AN4418 Application note 3 W, 5 V isolated flyback converter using the VIPer06HS, from the VIPer™ plus family By Alessandro Cannone Introduction This document describes the STEVAL-ISA137V1, a 5 V - 3 W power supply in isolated flyback topology with VIPer06HS: a new off-line high voltage converter by STMicroelectronics. The main features of the device are: 800 V avalanche rugged power section, PWM operation at 115 kHz with frequency jittering for lower EMI, cycle-by-cycle current limit with adjustable set point, on-board soft-start and safe auto-restart after a fault condition. The available protections are: thermal shutdown with hysteresis, delayed overload protection and open loop failure protection (only available if auxiliary winding is used). This flyback converter is suitable for different applications. It can be used as an external adapter or as an auxiliary power supply in consumer equipment. Figure 1. Evaluation board image: power supply board May 2016 DocID025636 Rev 2 1/22 www.st.com Contents AN4418 Contents 1 2 Test board: design and evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1 Output voltage characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2 Efficiency measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.3 No load consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.4 Light load consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Typical board waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.1 Dynamic step load regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3 Soft-start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 4 Protection features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.1 Overload and short-circuit protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.2 Open-loop failure protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 5 Conducted noise measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 6 Thermal measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 8 Evaluation tools and documentation . . . . . . . . . . . . . . . . . . . . . . . . . . 20 9 Revision history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2/22 DocID025636 Rev 2 AN4418 1 Test board: design and evaluation Test board: design and evaluation The electrical specifications of the test board are listed in Table 1. Table 1. Electrical specifications Parameter Symbol Value AC main input voltage VIN [85 VAC; 265 VAC] Main frequency fL [50 Hz; 60 Hz] Output voltage VOUT 5V Max output current IOUT 600 mA Precision of output regulation ∆VOUT_LF ±5% High frequency output voltage ripple ∆VOUT_HF 50mV Min active mode efficiency ηAV 66.89% Max ambient operating temperature TAMB 60 ºC The power supply is set in the isolated flyback topology. The schematic is given in Figure 3 and the bill of materials (BOM) in Table 2. The input section includes a resistor R1 to limit inrush current, a diode bridge (BR) and a Π filter for EMC suppression (C1, L1, C2). The transformer core is a standard E13. A clamp network (D1, R2, C3) is used for leakage inductance demagnetization. As the device is used in a secondary regulation isolated topology, the FB pin must be connected to ground in order to disable the internal error amplifier. In this case, the feedback signal is transferred to the primary side through an opto-isolator connected in parallel with the compensation network (R5, C6, C7) to the COMP pin. Figure 2. FB and COMP pin internal structure 3.3V Burst Mode Ref PWM STOP + 15K - Disabled FB VBU E\A From SenseFET + + 3.3V - To PWM latch nR R COMP The resistor connected between the LIM pin and ground lowers the default current limitation of the device (according to the IDLIM vs RLIM illustration given in the datasheet) to the value DocID025636 Rev 2 3/22 22 Test board: design and evaluation AN4418 required for the desired power throughput, thus avoiding unnecessary overstress on the power components. A small LC filter has been added at the output in order to filter the high frequency ripple. At power-up, the DRAIN pin supplies the internal HV start-up current generator which charges the C4 capacitor up to VDDon. At this point the power MOSFET starts switching, the generator is turned off and the IC is powered by the energy stored in C4. The IC is supplied by the auxiliary winding and the voltage delivered must always remain above the VDDcs_on threshold (11.5V max.), in order to avoid activating the HV start-up. Auxiliary winding is connected to the VDD pin through D3 and L2, where the inductor component is used to filter voltage spike on VDD pin, during power MOSFET turn-off. This solution is preferred because, using a resistor, the continuous voltage on the VDD pin drops and the voltage may fall below the VDDcs_on threshold. An external clamp on the VDD pin (Zener diode and resistor) is used to protect the pin when overvoltage, due to an increase of output voltage, occurs on the same pin. Figure 3. Electrical schematic 4/22 DocID025636 Rev 2 AN4418 Test board: design and evaluation Table 2. Bill of materials (BOM) Reference Part Description Note BR RMB6S 0.5A – 600V Bridge Taiwan Semiconductor R1 ROX1SJ10R 10Ω±5% - 1W Resistor TE Connectivity R2 ERJT08J224V 220kΩ±5% - 1/3W Resistor Panasonic R3 ERJT08J221V 220Ω±5% - 1/3W Resistor Panasonic R4 CRG0603F15K 15kΩ±1% - 1/10W Resistor TE Connectivity R5 ERJ3GEYJ682V 6.8kΩ±5% - 1/10W Resistor Panasonic R6 ERJ3GEYJ332V 3.3kΩ±5% - 1/10W Resistor Panasonic R7 ERJ3GEYJ334V 330kΩ±5% - 1/10W Resistor Panasonic R8 ERJ3GEYJ824V 820kΩ±5% - 1/10W Resistor Panasonic R9 CRG0603F100K 100kΩ±1% - 1/10W Resistor TE Connectivity R10 CRG0603F33K 33kΩ±1% - 1/10W Resistor TE Connectivity C1,C2 400LLE3R3MEFC8X11R5 3.3µF - Electrolytic capacitor 400V Rubycon C3 C3216C0G2J221J060AA 220pF - Capacitor 630V TDK C4 50YK10MEFCTA5X11 10µF - Electrolytic capacitor 50V Rubycon C5 GRM188R71H104KA93D 100nF - Capacitor 50V Murata C6 GRM188R71H153KA01D 15nF - Capacitor 50V Murata C7 GRM1885C1H222FA01D 2.2nF - Capacitor 50V Murata C8 16ZLH470MEFC8X11.5 470µF - Electrolytic capacitor 16V Rubycon C9 EEUEB1A101 100µF - Electrolytic capacitor 10V Panasonic C10 VJ0603Y472KNAAO 4.7nF - Capacitor 50V Vishay C11 DE2E3KY222MA2BM01 2.2nF - Capacitor Y2 Murata D1 STTH1L06A Ultrafast diode 1A – 600V STMicroelectronics D2 STPS2L60A Power Schottky 2A – 60V STMicroelectronics D3 BAT46ZFILM Signal Schottky 0.15A – 100V STMicroelectronics D4 MMSZ5248B-V-GS08 Zener diode 18V 0.5W Vishay 1921.0041 T1 Magnetica Flyback transformer 750370423 Rev. 6A Wurth IC1 VIPer06HS Offline primary controller STMicroelectronics IC2 TS432ILT Low voltage adjustable shunt reference STMicroelectronics IC3 SFH6106-2T Optocoupler Vishay L1 LPS4414 1mH - Power Inductor Coilcraft L2 LPS3008 4.7µH - Power Inductor Coilcraft L3 ME3220 4.7µH - Power Inductor Coilcraft DocID025636 Rev 2 5/22 22 Test board: design and evaluation AN4418 The transformer characteristics are listed in the table below. Table 3. Transformer characteristics 6/22 Parameter Value Test conditions Manufacturer MAGNETICA Part number 1921.0041 Primary inductance 1.5 mH ± 20% Measured at 1 kHz, TAMB = 20°C Leakage inductance 19 µH Nom. Measured at 10 kHz, TAMB = 20°C Primary to secondary turn ratio (3 - 4)/(5 8) 13.75 Measured at 10 kHz, TAMB = 20°C Primary to auxiliary turn ratio (3 - 4)/(2 - 1) 4.78 Measured at 10 kHz, TAMB = 20°C Saturation current 0.27 A Primary, BSAT = 0.3 T, TAMB = 20°C Operating current 0.14 A Primary, POUT = 1.5 W, TAMB = 20°C Figure 4. Dimensional drawing and pin placement diagram - bottom view Figure 5. Dimensional drawing and pin placement diagram - electrical diagram Figure 6. Dimensional drawing and pin placement diagram - side view 1 Figure 7. Dimensional drawing and pin placement diagram - side view 2 DocID025636 Rev 2 AN4418 1.1 Test board: design and evaluation Output voltage characteristics The output voltage of the board is measured under different line and load conditions. Table 4 clearly demonstrates that the output voltage is not affected by line and load variations. For this reason, Figure 8 shows the load regulation for only one input voltage (230 VAC). Table 4. Output voltage line-load regulation VOUT (V) VIN (VAC) No load 0.3 A 0.45 A 0.6 A 85 5.00 5.00 5.00 5.00 115 5.00 5.00 5.00 5.00 150 5.00 5.00 5.00 5.00 180 5.00 5.00 5.00 5.00 230 5.00 5.00 5.00 5.00 265 5.00 5.00 5.00 5.00 Figure 8. Output voltage load regulation at 230 VAC 1.2 Efficiency measurements Any external power supply (EPS) must be capable of meeting the international regulation agency limits. The European code of conduct (EC CoC version 5) and US department of energy (DoE-US EISA 2007) limits are taken as references. EPS limits are fixed up to 66.89%, where the average efficiency is measured. The efficiency of the converter has been measured under different load and line voltage conditions. DocID025636 Rev 2 7/22 22 Test board: design and evaluation AN4418 The efficiency measurements have been performed with loading at 25%, 50%, 75% and 100% of maximum rate at 115 VAC and 230 VAC. Table 5 and Table 6 show the results. Table 5. Efficiency at 115 VAC %Load IOUT (A) VOUT (V) PIN (W) POUT (W) Efficiency (%) 25% 0.15 5.00 1.074 0.750 69.83 50% 0.30 5.00 1.982 1.500 75.68 75% 0.45 5.00 3.012 2.250 74.70 100% 0.60 5.00 3.978 3.000 75.41 Average efficiency 73.91 Table 6. Efficiency at 230 VAC %Load IOUT (A) VOUT (V) PIN (W) POUT (W) Efficiency (%) 25% 0.15 5.00 1.203 0.750 62.34 50% 0.30 5.00 2.274 1.500 65.96 75% 0.45 5.00 3.117 2.250 72.18 100% 0.60 5.00 4.193 3.000 71.55 Average efficiency 8/22 DocID025636 Rev 2 68.01 AN4418 Test board: design and evaluation Figure 9. Efficiency vs. output current load 1.3 No load consumption The input power of the converter has been measured under no load condition; in this situation, the converter works in burst mode so that the average switching frequency is reduced. Figure 10. No load consumption vs. input voltage 1.4 Light load consumption Even if the EC CoC and DoE US EISA 2007 do not stipulate other requirements regarding light load performance, the input power of the demo-board under light load conditions is given in order to provide a complete picture. In particular, in order to comply with EuP Lot 6, the EPS requires an efficiency higher than 50% when the output load is 250 mW. The test board also satisfies this requirement. DocID025636 Rev 2 9/22 22 Test board: design and evaluation AN4418 Figure 11. Light load consumption at different output power 10/22 DocID025636 Rev 2 AN4418 2 Typical board waveforms Typical board waveforms Drain voltage and current waveforms under full load condition are reported for minimum and maximum input voltage in Figure 12 and Figure 13, and for the two nominal input voltages in Figure 14 and Figure 15 respectively. Figure 12. Drain current and voltage at full load Figure 13. Drain current and voltage at full load at 85 VAC at 265 VAC Figure 14. Drain current and voltage at full load Figure 15. Drain current and voltage at full load at 115 VAC at 230 VAC The output ripple at the switching frequency was also measured. The board is provided with an LC filter to further reduce the ripple without reducing the overall ESR of the output capacitor. The voltage ripple across the output connector (VOUT) and before the LC filter (VOUT_PRE) was measured in order to verify the effectiveness of the LC filter. The following two diagrams show the voltage ripple at 115 VAC (Figure 16) and at 230 VAC (Figure 17) under full load condition. DocID025636 Rev 2 11/22 22 Typical board waveforms AN4418 Figure 16. Output voltage ripple at full load at 115 VAC Figure 17. Output voltage ripple at full load at 230 VAC As the load is so low that the voltage at the COMP pin falls below the VCOMPL internal threshold (typically 1.1 V), the VIPer06HS is disabled. At this point, the feedback reaction to the energy delivery cutoff forces the COMP pin voltage to rise and, when it is 40 mV above the VCOMPL threshold, the device begins switching again. This results in a controlled on/off operation which is referred to as “burst mode”. This mode of operation reduces frequencyrelated losses when the load is very light or disconnected, facilitating compliance with energy saving regulations. The figures below show the output voltage ripple when the converter operates in burst mode and is supplied with 115 VAC and with 230 VAC respectively. Figure 18. Output voltage ripple during burst mode operation at 115 VAC 12/22 Figure 19. Output voltage ripple during burst mode operation at 230 VAC DocID025636 Rev 2 AN4418 2.1 Typical board waveforms Dynamic step load regulation In any power supply, it is important to measure the output voltage when the converter is subjected to dynamic load variations in order to ensure good stability and that no overvoltage or undervoltage occurs. The test has been performed for both nominal input voltages, varying the output load from 0 to 100% of the nominal value. In every tested condition, no abnormal oscillations were revealed on the output and over/under shoot were well within acceptable values. Figure 20. Dynamic step load (0 to 100% output Figure 21. Dynamic step load (0 to 100% output load) at 115 VAC load) at 230 VAC DocID025636 Rev 2 13/22 22 Soft-start 3 AN4418 Soft-start When the converter starts, the output capacitor has no charge and needs some time to reach the steady state condition. During this time, the power demand from the control loop is at its maximum, while the reflected voltage is low. These two conditions may lead to a deep continuous operating mode of the converter. Also, when the power MOSFET is switched on, it cannot be switched off immediately as the minimum on time (TON_MIN) must first elapse. Because of the deep continuous operating mode of the converter, during TON_MIN, an excessive drain current can overstress the component of the converter as well as the device itself, the output diode and the transformer. Transformer saturation is also possible under these conditions. To avoid all the above mentioned negative effects, the VIPer06HS implements an internal soft-start feature. As the device begins operation, regardless of the control loop request, the drain current is allowed to gradually increase from zero to the maximum value. The drain current limit is increased in steps, and the range from 0 to the fixed drain current limitation value (which can be adjusted through an external resistor) is divided in 16 steps. Each step length is 64 switching cycles, for a total duration of the soft-start phase around 8.5ms. Figure 22 shows the soft-start phase of the converter when it is operating at minimum line voltage and maximum load. Figure 22. Soft-start feature 14/22 DocID025636 Rev 2 AN4418 4 Protection features Protection features In order to increase end-product safety and reliability, VIPer06HS has the following protection mechanisms: overload and short-circuit protection and open loop failure protection. In the following sections, these protection mechanisms are tested and the results are presented. 4.1 Overload and short-circuit protection In case of overload or output short-circuit (see Figure 23), the drain current reaches the IDLIM value (or the one set by the user through the RLIM resistor). For each cycle that this condition is met, a counter increments; if this state is maintained continuously for the time tOVL (50 ms typical, internally fixed), the overload protection is tripped, the power section is turned off and the converter is disabled for a tRESTART time (typically 1 s). When this time has elapsed, the IC resumes switching and, if the short is still present, the protection again activates (Figure 24). This ensures that the restart attempts of the converter are at a low repetition rate, so that it works safely with extremely low power throughput and avoids the IC overheating in case of repeated overload events. Moreover, whenever the protection is tripped, the internal soft start function is invoked (Figure 25) in order to reduce the stress on the secondary diode. When the short is removed, the IC resumes normal operation. If the short is removed during tSS or tOVL, i.e., before the protection tripping, the counter decrements each cycle down to zero and the protection is thus not tripped. If the short-circuit is removed during tRESTART, the IC waits until tRESTART has elapsed before resuming switching (Figure 26). Figure 23. Overload protection: output shortcircuit applied Figure 24. Overload protection: continuous output short-circuit DocID025636 Rev 2 15/22 22 Protection features AN4418 Figure 25. Overload protection: soft start and tOVL 4.2 Figure 26. Overload protection: short-circuit removal Open-loop failure protection This kind of protection is useful when the device is supplied by an auxiliary winding and it is activated when feedback loop failure or auxiliary winding disconnection occurs. If R9 is open or R10 is shorted, the VIPer06HS works at its drain current limitation. The output voltage, VOUT, increases with the auxiliary voltage VAUX, which is coupled with the output according to the secondary-to-auxiliary turns ratio. As the auxiliary voltage rises to the internal VDD active clamp, VDDclamp (23.5 V min.), and the clamp current injected on the VDD pin exceeds the latch threshold, IDDol (4 mA min.), a fault signal is internally generated and the device stops switching even if tOVL hasn’t elapsed yet (see Figure 28). To verify the effectiveness of this protection, the external clamp on the VDD pin has been removed. Figure 27. Open loop failure protection: R9 open, tRESTART 16/22 Figure 28. Open loop failure protection: R9 open, tOVL DocID025636 Rev 2 AN4418 5 Conducted noise measurements Conducted noise measurements The VIPer06HS frequency jittering feature allows the spectrum to be spread over frequency bands rather than being concentrated on single frequency value. Especially when measuring conducted emission with the average detection method, the level reduction can be several dBµV. A pre-compliance test for the EN55022 (Class B) European normative was performed and peak measurements of the conducted noise emissions at full load and nominal mains voltages are shown in Figure 29 and Figure 30. The diagrams show that the measurements are well within the limits in all test conditions. Figure 29. CE peak measurement at 115 VAC full load Figure 30. CE peak measurement at 230 VAC full load DocID025636 Rev 2 17/22 22 Thermal measurements 6 AN4418 Thermal measurements Thermal analysis of the board was performed using an IR camera for the two nominal input voltages (115 VAC and 230 VAC) under full load condition. The results are shown in Figure 31 to Figure 34 and summarized in Table 7. Figure 31. Thermal map at 115 VAC full load. Top layer Figure 32. Thermal map at 115 VAC full load. Bottom layer Figure 33. Thermal map at 230 VAC full load. Top layer Figure 34. Thermal map at 230 VAC full load. Bottom layer Table 7. Temperature of key components (Tamb = 25 °C, emissivity = 0.95 for all points) Temp (°C) Point 18/22 Reference 115 VAC 230 VAC A 48.1 54.0 Transformer B 49.0 69.4 VIPer06HS C 59.4 63.7 Output diode D 45.0 59.2 Snubber diode DocID025636 Rev 2 AN4418 7 Conclusions Conclusions In this document, a flyback has been described and characterized. Special attention was paid to efficiency and low load performances and the bench results were good with very low input power under light load conditions. The efficiency performance was compared with the requirements of EC CoC and DoE regulation programs for external AC-DC adapters with very good results, with the measured active mode efficiency always higher than the required minimum. The EMI emissions are also quite low, even when only using a low cost input filter. DocID025636 Rev 2 19/22 22 Evaluation tools and documentation 8 AN4418 Evaluation tools and documentation The VIPer06HS evaluation board order code is: STEVAL-ISA137V1 Further information about this product is available in the VIPer06 datasheet at www.st.com. 20/22 DocID025636 Rev 2 AN4418 9 Revision history Revision history Table 8. Document revision history Date Revision Changes 20-Aug-2014 1 Initial release. 12-May-2016 2 Added: new T1 part 750370423 Rev 6A on Table 2. DocID025636 Rev 2 21/22 22 AN4418 IMPORTANT NOTICE – PLEASE READ CAREFULLY STMicroelectronics NV and its subsidiaries (“ST”) reserve the right to make changes, corrections, enhancements, modifications, and improvements to ST products and/or to this document at any time without notice. Purchasers should obtain the latest relevant information on ST products before placing orders. ST products are sold pursuant to ST’s terms and conditions of sale in place at the time of order acknowledgement. Purchasers are solely responsible for the choice, selection, and use of ST products and ST assumes no liability for application assistance or the design of Purchasers’ products. No license, express or implied, to any intellectual property right is granted by ST herein. Resale of ST products with provisions different from the information set forth herein shall void any warranty granted by ST for such product. ST and the ST logo are trademarks of ST. All other product or service names are the property of their respective owners. Information in this document supersedes and replaces information previously supplied in any prior versions of this document. © 2016 STMicroelectronics – All rights reserved 22/22 DocID025636 Rev 2