designfeature Nabil Sadiq, Senior Field Application Engineer Trina Noor, Field Applications/New Product Development Engineer, Cooper Bussmann Fuses protect against overcurrent events by melting their elements and opening the circuit. Fuses must be applied at or below their specified voltage rating, which differs between ac and dc current Selecting Fuses: Simple Procedures to Get the Right Overcurrent Protection for DC-DC Converters A lthough features and functionality attract the most attention for new electronic products, whether consumer, industrial, or medical, their reliability depends on protecting their power systems from overcurrent events. Internal, external, and nuisance threats can affect circuit and system reliability. Through proper fuse selection, you can minimize risks and failures so that an electronic product retains its competitive edge. Fuses are overcurrent devices that protect electrical and electronic devices by melting and opening a circuit to prevent excessive current from causing damage or starting fires. Fuses serve two main purposes: 1. To protect components, equipment and people from risk of fire and electric shock 2. To isolate sub systems from the main system. The fusing action begins when the circuit current is high enough to heat the fuse element and starts it to melt. Once melting begins, a gap is created that the current will “arc” across. Melting continues and the gap grows wider until it is too wide to 10 Power Electronics Technology | August 2010 www.powerelectronics.com Percentage of rating OVERCURRENTfuses 140 120 100 80 60 40 20 0 –55 –25 0 25 50 85 Temperature in Degrees C 105 125 Fig. 1: Typical fuse derating curves SADIQ_F1 sustain the arc. At that point, current ceases to flow and the overcurrent event is “cleared,” opening and making the circuit safe. 1. There are two types of overcurrent events: 1) Overload - simply drawing excessive current beyond the designed capacity of the circuit, 2) Short-circuit, or fault current. Regardless of the overcurrent event, fuses are designed and specified to be a circuit’s “weakest link.” These “thermally operated” devices typically employ a metal wire or strip element in their construction. Fuse Types Fast-acting fuses open very quickly when their current rating is exceeded. This action is needed when speed is important for sensitive electronics and for many dc power applications. They are generally used in resistive loads with low inrush current levels. Time-delay fuses have a time-delay mechanism. They are designed to open only on an excessive current draw for a defined period of time and are typically used to protect inductive and capacitive loads that experience heavy current draws upon initial powering. The time delay action prevents the fuse from needlessly blowing during a temporary heavy current draw or surge. Time-delay fuses tolerate higher inrush currents than fast-acting fuses and are often ideal for dc-dc converter input protection, as F1 L2 L1 AC C2 C1 L3 C3 PFC Boost Line Module 400VDC F2 COUT F3 Fig. 2. Fuse locations in a typical dc-dc converter most converters have an input capacitor that draws a large amount of current when initially charged. Selecting the right fuse is critical in all electronic and electrical system designs. Catastrophic system failure can be prevented with the proper fuse on the dc-dc converter input. In the event the converter’s internal circuitry can no longer withstand an overload condition, the fuse will prevent fire or further damage to the board, the converter, or neighboring components. Most dc-dc converters are protected from short-circuits on their outputs by either circuit-sensing current limit and/or thermal overload circuits. Fuses are required to protect against a catastrophic component failure (e.g., MOSFET failure) or if a component failure creates a short-circuit on the input side of the dc-dc converter. Proper selection of an input fuse for a dc-dc converter involves understanding and consideration of the following factors: 1. Voltage Rating 2. Current Rating 3. Interrupting Rating 4. Temperature Derating 5. Melting Integral (I2t) 6. Maximum Circuit Fault Current 7. Required Agency Approvals 8. Mechanical Considerations Voltage Rating Fuses are first rated by the ac and/or dc circuit voltage into which they can be safely applied. A fuse installed in an AC circuit performs differently than when installed in a DC circuit. With AC circuits, the current is crossing the zero potential at 60 or 50 cycles a second. This helps in breaking the arc that forms when the fuse element melts and creates a gap. In dc circuits, the voltage does not go to a zero potential, making it more difficult to suppress the arc in the melting element’s gap. Generally, fuse ac voltage ratings coincide with the utility supply, e.g., 110V, 240V, 415V, etc. This means that a fuse is suitable for use with these nominal voltages and is tested for voltage levels at least 3.3 VOUT 15% higher than the nominal rating. This is not true with dc voltage ratDC/DC ings, which are normally maximum converter ratings and should not be exceeded. More specifically, the voltage rating of 5 VOUT a fuse must be equal to or greater than the maximum voltage expected in the DC/DC application. converter Fuses are insensitive to voltage changes within their ratings so selecting the proper voltage rating is strictly a safety issue. Fuses can operate at any SADIQ_F2 www.powerelectronics.com August 2010 | Power Electronics Technology 11 OVERCURRENTfuses voltage below or equal to their rated voltage. Current Rating Although some power supplies are designed for constant current output regulation, most typical DC-DC converters are designed as constant power devices. This means that as the input voltage drops, the input current must increase to uphold the constant output power relationship of P= V*I. The fuse’s minimum current rating is determined by the maximum input current of a DC-DC converter. Typically, the maximum current consumption occurs at the maximum output load and the minimum input voltage. The magnitude of the input current can be determined from: POUT(MAX) IINPUT(MAX ) = (1) VIN (MIN ) × Efficiency Where: POUT(MAX) = Maximum dc-dc converter output INPUT (MAX) (2) power. IRATED = K TEMP VIN(MIN) = Minimum input voltage on the dc-dc converter input. 2 Efficiency = Efficiency dc-dc×converter at POUT(MAX) I2 t (FUSE) = Iof t (PULSE) FP (3) and VIN(MIN); can be determined from the dc-dc converter’s datasheet. To prevent damage to converter components, the fuse current rating is selected with a large enough current capability so that the fuse will not open under steady state conditions, yet will open during an abnormal (excessive) overload or short-circuit condition. Usually this results in selecting a fuse to be 150% to 200% percent of the maximum steady state input current at maximum load and minimum line input voltage. Interrupting Rating The fuse interrupting rating is the maximum amperage at rated voltage the fuse can safely interrupt. This rating must exceed the maximum fault (short-circuit) current the circuit can produce. Interrupting ratings for AC and DC currents are different and the fuse data sheet should be consulted before selection. Melting Integral The DC-DC converter peak inrush current is usually significantly greater than the steady state current. Additionally, periodic inrush currents can be sufficiently powerful to warm the fuse element. Though not large enough to melt the element, it can still cause significant thermal stress to the element. Cyclical expansions and contractions of the fuse element can lead to mechanical fatigue and premature failure. Selecting the appropriate fuse involves choosing the appropriate melting integral. The melting integral of a fuse, termed melting I2t, is the thermal energy required to melt a specific fuse element. The fuse element construction, materials and cross sectional area will determine this POUT(MAX) IINPUT(MAX ) = (1) value. VIN (MIN ) × Efficiency The task of a system designer is to select a fuse with the minimum I2t greater than the energy of the inrush current pulse. This rating ensures that the fuse will not cause a INPUT (MAX) conditions. For relinuisance openingI during transient (2) RATED = K TEMP able system operation for the required number of turn-on cycles, the following condition must be met: I2 t (FUSE) = I2 t (PULSE) × FP (3) Where: I2t (PULSE) = Energy of a current pulse I2t (FUSE) = Melting integral of a fuse FP = the pulse factor (dependent on fuse element construction in Table 1) I2t (FUSE) can be found in fuse datasheets. Do not use Table 1. pulse factor for solid matrix construction Pulse Factor for Solid Matrix Construction Temperature Derating When a fuse is applied in an ambient temperature exceeding the standard 23°C, the fuse current rating should be derated (a higher amp rating with higher temPOUT(MAX) peratures).I Conversely, operating at an ambient tempera(1) INPUT(MAX ) = VIN (MINstandard × Efficiency ture lower than the 23°C allows using a lower ) fuse amp rating. Fig.1 shows a typical fuse derating curve. The fuse rating is determined by: IRATED = Where: I INPUT(MAX) = Current determined from Equation (1) or a dc-dc converter datasheet KTEMP = Temperature derating factor determined from Fig. 1. The lowest suitable fuse rating is obtained by rounding up the calculated value to the next higher current rating shown in the fuse datasheet. INPUT (MAX) K TEMP I2 t (FUSE) = I2 t (PULSE) × FP 12 Power Electronics Technology | August 2010 (2) Number of Surge Pulses Pulse Factor, FP 1 to 100,000 1.25 Table 2. Fuse Pulse Factor Pulse Factor for Wire-in-Air Construction Number of Surge Pulses Pulse Factor, FP 100 2.1 1,000 2.6 10,000 3.4 100,000 4.5 (3) www.powerelectronics.com OVERCURRENTfuses the fuse’s maximum melting integral in Equation (3), and use either the minimum or nominal melting integral of the fuse. a fuse to open at lower current levels. Wire-in-air construction, as in the 3216TD and new S505H series, and many traditional ferrule fuses, provides high inrush withstand. Wire-inMaximum Circuit Fault Current air technology makes a smaller fuse Other selection considerations include possible without sacrificing I2t, temstart-up (inrush) currents and transient Fig. 3. Compact C310T Series 3.6mm x 10mm time- perature or operating voltage range. load conditions. When a dc-dc con- delay fuse for off-line protection Using a fuse with high surge-withstand verter is initially powered, the input capability means fewer open fuses durbulk capacitors of dc-dc must be charged. Current flowing momentary overloads. ing into the input terminals of a dc-dc converter is approximately I = V/R for typical power supplies AGENCY APROVALS with charge times less than 10 milliseconds. When North American UL/CSA and IEC standards V is the input voltage change, and R is a combinafor overcurrent protective devices require tion of wiring resistance, your source’s resistance under significantly different Time-versus-Current start-up, and the Equivalent Series Rating (ESR) of the characteristics. UL rated fuses are tested converter’s input bulk capacitors. to open at 135% of rated current while Larger dc-dc converters often use a large Fig. 4. PC-Tron® Fuses IEC fuse ratings are tested to carry 150% capacitor with very low ESR inside the con- operate up to 5A of rated current. Be aware of these difverter. This inrush current can have a sigferences as the fuses are tested and specified nificant effect on the fuse’s life. Size the fuse properly to differently between these standards for products sold in allow these inrush current pulses to pass without nuisance different parts of the world. openings or degrading the fuse element as discussed in The physical dimensions and materials for both UL and melting integral. IEC fuses are similar. However, fuses made to different To calculate current pulse energy, one must first deterstandards are not interchangeable. Their element melting mine the magnitude and duration of the current pulse. and opening times will differ when subjected to the same The most accurate way to determine parameters of a magnitude of current. The circuit designer must consider current pulse is to measure this current in the application that different world markets may require different fuse under minimum and maximum voltage conditions. agency standards. Note that the melting I2t values of the fuse must be To select a fuse that ensures system and agency complicalculated at the condition where the product of the peak ance the following conditions must be met: current squared and time the peak occurs is maximum. • Fuse current rating does not exceed the rating of the For example, the steady state current is maximum at low fuse used for safety testing of the dc-dc converter it is line so a transient load surge needs to be added to the low intended to protect. line current to establish the maximum peak current for • Fuse is installed on the ungrounded side of the circuit an operating condition. But the inrush current is usually to ensure uninterrupted ground connection in case the maximum at the highest input voltage. The fuse’s melting fuse opens. I2t must be evaluated at the condition with the highest • The input traces and chassis ground trace (if used) are calculated I2t to ensure that the fuse will not open during capable of conducting a current of 1.5 times the fuse these “normal” operating conditions. current rating. The pulse factor is dependent on the construction of the fuse element (see pulse factor tables under Melting Mechanical Considerations Integral). There are numerous physical sizes of fuses for electronThe patented solid matrix construction used in the ics, including subminiature fuses. The most common ferrule Cooper Bussmann® 0603FA, 3216FF, CC12H and CC06 designs are 5x15mm, 5x20mm and 6.3x32mm (¼ in. x fuse series provides excellent cycling and temperature 1¼ in.). Ferrule fuses are generally mounted in fuse clips performance while significantly reducing nuisance openor holders with some available with axial leads for solderings from high inrush currents. It also provides protection ing directly onto a PCB. Subminiature fuses are often used against unanticipated current surges from the system. The when board space is limited. For applications of this type, small physical size allows maximum protection without there are through-hole and surface mount devices available. oversizing the fuse rating. Solid matrix construction reducStandard package sizes for surface mount fuses are 0402 es heating from repeated surges that would normally cause (1005), 0603 (1608), 1206 (3216), 6125, and 1025. 13 Power Electronics Technology | August 2010 www.powerelectronics.com OVERCURRENTfuses These sizes are standard throughout the electronic industry. Throughhole axial and radial leaded products allow fuses to be PCB mounted. For example, Cooper Bussmann offers electronic fuses ranging from 32V to 450V. Voltage ratings can and do vary inside a fuse family or series, as well as interrupting ratings, I2t and agency approvals. Always consult data sheets for the ratings that apply to the desired voltage and amp rating of the application. Typical Fuse Locations in Power Supplies Product safety standards require fuses for primary ac power protection and secondary protection against any catastrophic failure in the input filter capacitors, Power Factor Correction (PFC) boost module, output capacitors, or within the dc-dc converters where fuse F1 in Fig. 2 is a typical ac fuse location. The fuse is placed near the input connector so that all other components are downstream and protected. The PFC boost module usually does not contain overcurrent protection. If a short-circuit is applied across the PFC output terminals, there is no internal circuit opening device to safely interrupt the power. The fuse in the AC input line (Fuse F1 in Fig. 2) protects the PFC boost converter. Although the primary input line fuse will eventually open, dc fuses positioned right at the input to the dc-dc converters will limit the energy delivered by the hold-up capacitors and prevent failure to the PFC boost module. DC fuses between the PFC and dc-dc converters protect against a catastrophic failure in the dc-dc converter (Fuses F2 and F3 in Fig.2). Fusing each dc-dc converter will allow the converter not subject to a fault to continue operating by isolating the failed converter. Fuses F2 and F3 have an added benefit during product development. www.powerelectronics.com Fig. 5. The S505H Series of time delay fuses By selectively removing these fuses, the various converters can be powered separately, or the PFC operated with an external load. In addition to facilitating testing of the different power sections during product development, the fuses can aid troubleshooting in production and in the event the product needs to be repaired. Fuses applied to overcurrent protection points of Fig. 2 include F1 providing primary overcurrent protection. Use ac line voltage rated fuses located on the transformer primary side (typically 125Vac / 250Vac line voltage) • SR-5 / SS-5 radial fuses • S501-2-R fast-acting fuse • C310T Series (coming soon) 3.6x10mm axial-leaded, timedelay, ceramic tube fuse (Fig. 3) • 5mm or ¼ in. ferrule fuses Fuses F2 and F3 that provide secondary overcurrent protection. Use 400Vdc or higher rated fuses on the secondary side of the transformer or on battery powered applications (ac or dc, typically lower voltages, but not always). • PC-Tron® (up to 2.5A) (Fig. 4) • S505H Series (coming soon) 400Vdc/500-600Vac, time-delay, 5x20mm (Fig. 5) August 2010 | Power Electronics Technology 14