FEATURES
Single Supply Operation: +2.5 V to +6 V
High Output Current: ⴞ250 mA
Extremely Low Shutdown Supply Current: 100 nA
Low Supply Current: 750 A/Amp
Wide Bandwidth: 3 MHz
Slew Rate: 5 V/s
No Phase Reversal
Very Low Input Bias Current
High Impedance Outputs When in Shutdown Mode
Unity Gain Stable
APPLICATIONS
Mobile Communication Handset Audio
PC Audio
PCMCIA/Modem Line Driving
Battery Powered Instrumentation
Data Acquisition
ASIC Input or Output Amplifier
LCD Display Reference Level Driver
GENERAL DESCRIPTION
The AD8591, AD8592 and AD8594 are single, dual and quad
rail-to-rail input and output single supply amplifiers featuring
250 mA output drive current and a power saving shutdown
mode. The AD8592 includes an independent shutdown function for each amplifier. When both amplifiers are in shutdown
mode the total supply current is reduced to less than 1 µA. The
AD8591 and AD8594 include a single master shutdown func-
tion that reduces total supply current to less than 1 µA. All
amplifier outputs are in a high impedance state when in shutdown mode.
These amplifiers have very low input bias currents, making them
suitable for integrators and diode amplification. Outputs are
stable with virtually any capacitive load. Supply current is less
than 750 µA per amplifier in active mode.
Applications for these amplifiers include audio amplification for
portable computers, portable phone headsets, sound ports, sound
cards and set-top boxes. The AD859x family is capable of driving
heavy capacitive loads such as LCD panel reference levels.
The ability to swing rail-to-rail at both the input and output
enables designers to buffer CMOS DACs, ASICs and other
wide output swing devices in single supply systems.
The AD8591, AD8592 and AD8594 are specified over the indus-
trial (–40°C to +85°C) temperature range. The AD8591, single,
is available in the tiny 6-lead SOT package. The AD8592, dual, is
available in the 10-lead µSOIC surface mount package. The
AD8594, quad, is available in 16-lead narrow SOIC and 16-lead
TSSOP packages.
REV. A
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
Lead Temperature Range (Soldering, 60 sec) . . . . . . . +300°C
NOTES
1
Stresses above those listed under Absolute Maximum Ratings may cause perma-
nent damage to the device. This is a stress rating only; functional operation of the
device at these or any other conditions above those listed in the operational sections
of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability.
2
For supplies less than ±5 V the differential input voltage is limited to the supplies.
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection. Although
the AD8591/AD8592/AD8594 features proprietary ESD protection circuitry, permanent damage
may occur on devices subjected to high energy electrostatic discharges. Therefore, proper ESD
precautions are recommended to avoid performance degradation or loss of functionality.
θJA is specified for worst case conditions, i.e., θ
for surface mount packages.
ORDERING GUIDE
TemperaturePackagePackage
ModelRangeDescriptionOption
AD8591ART –40°C to +85°C6-Lead SOT-23RT-6
AD8592ARM –40°C to +85°C10-Lead µSOICRM-10
AD8594AR–40°C to +85°C16-Lead SOICR-16A
AD8594ARU –40°C to +85°C16-Lead TSSOPRU-16
1
JA
JC
is specified for device in socket
JA
Units
Typical Performance Characteristics
1k
VS = +2.7V
= +258C
T
A
100
10
1
∆OUTPUT VOLTAGE – mV
0.1
SOURCE
0.1110
LOAD CURRENT – mA
SINK
100
1k0.01
Figure 1. Output Voltage to Supply
Rail vs. Load Current
10k
VS = +5V
= +258C
T
A
1k
100
10
∆OUTPUT VOLTAGE – mV
1
0.1
Figure 2. Output Voltage to Supply
Rail vs. Load Current
SOURCE
0.1110
LOAD CURRENT – mA
SINK
100
0.90
0.85
0.80
0.75
0.70
0.65
0.60
0.55
SUPPLY CURRENT/AMPLIFIER – mA
1k0.01
0.50
240 220
0 20406080
TEMPERATURE – 8C
VS = +5V
VS = +2.7V
100
Figure 3. Supply Current per
Amplifier vs. Temperature
–4–
REV. A
Page 5
AD8591/AD8592/AD8594
g
0.8
TA = +258C
0.7
0.6
0.5
0.4
0.3
0.2
0.1
SUPPLY CURRENT/AMPLIFIER – mA
0
0.751.253
1.752.252.75
SUPPLY VOLTAGE – 6Volts
Figure 4. Supply Current per
Amplifier vs. Supply Voltage
4
VS = +2.7V, +5V
V
= VS/2
CM
3
2
1
0
21
INPUT OFFSET CURRENT – pA
22
23
24
25
26
27
INPUT OFFSET VOLTAGE – mV
28
250 235
215
5254565
TEMPERATURE – 8C
VS = +5V
V
= +2.5V
CM
Figure 5. Input Offset Voltage vs.
Temperature
8
VS = +5V
7
T
= +258C
A
6
5
4
3
INPUT BIAS CURRENT – pA
2
8
VS = +2.7V, +5V
V
= VS/2
CM
7
6
5
4
INPUT BIAS CURRENT – pA
3
2
85
250 235
215
5254565
TEMPERATURE – 8C
85
Figure 6. Input Bias Current vs.
Temperature
80
60
40
20
0
GAIN – dB
VS = +2.7V
R
= NO LOAD
L
T
= +258C
A
45
90
135
180
PHASE SHIFT – Degrees
22
250 235
215
5254565
TEMPERATURE – 8C
85
Figure 7. Input Offset Current vs.
Temperature
80
60
40
20
0
GAIN – dB
1k10k100M
100k1M10M
FREQUENCY – Hz
VS = +5V
R
= NO LOAD
L
T
= +258C
A
Figure 10. Open-Loop Gain and
Phase vs. Frequency
45
90
rees
135
180
PHASE SHIFT – De
1
015
COMMON-MODE VOLTAGE – Volts
234
Figure 8. Input Bias Current vs.
Common-Mode Voltage
5
4
3
2
OUTPUT SWING – V p-p
1
0
1k10k10M
FREQUENCY – Hz
VS = +2.7V
R
= 2kV
L
T
= +258C
A
V
= 2.5V p-p
IN
100k1M
Figure 11. Closed-Loop Output
Voltage Swing vs. Frequency
1k10k100M
100k1M10M
FREQUENCY – Hz
Figure 9. Open-Loop Gain and Phase
vs. Frequency
5
VS = +5V
R
= 2kV
L
4
T
= +258C
A
V
= 4.9V p-p
IN
3
2
OUTPUT SWING – V p-p
1
0
1k10k10M
100k1M
FREQUENCY – Hz
Figure 12. Closed-Loop Output
Voltage Swing vs. Frequency
–5–REV. A
Page 6
AD8591/AD8592/AD8594
200
VS = +5V
180
T
= +258C
A
160
140
120
100
80
IMPEDANCE – V
60
40
20
0
1k10k100M
AV = 10
AV = 1
100k1M10M
FREQUENCY – Hz
Figure 13. Closed-Loop Output
Impedance vs. Frequency
140
120
100
80
60
40
20
PSRR – dB
0
220
240
260
100
+PSRR
1k10k
2PSRR
FREQUENCY – Hz
VS = +5V
T
= +258C
A
100k1M10M
Figure 16. Power Supply Rejection
Ratio vs. Frequency
110
VS = +5V
T
= +258C
A
100
90
80
CMRR – dB
70
60
50
1k10k10M
100k1M
FREQUENCY – Hz
Figure 14. Common-Mode Rejection
Ratio vs. Frequency
60
VS = +2.5V
R
= 2kV
L
50
T
= +258C
A
40
30
20
10
SMALL SIGNAL OVERSHOOT – %
0
1010010k1k
CAPACITANCE – pF
+OS
2OS
Figure 17. Small Signal Overshoot
vs. Load Capacitance
140
120
100
80
60
40
20
PSRR – dB
0
220
240
260
100
+PSRR
2PSRR
1k10k
FREQUENCY – Hz
VS = +2.5V
T
= +258C
A
100k1M10M
Figure 15. Power Supply Rejection
Ratio vs. Frequency
60
VS = +5V
R
= 2kV
L
50
T
= +258C
A
40
30
20
10
SMALL SIGNAL OVERSHOOT – %
0
1010010k
2OS
+OS
1k
CAPACITANCE – pF
Figure 18. Small Signal Overshoot
vs. Load Capacitance
0V
20mV/DIV
VS = 61.35V
V
= 650mV
IN
A
= 1
V
R
= 2kV
L
C
= 300pF
L
T
= +258C
A
500 ns/DIV
Figure 19. Small Signal Transient
Response
0V
20mV/DIV
VS = 62.5V
V
= 650mV
IN
= 1
A
V
RL = 2kV
= 300pF
C
L
= +258C
T
A
500 ns/DIV
Figure 20. Small Signal Transient
Response
–6–
VS = 61.35V
= 1
A
100
90
10
0%
500mV
R
T
V
L
A
500ns
= 2kV
= +258C
Figure 21. Large Signal Transient
Response
REV. A
Page 7
AD8591/AD8592/AD8594
VS = 62.5V
A
= 1
100
90
10
0%
500mV
V
R
L
T
A
500ns
= 2kV
= +258C
Figure 22. Large Signal Transient
Response
100
90
100mV/DIV
10
0%
VS = +5V
A
= 1000
V
T
= +258C
A
FREQUENCY = 1kHz
1V
100
90
10
0%
1V
VS = 62.5V
= 1
A
V
T
= +258C
A
10ms
Figure 23. No Phase Reversal
VS = +5V
100
AV= 1000
90
TA = +258 C
FREQUENCY = 10kHz
V/DIV
m
200
10
0%
1
0.1
CURRENT NOISE DENSITY – pA/ Hz
0.01
10100100k
FREQUENCY – Hz
VS = +5V
= +258C
T
A
1k
10k
Figure 24. Current Noise Density vs.
Frequency
VS = +2.7V
= +1.35V
V
500
400
300
200
QUANTITY – Amplifiers
100
CM
T
A
= +258C
MARKER 41mV/ Hz
Figure 25. Voltage Noise Density vs.
Frequency
MARKER 25.9 mV/ Hz
Figure 26. Voltage Noise Density vs.
Frequency
VS = +5V
V
= +2.5V
500
400
300
200
QUANTITY – Amplifiers
100
–12
–10 –8 –6 –4 –2 0 2 4
INPUT OFFSET VOLTAGE – mV
CM
T
A
= +258C
Figure 28. Input Offset Voltage
Distribution
–12
–10 –8 –6 –4 –2 02 4
INPUT OFFSET VOLTAGE – mV
Figure 27. Input Offset Voltage
Distribution
–7–REV. A
Page 8
AD8591/AD8592/AD8594
AD8591/AD8592/AD8594 APPLICATION SECTION
Theory of Operation
The AD859x family of amplifiers are all CMOS, high output drive,
rail-to-rail input and output single supply amplifiers designed for
low cost and high output current drive. The parts include a power
saving shutdown function making the AD8591/AD8592/AD8594
op amps ideal for portable multimedia and telecom applications.
Figure 29 shows the simplified schematic for an AD8591/AD8592/
AD8594 amplifier. Two input differential pairs, consisting of an
n-channel pair (M1-M2) and a p-channel pair (M3-M4), provide
a rail-to-rail input common-mode range. The outputs of the input
differential pairs are combined in a compound folded-cascode
stage, which drives the input to a second differential pair gain
stage. The outputs of the second gain stage provide the gate voltage drive to the rail-to-rail output stage.
The rail-to-rail output stage consists of M15 and M16, which are
configured in a complementary common-source configuration.
As with any rail-to-rail output amplifier, the gain of the output
stage, and thus the open-loop gain of the amplifier, is dependent
on the load resistance. Also, the maximum output voltage swing
is directly proportional to the load current. The difference between the maximum output voltage to the supply rails, known as
the dropout voltage, is determined by the AD8591/AD8592/
AD8594 output transistors’ on-channel resistance. The output
dropout voltage is given in Figure 1 and Figure 2.
V+
*
100mA
M5
M8
M6
M7M10
M9
V–
*NOTE: ALL CURRENT SOURCES GO
TO 0 mA IN SHUTDOWN MODE
Although not shown on the simplified schematic, ESD protection diodes are connected from each input to each power supply
rail. These diodes are normally reverse biased, but will turn on
if either input voltage exceeds either supply rail by more than
+0.6 V. Should this condition occur, the input current should
be limited to less than ±5 mA. This can be done by placing a
resistor in series with the input(s). The minimum resistor value
should be:
Output Phase Reversal
The AD8591/AD8592/AD8594 are immune to output voltage
phase reversal with an input voltage within the supply voltages
of the device. However, if either of the device’s inputs exceeds
+0.6␣ V outside of the supply rails, the output could exhibit
phase reversal. This is due to the ESD protection diodes becoming forward biased, thus causing the polarity of the input
terminals of the device to switch.
The technique recommended in the Input Overvoltage Protection
section should be applied in applications where the possibility of
input voltages exceeding the supply voltages exists.
Output Short Circuit Protection
To achieve high output current drive and rail-to-rail performance,
the outputs of the AD859x family do not have internal short circuit protection circuitry. Although these amplifiers are designed to
sink or source as much as 250␣ mA of output current, shorting the
output directly to the positive supply could damage or destroy the
device. To protect the output stage, the maximum output current
should be limited to ±250␣ mA.
By placing a resistor in series with the output of the amplifier as
shown in Figure 30, the output current can be limited. The
minimum value for R
R
V
≥
X
250
For a +5 V single supply application, R
Because R
is inside the feedback loop, V
X
tradeoff in using R
under heavy output current loads. R
tive output impedance of the amplifier to R
can be found from Equation 2.
X
SY
mA
should be at least 20␣ Ω.
X
is not affected. The
is a slight reduction in output voltage swing
X
OUT
will also increase the effec-
X
+ RX, where RO is
O
(2)
the output impedance of the device.
+5V
R
V
IN
AD8592
20V
X
V
OUT
Figure 30. Output Short Circuit Protection
Power Dissipation
Although the AD859x family of amplifiers are able to provide
load currents of up to 250␣ mA, proper attention should be
given to not exceeding the maximum junction temperature for
the device. The equation for finding the junction temperature is
given as:
TPT
=×+θ
DISSAAJJ
(3)
Where TJ = AD859x junction temperature
P
= AD859x power dissipation
DISS
θ
= AD859x junction-to-ambient thermal resistance
JA
of the package; and
= The ambient temperature of the circuit
T
A
V
IN MAX
R
IN
,
≥
5
mA
(1)
–8–
REV. A
Page 9
AD8591/AD8592/AD8594
U1-A
R2
2kV
4
C1
100mF
+5V
1
10
2
3
5
+5V
V
DD
V
DD
LEFT
OUT
AD1881
(AC97)
RIGHT
OUT
V
SS
R4
20V
+5V
R1
100kV
7
8
6
9
R5
20V
C2
100mF
NOTE: ADDITIONAL PINS
OMITTED FOR CLARITY
U1-B
U1 = AD8592
R3
2kV
NC
28
35
36
In any application, the absolute maximum junction temperature
must be limited to +150°C. If this junction temperature is ex-
ceeded, the device could suffer premature failure. If the output
voltage and output current are in phase, for example, with a
purely resistive load, the power dissipated by the AD859x can
be found as:
PI VV
ISSSYOUTDLOAD
WhereI
=×
= AD859x output load current
LOAD
V
= AD859x supply voltage; and
SY
V
= The output voltage
OUT
–
()
(4)
By calculating the power dissipation of the device and using the
thermal resistance value for a given package type, the maximum
allowable ambient temperature for an application can be found
using Equation 3.
Capacitive Loading
The AD859x exhibits excellent capacitive load driving capabilities
and can drive up to 10 nF directly. Although the device is stable
with large capacitive loads, there is a decrease in amplifier bandwidth as the capacitive load increases. Figure 31 shows a graph of
the AD8592 unity gain bandwidth under various capacitive loads.
4
3.5
3
2.5
2
VS = 62.5V
= 1kV
R
L
T
= +258C
A
50mV
ONLY
100
90
10
0%
50mV
10ms
47nF LOAD
SNUBBER
IN CIRCUIT
Figure 33. Snubber Network Reduces Overshoot and
Ringing Caused from Driving Heavy Capacitive Loads
The optimum values for the snubber network should be determined
empirically based on the size of the capacitive load. Table I shows a
few sample snubber network values for a given load capacitance.
Table I. Snubber Networks for Large Capacitive Loads
Load CapacitanceSnubber Network
(CL)(R
, CS)
S
0.47 nF300 Ω, 0.1 µF
4.7 nF30 Ω, 1 µF
47 nF5 Ω, 1 µF
A PC-98 Compliant Headphone/Speaker Amplifier
Because of its high output current performance and shutdown
feature, the AD8592 makes an excellent amplifier for driving an
audio output jack in a computer application. Figure 34 shows
how the AD8592 can be interfaced with an AC97 codec to drive
headphones or speakers.
Figure 31. Unity Gain Bandwidth vs. Capacitive Load
When driving heavy capacitive loads directly from the AD859x
output, a snubber network can be used to improve transient
response. This network consists of a series R-C connected from
the amplifier’s output to ground, placing it in parallel with the
capacitive load. The configuration is shown in Figure 32. Although this network will not increase the bandwidth of the amplifier, it will significantly reduce the amount of overshoot, as
shown in Figure 33.
1.5
BANDWIDTH – MHz
1
0.5
0
0.011000.1
V
100mV p-p
IN
Figure 32. Configuration for Snubber Network to
Compensate for Capacitive Loads
CAPACITIVE LOAD – nF
AD8592
110
+5V
R
S
5V
C
S
1mF
C
L
47nF
V
OUT
Figure 34. A PC-98 Compliant Headphone/Line Out Amplifier
When headphones are plugged into the jack, the normalizing contacts disconnect from the audio contacts. This allows the voltage to
the AD8592 shutdown pins to be pulled up to +5 V, activating the
amplifiers. With no plug in the output jack, the shutdown voltage is
pulled to 100 mV through the R1 and R3␣ +␣ R5 voltage divider.
This powers the AD8592 down when it is not needed, saving
current from the power supply or battery.
–9–REV. A
Page 10
AD8591/AD8592/AD8594
(
)
If gain is required from the output amplifier, four additional
resistors should be added as shown in Figure 35. The gain of
the AD8592 can be set as:
R
A
LEFT
AD1881
(AC97)
RIGHT
NOTE: ADDITIONAL PINS
OMITTED FOR CLARITY
7
=
V
R
6
+5V
V
DD
38
V
DD
35
OUT
R6
10kV
27
V
REF
R6
10kV
36
OUT
V
SS
R7
20kV
+5V
10
2
U1-A
3
5
6
7
U1-B
8
1
4
100kV
9
R7
U1 = AD8592
20kV
R7
AV == +6dB WITH VALUES SHOWN
R6
C1
100mF
2kV
+5V
R1
C2
100mF
2kV
R4
20V
NC
R2
R5
20V
R3
(5)
Figure 35. A PC-98 Compliant Headphone/Line Out
Amplifier With Gain
Input coupling capacitors are not required for either circuit as
the reference voltage is supplied from the AD1881.
R4 and R5 help protect the AD8592 output in case the output
jack or headphone wires accidentally get shorted to ground.
The output coupling capacitors C1 and C2 block dc current
from the headphones and create a high-pass filter with a corner
frequency of:
f
=
dBL–3
214
1
π
CR R
+
()
(6)
Where RL is the resistance of the headphones.
A Combined Microphone and Speaker Amplifier for
Cellphone and Portable Headsets
The dual amplifiers in the AD8592 make an efficient design for
interfacing with a headset containing a microphone and speaker.
Figure 36 demonstrates a simple method for constructing an
interface to a codec.
R3
100kV
+5V
10
2
U1-A
1
4
3
5
6
7
U1-B
9
8
R5
10kV
OPTIONAL
R4
10kV
R6
10kV
TO
CODEC
V
REF
FROM CODEC
FROM CODEC
MONO OUT
(OR LEFT OUT)
(RIGHT OUT)
MIC + SPEAKER
JACK
U1 = AD8592
1kV
2.2kV
R7
+5V
R1
NC
C1
0.1mF
100kV
R8
C2
10mF
R2
10kV
+5V
Figure 36. A Speaker/Mic Headset Amplifier Circuit
U1-A is used as a microphone preamplifier, where the gain of
the preamplifier is set as R3/R2. R1 is used to bias an electret
microphone and C1 blocks any dc voltages from the amplifier.
U1-B is the speaker amplifier, and its gain is set at R5/R4. To
sum a stereo output, R6 should be added, equal in value to R4.
Using the same principle as described in the previous section,
the normalizing contact on the microphone/speaker jack can be
used to put the AD8592 into shutdown when the headset is not
plugged in. The AD8592 shutdown inputs can also be controlled with TTL or CMOS compatible logic, allowing microphone or speaker muting if desired.
An Inexpensive Sample-and-Hold Circuit
The independent shutdown control of each amplifier in the
AD8592 allows a degree of flexibility in circuit design. One particular application for which this feature is useful is in designing a
sample-and-hold circuit for data acquisition. Figure 37 shows a
schematic of a simple, yet extremely effective sample-and-hold
circuit using a single AD8592 and one capacitor.
C1
1nF
8
U1-B
6
7
U1 = AD8592
+5V
9
SAMPLE
AND HOLD
OUTPUT
+5V
2
10
U1-A
V
IN
3
SAMPLE
5
CLOCK
1
4
Figure 37. An Efficient Sample-and-Hold Circuit
–10–
REV. A
Page 11
AD8591/AD8592/AD8594
The U1-A amplifier is configured as a unity gain buffer driving a
1 nF capacitor. The input signal is connected to the noninverting
input, while the sample clock controls the shutdown for that
amplifier. When the sample clock is high, the U1-A amplifier is
active and the output follows V
. Once the sample clock goes
IN
low, U1-A shuts down with the output of the amplifier going to
a high impedance state, holding the voltage on the C1 capacitor.
The U1-B amplifier is used as a unity gain buffer to prevent loading on C1. Because of the low input bias current of the U1-B
CMOS input stage and the high impedance state of the U1-A
output in shutdown, there is very little voltage droop from C1
during the Hold period. This circuit can be used with sample
frequencies as high as 500␣ kHz and as low as below 1␣ Hz. Even
lower voltage droop can be achieved for very low sample rates
by increasing the value of C1.
Direct Access Arrangement for PCMCIA Modems
(Telephone Line Interface)
Figure 38 illustrates a +5␣ V transmit/receive telephone line
interface for 600␣ Ω systems. It allows full duplex transmission of
signals on a transformer-coupled 600␣ Ω line in a differential
manner. Amplifier A1 provides gain that can be adjusted to
meet the modem output drive requirements. Both A1 and A2
are configured to apply the largest possible signal on a single
supply to the transformer. Because of the AD8594’s high output
current drive and low dropout voltages, the largest signal available on a single +5␣ V supply is approximately 4.5␣ V␣ p-p into a
600␣ Ω transmission system. Amplifier A3 is configured as a
difference amplifier for two reasons: (1)␣ It prevents the transmit
signal from interfering with the receive signal and (2)␣ it extracts
the receive signal from the transmission line for amplification by
A4. Amplifier A4’s gain can be adjusted in the same manner as
A1’s to meet the modem’s input signal requirements. Standard
resistor values permit the use of SIP (Single In-line Package)
format resistor arrays. Couple this with the AD8594 16-lead
TSSOP or SOIC footprint, and this circuit offers a compact,
cost effective solution.
Single Supply Differential Line Driver
Figure 39 shows a single supply differential line driver circuit that
can drive a 600␣ Ω load with less than 0.7% distortion from 20 Hz
to 15 kHz with an input signal of 4 V p-p and a single +5 V supply.
The design uses an AD8594 to mimic the performance of a fully
balanced transformer based solution. However, this design occupies much less board space while maintaining low distortion and
can operate down to dc. Like the transformer based design, either
output can be shorted to ground for unbalanced line driver applications without changing the circuit gain of 1.
R3
2
C1
22mF
V
IN
A1, A2 = 1/2 AD8592
GAIN =
SET: R7, R10, R11 = R2
SET: R6, R12, R13 = R3
3
R3
R2
+5V
A1
10
4
10kV
1
R1
10kV
R10
10kV
10kV
R6
10kV
8
7
100kV
R9
+5V
R14
50V
R5
50V
R8
100kV
1mF
C2
2
1
A2
3
R2
R7
10kV
+5V
10
7
A1
9
4
R11
R12
10kV
10kV
8
9
A2
7
R13
10kV
C3
47mF
600V
C4
47mF
V
O1
R
L
V
O2
Figure 39. A Low Noise, Single Supply Differential
Line Driver
R8 and R9 set up the common-mode output voltage equal to
half of the supply voltage. C1 is used to couple the input signal
and can be omitted if the input’s dc voltage is equal to half of
the supply voltage.
The circuit can also be configured to provide additional gain if
desired. The gain of the circuit is:
P1
Tx GAIN
TO TELEPHONE
LINE
1:1
Z
O
600V
T1
MIDCOM
671-8005
A1, A2 = 1/2 AD8592
A3, A4 = 1/2 AD8592
6.2V
6.2V
R11
10kV
R3
360V
R9
10kV
R12
10kV
2
3
ADJUST
R5
10kV
R6
10kV
R10
10kV
5
A3
2kV
1
9
1
5
9.09kV
A1
6
A2
R13
10kV
R2
2
3
8
7
R14
14.3kV
8
7
R1
10kV
6
A4
C1
0.1mF
10mF
P2
Rx GAIN
ADJUST
2kV
9
TRANSMIT
SHUTDOWN
+5V
RECEIVE
C2
0.1mF
TxA
R7
10kV
R8
10kV
RxA
Figure 38. A Single Supply Direct Access Arrangement for
PCMCIA Modems
Where:V
–11–REV. A
V
OUT
A
==
V
V
IN
OUT
R2 = R7 = R10 = R11 and,
R3 = R6 = R12 = R13
R
3
R
2
= VO1␣–␣VO2,
(7)
Page 12
AD8591/AD8592/AD8594
SPICE Model for the AD8591/AD8592/AD8594 Amplifier
The SPICE model for the AD8591/AD8592/AD8594 amplifier is
one of the more realistic computer simulation macro-models
available, providing a high degree of realism with respect to characteristics of the actual amplifier. This model, shown in Listing 1,
is based on typical values for the device and can be downloaded
from Analog Devices’ Internet site at www.analog.com.
The model uses a common source output stage to provide railto-rail performance. This allows realistic simulation of openloop gain dependency on load resistance as well as maximum
output voltage versus output current. Two differential pairs are
used in the input stage of the model, simulating the rail-to-rail
input stage of the AD8591/AD8592/AD8594 amplifier.
The EOS voltage source establishes the input offset voltage and
is also used to simulate the common-mode rejection power
supply rejection, and input voltage noise characteristics for the
model. In addition, G2, R2 and CF are used to help set the
open-loop gain and gain-bandwidth product of the model.
A number of secondary characteristics are also accurately portrayed in the SPICE model. Flicker noise is accurately modeled
with the 1/f corner frequency set through the KF and AF terms
in the input stage transistors. C1 and C2 are used in the input
section to create secondary poles to achieve an accurate phase
margin characteristic for the model.
The AD8591/AD8592/AD8594 shutdown circuitry is included
in the model. Switches S1 through S7 deactivate the op amp
circuitry in shutdown mode. The logic threshold for the shutdown circuitry is accurately modeled through the VSWITCH
model parameters near the end of the listing. The active supply
current versus supply voltage is also modeled through the voltage-controlled current source GSY.
Characteristics of this model are based on typical values for the
AD8591/AD8592/AD8594 amplifier at +27°C. The model’s
characteristics are optimized specifically at +27°C, and may lose
accuracy at different simulation temperatures.
–12–
REV. A
Page 13
Listing 1: AD859x SPICE Macro-Model
* AD8592 SPICE Macro-Model Typical Values
* 9/98, Ver. 1
* TAM / ADSC
*
* Copyright 1998 by Analog Devices
*
* Refer to “README.DOC” file for License
* Statement. Use of this
* model indicates your acceptance of the
* terms and provisions in
* the License Statement.
*
* Node Assignments
*noninverting input
*|inverting input
*|| positive supply
*|| | negative supply
*|| | |output
*|| | ||shutdown
*| | | || |
.SUBCKT AD85921299504580
*
* INPUT STAGE
*
M1 4 1 3 3 PIX L=0.8E-6 W=125E-6
M2 6 7 3 3 PIX L=0.8E-6 W=125E-6
RC1 4 50 4E3
RC2 6 50 4E3
C1 4 6 2E-12
I1 99 8 100E-6