Thermal load distribution near the tips of the ALT-II limiter roof on TEXTOR-94

Abstract

The aim of these investigations is the description of the power flux on roof-like limiters near the LCFS. Here, the specific subject is the toroidal pump limiter ALT-II of TEXTOR-94. Similar to other plasma facing objects, the surface of ALT-II is shaped to enhance the plasma wetted area; this is achieved by reducing the angle of incidence of the magnetic field to less than 1 • for the first 10 mm of the SOL. This small angle of incidence enhances all effects of toroidal non-uniformity as given, for example, by magnetic field ripple. Extensive modelling explains well the observed heating pattern on the limiter surface due to the ripple effect. In contrast to the expectations from density and temperature distributions in the SOL and at the edge of the confined region, an excessive power density is deposited on the first few millimetres near the tip of the limiter roof. Physical effects which could cause this phenomenon are discussed.

Thermal load distribution on the ALT-II limiter of TEXTOR-94 during RI mode operation and during disruptions

“…As we had already discussed previously [1], the temperature on the top and bottom row of the graphite tile is not simply symmetric around the tangent line in the centre of the blade (45 • ) even though the plasma is well adjusted. The problem arises from the scalar product n · b of the surface normal vector with the magnetic field vector, which contains field ripple as an additional complication [1]. This field ripple leads to a self-shadowing of the incoming power flux (private flux zone with a connection length of the order of one metre).…”

“…3 many different edge plasma conditions are shown but nevertheless the effect is always present and the power fraction in the 'short decay length part' is always substantial and amounts to roughly 50%. The 'short power decay length' effect was first described on TEXTOR-94 [1]; ALT-II probably had to have a very suitable tile shape in order to discover the effect. Nevertheless, other limiter tokamaks also showed a similar enhancement of the heat flux close to the last closed flux surface.…”

“…4 of Ref. [1]. The heating pattern is a combined effect of the power decay function away from the last closed flux surface and the angle of incidence with respect to the tile surface.…”

“…n · b (1) where n is the vector normal to the limiter surface and b is the unit vector in the direction of the magnetic field. The surface normal vector is obtained from knowledge of the tile geometry; as the surface normal vector is nearly orthogonal to the local magnetic field unit vector (the angle amounts to about 86-89.4 • ), many details of the magnetic field have to be taken into account such as I p , β pol and plasma shift.…”

“…As in the previous publication [1], it was tried to fit the flux density function of Fig. 2(c) again with two exponential functions (neglecting the excessive heat around ∆r = 15 mm, which we attribute to reflections); we subtract at first the radiative fraction of the tile heating which we obtain from bolometric measurements (0.05 MW/m 2 ).…”