are considered to be a
prerequisite for any economical fusion reactor. Although
the plasma is expected to be quiescent, long pulses (tens of
minutes) may give rise to local overheating. Such power
pulse-limiting events require implementation of detection
and protective measures to avoid operational problems.
The aim of the Event Detection Intelligent Camera (EDICAM)
project is to develop and operate a special fast camera
system for W7-X. The main challenge our scientists
had to face was the versatility such a system must have: it
must provide an overview video over the full cross section
of the stellarator; it must be fast enough to record plasma
phenomena changing in less than one millisecond; and the
video must cover the whole plasma discharge, lasting up
to half an hour. The camera system must be able to protect
the W7-X machine by detecting dangerous operational
states when the plasma touches structural elements and
rapidly communicating this to the plasma control system.
Additionally, the cameras must operate in the harsh environment
of a high magnetic field and irradiation by hot
surrounding surfaces.
Intelligent surveillance
Even if such a camera existed, the problem of storage
space would be overwhelming: usually fast cameras are
used for very short time intervals, fractions of a second,
but even then they can produce 10 GB of data – and W7-X
can operate for 1800 seconds. And to make things even
worse, we require 10 cameras to be able to monitor all
inner surfaces of the machine. These problems were overcome
by two innovative ideas.
The first idea was to completely change the way that a fast
camera should work: using modern data acquisition technology
(10 Gbits/s optical transmitter and PCI Express)
we built the system so that all data from the sensor can be
stored in the computer, making camera memory unnecessary.
In such a system, the problem of storage space is also
handled: the camera can be run at a standard frame rate but
speed up the readout of interesting areas when some event
happens, dropping all other frames containing no valuable
information. This functionality inspired the name of this
device: the Event Detection Intelligent CAMera, EDICAM.
The detected events, if regarded as dangerous, can
be communicated to the plasma control system, which can
take measures to protect the fusion machine.
The second innovative idea was to use a special CMOS
sensor with a so-called non-destructive readout (NDR)
capability. Normal camera sensors are erased when the
image is read out from them, and then the recording of the
next movie frame is started; these sensors are also erased
when only a small part of the sensor (‘region-of-interest’,
ROI) is read out. In EDICAM it is possible to read out a
small part of the sensor several times without affecting the
recording of the full frame. That is, a small-sized movie,
containing only a part of the whole picture, can be
recorded at the same time as a full-sized movie. The EDICAM
is able to handle up to 4 ROIs. It resembles a TVcamera,
transmitting a football game, with four built-in
slow-motion cameras, zooming into different areas of the
playing field. This feature solves the speed problem: while
the whole inner wall is monitored at a low speed, some
components of the machine can be viewed 100–1000
times faster, providing a really fast reaction time for the
event detection. Speed, light intensity, and the NDR capability
are much more important for scientists than color;
therefore, the special camera sensor gives a black-andwhite
image. However, it can differentiate between 4096
shades of gray, instead of the 256 in everyday cameras.
The camera is shown in Fig. 1.
Stellarator News -3- June 2015
However, processing fast camera video streams in real
time requires very high computational power. Therefore,
the EDICAM system uses field programmable gate arrays
(FPGAs) for data processing as well as the camera control
itself. These FPGA advanced microchips are as fast as real
hardware chips, but their internal structure and hence
functionality are determined by a program code loaded at
startup. This allows us to develop new features and apply
them without hardware modifications. An overview of the
camera system as installed on W7-X is shown in Fig. 2.
Figure 3 shows the view of a typical camera inside the
vacuum vessel.
The most important challenge our scientists now face is to
install the 10-camera system on W7-X and to get the system
running by the time the first plasma is formed in the
vacuum chamber of the stellarator.
T. Szepesi, G. Bodnár, G. Cseh, G. Kocsis, T. Ilkei, T. Szabolics,
S. Zoletnik
Wigner Research Centre for Physics, Budapest, Hungary
E-mail: szabolics.tamas@wigner.mta.hu
Fig. 1. The camera head of the EDICAM diagnostic system
will be situated close to the plasma. This hardware is
responsible for taking the images inside W7-X and transferring
image data through a 100-m-long optical cable to
the diagnostic hall, where one computer for each camera
gathers the data and processes it for transfer to a live
video stream in the control room.
Fig. 2. The video diagnostic system at W7-X: Ten cameras,
located in the ports with blue lines, will have the
viewing cones shown in green and observe the complete
interior of the vessel.
Fig. 3. View from the perspective of one camera taken
during assembly. The heat sinks of the plasma vessel wall
are visible at left. The rod in the center is the manipulator
for magnetic flux surface measurement.s The three red circles
mark fiducial points for calibration.