 |
|
| If you can't view the Datasheet, Please click here to try to view without PDF Reader . |
|
|
|
| If you can't view the Datasheet, Please click here to try to view without PDF Reader . |
|
|
| Datasheet File OCR Text: |
| TELEFUNKEN Semiconductors ANT016 Bipolar Power Transistors in Electronic Ballasts Selection Criteria and System Requirements Issue: 11. 95 ANT016 TELEFUNKEN Semiconductors Table of Contents Introduction ..................................................................................................................................................................1 Lighting Systems...........................................................................................................................................................2 Compact and Industrial Ballasts.................................................................................................................................2 Design Parameters ........................................................................................................................................................3 Equivalent Circuit ......................................................................................................................................................3 Ignition ................................................................................................................................................................4 Normal Operation................................................................................................................................................4 Requirements Regarding Power Transistors .............................................................................................................6 Power Capability........................................................................................................................................................6 Grouping System tx....................................................................................................................................................7 Working Voltage VCEW .............................................................................................................................................8 Power Factor Correction ...........................................................................................................................................10 Passive Power Factor Correction .............................................................................................................................10 Active Power Factor Correction...............................................................................................................................10 Measures for Protection .............................................................................................................................................12 Upper Transistor.......................................................................................................................................................12 Lower Transistor with b-e Short Circuits.................................................................................................................13 Lower Transistor with Interrupted Base Current .....................................................................................................13 Lower Transistor with Negative b-e Voltage ...........................................................................................................14 Circuit Examples ........................................................................................................................................................15 Ballast Unit for 8 W at 120-V Mains .......................................................................................................................15 Ballast Unit for 8 W at 230-V Mains .......................................................................................................................16 Ballast Unit for 40 W at 230-V Mains .....................................................................................................................17 Ballast Unit for 100 W at 230-V Mains ...................................................................................................................18 Issue: 11. 95 TELEFUNKEN Semiconductors ANT016 Introduction Better light efficiency and longer lifetime are becoming the main reasons for the substitution of incandescent bulbs with fluorescent lamps. Fluorescent lamps are supplied from the line voltage and must be driven via a ballast unit. There are three types of ballast units; the almost conventional type, the conventional type with lower power dissipation and the electronic ballast unit. By using the electronic ballast unit, the greatest economic profit is possible. A principal distinction can be made between the electronic ballasts for compact lamps - where the lamp and the ballast are mechanical units which can not be separated without destruction - and the electronic ballast for industrial lamps where lamps are independent and can be exchanged individually. TEMIC has been offering a type program of high-voltage bipolar transistors which is specially developed and optimized for the requirements in the field of electronic ballast units for many years. This paper describes the various requirements for the bipolar power transistors. Rough mathematical calculations will give an overview with regard to the device performance of the ballast unit. Application-specific data of bipolar power transistors and their benefits for the customer will be explained in order to make the devices suitable to the requirements in electronic lighting systems. In addition, some measures to increase the power factor will be discussed. Finally, some circuit design examples of electronic ballast units will be provided. Issue: 11. 95 1 ANT016 Lighting Systems Comparisons given in table 1 show the advantage of electronic ballast units for fluorescent lamps over all other lighting systems. This is explained through the better conversion of electrical to light power - a result of the lower power loss in the ballast as well as of the longer life time of the lamps. The column Fluorescent Lamp/Electronic ballast" includes compact and industrial lamps. * TELEFUNKEN Semiconductors No ignition because the glass of the lamp is broken or the tube is oxygen-contaminated while the lamp filament is intact. This leads to a permanent ignition mode, the transistors become overheated and, if the transistors are not switched off, are destroyed. If the lamp filament is burned down, there is normally no danger for the electronic ballast unit, because the series resonant circuit is open as shown in figure 1. * Compact and Industrial Ballasts In contrast to compact lamps, it is not acceptable that electronic ballasts for industrial lamps are destroyed when the fluorescent lamp fails. In principle, an emergency switch-off is needed for this type of ballast. This, however, increases the circuitry. In case of lamp failure, there are two possible operation modes which may occur: An additional essential feature of industrial ballasts is the possibility of lamp exchange during operation. This is practicable because the lamp interrupts the switching operation of the ballast. After re-installation, lamp ignition is started. Several possible configurations are shown in figure 1. Suitable protection measures in case of lamp failure are discussed in the chapter Measures for Protection". Table 1. Comparison of typical operating and system characteristics of the most common lighting systems Characteristics Incandescent Lamp Fluorescent Lamp Conventional ballast W/lm-ratio Life time in h Light Stroboscopic effect Humming noise Starting behavior Light efficiency Alternation numbers Frequency Series choke Choke loss Factor of phase angle Power factor 100% 1000 Flicker-free No No Immediately 5.60% - 50 Hz No No 1 1 40% 5360 Cathode flickering Yes Yes Fluorescent Lamp Low-power converter ballast < 40% 5360 Cathode flickering Yes Yes Fluorescent Lamp Electronic ballast 20% 8000 Flicker-free No No Without flickering 27.80% 500000 (60 s/ 150 s) 30 - 40 kHz Small/ light 9.90% 0.95 (cap.) 1 with PFC* Flickering when started Flickering when started 22.50% 15000 (60 s/ 150 s) 50 Hz Big/ heavy 18.30% Inductive Inductive 22.50% 15000 (60 s/ 150 s) 50 Hz Big/ heavy < 18.30% Inductive Inductive Note: * PFC = Power Factor Correction 2 Issue: 11. 95 TELEFUNKEN Semiconductors ANT016 Intact fluorescent lamp Fluorescent lamp with burned-down filament Non-ignitable fluorescent lamp Figure 1. Operation modes of electronic ballast units R2 F1 D1 L1 D2 R1 D5 L2 R4 R6 D7 C4 CP Mains C1 C2 L1 D3 D6 L2 C3 R3 L2 R5 T2 D8 LV D4 Figure 2. Typical circuitry of an electronic ballast unit Design Parameters When designing an electronic ballast for fluorescent lamps, the parameters of the lamp, such as power, ignition voltage, operating voltage and the resistance values of the filament, have to be known. Most of the European electronic ballasts work with a configuration that has a ring core as a saturation transformer (see figure 2). The converter is therefore self-oscillating. It can be assumed that the oscillating frequency is nearly constant in a given operating range. Most of the circuits work in the range of 30 - 40 kHz because the light efficiency of the fluorescent lamp is at its maximum value above 35 kHz. Higher frequencies, however, cause higher switching losses and higher expenditure of the line filter respectively of electromagnetic compatibility. The components of the series resonant circuit, ballast choke (LV) and parallel capacitor (CP) influence the Issue: 11. 95 operating performance of the ballast. The capacitor CV is necessary to avoid a dc portion being applied to the lamp as this would reduce the life time of the lamp. Equivalent Circuit Three different modes can be separated under correct operation: * * * Pre-heating of the filament with non-ignited lamp (figure 3) Ignition mode with non-ignited lamp (figure 3) Permanent operation with ignited lamp. The lamp characteristic is similar to a Zener diode (figure 7) 3 Lamp T1 CV ANT016 Ignition CV VS VC in Volt P TELEFUNKEN Semiconductors CP V Lp 2 VS LV VL IC 3000 2500 2000 1500 1000 500 0 30 31 32 33 34 35 36 37 38 f in kHz Figure 5. Voltage at parallel capacitor Cp as a function of the operating frequency (f0 = 35 kHz) Figure 3. Equivalent circuit in the ignition mode The filament has been neglected in all equivalent circuits. The following equation for the voltages is therefore valid: r r VS r = VLp + VL (1) 2 and then: r r VS 1 = j x IC x x L - 2 x C The voltage at the parallel capacitor must be limited to defined ranges for the operating modes pre-heating of the filament" and ignition",. In the pre-heating mode", the voltage must be below the smallest ignition voltage of the lamp (valid at room temperature). In the ignition mode", the peak value of the voltage calculated in (3) has to be higher than the highest ignition voltage of the lamp. The plot shown in figure 5 is calculated with (5). VCIgnit. = (2) f 2 - f2 VS 2 x 0 2 + (2 x R fil. x x C P ) f 2 0 where: f0 = = = resonant frequency operating frequency resistance of filament 2 (5) VLp VS 2 VL IC f Rfil. Figure 4. Example of a vector diagram with nonignited lamp Figure 4 shows the vector diagram in the ignition mode, assuming capacitive de-tuning of the resonant circuit. This means that the operating frequency of the converter is below the resonant frequency. The maximum attainable ignition voltage is defined as: r VC P = r VS 2 x 20 2 - 02 There are two possibilities to change the lamp voltage before ignition: * Change the operating frequency * Change the resonant frequency by variation of LV or CP max (3) Normal Operation Under normal operation, calculations describing the behavior of the ballast become more complex because there is no proceeding possibility to use equations and to calculate the parameters, like the lamp current in a closed loop. The maximum ignition current is the peak collector current and is given with: r $ I C max = 2 x VC P xxC (4) max 4 Issue: 11. 95 TELEFUNKEN Semiconductors ANT016 The real and the imaginary part of the inverter output voltage VS can be calculated with the actual angle 1. The real part, the imaginary part and the modulus of the vector of the voltage at the ballast choke LV (VL) can be calculated with these results. Il Ic Ilp 2 Vl Vlp Vs 2 1 V r Re(VL ) = Re S + VLp 2 V Im(VL ) = Im S 2 VL = Re(VL ) + Im(VL ) 2 2 (10) (11) (12) Figure 6. Example of a vector diagram during normal operation of the ballast CV VS 2 VS CP VLp LV VL IC I L I Lp The angle between the lamp voltage vector and ballast choke voltage vector can be calculated by using equations (9) and (12). VS Im 2 2 = arcsin VL (13) LP The modulus of the current flowing through the ballast choke LV can be calculated by using equation (12). r VL r IL = (14) xL r Im(I L ) = cos 2 x I L Figure 7. Equivalent circuit under normal operation The construction of the vector diagram shown in figure 6 is based on the following considerations: In the first step, the angle 1 must be fixed arbitrarily, then the solution can be found with iterative calculations by variation of 1. This iteration starts with the voltage equation in (1) and the current node equation: r r r (6) I L = IC + I Lp where: r r IC = x C P x VLp = IC ! r (15) The iteration can be stopped if the condition in (15) is valid. Only if the imaginary part of IL is equal to the modulus of IC can a possible solution be achieved. r Re(I L ) = sin 2 x I L = I Lp (16) Equation (16) helps to calculate the value of the lamp current, equation (17) to calculate the lamp power. PLp = U Lp x I Lp (17) (7) The possible lamp power can therefore be calculated with the knowledge of the four parameters LV, CP, and Vs. Otherwise, this is only possible if the operating frequency is the resonant frequency. In this case, the equations (1) and (2) from the application note 002 (November 1991) can be used: LV = CP = VS x VLp 2 x 0 x PLp 1 02 x LV (18) (19) The value of the current flowing through the parallel capacitor CP (IC) is now fixed. It is assumed that the vector of the lamp voltage (VLp) is parallel to the real axis of the vector. The vector of the current through the capacitor is therefore parallel to the imaginary axis. r VS VS Re (8) = sin 1 x 2 2 r VS VS Im (9) = cos 1 x 2 2 Issue: 11. 95 5 ANT016 TEMIC's aim is to optimize and to develop all of its bipolar power transistors for customer-defined requirements. It is therefore necessary to know the typical conditions wherein the components are able to work well. The most important system variables for electronic ballast units for fluorescent lamps are the following: * Current load in the case of ignition and normal operation with maximum expected value of line voltage * RBSOA, voltage requirement * Switching behavior, base-drive conditions, switching frequency * Cooling conditions, power dissipation, ambient temperature * Dependence of the case temperature through other parts of the circuit Standard electronic ballast units are low-cost applications", where the base drive conditions are in principle not optimum. The transistor is activated by a base current when the anti-parallel free-wheeling diode is active and the collector current is negative. This leads to an immense overcharge of the collector-base diode. Transistors which are not optimized react in such a case with long storage times and extreme fall times. Lowest values of the power dissipation are necessary to operate with such transistors under ambient temperatures in the range of 100C - 110C without any heatsink. TELEFUNKEN Semiconductors Requirements Regarding Power Transistors IC VCE VCE : 50 V/div. I C : 0.2 A/div. T : 2 s/div. Figure 8. Current and voltage course Power Capability TEMIC specifies the power transistors of the SWOT (Simple sWitch-Off Transistor) type range especially for the requirements of electronic lighting application to give the user the best possible support for selection. The maximum collector current a TEMIC bipolar power transistor can operate in application is therefore the propagated collector current given in the data sheet. TEMIC defines the typical collector current as a dccurrent gain of 4 and a collector-emitter saturation voltage of 2 V. A factor for the overcurrent for electronic ballast units can be fixed to be the quotient of the maximum ignition current and the normal operating current. This leads to a typical value of 6 and furthermore, the definition given below of the system-collector current": I Csys = IC 6 (20) TEMIC SWOTs, driven with a dc-current gain of 4, have a typical collector-emitter saturation voltage of 0.1 V for this collector current. An example of the BUF630 for this definition is shown in figure 9. 6 Issue: 11. 95 TELEFUNKEN Semiconductors ANT016 I Csys VCE = 2 V IC h FE 100 10 1 0.01 0.1 1 10 I C (A) Figure 9. Definition of collector- and system current P Lamp 160 140 120 100 80 60 40 20 0 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 I Csys T1 L VS Lamp CV VL CP I ~ 1/f T2 Figure 10. Relationship of system current / lamp power In general, a system current of 15 mA represents a lamp power of 1 W for electronic ballast units at 230 V mains. This relation is shown in figure 10. Figure 11. Output power to the series inductance, to a change of output power. Unfortunately, bipolar power transistors in electronic ballast units are driven in many different ways, depending on the special circuit dimensioning. The typical switching characteristics in the data sheets do not give any relevant information about the real dynamic behavior in an actual electronic ballast circuit. The dynamic characteristics are influenced by the dc-current gain, blocking voltage, the technology used and by the chip size. These parameters are given by the transistor design of the manufacturer (TEMIC). Other parameters like the working point, type of load and the switch-off conditions depend on the application which is given by the manufacturer of the electronic ballast unit. Grouping System tx Most of the circuits for electronic ballast units using the half-bridge configuration have a strong dependence on output power and oscillating frequency: V PL VL x I L (21) f A change of storage time in bipolar power transistors leads to a change of the oscillating frequency and, due Issue: 11. 95 7 ANT016 IB > 2 C V DC tP : T 1 : 10 IB TELEFUNKEN Semiconductors IC t DUT IC T tP t x Figure 12. Typical storage time tx for bipolar power transistors in the field of electronic ballast units As a conclusion, there are mainly two problems to be solved: * The spread of the storage time of the bipolar power transistor in the application must be as small as possible. The absolute value of the storage time of the bipolar power transistor in the application must be available to fit in the application. Base drive T t * Figure 13. Base emitter resistance Base drive TEMIC's solution: Common to all configurations of electronic ballast units is the fact that transistors are heavily overdriven and that the reverse base current during switch-off is very low. TEMIC therefore offers the most relevant switching parameter tX for their bipolar power transistors. This parameter is controlled and measured in production for 100%. The basic test circuit is shown in figure 12. T Figure 14. Negative base emitter voltage Working Voltage VCEW In over 90 % of the switching applications, the bipolar power transistors are blocked either by short circuit respectively resistance < 100 between base and emitter (see figure 13) or negative base emitter voltage, as shown in figure 14, but not with a base emitter open circuit as shown in figure 15. Such switching applications are: * Electronic ballast units for fluorescent lamps * Electronic transformers for halogen lamps * Switch-mode power supplies So the value of VCEO is not the parameter to determine the switching capability of a bipolar power transistor. TEMIC defines the switching capability of a bipolar power transistor with the working voltage, as the socalled VCEW. 8 Base drive T Figure 15. Base emitter open circuit A transistor switched off under the conditions shown in figures 13 and 14 can be switched on and off, depending on the base drive condition and of course on the collector current, up to VCES. Therefore, TEMIC defines VCEW as the maximum voltage at which a TEMIC bipolar power transistor can be switched on and off without any risk at a defined collector current (ICW) and base drive condition. This results in a collector current versus working voltage area which is also known as the RBSOA- or FBSOA diagram of a bipolar transistor. TEMIC defines these as Safe Working Area". The maximum collector current in the Issue: 11. 95 TELEFUNKEN Semiconductors diagram in figure 16, ICW, is the collector current propagated in the data sheet. The highest working voltage is achieved if the reverse base current (IB2) is greater than 10% and less than 50% of the collector current. It is assumed that the transistor should have a saturation voltage less than 2 V for the forward base current. I CW ANT016 TO 220 Package (Continued) Type VCEO VCEW VCES V BUF642 BUF672 400 450 450 450 450 V 500 550 550 550 550 V 900 900 1000 1000 1000 IC A 6 11 5 6 7 ICW A 6 11 5 6 7 0.1 x I C < I B2 < 0.5 x IC BUF636A BUF640A IC VCEsat < 2 V BUF646A TO 251 (DPAK) Package Type V CEO V CE VCEW VCES VCEO VCEW VCES V V 300 500 500 550 550 V 600 700 700 1000 1000 IC A 2 4 6 1.6 5 ICW A 2 4 6 1.6 5 BUD600 BUD620 BUD630 BUD616A 250 400 400 450 450 Figure 16. Principle diagram of Safe Working Area" The maximum working voltage VCEW of the SWOT types (BUF..) is 100 V above VCEO as shown in table 2: Table 2. Propagated values of the working voltage TO 220 Package Type VCEO VCEW VCES V BUF620 BUF630 BUF644 BUF650 BUF654 400 400 400 400 400 V 500 500 500 500 500 V 700 700 700 700 700 IC A 4 6 8 10 12 ICW A 4 6 8 10 12 BUD636A TO 252 (DPAK SMD) Package Type VCEO VCEW VCES V BUD600SMD BUD620SMD BUD630SMD BUD616ASMD BUD636ASMD 250 400 400 450 450 V 300 500 500 550 550 V 600 700 700 1000 1000 IC A 2 4 6 1.6 5 ICW A 2 4 6 1.6 5 Issue: 11. 95 9 ANT016 Power Factor Correction Nearly all mains-operated electronic devices without power factor correction (Fig. 17) have typical line currents like the one shown in figure 18. TELEFUNKEN Semiconductors Usually, the power factor will be described with harmonic analysis (Fourier analysis). An unity power factor can be achieved if the current wave contains only the fundamental wave. Harmonics are included if the power factor is less than 1. The procedures can be distinguished as active and passive ones. Mains Load, e.g. electronic ballast unit Passive Power Factor Correction At the moment, there are three different known principles: * Choke with large inductance (100 Hz) Disadvantages: high weight and enormous volume Figure 17. Typical peak value rectification of the mains supply voltage Without considering a phase angle between voltage and current, the power factor can be defined as follows: P (22) k= I RMS x VRMS If the current and voltage are sine waves, the power factor will be k = 1. Assuming an impressed mains voltage (which is not always the case), the power factor is determined by the wave form of the current. As shown in figure 18, the power factor is approximately 0.5. In other words: the RMS value of the mains current is, for this example, two times higher than it could be if the power factor had been 1, while the power consumption in both cases would be the same. This is a great disadvantage in the field of energy distribution for power cables, transformers and open wire lines. The reason for this is that the line-cross section must be dimensioned for double the value without power factor correction. * Special charging circuit for the smoothing capacitor where 2 capacitors were loaded in series and discharged by the load in parallel Disadvantage: a large ripple on the dc voltage * Rectification of the high frequent oscillation of the converter output using the boot-strap principle Disadvantage: dependence of the dc-output level on the RF pulse current The circuits 2. and 3. are protected by patent law and are therefore not described in detail here. Active Power Factor Correction Most of the circuits for active power factor correction use the boost converter configuration as shown in figure 19. The circuit is similar to a fly-back converter, but has no galvanic isolation between input and output. Mains rectifier Load, e.g. electronic ballast unit PFC IC Figure 19. Typical configuration of active power factor correction circuits The circuit regulates the mains supply current to sinewave form and the dc-output voltage to a constant value. In order to enable the regulation, the dc-output voltage must be higher than the peak value of the rectified mains voltage. Two different operation modes for active power factor correction are known: Issue: 11. 95 Figure 18. Typical wave forms of mains current and supply voltage without power factor correction 10 TELEFUNKEN Semiconductors * The so-called discontinuous mode, where the average value - not the momentary value of the mains current - is regulated to sine wave form. This procedure is suitable for loads with nearly constant output power, e.g., electronic ballast units. * The so-called continuous mode is where the momentary value of the mains supply current is regulated to sine wave form. This procedure is suitable for loads with greater changes in output power. A disadvantage compared to the discontinuous mode is the ANT016 necessity for a higher inductance of the boost. This mode is therefore more expensive. The regulated dc-output voltage is an enormous advantage of the active power factor correction which also simplifies the design and dimensioning of the ballast. On the other hand, more additional components are needed for the active power factor correction. Only a passive power-factor correction therefore makes sense in the field of low-cost compact lamps. Issue: 11. 95 11 ANT016 Measures for Protection An immediate switch-off within a few seconds is necessary in case of operating problems such as lamp failure or short circuit at the output terminals of the ballast. It is not easy, however, to switch off a self-oscillating converter. In addition, a switch-off may be dangerous for the power transistor. In the following sub-chapters, some examples for switch-off circuits are described and discussed taking the semiconductor manufacturer's point of view into account. TELEFUNKEN Semiconductors VCEO > VDC are therefore necessary to ensure a safe switch-off operation The advantage of the working voltage VCEW (see chapter Working Voltage VCEW") which is typically 100 V above VCEO for the SWOT types can not be used with this kind of switch-off procedure. C2 R1 D3 T1 D1 Lamp Upper Transistor R2 R3 C1 T1 D1 V dc Rk2 Th1 Rk1 L D2 S L T2 D2 R4 R5 Rk3 Figure 20. Switch-off T1 Figure 22. Example of switch-off circuit The use of cheap power transistors with good dynamic characteristics and high dc-current gain is impossible under the described circumstances. The RF oscillation represents a high SOA load for the transistor. C2 T1 D1 0.7 V Vdc Vrk Vrb Re Rk 0.7 V D2 L 0.7 V Th1 T2 (above) IC1 : 1 A/div, (below) IC2 : 1 A/div, T : 20 s/div Figure 23. Equivalent circuit during switch-off under the conditions of figure 20 The equivalent circuit shown in figure 23 shows that T1 can not be switched off via Th1. The base emitter voltage of T1 and the voltage drop at Re can be limited only to approximately 2.1 V. This does not prevent a positive base drive current of T1. If T1 is driven, VCB cannot reach values smaller than VDC - 1.4 V which represents a high power dissipation and a tremendous FBSOA (Forward-Biased Safe Operating Area) - load. Figure 21. Moment of switch-off Although the switch off is initialized (start of RF oscillations) as shown in figure 21, the converter is active for some periods in a decay process. This is much more critical because the upper transistor is not saturated during this operation when a collector current is flowing. The collector-emitter voltage of T1 is approximately the dc-supply voltage. Transistors with 12 Issue: 11. 95 Lamp Rb TELEFUNKEN Semiconductors ANT016 T1 D1 Lower Transistor with b-e Short Circuits T1 D1 Rb Rk S T2 D2 Re L S T2 L D2 Figure 25. Interruption of base current Figure 24. Switch-off T2 This switch-off procedure is particularly suitable for ballast configurations as shown in figure 22 where the resonant circuit is connected with the plus wire of the dc-supply voltage. Transistor T2 cuts off the energy supply for the resonant circuit, which means the decay process will be as short as possible. The Switch S has to carry on the one hand the reverse base current of T2 and the base drive current which will be supplied from the driving transformer. The saturation voltage must be below 0.3 V and this could therefore cause problems. The short circuit with S should be as good a fit as possible. Four different kinds of switches are possible: * A short circuit" with an SCR that includes a saturation voltage of minimum 0.7 V (disadvantage: no absolute security, even at high temperatures) * Due to the finite dc-current gain, a bipolar transistor switch may be a solution depending on the design of the electronic ballast unit * A power MOSFET is a suitable switch-off device, provided that RDSon is low enough (disadvantage: the limitation of the base emitter reverse voltage may have a negative influence on the dynamic behavior) * The lowest saturation is achieved with a relay contact (disadvantage: contact bouncing or operate delay might be possible) V dc R4 R5 R7 D3 C2 Th1 D2 R3 T1 Rk R6 R8 T3 D5 Rk L T2 D4 R1 Vdc C1 R2 D1 Rk C3 Lamp C4 Figure 26. Interruption with a bipolar transistor is necessary for the reverse base current,. R3 and R4 should have different values of the resistance to ensure symmetrical switching behavior of T2 and T3, Th1 has to be switched on in case of lamp failure. T1 will therefore be switched off and the base current is then interrupted. In addition, ignitions of the DIAC D2 are also interrupted. A power MOSFET can be used the same way instead of a bipolar transistor. R4 R5 R7 Rk D2 R3 C2 Rk D3 T1 R6 R8 T3 D5 L T2 C3 D4 Lamp R1 R2 D1 Rk C4 C1 Lower Transistor with Interrupted Base Current A configuration as shown in figure 25 can be used to avoid the strict requirements for the saturation voltage of the switch-off device. A base emitter resistance for T1 and T2 is necessary to ensure fast storage and fall times (tS, tF). Under normal operating conditions, T1 is driven via R1. The transistor must not limit the base current of T3. D3 Issue: 11. 95 Th1 Figure 27. Interruption with a MOSFET The functional principle of the circuit shown in figure 26 is almost similar to the one shown in figure 27. D3 is necessary to limit the gate-source voltage of T1. The maximum value of the negative voltage amplitude at Rk plus the Zener voltage must be smaller than VGSS of T1 for the selection of the Zener voltage. 13 ANT016 Lower Transistor with Negative b-e Voltage T1 D1 TELEFUNKEN Semiconductors The auxiliary voltage to switch off T2 is generated with C1 loaded via R3. The Z-diode D1 limits the voltage at C1 to values smaller than VEBO of the power transistor T2. The electronic ballast unit can be switched off immediately if a 0.7 V pulse is applied to the gate of Th1. If Th1 is triggered, the plus terminal of C1 will have approximately ground potential and D2 is blocked. T2 is triggered via R4 and reverse base current is flowing through the circuit of base emitter of T2, Th2, Re, Th1 and C1 simultaneously. The gatecathode voltage of Th1 is used to keep the starting capacitor discharged to avoid new trigger pulses from the DIAC (not shown in figure 29). T2 L D2 Vb S Figure 28. Switch-off T2 An emergency switch-off circuit, as shown in figure 28, is the safest procedure with regard to the bipolar power transistor. An auxiliary voltage, realized with an auxiliary winding on the ballast choke, for instance, is not accepted by most of the ballast manufacturers because of the high ratio of the number of turns to the main winding. A better solution is shown in figure 29. V dc Vce 2 Ic 2 I c 2 : 1 A/div V ce : 100 V/div T : 20 s/div Figure 30. Moment of switch-off R3 0.7 V pulse Th1 T1 R1 R2 D1 D2 R4 Re C1 Th2 T2 RStart CStart Figure 29. Example for a switch-off circuit comparable to the one shown in figure 25 If the upper power transistor is switched off as shown in figure 20, the collector current of T2 is zero. After the ballast unit has been switched off, the trace of the collector-emitter voltage may give the impression that T2 is switched on again. Nevertheless, only the parallel free-wheeling diode is switched on. The decay process takes more than 8 cycles of the switching frequency. In contrast to figure 21, it is not worthwhile mentioning power dissipation at T2 after switch-off. Under the described conditions, the maximum working voltage VCEW can be used. Under the conditions of figure 20, voltages up to only VCEO can be applied. 14 Issue: 11. 95 TELEFUNKEN Semiconductors ANT016 Circuit Examples The following examples do not intend to give the impression that they are dimensioned for production requirements. There is neither a power factor correction circuit nor a line filter included. They should be seen only as proposals in order to give the possibility for a first critical look at the application of electronic ballast units. Ballast Unit for 8 W at 120-V Mains D1 F1 L1 D2 R1 D5 L2 R4 R2 C5 R6 D7 C4 C6 Mains C1 C2 L2 L1 D3 D4 D6 C3 L2 T2 R3 R5 D8 L3 Figure 31. Circuit example 8 W / 120 V Parts list: T1 T2 D1 D2 D3 D4 D5 D6 D7 D8 R1 R2 R3 R4 R5 R6 Issue: 11. 95 BUD620 BUD620 BYT51G BYT51G BYT51G BYT51G BYT51G G-ST2 (DIAC) BYT52G BYT52G 330 k 68 68 3.3 3.3 330 k lamp 8W 15 C1 C2 C3 C4 C5 C6 F1 L1 core winding L2 winding L3 core winding 100 nF 10 F 22 nF 1 nF 100 nF 3.3 nF 250 mA 5 mH (2 x) E13/4N27 2 x 80 turns toroidal core 3/4/4 turns 0.9 mH E16/5 N27 140 turns 250 V 250 V 100 V 250 V 250 V 400 V d = 0 mm = 0.2 mm = 0.2 mm d = 1 mm = 0.3 mm Lamp T1 ANT016 Ballast Unit for 8 W at 230-V Mains D1 F1 L1 D2 R1 D5 L2 R4 R2 TELEFUNKEN Semiconductors C5 R6 D7 C4 C6 Mains C1 C2 L2 L1 D3 D4 D6 C3 L2 T2 R3 R5 D8 L3 Figure 32. Circuit example 8 W / 230 V Parts list: T1 T2 D1 D2 D3 D4 D5 D6 D7 D8 R1 R2 R3 R4 R5 R6 BUD620 BUD620 BYT51K BYT51K BYT51K BYT51K BYT51K G-ST2 (DIAC) BYT52K BYT52K 330 k 68 68 3.3 3.3 330 k L3 core winding lamp 2.8 mH E16/5 250 turns 8W N27 d = 1 mm = 0.3 mm C1 C2 C3 C4 C5 C6 F1 L1 core winding L2 winding 100 nF 4.7 F 22 nF 1 nF 100 nF 3.3 nF 250 mA 11.5 mH E13/4 2 x 120 turns toroidal core 3/4/4 turns (2 x) N27 250 V 250 V 100 V 250 V 250 V 400 V d = 0 mm = 0.2 mm = 0.3 mm 16 Issue: 11. 95 Lamp T1 TELEFUNKEN Semiconductors ANT016 Ballast Unit for 40 W at 230-V Mains CV T1 D1 R1 D5 VS , f S C1 R5 D6 D3 D4 C2 D8 R6 T2 D2 R2 D7 R4 LV CP Figure 33. Circuit example 40 W / 230 V Parts list: T1 T2 D1 D2 D3 D4 D5 D6 D7 D8 R1 R2 R3 TE13005D TE13005D BYT51K BYT51K BYT51K BYT51K BYT51K G-ST2 (DIAC) 1N4148 1N 4148 1 470 k 51 L1 core prim. wdgs sec. wdgs. lamp 1.94 mH E25/7 146 turns 2 x 3 turns 40 W N27 d = 0.5 mm = 0.5 mm = 0.5 mm R4 R5 R6 C1 C2 C3 C4 1 51 1 k 15 F 150 nF 100 nF 10 nF 350 V 250 V 100 V 250 V L40W/250 OSRAM Issue: 11. 95 Fluorescent lamp R3 17 ANT016 Ballast Unit for 100 W at 230-V Mains D1 F1 L1 D2 R1 D5 L2 R4 R2 TELEFUNKEN Semiconductors C5 R6 D7 C4 C6 Mains C1 C2 L2 L1 D3 D4 D6 C3 L2 T2 R3 R5 D8 L3 Figure 34. Circuit example 100 W / 230 V Parts list: T1 T2 D1 D2 D3 D4 D5 D6 D7 D8 R1 R2 R3 R4 R5 R6 BUF654 BUF654 BYT51K BYT51K BYT51K BYT51K BYT51K G-ST2 (DIAC) BYT52K BYT52K 330 k 15 15 0.22 0.22 330 k L3 core winding lamp 0.5 mH E42/15 43 turns 100 W N27 d = 1 mm = 0.8 mm C1 C2 C3 C4 C5 C6 F1 L1 core winding L2 winding 100 nF 47 F 22 nF 4.7 nF 220 nF 22 nF 2A 11.5 mH E25/7 2 x 81 turns toroidal core 4/4/4 turns (2 x) N27 400 V 350 V 100 V 400 V 400 V 400 V d = 0 mm = 0.45 mm = 0.65 mm UVA lamp 18 Issue: 11. 95 Lamp T1 |
Price & Availability of ANT016
 |
|
|
|
|
All Rights Reserved © IC-ON-LINE 2003 - 2022 |
| [Add Bookmark] [Contact Us] [Link exchange] [Privacy policy] |
|
Mirror Sites : [www.datasheet.hk]
[www.maxim4u.com] [www.ic-on-line.cn]
[www.ic-on-line.com] [www.ic-on-line.net]
[www.alldatasheet.com.cn]
[www.gdcy.com]
[www.gdcy.net] |