AN1028 April, 1996 Maximum Power Enhancement Techniques for SOT-223 Power MOSFETs Alan Li, Brij Mohan, Steve Sapp, Izak Bencuya, Linh Hong 1. Introduction As packages become smaller, achieving efficient thermal performance for power applications requires that the designers employ new methods of meliorating the heat flow out of devices. Thus the purpose of this paper is to aid the user in maximizing the power handling capability of the SOT223 Power MOSFET offered by Fairchild Semiconductor. This effort allows the user to take full advantage of the exceptional performance features of Fairchild’s state-of-the-art Power MOSFET which offers very low on-resistance and improved junction-to-case (RθJC) thermal resistance. Ultimately the user may achieve improved component performance and higher circuit board packing density by using the thermal solution suggested below. In natural cooling, the method of improving power performance should be focused on the optimum design of copper mounting pads. The design should take into consideration the size of the copper and its placement on either or both of the board surfaces. A copper mounting pad is important because the drain lead of the Power MOSFET is mounted directly onto the pad. The pad acts as a heatsink to reduce thermal resistance and leads to improved power performance. D S D G Figure 1. SOT-223 Power MOSFET achieves junction-to-case thermal resistance RθJC of 12oC/W. 2. Theory When a device operates in a system under the steady-state condition, the maximum power dissipation is determined by the maximum junction temperature rating, the ambient temperature, and the junction-to-ambient thermal resistance. PDmax = ( TJmax - TA ) / RθJA (2.1) The term junction refers to the point of thermal reference of the semiconductor. Equation 2.1 can also be applied to the transient-state: PDmax (t) = [ TJmax - TA] / RθJA(t) (2.2) Rev B, August 1998 1 where PDmax (t) and RθJA(t) are time dependent. By using the transient thermal resistance curves shown in the data sheet, a transient temperature change can be calculated. The transient thermal behavior is a complicated subject because RθJA(t) increases non-linearly with time and the conditions of the power pulse. A more thorough treatment of transient power analysis is beyond the scope of this document and the reader can refer to [13] for details. Nevertheless, Fairchild provides a Discrete SPICE Thermal Model (LIT# 570240-002) for general thermal evaluation. The user may find these models helpful in determining the dynamic power and temperature limits in the application. RθJA has two distinct elements, RθJC junction-to-case and RθCA case-to-ambient thermal resistance. RθJA = RθJC + RθCA (2.3) The case thermal reference of the SOT-223 Power MOSFET is defined as the point of contact between the drain lead of the package and the mounting surface. RθCA is influenced by many variables such as ambient temperature, board layout, and cooling method. Due to the lack of an industry standard, the value of RθCA is not easily defined and can affect RθJA significantly. In addition, the case reference may be defined differently by various manufacturers. Under such conditions, it becomes difficult to define RθCA from the component manufacturer standpoint. On the other hand, RθJC is independent of users’ conditions and can be accurately measured by the component manufacturer. Therefore, in this paper an effort has been made to define a procedure which can be used to quantify the junction-to-ambient thermal resistance RθJA which is more useful to the circuit board designer. 3. Result The scope of the investigation has been limited to the size of copper mounting pad and its relative surface placement on the board. In still air with no heatsink, the application of these heat dissipation methods is the most cost effective thermal solution. A total of sixteen different combinations of 2 Oz copper pad sizes and their placement were designed to study their influence on RθJA thermal resistance. The configurations of the board layout are shown in figure 2 and table 1. Layouts 1 to 6 have the copper pad sizes from 0.0123 to 1 square inches on the top side of the board (top side is defined as the component side of the board). Layouts 7 to 11 have copper pad sizes from 0.2 to 1 square inches on the bottom side of the board. Layouts 12 to 16 have copper pad sizes from 0.132 to 1 square inches divided equally on both sides of the board. BOTTOM VIEW TOP VIEW Figure 2. Both Sides of the 4.5”x5” SOT-223 Thermal Board. Complete scale drawings are shown in Section 5. 2 L ayo u t 2 Oz Copper Mounting Pad Area (in2) Relative Placement on Board 1-6 0.0123, 0.066, 0.03, 0.53, 0.76, 1 Top 7-11 0.2, 0.4, 0.6, 0.8, 1 Bottom 12-16 0.132, 0.35, 0.568, 0.784, 1 1/2 Top and 1/2 Bottom Table 1: Thermal Board Configurations RθJA was calculated from the relationship between power and the change of junction temperature. If readers are interested in the test conditions and method, they are encouraged to refer to appendix B for details. Rja, JUCTION-TO-AMBIENT o THERMAL RESISTANCE ( C/W) 120 4.5"x5" FR-4 Board o T A = 25 C S till Air 100 80 1/2T op+1/2Bottom Cu Bottom Cu 60 T op Cu 40 20 0 0.2 0.4 0.6 0.8 2oz COPPER MOUNT ING PAD AREA (in 2 ) 1 Figure 3. SOT-223 Junction-to-Ambient thermal resistance versus copper mounting pad area and its surface placement. Plots in figure 3 show the relationship of RθJA versus the copper mounting pad area and its surface placement on the board. It is apparent that increasing copper mounting pad area considerably lowers RθJA from approximately 110 to 40oC/W in the range from 0.0123 to 1 square inches. In addition, placing all the copper on the top side of the board further reduces RθJA by 10 to 15oC/W when compared with the other two placements. By substituting the thermal resistance, ambient temperature, and the maximum junction temperature rating into equation 2.1, the steady-state maximum power dissipation curves can be obtained and are shown in figure 4. A 18% increase in the power handling can be achieved by increasing the copper pad area on top of the board from 0.0123 to 0.066 in2, layout 2. This thermal pad fits directly under the package, so that no additional board space is required. For maximum performance, it is recommended to put extra copper on the bottom of the board connected to the top pad by through-hole thermal vias. 3 S T EADY-S T AT E POWER DIS S IPAT ION (W 3.5 3 T op Cu 2.5 1/2T op+1/2Bottom Cu Bottom Cu 2 1.5 1 Recommended package sized Cu to achieve 1.3W 4.5"x5" FR-4 Board T A = 25o C S till Air 0.5 0 0 0.2 0.4 0.6 0.8 2 2oz COPPER MOU NT ING PAD AREA (in ) 1 Figure 4. Maximum Power Dissipation Curves for SOT-223. In layout 2, 0.066 in² 2 Oz copper mounting pad area is recommended to achieve approximately 1.3W. 4. Conclusion Fairchild Semiconductor has attempted to define the thermal performance of the SOT-223 Power MOSFET, from a systems point of view. It has been demonstrated that significant thermal improvement can be achieved in the maximum power dissipation through the proper design of copper mounting pads on the circuit board. The results can be summarized as follows: • • • Enlarged copper mounting pads, on either one or both sides of the board, are effective in reducing the case-to-ambient thermal resistance R θCA. Placement of the copper pads on the top side of the board gives the best thermal performance. The most cost effective approach of designing layout 2 0.066 square inches copper pad directly under the package, without occupying additional board space, can increase the maximum power from approximately 1.1 to 1.3W. 4 5. SOT-223 Thermal Board Top View in Actual Scale 5 SOT-223 Thermal Board Bottom View in Actual Scale 6 Appendix A Heat Flow Theory Applied to Power MOSFETs When a Power MOSFET operates with an appreciable current, its junction temperature is elevated. It is important to quantify its thermal limits in order to achieve acceptable performance and reliability. This limit is determined by summing the individual parts consisting of a series of temperature rises from the semiconductor junction to the operating environment. A one dimensional steady-state model of conduction heat transfer is demonstrated in figure 5. The heat generated at the device junction flows through the die to the die attach pad, through the lead frame to the surrounding case material, to the printed circuit board, and eventually to the ambient environment. There are also secondary heat paths. One is from the package to the ambient air. The other is from the drain lead frame to the detached source and gate leads then to the printed circuit board. These secondary heat paths are assumed to be negligible contributors to the heat flow in this analysis. RθJ A = RθJ C + RθC A TJ-TA = P D * RθJ A Lead Frame Die Case Reference for thermal couple in RθJC measurement Junction Reference RθC A S our ce, G ate Mounting Pad (Poor Thermal Path) RθJ C MoldedP ackage o T A = 25 C Boa rd Via Extended Copper Pla ne RθJ C (C omponent Variables ) L eadframe Size & Material No. of Conduct ion Pins Die Si ze Die Attach Material Molding Compound Size & Material D rain Mounting Pad RθC A(Applications Variables ) Mounting Pad Size, Mater ial, Shape &L ocation P lacement of M ounting Pad PCB Size & Material Amount of thermal Via Traces Length & Width Adjacent Heat Sources Air Flow Rate and Volume of Air Ambient Temperature ......e tc Figure 5: Cross-sectional view of a Power MOSFET mounted on a printed circuit board. Note that the case temperature is measured at the point where the drain lead(s) contact with the mounting pad surface. The increase of junction temperature above the surrounding environment is directly proportional to dissipated power and the thermal resistance. The steady-state junction-to-ambient thermal resistance, RθJA, is defined as RθJA = ( TJ - TA ) / P where TJ is the average temperature of the device junction. The term junction refers to the point of thermal reference of the semiconductor device. TA is the average temperature of the ambient environment. P is the power applied to the device which changes the junction temperature. RθJA is a function of the junction-to-case RθJC and case-to-ambient RθCA thermal resistance RθJA = RθJC + RθCA Rev B, August 1998 7 where the case of a Power MOSFET is defined at the point of contact between the drain lead(s) and the mounting pad surface. RθJC can be controlled and measured by the component manufacturer independent of the application and mounting method and is therefore the best means of comparing various suppliers component specifications for thermal performance. On the other hand, it is difficult to quantify RθCA due to heavy dependence on the application. Before using the data sheet thermal data, the user should always be aware of the test conditions and justify the compatibility in the application. Appendix B Thermal Measurement Prior to any thermal measurement, a K factor must be determined. It is a linear factor related to the change of intrinsic diode voltage with respect to the change of junction temperature. From the slope of the curve shown in figure 6, K factor can be determined. It is approximately 2.2mV/oC for most Power MOSFET devices. NDS9956 VSD vs Temperature 0.7 VGS = 0V VSD (V) 0.6 0.5 I SD = 20mA 10mA 5mA 2mA 0.4 0.3 0.2 25 1mA = 2.39 mV/°C 2mA = 2.33 5mA = 2.25 10mA = 2.19 20mA = 2.13 50 1mA 75 100 125 150 Temperature (°C) Figure 6. K factors, slopes of a VSD vs temperature curves, of a typical Power MOSFET After the K factor calibration, the drain-source diode voltage of the device is measured prior to any heating. A pulse is then applied to the device and the drain-source diode voltage is measured 30us following the end of the power pulse. From the change of the drain-source diode voltage, the K factor, input power, and the reference temperature, the time dependent single pulsed junction-toreference thermal resistance can be calculated. From the single pulse curve on figure 7, duty cycle curves can be determined. Note: a curve set in which RθJA is specified indicates that the part was characterized using the ambient as the thermal reference. The board layout specified in the data sheet notes will help determine the applicability of the curve set. 8 1 r(t), NORMALIZED EFFECTIVE TRANSIENT THERMAL RESISTANCE 0.5 0.2 0.1 0.05 0.02 D = 0.5 0.2 RθJA (t) = r(t) * RθJA RθJA = SeeNote 1a, b, c 0.1 0.05 0.02 P(pk) 0.01 0.01 t1 SinglePulse 0.005 0.002 0.001 0.0001 t2 TJ - T A = P * RθJA (t) Duty Cycle, D = t1 /t2 0.001 0.01 0.1 t1, TIME (sec) 1 10 100 300 Figure 7. Normalized Transient Thermal Resistance Curves B.1 Junction-to-Ambient Thermal Resistance Measurement Equipment and Setup: • Tesec DV240 Thermal Tester • 1 cubic foot still air environment • Thermal Test Board with 16 layouts defined by the size of the copper mounting pad and their relative surface placement. For layouts with copper on the top and bottom planes, there are 0.02 inch copper plated vias (heat pipes) connecting the two planes. See figure 2 and table 1 on the thermal application note for board layout and description. The conductivity of the FR-4 PCB used is 0.29 W/m-C. The length is 5.00 inches ± 0.005; width 4.50 inches ± 0.005; and thickness 0.062 inches ± 0.005. 2Oz copper clad PCB. The junction-to-ambient thermal measurement was conducted in accordance with the requirements of MIL-STD-883 and MIL-STD-750 with the exception of using 2 Oz copper and measuring diode current at 10mA. A test device is soldered on the thermal test board with minimum soldering. The copper mounting pad reaches the remote connection points through fine traces. Jumpers are used to bridge to the edge card connector. The fine traces and jumpers do not contribute significant thermal dissipation but serve the purpose of electrical connections. Using the intrinsic diode voltage measurement described above, the junction-to-ambient thermal resistance can be calculated. B.2 Junction-to-Case Thermal Resistance Measurement Equipment and Setup: • Tesec DV240 Thermal Tester • large aluminum heat sink • type-K thermocouple with FLUKE 52 K/J Thermometer The drain lead(s) is soldered on a 0.5 x 1.5 x 0.05 copper plate. The plate is mechanically clamped to a heat sink which is large enough to be considered ideal. Thermal grease is applied in-between the two planes to provide good thermal contact. Theoretically the case temperature should be held constant regardless of the conditions. Thus a thermocouple is used and fixed at the point of contact between the drain lead(s) and the copper plate surface, to account for any heatsink temperature change. Using the intrinsic diode voltage measurement described earlier, the junction-to-case thermal resistance can be obtained. A plot of junction-to-case thermal resistance for 9 various packages is shown in figure 8. Note RθJC can vary with die size and the effect is more prominent as RθJC decreases. Junction-to-Case Thermal Resistance 80 * Dual Leadframes ** Triple Leadframes rjcall.pre 10/4/95 68 Typical R θ jc ( C/W) 60 * 53.3 o * 40 38.9 30 23.8 20 20.8 ** 17.6 15 13.3 7.4 5 1 0 l 6 K P al le le le -II -3 63 23 ua -1 PA ng ng ng -2 Du OT SO OP OT-2 IC Si Si Si 8D DrS TO TS TS T-6 S SO -8 -6 O-8 pe O T T S O u O O rS S S rS rS pe pe pe Su Su Su Figure 8. Junction-to-case thermal resistance RθJC of various surface mount Power MOSFET packages. 10 References [1] K. Azar, S.S. Pan, J. Parry, H. Rosten, “Effect of Circuit Board Parameters on Thermal Performance of Electronic Components in Natural Convection Cooling,” IEEE 10th annual Semi-Therm Conference, Feb. 1994. [2] A. Bar-Cohen, & A.D. Krauss, “Advances in Thermal Modeling of Electronic Components & Systems,” Vol 1, Hemisphere Publishing, Washington, D.C., 1988. [3] R.T. Bilson, M.R. Hepher, J.P. McCarthy, “The Impact of Surface Mounted Chip Carrier Packaging on Thermal Management in Hybrid Microcircuit,” Thermal Management Concepts in Microelectronics Packaging, InterFairchild Society for Hybrid Microelectronics, 1984. [4] R.A. Brewster, R.A. Sherif, “Thermal Analysis of A Substrate with Power Dissipation in the Vias,” IEEE 8th Annual Semi-Therm Conf., Austin, Tx , Feb. 1992. [5] D. Edwards, “Thermal Enhancement of IC Packages, “ IEEE 10th Annual Semi-Therm Conf., San Jose, Ca, Feb. 1994. [6] S.S. Furkay, “Convective Heat Transfer in Electronic Equipment: An Overview,” Thermal Management Concepts, 1984. [7] C. Harper, Electronic Packaging & Interconnection Handbook, McGraw-Hill, NY, 1991, Ch. 2. [8] Y.M. Kasem, R.K. Williams, “Thermal Design Principles and Characterization of Miniaturized Surface-Mount Packages for Power Electronics,” IEEE 10th annual Semi-Therm Conf., San Jose, Ca, Feb. 1994. [9] V. Manno, N.R. Kurita, K. Azar, “Experimental Characterization of Board Conduction Effect,” IEEE 9th Annual SemiTherm Conf., 1993. [10] J.W. Sofia, “Analysis of Thermal Transient Data with Synthesized Dynamic Models for Semiconductor Devices,” IEEE 10th Annual Semi-Therm Conf., San Jose, Ca, Feb. 1994. [11]G.R. Wagner, “Circuit Board Material/Construction and its Effect on Thermal Management,” Thermal Management Concepts, 1984. [12] M. Wills, “Thermal Analysis of Air-Cooled Cbs,” Electron Prod., pp. 11-18, May 1983. [13] Motorola Application Note AN-569. 11 TRADEMARKS The following are registered and unregistered trademarks Fairchild Semiconductor owns or is authorized to use and is not intended to be an exhaustive list of all such trademarks. ACEx™ CoolFET™ CROSSVOLT™ E2CMOSTM FACT™ FACT Quiet Series™ FAST® FASTr™ GTO™ HiSeC™ ISOPLANAR™ MICROWIRE™ POP™ PowerTrench™ QS™ Quiet Series™ SuperSOT™-3 SuperSOT™-6 SuperSOT™-8 TinyLogic™ DISCLAIMER FAIRCHILD SEMICONDUCTOR RESERVES THE RIGHT TO MAKE CHANGES WITHOUT FURTHER NOTICE TO ANY PRODUCTS HEREIN TO IMPROVE RELIABILITY, FUNCTION OR DESIGN. FAIRCHILD DOES NOT ASSUME ANY LIABILITY ARISING OUT OF THE APPLICATION OR USE OF ANY PRODUCT OR CIRCUIT DESCRIBED HEREIN; NEITHER DOES IT CONVEY ANY LICENSE UNDER ITS PATENT RIGHTS, NOR THE RIGHTS OF OTHERS. LIFE SUPPORT POLICY FAIRCHILD’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF FAIRCHILD SEMICONDUCTOR CORPORATION. As used herein: 1. Life support devices or systems are devices or 2. A critical component is any component of a life support device or system whose failure to perform can systems which, (a) are intended for surgical implant into be reasonably expected to cause the failure of the life the body, or (b) support or sustain life, or (c) whose support device or system, or to affect its safety or failure to perform when properly used in accordance with instructions for use provided in the labeling, can be effectiveness. reasonably expected to result in significant injury to the user. PRODUCT STATUS DEFINITIONS Definition of Terms Datasheet Identification Product Status Definition Advance Information Formative or In Design This datasheet contains the design specifications for product development. Specifications may change in any manner without notice. Preliminary First Production This datasheet contains preliminary data, and supplementary data will be published at a later date. Fairchild Semiconductor reserves the right to make changes at any time without notice in order to improve design. No Identification Needed Full Production This datasheet contains final specifications. Fairchild Semiconductor reserves the right to make changes at any time without notice in order to improve design. Obsolete Not In Production This datasheet contains specifications on a product that has been discontinued by Fairchild semiconductor. The datasheet is printed for reference information only.