Download LT1920 - Single Resistor Gain Programmable, Precision Instrumentation Amplifier

LT1920
Single Resistor Gain
Programmable, Precision
Instrumentation Amplifier
U
DESCRIPTIO
FEATURES
■
■
■
■
■
■
■
■
■
■
■
■
■
■
The LT ®1920 is a low power, precision instrumentation
amplifier that requires only one external resistor to set gains
of 1 to 10,000. The low voltage noise of 7.5nV/√Hz (at 1kHz)
is not compromised by low power dissipation (0.9mA typical
for ±2.3V to ±15V supplies).
The high accuracy of 30ppm maximum nonlinearity and
0.3% max gain error (G = 10) is not degraded even for load
resistors as low as 2k (previous monolithic instrumentation
amps used 10k for their nonlinearity specifications). The
LT1920 is laser trimmed for very low input offset voltage
(125µV max), drift (1µV/°C), high CMRR (75dB, G = 1) and
PSRR (80dB, G = 1). Low input bias currents of 2nA max are
achieved with the use of superbeta processing. The output
can handle capacitive loads up to 1000pF in any gain configuration while the inputs are ESD protected up to 13kV (human
body). The LT1920 with two external 5k resistors passes the
IEC 1000-4-2 level 4 specification.
The LT1920, offered in 8-pin PDIP and SO packages, is a pin
for pin and spec for spec improved replacement for the
AD620. The LT1920 is the most cost effective solution for
precision instrumentation amplifier applications. For even
better guaranteed performance, see the LT1167.
Single Gain Set Resistor: G = 1 to 10,000
Gain Error: G = 10, 0.3% Max
Gain Nonlinearity: G = 10, 30ppm Max
Input Offset Voltage: G = 10, 225µV Max
Input Offset Voltage Drift: 1µV/°C Max
Input Bias Current: 2nA Max
PSRR at G = 1: 80dB Min
CMRR at G = 1: 75dB Min
Supply Current: 1.3mA Max
Wide Supply Range: ± 2.3V to ±18V
1kHz Voltage Noise: 7.5nV/√Hz
0.1Hz to 10Hz Noise: 0.28µVP-P
Available in 8-Pin PDIP and SO Packages
Meets IEC 1000-4-2 Level 4 ESD Tests with
Two External 5k Resistors
U
APPLICATIO S
■
■
■
■
■
Bridge Amplifiers
Strain Gauge Amplifiers
Thermocouple Amplifiers
Differential to Single-Ended Converters
Medical Instrumentation
, LTC and LT are registered trademarks of Linear Technology Corporation.
U
TYPICAL APPLICATIO
Single Supply Barometer
VS
3
2
2
LUCAS NOVA SENOR
NPC-1220-015-A-3L
8
+
1/2
LT1490
1
LT1634CCZ-1.25
Gain Nonlinearity
–
4
1
–
4
5k
R6
1k
5
6
R8
100k
5k
2
+
7
R2
12Ω
5k
+
3
6
LT1920
G = 60
8
3
5
TO
4-DIGIT
DVM
+
4
5
1/2
LT1490
–
–
R1
825Ω
RSET
R3
50k
2
1
5k
6
R4
50k
VS
1
NONLINEARITY (100ppm/DIV)
R5
392k
OUTPUT VOLTAGE (2V/DIV)
G = 1000
RL = 1k
VOUT = ±10V
7
R7
50k
VS = 8V TO 30V
VOLTS
2.800
3.000
3.200
INCHES Hg
28.00
30.00
32.00
1167 TA02
1920 TA01
www.BDTIC.com/Linear
1
LT1920
W
U
PACKAGE/ORDER INFORMATION
U
W W
W
(Note 1)
Supply Voltage ...................................................... ±20V
Differential Input Voltage (Within the
Supply Voltage) ..................................................... ±40V
Input Voltage (Equal to Supply Voltage) ................ ±20V
Input Current (Note 3) ........................................ ±20mA
Output Short-Circuit Duration .......................... Indefinite
Operating Temperature Range ................ – 40°C to 85°C
Specified Temperature Range
LT1920C (Note 4) .................................... 0°C to 70°C
LT1920I .............................................. – 40°C to 85°C
Storage Temperature Range ................. – 65°C to 150°C
Lead Temperature (Soldering, 10 sec).................. 300°C
ELECTRICAL CHARACTERISTICS
ORDER PART
NUMBER
TOP VIEW
8
RG
–IN 2
–
7
+VS
+IN 3
+
6
OUTPUT
5
REF
RG 1
–VS 4
LT1920CN8
LT1920CS8
LT1920IN8
LT1920IS8
N8 PACKAGE
8-LEAD PDIP
S8 PACKAGE
8-LEAD PLASTIC SO
S8 PART MARKING
TJMAX = 150°C, θJA = 130°C/ W (N8)
TJMAX = 150°C, θJA = 190°C/ W (S8)
1920
1920I
Consult factory for Military grade parts.
VS = ±15V, VCM = 0V, TA = 25°C, R L = 2k, unless otherwise noted.
SYMBOL
PARAMETER
CONDITIONS (Note 6)
G
Gain Range
G = 1 + (49.4k/RG )
Gain Error
G=1
G = 10 (Note 2)
G = 100 (Note 2)
G = 1000 (Note 2)
G/T
U
ABSOLUTE MAXIMUM RATINGS
Gain vs Temperature
G < 1000 (Note 2)
Gain Nonlinearity (Note 5)
VO = ±10V, G = 1
VO = ±10V, G = 10 and 100
VO = ±10V, G = 100 and 1000
MIN
TYP
1
●
VOST
Total Input Referred Offset Voltage
VOST = VOSI + VOSO/G
VOSI
Input Offset Voltage
G = 1000, VS = ±5V to ±15V
G = 1000, VS = ±5V to ±15V
●
●
VOSI/T
Input Offset Drift (RTI)
(Note 3)
VOSO
Output Offset Voltage
G = 1, VS = ±5V to ±15V
G = 1, VS = ±5V to ±15V
●
(Note 3)
●
MAX
UNITS
10k
0.008
0.010
0.025
0.040
0.1
0.3
0.3
0.35
20
50
ppm/°C
10
10
20
30
ppm
ppm
ppm
30
125
185
µV
µV
400
1000
1500
1
%
%
%
%
µV/°C
µV
µV
VOSO /T
Output Offset Drift
5
15
µV/°C
IOS
Input Offset Current
0.3
1
nA
IB
Input Bias Current
0.5
2
nA
en
Input Noise Voltage, RTI
0.1Hz to 10Hz, G = 1
0.1Hz to 10Hz, G = 10
0.1Hz to 10Hz, G = 100 and 1000
2.00
0.50
0.28
µVP-P
µVP-P
µVP-P
Total RTI Noise = √eni 2 + (eno /G)2
eni
Input Noise Voltage Density, RTI
fO = 1kHz
7.5
nV/√Hz
eno
Output Noise Voltage Density, RTI
fO = 1kHz
67
nV/√Hz
in
Input Noise Current
fO = 0.1Hz to 10Hz
10
pAP-P
Input Noise Current Density
fO = 10Hz
124
fA/√Hz
RIN
Input Resistance
VIN = ±10V
200
GΩ
CIN(DIFF)
Differential Input Capacitance
fO = 100kHz
1.6
pF
2
www.BDTIC.com/Linear
LT1920
ELECTRICAL CHARACTERISTICS
VS = ±15V, V CM = 0V, TA = 25°C, R L = 2k, unless otherwise noted.
SYMBOL
PARAMETER
CIN(CM)
Common Mode Input Capacitance
fO = 100kHz
VCM
Input Voltage Range
G = 1, Other Input Grounded
VS = ±2.3V to ±5V
VS = ±5V to ±18V
VS = ±2.3V to ±5V
VS = ±5V to ±18V
CMRR
PSRR
CONDITIONS (Note 6)
Common Mode Rejection Ratio
Power Supply Rejection Ratio
MIN
1.6
●
●
–VS + 1.9
–VS + 1.9
–VS + 2.1
–VS + 2.1
80
100
120
120
120
135
140
150
dB
dB
dB
dB
RL = 10k
VS = ±2.3V to ±5V
VS = ±5V to ±18V
VS = ±2.3V to ±5V
VS = ±5V to ±18V
IOUT
Output Current
BW
Bandwidth
G=1
G = 10
G = 100
G = 1000
SR
Slew Rate
Settling Time to 0.01%
0.9
●
●
–VS + 1.1
–VS + 1.2
–VS + 1.4
–VS + 1.6
20
Reference Voltage Range
AVREF
Reference Gain to Output
V
V
V
V
VS = ±2.3 to ±18V
G=1
G = 10
G = 100
G = 1000
Output Voltage Swing
VREF
+VS – 1.2
+VS – 1.4
+VS – 1.3
+VS – 1.4
dB
dB
dB
dB
VOUT
Reference Input Current
pF
95
115
125
140
VS = ±2.3V to ±18V
Reference Input Resistance
UNITS
75
95
110
110
Supply Current
IREFIN
MAX
1k Source Imbalance,
VCM = 0V to ±10V
G=1
G = 10
G = 100
G = 1000
IS
RREFIN
TYP
1.3
+VS – 1.2
+VS – 1.3
+VS – 1.3
+VS – 1.5
mA
V
V
V
V
27
mA
1000
800
120
12
kHz
kHz
kHz
kHz
G = 1, VOUT = ±10V
1.2
V/µs
10V Step
G = 1 to 100
G = 1000
14
130
µs
µs
20
kΩ
50
µA
VREF = 0V
The ● denotes specifications that apply over the full specified
temperature range.
Note 1: Absolute Maximum Ratings are those values beyond which the life
of a device may be impaired.
Note 2: Does not include the effect of the external gain resistor RG.
Note 3: This parameter is not 100% tested.
Note 4: The LT1920C is designed, characterized and expected to meet the
industrial temperature limits, but is not tested at – 40°C and 85°C. I-grade
parts are guaranteed.
– VS + 1.6
+VS – 1.6
V
1 ± 0.0001
Note 5: This parameter is measured in a high speed automatic tester that
does not measure the thermal effects with longer time constants. The
magnitude of these thermal effects are dependent on the package used,
heat sinking and air flow conditions.
Note 6: Typical parameters are defined as the 60% of the yield parameter
distribution.
www.BDTIC.com/Linear
3
LT1920
U W
TYPICAL PERFOR A CE CHARACTERISTICS
Gain Nonlinearity, G = 100
NONLINEARITY (10ppm/DIV)
Gain Nonlinearity, G = 10
NONLINEARITY (1ppm/DIV)
NONLINEARITY (10ppm/DIV)
Gain Nonlinearity, G = 1
1167 G01
1167 G02
OUTPUT VOLTAGE (2V/DIV)
G = 10
RL = 2k
V OUT = ± 10V
Gain Error vs Temperature
NONLINEARITY (100ppm/DIV)
Gain Nonlinearity, G = 1000
Warm-Up Drift
0.20
14
0.15
12
GAIN ERROR (%)
0.10
0.05
G=1
0
– 0.05
– 0.10
G = 1000 OUTPUT VOLTAGE (2V/DIV)
RL = 2k
VOUT = ±10V
1167 G04
– 0.15
VS = ±15V
G = 10*
VOUT = ±10V
RL = 2k
G = 100*
*DOES NOT INCLUDE
G = 1000*
TEMPERATURE EFFECTS
OF R G
– 0.20
– 50
– 25
0
25
50
TEMPERATURE (°C)
S8
10
8
N8
6
4
2
0
1
2
3
4
TIME AFTER POWER ON (MINUTES)
0
100
75
VS = ± 15V
TA = 25°C
G=1
COMMON MODE REJECTION RATIO (dB)
160
INPUT BIAS CURRENT (pA)
400
300
200
100
0
85°C
70°C
–100
– 200
0°C
– 300
– 40°C
25°C
– 400
– 500
–15 –12 – 9 – 6 – 3 0 3 6 9 12 15
COMMON MODE INPUT VOLTAGE (V)
1920 G13
4
140
VS = ±15V
TA = 25°C
1k SOURCE
IMBALANCE
G = 1000
120
G = 100
G = 10
100
G=1
Negative Power Supply Rejection
Ratio vs Frequency
80
60
40
20
0
0.1
1
10
1k
100
FREQUENCY (Hz)
10k
100k
NEGATIVE POWER SUPPLY REJECTION RATIO (dB)
Common Mode Rejection Ratio
vs Frequency
500
5
1920 G09
1920 G06
Input Bias Current
vs Common Mode Input Voltage
1167 G03
G = 100 OUTPUT VOLTAGE (2V/DIV)
RL = 2k
VOUT = ±10V
CHANGE IN OFFSET VOLTAGE (µV)
G=1
OUTPUT VOLTAGE (2V/DIV)
RL = 2k
VOUT = ±10V
160
140
G = 100
120
G = 10
100
G=1
V + = 15V
TA = 25°C
G = 1000
80
60
40
20
0
0.1
1920 G14
www.BDTIC.com/Linear
1
10
1k
100
FREQUENCY (Hz)
10k
100k
1920 G15
LT1920
U W
TYPICAL PERFOR A CE CHARACTERISTICS
160
V – = – 15V
TA = 25°C
140
G=1
60
10
0
20
–10
1
10
1k
100
FREQUENCY (Hz)
10k
100k
G = 10
20
40
0
0.1
G = 100
30
GAIN (dB)
80
G = 1000
40
G = 100
100
1.50
50
G = 1000
G = 10
120
Supply Current vs Supply Voltage
Gain vs Frequency
60
SUPPLY CURRENT (mA)
POSITIVE POWER SUPPLY REJECTION RATIO (dB)
Positive Power Supply Rejection
Ratio vs Frequency
G=1
– 20
0.01
0.1
100
1
10
FREQUENCY (kHz)
0.75
0.50
1000
10
15
5
SUPPLY VOLTAGE (± V)
0
0.1Hz to 10Hz Noise Voltage, RTI
G = 1000
VS = ±15V
TA = 25°C
NOISE VOLTAGE (0.2µV/DIV)
VS = ±15V
TA = 25°C
1/fCORNER = 10Hz
GAIN = 1
1/fCORNER = 9Hz
GAIN = 10
1/fCORNER = 7Hz
20
1920 G18
NOISE VOLTAGE (2µV/DIV)
VOLTAGE NOISE DENSITY (nV√Hz)
– 40°C
0.1Hz to 10Hz Noise Voltage,
G=1
VS = ±15V
TA = 25°C
10
25°C
1.00
1920 G17
Voltage Noise Density
vs Frequency
100
85°C
VS = ± 15V
TA = 25°C
1920 G16
1000
1.25
GAIN = 100, 1000
BW LIMIT
GAIN = 1000
0
1
10
100
1k
FREQUENCY (Hz)
10k
100k
2
1
0
3
4 5 6
TIME (SEC)
7
8
9
1920 G19
VS = ±15V
TA = 25°C
1000
1920 G22
7
8
9
10
VS = ±15V
OUTPUT CURRENT (mA)
(SINK)
(SOURCE)
40
10
10
100
FREQUENCY (Hz)
4 5 6
TIME (SEC)
Short-Circuit Current vs Time
30
20
10
TA = – 40°C
TA = 25°C
TA = 85°C
0
– 10
– 20
– 30
– 40
1
3
50
CURRENT NOISE (5pA/DIV)
CURRENT NOISE DENSITY (fA/√Hz)
VS = ±15V
TA = 25°C
RS
2
1920 G21
0.1Hz to 10Hz Current Noise
100
1
1920 G20
Current Noise Density
vs Frequency
1000
0
10
TA = 85°C
TA = – 40°C
TA = 25°C
– 50
0
1
2
3
4 5 6
TIME (SEC)
7
8
9
10
2
1
0
3
TIME FROM OUTPUT SHORT TO GROUND (MINUTES)
1920 G23
www.BDTIC.com/Linear
1920 G24
5
LT1920
U W
TYPICAL PERFOR A CE CHARACTERISTICS
Small-Signal Transient Response
Large-Signal Transient Response
Overshoot vs Capacitive Load
100
VS = ±15V
VOUT = ± 50mV
RL = ∞
90
G=1
VS = ±15V
RL = 2k
CL = 60pF
10µs/DIV
1167 G28
OVERSHOOT (%)
5V/DIV
20mV/DIV
80
G=1
V S = ±15V
R L = 2k
C L = 60pF
10µs/DIV
70
60
50
AV = 1
40
30
AV = 10
20
1167 G29
10
AV ≥ 100
0
10
100
1000
CAPACITIVE LOAD (pF)
10000
1920 G25
Large-Signal Transient Response
Output Impedance vs Frequency
Small-Signal Transient Response
5V/DIV
20mV/DIV
OUTPUT IMPEDANCE (Ω)
1000
G = 10
VS = ± 15V
RL = 2k
CL = 60pF
10µs/DIV
1167 G31
G = 10
VS = ±15V
RL = 2k
CL = 60pF
10µs/DIV
VS = ± 15V
TA = 25°C
G = 1 TO 1000
100
10
1
1167 G32
0.1
1
10
100
FREQUENCY (kHz)
1000
1920 G26
Small-Signal Transient Response
10µs/DIV
1167 G34
35
PEAK-TO-PEAK OUTPUT SWING (V)
5V/DIV
20mV/DIV
Large-Signal Transient Response
G = 100
VS = ± 15V
RL = 2k
CL = 60pF
Undistorted Output Swing
vs Frequency
G = 100
VS = ±15V
RL = 2k
CL = 60pF
10µs/DIV
1167 G35
30 G = 10, 100, 1000
G=1
25
VS = ± 15V
TA = 25°C
20
15
10
5
0
1
10
100
FREQUENCY (kHz)
1000
1920 G27
6
www.BDTIC.com/Linear
LT1920
U W
TYPICAL PERFOR A CE CHARACTERISTICS
Large-Signal Transient Response
Small-Signal Transient Response
Settling Time vs Gain
5V/DIV
20mV/DIV
SETTLING TIME (µs)
1000
1167 G37
50µs/DIV
G = 1000
V S = ±15V
RL = 2k
C L = 60pF
50µs/DIV
G = 1000
VS = ±15V
RL = 2k
CL = 60pF
VS = ± 15V
TA = 25°C
∆VOUT = 10V
1mV = 0.01%
100
10
1167 G38
1
1
10
100
1000
GAIN (dB)
1920 G30
Settling Time vs Step Size
VS = ±15
G=1
TA = 25°C
CL = 30pF
RL = 1k
8
2
TO 0.1%
1.6
VOUT
0V
0
0V
–2
VOUT
–4
TO 0.01%
–6
–8
–10
3
4
1.4
+ SLEW
1.2
– SLEW
1.0
TO 0.1%
2
VS = ± 15V
VOUT = ±10V
G=1
TO 0.01%
0.8
– 50 –25
5 6 7 8 9 10 11 12
SETTLING TIME (µs)
50
0
25
75
TEMPERATURE (°C)
100
125
1920 G36
1920 G33
Output Voltage Swing
vs Load Current
+ VS
OUTPUT VOLTAGE SWING (V)
(REFERRED TO SUPPLY VOLTAGE)
OUTPUT STEP (V)
6
4
Slew Rate vs Temperature
1.8
SLEW RATE (V/µs)
10
VS = ± 15V
85°C
25°C
– 40°C
+ VS – 0.5
+ VS – 1.0
+ VS – 1.5
SOURCE
+ VS – 2.0
– VS + 2.0
– VS + 1.5
SINK
– VS + 1.0
– VS + 0.5
– VS
0.01
0.1
1
10
OUTPUT CURRENT (mA)
100
1920 G39
www.BDTIC.com/Linear
7
LT1920
W
BLOCK DIAGRAM
V+
VB
R5
10k
+
R6
10k
6 OUTPUT
A1
–
R3
400Ω
–IN
C1
2
Q1
R1
24.7k
V
–
–
+
A3
RG 1
RG 8
V–
VB
V+
+
R7
10k
R8
10k
5 REF
A2
–
R4
400Ω
+IN
3
C2
V–
Q2
R2
24.7k
7
V+
4
V–
V–
PREAMP STAGE
DIFFERENCE AMPLIFIER STAGE
1920 F01
Figure 1. Block Diagram
U
THEORY OF OPERATIO
The LT1920 is a modified version of the three op amp
instrumentation amplifier. Laser trimming and monolithic
construction allow tight matching and tracking of circuit
parameters over the specified temperature range. Refer to
the block diagram (Figure 1) to understand the following
circuit description. The collector currents in Q1 and Q2 are
trimmed to minimize offset voltage drift, thus assuring a
high level of performance. R1 and R2 are trimmed to an
absolute value of 24.7k to assure that the gain can be set
accurately (0.3% at G = 100) with only one external
resistor RG. The value of RG in parallel with R1 (R2)
determines the transconductance of the preamp stage. As
RG is reduced for larger programmed gains, the transconductance of the input preamp stage increases to that of the
input transistors Q1 and Q2. This increases the open-loop
gain when the programmed gain is increased, reducing
the input referred gain related errors and noise. The input
voltage noise at gains greater than 50 is determined only
by Q1 and Q2. At lower gains the noise of the difference
amplifier and preamp gain setting resistors increase the
noise. The gain bandwidth product is determined by C1,
C2 and the preamp transconductance which increases
8
with programmed gain. Therefore, the bandwidth does not
drop proportional to gain.
The input transistors Q1 and Q2 offer excellent matching,
which is inherent in NPN bipolar transistors, as well as
picoampere input bias current due to superbeta processing. The collector currents in Q1 and Q2 are held constant
due to the feedback through the Q1-A1-R1 loop and
Q2-A2-R2 loop which in turn impresses the differential
input voltage across the external gain set resistor RG.
Since the current that flows through RG also flows through
R1 and R2, the ratios provide a gained-up differential voltage,G = (R1 + R2)/RG, to the unity-gain difference amplifier
A3. The common mode voltage is removed by A3, resulting in a single-ended output voltage referenced to the
voltage on the REF pin. The resulting gain equation is:
VOUT – VREF = G(VIN+ – VIN–)
where:
G = (49.4kΩ / RG) + 1
solving for the gain set resistor gives:
RG = 49.4kΩ /(G – 1)
www.BDTIC.com/Linear
LT1920
U
THEORY OF OPERATIO
Input and Output Offset Voltage
Output Offset Trimming
The offset voltage of the LT1920 has two components: the
output offset and the input offset. The total offset voltage
referred to the input (RTI) is found by dividing the output
offset by the programmed gain (G) and adding it to the
input offset. At high gains the input offset voltage dominates, whereas at low gains the output offset voltage
dominates. The total offset voltage is:
The LT1920 is laser trimmed for low offset voltage so that
no external offset trimming is required for most applications. In the event that the offset needs to be adjusted, the
circuit in Figure 2 is an example of an optional offset adjust
circuit. The op amp buffer provides a low impedance to the
REF pin where resistance must be kept to minimum for
best CMRR and lowest gain error.
Total input offset voltage (RTI)
= input offset + (output offset/G)
2
–IN
–
1
Total output offset voltage (RTO)
= (input offset • G) + output offset
RG
+IN
3
OUTPUT
+
5
–
1
±10mV
ADJUSTMENT RANGE
1/2
LT1112
+
The reference terminal is one end of one of the four 10k
resistors around the difference amplifier. The output voltage of the LT1920 (Pin 6) is referenced to the voltage on
the reference terminal (Pin 5). Resistance in series with
the REF pin must be minimized for best common mode
rejection. For example, a 2Ω resistance from the REF pin
to ground will not only increase the gain error by 0.02%
but will lower the CMRR to 80dB.
V+
REF
8
Reference Terminal
6
LT1920
2
10mV
100Ω
3
10k
100Ω
–10mV
V–
1920 F02
Figure 2. Optional Trimming of Output Offset Voltage
Single Supply Operation
Input Bias Current Return Path
For single supply operation, the REF pin can be at the same
potential as the negative supply (Pin 4) provided the
output of the instrumentation amplifier remains inside the
specified operating range and that one of the inputs is at
least 2.5V above ground. The barometer application on the
front page of this data sheet is an example that satisfies
these conditions. The resistance RSET from the bridge
transducer to ground sets the operating current for the
bridge and also has the effect of raising the input common
mode voltage. The output of the LT1920 is always inside
the specified range since the barometric pressure rarely
goes low enough to cause the output to rail (30.00 inches
of Hg corresponds to 3.000V). For applications that require the output to swing at or below the REF potential, the
voltage on the REF pin can be level shifted. An op amp is
used to buffer the voltage on the REF pin since a parasitic
series resistance will degrade the CMRR. The application
in the back of this data sheet, Four Digit Pressure Sensor,
is an example.
The low input bias current of the LT1920 (2nA) and the
high input impedance (200GΩ) allow the use of high
impedance sources without introducing additional offset
voltage errors, even when the full common mode range is
required. However, a path must be provided for the input
bias currents of both inputs when a purely differential
signal is being amplified. Without this path the inputs will
float to either rail and exceed the input common mode
range of the LT1920, resulting in a saturated input stage.
Figure 3 shows three examples of an input bias current
path. The first example is of a purely differential signal
source with a 10kΩ input current path to ground. Since the
impedance of the signal source is low, only one resistor is
needed. Two matching resistors are needed for higher
impedance signal sources as shown in the second
example. Balancing the input impedance improves both
common mode rejection and DC offset. The need for input
resistors is eliminated if a center tap is present as shown
in the third example.
www.BDTIC.com/Linear
9
LT1920
U
THEORY OF OPERATIO
–
RG
THERMOCOUPLE
–
MICROPHONE,
HYDROPHONE,
ETC
LT1920
RG
–
RG
LT1920
+
+
200k
10k
LT1920
+
200k
CENTER-TAP PROVIDES
BIAS CURRENT RETURN
1920 F03
Figure 3. Providing an Input Common Mode Current Path
U
W
U
U
APPLICATIONS INFORMATION
The LT1920 is a low power precision instrumentation
amplifier that requires only one external resistor to accurately set the gain anywhere from 1 to 1000. The output
can handle capacitive loads up to 1000pF in any gain
configuration and the inputs are protected against ESD
strikes up to 13kV (human body).
Input Protection
VCC
J1
2N4393
J2
2N4393
RIN
RFI Reduction
In many industrial and data acquisition applications,
instrumentation amplifiers are used to accurately amplify
small signals in the presence of large common mode
voltages or high levels of noise. Typically, the sources of
these very small signals (on the order of microvolts or
millivolts) are sensors that can be a significant distance
from the signal conditioning circuit. Although these sen-
OPTIONAL FOR HIGHEST
ESD PROTECTION
+
RG
RIN
The LT1920 can safely handle up to ±20mA of input
current in an overload condition. Adding an external 5k
input resistor in series with each input allows DC input
fault voltages up to ±100V and improves the ESD immunity to 8kV (contact) and 15kV (air discharge), which is the
IEC 1000-4-2 level 4 specification. If lower value input
resistors are needed, a clamp diode from the positive
supply to each input will maintain the IEC 1000-4-2
specification to level 4 for both air and contact discharge.
A 2N4393 drain/source to gate is a good low leakage diode
for use with 1k resistors, see Figure 4. The input resistors
should be carbon and not metal film or carbon film.
10
VCC
VCC
LT1920
OUT
REF
–
VEE
1920 F04
Figure 4. Input Protection
sors may be connected to signal conditioning circuitry,
using shielded or unshielded twisted-pair cabling, the cabling may act as antennae, conveying very high frequency
interference directly into the input stage of the LT1920.
The amplitude and frequency of the interference can have
an adverse effect on an instrumentation amplifier’s input
stage by causing an unwanted DC shift in the amplifier’s
input offset voltage. This well known effect is called RFI
rectification and is produced when out-of-band interference is coupled (inductively, capacitively or via radiation)
and rectified by the instrumentation amplifier’s input transistors. These transistors act as high frequency signal
detectors, in the same way diodes were used as RF
envelope detectors in early radio designs. Regardless of
the type of interference or the method by which it is
coupled into the circuit, an out-of-band error signal appears in series with the instrumentation amplifier’s inputs.
www.BDTIC.com/Linear
LT1920
U
U
W
U
APPLICATIONS INFORMATION
To significantly reduce the effect of these out-of-band
signals on the input offset voltage of instrumentation
amplifiers, simple lowpass filters can be used at the
inputs. This filter should be located very close to the input
pins of the circuit. An effective filter configuration is
illustrated in Figure 5, where three capacitors have been
added to the inputs of the LT1920. Capacitors CXCM1 and
CXCM2 form lowpass filters with the external series resistors RS1, 2 to any out-of-band signal appearing on each of
the input traces. Capacitor CXD forms a filter to reduce any
unwanted signal that would appear across the input traces.
An added benefit to using CXD is that the circuit’s AC
common mode rejection is not degraded due to common
mode capacitive imbalance. The differential mode and
common mode time constants associated with the capacitors are:
mode time constant close to the sensor’s BW also minimizes any noise pickup along the leads. To avoid any
possibility of inadvertently affecting the signal to be processed, set the common mode time constant an order of
magnitude (or more) larger than the differential mode time
constant. To avoid any possibility of common mode to
differential mode signal conversion, match the common
mode time constants to 1% or better. If the sensor is an
RTD or a resistive strain gauge, then the series resistors
RS1, 2 can be omitted, if the sensor is in proximity to the
instrumentation amplifier.
IN +
V+
CXCM1
0.001µF
RS1
1.6k
tDM(LPF) = (2)(RS)(CXD)
+
CXD
0.1µF
tCM(LPF) = (RS1, 2)(CXCM1, 2)
Setting the time constants requires a knowledge of the
frequency, or frequencies of the interference. Once this
frequency is known, the common mode time constants
can be set followed by the differential mode time constant.
Set the common mode time constants such that they do
not degrade the LT1920’s inherent AC CMR. Then the
differential mode time constant can be set for the bandwidth required for the application. Setting the differential
U
PACKAGE DESCRIPTION
IN –
RG
RS2
1.6k
LT1920
VOUT
–
CXCM2
0.001µF
V–
1920 F05
EXTERNAL RFI
FILTER
f(–3dB) ≈ 500Hz
Figure 5. Adding a Simple RC Filter at the Inputs to an
Instrumentation Amplifier is Effective in Reducing Rectification
of High Frequency Out-of-Band Signals
Dimensions in inches (millimeters) unless otherwise noted.
N8 Package
8-Lead PDIP (Narrow 0.300)
(LTC DWG # 05-08-1510)
0.300 – 0.325
(7.620 – 8.255)
0.009 – 0.015
(0.229 – 0.381)
(
+0.035
0.325 –0.015
8.255
+0.889
–0.381
)
0.045 – 0.065
(1.143 – 1.651)
0.130 ± 0.005
(3.302 ± 0.127)
0.065
(1.651)
TYP
0.100 ± 0.010
(2.540 ± 0.254)
0.400*
(10.160)
MAX
8
7
6
5
1
2
3
4
0.255 ± 0.015*
(6.477 ± 0.381)
0.125
(3.175) 0.020
MIN (0.508)
MIN
0.018 ± 0.003
(0.457 ± 0.076)
N8 1197
*THESE DIMENSIONS DO NOT INCLUDE MOLD FLASH OR PROTRUSIONS.
MOLD FLASH OR PROTRUSIONS SHALL NOT EXCEED 0.010 INCH (0.254mm)
www.BDTIC.com/Linear
Information furnished by Linear Technology Corporation is believed to be accurate and reliable.
However, no responsibility is assumed for its use. Linear Technology Corporation makes no representation that the interconnection of its circuits as described herein will not infringe on existing patent rights.
11
LT1920
U
TYPICAL APPLICATION
Nerve Impulse Amplifier
3V
PATIENT/CIRCUIT
PROTECTION/ISOLATION
3
8
+IN
C1
0.01µF
R2
1M
RG
6k
R4
30k
–
PATIENT
GROUND
1
3V
6
LT1920
G = 10
1
5
R6
1M
5
2
1/2
LT1112
0.3Hz
HIGHPASS
7
C2
0.47µF
R3
30k
R1
12k
+
–
2
6
4
8
+
–
4
–3V
–3V
+
AV = 101
POLE AT 1kHz
U
PACKAGE DESCRIPTION
OUTPUT
1V/mV
R7
10k
R8
100Ω
3
–IN
7
1/2
LT1112
C3
15nF
1920 TA03
Dimensions in inches (millimeters) unless otherwise noted.
S8 Package
8-Lead Plastic Small Outline (Narrow 0.150)
(LTC DWG # 05-08-1610)
0.189 – 0.197*
(4.801 – 5.004)
0.010 – 0.020
× 45°
(0.254 – 0.508)
0.008 – 0.010
(0.203 – 0.254)
0.053 – 0.069
(1.346 – 1.752)
0.004 – 0.010
(0.101 – 0.254)
7
8
6
5
0°– 8° TYP
0.016 – 0.050
0.406 – 1.270
0.014 – 0.019
(0.355 – 0.483)
0.050
(1.270)
TYP
0.150 – 0.157**
(3.810 – 3.988)
0.228 – 0.244
(5.791 – 6.197)
*DIMENSION DOES NOT INCLUDE MOLD FLASH. MOLD FLASH
SHALL NOT EXCEED 0.006" (0.152mm) PER SIDE
**DIMENSION DOES NOT INCLUDE INTERLEAD FLASH. INTERLEAD
FLASH SHALL NOT EXCEED 0.010" (0.254mm) PER SIDE
SO8 0996
1
3
2
4
RELATED PARTS
PART NUMBER
DESCRIPTION
COMMENTS
LTC1100
Precision Chopper-Stabilized Instrumentation Amplifier
Best DC Accuracy
LT1101
Precision, Micropower, Single Supply Instrumentation Amplifier
Fixed Gain of 10 or 100, IS < 105µA
LT1102
High Speed, JFET Instrumentation Amplifier
Fixed Gain of 10 or 100, 30V/µs Slew Rate
LT1167
Single Resistor Gain Programmable Precision
Instrumentation Amplifier
Upgraded Version of the LT1920
LTC®1418
14-Bit, Low Power, 200ksps ADC with Serial and Parallel I/O
Single Supply 5V or ±5V Operation, ±1.5LSB INL
and ±1LSB DNL Max
LT1460
Precision Series Reference
Micropower; 2.5V, 5V, 10V Versions; High Precision
LTC1562
Active RC Filter
Lowpass, Bandpass, Highpass Responses; Low Noise,
Low Distortion, Four 2nd Order Filter Sections
LTC1605
16-Bit, 100ksps, Sampling ADC
Single 5V Supply, Bipolar Input Range: ±10V,
Power Dissipation: 55mW Typ
12
www.BDTIC.com/Linear
Linear Technology Corporation
1920f LT/TP 0299 4K • PRINTED IN USA
1630 McCarthy Blvd., Milpitas, CA 95035-7417
(408)432-1900 ● FAX: (408) 434-0507 ● www.linear-tech.com
 LINEAR TECHNOLOGY CORPORATION 1998