NSC LMP7731MFX

LMP7731
2.9 nV/sqrt(Hz) Low Noise, Precision, RRIO, Operational
Amplifier in SOT23-5
General Description
Features
The LMP7731 is a single, low noise, low offset voltage, railto-rail input and output, low voltage precision amplifier. The
LMP7731 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 LMP7731 has bipolar 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 LMP7731 an excellent choice for photometry applications.
The LMP7731 provides a wide GBW of 22 MHz while consuming only 2 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 LMP7731 has a supply voltage range of 1.8V to 5.5V,
making it an ideal choice for battery operated portable applications.
The LMP7731 is offered in a 5-pin SOT23 package.
(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 per channel
■ Supply voltage range
■ Operating temperature range
■ Input bias current
■ RRIO
■ 5-Pin SOT23 package
3.3 nV/√Hz
2.9 nV/√Hz
±40 µV
±1.0 µV/°C
130 dB
130 dB
22 MHz
2.4 V/µs
0.001%
2.2 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
20175201
LMP® is a registered trademark of National Semiconductor Corporation.
© 2007 National Semiconductor Corporation
201752
www.national.com
LMP7731, Low Noise, Precision, RRIO, Amplifier in SOT23–5
July 2007
LMP7731
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
Inputs pins only
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)
5-Pin SOT23
265°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
Typ
(Note 5)
Max
(Note 6)
VCM = 2.0V
±9
±50
±120
VCM = 0.5V
±9
±40
±100
VCM = 2.0V
±0.5
±1.0
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 Voltage
(Note 7)
Input Offset Voltage Drift
Input Offset Current
Conditions
TCIOS
Input Offset Current Drift
VCM = 0.5V and VCM = 2.0V
CMRR
Common Mode Rejection Ratio
0.15V ≤ VCM ≤ 0.7V
120
1.5V ≤ VCM ≤ 2.35V
105
99
129
2.5V ≤ V+ ≤ 5V
111
105
129
1.5V ≤ VCM ≤ 2.27V
Power Supply Rejection Ratio
1.8V ≤ V+ ≤ 5.5V
μV
μV/°C
nA
nA
nA/°C
dB
dB
117
CMVR
Input Common-Mode Voltage
Range
Large Signal CMRR ≥ 80 dB
AVOL
Large Signal Voltage Gain
RL = 10 kΩ to V+/2
VO = 0.5V to 2.0V
112
104
130
RL = 2 kΩ to V+/2
VO = 0.5V to 2.0V
109
90
119
www.national.com
Units
μV/month
0.0474
101
89
0.23V ≤ VCM ≤ 0.7V
PSRR
Min
(Note 6)
2
0
2.5
V
dB
VO
Parameter
Output Swing High
Output Swing Low
IO
Output Short Circuit Current
IS
Supply Current
(Per Channel)
Conditions
Min
(Note 6)
Typ
(Note 5)
Max
(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, VO = V+/2
VIN (diff) = 100 mV
22
12
31
Sinking, VO = V+/2
VIN (diff) = −100 mV
15
10
44
Units
mV
from either
rail
mA
VCM = 2.0V
2.0
2.7
3.4
VCM = 0.5V
2.3
3.1
3.9
mA
SR
Slew Rate
AV = +1, CL = 10 pF, RL = 10 kΩ to V+/2,
VO = 2 VPP
2.4
V/μs
GBW
Gain Bandwidth Product
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
AV = 1, f = 1 kHz, Amplitude = 1V
en
Input-Referred Voltage Noise
f = 1 kHz, VCM = 2.0V
3
f = 1 kHz, VCM = 0.5V
3
0.1 Hz to 10 Hz
in
Input-Referred Current Noise
nV/
75
f = 1 kHz, VCM = 2.0V
1.1
f = 1 kHz, VCM = 0.5V
2.3
3.3V Electrical Characteristics
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
Typ
(Note 5)
Max
(Note 6)
VCM = 2.5V
±6
±50
±120
VCM = 0.5V
±6
±40
±100
VCM = 2.5V
±0.5
±1.0
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 Voltage
(Note 7)
Input Offset Voltage Drift
Input Offset Current
Conditions
3
Min
(Note 6)
Units
μV
μV/°C
μV/month
nA
nA
www.national.com
LMP7731
Symbol
LMP7731
Symbol
Parameter
Conditions
Min
(Note 6)
TCIOS
Input Offset Current Drift
VCM = 0.5V and VCM = 2.5V
CMRR
Common Mode Rejection Ratio
0.15V ≤ VCM ≤ 0.7V
0.23V ≤ VCM ≤ 0.7V
1.5V ≤ VCM ≤ 3.15V
1.5V ≤ VCM ≤ 3.07V
PSRR
Power Supply Rejection Ratio
2.5V ≤ V+ ≤ 5.0V
Typ
(Note 5)
0.048
101
89
120
105
99
130
111
105
129
1.8V ≤ V+ ≤ 5.5V
Large Signal CMRR ≥ 80 dB
AVOL
Large Signal Voltage Gain
RL = 10 kΩ to V+/2
VO = 0.5V to 2.8V
112
104
130
RL = 2 kΩ to V+/2
VO = 0.5V to 2.8V
110
92
119
Output Swing Low
IO
IS
Output Short Circuit Current
Supply Current
(Per Channel)
nA/°C
dB
dB
Input Common-Mode Voltage
Range
Output Swing High
Units
117
CMVR
VO
Max
(Note 6)
0
3.3
dB
RL = 10 kΩ to V+/2
5
50
75
RL = 2 kΩ to V+/2
14
50
75
RL = 10 kΩ to V+/2
9
50
75
RL = 2 kΩ to V+/2
13
50
75
Sourcing, VO = V+/2
VIN (diff) = 100 mV
28
22
45
Sinking, VO = V+/2
VIN (diff) = -100 mV
25
20
48
V
mV
from either
rail
mA
VCM = 2.5V
2.1
2.8
3.5
VCM = 0.5V
2.4
3.2
4.0
mA
SR
Slew Rate
AV = +1, CL = 10 pF, RL = 10 kΩ to V+/2,
VO = 2 VPP
2.4
V/μs
GBW
Gain Bandwidth Product
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 V+/2
62
deg
RIN
Input Resistance
Differential Mode
38
kΩ
Common Mode
151
MΩ
0.002
%
THD+N
Total Harmonic Distortion
AV = 1, f = 1 kHz, Amplitude = 1V,
en
Input-Referred Voltage Noise
f = 1 kHz, VCM = 2.5V
2.9
f = 1 kHz, VCM = 0.5V
2.9
0.1 Hz to 10 Hz
in
Input-Referred Current Noise
www.national.com
65
f = 1 kHz, VCM = 2.5V
1.1
f = 1 kHz, VCM = 0.5V
2.1
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
Typ
(Note 5)
Max
(Note 6)
VCM = 4.5V
±6
±50
±120
VCM = 0.5V
±6
±40
±100
VCM = 4.5V
±0.5
±1.0
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 Voltage
(Note 7)
Input Offset Voltage Drift
Input Offset Current
Conditions
TCIOS
Input Offset Current Drift
VCM = 0.5V and VCM = 4.5V
CMRR
Common Mode Rejection Ratio
0.15V ≤ VCM ≤ 0.7V
101
89
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
Large Signal CMRR ≥ 80 dB
AVOL
Large Signal Voltage Gain
RL = 10 kΩ to V+/2
VO = 0.5V to 4.5V
112
104
130
RL = 2 kΩ to V+/2
VO = 0.5V to 4.5V
110
94
119
Output Swing Low
IO
IS
SR
Output Short Circuit Current
Supply Current
(Per Channel)
Slew Rate
μV/°C
nA
nA
nA/°C
dB
dB
Input Common-Mode Voltage
Range
Output Swing High
μV
117
CMVR
VO
Units
μV/month
0.0482
120
0.23V ≤ VCM ≤ 0.7V
PSRR
Min
(Note 6)
0
5
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, VO = V+/2
VIN (diff) = 100 mV
33
27
47
Sinking, VO = V+/2
VIN (diff) = -100 mV
30
25
49
mV
from either
rail
mA
VCM = 4.5V
2.2
3.0
3.7
VCM = 0.5V
2.5
3.4
4.2
AV = +1, CL = 10 pF, RL = 10 kΩ to V+/2,
VO = 2 VPP
2.4
5
V
mA
V/μs
www.national.com
LMP7731
5V Electrical Characteristics
LMP7731
Symbol
Parameter
Conditions
GBW
Gain Bandwidth Product
GM
Gain Margin
CL = 20 pF, RL = 10 kΩ to
ΦM
RIN
Min
(Note 6)
Typ
(Note 5)
Units
CL = 20 pF, RL = 10 kΩ to V+/2
22
MHz
V+/2
12
dB
Phase Margin
CL = 20 pF, RL = 10 kΩ to V+/2
65
deg
Input Resistance
Differential Mode
38
kΩ
Common Mode
151
MΩ
0.001
%
THD+N
Total Harmonic Distortion
AV = 1, f = 1 kHz, Amplitude = 1V
en
Input-Referred Voltage Noise
f = 1 kHz, VCM = 4.5V
2.9
f = 1 kHz, VCM = 0.5V
2.9
0.1 Hz to 10 Hz
in
Max
(Note 6)
nV/
78
Input-Referred Current Noise
f = 1 kHz, VCM = 4.5V
1.1
f = 1 kHz, VCM = 0.5V
2.2
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)
Field-Induced 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. 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
5-Pin SOT23
20175202
Top View
Ordering Information
Package
5-Pin SOT23
www.national.com
Part Number
LMP7731MF
LMP7731MFX
Package Marking
Transport Media
1k Units Tape and Reel
AY3A
3k Units Tape and Reel
6
NSC Drawing
MF05A
Unless otherwise noted: TA = 25°C, RL > 10 kΩ, VCM = VS/2.
Offset Voltage Distribution
TCVOS Distribution
20175238
20175234
Offset Voltage Distribution
TCVOS Distribution
20175236
20175239
Offset Voltage Distribution
TCVOS Distribution
20175240
20175235
7
www.national.com
LMP7731
Typical Performance Characteristics
LMP7731
Offset Voltage Distribution
TCVOS Distribution
20175241
20175237
Offset Voltage vs. Temperature
Offset Voltage vs. Temperature
20175251
20175252
PSRR vs. Frequency
CMRR vs. Frequency
20175256
20175229
www.national.com
8
LMP7731
Offset Voltage vs. Supply Voltage
Offset Voltage vs. VCM
20175242
20175243
Offset Voltage vs. VCM
Offset Voltage vs. VCM
20175245
20175244
Input Offset Voltage Time Drift
Slew Rate vs. Supply Voltage
20175220
20175230
9
www.national.com
LMP7731
Input Voltage Noise vs. Frequency
Input Current Noise vs. Frequency
20175261
20175262
Time Domain Voltage Noise
Time Domain Voltage Noise
20175267
20175269
Time Domain Voltage Noise
Output Voltage vs. Output Current
20175268
20175259
www.national.com
10
LMP7731
Input Bias Current vs. VCM
Input Bias Current vs. VCM
20175226
20175225
Input Bias Current vs. VCM
Open Loop Frequency Response Over Temperature
20175218
20175227
Open Loop Frequency Response
Open Loop Frequency Response
20175228
20175219
11
www.national.com
LMP7731
THD+N vs. Frequency
THD+N vs. Output Voltage
20175257
20175258
Large Signal Step Response
Small Signal Step Response
20175222
20175221
Large Signal Step Response
Small Signal Step Response
20175224
www.national.com
20175223
12
Output Swing High vs. Supply Voltage
20175246
20175250
Output Swing Low vs. Supply Voltage
Sinking Current vs, Supply Voltage
20175249
20175247
Sourcing Current vs. Supply Voltage
20175248
13
www.national.com
LMP7731
Supply Current vs. Supply Voltage
LMP7731
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 Information
LMP7731
The LMP7731 is a single, low noise, low offset voltage, railto-rail input and output, and low voltage precision amplifier.
with a 1/f corThe low input voltage noise of only 2.9 nV/
ner at 3 Hz makes the LMP7731 ideal for sensor applications
where DC accuracy is of importance.
The LMP7731 has a 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 as the error contribution by the amplifier
is at a minimum.
The LMP7731 has a high gain bandwidth of 22 MHz. This
wide bandwidth enables use of the amplifier at higher gain
settings while retaining 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 the signal to
noise ratio.
The LMP7731 has proprietary input bias cancellation circuitry
on the input stages. This allows the LMP7731 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 LMP7731 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 are available for
sensitive applications or applications with a constrained error
budget.
The LMP7731 is offered in the space saving 5-Pin SOT23
package. This small package is an ideal solution for area
constrained PC boards and portable electronics.
20175206
FIGURE 1. Input Bias Current Cancellation
INPUT VOLTAGE NOISE MEASUREMENT
The LMP7731 has very low input voltage noise. The peak-topeak input voltage noise of the LMP7731 can be measured
using the test circuit shown in Figure 2
INPUT BIAS CURRENT CANCELLATION
The LMP7731 has proprietary input bias current cancellation
circuitry on their input stages.
The LMP7731 has rail-to-rail input. This is achieved by having
two input stages in parallel. Figure 1 shows only one of the
input stages as the circuitry is symmetrical for both stages.
20175255
FIGURE 2. 0.1 Hz to 10 Hz Noise Test Circuit
www.national.com
14
DIODES BETWEEN THE INPUTS
The LMP7731 has a set of anti-parallel diodes between the
input pins as shown in Figure 5. These diodes are present to
protect the input stage of the amplifier. 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.
20175268
FIGURE 3. 0.1 Hz to 10 Hz Input Voltage Noise
Measuring the very low peak-to-peak noise performance of
the LMP7731, 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
LMP7731 units.
20175204
FIGURE 5. Anti-Parallel Diodes between Inputs
20175230
FIGURE 4. Start-Up Input Offset Voltage Drift
15
www.national.com
LMP7731
During the peak-to-peak noise measurement, the LMP7731
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 noise contributions
of noise from frequencies below 0.1 Hz.
Figure 3 shows typical peak-to-peak noise for the LMP7731
measured with the circuit in Figure 2 for the LMP7731.
LMP7731
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 voltage 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 drawback to 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 the 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 temperature of their surrounding environment 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, whereas 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 the 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.
20175205
FIGURE 6. Driving an ADC
www.national.com
16
LMP7731
and
As shown above, the NEP of two thermopiles cannot be compared 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:
20175207
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.
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 varies 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 thermopiles 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.
The noise spectral density of a resistor is calculated by:
The 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 is 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.
Sensor's Impedance and Op Amp'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 the amplifier's input bias current will be. For very
high impedance sensors, it is imperative that op amps 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 op amp 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.
The lowest expected level of the sensor's output is very important. It is necessary to compare this level with the 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 op amp's total input noise, then signal integrity at the op amp's output and the ADC's input will be
compromised.
Where k is the Boltzman constant and T is absolute temperature. The unit of noise spectral density is: V/
For the thermopile sensor, this noise is usually represented
by VNOISE where:
A typical value for this voltage noise is 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 bandwidth. A smaller NEP is desired, however NEP is dependent
on the thermopile active area, AD. For a thermopile
17
www.national.com
LMP7731
20175260
FIGURE 8. Thermopile Amplifier
Figure 8 shows the LMP7731 used as a thermopile amplifier.
The LMP7731 is a great choice for use with thermopile sensors. The LMP7731 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 LMP7731 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 LMP7731 is connected
to the thermopile. This circuit is comprised of two LMP7731
amplifiers, the LM4140A-2.5 which is a precision voltage reference, the ADC122S021 which is a 2 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 LMP7731 amplifiers are referred to as amplifier 1
and amplifier 2 per Figure 8.
In Figure 8 the 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 LMP7731 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
LMP7731
Physical Dimensions inches (millimeters) unless otherwise noted
5-Pin SOT23
NS Package Number MF05A
19
www.national.com
LMP7731, Low Noise, Precision, RRIO, Amplifier in SOT23–5
Notes
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© 2007 National Semiconductor Corporation
For the most current product information visit us at www.national.com
National Semiconductor
Americas Customer
Support Center
Email:
[email protected]
Tel: 1-800-272-9959
www.national.com
National Semiconductor Europe
Customer Support Center
Fax: +49 (0) 180-530-85-86
Email: [email protected]
Deutsch Tel: +49 (0) 69 9508 6208
English Tel: +49 (0) 870 24 0 2171
Français Tel: +33 (0) 1 41 91 8790
National Semiconductor Asia
Pacific Customer Support Center
Email: [email protected]
National Semiconductor Japan
Customer Support Center
Fax: 81-3-5639-7507
Email: [email protected]
Tel: 81-3-5639-7560