Concept to reduce the measurement error of the metal resistive bolometer through remedying the neutral gas pressure effect

Summary form only given. The thin detector foil of the widely used metal resistive-type bolometer is composed of an absorber, a substrate and two meander resistors. The two resistors on the detector foil together with other two identical ones coated on a reference foil form a Wheatstone bridge. Absorption of plasma radiation causes a temperature-induced resistivity increase of the detector foil, driving an unbalance in the Wheatstone bridge. Since the bolometers sometimes lie in an environment of a significant neutral gas background, the bridge can be affected by the neutral gas pressure. To minimize this effect, the bolometer heads are constructed with a pressure compensation mechanism. In this work, the metal resistive bolometers, including the types being used on ASDEX-upgrade and JET, as well as those planned for W7-X and ITER, are investigated under different neutral gas pressure conditions. The foil either possesses a Kapton or mica substrate, carrying a gold absorber, or a silicon nitride substrate, carrying a platinum absorber. The gas pressure is increased up to 20Pa, covering the divertor conditions expected for W7-X and ITER. An evident pressure effect has been observed for all the kinds of the investigated bolometers. The induced bridge signals are converted to an equivalent plasma-radiated power flux based on the bolometer equation and the detector parameters, which are calibrated for each measurement condition by ohmic heating. An error factor, defined as signal equivalent power flux per unit pressure change of about 2.0-8.0Wm^-2Pa^-1 is obtained. For the W7-X and ITER divertor bolometers, presumably suffering from a pressure change up to 0.5Pa and 10Pa, respectively, non-negligible measurement errors were derived. The study results indicate that the compensation still needs to be improved further for cases with high background pressures, as in the divertor region. This effect usually is negligible for the bulk plasma bolometers lo- ated outside the divertor where the neutral gas pressure is usually much lower. The work presents also an analysis of the reasons for the effect described above. It is concluded that the observed neutral gas pressure effect arises from strain and gas-cooling induced meander resistance deviations between the detector and the reference foil. Cooling rates of the foils at different gas pressures have been calculated according to the realistic geometry of the bolometer heads. Concepts for remedying the effect, based on modification of the bolometer head structure as well as reducing the gas-cooling contribution to the total cooling rates of the foils, are proposed.