《LTC6655应用指南.docx》由会员分享，可在线阅读，更多相关《LTC6655应用指南.docx（14页珍藏版）》请在三一文库上搜索。
1、Application Note 124July 2009775 Nanovolt Noise Measurement for A Low Noise Voltage ReferenceQuantifying SilenceJim WilliamsIntroductionFrequently, voltage reference stability and noise define measurement limits in instrumentation systems. In par-ticular, reference noise often sets stable resolution
2、 limits. Reference voltages have decreased with the continuing drop in system power supply voltages, making reference noise increasingly important. The compressed signal processing range mandates a commensurate reduction in reference noise to maintain resolution. Noise ultimately translates into qua
3、ntization uncertainty in A to D converters, introducing jitter in applications such as scales, inertial navigation systems, infrared thermography, DVMs and medical imaging apparatus. A new low voltage reference, the LTC6655, has only 0.3ppm (775nV) noise at 2.5VOUT. Figure 1 lists salient specificat
4、ions in tabular form. Ac-curacy and temperature coefficient are characteristic of high grade, low voltage references. 0.1Hz to 10Hz noise, particularly noteworthy, is unequalled by any low voltage electronic reference.Noise MeasurementSpecial techniques are required to verify the LTC6655s ex-tremely
5、 low noise. Figure 2s approach appears innocently straightforward but practical implementation represents a high order difficulty measurement. This 0.1Hz to 10Hz noiseLTC6655 Reference Tabular Specificationstesting scheme includes a low noise pre-amplifier, filters and a peak-to-peak noise detector.
6、 The pre-amplifiers 160nV noise floor, enabling accurate measurement, requires special design and layout techniques. A forward gain of 106 permits readout by conventional instruments.Figure 3s detailed schematic reveals some considerations required to achieve the 160nV noise floor. The references DC
7、 potential is stripped by the 1300F, 1.2k resistor combination; AC content is fed to Q1. Q1-Q2, extraordi-narily low noise J-FETs, are DC stabilized by A1, with A2 providing a single-ended output. Resistive feedback from A2 stabilizes the configuration at a gain of 10,000. A2s output is routed to am
8、plifier-filter A3-A4 which provides 0.1Hz to 10Hz response at a gain of 100. A5-A8 comprise a peak-to-peak noise detector read out by a DVM at a scale factor of 1 volt/microvolt. The peak-to-peak noise detector provides high accuracy measurement, eliminating tedious interpretation of an oscilloscope
9、 display. Instanta-neous noise value is supplied by the indicated output to a monitoring oscilloscope. The 74C221 one-shot, triggered by the oscilloscope sweep gate, resets the peak-to-peak noise detector at the end of each oscilloscope 10-second sweep.L, LT, LTC, LTM, Linear Technology and the Line
10、ar logo are registered trademarks of Linear Technology Corporation. All other trademarks are the property of their respective owners.SPECIFICATIONLIMITSOutput Voltages1.250, 2.048, 2.500, 3.000, 3.300, 4.096, 5.000Initial Accuracy0.025%, 0.05%Temperature Coefficient2ppm/C, 5ppm/C0.1Hz to 10Hz Noise0
11、.775V at VOUT = 2.500V, Peak-to-Peak Noise is within this Figure in 90% of 1000 Ten Second Measurement IntervalsAdditional Characteristics5ppm/Volt Line Regulation, 500mV Dropout, Shutdown Pin, ISUPPLY = 5mA, VIN = VO + 0.5V to 13.2VMAX,IOUT(SINK/SOURCE) = 5mA, ISHORT Circuit = 15mA.Figure 1. LTC665
12、5 Accuracy and Temperature Coefficient Are Characteristic of High Grade, Low Voltage References. 0.1Hz to 10Hz Noise, Particularly Noteworthy, Is Unequalled by Any Low Voltage Electronic Referencean124fAN124-1Application Note 124A = 106LOW NOISE0.1Hz TO 10Hz FILTER ANDLTC6655AC PRE-AMPPEAK TO PEAK N
13、OISE DETECTOR2.5V REFERENCEEN, 0.1Hz TO 10Hz = 160nV0V TO 1V = 0V TO 1V, A = 100A = 10,000OUTPUTRESETDC OUT0V TO 1V = 0VP-P TO1VP-P AT INPUT700nVOSCILLOSCOPENOISE0.1Hz TO 10HzSWEEPGATE OUTVERTICALINPUTAN124 F02Figure 2. Conceptual 0.1Hz to 10Hz Noise Testing Scheme Includes Low Noise Pre-Amplifier,
14、Filter and Peak to Peak Noise Detector. Pre-Amplifiers 160nV Noise Floor, Enabling Accurate Measurement, Requires Special Design and Layout TechniquesNumerous details contribute to the circuits performance. The 1300F capacitor, a highly specialized type, is selected for leakage in accordance with th
15、e procedure given in Appendix B. Further, it, and its associated low noise 1.2k resistor, are fully shielded against pick-up. FETs Q1 and Q2 differentially feed A2, forming a simple low noise op amp. Feedback, provided by the 100k - 10 pair, sets closed loop gain at 10,000. Although Q1 and Q2 have e
16、xtraordinarily low noise characteristics, their offset and drift are uncontrolled. A1 corrects these deficiencies by adjusting Q1s channel current via Q3 to minimize the Q1-Q2 input difference. Q1s skewed drain values ensure that A1 is able to capture the offset. A1 and Q3 supply whatever current is
17、 required into Q1s channel to force offset within about 30V. The FETs VGS can vary over a 4:1 range. Because of this, they must be selected for 10% VGS matching. This matching allows A1 to capture the offset without introducing significant noise. Q1 and Q2 are thermally mated and lagged in epoxy at
18、a time constant much greater than A1s DC stabilizing loop roll-off, preventing offset instability and hunting. The entire A1-Q1-Q2-A2 assembly and the reference under test are completely enclosed within a shielded can.1 The reference is powered by a 9V battery to minimize noise and insure freedom fr
19、om ground loops.Peak-to-peak detector design considerations include J-FETs used as peak trapping diodes to obtain lower leakage thanafforded by conventional diodes. Diodes at the FET gates clamp reverse voltage, further minimizing leakage.2 The peak storage capacitors highly asymmetric charge-discha
20、rge profile necessitates the low dielectric absorption polypro-pelene capacitors specified.3 Oscilloscope connections via galvanically isolated links prevent ground loop induced corruption. The oscilloscope input signal is supplied by an isolated probe; the sweep gate output is interfaced with an is
21、olation pulse transformer. Details appear in Appendix C.Noise Measurement Circuit PerformanceCircuit performance must be characterized prior to mea-suring LTC6655 noise. The pre-amplifier stage is verified for 10Hz bandwidth by applying a 1V step at its input (reference disconnected) and monitoring
22、A2s output. Figure 4s 10ms risetime indicates 35Hz response, insuring the entire 0.1Hz to 10Hz noise spectrum is supplied to the succeeding filter stage.Note 1. The pre-amplifier structure must be carefully prepared. See Appendix A, “Mechanical and Layout Considerations”, for detail on pre-amplifier
23、 construction.Note 2. Diode connected J-FETs superior leakage derives from their extremely small area gate-channel junction. In general, J-FETs leak a few picoamperes (25C) while common signal diodes (e.g. 1N4148) are about 1,000X worse (units of nanoamperes at 25C).Note 3. Teflon and polystyrene di
24、electrics are even better but the Real World intrudes. Teflon is expensive and excessively large at 1F. Analog types mourn the imminent passing of the polystyrene era as the sole manufacturer of polystyrene film has ceased production.an124fAN124-23-AN124an124f1N46970.1F15V0.15F10V1F+10kA = 100 ANDA1
25、0.1Hz TO 10Hz FILTERLT1012Q30.1F15V2kA = 1042N2907+A415V15V1k*200*1F124k*124k*LT1012LOW NOISEA3+PRE-AMPLT10120.1F1F450*900*1M*330F +9V100k100kA216V1300F TLT109710k*SHIELD+330F516V+SQ1 Q20.022FINLTC6655+100k*330*100*2.5VF*1.2kSDINOUT10k INPUT+1F750*10*330FROOT-SUM-SQUAREREFERENCE16VCORRECTION15V330FS
26、EE TEXTUNDER TESTAC LINE GROUNDSHIELDED CAN16V+RESET PULSE0.22F10kGENERATORPEAK TO PEAK4.7k+15NOISE DETECTOR+15C2RC2RST = Q2BAT-85+VB2RST+ PEAK74C221A5A71FPBAT-851/4 LT10581/4 LT1058+1kCLR2A210k+1510k+1515+TO OSCILLOSCOPE INPUTFROM OSCILLOSCOPE0.005F0.1FDVMO TO 1V =VIA ISOLATED PROBE,100k1V/DIV = 1V
27、/DIV,SWEEP GATE OUTPUTO TO 1VREFERRED TO INPUT,VIA ISOLATIONSWEEP = 1s/DIVPULSE TRANSFORMER4.7k PEAK+1k* = 1% METAL FILMQ1, Q2 = THERMALLY MATEDA81/4 LT1058* = 1% WIREWOUND, ULTRONIX105A2SK369 (MATCH VGS 10%)T = TANTALUM,WET SLUGOR LSK389 DUALRST= 1N4148THERMALLY LAGILEAK 5nA1FP100kA6SEE TEXTSEE TEX
28、T/APPENDIX B1/4 LT1058P= POLYPROPELENE+= 2N4393A4 330F OUTPUT CAPACITORS = 200nA LEAKAGE10k= 1/4 LTC202AT 1VDC AT 25C15SEE APPENDIX C FOR POWER, SHIELDING0.005FAND GROUNDING SCHEMEAN124 F03Figure 3. Detailed Noise Test Circuitry. Thermally Lagged Q1-Q2 Low Noise J-FET Pair Is DC Stabilized by A1-Q3;
29、 A2 Delivers A = 10,000 Pre-Amplifier Output. A3-A4 form 0.1Hz to 10Hz ,A = 100, Bandpass Filter; Total Gain Referred to Pre-Amplifier Input Is 106. Peak to Peak Noise Detector, Reset by Monitoring Oscilloscope Sweep Gate, Supplies DVM OutputApplication Note 124Application Note 124Figure 5 describes
30、 peak-to-peak noise detector operation. Waveforms include A3s input noise signal (Trace A), A7 (Trace B) positive/A8 (Trace C) negative peak detector outputs and DVM differential input (Trace D). Trace Es oscilloscope supplied reset pulse has been lengthened for photographic clarity.Circuit noise fl
31、oor is measured by replacing the LTC6655 with a 3V battery stack. Dielectric absorption effects in the large input capacitor require a 24-hour settling period before measurement. Figure 6, taken at the circuits oscil-100nV/DIV1s/DIVAN124 F06loscope output, shows 160nV 0.1Hz to 10Hz noise in a 10 sec
32、ond sample window. Because noise adds in root-sum-square fashion, this represents about a 2% error inFigure 6. Low Noise Circuit/Layout Techniques Yield 160nV 0.1Hz to 10Hz Noise Floor, Ensuring Accurate Measurement. Photograph Taken at Figure 3s Oscilloscope Output with 3V Battery Replacing LTC6655
33、 Reference. Noise Floor Adds 2% Error to Expected LTC6655 Noise Figure Due to Root-Sum-Square Noise Addition Characteristic; Correction is Implemented at Figure 3s A32mV/DIV10ms/DIVAN124 F04the LTC 6655s expected 775nV noise figure. This term is accounted for by placing Figure 3s “root-sum-square co
34、r-rection” switch in the appropriate position during reference testing. The resultant 2% gain attenuation first order cor-rects LTC6655 output noise reading for the circuits 160nV noise floor contribution. Figure 7, a strip-chart recording of the peak-to-peak noise detector output over 6 minutes, sh
35、ows less than 160nV test circuit noise.4 Resets occurFigure 4. Pre-Amplifier Rise Time Measures 10ms; Indicated 35Hz Bandwidth Ensures Entire 0.1Hz to 10Hz Noise Spectrum Is Supplied to Succeeding Filter Stageevery 10 seconds. A 3V battery biases the input capacitor, replacing the LTC6655 for this test.A = 5mV/DIVB = 0.5V/DIVC = 0.5V/DIVD = 1V/DIVE = 20V/DIV1s/DIVAN124 F05Figure 8 is LTC6655 noise after the indicated 24-hour dielectric absorption soak time. Noise is within 775nV peak-to-peak in this 10 second sample window with the root-sum-square correction