The non-uniformity of the target temperature due to micrometric hot spots (Hermann et al 2004 Phys. Scr. T 111 98) is an explanation for the experimental fact that near-infrared measurements yield higher temperature values than mid-infrared measurements (Hildebrandt et al 2003 InfraMation 2003 Proc. (Las Vegas, USA, October 2003), Delchambre et al 2005 J. Nucl. Mater.337–339 1069). The issue of micrometric hot spot disturbance in the surface temperature (Tsurf) measurement and heat load calculation is addressed in this paper. The theoretical investigation at 3, 5 and 12 µm and experiments in the range 3.5–5 µm indicate that the surface state can play an important role in the non-uniform heating surface and consequently in the overestimation of the bulk temperature. The contribution of the hot spots to temperature measurements and flux calculations has been simulated at different wavelengths. Calculations show that (1) the overestimation of the bulk temperature decreases with the wavelength and (2) the overestimation depends on the temperature difference, ΔT, between the bulk and the micrometric hot spots. In addition, experiments have been carried out in order to compare the flux calculations at different wavelengths on different graphite (polished, dusty). The results obtained are very sensitive to the surface state pointing out the difficulties in improving the heat flux calculation model, since the surface state can change during the plasma discharges. This paper shows that the problem of non-homogenous surface temperature can be significantly diminished on working at longer wavelengths.
Evidence of hot spot formation on carbon limiters due to thermal electron emission