Electromagnetic analysis for fusion reactors: status and needs Page: 2 of 6
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
ELECTROMAGNETIC ANALYSIS FOR FUSION REACTORS: STATUS AND NEEDS*
Larry R. Turner
Argonne National Laboratory
9700 South Cass Avenue
Argonne, Illinois 60439
Electromagnetic effects have far-reaching
implications for the design, operation, and maintenance
of future fusion reactors. Two-dimensional (2-D) eddy
current computer codes are available, but are of
limited value in analyzing reactors, Three-dimensional
(3-D) codes are needed, but are only beginning to be
developed. Both 2-D and 3-D codes need verification
against experimental data, such as that provided by the
upcoming FELIX experiments. Coupling between eddy cur-
rents and deflections has application In fusion reactor
design and is being studied both by analysis and exper-
iment.
Introduction
The consequences of electromagnetic effects
In future fusion reactors will be qualitatively dif-
ferent from those of existing fusion experiments„ even
from such large-scale experiments as TFTR and JET.
Design of the reactor in response to these electromag-
netic concerns has far-reaching implications In the
configuration, operation, and maintenance of the
reactor.
This paper describes the Impact of electro-
magnetic effects on future reactors, the status of com-
puter codes for analyzing those effects, the need for
experiments to validate the codes, and the FELIX exper-
iments which will provide the needed data for code
validation. Finally, the coupling between eddy cur-
rents and deflections is discussed, a topic of impor-
tance for fusion reactor design and one for which
analysis and experimentation is underway.
Impact of Electromagnetic Effects on Future Reactors
There are two major differences between the
electromagnetic effects in existing devices and in
future fusion experiments and reactors. First, in
existing devices, the induced eddy currents are almost
entirely toroidal while in future tokamaks, the cur-
rents will be more complex, with pololdal components
and possibly radial components as well. Second, plas-
mas In existing tokamaks are surrounded by thin vacuum
vessels, with short time constants with very little
delay and perturbation of the pololdal field. Plasmas
In future tokamaks will be surrounded by thick FWBS
structures, which will introduce sizable pololdal field
delays and perturbations.
Most existing fusion experiments, TFTR and
JET in particular, have bellows or other segments to
provide continuous but high resistance first wall
current paths. The high resistance bellows decrease
the L/R time constant of the first wall, facilitate
magnetic flux penetration, and tend to limit the in-
duced current flowing in the first wall; but to first
order they do not change or constrict the current
path. Induced currents flow toroldally, do not inter-
act with the toroidal field, and can to fair approxima-
tion be modelled by a relatively small number of
coaxial current loops. Other fusion devices, such as
PLT, have a dielectric break In the vacuum vessel to
prevent circulating currents. Opposite currents flow
on the inside and outside surfaces of the vessel, and
the currents decay even more quickly, but their analy-
sis is similarly fairly straightforward.
* Work supported by the U. S. Department of Energy.
By contrast, future fusion reactors will have
segmented FWBS components to permit remote maintenance
and rapid replacement of FWBS sectors. Induced cur-
rents in the first wall will flow from sector to sector
at a limited nunber of electrical connectors. Current
paths will not be strictly toroidal: there wilJ be
pololdal current components where the current con-
stricts near the connectors. If maintenance and
materials considerations force the connectors to be
located at the back of the FWBS sectors rather than at
the first wall, there will be radial currents as
well, (Concerns about arcing may eliminate this design
alternative, as discussed below,) These radial and
pololdal currents and the non-axisymmetric toroidal
currents will introduce transient field ripple Into the
plasma region, possibly up to many hundreds of gauss in
magnitude. Whereas toroidal currents interact only
with the pololdal field, the radial and pololdal
currents will Interact with the toroidal field,
typically ten times stronger, and produce severe and
complex distribution of forces and torques.
The segmentation of the FWBS iystem to faci-
litate remote maintenance also introduces the possi-
bility of electrical arcing between segments during a
plasma disruption. If arcing welds neighboring FWBS or
limiter sectors together, removal and replacement of
the sectors may result in months of delay. The danger
of arcing may well require that the electrical con-
nectors between sectors lie no more than 10 cm from the
rst wall despite the increased thermal, radiation,
and maintenance difficulties that result. Even if
arcing does not weld neighboring sectors together, it
can provide a low-resistance current path between them
and produce eddy currents, forces, and torques far
higher than those designed for the nominal case of no
arcing.
The field from the EF colls and flux from the
OH colls must penetrate Into the plasma region without
undue delay or distortion. In existing fusion experi-
ments, the major encumbrance to field penetration Is
the thin and hlgh-reslstance first wall, deliberately
designed to have a short L/R time constant (0.5 ms in
JET, 3 ms for TFTR) in order to minimize electromag-
netic delay.
However, in future fusion reactors, the major
encumbrance to field penetration will be the thick
blanket and shield sectors, truly three-dimensional
bodies, with large thickness relative to the skin depth
and long L/R time constants (100 - 300 ms for INTOR and
STARFIRE). Moreover, the gaps between sectors will be
small and convoluted to prevent neutron streaming. The
computation of a complete 3-D system with narrow gaps
is beyond existing capability.
Finally, non-axisymmetric conducting first
wall segments, electrical connectors between those
segments, and active or passive saddle coils enclosing
the plasma region may be needed for the control of
plasma position. However, these conducting elements
may interact electroroagnetically with the changing
magnetic fields from the PF colls or from a plasma
disruption to create problems of force restraint or
power demand which would be highly expensive or even
impossible to solve through conservative design. At
present, we cannot adequately determine either if such
conducting elements are needed or what complications
they may introduce.
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Turner, L. R. Electromagnetic analysis for fusion reactors: status and needs, article, January 1, 1983; Illinois. (https://digital.library.unt.edu/ark:/67531/metadc1067015/m1/2/?rotate=270: accessed March 18, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.