PRODUCT SPECIFICATION RC4391 Pin Descriptions Pin Assignments LBR 1 8 VFB Pin Number Pin Function Description LBD 2 7 VREF 1 Low Battery Resistor (LBR) CX 3 6 +VS 2 Low Battery Detector (LBD) 3 Timing Capacitor (CX) 4 Ground 5 External Inductor (LX) 6 +Supply Voltage (+VS) GND 4 LX 5 65-3471-02 7 +1.25V Reference Voltage (VREF) 8 Feedback Voltage (VFB) Absolute Maximum Ratings Parameter Conditions Min Typ Internal Power Dissipation Max Unit 500 mW +30 V 0 70 °C RV4391 -25 85 °C RM4391 -55 125 °C -65 Supply Voltage1 (Pin 6 to Pin 4 or Pin 6 to Pin 5) Operating Temperature RC4391 150 °C PDIP, SOIC 125 °C CerDIP 175 °C Switch Current (IMAX) Peak 375 mA PD TA <50˚C PDIP 468 mW CerDIP 833 mW SOIC 300 mW (10 seconds) 300 °C Storage Temperature Junction Temperature Lead Soldering Temperature Note: 1. The maximum allowable supply voltage (+VS) in inverting applications will be reduced by the value of the negative output voltage, unless an external power transistor is used in place of Q1. Thermal Characteristics Therm. Res qJC Therm. Res. qJA For TA >50˚C Derate at 2 8-Lead Plastic DIP 8-Lead Ceramic DIP Small Outline SO-8 — 45°C/W — 160°C/W 150°C/W 240°C/W 6.25 mW/°C 8.33 mW/°C 4.17 mW/°C RC4391 PRODUCT SPECIFICATION Electrical Characteristics (VS = +6.0V, over the full operating temperature range unless otherwise noted) Symbol Parameters Condition Min +VS Supply Voltage (Note 1) 4.0 ISY Supply Current VS = +25V VREF Reference Voltage VOUT Output Voltage LI1 Line Regulation Typ Max Units 30 V 300 500 mA 1.13 1.25 1.36 V VOUT nom = -5.0V -5.5 -5.0 -4.5 V VOUT nom = -15V -16.5 -15.0 -13.5 VOUT nom = -5.0V, %VOUT CX = 150pF VS = +5.8V to +15V 2.0 4.0 1.5 3.0 VOUT nom = -15V, CX = 150pF VS = +5.8V to +15 L01 Load Regulation VOUT nom = -5.0V, %VOUT CX = 350pF, VS = +4.5V, PLOAD = 0mW to 75mW 0.2 0.5 PLOAD = 0mW to 75mW 0.2 0.3 Pin 5 = -20V 0.1 30 VOUT nom = -15V, CX = 350pF, VS = +4.5V, ICO Switch Leakage Current mA Note: 1. The maximum allowable supply voltage (+VS) in inverting applications will be reduced by the value of the negative output voltage, unless an external power transistor is used. 3 PRODUCT SPECIFICATION RC4391 Electrical Characteristics (VS = +6.0V, TA = +25°C unless otherwise noted) Symbol Parameters Condition ISY Supply Voltage VS = +4.0V, Min Typ Max Units 170 250 mA 300 500 No External Loads VS = +25V No External Loads VOUT LI1 Output Voltage Line Regulation VOUT nom = -5.0V -5.35 -5.0 -4.65 VOUT nom = -15V -15.85 -15.0 -14.15 VOUT nom = -5.0V V %VOUT CX = 150pF, 1.5 3.0 1.0 2.0 VS = +5.8V to +15V VOUT nom = -15V, CX = 150pF VS = +5.8V to +15V L01 Load Regulation VOUT nom = -5.0V, %VOUT CX = 350pF, VS = +4.5V, 0.2 0.4 0.07 0.14 1.25 1.32 PLOAD = 0mW to 75mW VOUT nom = -15V, CX = 350pF, VS = +4.5V, PLOAD = 0mW to 75mW 4 VREF Reference Voltage 1.18 ISW Switch Current Pin 5 = 5.5V ICO Switch Leakage Current Pin 5 = -24V ICX Cap. Charging Current Pin 3 = 0V ILBDL LBD Leakage Current Pin 1 = 1.5V, Pin 2 = 6.0V ILBD0 LBD On Current Pin 1 = 1.1V, Pin 2 = 0.4V ILBRB LBR Bias Current Pin 1 = 1.5V 75 6.0 210 100 V mA 0.01 5.0 mA 10 14 mA 0.01 5.0 mA 600 mA 0.7 mA RC4391 PRODUCT SPECIFICATION 6.5 8 6.0 7 5.5 6 FO (kHz) 5.0 4.5 4.0 5 4 3 2 65-3268 3.5 3.0 0 5 10 15 20 65-3272 FO (kHz) Typical Performance Characteristics 1 0 25 -55 0 70 125 TA (¡C) +VS (V) Figure 1. Oscillator Frequency vs. Supply Voltage Figure 2. Oscillator Frequency vs. Temperature 1.260 1.260 1.255 1.255 VREF (V) 1.250 1.250 1.245 65-3269 1.245 1.240 -55 0 25 70 125 65-3273 VREF (V) 25 1.240 4 6 TA (¡C) 10 20 30 +VS (V) Figure 3. Reference Voltage vs. Temperature Figure 4. Reference Voltage vs. Supply Voltage 4 600 500 3 +VS (V) 300 2 200 0 1 2 3 4 5 6 7 8 VCE (SAT) (V) Figure 5. Collector Current vs. Q1 Saturation Voltage 0 65-3271 1 100 20 10 65-3270 IC (mA) 400 -55 0 25 70 125 TA (¡C) Figure 6. Minimum Supply Voltage vs. Temperature 5 PRODUCT SPECIFICATION RC4391 Principles of Operation The basic switching inverter circuit is the building block on which the complete inverting application is based. A simplified diagram of the voltage inverter circuit with ideal components and no feedback circuitry is shown in Figure 7. When the switch S is closed, charging current from the battery flows through the inductor L, which builds up a magnetic field, increasing as the switch is held closed. When the switch is opened, the magnetic field collapses, and the energy stored in the magnetic field is converted into a current which flows through the inductor in the same direction as the changing current. Because there is no path for this current to flow through the switch, the current must flow through the diode to charge the capacitor C. The key to the inversion is the ability of the inductor to become a source when the charging current is removed. reference. Because CF is initially discharged a positive voltage is applied to the comparator, and the output of the comparator gates the squarewave oscillator. This gated squarewave signal turns on, then off, the PNP output transistor. This turning on and off of the output transistor performs the same function as opening and closing the ideal switch in the simplified diagram; i.e., it stores energy in the inductor during the on time and releases it into the capacitor during the off time. The comparator will continue to allow the oscillator to turn the switch transistor on and off until enough energy has been stored in the output capacitor to make the comparator input voltage decrease to less than 0V. The voltage applied to the comparator is set by the output voltage, the reference voltage, and the ratio of R1 to R2. The equation V = L (di/dt) gives the maximum possible voltage across the inductor; in the actual application, feedback circuitry and the output capacitor will decrease the output voltage to a regulated fixed value. S +VS L A complete schematic for the standard inverting application is shown in Figure 8. The ideal switch in the simplified diagram is replaced by the PNP transistor switch between pins 5 and 6. CF functions as the output filter capacitor, and D1 and LX replace D and L. 65-1601 Figure 7. Simple Inverting Regulator C F* 33µF R3 260K -V OUT LBR R6 100K Parts List R1 = R1 C2 R4 590K Q2 C1 VFB R2 = Cx = Lx = R2 VREF +1.25V REF/Bias LBD VREF C1 0.1µF LBD Output D1 +VS B OSC 900 k W 75 k W 300 k W 75 k W 150 pF 150 pF 1.0 mH Dale TE3 Q4 TA = Optional +Vs -VOUT = (1.25V) ( A Cx -15V Output -5.0V Output F 1N914 CX VOUT C (+) When power is first applied, the ground sensing comparator (pin 8) compares the output voltage to the +1.25V voltage To +Vs (–) D R1 ) R2 D Q1 GND LX RC4391 Lx E 65-1602 *Caution: Use current limiting protection circuit for high values of CF (Figure 13) Figure 8. Inverting Regulator – Standard Circuit 6 RC4391 PRODUCT SPECIFICATION 1.78V CX A 0.62V O SC (Internal) B IL I LOAD 0 mA C +V S (Internal) D +V S - 0.7V VBAT LX V OUT LX V BEQ1 Max I LX 0 mA I MAX E ID 0 mA F +V S - V SW V LX Ground G -V OUT - V D 65-2472 Figure 9. Inverting Regulator Waveforms This feedback system will vary the duration of the on time in response to changes in load current or battery voltage (see Figure 9). If the load current increases (waveform C), then the transistor will remain on (waveform D) for a longer portion of the oscillator cycle, (waveform B) to build up to a higher peak value. The duty cycle of the switch transistor varies in response to changes in load and line. Step-Down Regulator The step-down circuit function is similar to inversion; it uses the same components (switch, inductor, diode, filter capacitor), and charges and discharges the inductor by closing and opening the switch. The great difference is that the inductor is in series with the load; therefore, both the charging current and the discharge current flow into the load. In the inverting circuit only the discharge current flows into the load. Refer to Figure 10. When the switch S is closed, current flows from the battery, through the inductor, and through the load resistor to ground. After the switch is opened, stored energy in the inductor causes current to keep flowing through the load, the circuit being completed by the catch diode D. Since current flows to the load during charge and discharge, the average load cur- L S +VS D (+) C RL VOUT (-) 65-2473 Figure 10. Simple Step-Down Regulator rent will be greater than in an inverting circuit. The significance of that is that for equal load currents the step-down circuit will require less peak inductor current than an inverting circuit. Therefore, the inductor will not require as large of a core, and the switch transistor will not be stressed as heavily for equal load currents. Figure 11 depicts a complete schematic for a step-down circuit using the RC4391. Observe that the ground lead of the 4391 is not connected to circuit ground; instead, it is tied to the output voltage. It is by this rearrangement that the feedback system, which senses voltages more negative than the ground lead, can be used to regulate a non-negative output voltage. 7 PRODUCT SPECIFICATION RC4391 When power is first applied, the output filter capacitor is discharged so the ground lead potential starts at 0V. The reference voltage is forced to +1.25V above the ground lead and pulls the feedback input (pin 8) more positive than the ground lead. This positive voltage forces the control network to begin pulsing the switch transistor. As the switching action pumps up the output voltage, the ground lead rises with the output until the voltage on the ground lead is equal to the feedback voltage. At that point, the control network reduces the time on time of the switch to maintain a constant output. This control network will vary the on time of the switch in response to changes in load current or battery voltage (see Figure 12). If the load current increases (waveform C), then the transistor will remain on (waveformD) for a longer portion of the oscillator cycle, (waveform B), thus allowing the inductor current (waveform E) to build up to a higher peak value. The duty cycle of the switch transistor varies in response to changes in load and line. Design Equations The inductor value and timing capacitor (CX) value must be carefully tailored to the input voltage, input voltage range, output voltage, and load current requirements of the application. The key to the problem is to select the correct inductor value for a given oscillator frequency, such that the inductor current rises to a high enough peak value (IMAX) to meet the average load current drain. The selection of this inductor value must take into account the variation of oscillator frequency from unit to unit and the drift of frequency over temperature. Use ±30% as a maximum variation of oscillator frequency. R1 LBR VFB C2 Q2 C1 R2 LBD CX VREF +1.25V REF/Bias VREF +VS B OSC +Vs A Cx D Q1 GND LX +V OUT RC4391 E Lx CF +VOUT R1 = (1.25V) ( ) R2 F D1 1N914 65-2475 Important Note: This circuit must have a minimum load ³ 1 mA always connected. Figure 11. Step-Down Regulator – Standard Circuit 8 RC4391 PRODUCT SPECIFICATION 1.78V A CX 0.62V (Internal) B OSC IL I LOAD C 0 mA +VS D (Internal) +V - 0.7V VBAT LX V OUT - V BAT LX VBEQ1 I MAX I LX E 0 mA +V S - VSW VOUT F VLX VS ( -0.7V) 65-2474 Figure 12. Step-Down Regulator Waveforms The oscillator creates a squarewave using a method similar to the 555 timer IC, with a current steering flip-flop controlled by two voltage sensing comparators. The oscillator frequency is set by the timing capacitor (CX) according to the following equation. –6 4.1x10 F O ( Hz ) = ----------------------C ( pF ) x The squarewave output of the oscillator is internal and cannot be directly measured, but is equal in frequency to the triangle waveform measurable at pin 3. The switch transistor is normally on when the triangle waveform is ramping up and off when ramping down. Capacitor selection depends on the application; higher operating frequencies will reduce the output voltage ripple and will allow the use of an inductor with a physically smaller inductor core, but excessively high frequencies will reduce load driving capability and efficiency. Inverting Design Procedure 1. Select an operating frequency and timing capacitor value as shown above (frequencies from 10kHz to 50kHz are typical). 2. Find the maximum on time TON (add 3mS for the turn off base recombination delay of Q1): 1 T ON = ---------- + 3mS 2F O 3. Calculate the peak inductor current IMAX (if this value is greater than 375mA then an external power transistor must be used in place of Q1): ( V OUT + V D )2I L I MAX = --------------------------------------------------------( F O ) ( T ON ) ( V S – V SW ) Where: VS = Supply Voltage VSW = Saturation Voltage of Q1 (typically 0.5V) VD = Diode Forward Voltage (typically 0.7V) IL = DC Load Current 4. Find an inductance value for LX: V S – V SW L X ( Henries ) = æ -------------------------ö ( T ON ) è I MAX ø The inductor chosen must exhibit this value of inductance and have a current rating equal to IMAX. 9 PRODUCT SPECIFICATION RC4391 Step-Down Design Procedure Compensation 1. Select an operating frequency. When large values (> 50 kW) are used for the voltage setting resistors (R1 and R2 of Figure 8) stray capacitance at the VFB input can add lag to the feedback response, destabilizing the regulator, increasing low frequency ripple, and lowering efficiency. This can often be avoided by minimizing the stray capacitance at the VFB node. It can also be remedied by adding a lead compensation capacitor of 100 pF to 10 nF. In inverting applications, the capacitor connects between -VOUT and VFB; for step-down circuits it connects between ground and VFB. Most applications do not require this capacitor. 2. Determine the maximum on time TON as in the inverting design procedure. 3. Calculate IMAX: 2I L I MAX = ---------------------------------------------------------------------------( V S – V OUT ) ( F O ) ( T ON ) --------------------------------- + 1 ( V OUT – V D ) 4. Calculate LX: V S – V SW L X (Henries) = æ -------------------------ö ( T ON ) è I MAX ø Alternate Design Procedure The design equations above will not work for certain input/ output voltage ratios, and for these circuits another method of defining component values must be used. If the slope of the current discharge waveform is much less than the slope of the current charging waveform, then the inductor current will become continuous (never discharging completely), and the equations will become extremely complex. So, if the voltage applied across the inductor during the charge time is greater than during the discharge time, use the design procedure below. For example, a step-down circuit with 20V input and 5V output will have approximately 15V across the inductor when charging, and approximately 5V when discharging. So in this example the inductor current will be continuous and the alternate procedure will be necessary. The alternate procedure may also be used for discontinuous circuits. 1. Select an operating frequency based on efficiency and component size requirements (a value between 10kHz and 50kHz is typical). 2. Build the circuit and apply the worst case conditions to it, i.e., the lowest battery voltage and the highest load current at the desired output voltage. 3. Adjust the inductor value down until the desired output voltage is achieved, then decrease its value by 30% to cover manufacturing tolerances. 4. Check the output voltage with an oscilloscope for ripple, at high supply voltages, at voltages as high as are expected. Also check for efficiency by monitoring supply and output voltages and currents: ( V OUT ) ( I OUT ) ö æ eff = ----------------------------------------è ( +V S ) ( I SY )x100ø 5. If the efficiency is poor, go back to Step 1 and start over. If the ripple is excessive, then increase the output filter capacitor value or start over. Inductors Efficiency and load regulation will improve if a quality high Q inductor is used. A ferrite pot core is recommended; the wind-yourself type with an air gap adjustable by washers or spacers is very useful for bread-boarding prototypes. Care must be taken to choose a core with enough permeability to handle the magnetic flux produced at IMAX. If the core saturates, then efficiency and output current capability are severely degraded and excessive current will flow through the switch transistor. A pot core inductor design section is provided later in this datasheet. An isolated AC current probe for an oscilloscope (example: Tektronix P6042) is an excellent tool for saturation problems; with it the inductor current can be monitored for nonlinearity at the peaks (a sign of saturation). Low Battery Detector An open collector signal transistor Q2 with comparator C2 provides the designer with a method of signaling a display or computer whenever the battery voltage falls below a programmed level (see Figure 13). This level is determined by the +1.25V reference level and by the selection of two external resistors according to the equation: R4 V TH = V REF æ ------- + 1ö è R5 ø When the battery drops below this threshold Q2 will turn on and sink typically 600mA. The low battery detection circuit can also be used for other less conventional applications such as the voltage dependent oscillator circuit of Figure 18. +Vs R4 LBD 1 LBR C2 2 Q2 R5 V REF 1.25V 65-1651A Figure 13. Low Battery Detector 10 I LBD RC4391 PRODUCT SPECIFICATION Device Shutdown The entire device may be shut down to an extremely low current non-operating condition by disconnecting the ground (pin 4). This can be easily done by putting an NPN transistor in series with ground pin and switching it with an external signal. This switch will not affect the efficiency of operation, but will add to and increase the reference voltage by an amount equal to the saturation voltage of the transistor used. A mechanical switch can also be used in series between circuit ground and pin 4, without introducing any reference offset. Power Transistor Interfaces The most important consideration in selecting an external power transistor is the saturation voltage at IC = IMAX. The lower the saturation voltage is, the better the efficiency will be. Also, a higher beta transistor requires less base drive and therefore less power will be. Also, a higher beta transistor requires less base drive and therefore less power will be consumed in driving it, improving efficiency losses in the interface. The part numbers given in the following applications are recommended, but other types may be more appropriate depending on voltage and power levels. When troubleshooting external power transistor circuits, ensure that clean, sharp-edged waveforms are driving the interface and power transistors. Monitor these waveforms with an oscilloscop—disconnect the inductor, and tie the VFB input (pin 8) high through a 10K resistor. This will cause the regulator to pulse at maximum duty cycle without drawing excessive inductor currents. Check for expected on time and off time, and look for slow rise times that might cause the power transistor to enter its linear operating region. The following external power transistor circuits may demand some adjustment to resistor values to satisfy various power levels and input/output voltages. CX and LX values must be selected according to the design equations (pages 2-213 and 2-214). Inverting Medium Power Application Figure 8 is a schematic of an inverting medium power supply (250mW to 1W) using an external PNP switch transistor. Supply voltage is applied to the IC via R3: when the internal switch transistor is turned on current through R4 is also drawn through R3; creating a voltage drop from base to emitter of the external switch transistor. This drop turns on the external transistor. Voltage pulses on the supply lead (pin 6) do not affect circuit operation because the internal reference and bias circuitry have good supply rejection capabilities. A power Schottky diode is used for higher efficiency. Inverting High Power Application For higher power applications (500mW to 5W), refer to Figure 9. This circuit uses an extra external transistor to provide well controlled drive current in the correct phase to the power switch transistor. The value of R3 sets the drive current to the switch by making the interface transistor act as a current source. R4 and R5 must be selected such that the RC time constant of R4 and the base capacitance of Q2 do not slow the response time (and affect duty cycle), but not so low in value that excess power is consumed and efficiency suffers. The resistor values chosen should be proportional to the supply voltage (values shown are for +5V). Step-Down Power Applications Figures 16 and 17 show medium and high power interfaces modified to perform step-down functioning. The design +5V C1 0.1µF R3 1k½ R2 62 k½ Q1 2N3635 -24V 7 6 5 VFB VREF +Vs Lx 5 4391 Cx R4 50½ CF 100µF 220µH GND 4 3 150 pF Motorola MBR030 R1 1.2 M½ Cx 65-2476 Figure 14. Inverting Medium Power Application 11 PRODUCT SPECIFICATION RC4391 +Vs R6 1K C1 0.1µF R5 2K R2 Q1 TIP116 Q2 2N33904 MBR140P 5 VFB 7 6 5 VREF +Vs L x 4391 Cx -V OUT R4 4.7K R3 750½ R1 GND 4 3 CF Lx Cx 65-2478 Figure 9. Inverting High Power Application equations and suggestions for the circuits of Figures 14 and 15 also apply to these circuits. For a certain range of load power, the RC4193 can be used for step-sown applications. A load range from 400mW to 2W can be sustained with fewer components (especially when stepping down greater than 30V) than the comparable RC4391 circuit. Refer to Fairchild Semiconductor's RC4191/4192/4193 data sheet for a schematic of this medium power step-down application. Voltage Dependent Oscillator The RC4391's ability to supply load current at low battery voltages depends on the inductor value and the oscillator frequency. Low values of inductance or a low oscillator frequency will cause a higher peak inductor current and therefore increase the load current capability. A large inductor current is not necessarily best , however, because the large amount of energy delivered with each cycle will cause a large voltage ripple at the output, especially at high input voltages. This trade-off between load current capability and output ripple can be improved with the circuit connection shown in Figure 18. This circuit uses the low battery detector to sense for a low battery voltage condition and will decrease the oscillator frequency after a pre- programmed threshold is reached. 12 The threshold is programmed exactly as the normal low battery detector connection: R4 V TH = V REF æ ------- + 1ö è R5 ø When the battery voltage reaches this threshold the comparator will turn on the open collector transistor at pin 2, effectively pulling CY in parallel with CX. This added capacitance will reduce the oscillator frequency, according to the following equation: –6 4.1x10 F O ( Hz ) = ------------------------------------------------C X ( pF ) + C Y ( pF ) Current Limiting The oscillator (CX) pin can be used to add short circuit protection and to protect against over current at start-up (when using large values for the output filter capacitor —greater than 100 mF). A transistor VBE is used as a current sensing comparator which resets the oscillator upon sensing an over current condition, thus providing cycle-by-cycle current limiting. Figure 19 shows how this is applied. RC4391 PRODUCT SPECIFICATION +Vs C1 0.1 µF R3 1K R2 2N3635 8 VFB R1 7 6 VREF +Vs 5 Lx MBR030 R4 30 - 100½ 4391 Cx Lx GND 4 3 Cx +V OUT CF Note: A minimum load ³1mA must be connected. 65-2479 Figure 16. Step-Down Medium Power Application TIP116 * MBR140P 500½ 6 +1.3V VBAT 7 +VS VREF R2 5K 8 V FB R1 5K 470 pF LX 4391 CX 5 2N3904 GND 3 CX R3 1K R4 20K 4 470 pF 250µH (+) CF 470µF V OUT (+5V at 1A as shown) (-) Note: A minimum load ³1mA must be connected. *Optional — Extends supply voltage range. 65-2077 Figure 17. Step-Down High Power Application To +VS OSC CX +VS 3 +1.25V LBD 2 CV 1W CX +VS R4 1 LBR C2 2N3906 or Equivalent Q2 3 CX 4391 CX R5 65-2053 Figure 18. Voltage Dependent Oscillator 65-2159 Figure 18. Current Limiting 13 14 Q46 R1 540K Q7 4X Q1 R4 76K R3 8.2K Q8 C1 25 pF Q4 0.6 Q5 Q45 0.3 R9 60K +VS (6) R10 50K R2 3K Q6 Q9 R8 150K Q44 Q2 (7) VREF R11 160K Q4B D7 Q14 Q11 Q10 Q3A 0.5 D8 Q12 Q3B 0.5 LBR (1) Q16 2X Q15 Q22 D9 Q20 Q24 D18 (8) VFB D16 Q25 D17 Q23 Q21 Q17 Q18 Q19 0.1 0.1 0.1 D15 LBD (2) D10 Q26 40X R6 20K Q27 10X R5 10K Q3 Q35 Q36 Q13 Q41 Q29 50X (5) LX R7 10K Q28 CX (3) Q42 2X Q37 Q31 0.9 D12 Q38 Q32 0.9 Q43 2X Q40 Q39 Q33 0.9 65-6364 (4) GND D14 D13 Q34 1.1 PRODUCT SPECIFICATION RC4391 Simplified Schematic Diagram RC4391 PRODUCT SPECIFICATION Troubleshooting Chart Symptom Draws excessive supply current on star-up. Possible Problems Inductance value too low. Output frequency (FO) too low. Combination of low resistance inductor and high value filter capacitor — needs current limiting circuit (Figure 13). Output voltage is low. Inductance value too high for FO or core saturating. Inductor "sings" with audible hum. Not potted well or bolted loosely. LX pin appears noisy — scope will not synchronize. Normal operating condition. Inductor is saturating: -IMAX ILX 1. Core too small. 2. Core too hot. 3. Operating frequency too low. Time Inductor current shows nonlinear waveform. Waveform has resistive component: -IMAX ILX 1. Wire size too small. 2. Power transistor lacks base drive. 3. Components not rated high enough. Time 4. Battery has high series resistance. Inductor current shows nonlinear waveform. -IMAX External transistor lacks base drive or beta is too low. ILX Time Inductor current is linear until high current is reached. Poor efficiency. Core saturating. Diode or transistor: 1. Not fast enough. 2. Not rated for current level (high VCESAT). High series resistance. Operating frequency too high. Motorboating (erratic current pulses). Loop stability problem — needs feedback from VOUT to VFB (pin 8), 100pF to 1000pF 15 PRODUCT SPECIFICATION RC4391 Pot Core Inductor Design Electrical Circuit Magentic Circuit I E=I*R E H =B • R 1 U North South Flux 65-3464-07 Figure 20. Electricity vs. Magnetism Electricity Versus Magnetism Electrically the inductor must meet just one requirement, but that requirement can be hard to satisfy. The inductor must exhibit the correct value of inductance (L, in Henrys) as the inductor current rises to its highest operating value (IMAX). This requirement can be met most simply by choosing a very large core and winding it until it reaches the correct inductance value, but that brute force technique wastes size, weight and money. A more efficient design technique must be used. the point where the permeability decreases, the magnetic field has realigned all of the magnetic domains in the core material. Once all of the domains have been aligned the core will then carry no more flux than just air, it becomes as if there were no core at all. This phenomenon is called saturation. Because the inductance value, L, is dependent on the amount of flux, core saturation will cause the value of L to decrease dramatically, in turn causing excessive and possibly destructive inductor current. 6000 First Difference: Permeability instead of being analogous to resistance, is actually more like conductance (1/R). As permeability increases, flux increases. Second Difference: Resistance is a linear function. As voltage increases, current increases proportionally, and the resistance value stays the same. In a magnetic circuit the value of permeability varies as the applied magnetic force varies. This nonlinear characteristic is usually shown in graph form in ferrite core manufacturer's data sheet. As the applied magnetizing force increases, at some point the permeability will start decreasing, and therefore the amount of magnetic flux will not increase any further, even as the magnetizing force increases. The physical reality is that, at 16 B Gauss First, one must understand how the inductor's magnetic field works. The magnetic circuit in the inductor is very similar to a simple resistive electrical circuit. There is a magnetizing force (H, in oersteds), a flow of magnetism, or flux density (B, in Gauss), and a resistance to the flux, called permeability (U, in Gauss per oersted). H is equivalent to voltage in the electrical model, flux density is like current flow, and permeability is like resistance (except for two important differences discussed to the right). +25¡C 5000 +85¡C 4000 +125¡C 3000 2000 1000 Stackpole Ceramag 24B Hysteresis Loop vs. Temperature 0 -0.5 0 0.5 1 2 2.5 3 5 7 65-2170 Question: What happens if too small a core is used? 9 H Oersteds Figure 21. Typical Manufacturer’s Curve Showing Saturation Effects Pot Cores for RC4391 Pot core inductors are best suited for the RC4391 switching regulator for several reasons: 1. They are available in a wide range of sizes. RC4391 applications are usually low power with relatively low peak currents (less than 500mA). A small inexpensive pot core can be chosen to meet the circuit requirements. 2. Pot cores are easily mounted. They can be bolted directly to the PC card adjacent to the regulator IC. RC4391 PRODUCT SPECIFICATION Use of the Design Aid Graph 3. Pot cores can be easily air-gapped. The length of the gap is simply adjusted using different washer thicknesses. cores are also available with predetermined air gaps. 1. From the application requirement, determine the inductor value (L) and the required peak current (IMAX). 2. Observe the curves of the design aid graph and determine the smallest core that meets both the L and I requirements. 4. Electromagnetic interference (EMI) is kept to a minimum. the completely enclosed design of a pot core reduces stray electromagnetic radiation—an important consideration if the regulator circuit is built on a PC card with other circuitry. 3. Note the approximate air gap at IMAX for the selected core, and order the core with the gap. (If the gapping is done by the user, remember that a washer lspacer results in an air gap of twice the washer thickness, because two gaps will be created, one at the center post and one at the rim, like taking two bites from a doughnut.) Not quite. Core size is dependent on the amount of energy stored, not on load power. Raising the operating frequency allows smaller cores and windings. Reduction of the size of the magnetics is the main reason switching regulator design tends toward higher operating frequency. Designs with the RC4391 should use 75 kHz as a maximum running frequency, because the turn off delay of the power transistor and stray capacitive coupling begin to interfere. Most applications are in the 10 to 50 kHz range, for efficiency and EMI reasons. 4. If the required inductance is equal to the indicated value on the graph, then wind the core with the number of turns shown in the table of sizes. The turns given are the maximum number for that gauge of wire that can be easily wound in cores winding area. 5. If the required inductance is less than the value indicated on the graph, a simple calculation must be done to find the adjusted number of turns. Find AL (inductance index) for a specific air gap. The peak inductor current (IMAX) must reach a high enough value to meet the load current and simultaneously the inductor value is decreased, then the core can be made smaller. For a given core size and winding, an increase in air gap spacing (an air gap is a break in the material in the magnetic path, like a section broken off a doughnut) will cause the inductance to decrease and IMAX (the usable peak current before saturation )to increase. inHenriesö L ( indicated ) --------------------------------- = AL æ ------------------------2 ø 2 è Turn Turns Then divide the required inductance value by AL to give the actual turns squared, and take the square root to find the actual turns needed. The curves shown are typical of the ferrite manufacturer's power HF material, such as Siemens N27 or Stackpole 24B, which are usually offered in standard millimeter sizes including the sizes shown. #1 22X 13 mm 24 Gauge 70 Turns DCW = 0.5W L ( required ) ActualTurns = ------------------------------AL Air Gap = 0.02" 3A #3 #4 14X 8 mm 28 Gauge 60 Turns DCW = 0.6W 11X 7 mm 30 Gauge 50 Turns DCW = 1W 2A Air Gap = 0.006" #1 1A #2 No Air Gap #3 0 65-2171 #2 18X 11 mm 26 Gauge 70 Turns DCW = 0.7W IMAX (Amperes*) Air Gap = 0.012" #4 1 mH 2 mH 3 mH Inductor Value (Henries) *Includes safety margin (25%) to ensure nonsaturation Figure 22. Inductor Design Aid 17 PRODUCT SPECIFICATION RC4391 If the actual number of turns is significantly less than the number from the table then the wire size can be increased to use up the leftover winding area and reduce resistive losses. 6. Wind and gap the core as per calculations, and measure the value with an inductance meter. Some adjustment of the number of turns may be necessary. Where: N = number of turns Ae = core area from data sheet (in cm2) le = magnetic path length from data sheet (in cm) ue =permeability of core from manufacturer's graph g = center post air gap (in cm) The saturation characteristics may be checked with the inductor wired into the switching regulator application circuit. To do so, build and power up the circuit. Then clamp an oscilloscope current probe (recommend Tektronix P6042 or equivalent) around the inductor lead and monitor the current in the inductor. Draw the maximum load current from the application circuit so that the regulator is running at close to full duty cycle. Compare the waveform you see to those pictured. Manufacturers Check for saturation at the highest expected ambient temperature. Siemens Company 186 Wood Avenue South Iselin, NJ 08830 7. After the operation in circuit has been checked, reassemble and pot the core using a potting compound recommended by the manufacturer. If the core material differs greatly in magnetic characteristics from the standard power material shown in Figure 16, then the following general equation can be used to help in winding and gapping. This equation can be used for any core geometry, such as an E-E core. 2 Below is a list of several pot core manufacturers: Ferroxcube Company 5083 Kings Highway Saugerties, NY 12477 Indiana General Electronics Keasley, NJ 08832 Stackpole Company 201 Stackpole Street St. Mary, PA 15857 TDK Electronics 13-1, 1-Chrome Nihonbaski, Chuo-ku, Tokyo 8 ( 1.26 ) ( N ) ( Ae ) ( 10 ) L X = ----------------------------------------------------g = ( le/ue ) Proper Operation (Waveform is Fairly Linear) Improper Operation (Waveform is Nonlinear, Inductor Is Saturating) IMAX 0 IMAX 0 65-3464-08 Figure 23. Inductor Current Waveforms 18 RC4391 PRODUCT SPECIFICATION Mechanical Dimensions 8-Lead Ceramic DIP Package Inches Symbol Min. A b1 b2 c1 D E e eA L Q s1 a Millimeters Max. Min. — .200 .014 .023 .045 .065 .008 .015 — .405 .220 .310 .100 BSC .300 BSC .125 .200 .015 .060 .005 — 90¡ 105¡ Notes: Notes Max. — 5.08 .36 .58 1.14 1.65 .20 .38 — 10.29 5.59 7.87 2.54 BSC 7.62 BSC 3.18 5.08 .38 1.52 .13 — 90¡ 105¡ 1. Index area: a notch or a pin one identification mark shall be located adjacent to pin one. The manufacturer's identification shall not be used as pin one identification mark. 8 2, 8 2. The minimum limit for dimension "b2" may be .023 (.58mm) for leads number 1, 4, 5 and 8 only. 8 4 4 5, 9 7 3. Dimension "Q" shall be measured from the seating plane to the base plane. 3 6 4. This dimension allows for off-center lid, meniscus and glass overrun. 5. The basic pin spacing is .100 (2.54mm) between centerlines. Each pin centerline shall be located within ±.010 (.25mm) of its exact longitudinal position relative to pins 1 and 8. 6. Applies to all four corners (leads number 1, 4, 5, and 8). 7. "eA" shall be measured at the center of the lead bends or at the centerline of the leads when "a" is 90¡. 8. All leads – Increase maximum limit by .003 (.08mm) measured at the center of the flat, when lead finish applied. 9. Six spaces. D 4 1 Note 1 E 5 8 s1 eA e A Q a c1 L b2 b1 19 PRODUCT SPECIFICATION RC4391 Mechanical Dimensions (continued) 8-Lead Plastic DIP Package Inches Symbol A A1 A2 B B1 C D D1 E E1 e eB L Millimeters Min. Max. Min. Max. — .015 .115 .014 .045 .008 .348 .005 .300 .240 .210 — .195 .022 .070 .015 .430 — .325 .280 — .38 2.93 .36 1.14 .20 8.84 .13 7.62 6.10 5.33 — 4.95 .56 1.78 .38 10.92 — 8.26 7.11 .100 BSC — .430 .115 .160 2.54 BSC — 10.92 2.92 4.06 8¡ 8¡ N Notes: Notes 1. Dimensioning and tolerancing per ANSI Y14.5M-1982. 2. "D" and "E1" do not include mold flashing. Mold flash or protrusions shall not exceed .010 inch (0.25mm). 3. Terminal numbers are for reference only. 4. "C" dimension does not include solder finish thickness. 5. Symbol "N" is the maximum number of terminals. 4 2 2 5 D 4 1 5 8 E1 D1 E e A2 A A1 C L B1 20 B eB RC4391 PRODUCT SPECIFICATION Mechanical Dimensions (continued) 8-Lead SOIC Package Inches Symbol Min. A A1 B C D E e H h L N a ccc Millimeters Max. Min. Max. .053 .069 .004 .010 .013 .020 .008 .010 .189 .197 .150 .158 .050 BSC 1.35 1.75 0.10 0.25 0.33 0.51 0.20 0.25 4.80 5.00 3.81 4.01 1.27 BSC .228 .010 .016 5.79 0.25 0.40 .244 .020 .050 8 6.20 0.50 1.27 8 0¡ 8¡ 0¡ 8¡ — .004 — 0.10 8 Notes: Notes 1. Dimensioning and tolerancing per ANSI Y14.5M-1982. 2. "D" and "E" do not include mold flash. Mold flash or protrusions shall not exceed .010 inch (0.25mm). 3. "L" is the length of terminal for soldering to a substrate. 4. Terminal numbers are shown for reference only. 5 2 2 5. "C" dimension does not include solder finish thickness. 6. Symbol "N" is the maximum number of terminals. 3 6 5 E 1 H 4 h x 45¡ D C A1 A SEATING PLANE e B –C– LEAD COPLANARITY a L ccc C 21 PRODUCT SPECIFICATION RC4391 Ordering Information Part Number Package Operating Temperature Range RC4391N 8 Lead Plastic DIP 0˚C to +70°C RC4391M 8 Lead Plastic SOIC 0˚C to +70°C RV4391N 8 Lead Plastic DIP -25°C to +85°C RM4391D 8 Lead Ceramic DIP -55˚C to +125°C 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 THE PRESIDENT OF FAIRCHILD SEMICONDUCTOR CORPORATION. As used herein: 1. Life support devices or systems are devices or systems which, (a) are intended for surgical implant into the body, or (b) support or sustain life, and (c) whose failure to perform when properly used in accordance with instructions for use provided in the labeling, can be reasonably expected to result in a significant injury of the user. 2. A critical component in any component of a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system, or to affect its safety or effectiveness. www.fairchildsemi.com 5/20/98 0.0m 001 Stock#DS30004391 Ó 1998 Fairchild Semiconductor Corporation