AD795JN [ADI]

Low Power, Low Noise Precision FET Op Amp; 低功耗,低噪声精密FET运算放大器
AD795JN
型号: AD795JN
厂家: ADI    ADI
描述:

Low Power, Low Noise Precision FET Op Amp
低功耗,低噪声精密FET运算放大器

运算放大器
文件: 总16页 (文件大小:333K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
Low Power, Low Noise  
Precision FET Op Amp  
a
AD795  
FEATURES  
CO NNECTIO N D IAGRAMS  
Low Pow er Replacem ent for Burr-Brow n  
OPA-111, OPA-121 Op Am ps  
Low Noise  
8-P in P lastic Mini-D IP (N) P ackage  
30  
V
s
= ±15V  
25  
2.5 V p-p m ax, 0.1 Hz to 10 Hz  
11 nV/ Hz m ax at 10 kHz  
0.6 fA/ Hz at 1 kHz  
20  
15  
10  
High DC Accuracy  
250 V m ax Offset Voltage  
3 V/ ؇C m ax Drift  
5
0
10  
100  
LOAD RESISTANCE –  
1k  
10k  
1 pA m ax Input Bias Current  
Low Pow er: 1.5 m A m ax Supply Current  
Available in Low Cost Plastic Mini-DIP and Surface  
Mount (SOIC) Packages  
8-P in SO IC (R) P ackage  
1
2
3
4
8
7
6
5
NC  
–IN  
+IN  
NC  
+V  
S
OUTPUT  
NC  
APPLICATIONS  
Low Noise Photodiode Pream ps  
CT Scanners  
–V  
S
AD795  
NC = NO CONNECT  
Precision l-to-V Converters  
P RO D UCT D ESCRIP TIO N  
Furthermore, the AD795 features a guaranteed low input noise  
of 2.5 µV p-p (0.1 Hz to 10 Hz) and a 11 nV/Hz max noise  
level at 10 kHz. T he AD795 has a fully specified and tested  
input offset voltage drift of only 3 µV/°C max.  
T he AD795 is a low noise, precision, FET input operational  
amplifier. It offers both the low voltage noise and low offset drift  
of a bipolar input op amp and the very low bias current of a  
FET -input device. T he 1014 common-mode impedance  
insures that input bias current is essentially independent of  
common-mode voltage and supply voltage variations.  
T he AD795 is useful for many high input impedance, low noise  
applications. T he AD795J and AD795K are rated over the  
commercial temperature range of 0°C to +70°C.  
T he AD795 has both excellent dc performance and a guaran-  
teed and tested maximum input voltage noise. It features 1 pA  
maximum input bias current and 250 µV maximum offset volt-  
age, along with low supply current of 1.5 mA max.  
T he AD795 is available in 8-pin plastic mini-DIP and 8-pin  
surface mount (SOIC) packages.  
1k  
100  
10  
1
50  
SAMPLE SIZE = 570  
40  
30  
20  
10  
0
10  
100  
FREQUENCY – Hz  
1k  
10k  
–5  
–4  
–3  
–2  
–1  
0
1
2
3
4
5
INPUT OFFSET VOLTAGE DRIFT – µV/°C  
AD795 Voltage Noise Spectral Density  
Typical Distribution of Average Input Offset Voltage Drift  
REV. A  
Inform ation furnished by Analog Devices is believed to be accurate and  
reliable. However, no responsibility is assum ed by Analog Devices for its  
use, nor for any infringem ents of patents or other rights of third parties  
which m ay result from its use. No license is granted by im plication or  
otherwise under any patent or patent rights of Analog Devices.  
One Technology Way, P.O. Box 9106, Norw ood, MA 02062-9106, U.S.A.  
Tel: 617/ 329-4700  
Fax: 617/ 326-8703  
AD795–SPECIFICATIONS(@ +25؇C and ؎15 V dc unless otherwise noted)  
AD 795JN/JR  
AD 795K  
Typ  
P aram eter  
Conditions  
Min  
Typ  
Max  
Min  
Max  
Units  
INPUT OFFSET VOLT AGE1  
Initial Offset  
100  
300  
3
110  
100  
500  
1000  
10  
50  
100  
1
110  
100  
250  
400  
3
µV  
µV  
µV/°C  
dB  
dB  
Offset  
T
MIN–T MAX  
vs. T emperature  
vs. Supply (PSRR)  
vs. Supply (PSRR)  
86  
84  
90  
87  
T MIN–T MAX  
2
INPUT BIAS CURRENT  
Either Input  
Either Input @ T MAX  
Either Input  
Offset Current  
Offset Current @ T MAX  
VCM = 0 V  
VCM = 0 V  
VCM = +10 V  
VCM = 0 V  
VCM = 0 V  
1
23  
1
0.1  
2
2/3  
1.0  
1
23  
1
0.1  
2
1
pA  
pA  
pA  
pA  
pA  
=
0.6  
=
OPEN-LOOP GAIN  
VO = ±10 V  
RLOAD 10 kΩ  
RLOAD 10 kΩ  
110  
100  
120  
108  
110  
100  
120  
108  
dB  
dB  
INPUT VOLT AGE NOISE  
0.1 Hz to 10 Hz  
f = 10 Hz  
f = 100 Hz  
f = 1 kHz  
1.0  
20  
12  
11  
9
3.3  
50  
40  
17  
11  
1.0  
20  
12  
11  
9
2.5  
40  
30  
15  
11  
µV p-p  
nV/Hz  
nV/Hz  
nV/Hz  
nV/Hz  
f = 10 kHz  
INPUT CURRENT NOISE  
f = 0.1 Hz to 10 Hz  
f = 1 kHz  
13  
0.6  
13  
0.6  
fA p-p  
fA/Hz  
FREQUENCY RESPONSE  
Unity Gain, Small Signal  
Full Power Response  
G = –1  
1.6  
16  
1
1.6  
16  
1
MHz  
kHz  
VO = 20 V p-p  
RLOAD = 2 kΩ  
VOUT = 20 V p-p  
RLOAD = 2 kΩ  
Slow Rate, Unity Gain  
V/µs  
SET T LING T IME3  
T o 0.1%  
10 V Step  
10 V Step  
50% Overdrive  
f = 1 kHz  
R1 10 kΩ  
VO = 3 V rms  
10  
11  
2
10  
11  
2
µs  
µs  
µs  
T o 0.01%  
Overload Recovery4  
T otal Harmonic  
Distortion  
–108  
–108  
dB  
INPUT IMPEDANCE  
Differential  
Common Mode  
VDIFF = ±1 V  
1012ʈ2  
1012ʈ2  
ʈpF  
ʈpF  
1014ʈ2.2  
1014ʈ2.2  
INPUT VOLT AGE RANGE  
Differential5  
Common-Mode Voltage  
Over Max Operating Temperature  
Common-Mode Rejection Ratio  
±20  
±11  
±20  
±11  
V
V
V
dB  
dB  
±10  
±10  
90  
±10  
±10  
94  
VCM = ±10 V  
T MIN–T MAX  
110  
100  
110  
100  
86  
90  
OUT PUT CHARACT ERIST ICS  
Voltage  
RLOAD 2 kΩ  
T MIN–T MAX  
VOUT = ±10 V  
Short Circuit  
VS –4  
VS –4  
±5  
VS –2.5  
VS –4  
VS –4  
±5  
VS –2.5  
V
V
mA  
mA  
Current  
±10  
±15  
±10  
±15  
POWER SUPPLY  
Rated Performance  
Operating Range  
Quiescent Current  
±15  
±15  
V
V
mA  
±4  
±18  
1.5  
±4  
±18  
1.5  
1.3  
1.3  
–2–  
REV. A  
AD795  
NOT ES  
1Input offset voltage specifications are guaranteed after 5 minutes of operation at T A = +25°C.  
2Bias current specifications are guaranteed maximum at either input after 5 minutes of operation at T A = +25°C. For higher temperature, the current doubles every 10 °C.  
3Gain = –1, R1 = 10 k.  
4Defined as the time required for the amplifier’s output to return to normal operation after removal of a 50% overload from the amplifier input.  
5Defined as the maximum continuous voltage between the inputs such that neither input exceeds ±10 V from ground.  
All min and max specifications are guaranteed.  
Specifications subject to change without notice.  
ABSO LUTE MAXIMUM RATINGS1  
ESD SUSCEP TIBILITY  
Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ±18 V  
ESD (electrostatic discharge) sensitive device. Electrostatic  
charges as high as 4000 volts, which readily accumulate on the  
human body and on test equipment, can discharge without  
detection. Although the AD795 features proprietary ESD pro-  
tection circuitry, permanent damage may still occur on these  
devices if they are subjected to high energy electrostatic dis-  
charges. T herefore, proper ESD precautions are recommended  
to avoid any performance degradation or loss of functionality.  
Internal Power Dissipation2 (@ TA = +25°C)  
SOIC Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500 mW  
8-Pin Mini-DIP Package . . . . . . . . . . . . . . . . . . . . 750 mW  
Input Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ±VS  
Output Short Circuit Duration . . . . . . . . . . . . . . . . Indefinite  
Differential Input Voltage . . . . . . . . . . . . . . . . . . +VS and –VS  
Storage T emperature Range (N, R) . . . . . . . –65°C to +125°C  
Operating T emperature Range  
AD795J/K . . . . . . . . . . . . . . . . . . . . . . . . . . . 0°C to +70°C  
O RD ERING GUID E  
NOT ES  
1Stresses above those listed under “Absolute Maximum Ratings” may cause  
permanent damage to the device. T his is a stress rating only and functional  
operation of the device at these or any other conditions above those indicated in the  
operational section of this specification is not implied. Exposure to absolute  
maximum rating conditions for extended periods may affect device reliability.  
28-Pin Plastic Mini-DIP Package: θJA = 100°C/Watt  
Model  
Tem perature Range  
P ackage O ption*  
AD795JN  
AD795KN  
AD795JR  
0°C to +70°C  
0°C to +70°C  
0°C to +70°C  
N-8  
N-8  
R-8  
8-Pin Small Outline Package: θJA = 155°C/Watt  
*N = Plastic mini-DIP; R = SOIC package.  
REV. A  
–3–  
AD795–Typical Characteristics  
20  
20  
R
= 10kΩ  
L
R
= 10k  
L
15  
10  
5
+V  
15  
10  
5
IN  
+V  
OUT  
–V  
IN  
–V  
OUT  
0
0
0
5
10  
SUPPLY VOLTAGE – ±Volts  
15  
20  
0
5
10  
SUPPLY VOLTAGE – ±Volts  
15  
20  
Figure 1. Com m on-Mode Voltage Range vs. Supply  
Figure 2. Output Voltage Range vs. Supply Voltage  
30  
1.0  
0.95  
0.90  
0.85  
0.80  
0.75  
0.70  
0.65  
V
= ±15V  
s
25  
20  
15  
10  
5
0
0.60  
0
5
10  
15  
20  
10  
100  
1k  
10k  
SUPPLY VOLTAGE – ±Volts  
LOAD RESISTANCE – Ω  
Figure 4. Input Bias Current vs. Supply  
Figure 3. Output Voltage Swing vs. Load Resistance  
10 –9  
50  
SAMPLE SIZE = 1058  
40  
10 –10  
10 –11  
30  
20  
10  
0
10 –12  
10 –13  
10 –14  
–60 –40 –20  
0
20  
40  
60  
80  
100 120 140  
0
.5  
1
1.5  
2
TEMPERATURE –  
°
C
INPUT BIAS CURRENT – pA  
Figure 5. Typical Distribution of Input Bias Current  
Figure 6. Input Bias Current vs. Tem perature  
–4–  
REV. A  
AD795  
10–4  
1.00  
0.95  
10–5  
10–6  
10–7  
–I  
+I  
IN  
IN  
0.90  
0.85  
0.80  
0.75  
0.70  
0.65  
0.60  
10–8  
10–9  
10–10  
10–11  
10–12  
10–13  
10–14  
–5  
–6  
4
–2  
–1  
1
2
3
5
6
–4 –3  
0
0
–15  
–10  
–5  
+5  
+10  
+15  
DIFFERENTIAL INPUT VOLTAGE – ±Volts  
COMMON MODE VOLTAGE – Volts  
Figure 8. Input Bias Current vs. Differential Input Voltage  
Figure 7. Input Bias Current vs. Com m on-Mode Voltage  
100  
15  
1k  
f = 1kHz  
Noise Bandwidth: 0.1 to 10Hz  
10  
12.5  
10  
VOLTAGE NOISE  
100  
10  
1.0  
CURRENT NOISE  
7.5  
5
0.1  
0.01  
1.0  
0
20  
40  
60  
80  
100 120 140  
–60 –40 –20  
3
4
5
6
7
8
9
10  
10  
10  
10  
10  
10  
10  
TEMPERATURE –  
°C  
SOURCE RESISTANCE –  
Figure 10. Input Voltage Noise vs. Source Resistance  
Figure 9. Voltage and Current Noise Spectral Density vs.  
Tem perature  
1k  
50  
SAMPLE SIZE = 344  
40  
f = 0.1 TO 10Hz  
100  
10  
30  
20  
10  
0
1.0  
1
10  
100  
1k  
10k  
100k  
1M  
10M  
0
1
2
3
FREQUENCY – Hz  
0.1 TO 10Hz INPUT VOLTAGE NOISE p-p – µV  
Figure 11. Typical Distribution of Input Voltage Noise  
Figure 12. Input Voltage Noise Spectral Density  
REV. A  
–5–  
AD795–Typical Characteristics  
30  
10  
8
25  
6
0.1%  
– OUTPUT CURRENT  
4
0.01%  
2
20  
ERROR  
0
–2  
–4  
15  
+ OUTPUT CURRENT  
0.1%  
0.01%  
10  
–6  
–8  
–10  
5
3
4
5
6
7
8
9
10  
11  
–60 –40 –20  
0
20  
40  
60  
80  
100 120 140  
TEMPERATURE –  
°C  
SETTING TIME – µs  
Figure 13. Short Circuit Current Lim it vs. Tem perature  
Figure 14. Output Swing and Error vs. Settling Tim e  
120  
1000  
900  
800  
700  
600  
500  
400  
300  
200  
100  
0
100  
+PSRR  
–PSRR  
80  
60  
40  
20  
0
1
10  
100  
1k  
10k  
100k  
1M  
10M  
–15  
–10  
–5  
0
5
10  
15  
FREQUENCY – Hz  
INPUT COMMON MODE VOLTAGE – Volts  
Figure 15. Absolute Input Error Voltage vs. Input  
Com m on-Mode Voltage  
Figure 16. Power Supply Rejection vs. Frequency  
120  
120  
120  
100  
80  
60  
40  
20  
0
100  
80  
60  
40  
20  
0
PHASE  
100  
80  
60  
40  
20  
0
GAIN  
–20  
10M  
–20  
10  
100  
1k  
10k  
100k  
1M  
1
10  
100  
1k  
10k  
100k  
1M  
10M  
FREQUENCY – Hz  
FREQUENCY – Hz  
Figure 17. Com m on-Mode Rejection vs. Frequency  
Figure 18. Open-Loop Gain & Phase Margin vs. Frequency  
–6–  
REV. A  
AD795  
30  
1000  
100  
10  
R
= 10k  
L
25  
20  
15  
10  
5
1.0  
0.1  
0
1k  
10k  
100k  
1M  
1k  
10k  
100k  
FREQUENCY – Hz  
1M  
10M  
FREQUENCY – Hz  
Figure 19. Large Signal Frequency Response  
Figure 20. Closed-Loop Output Im pedance vs. Frequency  
2.0  
–60  
V
= 3Vrms  
= 10k  
IN  
R
L
–70  
–80  
1.5  
1.0  
–90  
–100  
–110  
–120  
0.5  
0
0
5
10  
15  
20  
100  
1k  
10k  
FREQUENCY – Hz  
100k  
SUPPLY VOLTAGE ± Volts  
Figure 21. Total Harm onic Distortion vs. Frequency  
Figure 22. Quiescent Supply Current vs. Supply  
Voltage Drift  
50  
SAMPLE SIZE = 1419  
40  
30  
20  
10  
0
–500 –400 –300 –200 –100  
0
100 200 300 400 500  
INPUT OFFSET VOLTAGE – µV  
Figure 23. Typical Distribution of Input Offset Voltage  
REV. A  
–7–  
AD795  
10kΩ  
+V  
S
0.1µF  
20V  
5µs  
10mV  
500ns  
100  
90  
100  
90  
10kΩ  
7
V
2
3
IN  
V
6
OUT  
AD795  
C
R
L
L
100pF  
10kΩ  
4
0.1µF  
10  
10  
0%  
0%  
–V  
S
5V  
Figure 24. Unity Gain Inverter  
Figure 25. Unity Gain Inverter  
Large Signal Pulse Response  
Figure 26. Unity Gain Inverter  
Sm all Signal Pulse Response  
+VS  
0.1µF  
20mV  
500ns  
20V  
5µs  
100  
90  
100  
90  
7
2
VOUT  
6
AD795  
VIN  
CL  
100pF  
RL  
3
10kΩ  
4
0.1µF  
10  
10  
0%  
0%  
–VS  
5V  
Figure 27. Unity Gain Follower  
Figure 28. Unity Gain Follower  
Large Signal Pulse Response  
Figure 29. Unity Gain Follower  
Sm all Signal Pulse Response  
current will double for every 10°C rise in junction temperature  
(illustrated in Figure 6). On-chip power dissipation will raise the  
device operating temperature, causing an increase in input  
current. Reducing supply voltage to cut power dissipation will  
reduce the AD795s input current (Figure 4). Heavy output  
loads can also increase chip temperature, maintaining a  
minimum load resistance of 10 kis recommended.  
MINIMIZING INP UT CURRENT  
T he AD795 is guaranteed to 1 pA max input current with ±15  
volt supply voltage at room temperature. Careful attention to  
how the amplifier is used will maintain or possibly better this  
performance.  
T he amplifier’s operating temperature should be kept as low as  
possible. Like other JFET input amplifier’s, the AD795s input  
–8–  
REV. A  
AD795  
CIRCUIT BO ARD NO TES  
T he AD795 is designed for throughhole mounting on PC  
boards, using either mini-DIP or surface mount (SOIC).  
Maintaining picoampere resolution in those environments  
requires a lot of care. Both the board and the amplifier’s  
package have finite resistance. Voltage differences between the  
input pins and other pins as well as PC board metal traces will  
cause parasitic currents (Figure 30) larger than the AD795’s  
input current unless special precautions are taken. T wo methods  
of minimizing parasitic leakages are guarding of the input lines  
and maintaining adequate insulation resistance.  
C
F
R
F
V
E
2
3
6
+
AD795  
I
S
V
OUT  
I
P
R
P
C
P
V
R
dC  
P
dV  
S
S
I
=
+
V +  
S
C
P
P
Figures 31 and 32 show the recommended guarding schemes for  
follower and inverted topologies. Note that for the mini-DIP,  
the guard trace should be on both sides of the board. On the  
SOIC, Pin 1 is not connected, and can be safely connected to  
the guard. T he high impedance input trace should be guarded  
on both edges for its entire length.  
dT  
dT  
P
V
S
Figure 30. Sources of Parasitic Leakage Currents  
GUARD TRACES PARALLEL  
TO BOTH EDGES OF  
INPUT TRACE  
CF  
GUARD  
1
8
RF  
INPUT  
TRACE  
7
2
3
2
3
AD795  
TOP VIEW  
6
+
TO ANALOG  
COMMON  
6
5
AD795  
IS  
VOUT  
("N" PACKAGE)  
4
–VS  
8
1
1
8
7
6
BOTTOM  
VIEW  
("N"  
NOTE:  
ON THE "R" PACKAGE  
PINS 1, 5 AND 8 ARE OPEN  
AND CAN BE CONNECTED  
TO ANALOG COMMON OR  
TO THE DRIVEN GUARD TO  
REDUCE LEAKAGE.  
2
3
7
6
2
3
PACKAGE)  
TOP VIEW  
("R" PACKAGE)  
5
4
4
5
Figure 31. Guarding Schem e–lnverter  
GUARD  
GUARD TRACES  
3
2
1
8
+
AD795  
TOP VIEW  
6
+
AD795  
V
OUT  
V
7
6
2
3
S
R
F
INPUT  
TRACE  
–V  
S
4
5
CONNECT TO JUNCTION  
OF R AND R , OR TO PIN 6  
F
I
R
FOR UNITY GAIN.  
I
Figure 32. Guard Schem e–Follower  
REV. A  
–9–  
AD795  
Leakage through the bulk of the circuit board will still occur  
with the guarding schemes shown in Figures 31 and 32. Stan-  
dard “G10” type printed circuit board material may not have  
high enough volume resistivity to hold leakages at the sub-  
picoampere level particularly under high humidity conditions.  
One option that eliminates all effects of board resistance is  
shown in Figure 33. T he AD795’s sensitive input pin (either  
Pin 2 when connected as an inverter, or Pin 3 when connected  
as a follower) is bent up and soldered directly to a T eflon*  
insulated standoff. Both the signal input and feedback compo-  
nent leads must also be insulated from the circuit board by  
T eflon standoffs or low-leakage shielded cable.  
Both proper shielding and rigid mechanical mounting of  
components help minimize error currents from both of these  
sources.  
O FFSET NULLING  
T he AD795s input offset voltage can be nulled (mini-DIP  
package only) by using balance Pins 1 and 5, as shown in  
Figure 34. Nulling the input offset voltage in this fashion will  
introduce an added input offset voltage drift component of  
2.4 µV/°C per millivolt of nulled offset.  
+V  
S
INPUT PIN:  
PIN 2 FOR INVERTER  
OR PIN 3 FOR FOLLOWER  
7
2
3
V
OUT  
INPUT SIGNAL  
+
AD795  
1
6
LEAD  
1
8
7
6
5
5
AD795  
2
3
AD795  
100kΩ  
4
PC  
BOARD  
4
–V  
S
TEFLON INSULATED STANDOFF  
Figure 34. Standard Offset Null Circuit  
Figure 33. Input Pin to Insulating Standoff  
T he circuit in Figure 35 can be used when the amplifier is used  
as an inverter. T his method introduces a small voltage in series  
with the amplifier’s positive input terminal. T he amplifier’s  
input offset voltage drift with temperature is not affected.  
However, variation of the power supply voltages will cause  
offset shifts.  
Contaminants such as solder flux on the board’s surface and on  
the amplifier’s package can greatly reduce the insulation resis-  
tance between the input pin and those traces with supply or  
signal voltages. Both the package and the board must be kept  
clean and dry. An effective cleaning procedure is to first swab  
the surface with high grade isopropyl alcohol, then rinse it with  
deionized water and, finally, bake it at 100°C for 1 hour. Poly-  
propylene and polystyrene capacitors should not be subjected to  
the 100°C bake as they will be damaged at temperatures greater  
than 80°C.  
R
F
R
I
2
3
+
Other guidelines include making the circuit layout as compact  
as possible and reducing the length of input lines. Keeping  
circuit board components rigid and minimizing vibration will  
reduce triboelectric and piezoelectric effects. All precision high  
impedance circuitry requires shielding from electrical noise and  
interference. For example, a ground plane should be used under  
all high value (i.e., greater than 1 M) feedback resistors. In  
some cases, a shield placed over the resistors, or even the entire  
amplifier, may be needed to minimize electrical interference  
originating from other circuits. Referring to the equation in  
Figure 30, this coupling can take place in either, or both, of two  
different forms—coupling via time varying fields:  
6
+
OUT  
AD795  
V
V
I
+V  
–V  
S
499k  
0.1µF  
499kΩ  
200Ω  
100kΩ  
S
Figure 35. Alternate Offset Null Circuit for Inverter  
dV  
CP  
dT  
or by injection of parasitic currents by changes in capacitance  
due to mechanical vibration:  
dCp  
V
dT  
*T eflon is a registered trademark of E.I. du Pont Co.  
–10–  
REV. A  
AD795  
AC RESP O NSE WITH H IGH VALUE SO URCE AND  
FEED BACK RESISTANCE  
10mV  
5µs  
Source and feedback resistances greater than 100 kwill  
magnify the effect of input capacitances (stray and inherent to  
the AD795) on the ac behavior of the circuit. T he effects of  
common-mode and differential input capacitances should be  
taken into account since the circuit’s bandwidth and stability  
can be adversely affected.  
100  
90  
10  
0%  
In a follower, the source resistance, RS, and input common-  
mode capacitance, CS (including capacitance due to board and  
capacitance inherent to the AD795), form a pole that limits  
circuit bandwidth to 1/2 π RSCS. Figure 36 shows the follower  
pulse response from a 1 Msource resistance with the  
amplifier’s input pin isolated from the board, only the effect of  
the AD795s input common-mode capacitance is seen.  
Figure 38. Inverter Pulse Response with 1 MSource and  
Feedback Resistance, 1 pF Feedback Capacitance  
O VERLO AD ISSUES  
Driving the amplifier output beyond its linear region causes  
some sticking; recovery to normal operation is within 2 µs of the  
input voltage returning within the linear range.  
10mV  
5µs  
100  
90  
If either input is driven below the negative supply, the amplifiers  
output will be driven high, causing a phenomenon called phase  
reversal. Normal operation is resumed within 30 µs of the input  
voltage returning within the linear range.  
10  
Figure 39 shows the AD795’s input currents versus differential  
input voltage. Picoamp level input current is maintained for  
differential voltages up to several hundred millivolts. T his  
behavior is only important if the AD795 is in an open-loop  
application where substantial differential voltages are produced.  
0%  
Figure 36. Follower Pulse Response from 1 MΩ  
Source Resistance  
–4  
10  
In an inverting configuration, the differential input capacitance  
forms a pole in the circuit’s loop transmission. T his can create  
peaking in the ac response and possible instability. A feedback  
capacitance can be used to stabilize the circuit. T he inverter  
pulse response with RF and RS equal to 1 M, and the input pin  
isolated from the board appears in Figure 37. Figure 38 shows  
the response of the same circuit with a 1 pF feedback  
capacitance. T ypical differential input capacitance for the  
AD795 is 2 pF.  
–5  
10  
–I  
+I  
N
N
–6  
10  
–7  
–8  
–9  
10  
10  
10  
–10  
–11  
–12  
10  
10  
10  
10mV  
5µs  
–13  
–14  
10  
10  
100  
90  
–6 –5 –4  
–3 –2 –1  
0
1
2
3
4
5
6
DIFFERENTIAL INPUT VOLTAGE – ±Volts  
Figure 39. Input Bias Current vs. Differential Input Voltage  
10  
0%  
Figure 37. Inverter Pulse Response with 1 MSource and  
Feedback Resistance  
REV. A  
–11–  
AD795  
INP UT P RO TECTIO N  
T he AD795 safely handles any input voltage within the supply  
voltage range. Some applications may subject the input  
terminals to voltages beyond the supply voltages—in these  
cases, the following guidelines should be used to maintain the  
AD795s functionality and performance.  
R
P
SOURCE  
3
2
6
AD795  
If the inputs are driven more than a 0.5 V below the minus sup-  
ply, milliamp level currents can be produced through the input  
terminals. T hat current should be limited to 10 mA for “tran-  
sient” overloads (less than 1 second) and 1 mA for continuous  
overloads, this can be accomplished with a protection resistor in  
the input terminal (as shown in Figures 40 and 41). T he pro-  
tection resistor’s Johnson noise will add to the amplifier’s input  
voltage noise and impact the frequency response.  
Figure 41. Follower with Input Current Lim it  
Figure 42 is a schematic of the AD795 as an inverter with an  
input voltage clamp. Bootstrapping the clamp diodes at the  
inverting input minimizes the voltage across the clamps and  
keeps the leakage due to the diodes low. Low leakage diodes  
(less than 1 pA), such as the FD333s should be used, and  
should be shielded from light to keep photocurrents from being  
generated. Even with these precautions, the diodes will mea-  
surably increase the input current and capacitance.  
Driving the input terminals above the positive supply will cause  
the input current to increase and limit at 40 µA. T his condition  
is maintained until 15 volts above the positive supply—any  
input voltage within this range does not harm the amplifier.  
Input voltage above this range causes destructive breakdown  
and should be avoided.  
In order to achieve the low input bias currents of the AD795, it  
is not possible to use the same on-chip protection as used in  
other Analog Devices op amps. T his makes the AD795  
sensitive to handling and precautions should be taken to  
minimize ESD exposure whenever possible.  
RF  
RF  
CF  
RP  
SOURCE  
2
3
SOURCE  
2
3
6
AD795  
6
AD795  
PROTECT DIODES  
(LOW LEAKAGE)  
Figure 42. Input Voltage Clam p with Diodes  
Figure 40. Inverter with Input Current Lim it  
–12–  
REV. A  
AD795  
will typically drop by a factor of two for every 10°C rise in  
temperature. In the AD795, both the offset voltage and drift are  
low, this helps minimize these errors.  
10pF  
10 9  
Minim izing Noise Contr ibutions  
T he noise level limits the resolution obtainable from any pre-  
amplifier. T he total output voltage noise divided by the  
feedback resistance of the op amp defines the minimum  
detectable signal current. T he minimum detectable current  
divided by the photodiode sensitivity is the minimum detectable  
light power.  
GUARD  
2
3
OUTPUT  
6
AD795  
PHOTODIODE  
8
FILTERED  
OUTPUT  
Sources of noise in a typical preamp are shown in Figure 45.  
T he total noise contribution is defined as:  
2
2
Rf  
Rf 1+ s (Cd ) Rd  
Rd 1+ s (Cf ) Rf  
OPTIONAL 26Hz  
FILTER  
2
2
2
V
=
(i  
+
i
+ i )  
s
+(en2  
) 1+  
n
f
OUT  
1+ s (Cf ) Rf  
Figure 43. The AD795 Used as a Photodiode Pream plifier  
Cf  
10pF  
P r eam plifier Applications  
T he low input current and offset voltage levels of the AD795  
together with its low voltage noise make this amplifier an  
excellent choice for preamplifiers used in sensitive photodiode  
applications. In a typical preamp circuit, shown in Figure 43,  
the output of the amplifier is equal to:  
Rf  
109  
PHOTODIODE  
en  
i f  
VOUT = ID (Rf) = Rp (P) Rf  
in  
where:  
iS  
Rd  
Cd  
50pF  
OUTPUT  
iS  
ID = photodiode signal current (Amps)  
Rp = photodiode sensitivity (Amp/Watt)  
Rf = the value of the feedback resistor, in ohms.  
P
= light power incident to photodiode surface, in watts.  
Figure 45. Noise Contributions of Various Sources  
An equivalent model for a photodiode and its dc error sources is  
shown in Figure 44. T he amplifier’s input current, IB, will  
contribute an output voltage error which will be proportional to  
Figure 46, a spectral density versus frequency plot of each  
source’s noise contribution, shows that the bandwidth of the  
amplifier’s input voltage noise contribution is much greater than  
its signal bandwidth. In addition, capacitance at the summing  
junction results in a “peaking” of noise gain in this configura-  
tion. T his effect can be substantial when large photodiodes with  
large shunt capacitances are used. Capacitor Cf sets the signal  
bandwidth and also limits the peak in the noise gain. Each  
source’s rms or root-sum-square contribution to noise is ob-  
tained by integrating the sum of the squares of all the noise  
sources and then by obtaining the square root of this sum.  
Minimizing the total area under these curves will optimize the  
preamplifier’s overall noise performance.  
the value of the feedback resistor. T he offset voltage error, VOS  
will cause a “dark” current error due to the photodiode’s finite  
shunt resistance, Rd. T he resulting output voltage error, VE, is  
equal to:  
,
VE = (1 + Rf/Rd) VOS + Rf IB  
A shunt resistance on the order of 109 ohms is typical for a  
small photodiode. Resistance Rd is a junction resistance which  
Cf  
10pF  
An output filter with a passband close to that of the signal can  
greatly improve the preamplifier’s signal to noise ratio. T he  
photodiode preamplifier shown in Figure 45—without a  
bandpass filter—has a total output noise of 50 µV rms. Using a  
26 Hz single pole output filter, the total output noise drops to  
23 µV rms, a factor of 2 improvement with no loss in signal  
bandwidth.  
Rf  
9
10  
PHOTODIODE  
V
OS  
I
B
OUTPUT  
Cd  
50pF  
I
Rd  
D
Figure 44. A Photodiode Model Showing DC Error  
Sources  
REV. A  
–13–  
AD795  
voltage contributions are also amplified by the “T ” network  
gain. A low noise, low offset voltage amplifier, such as the  
AD795, is needed for this type of application.  
10µV  
&
i
i
f
s
SIGNAL BANDWIDTH  
A pH P r obe Buffer Am plifier  
A typical pH probe requires a buffer amplifier to isolate its 106  
to 109 source resistance from external circuitry. Just such an  
amplifier is shown in Figure 48. T he low input current of the  
AD795 allows the voltage error produced by the bias current  
and electrode resistance to be minimal. T he use of guarding,  
shielding, high insulation resistance standoffs, and other such  
standard methods used to minimize leakage are all needed to  
maintain the accuracy of this circuit.  
1µV  
i
n
WITH FILTER  
NO FILTER  
100nV  
e
n
en  
T he slope of the pH probe transfer function, 50 mV per pH  
unit at room temperature, has a +3300 ppm/°C temperature  
coefficient. T he buffer of Figure 48 provides an output voltage  
equal to 1 volt/pH unit. T emperature compensation is provided  
by resistor RT which is a special temperature compensation  
resistor, part number Q81, 1 k, 1%, +3500 ppm/°C, available  
from T el Labs Inc.  
10nV  
1
10  
100  
1k  
10k  
100k  
FREQUENCY – Hz  
Figure 46. Voltage Noise Spectral Density of the Circuit of  
Figure 45 With and Without an Output Filter  
10pF  
R
G
10k  
VOS ADJUST  
100kΩ  
Rf  
+VS  
+15V  
COM  
–15V  
0.1µF  
0.1µF  
8
R
1.1kΩ  
10  
i
–VS  
–VS  
V
1
OUT  
GUARD  
AD795  
4
3
2
OUTPUT  
1VOLT/pH UNIT  
5
PHOTODIODE  
R
R
G
AD795  
6
V
= I Rf (1+  
D
)
OUT  
i
7
PH  
PROBE  
19.6kΩ  
8
+VS  
Figure 47. A Photodiode Pream p Em ploying a “T”  
Network for Added Gain  
RT  
1kΩ  
+3500ppm/ °C  
Using a “ T” Networ k  
A “T ” network, shown in Figure 47, can be used to boost the  
effective transimpedance of an I-to-V converter, for a given  
feedback resistor value. However, amplifier noise and offset  
Figure 48. A pH Probe Am plifier  
–14–  
REV. A  
AD795  
LOW NOISE  
OP AMPS  
Low V  
Low I  
N
N
(V 10 nV/Hz @ 1 kHz)  
(I 10 fA/Hz @ 1 kHz)  
N
N
Fast  
Audio  
Amplifiers  
Precision  
Low V  
Fast  
FET Input  
Low Power  
Electrometer  
N
(SR 45 V/µs)  
AD OP27  
OP27  
AD OP37  
OP37  
AD645  
AD711  
AD645  
AD743  
AD795  
AD548  
AD648  
OP80  
OP61  
OP275  
Low  
Power  
AD795  
AD712 (Dual)  
SSM2015  
SSM2016  
SSM2017  
SSM2134  
SSM2139  
OP249 (Dual)  
AD713 (Quad)  
Faster  
Lower V  
N
OP80  
OP227 (Dual)  
Faster  
(SR 230 V/µs)  
Fast  
AD743  
OP270 (Dual)  
OP271 (Dual)  
OP470 (Quad)  
OP471 (Quad)  
(SR 8 V/µs)  
Faster  
AD745  
General  
Purpose  
AD5539  
AD829  
AD840  
AD844  
AD846  
AD848  
AD849  
Faster  
OP282 (Dual)  
OP482 (Quad)  
AD744  
OP42  
AD745  
AD546  
OP44  
AD746 (Dual)  
High Output  
Current  
Lowest I  
B
60 fA Max  
OP50  
AD549  
Ultrafast  
(SR 1000 V/µs)  
AD811  
AD844  
AD9610  
AD9617  
AD9618  
Low Noise Op Am p Selection Tree  
REV. A  
–15–  
AD795  
O UTLINE D IMENSIO NS  
D imensions shown in inches and (mm).  
P lastic Mini-D IP (N) P ackage  
8
5
0.25  
(6.35)  
0.31  
(7.87)  
PIN 1  
1
4
0.30 (7.62)  
REF  
0.39 (9.91) MAX  
0.035±0.01  
(0.89±0.25)  
0.165±0.01  
(4.19±0.25)  
0.011±0.003  
(0.28±0.08)  
0.18±0.03  
(4.57±0.76)  
0.125  
(3.18)  
MIN  
15  
°
0
°
0.10  
(2.54)  
0.018±0.003  
(0.46±0.08)  
0.033  
(0.84)  
NOM  
SEATING  
PLANE  
BSC  
8-P in SO IC (R) P ackage  
0.150 (3.81)  
8
5
0.244 (6.20)  
0.228 (5.79)  
0.157 (3.99)  
0.150 (3.81)  
PIN 1  
1
4
0.020 (0.051) x 45  
CHAMF  
°
0.190 (4.82)  
0.170 (4.32)  
0.197 (5.01)  
0.189 (4.80)  
8
0
°
°
0.090  
(2.29)  
0.102 (2.59)  
0.094 (2.39)  
0.010 (0.25)  
0.004 (0.10)  
10  
°
0°  
0.019 (0.48)  
0.014 (0.36)  
0.050  
(1.27)  
BSC  
0.030 (0.76)  
0.018 (0.46)  
0.098 (0.2482)  
0.075 (0.1905)  
All brand or product nam es m entioned are tradem arks or registered tradem arks of their respective holders.  
–16–  
REV. A  

相关型号:

AD795JR

Low Power, Low Noise Precision FET Op Amp
ADI

AD795JR-REEL

Low Power, Low Noise Precision FET Op Amp
ADI

AD795JR-REEL7

Low Power, Low Noise Precision FET Op Amp
ADI

AD795JRZ

Low Power, Low Noise Precision FET Op Amp
ADI

AD795JRZ-REEL

Low Power, Low Noise Precision FET Op Amp
ADI

AD795JRZ-REEL7

Low Power, Low Noise Precision FET Op Amp
ADI

AD795KN

Low Power, Low Noise Precision FET Op Amp
ADI

AD795SH-883B

IC OP-AMP, 500 uV OFFSET-MAX, 1.6 MHz BAND WIDTH, MBCY8, HERMETIC SEALED, METAL CAN, HEADER, TO-99, 8 PIN, Operational Amplifier
ADI

AD795_02

Low Power, Low Noise Precision FET Op Amp
ADI

AD795_09

Low Power, Low Noise Precision FET Op Amp
ADI

AD7960

18-Bit, 5 MSPS PulSAR Differential ADC
ADI

AD7960BCPZ

18-Bit, 5 MSPS PulSAR Differential ADC
ADI