LTC1042C

LTC1042 LINER

Document overview

• Manufacturer or author: Provided By www.digicamel.com(FREE DATASHEET DOWNLOAD SITE)
• PDF pages: 8

Technical content

TECHNOLOGY Window Comparator FEATURES DESCRIPTION = Micropower 1.5nW (1 Sample/Second) The LTC1042 is a monolithic CMOS window comparator = Wide Supply Range— + 2.8V to + 16V manufactured using Linear Technology’s enhanced = High Accuracy LTCMOS™ silicon gate process. Two high impedance Center Error + 1mV Max voltage inputs, CENTER and WIDTH/2, define the middle Width Error + 0.15% Max and width of the comparison window. Whenever the input = Wide Input Voltage Range voltage, Vin, is inside the window the WITHIN WINDOW V* to Ground output is high. The ABOVE WINDOW output is high = TTL Outputs with 5V Supply whenever Vin is above the window. By interchanging Vin = Two Independent Ground-Referred Control Inputs and CENTER the ABOVE WINDOW output becomes = Small Size 8-Pin MiniDIP BELOW WINDOW andis high if Vix is below the window. Sampling techniques provide high impedance voltage in- puts that can common-mode to both supply rails (V+ and GND). An important feature of the inputs is their non- APPLICATIONS interaction. Also the device is effectively “chopper = Fault Detectors Stabilized”, giving it extremely high accuracy over all = GoJNo-Go Testing conditions of temperature, power supply and input voltage = Microprocessor Power Supply Monitor range. Another benefit of the sampling techniques used to de- sign the LTC1042 is the extremely low power consump- tion. When the device is strobed, it internally turns on the 6 | power to the comparators, samples the inputs, stores the outputs in CMOS latches and then turns off power to the comparators. This all happens in about 80us. Average power can be made small, almost arbitrarily, by lowering the strobe rate. The device can be self-strobed using an external RC network or strobed externally by driving the LTCMos™ is a trademark of Linear Technology Corp. OSC pin with a CMOS gate. SSS Battery Powered Remote Freezer Alarm Total Supply Current vs Sampling v+ Frequency = “HI" == TEMPERATURE 10000 yt _—_ n ia “W" = TEMPERATURE = arin 4 0.05 6F = | | K tTC1042 SUPPLY an ee ‘T=YELLOW SPRINGS INSTRUMENT oo ., INC. P/N 44007. 0.01 cai | ‘ALL RESISTORS + 1% UNLESS OTHERWISE SPECIFIED. 0.4 1 10 100 1000 10000 “OTHER TEMPERATURE BANDS MAY BE SELECTED BY CHOOSING APPROPRIATE VALUES FOR RY AND R2. SAMPLING FREQUENCY, fs (Hz) SSS AY Wee 6-77

Operating Temperature Range vanes a aso - 40° >| C1042 LTCHOERG oo. .seeeeeeesssssseeeeees 40°C t085°C wo] °° FElanemeor | i rotouonue LTC1042M ooo cececeesesesesesese, 55°C to 125°C owed FI worse Tre0a2CNe Nene ee eee EEE an neal ELECTRICAL CHARACTERISTICS Test Conditions: Tyin<Ta< Tmax unless otherwise specified SYMBOL PARAMETER TEST CONDITIONS MIN TYP MAX UNITS Center Error (Note 2) Vt =2.8V to 6V (Note 1) +03 +1 mV + + +0.05 20.15 % WIDTHI2 V+ =6V to 18V (Note 1) +1 +3 mV + + : +£0.05 £0.15 % WIDTHI2 Width Error (Note 3) V* =2.8V to 6V (Note 1) +06 +2 mV + + £04 £03 % WIDTHI2 V+ =6V to 15V (Note 1) +2 26 mV e + + +01 +03 % WIDTHI2 leus Input Bias Current V+ =5V, Ty = 25°C, OSC=GND nA Vin, CENTER and WIDTHI2 Inputs Rw Average Input Resistance fg = 1kHz (Note 4) je/ o 6 °° | Mo Input Voltage Range poe an v PSR Power Supply Range ee 2 v Igon) Power Supply ON Vt+=5V 1.2 3 mA Current (Note 5) Igorr) Power Supply OFF V+=SV LTC1042C 0.001 05 vA Current (Note 5) LTC1042M 0.001 5.0 vA tb Response Time (Note 6) Povey oto us Output Levels Vou Logic 1 Output V+ =4,75V, loyr= — 360A 24 44 v Vo. Logical 0 Output V+ =4,75V, lout = 1.6mMA 0.25 0.45 Vv Rext External Timing Resistor Resistor Connected between V* 100 10,000 ka and OSC Pin fg Sampling Frequency V+ =5V, T,=25°C Hz Rext= 1M0, Cer = 0.1uF The @ denotes the spectications which apply over the full operating Note 4: Rjy is guaranteed by design and is not tested. Riy = 1i(fs x 669F). temperature range. ; oo Note 5: Average supply current = Tp x lyon) X fs + (1 ~Tp fy) Igor Note 1: Applies over input voltage range limit and includes gain Note 6: Response time is set by an internal oscillator and is independent of uncertainty. ‘overdrive voltage. Tp is guaranteed by correlation test and is not directly Note 2: Center error= {(Vy + Vi/2- CENTER] (where Vy = upper band limit measured. and V; = lower band limit), Note 3: Width error = (Vy - Vi - 2x WIDTHI2) (where Vy = upper band limit and V, = lower band limit). eee

TYPICAL PERFORMANCE CHARACTERISTICS Normalized Sampling Frequency Ison) vs V* vs V*, Temperature Sampling Rate vs Rext, Cext 20 22 eee % > 20 SS Sas Saati exLt Tat A esL LA TT et og. Stes. Saul

444 RIERA CE 3 SiS

Ae lH 8 MNS ' =a a Za rd a — cei ess Teeter LL ee | Ceo SS 2 4 6 8 10 12 4 16 0 2 4 6 8 0 12 «14 «16 100k 1M 10M ‘SUPPLY VOLTAGE, V+ (V) ‘SUPPLY VOLTAGE, V+ (V) Rexr (9) Response Time vs Supply Voltage Response Time vs Temperature _ Rin vs Sampling Frequency 10 130 G10) See

100 Laer | TL "0 e+ 4 s FRR

Pi itty iy 10 a a Soo SEU TTT TTT Zz — = 2 3 yoo LURE

3 PEt ety gro tt tt | g ero r

wo titTitti| 3 KEE d EEE Ci a9 2 ; ‘ surety vole. V+ w “oe ° 2 MBIENT TEMPERATURE, mee * ' SAMPLING FREQUENCY, ‘cine ° oo APPLICATIONS INFORMATION The LTC1042 uses sampled data techniques to achieve its | when (CENTER-WIDTH/2) < Vin < (CENTER + WIDTH/2) unique characteristics. It consists of two comparators, both comparator outputs are low, In this condition Viy is each of which has two differential inputs (Figure 1). When within the window and the WITHIN WINDOW output is the sum of the voltages on a comparator’s inputs is posi: high. When Viy>CENTER + WIDTH/2, Vin is above the win- tive, the output is high; when the sum is negative, the out- dow and the ABOVE WINDOW output is high. put is low. The inputs are interconnected such that LT We 6-79

\\TC1042 a APPLICATIONS INFORMATION An important feature of the LTC1042is the non-interaction At low sampling rates, Rext dominates the power con- of the inputs. This means the center and width of the win- sumption. Rext consumes power continuously. The aver- dow can be changed without one affecting the other. Also age voltage at the OSC pin is approximately V*/2. The note that the width of the window is set by a ground re- power consumed by Reyris: ferred signal (WIDTH/2). P(Rexq)=(V* 2)2Rexr Strobing EXAMPLE: Assume Rext = 1MQ and V+ =5V. Then: An internal oscillator allows the LTC1042 to strobe itself. P(Rext) = (2.5)2/1M0= 6.25,W The frequency of oscillation sets the sampling rate and is Pan ‘ f set with an external RC network (see typical curve, OSC This is more than ten times the typical power consumed fi age by the LTC1042 at V+ =5V and 1 sample/second. Where requency vs Rext, Cext). To assure oscillation, under all + A conditions, Rexr must be between 100kQ and 10MO. There power is a premium, Rext should be made as large as isno limit to the size of C " possible. Note that the power dissipated by Rexr is not a EXT. function of the sampling frequency or Ceyr. A sampling cycle is initiated on the positive going transi- jf yj H i : : y a, igh sampling rates are needed and power consumption tion or tne nage on ne O86 pin. Winen this voltage is ig of secondary importance, a convenient way to get the fate Pos! Mn ae Cc nl tigger trips and ini- maximum possible sampling rate is to make Rexr = 100k0 tiates the sampling cycle. A sampling cycle consists of ap- ang Ceyr=0. The sampling rate, set by the LTC1042's ac- plying power to both comparators, sampling the inputs, tive time, will nominally be = 10kHz storing the results in CMOS output latches and turning , . power off. This whole process takes approximately 80us. To synchronize the sampling of the LTC1042 to an external During the 80us “active” time, the LTC1042 draws typically frequency source, the OSC pin can be driven by a CMOS 1.2mA (Is(on)) at V* =5V. Because power is consumed gate. A CMOS gate is necessary because the input trip only during the “active” time, extremely low average points of the oscillator are close to the supply rails and power consumption can be achieved at low sample rates. | TTL does not have enough output swing. Externally driven, For example at a sample rate of 1 sample/second the aver- _ there will be a delay from the rising edge of the OSC input age power consumption is: and the start of the sampling cycle of approximately 5ys. Power = (V*) (Isav)) = 5V x 1.2mA x 80ys/1sec. =0.48.W WINDOW ° WITHIN WINDOW: Ps y owwoott name | i Vtp--- ; a CENTER) rH COMP B [Ee] ABOVE WINDOW & 1 a * (BELOW WINDOW) & i) wiota/2 [5] = ~wora/2 | wioTH/2 Go [4] ° j i ow ee —~——POWER ON M Vu osc | POWER OFF INPUT VOLTAGE, Vin, Leo “) Figure 1. LTC1042 Block Diagram (8) SSS, 6-80 AT Wear

classic linear comparator. CMOS switches anda precision —_que to voltage divider between Rg and Rwy. the equivalent circuit shown in Figure 2. The input capaci- Veron = Vin[2Cin/(2Cjy + Cs) + Re/(Rg + Rin)]. For Rg<10ka from input impedance effects is reduced to 136,V. quently, the input voltage range includes both supply rails. For source resistances greater than 10k@, Cjy cannot fully Thisis a further benefit of the input sampling structure. shared between Cjy and Cs, causing a voltage error. The . another error term. Each time the input switches cycle be- ‘rors may be expressed in terms of Vy and V,. an average input resistance of Rin = Vin/lave = 1ifs x Cin). time and temperature. Figure 2. Equivalent Input Circuit

+ v Vv a= “ HI" = SUPPLY IN ice oa os 100k (45<V+<55) asueu let toss iT1004-2.5 ea 75) on 10k ALL RESISTORS 25% UNLESS OTHERWISE NOTED. “SUPPLY TOLERANCE EQUALS R2 IN kQ. 1.E., 10k= + 10%. Single 5V Thermocouple Over Temperature Alarm H COLD JUNCTION COMPENSATOR 1 w ! RS | i] I sk @2sect 1 \\ ‘00 G) r-t4-5 1 TEMPERATURE | Berove12 ! wero F a a 1N wNDOW Q of] z wie) ae 8 ! 1870 rezoe St 1 | 6 Bie a | to wey ool pemmor dA a) 1 = I ! 1 R2 _— 7 | ! ! | Fo ERMOCOUPL H ' . ~ om 1.235 (82-43) 21.235 (R24 0.0047 F 2x 1.235 2¢R3 H Tr aaa ce I ‘YELLOW SPRINGS INST. CO. P/N 44007 I “CHOOSE C5 TO FILTER NOISE bo ee ee ee ed **CHOOSE Re. Ry, AY, R2 AND R3 TO SET WINDOW 1k ALL RESISTORS 4: 1% UNLESS OTHERWISE NOTED My a

Wind Powered Battery Charger A simple wind powered battery charger can be con- 2) If the generator voltage output is between 13.8V and structed using the new LTC1042, a 12V DC permanent 15.1V, the 12V lead acid battery is being charged at magnet motor, and low cost power FET transistor. about a 1 amp/hour rate (limited by the power FET). The DC motor is used as a generator with the voltage out- 3) If generator voltage exceeds 15.1V (a condition caused put being proportional to its RPM. The LTC1042 monitors by excessive wind speed or 12V battery being fully the voltage output and provides the following control charged) then a fixed load is connected thus limiting functions. the generator RPM to prevent damage. 1) If generator voltage output is below 13.8V, the control This charger can be used as a remote source of power circuit is active and the NiCad battery is charging — where wind energy is plentiful such as on sailboats or re- through the LM334 current source. The lead acid bat- — mote radio repeater sites, Unlike solar powered panels, tery is not being charged. this system will function in bad weather and at night. — GENERATOR = | _ 1N4001 10k: O.1pF OQ a = I so L334 fo TT _ WITHIN: ‘WINDOW 13.8V TO 15.1V MTPBNOS Ly 16] OVER VOLTAGE fey, MTPANOS 40k S 171004-1.2" | oF ki P| (15.1) SS LIT WER 6-83

PACKAGE DESCRIPTION bimensions in inches (millimeters) unless otherwise noted.

8 Lead Hermetic DIP

9.005 oa) fr 0.405, ni (10.287) | MAX 0.025 RAD TYP t 0.220-0.310 (5.588 7.874) 0.290 -0.320 | (7.37 =8.13) CTR oa 0.055, (5.080) ir87) mx Max 0.015 -0.060 (0.381 - 1.524) 0.008 - 0.018 yo45¢ (977920. a {0:360—0.660) h a MIN ee T, 6

8 Lead Plastic

0.400 (10.160) MAX [a] Cs} [5] —F D) 0.250 + 0.010 . (6.350 +0.254) WRT GI 0.300 0.320 0.130 + 0.005 1.045 — | ——ee 0.065 Eo (1.651) TYP _—_ MIN 0.009 -0.015 0.125 0.925 +9.025 MIN = 0.015 0.045 + 0.015 ced 0.100:+ 0.010 0.018 20.003, esta 110°C 150°C ea