Alcator C-Mod Weekly Highlights Oct 4, 1999 Last week began a maintenance period at Alcator C-Mod. No plasma runs were scheduled. This week will also be a maintenance week. A detailed survey of TF magnet resistance measurements, using a quiet 100A DC supply, was carried out as part of the investigation of the increased values noted during operation the previous week. These measurements showed a pattern consistent with the post-shot scanner readings, although the maximum observed value of 2uOhm was somewhat lower. A series of 50kA and 100kA TF pulses were run, with instrumentation on 8 turns of the magnet in the vicinity of H-port, where the elevated readings occurred. The resistances measured during these high-current pulses are all similar, and within a range of 1.2 to 1.8uOhm. These measurements remained essentially constant over the 3 second long flat-top of the pulse and showed no abnormal behavior during ramp-up and ramp-down. A visual inspection revealed the presence of frost on several metal fittings inside the C-Mod cryostat. The inspection was limited to the area visible through the double plexiglass window at the top of the cryostat; the magnet itself is not viewable through this port. It is believed that a leaky gasket is responsible for allowing some moisture into the dewar. We are presently warming up the machine, and will inspect and repair any leaky seals during the present maintenance interval. Warming the entire machine to room temperature is expected to take of order one week. Physics & Analysis ------------------- The x-ray emissivity shows a pronounced pedestal in H-mode, which, at the outboard midplane, is typically located 5 mm inside the pedestals of electron density and temperature. This shift can be explained by a strong inward convection of impurities, localized to the edge of the plasma. Neoclassical theory predicts a strong inward pinch of impurities where the plasma density gradient is large, i.e. in the electron density pedestal region. One-dimensional simulations using the impurity transport code MIST show that the strong pinch must extend to roughly the top of the x-ray emissivity profile but not further inward than that, and that the width of the x-ray emissivity pedestal is mostly a function of D, the impurity diffusion coefficient. Using measurements of the top of the xray pedestal, it is found that the inward pinch region does coincide with the region of strong electron density gradients. D values inferred from the x-ray pedestal width are in good agreement with D values inferred from previous impurity injection experiments. During several recent runs, CaF2 was injected by laser blowoff. The edge xray arrays are sensitive to radiation from fluorine, and can therefore detect the CaF2 injection and observe its transport on a fast timescale (12 us sampling). During the past year, obvious asymmetries between the top and outboard xray pedestals have been consistently observed in steady-state, and this is thought to be due to the same neoclassical return flow that was proposed to explain the strong top/bottom asymmetry of Ar17+ reported previously (Rice, et al.) The recent CaF2 injection experiments show a transient (200 us) burst of xrays in the pedestal region at the top of the plasma, but no simultaneous burst from the outboard pedestal. The injected impurities pass through the views of both xray arrays, but apparently the flow quickly sweeps the impurities up to the top of the plasma before they begin to emit xrays. Given the relevant ionization rates, we find that the neoclassical flow must therefore be of the order of 10 km/s. Central impurity toroidal rotation was measured during a current scan of plasmas with constant target density and ICRF power at 2 MW. The range of currents achieved was from 0.4 to 1.4 MA. The ratio of the observed rotation to the plasma stored energy increase during the ICRF pulse was determined to be a decreasing function of plasma current. This is qualitatively similar to the predictions of neo-classical theory which has the toroidal rotation velocity inversely proportional to the poloidal magnetic field. ICRF System ------------ Based upon a cavity model, we have proceeded to make a modification to FMIT#1. This modification requires replacing the bottom third of the cavity with a 3/8" greater radius than the original. This change is expected to result in an ~30% decrease in plate impedance, allowing higher power for a given voltage. If this modification works as expected, we will make the same modification to FMIT#2. Testing will begin this week. The modification to J-port resonant loops is nearing completion. Modeling indicates that the addition of 1/2 wavelength to the loops, as well as modifications to the decoupling loop, will permit operation with (0,pi,0,pi) phasing, which should correspond to an improvement in the k-parallel spectrum for heating. The physical layout is also ready to proceed with the modification. We expect to have the new J-port configuration assembled and tested before next plasma operation. The low power RF chain in FMIT#3 and #4 is also being upgraded. New power supplies have been obtained and are being installed to eliminate a temporary power supply. This should increase the system's reliability and simplify the overall circuit. Travel and Visitors ------------------- Amanda Hubbard, Martin Greenwald, and Joe Snipes took part in the IAEA Technical Committee Meeting on H-mode and Internal Transport Barrier Physics in Oxford, held 27-29 September. They also participated in the Confinement, Threshold, and Edge Pedestal database meetings at JET on 30 Sept- 1 Oct.