Abstract: A protective control unit for controlling a highside-output transistor and a lowside-output transistor connected in series is provided. The highside-output transistor has a first main electrode region connected to a power supply, a second main electrode region and a first control electrode. The lowside-output transistor has a third main electrode region connected to the second main electrode region, a fourth electrode region connected to ground and a second control electrode. And an inductive load is connected to a connecting point between the second and the third electrode regions. The protective control unit of the present invention has a highside-drive circuit. The highside-drive circuit pulls out charges stored in the highside-output transistor, through the first control electrode, during the periods when the highside-output transistor is in the end of reverse conducting state and reverse recovery state.

Abstract: In the initial stage of the operation of turning off a semiconductor device, the impedance of a carrier pull out circuit remains low for rapidly pulling out the stored carriers from the control electrode of the device. When the turn-off transition of the device proceeds and becomes close to its completion, the impedance of the carrier pull out circuit is shifted to a higher level for retarding the carrier pull out speed. A detector is provided for detecting a control current developed by pulling out the carriers. When the current measured by the detector drops down to below a predetermined level, it is judged that the turn-off transition is approaching to its end. This permits the turn-off transition to be smoothly finished and can thus prevent unwanted oscillation of the control electrode voltage.

Abstract: In the initial stage of the operation of turning off a semiconductor device, the impedance of a carrier pull out circuit remains low for rapidly pulling out the stored carriers from the control electrode of the device. When the turn-off transition of the device proceeds and becomes close to its completion, the impedance of the carrier pull out circuit is shifted to a higher level for retarding the carrier pull out speed. A detector is provided for detecting a control current developed by pulling out the carriers. When the current measured by the detector drops down to below a predetermined level, it is judged that the turn-off transition is approaching to its end. This permits the turn-off transition to be smoothly finished and can thus prevent unwanted oscillation of the control electrode voltage.

Abstract: A protective control unit for controlling a highside-output transistor and a lowside-output transistor connected in series is provided. The highside-output transistor has a first main electrode region connected to a power supply, a second main electrode region and a first control electrode. The lowside-output transistor has a third main electrode region connected to the second main electrode region, a fourth electrode region connected to ground and a second control electrode. And an inductive load is connected to a connecting point between the second and the third electrode regions. The protective control unit of the present invention has a highside-drive circuit. The highside-drive circuit pulls out charges stored in the highside-output transistor, through the first control electrode, during the periods when the highside-output transistor is in the end of reverse conducting state and reverse recovery state.

Abstract: The method disclosed here comprises the steps of (a) mixing metal powder, a resin, abrasive grains, and a solid reducing agent at the normal (room) temperature through the melting point of the reducing agent to form a mixture and (b) molding and baking the mixture at the melting point of the reducing agent through that of the metal powder. The solid reducing agent is a fatty acid, preferably stearic acid having a volume ratio of 5 to 20% with respect to the metal powder. With is, it is possible to make metal-resin bond grindstones that give such high-quality mirror surfaces that have conductivity fit for ELID grinding and are not liable to have chippings or scratches and also have an Rmax value of approximately 3 nm or less.

Abstract: A polysilicon gate layer, a first n.sup.+ diffusion region serving as a drain region, and a second n.sup.+ diffusion region serving as a source region form a MOSFET, and then an operating point of the MOSFET is set into its saturation region by connecting a gate layer and a drain region of the MOSFET. The first and second n.sup.+ diffusion regions provide a first and a second leakage paths, respectively. A temperature sensor can be provided by use of the event that a leakage current flowing through the second leakage path is varied according to a substrate temperature. According to such configuration, scatter of detected temperatures due to scattering in manufacturing process can be reduced even if all scattering parameters in manufacturing process are considered. In addition, an required area of the temperature sensor can be made smaller since a high resistance value is not needed.