One of the main missions of KSTAR is to develop long-pulse operation capability relevant to the production of fusion energy. After a full metal wall configuration was decided for ITER, a major upgrade for KSTAR was planned, to a tungsten first wall similar to the JET ITER-like wall (coatings and bulk tungsten plasma-facing components). To accomplish the upgrade, tungsten bonding technology has been developed and tested. Since the leading edges of each castellation structure have to be protected, shaping of tungsten blocks has been studied by ANSYS simulation, and the miniaturized castellation has been exposed to Ohmic plasma to confirm the simulation results. It is found that a shaped castellation block has more heat handling capability than a conventional block. For more dedicated experiments, a multipurpose castellation block is fabricated and exposed to Ohmic, L- and H-mode plasmas and observed by IR camera from the top. During the fabrication and assembly of the blocks, leading edges caused by “naturally misaligned” blocks due to engineering limits with a maximum level up to 0.5 mm have been observed, and these have to be minimized for the future fusion machine.