UC3852

BILL ANDREYCAK

U-132INTRODUCTION

The controlled on-time, zero current switching technique provides a simple and efficient so-lution to obtaining high power factor correction. This discontinuous inductor current approachessentially programs a constant switch on-time during one line half-cycle. It does not require any“complex” analog square, multiply and divide functions to control the instantaneous switch cur-rent as with other PFC techniques. Additionally, zero current switching limits the peak current toexactly twice that of the average inductor current over all line and load combinations. High effi-ciency operation is also achieved with no boost rectifier recovery concerns and power loss. In atypical 80 Watt application the UC3852 PFC technique delivers a power factor of 0.998 with5.8% Total Harmonic Distortion at nearly 94% efficiency.CIRCUIT SCHEMATIC3-235APPLICATION NOTEUC3852 FEATURES

The UC3852 PFC controller contains several fea-tures which minimize external parts count whileproviding excellent performance and protection.Optimized for this off-line PFC application, theUC3852 delivers high power factor (0.997 typical)and a low cost overall solution.U-132CONTROL CIRCUIT ATTRIBUTES

programmable maximum frequency [3, 5]programmable maximum on-time [3, 5]overcurrent indication output [7]OPERATIONAL CHARACTERISTICS

low operating current [8]low start-up current (0.4 mA) [1]few external required components30 V maximum supply input [7]OFF-LINE PROTECTION

undervoltage lockout with hysteresis 16V turn-on, 11 V turn-off [1]clamped 12V gate drive output [2]active low, self biasing output [3]overcurrent protection [4]CONTROL TECHNIQUE

Zero Current Switching [9]controlled on-time [6] high noise immunity [6]UC3852 POWER FACTOR CORRECTION CONTROL IC BLOCK DIAGRAMFigure 2.3-236APPLICATION NOTEUC3852 POWER FACTOR CORRECTION CONTROL U-132PFC TECHNIQUE OVERVIEW

Most power factor correction techniques incorpo-rate the boost topology which can be operated ineither the continuous or discontinuous inductor cur-rent modes and switched at a fixed or variable fre-quency. Generally, the fixed frequency, continuousinductor current variety is preferred for higherpower applications to minimize the peak current.current version operated in a variable frequencyBelow about 500 Watts, the discontinuous inductormode offers several advantages. Benefits includereduced inductor size, minimal parts count and lowcost of implementation. This paper will highlight thecontrolled on-time, zero current switched variety ofdiscontinuous inductor current PFC operation.FUNDAMENTALSCONTROLLED ON-TIME

On-time of the time varies with line and load conditions but shouldPFC switch on-be considered constant for one line half-cycle. Alow frequency bandwidth is necessary in the volt-age error amplifier loop compensation which istypically rolled off to cross zero dB below the linefrequency.ZERO CURRENT SWITCHING

Zero current switching facilitates three importantadvantages in this application. First, the inductorcurrent must be zero before the next switching cy-cle is initiated inferring high efficiency and elimina-tion of the boost rectifier recovery loss. Secondly,the change in inductor current (delta IL) is equal tothe peak inductor current (IL(pk)) since currentstarts and returns to zero each cycle. The discon-tinuous boost converter current waveform has a tri-angular shape with an area (charge) equal toone-half of the product of its height (peak current)multiplied by its base (time). Since the timebasecan be considered as a series of consecutive trian-gles, the peak current is therefore limited to exactlytwice that of the average current. This is valid forboth the steady state and instantaneous switchingcycle relationships. The converter operates right onthe border between continuous and discontinuouscurrent modes which results in variable frequencyoperation.The “fixed” on-time in conjunction with zero currentswitching provide automatic power factor correc-tion of the input current. This can be demonstratedby analyzing the basic inductor waveform usingspecific attributes of this charging and discharging of the inductor current.PFC technique for eithertrolled by the UC3852 circuitry it will be used forSince the inductor charging condition is being con-the analysis.INDUCTOR WAVEFORMFigure 3.For the APPLICATION NOTEPFC POWER STAGE DESIGN

from the AC line input of the preregulator and workIt is advantageous to begin the power relationshipstowards the DC output section. The instantaneousprimary voltage (VP(t)) is related to the steadystate peak input (VP) by the following relationship:3.VP(t)=VPsin(wt)where VP = U-132APPLICATION NOTEthe high voltage DC output, and resistive losses atthese lower powers and currents are minimal.U-132timing circuit. Both t(on)max and t(off)max will beindividually calculated and added together to ob-tain the maximum conversion period, t(per)max.This is required to obtain the inductor value. Equa-tions 12A and 13A will be solved for their respec-tive maximums.12B. t (on) max =13B. t (off) max =CONVERSION PERIOD

The total time for one switching cycle is obtainedby adding the on-time with the instantaneous off-time. Switching frequency is the reciprocal of thecyclical switching period which varies with line,load and instantaneous line voltage.14. t(per) = t(on) + t(off)SWITCHING FREQUENCY

15. f(conv) = 1 / t(per)Switching frequency varies with the steady stateline and load operating conditions along with theinstantaneous input line voltage. Generally, thePFC converter is designed to operate above theaudible range after accommodating all circuit andcomponent tolerances. Many applications can usethirty kiloHertz (30 kHz) as a good first approxima-tion. Higher frequency operation should also beevaluated as this can significantly reduce the in-ductor size without negatively impacting efficiencyor cost. In most applications, the minimum switch-ing frequency will coincide with full load operationduring the peak of the input voltage waveform atlow line. In contrast, the highest frequency conver-sion occurs at light load and high line conditions,just as the input voltage waveform nears the zerocrossing point. A plot of t(on), t(off), t(per) andswitching frequency versus instantaneous line volt-age is shown in figure 4 and for the specific appli-cation circuit of figure 1. Figure 5 demonstrates thetypical changes incurred in conversion frequencyfrom low to high line inputs.INSTANTANEOUS LINE VOLTAGE (VAC)

Conversion Times vs Instantaneous LineNominal Line VoltageFig.4

SELECTING THE OUTPUT VOLTAGE

The boost converter output voltage should be de-signed to be at least thirty volts higher than thepeak of the input voltage at high line. This will pre-vent long conversion cycles due to the small volt-age across the discharging boost inductor. Whenthis thirty volt margin is ignored, the minimumswitching frequency will occur at the peak of highline operation and not at low line, but also at fullload. This will require recalculation of the timing in-tervals.INSTANTANEOUS LINE VOLTAGE (VAC)

Conversion Frequency vs Instantaneous LineFig. 5.INDUCTOR CONSIDERATIONS

The exact inductor value can determined by solv-ing equation 14 for the required inductance at theselected minimum operating frequency. Maximumon-time needs to be programmed into the UC385214A. t(per)max = t(on)max + t(off)maxThe minimum conversion frequency (F(conv)min)corresponds to the reciprocal of the maximum con-version period, t(per)max.15A. F(conv)min = 1 / t(per)max3-239APPLICATION NOTEINDUCTOR VALUEThe inductance value necessary for an applicationcan be obtained by substituting equations 12B and13B into 15A. using the relationship of 14A.This equation provides insight as to the possibleways to reduce the inductor value (size and cost)for a given set of design specifications. The mostobvious approach is to increase the minimum con-version frequency above thirty kiloHertz if none ofthe other parameters (Vo, Po) can be varied.INDUCTOR DESIGN SUMMARY

Generally, the size and cost of an inductor varywith its energy storage capacity, W(L). Althoughmost of the energy is stored in the air gap (with agapped ferrite design), the core set must supportthe necessary flux density (B) without saturating orexhibiting high core loss. The required energy stor-age of the boost inductor is:The number of turns required for a selected coresize and material is:where Bmax is in Teslas and Ae is in square centi-meters APPLICATION NOTEtions can be used:For many applications, the following approxima-ICC = 10mAI(charge) = 10 ms (one-half cycle at 50 Hz)UVLO hysteresis = 5 voltsV(turn-on) U-132APPLICATION NOTEMany other compensation arrangements are possi-ble.Using this compensation network, a low frequencyoff with a single pole (-20 dB/decade) responsegain of approximately 34 dB is achieved. This rollscentering at 1.6 Hz. The gain curve will intersectzero dB at about 120 Hz and result in excellentpower factor correction. Better dynamic responseand less overshoot of the output voltage can beobtained by adjusting the 20 K ohm input resistorto increase low frequency gain and move the zerodB crossing out to a higher frequency. Some slightdegradation of the power factor is to be expectedby increasing the loop response.SOFT START

Soft starting of the output is optional, but recom-mended to minimize the output voltage overshootupon power-up. This does not occur in applicationswhich will always have some load on the output.However, most electronic ballast have either noload, or a very light load on the output at power-upand will see the overshoot. Soft start implementa-tion requires only a diode and capacitor from thecompensation pin to ground. Another diode fromthe capacitor to VCC discharges the soft start ca-pacitor to the falling Vcc voltage when the AC linepower is removed. This will guarantee that the cir-cuit will always start up in soft start if the line is ACplug is removed for a few seconds. Again, this isan optional feature which depends on the applica-tion.One “trick” to significantly reduce the size of thesoft start capacitor is to replace the diode with acheap PNP transistor. A capacitance multiplier canbe obtained by connecting the error amplifier output and soft start capacitor fromPNP emitter to thethe base to ground. The collector of the transistorthe capacitance value up by beta of the transistoris connected to ground. This adaptation will scaleat the amplifier output. A 2N2907 or equivalent is apopular choice and will reduce the capacitancevalue by a factor of approximately 50.A 1N914 or 1N4148 signal diode should be usedfrom the base to emitter to prevent negative base-emitter voltages from damaging the transistor. Ad-ditionally, this transistor can easily be interfacedwith any optional fault protection schemes to softstart the controller following a fault.SOFT START IMPLEMENTATIONCURRENT SENSE

Current in the PFC design is sensed in the returnline of the preregulator circuitry at the AC inputbridge rectifiers. One side of the current sense re-sistor is referenced to the UC3852 “ground” con-U-132nection. The other end of the resistor develops thecurrent sense voltage which is equivalent to minuscircuitry incorporates two comparators, one forIL(t) * Rsense. The UC3852 zero current detectionzero current detection and another for over currentprotection.ZERO CURRENT DETECTION

The zero current detection circuitry uses a negative10 negative threshold guarantees that there are nomillivolt (-10mV) threshold as its reference. ThisFigure 7.startup problems since this input must be pulledbelow ground for normal operation. Whenever thezero detect input is raised above the minus ten mil-livolt threshold, the comparator is triggered and thenext switching cycle begins.Inductor current can be sensed by a current senseduring an overcurrent condition. This should onlyresistor which develops minus 400mV maximumoccur at a twenty percent overload, or 1.2 APPLICATION NOTEduce the amount of EMI/RFI filtering required byminimizing the rectifier recovery noise. For best re-sults, the filter delay time should match the rectifi-ers recovery time. A ten ohm resistor and a onenanoFarad (1 nF) capacitor are good starting val-ues.U-132inductor current and have an insignificant impacton power factor. However, this modification can reADVANCED PROTECTION CIRCUITRY

Certain applications of the UC3852 control IC mayrequire sophisticated protection features. Some ex-amples of these options are overvoltage protectionand restart delay, soft start or latch-off following afault. Each of these features can be added to thecontrol circuit with a minimal amount of externalparts, and often combined using shared compo-nents.OVERCURRENT FAULT PROTECTION

The UC3852 contains and overcurrent comparator(-400mV) which quickly terminates the PWM out-put. This comparator also drives circuitry con-nected to the ISET pin which raises its normal 5volt amplitude to 9 volts during the overcurrentcondition. In addition to programming the ramp ca-pacitor charging current, the ISET pin can be usedto drive external fault protection circuits. A resistorin series with a 5.6 volt zener diode to the ISET pinwill develop approximately 3.4 volts across the re-sistor when an overcurrent fault is detected. Thissignal can be used to trigger external shutdown orhiccup circuitry.Figure 8.Figure 9.GATE DRIVEThe UC3852 PWM output section is MOSFETcompatible and rated for a one amp peak current.This totem pole design also features a twelve volt(12V) clamped output voltage to prevent excessivegate voltage when used with unregulated (Vcc)supply voltages. A twelve ohm resistor betweenthe UC3852 and the MOSFET switch gate will limitthe peak output current to its one amp maximumduring normal operation.Additionally, the UC3852 self biasing active low to-tem-pole design holds the MOSFET gate low dur-ing undervoltage lockout, preventing catastrophicproblems at power-up and removal of the AC input.LIST OF COMPONENTS

C6 = 1 uF, 35VD5,6 = IN4148D7= 6.2V ZENERD8 = 40 V ZENERQ2,4 = 2N2907Q3 = 2N2222R9,10=10KR11 = 1 MEGR12 = 24 KR13 = Calculate for OVPR14 = 1 K3-243APPLICATION NOTETRANSFORMER COUPLED

CURRENT SENSE

Figure 10.Soft start is programmed by R11, C6 and the betaof Q2. Overcurrent protection starts at the UC3852U-132APPLICATION NOTEU-132tablished, typically 80 VAC. Capacitor C11, a smallfilter capacitor and the base of transistor Q10reach a voltage of V(C10) minus the Zener forwardvoltage drop of diode D11. As this voltage rises,the emitter of Q10 and voltage across resistor R13follows, offset by the base-emitter diode drop ofQ10. This increasing bias pulls more current fromDual AC Input Range (110/220 VAC)Feedforward Circuitthe UC3852 Figure 14.OTHER PFC APPLICATIONS

The basic PFC schematic of Figure 1 can be usedas a template for other PFC applications with dif-ferent input voltage ranges and output power lev-els. A majority of the changes will be toaccommodate higher ( or lower ) voltages and cur-rents. Once familiar with the complete design pro-cedure as outlined in this application note,designers are encouraged to recalculate the val-ues for their applications using the same guide-lines.UNIVERSAL AC INPUT RANGE

The UC3852 controlled-on time, zero currentswitched PFC technique can be used to accom-modate wide AC input voltages with the addition ofa simple feedforward circuit. This external circuitryis required to cancel out the line dependentchanges in the switch on-time over the three-to-one input range from 85 to 2 volts. Otherwise,the approximate nine-to-one control range of theUC3852 on-time would be fully used for line regu-lation allowing no accommodation for loadchanges.CIRCUIT OPERATION

The rectified input voltage is applied across thenetwork consisting of R10 through R12, D10 andC10. Capacitor C10 charges to the peak of the di-vided input voltage and is large enough to maintainthis level over one line cycle. Diode D11 serves asan offset to bypass the range extender circuitry un-til a sufficient minimum line voltage has been es-APPLICATION NOTECONTINUOUS CURRENT PFC BOOST CONVERTERU-132Figure 15.CONTINUOUS PFC CURRENT IMPLEMENTATIONFigure 16.Figure 17.3-246APPLICATION NOTEUC3852 CONTROLLED PFC FLYBACK CONVERTER

U-132Figure 18.UC3852 AS A CAPACITIVE DISCHARGE DRIVER

Figure 19.3-247APPLICATION NOTEAUTORANGE (110/220) VOLTAGEFEEDFORWARD CIRCUIT

Input line voltage U-132tual inductor current sense more positive. Therfore, the zero current detection threshold icrossed before the inductor current is actually zerand the PFC circuitry. Basically, the TL431 is

used as a comparator to switch in a second timingresistor (RSET’) when the input voltage exceeds apreset threshold.The AC input voltage is rectified by diode D20 and‘divided down by resistors R20 and R21. CapacitorC20 peak charges and filters this waveform to de-velop a DC voltage proportional to the input line.RSET is programming the initial charging currentto the timing capacitor CRAMP. When the voltageacross C20 exceeds the 2.5 V threshold of theTL431 comparator, its output goes low. This placesFigure 20.a second timing resistor, RSET’, in parallel with theoriginal one thus increasing the current to CRAMPand performing line feedforwardcompensation at approximately 155 VAC which ismid-range between high line of a 110 VAC) and low line for a 220 RSET’ must be selected to ac-count for the TL431 output saturation voltage.CONTINUOUS CURRENT

PFC technique can alsobe modified to operate in the continuous inductorcurrent mode. A positive amplitude, small offsetsignal is derived from the input voltage waveform.It gets added to the normal current sense signalwhich is negative with respect to ground. Summingthese two signals to the ZERO input biases the ac-APPLICATION NOTEgrammed by Ra, Rb and Rc according to the fol-lowing formulas.Vout(max) Vout(min) (5*(Ra+Rb))/Rbwhere Rx=(Rb*Rc)/(Rb+Rc)NON

PWM (non PFC) applications using avariable frequency control techniques can also beimplemented with the UC3852. This applies to bothcurrent mode and variable ON-Time control meth-ods. Typical examples of these are discontinuouscurrent boost and APPLICATION NOTEPERFORMANCE EVALUATION

The UC3852 controlled PFC circuit shown in Fig-ure 1 was constructed using the list of materialsprovided for this application. Power Factor and To-tal Harmonic Distortion to the 50th harmonic wereanalyzer. Test results indicated a power factor ofmeasured using a VOLTEC PM- 3000 AC power0.998 and T.H.D. below 6% at nominal line and fullload. Very similar readings were obtained over thecomplete input voltage range and a moderate loadchange. Zero Current Switching (ZCS) facilitateshigh overall efficiency with this PFC technique.UC3852 PFC TEST CIRCUITSPECIFICATIONS:VIN = 350 VDCPOUT = 86 WMEASURED PERFORMANCE:

P.F. = 5.81%TEST CONDITIONS;(nominal line)VIN = 0.799 AACPIN = 91.84INRUSH Ipk = 355.6 VDCIOUT = 86.1 WEFFICIENCY 3.91%5th 0.38%9th 0.21%LIST OF MATERIALSCAPACITORSC2 = 0.47 uF / 200 VC3 = 82uF / 400 VC4 = 22uF / 35 VC5 = 0.1uF / 35VC6 = 1nF / 16Vc7 = 0.1uF / 16V

DIODESD1-4 = 1N4937, 1A / 600 Vtrr = 1N4148, 0.2 A / 50 VINDUCTORSL2 = 1 mH Boost inductorL3 = 100 k ohms 1 WattR2 = 18.2 k ohms 1% 1/2 = 1 meg ohm 1/4 = 330 k ohms 1% 1/2 = 20 k ohms 1/4 IMPORTANT NOTICE

Texas Instruments and its subsidiaries (TI) reserve the right to make changes to their products or to discontinueany product or service without notice, and advise customers to obtain the latest version of relevant informationto verify, before placing orders, that information being relied on is current and complete. All products are soldsubject to the terms and conditions of sale supplied at the time of order acknowledgement, including thosepertaining to warranty, patent infringement, and limitation of liability.

TI warrants performance of its semiconductor products to the specifications applicable at the time of sale inaccordance with TI’s standard warranty. Testing and other quality control techniques are utilized to the extentTI deems necessary to support this warranty. Specific testing of all parameters of each device is not necessarilyperformed, except those mandated by government requirements.

CERTAIN APPLICATIONS USING SEMICONDUCTOR PRODUCTS MAY INVOLVE POTENTIAL RISKS OFDEATH, PERSONAL INJURY, OR SEVERE PROPERTY OR ENVIRONMENTAL DAMAGE (“CRITICALAPPLICATIONS”). TI SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED, AUTHORIZED, ORWARRANTED TO BE SUITABLE FOR USE IN LIFE-SUPPORT DEVICES OR SYSTEMS OR OTHERCRITICAL APPLICATIONS. INCLUSION OF TI PRODUCTS IN SUCH APPLICATIONS IS UNDERSTOOD TOBE FULLY AT THE CUSTOMER’S RISK.

In order to minimize risks associated with the customer’s applications, adequate design and operatingsafeguards must be provided by the customer to minimize inherent or procedural hazards.

TI assumes no liability for applications assistance or customer product design. TI does not warrant or representthat any license, either express or implied, is granted under any patent right, copyright, mask work right, or otherintellectual property right of TI covering or relating to any combination, machine, or process in which suchsemiconductor products or services might be or are used. TI’s publication of information regarding any thirdparty’s products or services does not constitute TI’s approval, warranty or endorsement thereof.

Copyright © 1999, Texas Instruments Incorporated