LMP7708MMX [NSC]
Precision, CMOS Input, RRIO, Wide Supply Range Decompensated Amplifiers; 精密， CMOS输入， RRIO ，宽电源范围代偿放大器
| 型号： | LMP7708MMX |
| 厂家： | 
National Semiconductor |
| 描述： | Precision, CMOS Input, RRIO, Wide Supply Range Decompensated Amplifiers |
| 文件： | 总24页 (文件大小：1536K) |
| 中文： | 中文翻译 | 下载： | 下载PDF数据表文档文件 |
June 2007
LMP7707/LMP7708/LMP7709
Precision, CMOS Input, RRIO, Wide Supply Range
Decompensated Amplifiers
General Description
Features
The LMP7707/LMP7708/LMP7709 devices are single, dual,
and quad low offset voltage, rail-to-rail input and output pre-
cision amplifiers which each have a CMOS input stage and a
wide supply voltage range. The LMP7707/LMP7708/
LMP7709 are part of the LMP® precision amplifier family and
are ideal for sensor interface and other instrumentation ap-
plications. These decompensated amplifiers are stable at a
gain of 6 and higher.
Unless otherwise noted, typical values at VS = 5V.
Input offset voltage (LMP7707)
Input offset voltage (LMP7708/LMP7709) ±220 µV (max)
Input bias current
Input voltage noise
CMRR
Open loop gain
Temperature range
Gain bandwidth product (AV =10)
Stable at a gain of 10 or higher
Supply current (LMP7707)
Supply current (LMP7708)
Supply current (LMP7709)
Supply voltage range
Rail-to-rail input and output
±200 µV (max)
■
■
■
■
■
■
■
■
■
■
■
■
■
■
±200 fA
9 nV/√Hz
130 dB
130 dB
−40°C to 125°C
14 MHz
The guaranteed low offset voltage of less than ±200 µV along
with the guaranteed low input bias current of less than ±1 pA
make the LMP7707/LMP7708/LMP7709 ideal for precision
applications. The LMP7707/LMP7708/LMP7709 are built uti-
lizing VIP50 technology, which allows the combination of a
CMOS input stage and a supply voltage range of 12V with
rail-to-rail common mode voltage capability. The LMP7707/
LMP7708/LMP7709 are the perfect choice in many applica-
tions where conventional CMOS parts cannot operate due to
the voltage conditions.
715 µA
1.5 mA
2.9 mA
2.7V to 12V
Applications
The unique design of the rail-to-rail input stage of each of the
LMP7707/LMP7708/LMP7709 significantly reduces the CM-
RR glitch commonly associated with rail-to-rail input ampli-
fiers. Both sides of the complimentary input stage have been
trimmed, thereby reducing the difference between the NMOS
and PMOS offsets. The output swings within 40 mV of either
rail to maximize the signal dynamic range in applications re-
quiring low supply voltage.
High impedance sensor interface
■
■
■
■
■
■
Battery powered instrumentation
High gain amplifiers
DAC buffer
Instrumentation amplifier
Active filters
The LMP7707 is offered in the space saving 5-Pin SOT23
package, the LMP7708 is offered in the 8-Pin MSOP and the
quad LMP7709 is offered in the 14-Pin TSSOP package.
These small packages are ideal solutions for area con-
strained PC boards and portable electronics.
Open Loop Frequency Response
Increased Bandwidth for Same Supply Current2a02t03A76V4> 10
LMP® is a registered trademark of National Semiconductor Corporation.
© 2007 National Semiconductor Corporation
202037
www.national.com
Junction Temperature (Note 3)
Soldering Information
+150°C
Absolute Maximum Ratings (Note 1)
If Military/Aerospace specified devices are required,
please contact the National Semiconductor Sales Office/
Distributors for availability and specifications.
ꢀInfrared or Convection (20 sec)
235°C
260°C
ꢀWave Soldering Lead Temp. (10
sec)
ESD Tolerance (Note 2)
ꢀHuman Body Model
ꢀMachine Model
ꢀCharge Device Model
VIN Differential
Supply Voltage (VS = V+ – V−)
Voltage at Input/Output Pins
Input Current
2000V
200V
Operating Ratings (Note 1)
Temperature Range (Note 3)
Supply Voltage (VS = V+ – V−)
−40°C to +125°C
2.7V to 12V
1000V
±300 mV
Package Thermal Resistance (θJA) (Note 3)
5-Pin SOT23
8-Pin MSOP
13.2V
V++ 0.3V to V− − 0.3V
265°C/W
235°C/W
122°C/W
10 mA
−65°C to +150°C
14-Pin TSSOP
Storage Temperature Range
3V Electrical Characteristics (Note 4)
Unless otherwise specified, all limits are guaranteed for TA = 25°C, V+ = 3V, V− = 0V, VCM = V+/2, and RL > 10 kΩ to V+/2.
Boldface limits apply at the temperature extremes.
Symbol
Parameter
Conditions
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
VOS
Input Offset Voltage
LMP7707
±37
±56
±200
±500
μV
LMP7708/LMP7709
±220
±520
TCVOS
IB
Input Offset Voltage Drift (Note 7)
Input Bias Current (Notes 7, 8)
±1
±5
±1
μV/°C
pA
±0.2
±50
−40°C ≤ TA ≤ 85°C
−40°C ≤ TA ≤ 125°C
±400
IOS
Input Offset Current
40
fA
CMRR
Common Mode Rejection Ratio
86
80
130
0V ≤ VCM ≤ 3V
LMP7707
dB
84
78
130
98
0V ≤ VCM ≤ 3V
LMP7708/LMP7709
2.7V ≤ V+ ≤ 12V, Vo = V+/2
PSRR
CMVR
Power Supply Rejection Ratio
86
82
dB
V
Input Common-Mode Voltage Range
−0.2
3.2
CMRR ≥ 80 dB
CMRR ≥ 77 dB
−0.2
3.2
AVOL
Open Loop Voltage Gain
100
96
114
114
124
RL = 2 kΩ (LMP7707)
VO = 0.3V to 2.7V
100
94
RL = 2 kΩ (LMP7708/LMP7709)
VO = 0.3V to 2.7V
dB
100
96
RL = 10 kΩ
VO = 0.2V to 2.8V
www.national.com
2
Symbol
VO
Parameter
Output Swing High
Conditions
RL = 2 kΩ to V+/2
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
40
40
30
35
40
45
20
20
42
42
42
80
120
LMP7707
RL = 2 kΩ to V+/2
LMP7708/LMP7709
80
150
mV
from V+
RL = 10 kΩ to V+/2
LMP7707
40
60
RL = 10 kΩ to V+/2
LMP7708/LMP7709
50
100
RL = 2 kΩ to V+/2
LMP7707
Output Swing Low
60
80
RL = 2 kΩ to V+/2
LMP7708/LMP7709
100
170
mV
RL = 10 kΩ to V+/2
LMP7707
40
50
RL = 10 kΩ to V+/2
LMP7708/LMP7709
Sourcing VO = V+/2
VIN = 100 mV
50
90
IO
Output Short Circuit Current
(Notes 3, 9)
25
15
Sinking VO = V+/2
25
20
VIN = −100 mV (LMP7707)
mA
mA
Sinking VO = V+/2
VIN = −100 mV (LMP7708/
LMP7709)
25
15
IS
Supply Current
LMP7707
0.670
1.4
1.0
1.2
LMP7708
LMP7709
1.8
2.1
2.9
3.5
4.5
SR
Slew Rate (Note 10)
VO = 2 VPP,10% to 90%
AV = 10
5.1
13
V/μs
MHz
%
GBWP
THD+N
Gain Bandwidth Product
Total Harmonic Distortion + Noise
f = 1 kHz, AV = 10, VO = 2.5V,
RL = 10 kΩ
0.024
en
in
Input-Referred Voltage Noise
Input-Referred Current Noise
f = 1 kHz
9
1
nV/
fA/
f = 100 kHz
5V Electrical Characteristics (Note 4)
Unless otherwise specified, all limits are guaranteed for TA = 25°C, V+ = 5V, V− = 0V, VCM = V+/2, and RL > 10 kΩ to V+/2.
Boldface limits apply at the temperature extremes.
Symbol
Parameter
Conditions
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
VOS
Input Offset Voltage
LMP7707
±37
±32
±200
±500
μV
LMP7708/LMP7709
±220
±520
TCVOS
IB
Input Offset Voltage Drift (Note 7)
Input Bias Current (Notes 7, 8)
±1
±5
±1
μV/°C
±0.2
±50
−40°C ≤ TA ≤ 85°C
−40°C ≤ TA ≤ 125°C
pA
±400
3
www.national.com
Symbol
IOS
Parameter
Input Offset Current
Conditions
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
40
fA
CMRR
Common Mode Rejection Ratio
88
83
130
130
100
0V ≤ VCM ≤ 5V
LMP7707
dB
86
81
0V ≤ VCM ≤ 5V
LMP7708/LMP7709
2.7V ≤ V+ ≤ 12V, VO = V+/2
PSRR
CMVR
Power Supply Rejection Ratio
86
82
dB
V
Input Common-Mode Voltage Range
−0.2
5.2
CMRR ≥ 80 dB
CMRR ≥ 78 dB
−0.2
5.2
AVOL
Open Loop Voltage Gain
100
96
119
119
130
60
60
40
40
50
50
30
30
66
66
76
76
RL = 2 kΩ (LMP7707)
VO = 0.3V to 4.7V
100
94
RL = 2 kΩ (LMP7708/LMP7709)
VO = 0.3V to 4.7V
dB
100
96
RL = 10 kΩ
VO = 0.2V to 4.8V
RL = 2 kΩ to V+/2
LMP7707
VO
Output Swing High
110
130
RL = 2 kΩ to V+/2
LMP7708/LMP7709
120
200
mV
from V+
RL = 10 kΩ to V+/2
LMP7707
50
70
RL = 10 kΩ to V+/2
LMP7708/LMP7709
60
120
RL = 2 kΩ to V+/2
LMP7707
Output Swing Low
80
90
RL = 2 kΩ to V+/2
LMP7708/LMP7709
120
190
mV
RL = 10 kΩ to V+/2
LMP7707
40
50
RL = 10 kΩ to V+/2
50
100
LMP7708/LMP7709
Sourcing VO = V+/2
VIN = 100 mV (LMP7707)
IO
Output Short Circuit Current
(Notes 3, 9)
40
28
Sourcing VO = V+/2
38
25
VIN = 100 mV (LMP7708/LMP7709)
Sinking VO = V+/2
40
28
mA
VIN = −100 mV (LMP7707)
Sinking VO = V+/2
VIN = −100 mV (LMP7708/
LMP7709)
40
23
IS
Supply Current
LMP7707
0.715
1.5
1.0
1.2
LMP7708
LMP7709
1.9
2.2
mA
2.9
3.7
4.6
SR
Slew Rate (Note 10)
VO = 4 VPP, 10% to 90%
AV = 10
5.6
14
V/μs
MHz
GBWP
Gain Bandwidth Product
www.national.com
4
Symbol
Parameter
Conditions
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
THD+N
Total Harmonic Distortion + Noise
f = 1 kHz, AV = 10, VO = 4.5V,
RL = 10 kΩ
0.024
%
en
in
Input-Referred Voltage Noise
Input-Referred Current Noise
f = 1 kHz
9
1
nV/
fA/
f = 100 kHz
±5V Electrical Characteristics (Note 4)
Unless otherwise specified, all limits are guaranteed for TA = 25°C, V+ = 5V, V− = −5V, VCM = 0V, and RL > 10 kΩ to 0V. Bold-
face limits apply at the temperature extremes.
Symbol
Parameter
Conditions
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
VOS
Input Offset Voltage
LMP7707
±37
±37
±200
±500
μV
LMP7708/LMP7709
±220
±520
TCVOS
IB
Input Offset Voltage Drift (Note 7)
Input Bias Current (Notes 7, 8)
±1
±5
1
μV/°C
pA
±0.2
±50
−40°C ≤ TA ≤ 85°C
−40°C ≤ TA ≤ 125°C
±400
IOS
Input Offset Current
40
fA
CMRR
Common Mode Rejection Ratio
92
88
138
−5V ≤ VCM ≤ 5V
LMP7707
dB
90
86
138
98
−5V ≤ VCM ≤ 5V
LMP7708/LMP7709
2.7V ≤ V+ ≤ 12V, V- = 0V, VO = V+/2
PSRR
CMVR
Power Supply Rejection Ratio
86
82
dB
V
Input Common-Mode Voltage Range
−5.2
5.2
CMRR ≥ 80 dB
CMRR ≥ 78 dB
−5.2
5.2
AVOL
Open Loop Voltage Gain
100
98
121
121
134
134
RL = 2 kΩ (LMP7707)
VO = −4.7V to 4.7V
100
94
RL = 2 kΩ (LMP7708/LMP7709)
VO = −4.7V to 4.7V
dB
100
98
RL = 10 kΩ (LMP7707)
VO = −4.8V to 4.8V
100
97
RL = 10 kΩ (LMP7708/LMP7709)
VO = −4.8V to 4.8V
5
www.national.com
Symbol
VO
Parameter
Output Swing High
Conditions
RL = 2 kΩ to 0V
Min
Typ
Max
Units
(Note 6) (Note 5) (Note 6)
90
90
40
40
90
90
40
40
86
86
84
150
170
LMP7707
180
290
RL = 2 kΩ to 0V
LMP7708/LMP7709
mV
from V+
80
100
RL = 10 kΩ to 0V
LMP7707
80
150
RL = 10 kΩ to 0V
LMP7708/LMP7709
Output Swing Low
130
150
RL = 2 kΩ to 0V
LMP7707
180
290
RL = 2 kΩ to 0V
LMP7708/LMP7709
mV
from V–
50
60
RL = 10 kΩ to 0V
LMP7707
60
110
RL = 10 kΩ to 0V
LMP7708/LMP7709
IO
Output Short Circuit Current
(Notes 3, 9)
Sourcing VO = 0V
50
35
VIN = 100 mV (LMP7707)
Sourcing VO = 0V
48
33
mA
mA
VIN = 100 mV (LMP7708/LMP7709)
Sinking VO = 0V
VIN = −100 mV
50
35
IS
Supply Current
LMP7707
LMP7708
LMP7709
0.790
1.7
1.1
1.3
2.1
2.5
3.2
4.2
5.0
SR
Slew Rate (Note 10)
VO = 9 VPP, 10% to 90%
AV = 10
5.9
15
V/μs
MHz
%
GBWP
THD+N
Gain Bandwidth Product
Total Harmonic Distortion + Noise
f = 1 kHz, AV = 10, VO = 9V,
RL = 10 kΩ
0.024
en
in
Input-Referred Voltage Noise
Input-Referred Current Noise
f = 1 kHz
9
1
nV/
fA/
f = 100 kHz
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.
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: Limits are 100% production tested at 25°C. Limits over the operating temperature range are guaranteed through correlations using the Statistical Quality
Control (SQC) method.
Note 7: This parameter is guaranteed by design and/or characterization and is not tested in production.
Note 8: Positive current corresponds to current flowing into the device.
Note 9: The short circuit test is a momentary test.
Note 10: The number specified is the slower of positive and negative slew rates.
www.national.com
6
Connection Diagrams
5-Pin SOT23
8-Pin MSOP
14-Pin TSSOP
20203703
20203702
20203704
Top View
Top View
Top View
Ordering Information
Package
Part Number
Package Marking
Transport Media
NSC Drawing
LMP7707MF
LMP7707MFX
LMP7708MM
LMP7708MMX
LMP7709MT
LMP7709MTX
1k Units Tape and Reel
3k Units Tape and Reel
1k Units Tape and Reel
3.5k Units Tape and Reel
94 Units/Rail
5-Pin SOT23
AH4A
AJ4A
MF05A
8-Pin MSOP
MUA08A
MTC14
14-Pin TSSOP
LMP7709MT
2.5k Units Tape and Reel
7
www.national.com
Typical Performance Characteristics Unless otherwise specified, TA = 25°C, VCM = VS/2, RL > 10 kΩ
connected to (V++V-)/2
Offset Voltage Distribution
Offset Voltage Distribution
Offset Voltage Distribution
TCVOS Distribution
TCVOS Distribution
TCVOS Distribution
20203736
20203737
20203738
20203741
20203742
20203743
www.national.com
8
Offset Voltage vs. Temperature
Offset Voltage vs. Supply Voltage
Offset Voltage vs. VCM
CMRR vs. Frequency
Offset Voltage vs. VCM
Offset Voltage vs. VCM
20203706
20203750
20203707
20203710
20203708
20203709
9
www.national.com
Input Bias Current vs. VCM
Input Bias Current vs. VCM
Input Bias Current vs. VCM
Input Bias Current vs. VCM
Input Bias Current vs. VCM
Input Bias Current vs. VCM
20203730
20203746
20203731
20203747
20203748
20203749
www.national.com
10
PSRR vs. Frequency
Supply Current vs. Supply Voltage (Per Channel)
20203745
20203711
Sinking Current vs. Supply Voltage
Sourcing Current vs. Supply Voltage
20203713
20203712
Output Voltage vs. Output Current
Slew Rate vs. Supply Voltage
20203717
20203716
11
www.national.com
Open Loop Frequency Response
Open Loop Frequency Response
20203714
20203715
Small Signal Step Response, AV = 10
Large Signal Step Response, AV = 10
20203719
20203718
Small Signal Step Response, AV = 100
Large Signal Step Response, AV = 100
20203726
20203720
www.national.com
12
Input Voltage Noise vs. Frequency
Open Loop Gain vs. Output Voltage Swing
20203727
20203752
Output Swing High vs. Supply Voltage
Output Swing Low vs. Supply Voltage
20203733
20203735
Output Swing High vs. Supply Voltage
Output Swing Low vs. Supply Voltage
20203732
20203734
13
www.national.com
THD+N vs. Frequency
THD+N vs. Output Voltage
20203728
20203729
Crosstalk Rejection Ratio vs. Frequency
(LMP7708/LMP7709)
20203753
www.national.com
14
Application Information
LMP7707/LMP7708/LMP7709
The LMP7707/LMP7708/LMP7709 devices are single, dual
and quad low offset voltage, rail-to-rail input and output pre-
cision amplifiers each with a CMOS input stage and the wide
supply voltage range of 2.7V to 12V. The LMP7707/
LMP7708/LMP7709 have a very low input bias current of only
±200 fA at room temperature.
The wide supply voltage range of 2.7V to 12V over the ex-
tensive temperature range of −40°C to 125°C makes either
the LMP7707, LMP7708 or LMP7709 an excellent choice for
low voltage precision applications with extensive temperature
requirements.
20203721
FIGURE 1. Isolating Capacitive Load
INPUT CAPACITANCE
The LMP7707/LMP7708/LMP7709 have only ±37 µV of typ-
ical input referred offset voltage and this offset is guaranteed
to be less than ±500 µV for the single and ±520 µV for the
dual and quad over temperature. This minimal offset voltage
allows more accurate signal detection and amplification in
precision applications.
CMOS input stages inherently have low input bias current and
higher input referred voltage noise. The LMP7707/LMP7708/
LMP7709 enhances this performance by having the low input
bias current of only ±200 fA, as well as a very low input re-
ferred voltage noise of 9 nV/
. In order to achieve this a
large input stage has been used. This large input stage in-
creases the input capacitance of the LMP7707/LMP7708/
LMP7709. The typical value of this input capacitance, CIN, for
the LMP7707/LMP7708/LMP7709 is 25 pF. The input capac-
itance will interact with other impedances such as gain and
feedback resistors, which are seen on the inputs of the am-
plifier, to form a pole. This pole will have little or no effect on
the output of the amplifier at low frequencies and DC condi-
tions, but will play a bigger role as the frequency increases.
At higher frequencies, the presence of this pole will decrease
phase margin and will also cause gain peaking. In order to
compensate for the input capacitance, care must be taken in
choosing the feedback resistors. In addition to being selective
in picking values for the feedback resistor, a capacitor can be
added to the feedback path to increase stability.
The low input bias current of only ±200 fA along with the low
input referred voltage noise of 9 nV/
give the LMP7707/
LMP7708/LMP7709 superior qualities for use in sensor ap-
plications. Lower levels of noise introduced by the amplifier
mean better signal fidelity and a higher signal-to-noise ratio.
The LMP7707/LMP7708/LMP7709 are stable for a gain of 6
or higher. With proper compensation though, the LMP7707,
LMP7708 or LMP7709 can be operational at a gain of ±1 and
still maintain much faster slew rates than comparable fully
compensated amplifiers. The increase in bandwidth and slew
rate is obtained without any additional power consumption.
National Semiconductor is heavily committed to precision
amplifiers and the market segment they serve. Technical sup-
port and extensive characterization data is available for sen-
sitive applications or applications with a constrained error
budget.
The LMP7707 is offered in the space saving 5-Pin SOT23
package, the LMP7708 comes in the 8-pin MSOP and the
LMP7709 is offered in the 14-Pin TSSOP package. These
small packages are ideal solutions for area constrained PC
boards and portable electronics.
CAPACITIVE LOAD
The LMP7707/LMP7708/LMP7709 devices can each be con-
nected as a non-inverting voltage follower. This configuration
is the most sensitive to capacitive loading.
20203744
The combination of a capacitive load placed on the output of
an amplifier along with the amplifier’s output impedance cre-
ates a phase lag which in turn reduces the phase margin of
the amplifier. If the phase margin is significantly reduced, the
response will be either underdamped or it will oscillate.
FIGURE 2. Compensating for Input Capacitance
Using this compensation method will have an impact on the
high frequency gain of the op amp, due to the frequency de-
pendent feedback of this amplifier. Low gain settings can,
again, introduce instability issues.
In order to drive heavier capacitive loads, an isolation resistor,
RISO, as shown in the circuit in Figure 1 should be used. By
using this isolation resistor, the capacitive load is isolated
from the amplifier’s output, and hence, the pole caused by
CL is no longer in the feedback loop. The larger the value of
RISO, the more stable the output voltage will be. If values of
RISO are sufficiently large, the feedback loop will be stable,
independent of the value of CL. However, larger values of
RISO result in reduced output swing and reduced output cur-
rent drive.
DIODES BETWEEN THE INPUTS
The LMP7707/LMP7708/LMP7709 have a set of anti-parallel
diodes between the input pins, as shown in Figure 3. 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 one diode voltage drop might damage the diodes.
The differential signal between the inputs needs to be limited
to ±300 mV or the input current needs to be limited to ±10 mA.
Exceeding these limits will damage the part.
15
www.national.com
HIGH IMPEDANCE SENSOR INTERFACE
Many sensors have high source impedances that may range
up to 10 MΩ. The output signal of sensors often needs to be
amplified or otherwise conditioned by means of an amplifier.
The input bias current of this amplifier can load the sensor’s
output and cause a voltage drop across the source resistance
as shown in Figure 5, where VIN + = VS – IBIAS*RS
The last term, IBIAS*RS, shows the voltage drop across RS. To
prevent errors introduced to the system due to this voltage,
an op amp with very low input bias current must be used with
high impedance sensors. This is to keep the error contribution
by IBIAS*RS less than the input voltage noise of the amplifier,
so that it will not become the dominant noise factor. The
LMP7707/LMP7708/LMP7709 have very low input bias cur-
rent, typically 200 fA.
20203725
FIGURE 3. Input of the LMP7707
TOTAL NOISE CONTRIBUTION
The LMP7707/LMP7708/LMP7709 have very low input bias
current, very low input current noise and very low input volt-
age noise. As a result, these amplifiers are ideal choices for
circuits with high impedance sensor applications.
Figure 4 shows the typical input noise of the LMP7707/
LMP7708/LMP7709 as a function of source resistance. The
total noise at the input can be calculated using Equation 1.
(1)
Where:
eni is the total noise on the input.
en denotes the input referred voltage noise
ei is the voltage drop across source resistance due to input
referred current noise or ei = RS * in
20203759
et is the thermal noise of the source resistance
FIGURE 5. Noise Due to IBIAS
The input current noise of the LMP7707/LMP7708/LMP7709
is so low that it will not become the dominant factor in the total
noise unless source resistance exceeds 300 MΩ, which is an
unrealistically high value.
USAGE OF DECOMPENSATED AMPLIFIERS
This section discusses the differences between compensated
and decompensated op amps and presents the advantages
of decompensated amplifiers. In high gain applications de-
compensated amplifiers can be used without any changes
compared to standard amplifiers. However, for low gain ap-
plications special frequency compensation measures have to
be taken to ensure stabilitiy.
As is evident in Figure 4, at lower RS values, the total noise
is dominated by the amplifier’s input voltage noise. Once RS
is larger than a few kilo-Ohms, then the dominant noise factor
becomes the thermal noise of RS. As mentioned before, the
current noise will not be the dominant noise factor for any
practical application.
Feedback circuit theory is discussed in detail, in particular as
it applies to decompensated amplifiers. Bode plots are pre-
sented for a graphical explanation of stability analysis. Two
solutions are given for creating a feedback network for de-
compensated amplifiers when relatively low gains are re-
quired: A simple resistive feedback network and a more
advanced frequency dependent feedback network with im-
proved noise performance. Finally, a design example is pre-
sented resulting in a practical application. The results are
compared to fully compensated amplifiers (National Semi-
conductors LMP7701/LMP7702/LMP7704).
COMPENSATED AMPLIFIERS
A (fully) compensated op amp is designed to operate with
good stability down to gains of ±1. For this reason, the com-
pensated op amp is also called a unity gain stable op amp.
Figure 6 shows the Open Loop Response of a compensated
amplifier.
20203758
FIGURE 4. Total Input Noise
www.national.com
16
202037aa
202037a9
FIGURE 6. Open Loop Frequency Response
FIGURE 7. Open Loop Frequency Response
Compensated Amplifier (LMP7701)
Decompensated Amplifier (LMP7707)
This amplifier is unity gain stable, because the phase shift is
still < 180°, when the gain crosses 0 dB (unity gain).
As shown in Figure 7, the reduced internal compensation
moves the first pole to higher frequencies. The second open
loop pole for the LMP7707/LMP7708/LMP7709 occurs at 4
MHz. The extrapolated unity gain (see dashed line in Figure
7) occurs at 14 MHz. An ideal two pole system would give a
phase margin of > 45° at the location of the second pole. Un-
fortunately, the LMP7707/LMP7708/LMP7709 have parasitic
poles close to the second pole, giving a phase margin closer
to 0°. The LMP7707/LMP7708/LMP7709 can be used at fre-
quencies where the phase margin is > 45°. The frequency
where the phase margin is 45° is at 2.4 MHz. The corre-
sponding value of the open loop gain (also called GMIN) is 6
times.
Stability can be expressed in two different ways:
Phase Margin
This is the phase difference between the ac-
tual phase shift and 180°, at the point where
the gain is 0 dB.
Gain Margin
This is the gain difference relative to 0 dB, at
the frequency where the phase shift crosses
the 180°.
The amplifier is supposed to be used with negative feedback
but a phase shift of 180° will turn the negative feedback into
positive feedback, resulting in oscillations. A phase shift of
180° is not a problem when the gain is smaller than 0 dB, so
the critical point for stability is 180° phase shift at 0 dB gain.
The gain margin and phase margin express the margin en-
hancing overall stability between the amplifiers response and
this critical point.
Stability has only to do with the loop gain and not with the
forward gain (G) of the op amp. For high gains, the feedback
network is attenuating and this reduces the loop gain; there-
fore the op amp will be stable for G > GMIN and no special
measures are required. For low gains the feedback network
attenuation may not be sufficient to ensure loop stability for a
decompensated amplifier. However, with an external com-
pensation network decompensated amplifiers can still be
made stable while maintaining their advantages over unity
gain stable amplifiers.
DECOMPENSATED AMPLIFIERS
Decompensated amplifiers, such as the LMP7707/LMP7708/
LMP7709, are designed to maximize the bandwidth and slew
rate without any additional power consumption over the unity
gain stable op amp. That is, a decompensated op amp has a
higher bandwidth to power ratio than an equivalent compen-
sated op amp. Compared with the unity gain stable amplifier,
the decompensated version has the following advantages:
EXTERNAL COMPENSATION FOR GAINS LOWER THAN
GMIN.
This section explains how decompensated amplifiers can be
used in configurations requiring a gain lower than GMIN. In the
next sections the concept of the feedback factor is introduced.
Subsequently, an explanation is given how stability can be
determined using the frequency response curve of the op amp
together with the feedback factor. Using the circuit theory, it
will be explained how decompensated amplifiers can be sta-
bilized at lower gains.
1. A wider closed loop bandwidth.
2. Better slew rate due to reduced compensation
capacitance within the op amp.
3. Better Full Power Bandwidth, given with Equation 2.
(2)
FEEDBACK THEORY
Figure 7 shows the frequency response of the decompensat-
ed amplifier.
Stability issues can be analyzed by verifying the loop gain
function GF, where G is the open loop gain of the amplifier
and F is the feedback factor of the feedback circuit.
The feedback function (F) of arbitrary electronic circuits, as
shown in Figure 8, is defined as the ratio of the input and out-
put signal of the same circuit.
17
www.national.com
(20 dB). This is shown as the dashed line in Figure 9. The
resistor choice of RF = R1 = 2 kΩ makes the inverse feedback
equal 2 V/V (6 dB), shown in Figure 9 as the solid line. The
intercept of G and 1/F represents the frequency for which the
loop gain is identical to 1 (0 dB). Consequently, the total phase
shift at the frequency of this intercept determines the phase
margin and the overall system stability. In this system exam-
ple 1/F crosses the open loop gain at a frequency which is
larger than the frequency where GMIN occurs, therefore this
system has less than 45° phase margin and is most likely in-
stable.
20203796
FIGURE 8. Op Amp with Resistive Feedback. (a) Non-
inverting (b) Inverting
The feedback function for a three-terminal op amp as shown
in Figure 8 is the feedback voltage VA – VB across the op amp
input terminals relative to the op amp output voltage, VOUT
.
That is
(3)
GRAPHICAL EXPLANATION OF STABILITY ANALYSIS
Stability issues can be observed by verifying the closed loop
gain function GF. In the frequencies of interest, the open loop
gain (G) of the amplifier is a number larger than 1 and there-
fore positive in dB. The feedback factor (F) of the feedback
circuit is an attenuation and therefore negative in dB. For cal-
culating the closed loop gain GF in dB we can add the values
of G and F (both in dB).
202037a2
FIGURE 9. 1/F for RF = R1 and Open Loop Gain Plot
One practical approach to stabilizing the system, is to assign
a value to the feedback factor F such that the remaining loop
gain GF equals one (unity gain) at the frequency of GMIN. This
realizes a phase margin of 45° or greater. This results in the
following requirement for stability: 1/F > GMIN. The inverse
feedback factor 1/F is constant over frequency and should
intercept the open loop gain at a value in dB that is greater
RESISTIVE COMPENSATION
A straightforward way to achieve a stable amplifier configu-
ration is to add a resistor RC between the inverting and the
non-inverting inputs as shown in Figure 10.
than or equal to GMIN
.
The inverse feedback factor for both configurations shown in
Figure 8, is given by:
(4)
20203797
The closed loop gain for the non-inverting configuration (a) is:
FIGURE 10. Op Amp with Compensation Resistor
between Inputs
(5)
This additional resistor RC will not affect the closed loop gain
of the amplifier but it will have positive impact on the feedback
network.
The closed loop gain for the inverting configuration (b) is:
The inverse feedback function of this circuit is:
(6)
For stable operation the phase margin must be equal to or
greater than 45° . The corresponding closed loop gain GMIN
for a non-inverting configuration, is
,
(9)
Proper selection of the value of RC results in the shifting of
the 1/F function to GMIN or greater, thus fulfilling the condition
for circuit stability. The compensation technique of reducing
the loop gain may be used to stabilize the circuit for the values
given in the previous example, that is GMIN = 20 dB and RF =
R1 = 2 kΩ. A resistor value of 250 Ω applied between the
amplifier inputs shifts the 1/F curve to the value GMIN (20 dB)
as shown by the dashed line in Figure 11. This results in
overall stability for the circuit. This figure shows a combination
of the open and closed loop gain and the inverse feedback
function.
(7)
For an inverting configuration:
(8)
If R1 and RF and are chosen so that the closed loop gain is
lower than the minimum gain required for stability, then 1/F
intersects the open loop gain curve for a value that is lower
than GMIN. For example, assume the GMIN is equal to 10 V/V
www.national.com
18
This example, represented by Figure 8 and Figure 9, is gener-
ic in the sense that the GMIN as specified did not distinguish
between inverting and non-inverting configurations.
By inspection of Equation 12, RC does not affect the ideal
closed loop gain. In this example where RF = R1, the closed
loop gain remains at 6 dB as long as GF >> 1. The closed
loop gain curve is shown as the solid line in Figure 11.
The addition of RC affects the circuit in the following ways:
1.
1/F is moved to a higher gain, resulting in overall system
stability.
However, adding RC results in reduced loop gain and in-
creased noise gain. The noise gain is defined as the inverse
of the feedback factor, F. The noise gain is the gain from the
amplifier input referred noise to the output. In effect, loop gain
is traded for stability.
2.
The ideal closed loop gain retains the same value as the
circuit without the compensation resistor RC.
LEAD-LAG COMPENSATION
This section presents a more advanced compensation tech-
nique that can be used to stabilize amplifiers. The increased
noise gain of the prior circuit is prevented by reducing the low
frequency attenuation of the feedback circuit. This compen-
sation method is called Lead-Lag compensation. Lead-lag
compensation components will be analyzed and a design ex-
ample using this procedure will be discussed.
202037a3
FIGURE 11. Compensation with Reduced Loop Gain
The technique of reducing loop gain to stabilize a decompen-
sated op amp circuit will be illustrated using the non-inverting
input configuration shown in Figure 12.
The feedback function in a lead-lag compensation circuit is
shaped using a resistor and a capacitor. They are chosen in
a way that ensures sufficient phase margin.
Figure 13 shows a Bode plot containing: the open loop gain
of the decompensated amplifier, a feedback function without
compensation and a feedback function with lead-lag com-
pensation.
20203798
FIGURE 12. Closed Loop Gain Analysis with RC
The effect of the choice of resistor RC in Figure 12 on the
closed loop gain can be analyzed in the following manner:
Assume the voltage at the inverting input of the op amp is
VX. Then,
(10)
Where G is the open loop gain of the op amp.
202037a5
FIGURE 13. Bode Plot of Open Loop gain G and 1/F with
and without Lead-Lag Compensation
(11)
Combining Equation 10, Equation 11, and Equation 9 pro-
duces the following equation for closed loop gain,
The shaped feedback function presented in Figure 13 can be
realized using the amplifier configuration in Figure 14. Note
that resistor RP is only used for compensation of the input
voltage caused by the IBIAS current. RP can be used to intro-
duce more freedom for calculating the lead-lag components.
This will be discussed later in this section.
(12)
19
www.national.com
Note that the constraint 1/F ≥ Gmin needs to be satisfied only
in the vicinity of the intersection of G and 1/F; 1/F can be
shaped elsewhere as needed. Two rules must be satisfied in
order to maintain adequate phase margin.
Rule 1
The plot of 1/F should intersect with the plot of the
open loop gain at a value larger than GMIN. At that
point, the open loop gain G has a phase margin of
45°.
The location f2 in Figure 15 illustrates the proper in-
tersection point for the LMP7707/LMP7708/
LMP7709 using the circuit of Figure 14. The inter-
section of G and 1/F at the op amp's second pole
location is the 45° phase margin reference point.
Rule 2
The 1/F pole (see Figure 15) should be positioned at
the frequency that is at least one decade below the
intersection point f2 of 1/F and G. This positioning
takes full advantage of the 90° of phase lead brought
about by the 1/F pole. This additional phase lead ac-
companies the increase in magnitude of 1/F ob-
served at frequencies greater than the 1/F pole.
20203765
FIGURE 14. LMP7707 with Lead-Lag Compensation for
Inverting Configuration
The inverse feedback factor of the circuit in Figure 14 is:
The resulting system has approximately 45° of phase margin,
based upon the fact that the open loop gain's dominant pole
and the second pole are more than one decade apart and that
the open loop gain has no other pole within one decade of its
intersection point with 1/F. If there is a third pole in the open
loop gain G at a frequency greater than f2 and if it occurs less
than a decade above that frequency, then there will be an
effect on phase margin.
(13)
The pole of the inverse feedback function is located at:
(14)
The zero of the inverse feedback function is located at:
DESIGN EXAMPLE
The input lead-lag compensation method can be applied to
an application using the LMP7707, LMP7708 or LMP7709 in
an inverting configuration, as shown in Figure 14.
(15)
The low frequency inverse feedback factor is given by:
(16)
The high frequency inverse feedback factor is given by:
(17)
From these formulas, we can tell that
1. The 1/F's zero is located at a lower frequency compared
to 1/F's pole.
2. The intersection point of 1/F and the open loop gain G is
determined by the choice of resistor values for RP and
RC if the values of R1 and RF are set before
compensation.
3. This procedure results in the creation of a pole-zero pair,
the positions of which are interdependent.
202037ab
4. This pole-zero pair is used to:
FIGURE 15. LMP7707 Open Loop Gain and 1/F Lead-Lag
Feedback Network.
Raise the 1/F value to a greater value in the region
immediately to the left of its intercept with the A
function in order to meet the Gmin requirement.
—
Figure 15 shows that GMIN = 16 dB and f2 (intersection point)
= 2.4 MHz.
Achieve the preceding with no additional loop phase
delay.
—
A gain of 6 dB (or a magnitude of –1) is well below the
LMP7707’s GMIN. Without external lead-lag compensation,
the inverse feedback factor is found using Equation 4 which
applies to both inverting and non-inverting configurations.
Unity gain implementation for the inverting configuration
means RF = R1, and 1/F = 2 (6 dB).
5. The location of the 1/F zero is determined by the following
conditions:
The value of 1/F at low frequency.
The value of 1/F at the intersection point.
The location of 1/F pole.
—
—
—
www.national.com
20
Procedure:
Figure 16 shows the results of the compensation of the
LMP7707.
The compensation circuit shown in Figure 14 is implemented.
The inverse feedback function is shaped by the solid line in
Figure 15. The 1/F plot is 6 dB at low frequencies. At higher
frequencies, it is made to intersect the loop gain G at fre-
quency f2 with gain amplitude of 16 dB (GMIN), which equals
a magnitude of six times. This follows the recommendations
in Rule 1. The 1/F pole fp is set one decade below the inter-
section point (f2 = 2.4 MHz) as given in Rule 2, and results in
a frequency fp = 240 kHz. The next steps should be taken to
calculate the values of the compensation components:
Step 1)
Set 1/F equal to GMIN using Equation 17. This gives
a value for resistor RC.
Step 2)
Set the 1/F pole one decade below the intersection
point using Equation 14. This gives a value for ca-
pacitor C.
This method uses bode plot approximation. Some fine-tuning
may be needed to get the best results.
Calculations:
202037a7
As described in Step 1, use Equation 17.
FIGURE 16. Bench Results for Lead- Lag Compensation
The top waveform shows the output response of a uncom-
pensated LMP7707 using no external compensation compo-
nents. This trace shows ringing and is unstable (as expected).
The middle waveform is the response of a compensated
LMP7707 using the compensation components calculated
with the described procedure. The response is reasonably
well behaved. The bottom waveform shows the response of
an overcompensated LMP7707.
(18)
Now substitute RF/R1 = 1 into the equation above since this
is a unity gain, inverting amplifier, then
(19)
According to Step 2 use Equation 14
Finally, Figure 17 compares the step response of the com-
pensated LMP7707 to that of the unity gain stable LMP7701.
The increase in dynamic performance is clear.
(20)
which leads to:
(21)
Choose a value of RF that is below 2 kΩ, in order to minimize
the possibility of shunt capacitance across high value resis-
tors producing a negative effect on high frequency operation.
If RF = R1 = 1 kΩ, then RF // R1 = 500 Ω. For simplicity, choose
RP = 0 Ω . The value of RC is derived from Equation 19 and
has a value of RC = 250 Ω. This is not a standard value. A
value of RC = 330 Ω is a first choice (using 10% tolerance
components).
The value of capacitor C is 2.2 nF. This value is significantly
higher than the parasitic capacitances associated with pas-
sive components and board layout, and is therefore a good
solution.
202037a6
Bench results:
For bench evaluation the LMP7707 in an inverting configura-
tion has been verified under three different conditions:
FIGURE 17. Bench Results for Comparison of LMP7701
and LMP7707
•
•
•
Uncompensated.
Lead-lag compensation resulting in a phase margin of 45°.
Lead lag overcompensation resulting in a phase margin
larger than 45°.
The application of input lead-lag compensation to a decom-
pensated op amp enables the realization of circuit gains of
less than the minimum specified by the manufacturer. This is
accomplished while retaining the advantageous speed versus
power characteristic of decompensated op amps.
The calculated components for these three conditions are
Condition
RC
C
Uncompensated
Compensated
NA
NA
2.2 nF
3.3 nF
330 Ω
240 Ω
Overcompensated
21
www.national.com
Physical Dimensions inches (millimeters) unless otherwise noted
5-Pin SOT23
NS Package Number MF05A
8-Pin MSOP
NS Package Number MUA08A
www.national.com
22
14-Pin TSSOP
NS Package Number MTC14
23
www.national.com
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
National Semiconductor Europe
Customer Support Center
Fax: +49 (0) 180-530-85-86
National Semiconductor Asia
Pacific Customer Support Center
Email: ap.support@nsc.com
National Semiconductor Japan
Customer Support Center
Fax: 81-3-5639-7507
Email:
new.feedback@nsc.com
Tel: 1-800-272-9959
Email: europe.support@nsc.com
Deutsch Tel: +49 (0) 69 9508 6208
English Tel: +49 (0) 870 24 0 2171
Français Tel: +33 (0) 1 41 91 8790
Email: jpn.feedback@nsc.com
Tel: 81-3-5639-7560
www.national.com
相关型号：
©2020 ICPDF网 联系我们和版权申明