DESCRIPTION The A78L00 series of 3-Terminal Positive Voltage Regulators employ internal current limiting and thermal shutdown, making them essentially indestructible. lf adequate heat sinking is provided, they can deliver up to 100mA output current. They are intended as fixed voltage regulators in a wide range of applications including local or on card reg- ulation for elimination of noise and distri- bution problems associated with single point regulation. In addition, they can be used with power pass elements to make high current voltage regulators. The pwA78LO0O0 used as a Zener diode/resistor combination replacement, offers an effec- tive output impedance improvement of typi- cally two orders of magnitude, along with lower quiescent current and lower noise. EQUIVALENT CIRCUIT FEATURES Output curent up to 100mA No external components Internal thermal overload protection e Internal short circuit current limiting Available in JEDEC TO-92 and low pro- file TO-39 packages Output voltages of 2.6V, 5V, 6.2V, 8.2V, 12V and 15V Output voltage tolerances or +5% (78L00-A) and +10% (78L00) over the temperature range , oO ae R13 20k 3 os aQ10 R14 4k R10 (2) OUTPUT Kors se ns 500 TO 7k | 1k TO 14k {oe Q11 Ra aa q RB 3 R7 3 4k $ 1.4k ot (3) <) COMMON ABSOLUTE MAXIMUM RATINGS PARAMETER RATING UNIT Input voltage 2.6V, 5V, 6.2V and 8.2V 30 Vv 12V and 15V 35 Vv Internal power dissipation Internally limited Storage temperature range Metal can (TO-39 type) -65 to +150 C Moided TO-92 -55 to +150 C Operating junction temperature range 0 to +150 C Lead temperatures Metal can (soldering, 60s time limit) 300 C Molded TO-92 (soldering, 10s time limit) 260 C wA78L02/5/6/8/12/15-DB,S CONNECTION DIAGRAMS / ) INPUT @) OUTPUT VOLTAGE 15V 15V DB PACKAGE (TO-39) (1) ouTeUT Q) ORDER INFORMATION COMMON PART NO. 78L02COB 78LO2ACDB 78LOSCDB 78LO5ACDB 78LO6CDB 78LO6ACDB 78LO8CDB 78LO8ACDB 78L12CDB 78L12ACDB 78L15CDB 78L15ACDB COMMON OUTPUT OUTPUT VOLTAGE S PACKAGE (JEDEC TO-92) ORDER INFORMATION INPUT PART NO. 78L02CS 78LO2ACS 78LO5CS 78LOSACS 78L06CS 78LO6ACS 78LO8CS 78LO8ACS 78L12CS 78L12ACS 7BL15CS 78L15ACS 160DC ELECTRICAL CHARACTERISTICS uA78L02/5/6/8/12/15-DB,S lout = 40mA, 0C < Ty < +125C, Cin = 0.33 uF, Cout = 0.1uF unless otherwise specified. SiN0tES 78LO2AC: 78LO2C PARAMETER TEST CONDITIONS UNIT Min | Typ | Max Min | Typ Max Vin = 9.0V Vin = 9.0V Vout Output voltage Ty = 25C 25 | 26 [| 27 24 | 26 | 28 Vv 4.75V = Vin <=: 20V 4.75V <= Vin = 20V 1mA < lout = 70mMA 2.45 2.75 2.35 2.85 Vv 1mA < lout S 40mMA 2.45 2.75 2.35 2.85 Vv 4.75 = Vin = 20V 4.75V < Vin S 20V Line regulation Ty = 25C | 40 | 100 | 40 | 125 | mv 5V < Vin = 20V 5V <= Vin = 20V 30 75 30 100 mV Load regulation 1mA <= lout <= 100mMA, Ty = 25C 10 50 10 50 mV imA < lout = 40mA, Ty = 25C 40 25 4.0 25 mw loc Ty = 25C 3.6 6.0 3.6 6.0 mA Ty = 125C 5.5 6.5 mA 5V < Vin = 20V 5V <= Vin = 20V Alcc With line 2.5 2.5 mA With load, 1mA = lout = 40mA 0.1 0.2 mA Output noise voltage Ty = 25C, 10Hz < f < 100kHz 30 30 BV Long term stability 10 10 mV 6V < VIN = 16V 6V < Vin <= 16V Ripple rejection Ty = 25C, f = 120Hz 43 51 42 51 dB Dropout voltage Ta = 25C 1.7 1.7 Vv Vout Output temperature drift lout = 5mA -0.4 -0.4 mvV/C Isc Ty = 25C 140 140 mA DC ELECTRICAL CHARACTERISTICS (Cont'd) 78LO5AG 78L05C PARAMETER TEST CONDITIONS - UNIT Min Typ | Max Min Typ | Max Vin = 10V Vin = 10V Vout Output voltage Ty = 25C 48 | 50 | 52 | 46 | 50 | 54 Vv 7V < Vin S 20V 7V <= Vin = 20V 1mA <= lout < 70mMA 4.75 5.25 4.50 5.50 Vv mA < lout = 40mA 4.75 5.25 | 4.50 5.50 Vv 7V < VIN = 20V 7V <= Vin = 20V Line regulation Ty = 26C | 55 | 150 {| 55 | 200 | mv 8V < Vin = 20V 8V < Vin = 20V 45 100 45 150 mV Load regulation tmA <= lout = 100mMA, Ty = 25C 1 60 11 60 mV 1mA < lout Ss 40mA, Ty = 25C 5.0 30 5.0 30 mw lec Ty = 25C 3.8 6.0 3.8 6.0 mA Ty = 125C 5.5 5.5 mA 8V < Vin <= 20V 8V < Vin = 20V Alcc With line 1.5 1.5 mA With load, 1mA < lout = 40mA 0.1 0.2 mA Output noise voltage Ty = 25C, 10Hz < f < 100kHz 40 40 BV Long term stability 12 12 mV 8V <= Vin S 18V 8V <= Vin S 18V Ripple rejection Ty = 25C, f = 120Hz 41 49 40 49 dB Dropout voltage Ta = 26C 1.7 17 Vv Vout Output temperature drift lout = 5mA -0.65 ~0.65 mvV/C Isc Ty = 26C 140 140 mA 161DC ELECTRICAL CHARACTERISTICS uA781L.02/5/6/8/12/15-DB,S (Cont'd) lout = 40mA, 0C <= Ty <= +128C, Cin = 0.33yF, Cout = 0.14F unless otherwise specified. 78LO6AC 78LO6C PARAMETER TEST CONDITIONS ~ UNIT Min [ Typ [ Max Min Typ [ Max T T T Vin = 12V Vin = 12V Vout Output voltage Ty = 25C 595| 62 | 6.45 57 | 62 | 67 Vv 8.5V < VIN = 20V 8.5V < Vin = 20V imA = lout = 70mA 5.90 6.50 5.60 6.80 Vv 1ImA = lout < 40mA 5.90 6.50 5.6 6.8 V 8.5V <= Vin = 20V 8.5V < Vin <= 20V Line regulation Ty = 25C | 65 | 175 | 65 | 200 | mv 9V = VIN = 20V QV < Vin = 20V 55 125 55 150 mV Load regulation 1mA = lout = 100mA, Ty = 28C 13 80 13 80 mV 4mA <= lout = 40mA, Ty = 25C 6.0 40 6.0 40 mV lec Ty = 25C 3.9 6.0 3.9 6.0 mV Ty = 125C 5.5 5.5 mA 9.0V < VIN = 20V 9.0V <= Vin = 20V Alcc With line 1.5 4.5 mA With load, ImA -OVout = Line regulation of this circuit is typically 110mvV for an input range of 25-35V at a constant load current, i.e. 11mV/V. Load regulation = constant V1 load regulation (typically 10mV, 10-30mA I.) + (1ImV/V) X 0.24 X 20mA (typically 53mV) = 63mV for a load current change of 20mA ataconstant Vin of 30V. THERMAL CONSIDERATIONS The TO-92 molded package is capable of unusually high power dissipation due to the lead frame design. However, its thermal capabilities are generally overlooked be- cause of a lack of understanding of the thermal paths from the semiconductor junc- tion to ambient temperature. While thermal resistance is normally specified for the de- vice mounted 1cm above an infinite heat sink, very little has been mentioned of the options available to improve on the conser- vatively rated thermal! capability. An explanation of the thermal paths of the TO-92 and comparison of the thermal equivalent circuit of the TO-39 metal pack- age with that of the TO-92 will allow the designer to determine the thermal stress he is applying in any give application. THE METAL CAN THERMAL MODEL In the TO-39 case, where the die is attached directly to the base of a metal package, the thermal equivalent circuit is often repre- sented simply as a series connection of the juntion-to-case thermal resistance, 6c, and the case-to-ambient thermal! resistance, 8ca, as shown in Figure 1. In this model, the current source represents the thermal energy source; TJ is the junc- tion temperature, assuming a constant sur- face temperature across the die; 6c is the junction-to-case thermal resistance, meas- ured at a point on the case directly beneath the die location; 6ca is the thermal resist- ance from the case to the ultimate heat sink, ambient temperature, as represented by the battery. The heat flow is analogous to elec- trical current, and temperature to voltage. The total thermal resistance from junction to ambient is then: Osa = BC + OCA The maximum power dissipation is a func- tion of the maximum permissible junction temperature (which is a function of the package materials and construction) and the total thermal! resistance from the junc- tion to ambient temperature. Junction temperature is assumed to the limiting fac- tor. Binnoties Where: A78L02/5/6/8/12/15-DB.S Thus: maximum power dissipation Ty(Max) ~ TA uc + OCA Since BJA = uc + OCA Ta(Max) TA then VA = Po or Qua Pp=Ty -Ta Ty - Ta Pp = BJA Therefore, using the Vee method of junction temperature sensing, and attaching a ther- mocouple to the case at the location speci- fied, the relative values of 0yc, and ca can readily be determined. The thermal ratings of the metal can pack- age are normally presented with the case attached to an infinite heat sink at still air ambient temperature. This causes 6ca to go to zero resulting in @sc representing the total 6ya. The infinite heat sink is an unreal- izable condition in the practical world, but serves to project a goal. THE TO-92 PACKAGE The TO-92 package thermal paths are con- siderably more compiex than those of the TO-39 metal can package. In addition to the path through the molding compound to ambient temperature, there is another path through the leads in parallel with the case path, to ambient temperature, as shown in Figure 2. The total thermal resistance in this model is then: (uc + @ca) (Ou + OLA) OA = 6yc + Oca + OIL OLA thermal resistance of the case between the regula- tor die and a point on the case directly above the die location. thermal resistance be- tween the case and air at ambient temperature. thermal resistance from transistor die through the collector lead to a point 1/16 below the regulator case. total thermal resistance of the collector-base- emiiter leads to ambient temperature. duc = 6CA = OL = OLA = 165As one can see from Figure 1, the metal can package generally does not have the lead cooling path because of the high thermal resistances resulting from the construction of the header, case and leads. Normally, this material is kovar. Now, @uc and 6. are within the package and not variable by the user. However, 6ca and @La are outside the package and can be effectively used to control the total thermal resistance and, therefore, junction temperature. Replacing @ja of equation (1) with @ya equa- tion (3) gives: (yc + Oca} (Au @ OLA) = Tu - TA duc + Oca + BL + OLA Po The maximum Ty allowed in equation (4) is 150C. The maximum power dissipation is determined by the net total thermal resist- ance @ya, the parallel equivalent networks of the case series path and lead series path, divided into the difference of the maximum junction temperature, 150C, and ambient temperature generally specified as 25C. In the case of the 78LXX, the maximum dissi- pation of a .4 inch condition is: Pp = 150-25 oi, = 180C/W OJA Pp = 0.7W If lead length is reduced to .125 inch 6a becomes 160C, and Pp(MAX) = 0.78W. METHODS OF HEAT SINKING With two external thermal resistances in each leg of a parallel network available to the circuit designer as variabies, he can choose the method of heat sinking most applicable to his particular situation. To demonstrate, consider the effect of placing a small 72C/W flag type heat sink, such as the Staver F1-7D-2, on the 78LXX moided case. The heat sink effectively replaces the 6ca (Figure 2) and the new thermal resist- 166 ance, 9a, is: 6a = 145C/W (assuming .125 inch lead length) THERMAL EQUIVALENT CIRCUIT TO-39 PACKAGE (DIE ATTACHED TO METAL PACKAGE BASE) Ty uc an Te 8 { Pg (WATTS) 9a Ta ij Figure 1 TO-92 THERMAL EQUIVALENT CIRCUIT Ty 3 6yc 2 9a | fre (WATTS) 2 4ca 2 La < Ta ann {+ Figure 2 The net change of 15C/W increases the allowable power dissipation to 0.86W with an inserted cost of 1-2 cents. A still further decrease in 6a could be achieved by using a sink rated at 46C/W, such as the Staver FS-7A. Also, if the case sinking does not provide an adequate reduction in total 6Ja, the other external thermal resistance, @La, Si}NGticS pA78L02/5/6/8/12/15-DB.S may be reduced by shortening the lead length from package base to mounting me- dium. However, one point must be kept in mind. The tead thermal path includes a thermal resistance, @sa, from the leads at the mounting point to ambient, that is, the mounting medium, 6a is then equal to Ls + 63a. The new model is shown in Figure 3. TO-92 THERMAL EQUIVALENT CIRCUIT (LEAD AT OTHER THAN AMBIENT TEMPERATURE) Ty ; Oat Suc 9s | P_ (WATTS) ICA sa Ta i Figure 3 In the case of a socket, @sa could be as high as 270C/W, thus causing a net increase in @s4 and a consequent decrease in the maximum dissipation capability. Shorten- ing the lead length may return the net 6a to the original value, but lead sinking would riot be accomplished. In those cases where the regulator is insert- ed into acopper clad printed circuit board, it is advantageous to have a maximum area of copper at the entry points of the leads. While it would be desirable to rigorously define the effect of PC board copper, the real world variables are too great to allow anything more than a few general observations. The best analogy for PC board copper is to compare it with parallel resistors. Beyond some point, additonal resistors are not sig- nificantly effective; beyond some point, ad- ditional copper area is not effective.