LMP7732 2.9 nV/sqrt(Hz) Low Noise, Precision, RRIO Amplifier General Description Features The LMP7732 is a dual low noise, low offset voltage, rail-torail input and output, low voltage precision amplifier. The LMP7732 is part of the LMP® precision amplifier family and is ideal for precision and low noise applications with low voltage requirements. This operational amplifier offers low voltage noise of 2.9 nV/ with a 1/f corner of only 3 Hz and low DC offset with a maximum value of ±40 µV, targeting high accuracy, low frequency applications. The LMP7732 has bipolar junction input stages with a bias current of only 1.5 nA. This low input bias current, complemented by the very low AC and DC levels of voltage noise, makes the LMP7732 an excellent choice for photometry applications. The LMP7732 provides a wide GBW of 22 MHz while consuming only 4 mA of current. This high gain bandwidth along with the high open loop gain of 130 dB enables accurate signal conditioning in applications with high closed loop gain requirements. The LMP7732 has a supply voltage range of 1.8V to 5.5V, making it an ideal choice for battery operated portable applications. The LMP7732 is offered in the 8-Pin SOIC and MSOP packages. The LMP7731 is the single version of this product and is offered in the 5-Pin SOT-23 and 8-Pin SOIC packages. (Typical values, TA = 25°C, VS = 5V) ■ Input voltage noise — f = 3 Hz — f = 1 kHz ■ Offset voltage (max) ■ Offset voltage drift (max) ■ CMRR ■ Open loop gain ■ GBW ■ Slew rate ■ THD @ f = 10 kHz, AV = 1, RL = 2 kΩ ■ Supply current ■ Supply voltage range ■ Operating temperature range ■ Input bias current ■ RRIO 3.3 nV/√Hz 2.9 nV/√Hz ±40 µV ±1.3 µV/°C 130 dB 130 dB 22 MHz 2.4 V/µs 0.001% 4.4 mA 1.8V to 5.5V −40°C to 125°C ±1.5 nA Applications ■ ■ ■ ■ Thermopile amplifier Gas analysis instruments Photometric instrumentation Medical instrumentation Typical Application Thermopile Signal Amplifier 30015001 LMP® is a registered trademark of National Semiconductor Corporation. © 2008 National Semiconductor Corporation 300150 www.national.com LMP7732 Low Noise, Precision, RRIO Amplifier June 16, 2008 LMP7732 Storage Temperature Range Junction Temperature (Note 3) Soldering Information Infrared or Convection (20 sec) Wave Soldering Lead Temp. (10 sec) Absolute Maximum Ratings (Note 1) If Military/Aerospace specified devices are required, please contact the National Semiconductor Sales Office/ Distributors for availability and specifications. ESD Tolerance (Note 2) Human Body Model For inputs pins only For all other pins Machine Model Charge Device Model VIN Differential Supply Voltage (VS = V+ – V−) Operating Ratings 2000V 2000V 200V 1000V ±2V 6.0V 2.5V Electrical Characteristics −65°C to 150°C +150°C max 235°C 260°C (Note 1) Temperature Range Supply Voltage (VS = V+ – V–) −40°C to 125°C 1.8V to 5.5V Package Thermal Resistance (θJA) 8-Pin SOIC 8-Pin MSOP 190 °C/W 235°C/W (Note 4) Unless otherwise specified, all limits are guaranteed for TA = 25°C, V+ = 2.5V, V− = 0V, VCM = V+/2, RL >10 kΩ to V+/2. Boldface limits apply at the temperature extremes. Symbol VOS TCVOS IB IOS Parameter Input Offset Voltage (Note 7) Conditions Min Typ Max (Note 6) (Note 5) (Note 6) VCM = 2.0V ±9 ±50 ±150 VCM = 0.5V ±9 ±40 ±125 Input Offset Voltage Temperature Drift VCM = 2.0V ±0.5 ±1.3 VCM = 0.5V ±0.2 ±0.8 Input Offset Voltage Time Drift VCM = 0.5V and VCM = 2.0V 0.35 Input Bias Current VCM = 2.0V ±1 ±30 ±45 VCM = 0.5V ±12 ±50 ±75 VCM = 2.0V ±1 ±50 ±75 VCM = 0.5V ±11 ±60 ±80 Input Offset Current TCIOS Input Offset Current Drift VCM = 0.5V and VCM = 2.0V CMRR Common Mode Rejection Ratio 0.15V ≤ VCM ≤ 0.7V 101 89 120 1.5V ≤ VCM ≤ 2.35V 105 99 129 2.5V ≤ V+ ≤ 5V 111 105 129 0.23V ≤ VCM ≤ 0.7V 1.5V ≤ VCM ≤ 2.27V PSRR Power Supply Rejection Ratio CMVR Common Mode Voltage Range AVOL Open Loop Voltage Gain www.national.com Large Signal CMRR ≥ 80 dB nA nA nA/°C dB dB 0 2.5 112 104 130 RL = 2 kΩ to V+/2 VOUT = 0.5V to 2.0V 109 90 119 2 μV/°C 117 RL = 10 kΩ to VOUT = 0.5V to 2.0V V+/2 μV μV/month 0.0474 1.8V ≤ V+ ≤ 5.5V Units V dB VOUT Parameter Output Voltage Swing High Output Voltage Swing Low IOUT IS Output Current Supply Current Conditions Min Typ Max (Note 6) (Note 5) (Note 6) RL = 10 kΩ to V+/2 4 50 75 RL = 2 kΩ to V+/2 13 50 75 RL = 10 kΩ to V+/2 6 50 75 RL = 2 kΩ to V+/2 9 50 75 Sourcing, VOUT = V+/2 VIN (diff) = 100 mV 22 12 31 Sinking, VOUT = V+/2 VIN (diff) = −100 mV 15 10 44 Units mV from either rail mA VCM = 2.0V 4.0 5.4 6.8 VCM = 0.5V 4.6 6.2 7.8 mA SR Slew Rate AV = +1, CL = 10 pF, RL = 10 kΩ to V+/2 VOUT = 2 VPP 2.4 V/μs GBW Gain Bandwidth CL = 20 pF, RL = 10 kΩ to V+/2 21 MHz GM Gain Margin CL = 20 pF, RL = 10 kΩ to V+/2 14 dB ΦM Phase Margin CL = 20 pF, RL = 10 kΩ to V+/2 60 deg RIN Input Resistance Differential Mode 38 kΩ Common Mode 151 MΩ 0.002 % THD+N Total Harmonic Distortion + Noise AV = 1, fO = 1 kHz, Amplitude = 1V en Input Referred Voltage Noise Density f = 1 kHz, VCM = 2.0V 3.0 f = 1 kHz, VCM = 0.5V 3.0 Input Voltage Noise 0.1 Hz to 10 Hz 75 Input Referred Current Noise Density f = 1 kHz, VCM = 2.0V 1.1 f = 1 kHz, VCM = 0.5V 2.3 in 3.3V Electrical Characteristics nV/ nVPP pA/ (Note 4) Unless otherwise specified, all limits are guaranteed for TA = 25°C, V+ = 3.3V, V− = 0V, VCM = V+/2, RL > 10 kΩ to V+/2. Boldface limits apply at the temperature extremes. Symbol VOS TCVOS IB IOS Parameter Input Offset Voltage (Note 7) Conditions Min Typ Max (Note 6) (Note 5) (Note 6) VCM = 2.5V ±6 ±50 ±150 VCM = 0.5V ±6 ±40 ±125 Input Offset Voltage Temperature Drift VCM = 2.5V ±0.5 ±1.3 VCM = 0.5V ±0.2 ±0.8 Input Offset Voltage Time Drift VCM = 0.5V and VCM = 2.5V 0.35 Input Bias Current VCM = 2.5V ±1.5 ±30 ±45 VCM = 0.5V ±13 ±50 ±77 VCM = 2.5V ±1 ±50 ±70 VCM = 0.5V ±11 ±60 ±80 Input Offset Current 3 Units μV μV/°C μV/month nA nA www.national.com LMP7732 Symbol LMP7732 Symbol Parameter Conditions Min Typ Max (Note 6) (Note 5) (Note 6) Units 0.048 nA/°C TCIOS Input Offset Current Drift VCM = 0.5V and VCM = 2.5V CMRR Common Mode Rejection Ratio 0.15V ≤ VCM ≤ 0.7V 101 89 120 1.5V ≤ VCM ≤ 3.15V 105 99 130 2.5V ≤ V+ ≤ 5.0V 111 105 129 0.23V ≤ VCM ≤ 0.7V 1.5V ≤ VCM ≤ 3.07V PSRR Power Supply Rejection Ratio 1.8V ≤ V+ ≤ 5.5V CMVR Common Mode Voltage Range AVOL Open Loop Voltage Gain VOUT Output Voltage Swing High Output Voltage Swing Low IS Output Current Supply Current dB 117 Large Signal CMRR ≥ 80 dB 0 3.3 RL = 10 kΩ to VOUT = 0.5V to 2.8V 112 104 130 RL = 2 kΩ to V+/2 VOUT = 0.5V to 2.8V 110 92 119 V+/2 5 50 75 RL = 2 kΩ to V+/2 14 50 75 9 50 75 13 50 75 RL = 10 kΩ to V+/2 Sourcing, VOUT = V+/2 VIN (diff) = 100 mV 28 22 45 Sinking, VOUT = V+/2 VIN (diff) = −100 mV 25 20 48 V dB RL = 10 kΩ to V+/2 RL = 2 kΩ to V+/2 IOUT dB mV from either rail mA VCM = 2.5V 4.2 5.6 7.0 VCM = 0.5V 4.8 6.4 8.0 mA SR Slew Rate AV = +1, CL = 10 pF, RL = 10 kΩ to V+/2 VOUT = 2 VPP 2.4 GBW Gain Bandwidth CL = 20 pF, RL = 10 kΩ to V+/2 22 MHz GM Gain Margin CL = 20 pF, RL = 10 kΩ to V+/2 14 dB ΦM Phase Margin CL = 20 pF, RL = 10 kΩ to THD+N Total Harmonic Distortion + Noise AV = 1, fO = 1 kHz, Amplitude = 1V RIN Input Resistance en in Input Referred Voltage Noise Density 62 deg 0.002 % Differential Mode 38 kΩ Common Mode 151 MΩ f = 1 kHz, VCM = 2.5V 2.9 f = 1 kHz, VCM = 0.5V 2.9 V+/2 Input Voltage Noise 0.1 Hz to 10 Hz 75 Input Referred Current Noise Density f = 1 kHz, VCM = 2.5V 1.1 f = 1 kHz, VCM = 0.5V 2.1 www.national.com V/μs 4 nV/ nVPP pA/ (Note 4) Unless otherwise specified, all limits are guaranteed for TA = 25°C, V+ = 5V, V− = 0V, VCM = V+/2, RL > 10 kΩ to V+/2. Boldface limits apply at the temperature extremes. Symbol VOS TCVOS IB IOS Parameter Input Offset Voltage (Note 7) Conditions VCM = 4.5V ±6 ±50 ±150 VCM = 0.5V ±6 ±40 ±125 Input Offset Voltage Temperature Drift VCM = 4.5V ±0.5 ±1.3 VCM = 0.5V ±0.2 ±0.8 Input Offset Voltage Time Drift VCM = 0.5V and VCM = 4.5V 0.35 Input Bias Current VCM = 4.5V ±1.5 ±30 ±50 VCM = 0.5V ±14 ±50 ±85 VCM = 4.5V ±1 ±50 ±70 VCM = 0.5V ±11 ±65 ±80 Input Offset Current TCIOS Input Offset Current Drift VCM = 0.5V and VCM = 4.5V CMRR Common Mode Rejection Ratio 0.15V ≤ VCM ≤ 0.7V 120 1.5V ≤ VCM ≤ 4.85V 105 99 130 2.5V ≤ V+ ≤ 5V 111 105 129 1.5V ≤ VCM ≤ 4.77V Power Supply Rejection Ratio 1.8V ≤ V+ ≤ 5.5V CMVR Common Mode Voltage Range AVOL Open Loop Voltage Gain VOUT Output Voltage Swing High Output Voltage Swing Low IOUT IS Output Current Supply Current Large Signal CMRR ≥ 80 dB μV μV/°C nA nA nA/°C dB dB 117 0 5 RL = 10 kΩ to VOUT = 0.5V to 4.5V 112 104 130 RL = 2 kΩ to V+/2 VOUT = 0.5V to 4.5V 110 94 119 V+/2 Units μV/month 0.0482 101 89 0.23V ≤ VCM ≤ 0.7V PSRR Min Typ Max (Note 6) (Note 5) (Note 6) dB RL = 10 kΩ to V+/2 8 50 75 RL = 2 kΩ to V+/2 24 50 75 RL = 10 kΩ to V+/2 9 50 75 RL = 2 kΩ to V+/2 23 50 75 Sourcing, VOUT = V+/2 VIN (diff) = 100 mV 33 27 47 Sinking, VOUT = V+/2 VIN (diff) = −100 mV 30 25 49 V mV from either rail mA VCM = 4.5V 4.4 6.0 7.4 VCM = 0.5V 5.0 6.8 8.4 mA SR Slew Rate AV = +1, CL = 10 pF, RL = 10 kΩ to V+/2 VOUT = 2 VPP 2.4 V/μs GBW Gain Bandwidth CL = 20 pF, RL = 10 kΩ to V+/2 22 MHz 5 www.national.com LMP7732 5V Electrical Characteristics LMP7732 Symbol Parameter Conditions GM Gain Margin ΦM Phase Margin CL = 20 pF, RL = 10 kΩ to RIN Input Resistance Differential Mode Min Typ Max (Note 6) (Note 5) (Note 6) CL = 20 pF, RL = 10 kΩ to V+/2 12 dB V+/2 65 deg 38 kΩ 151 MΩ 0.001 % Common Mode THD+ N Total Harmonic Distortion + Noise AV = 1, fO = 1 kHz, Amplitude = 1V en f = 1 kHz, VCM = 4.5V 2.9 f = 1 kHz, VCM = 0.5V 2.9 in Input Referred Voltage Noise Density Units Input Voltage Noise 0.1 Hz to 10 Hz 75 Input Referred Current Noise Density f = 1 kHz, VCM = 4.5V 1.1 f = 1 kHz, VCM = 0.5V 2.2 nV/ nVPP pA/ Note 1: Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. Operating Ratings indicate conditions for which the device is intended to be functional, but specific performance is not guaranteed. For guaranteed specifications and the test conditions, see the Electrical Characteristics Tables. Note 2: Human Body Model, applicable std. MIL-STD-883, Method 3015.7. Machine Model, applicable std. JESD22-A115-A (ESD MM std. of JEDEC) FieldInduced Charge-Device Model, applicable std. JESD22-C101-C (ESD FICDM std. of JEDEC). Note 3: The maximum power dissipation is a function of TJ(MAX), θJA, and TA. The maximum allowable power dissipation at any ambient temperature is PD = (TJ(MAX) - TA)/ θJA. All numbers apply for packages soldered directly onto a PC board. Note 4: Electrical Table values apply only for factory testing conditions at the temperature indicated. Factory testing conditions result in very limited self-heating of the device such that TJ = TA. No guarantee of parametric performance is indicated in the electrical tables under conditions of internal self-heating where TJ > TA. Absolute maximum Ratings indicate junction temperature limits beyond which the device maybe permanently degraded, either mechanically or electrically. Note 5: Typical values represent the most likely parametric norm as determined at the time of characterization. Actual typical values may vary over time and will also depend on the application and configuration. The typical values are not tested and are not guaranteed on shipped production material. Note 6: All limits are guaranteed by testing, statistical analysis or design. Note 7: Ambient production test is performed at 25°C with a variance of ±3°C. Connection Diagram 8-Pin SOIC/MSOP 30015003 Top View Ordering Information Package 8-Pin SOIC Part Number LMP7732MA LMP7732MAX Package Marking 95 units/Rails LMP7732MA 2.5k Units Tape and Reel LMP7732MM 8-Pin MSOP LMP7732MME NSC Drawing M08A 1k Units Tape and Reel AZ3A 250 Units Tape and Reel LMP7732MMX www.national.com Transport Media 3.5k Units Tape and Reel 6 MUA08A Unless otherwise noted: TA = 25°C, RL > 10 kΩ, VCM = VS/2. Offset Voltage Distribution TCVOS Distribution 30015076 30015071 Offset Voltage Distribution TCVOS Distribution 30015074 30015073 Offset Voltage Distribution TCVOS Distribution 30015077 30015070 7 www.national.com LMP7732 Typical Performance Characteristics LMP7732 Offset Voltage Distribution TCVOS Distribution 30015075 30015072 Offset Voltage vs. Temperature Offset Voltage vs. Temperature 30015082 30015083 PSRR vs. Frequency CMRR vs. Frequency 30015062 30015029 www.national.com 8 LMP7732 Offset Voltage vs. Supply Voltage Offset Voltage vs. VCM 30015053 30015054 Offset Voltage vs. VCM Offset Voltage vs. VCM 30015056 30015055 Input Offset Voltage Time Drift Slew Rate vs. Supply Voltage 30015080 30015020 9 www.national.com LMP7732 Input Voltage Noise vs. Frequency Input Current Noise vs. Frequency 30015063 30015064 Time Domain Voltage Noise Time Domain Voltage Noise 30015065 30015067 Time Domain Voltage Noise Output Voltage vs. Output Current 30015066 30015057 www.national.com 10 LMP7732 Input Bias Current vs. VCM Input Bias Current vs. VCM 30015026 30015025 Input Bias Current vs. VCM Open Loop Frequency Response Over Temperature 30015018 30015027 Open Loop Frequency Response Open Loop Frequency Response 30015028 30015019 11 www.national.com LMP7732 THD+N vs. Frequency THD+N vs. Output Voltage 30015085 30015069 Large Signal Step Response Small Signal Step Response 30015022 30015021 Large Signal Step Response Small Signal Step Response 30015024 www.national.com 30015023 12 Output Swing High vs. Supply Voltage 30015081 30015058 Output Swing Low vs. Supply Voltage Sinking Current vs, Supply Voltage 30015059 30015060 Sourcing Current vs. Supply Voltage 30015061 13 www.national.com LMP7732 Supply Current vs. Supply Voltage LMP7732 one of the input stages as the circuitry is symmetrical for both stages. Figure 1 shows that as the common mode voltage gets closer to one of the extreme ends, current I1 significantly increases. This increased current shows as an increase in voltage drop across resistor R1 equal to I1*R1 on IN+ of the amplifier. This voltage contributes to the offset voltage of the amplifier. When common mode voltage is in the mid-range, the transistors are operating in the linear region and I1 is significantly small. The voltage drop due to I1 across R1 can be ignored as it is orders of magnitude smaller than the amplifier's input offset voltage. As the common mode voltage gets closer to one of the rails, the offset voltage generated due to I1 increases and becomes comparable to the amplifiers offset voltage. Application Notes LMP7732 The LMP7732 is a dual low noise, low offset voltage, rail-torail input and output, low voltage precision amplifier. with a 1/f corThe low input voltage noise of only 2.9 nV/ ner at 3 Hz makes the LMP7732 ideal for sensor applications where DC accuracy is of importance. The LMP7732 has very low guaranteed offset voltage of only ±40 µV. This low offset voltage along with the very low input voltage noise allows higher signal integrity and higher signal to noise ratios since the error contribution by the amplifier is at a minimum. The LMP7732 has high gain bandwidth of 22 MHz. This wide bandwidth enables the use of the amplifier at higher gain settings while retaining ample usable bandwidth for the application. This is particularly beneficial when system designers need to use sensors with very limited output voltage range as it allows larger gains in one stage which in turn increases signal to noise ratio. The LMP7732 has a proprietary input bias cancellation circuitry on the input stages. This allows the LMP7732 to have only about 1.5 nA bias current with a bipolar input stage. This low input bias current, paired with the inherent lower input voltage noise of bipolar input stages makes the LMP7732 an excellent choice for precision applications. The combination of low input bias current, low input offset voltage, and low input voltage noise enables the user to achieve unprecedented accuracy and higher signal integrity. National Semiconductor is heavily committed to precision amplifiers and the market segment they serve. Technical support and extensive characterization data is available for sensitive applications or applications with a constrained error budget. The LMP7732 comes in the 8-Pin SOIC and MSOP packages. These small packages are ideal solutions for area constrained PC boards and portable electronics. 30015006 FIGURE 1. Input Bias Current Cancellation INPUT VOLTAGE NOISE MEASUREMENT The LMP7732 has very low input voltage noise. The peak-topeak input voltage noise of the LMP7732 can be measured using the test circuit shown in Figure 2 INPUT BIAS CURRENT CANCELLATION The LMP7732 has proprietary input bias current cancellation circuitry on its input stage. The LMP7732 has rail-to-rail input. This is achieved by having a p-input and n-input stage in parallel. Figure 1 only shows 30015079 FIGURE 2. 0.1 Hz to 10 Hz Noise Test Circuit www.national.com 14 DIODES BETWEEN THE INPUTS The LMP7732 has a set of anti-parallel diodes between their input pins, as shown in Figure 5. These diodes are present to protect the input stage of the amplifiers. At the same time, they limit the amount of differential input voltage that is allowed on the input pins. A differential signal larger than the voltage needed to turn on the diodes might cause damage to the diodes. The differential voltage between the input pins should be limited to ±3 diode drops or the input current needs to be limited to ±20 mA. 30015066 FIGURE 3. 0.1 Hz to 10 Hz Input Voltage Noise Measuring the very low peak-to-peak noise performance of the LMP7732, requires special testing attention. In order to achieve accurate results, the device should be warmed up for at least five minutes. This is so that the input offset voltage of the op amp settles to a value. During this warm up period, the offset can typically change by a few µV because the chip temperature increases by about 30°C. If the 10 seconds of the measurement is selected to include this warm up time, some of this temperature change might show up as the measured noise.Figure 4 shows the start-up drift of five typical LMP7732 units. 30015004 FIGURE 5. Anti-Parallel Diodes between Inputs 30015080 FIGURE 4. Start-Up Input Offset Voltage Drift 15 www.national.com LMP7732 During the peak-to-peak noise measurement, the LMP7732 must be shielded. This prevents offset variations due to airflow. Offset can vary by a few nV due to this airflow and that can invalidate measurements of input voltage noise with a magnitude which is in the same range. For similar reasons, sudden motions must also be restricted in the vicinity of the test area. The feed-through which results from this motion could increase the observed noise value which in turn would invalidate the measurement. The frequency response of this noise test circuit at the 0.1 Hz corner is defined by only one zero. The test time for the 0.1 Hz to 10 Hz noise measurement using this configuration should not exceed 10 seconds, as this time limit acts as an additional zero to reduce or eliminate the contributions of noise from frequencies below 0.1 Hz. Figure 3 shows typical peak-to-peak noise for the LMP7732 measured with the circuit in Figure 2. LMP7732 THERMOPILE AMPLIFIER DRIVING AN ADC Analog to Digital Converters, ADCs, usually have a sampling capacitor on their input. When the ADC's input is directly connected to the output of the amplifier a charging current flows from the amplifier to the ADC. This charging current causes a momentary glitch that can take some time to settle. There are different ways to minimize this effect. One way is to slow down the sampling rate. This method gives the amplifier sufficient time to stabilize its output. Another way to minimize the glitch, caused by the switch capacitor, is to have an external capacitor connected to the input of the ADC. This capacitor is chosen so that its value is much larger than the internal switching capacitor and it will hence provide the charge needed to quickly and smoothly charge the ADC's sampling capacitor. Since this large capacitor will be loading the output of the amplifier as well, an isolation resistor is needed between the output of the amplifier and this capacitor. The isolation resistor, RISO, separates the additional load capacitance from the output of the amplifier and will also form a low-pass filter and can be designed to provide noise reduction as well as anti-aliasing. The draw back of having RISO is that it reduces signal swing since there is some voltage drop across it. Figure 6 (a) shows the ADC directly connected to the amplifier. To minimize the glitch in this setting, a slower sample rate needs to be used. Figure 6 (b) shows RISO and an external capacitor used to minimize the glitch. Thermopile Sensors Thermopiles are arrays of interconnected thermocouples which can detect surface temperature of an object through radiation rather than direct contact. The hot and cold junctions of the thermocouples are thermally isolated. The hot junctions are exposed to IR radiation emitted from the measurement surface and the cold junctions are connected to a heat sink. The incident IR changes the temperature of the hot junctions of the thermopile and produces an output voltage proportional to this change. The hot junction of the thermopile is covered with a highly emissive coating. The IR radiation incident to this highly emissive material changes the temperature of this coating. The temperature change is converted to a voltage by the thermopile. Emissivity represents the radiation or absorption efficiency of a material relative to a black body. An ideal black body has an emissivity of 1.0. Excluding shiny metals, most objects have emissivities above 0.85. As a practical matter, shiny metals are not good candidates for IR sensing because of their low emissivity. The low emissivity means that the material is highly reflective. Reflective materials often “reflect” the surrounding environment's temperature rather than their own heat radiation. This makes them not suitable for thermopile applications. The output voltage of a thermopile is related to temperature and emissivity by the following formula: Where: VOUT : Output voltage of the thermopile K : Proportionality constant εOBJ: Emissivity of object being measured TOBJ: Temperature of object being measured δ : Correction factor. This is needed since thermopile filters do not allow all wavelengths to enter the sensor εTP: Emissivity of the thermopile TTP: Temperature of the thermopile As mentioned above, the IR radiation generates a static voltage across the pyroelectric material. If the illumination is constant, the signal level detection declines. This is why the radiation needs to be periodically refreshed. This task is usually achieved by the means of a mechanical chopper in front of the detector. Thermopiles offer much faster response time compared to other temperature measurement devices. Packaged thermistors and thermocouples have response times that can range up to a few seconds, where as packaged thermopiles can easily achieve response times in the order of tens of milliseconds. Thermopiles also provide superior thermal isolation compared to their contact temperature measurement counterparts. Physical contact disturbs the systems temperature and also creates temperature gradients. Figure 7 shows a simplified schematic of a thermopile. The cold junctions are connected to a heat sink, and the absorber material covers the hot junction. The output voltage resulting from temperature difference between the two junctions is measured at the two ends of the array of thermocouples. As is evident in Figure 7, increasing the number of thermocouples in a thermopile increases the output voltage range. This also increases the active area of the thermopile sensor. 30015005 FIGURE 6. Driving An ADC www.national.com 16 And As it is shown above, one cannot just compare the NEP of two thermopiles without considering the corresponding active areas. A better way to compare thermopiles is to look at their specific detectivity, D*. Specific detectivity includes both the device noise and its sensitivity. It is normalized with respect to the detector's active area and also noise bandwidth. D* is given by: 30015007 FIGURE 7. Thermopile Thermopiles have very wide temperature ranges of -100°C to 1000°C When choosing a thermopile for a certain application, one must pay attention to several parameters. Some of these parameters are discussed below: Thermopiles' sensitivity, or responsivity, is determined by the ratio of output voltage to the absorbed input signal power and is usually specified in V/W. Typical sensitivity of thermopiles ranges from 10s of V/W to about 100 V/W. Generally, higher values of sensitivity are desirable. Sensitivity is dependent on the absorber's area and number of thermocouples used in the sensor. Sensitivity is often represented by S where: S = VOUT/PIN The sensitivity of a thermopile changes with change in temperature. This change is usually specified as the Temperature Coefficient, TC, of sensitivity. Lower numbers are desired for this parameter. Resistance of the thermopile is usually specified in the datasheet. This is the impedance which will be seen by the input of the amplifier used to process the thermopile's output signal. Typical values for thermopile resistance, RTP, range from 10s of kilo-ohms to about 100 kΩ. This resistance is also a function of temperature. The temperature coefficient of the resistance is usually specified in a thermopile's datasheet. As with any other parameter, minimum variation with temperature is desired. The dominant noise source for a thermopile is its resistance. Noise spectral density of a resistor is calculated by: Unit of D* is cm / W. Typical values for specific detectivity range from 108 to 3*108 cm / W. After receiving radiation, the thermopile takes some time before it comes to thermal equilibrium. The time it takes for the sensor to achieve this equilibrium is called response time or time constant of the sensor. Clearly, lower time constants are very desirable. Precision Amplifier Since the output of thermopiles are usually very small and at most in the order of only a few millivolts, the first part of the signal conditioning path should involve amplification. In choosing an amplifier for this purpose, a few different sensor characteristics and the way they interface with the amplifier should be considered. These are: Sensor's Impedance and Opamp's Input Bias Current The input bias current causes a voltage drop across the sensor and the amount of this voltage is equal to the sensor's impedance multiplied by the magnitude of bias current. The higher the sensor's input impedance, the more accentuated the effect of amplifier's input bias current will be. For very high impedance sensors, it is imperative that opamps with very low input bias currents be used. Thermopiles have input impedances in the range of 100 kΩ, so input bias current is not as critical as in some other applications. Sensor's output voltage range: The output signal of the sensor is fed into the opamp where it will be amplified or otherwise conditioned, (e.g. level shifted, buffered.) It is important to pay attention to different parameters of this output signal. One important aspect is the lowest expected level of the sensor's output and compare that with different parameters contributing to the amplifier's total input noise. If the sensor's output level is in the same order of magnitude or smaller than the opamp's total input noise, then signal integrity at the opamp's output and the ADC's input will be compromised. Where k is the Boltzman constant and T is absolute temperature. Unit of noise spectral density is: V/ For the thermopile sensor, this noise is usually represented by VNOISE where: Typical values for this voltage noise are in the order of a few tens of nV/ . The Noise Equivalent Power, NEP, is often used to specify the minimum detectable signal level per square root band- 17 www.national.com LMP7732 width. A smaller NEP is desired, however NEP is dependent on the thermopile active area, AD. For a thermopile: LMP7732 30015078 FIGURE 8. Thermopile Amplifier Figure 8 shows the LMP7732 used as a thermopile amplifier. The LMP7732 is a great choice for use with thermopile sensors. The LMP7732 provides unprecedented accuracy and precision because of its very low input voltage noise and the very low 1/f corner frequency. The 1/f noise is one of the main sources of error in DC operating mode. Since thermopiles and most other sensors operate on DC signals, signal integrity at the DC level is very important. The LMP7732 also has very low offset voltage and offset voltage drift which greatly reduces the effects of input offset voltage of the amplifier on the thermopile signal. The thermopile used in this circuit is TPS332 from PerkinElmer Optoelectronics, PKI. This thermopile has an internal resistance, RTP, of 75 kΩ. The output voltage of the thermopile is represented with a DC voltage source. The TPS332 has a thermistor integrated in the package. The thermistor is used to measure the ambient temperature of the thermopile at the time of measurement. The thermistor's resistance at room temperature is 30 kΩ. More information about this thermopile and other sensors from PKI can be found on http://www.perkinelmer.com/ The circuit in Figure 8 shows how the LMP7732 is connected to the thermopile. This circuit is comprised of two LMP7732 amplifiers, the LM4140A-2.5 which is a precision voltage reference, the ADC122S021 which is a two channel Analog to Digital converter, and the thermopile sensor. Note that the two amplifiers used in this circuit are numbered for ease of refer- www.national.com ence. The LMP7732 amplifiers are referred to as amplifier 1 and amplifier 2 per Figure 8. In Figure 8the LM4140A is providing a precision voltage reference of 2.5V. This reference voltage is applied to the thermistor via the 30 kΩ resistor. The thermistor's resistance is converted to a voltage using this set up. This voltage is fed into the ADC's channel one. The ADC uses this voltage and the thermistor's look up table to convert this voltage to temperature. The 2.5V reference is also fed into amplifier 1, which is configured as a buffer. This LMP7732 transfers the 2.5V signal to both inputs of amplifier 2. This means the 2.5V will show up on the output of amplifier 2. Having an output level that is mid-supply is important since the thermopile sensor has a bipolar output signal and this way the amplifier can accurately gain the thermopile voltage, whether its polarity is positive or negative. It is also important because the output signal of amplifier 2 is only positive. ADCs can only handle positive signals on their inputs. Amplifier 2 is used to gain and filter the thermopile signal. The low pass filter ensures that AC noise will not be gained up and, as a result, the output signal will be cleaner. The output of amplifier 2 is fed into the ADC's channel 0. The ADC uses the ambient temperature, which was calculated using the voltage on Channel 1 and the thermistor's look up table, along with the thermopiles' gained output voltage available on channel 0 and the thermopile's look up table to determine the object's temperature. 18 LMP7732 Physical Dimensions inches (millimeters) unless otherwise noted 8-Pin SOIC NS Package Number M08A 8-Pin MSOP NS Package Number MUA08A 19 www.national.com LMP7732 Low Noise, Precision, RRIO Amplifier Notes For more National Semiconductor product information and proven design tools, visit the following Web sites at: Products Design Support Amplifiers www.national.com/amplifiers WEBENCH www.national.com/webench Audio www.national.com/audio Analog University www.national.com/AU Clock Conditioners www.national.com/timing App Notes www.national.com/appnotes Data Converters www.national.com/adc Distributors www.national.com/contacts Displays www.national.com/displays Green Compliance www.national.com/quality/green Ethernet www.national.com/ethernet Packaging www.national.com/packaging Interface www.national.com/interface Quality and Reliability www.national.com/quality LVDS www.national.com/lvds Reference Designs www.national.com/refdesigns Power Management www.national.com/power Feedback www.national.com/feedback Switching Regulators www.national.com/switchers LDOs www.national.com/ldo LED Lighting www.national.com/led PowerWise www.national.com/powerwise Serial Digital Interface (SDI) www.national.com/sdi Temperature Sensors www.national.com/tempsensors Wireless (PLL/VCO) www.national.com/wireless THE CONTENTS OF THIS DOCUMENT ARE PROVIDED IN CONNECTION WITH NATIONAL SEMICONDUCTOR CORPORATION (“NATIONAL”) PRODUCTS. NATIONAL MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE ACCURACY OR COMPLETENESS OF THE CONTENTS OF THIS PUBLICATION AND RESERVES THE RIGHT TO MAKE CHANGES TO SPECIFICATIONS AND PRODUCT DESCRIPTIONS AT ANY TIME WITHOUT NOTICE. NO LICENSE, WHETHER EXPRESS, IMPLIED, ARISING BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. TESTING AND OTHER QUALITY CONTROLS ARE USED TO THE EXTENT NATIONAL DEEMS NECESSARY TO SUPPORT NATIONAL’S PRODUCT WARRANTY. EXCEPT WHERE MANDATED BY GOVERNMENT REQUIREMENTS, TESTING OF ALL PARAMETERS OF EACH PRODUCT IS NOT NECESSARILY PERFORMED. NATIONAL ASSUMES NO LIABILITY FOR APPLICATIONS ASSISTANCE OR BUYER PRODUCT DESIGN. BUYERS ARE RESPONSIBLE FOR THEIR PRODUCTS AND APPLICATIONS USING NATIONAL COMPONENTS. PRIOR TO USING OR DISTRIBUTING ANY PRODUCTS THAT INCLUDE NATIONAL COMPONENTS, BUYERS SHOULD PROVIDE ADEQUATE DESIGN, TESTING AND OPERATING SAFEGUARDS. EXCEPT AS PROVIDED IN NATIONAL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, NATIONAL ASSUMES NO LIABILITY WHATSOEVER, AND NATIONAL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY RELATING TO THE SALE AND/OR USE OF NATIONAL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. LIFE SUPPORT POLICY NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS PRIOR WRITTEN APPROVAL OF THE CHIEF EXECUTIVE OFFICER AND GENERAL COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein: Life support devices or systems are devices which (a) are intended for surgical implant into the body, or (b) support or sustain life and whose failure to perform when properly used in accordance with instructions for use provided in the labeling can be reasonably expected to result in a significant injury to the user. A critical component is any component in a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system or to affect its safety or effectiveness. National Semiconductor and the National Semiconductor logo are registered trademarks of National Semiconductor Corporation. All other brand or product names may be trademarks or registered trademarks of their respective holders. Copyright© 2008 National Semiconductor Corporation For the most current product information visit us at www.national.com National Semiconductor Americas Technical Support Center Email: [email protected] Tel: 1-800-272-9959 www.national.com National Semiconductor Europe Technical Support Center Email: [email protected] German Tel: +49 (0) 180 5010 771 English Tel: +44 (0) 870 850 4288 National Semiconductor Asia Pacific Technical Support Center Email: [email protected] National Semiconductor Japan Technical Support Center Email: [email protected]