APPENDIX A Thermal Response Measurements for Semiconductor Device Die Attachment Evaluation A.1 General Considerations Steady-state thermal response (or thermal resistance) and transient thermal response of semiconductor devices are sensitive to the presence of voids in the die attachment material between the semiconductor chip and header since these voids impede the flow of heat from the chip to the case (header). Due to the difference in the thermal time constants of the chip and case, the measurement of transient thermal response can be made more sensitive to the presence of voids than can the measurement of steady-state thermal response. This is because the chip thermal time constant is generally several orders of magnitude shorter than that of the case. Thus, the heating power pulse width can be selected so that only the chip, and the chip-to-case interface, are heated during the pulse by using a pulse width somewhat greater than the chip thermal time constant but less than that of the case. Heating power pulse widths ranging from 5 to 10 ms have been found to satisfy this criterion. This enables the detection of voids to be greatly enhanced, with the added advantage of not having to heatsink the device under test. A.2 Procedure To compare relative quality of the die bond, if it is assumed that the temperature coefficient of the TSP is the same for all members of a given group of devices of the same type or lot, it is possible to eliminate the conversion of the TSP to temperature in the calculations. This assumption is generally valid for semiconductor devices where the forward voltage of a p-n junction is used as the TSP. It is therefore necessary only to measure the TSP first under conditions of no internal power dissipation and then at some specified time after the termination of the heating power pulse. A quantity proportional to the junction-to-reference point temperature difference is obtained by subtracting the second voltage, V M2' from the first, VMI. This allows the transient thermal response technique to be made less time consuming for use as a manufacturing screen or process control measurement for die attachment evaluation. APPENDIX B Reference Point Temperature Measurements of Conduction Cooled Power Transistors. B.1 Case Temperature (TC) The case temperature of a stud-mounted, hexagonal-base transistor is measured at the center of any flat on the hex. The case temperature of other base-mounted transistors is measured at a point specified by the manufacturer. The recommended procedure for measuring the case temperature of a transistor employs the use of a thermocouple and is defined as follows: B.1.1 Type of Thermocouple - The thermocouple material shall be copper-constantan (Type T). Its useful temperature range for standard temperature measurements is -183 to +371°C. The wire size shall be no larger than #30 AWG. The junction of the thermocouple shall be welded together to form a bead rather than soldered or twisted. [See "1971 Annual Book of ASTM StandardsPart 30, Method E220 for Calibration of Thermocouples by Comparison Techniques" for information on construction and usage of thermocouples.] B.1.2 Mounting Method - A small hole, just large enough to insert the thermocouple, shall be drilled approximately 0.762 mm (0.03 in) deep in the base plate of the semiconductor device at the point specified by the manufacturer. The edge of the hole should then be peened with a small center punch to form a rigid mechanical contact with the welded bead of the thermocouple. In the event that drilling into the base plate of the device case becomes impractical because of case material or case dimensions, the thermocouple wire may be welded or soldered directly to a specified point on the case. B.1.3 Accuracy - An accuracy of ±0.5°C should be expected of the thermocouple and associated measuring system. Under load condition, slight variations in the temperature of different points on the case may reduce the accuracy to ±1.0°C for still-air cooling, and ±2.0°C for forced-air cooling. B.1.4 Other Thermocouple Mounting Methods - Other methods of mounting thermocouples, with the possible exception of the thermocouple welded or soldered directly to the case, usually result in temperature readings lower than the actual temperature. Such deviations result from: A. Inadequate contact with the case when using cemented thermocouples. B. External heat dissipator in contact with the thermocouple when using pressure contacts. B.2 Mounting Surface Temperature (TM) The mounting surface temperature for EIA Registered Power Transistor packages is measured using a thermocouple imbedded in a washer. The use of a washer, with a thermocouple imbedded, offers the best single method of non-destructive testing that is compatible with most package types. [The case and mounting surface temperatures are sometimes used interchangeably.] B.2.1 Type of Thermocouple - The thermocouple material shall be copper-constantan (Type T). Its useful temperature range for standard temperature measurements is -183 to +371 °C. The wire size shall be no larger than #30 AWG. The junction of the thermocouple shall be welded together to form a bead rather than soldered or twisted. [See “1971 Annual Book of ASTM Standards Part 30, Method E220 for Calibration of Thermocouples by Comparison Techniques" for information on construction and usage of thermocouples.] B.2.2 Mounting Washer Construction - For all registered TO outlines the following general rules apply: a. The base material of the washer shall be copper. (Preferred half hard or softer). b. The thickness of the washer shall be 3.175 ± 0.127 mm(0.125 ± 0.005 in). c. The outline of the washer shall be larger by 0.762 to 1.524 mm (0.03 to 0.06 in) than the outline of the scating surface of the package for which the washer is intended. d. Clearance holes shall be 0.406 to 0.787 mm(0.016 to 0.031 in) larger than the maximum outside diameter of the studs or screws intended to pass through the holes. e. Surfaces of the washer shall be flat within 25 μm per 25 mm (0.001 in per in) and parallel within 75 μm per 25 mm (0.003 in per in) and shall be nickel plated to a thickness of 1.27 to 2.54 μm (50 to 100 μin). [See "ANSI B46.1 – 1962, Surface Texture" for further details.] f. The surface of the washer shall be free from burrs, but the maximum chamfering of edges or holes shall not exceed 0.406 mm(0.016 in) by 45 deg. so as not to effectively reduce the contact area of the washer. g. Both surfaces of the washer shall have a 1.60 μm (63 μin) finish or better and be free of oxides. [See “ANSI B46.1 – 1962, Surface Texture" for further details.] h. The thermocouple hole shall be drilled into the washer midway between and parallel to the top and bottom surfaces. The size of the thermocouple hole shall be no greater than 1.52 mm (0.06 in) in diam. but it is recommended that it be no larger than necessary to accept the thermocouple. i. For flat type packages (such as TO-3) the bottom of the thermocouple hole shall extend approximately 0.76 mm(0.03 in) beyond the geometric center of the washer. Radial orientation of thermocouple hole is arbitrary. j. For stud type packages (such as T0-61) the bottom of the thermocouple hole shall be approximately 0.76 mm(0.03 in) from the inside hole of the washer. k. For tab type packages the bottom of the thermocouple hole shall extend approximately 0.76 mm(0.03 in) beyond the geometric center of the seating surface. 1. It is recommended that the thermocouple be secured into the washer with a thermal conducting adhesive and that particular attention be paid to minimizing air voids around the ball of the thermocouple. (The thermocouple bead should be in direct contact with the copper washer.) m. Clearance holes for device leads should allow suitable clearance to prevent electrical shorting to the washer. It is recommended that this clearance hole be approximately 1.52 mm (0.06 in) larger in diameter than the lead to allow clearance for insulating sleeving to be used on the leads. n. Device mounting torque should comply with the manufacturer's recommendations. o. A thermal conducting compound at the interface of the washer and device should be used. p. Special precautions must be taken so that only the bead of the thermocouple is allowed to come in mechanical contact with the washer. APPENDIX C Extrapolation Procedure for Power Transistors C.1 Extrapolation Procedure This extrapolation procedure is based on the assumption that the heat source thickness is small compared to the total chip thickness and that for approximately the first 200 or 250 μs of cooling, one dimensional heat flow occurs. Therefore, the first 200 or 250 microseconds of cooling can be represented by TJ (Cooling)(t) = TJ (S.S.) —Kt1⁄2 where K is approximately constant and TJ(S.S.) is the steady state junction temperature. (1) If, for equation (1), TJ(Cooling)(t) versus t2 is plotted on linear graph paper, the generated curve will be a straight line with TJ(S.S.) as the Y - intercept, i.c., at t=0. Plotting this curve for an actual device also provides a means for determining when non-thermal switching transients are significant since the curve, as plotted, will be non-linear under this condition. Using equation (1), a mathematical expression which can be used to generate the extrapolated valur TJ(S.S.) (t = 0) is a follows: |