REV. A
OP227
– 1 1 –
Dynamic range is limited by A1 as well as A2. The output of A1
is:
V
R
R
V
V
N
O
d
CM
1
1
2
=
+
ê
ˉ
+
–
If the instrumentation amplifier was designed for a gain of 10
and maximum V
d
of
±
1 V, then R
N
/R
O
would need to be four
and V
O
would be a maximum of
±
10 V. Amplifier A1 would have
a maximum output of
±
5 V plus 2 V
CM
, thus a limit of
±
10 V
on the output of A1 would imply a limit of
±
2.5 V on V
CM
. A
nominal value of 10 k
W
for R
N
is suitable for most applications.
A range of 20
W
to 2.5 k
W
for R
O
will then provide a gain range
of 10 to 1000. The current through R
O
is V
d
/R
O
, so the amplifiers
must supply
±
10 mV/20
W
(or
±
0.5 mA) when the gain is at the
maximum value of 1000 and V
d
is at
±
10 mV.
Rejecting common-mode inputs is important in accurately
amplifying low level differential signals. Two factors determine
the CMR in this instrumentation amplifier configuration (assuming
infinite gain):
∑
CMR of the op amps
∑
Matching of the resistor network ratios (R3/R4 = R2/R1)
In this instrumentation amplifier configuration error due to CMR
effect is directly proportional to the CMR match of the op
amps.
For the OP227, this DCMR is a minimum of 97 dB for the
“G”
and 110 dB for the “E” grades. A DCMR value of 100 dB and a
common-mode input range of
±
2.5 V indicates a peak input-
referred error of only
±
25
m
V. Resistor matching is the other
factor affecting CMR. Defining A
d
as the differential gain of the
instrumentation amplifier and assuming that R1, R2, R3, and R4
are approximately equal (R
N
will be the nominal value), then CMR
for this instrumentation amplifier configuration will be approxi-
mately A
d
divided by 4 R/R
N
. CMR at differential gain of 100
would be 88 dB with resistor matching of 0.01%. Trimming R1
to make the ratio R3/R4 equal to R2/R1 will raise the CMR
until limited by linearity and resistor stability considerations.
The high open-loop gain of the OP227 is very important to
achieving high accuracy in the two op amp instrumentation
amplifier configuration. Gain error can be approximated by:
Gain Error
A
A
A
A A
d
O
d
O
1
1
2
1
2
1
+
<
,
where A
d
is the instrumentation amplifier differential gain and
A
O2
is the open loop gain of op amp A2. This analysis assumes
equal values of R1, R2, R3, and R4. For example, consider an
OP227 with A
O2
of 700 V/mV. Id the differential gain A
d
were
set to 700, then the gain error would be 1/1.001, which is
approximately 0.1%.
Another effect of finite op amp gain is undesired feedthrough of
common-mode input. Defining A
O1
as the open-loop gain of op
amp A1, then the common-mode error (CME) at the output
due to this effect would be approximately:
CME
A
A
A
A
V
d
d
O
O
CM
2
1
1
2
1
+
,
For A
d
/A
01
< 1, this simplifies to (2A
d
/A
01
) 3 V
CM
. If the op amp
gain is 700 V/mV, V
CM
is 2.5 V, and A
d
is set to 700, then the
error at the output due to this effect will be approximately 5 mV.
A compete instrumentation amplifier designed for a gain of 100
is shown in Figure 3. It has provision for trimming of input
offset
voltage, CMR, and gain. Performance is excellent due to
the high
gain, high CMR, and low noise of the individual ampli-
fiers combined
with the tight matching characteristics of the
OP227 dual.
3
4
10
11
2
1
14
13
12
7
6
10k
OFFSET
V+
V–
V+
V
O
= 100V
d
V–
5
ADJUST
CMR
50
9.95k
2.5k
191
10k
0.1%
V
CM
– 1/2V
d
GAIN
V
CM
– 1/2V
d
10k , 0.1%
10k , 0.1%
OP227
Figure 3. Two Op Amp Instrumentation Amplifier Using
OP227 Dual
A three op amp instrumentation amplifier configuration using
the OP227 and OP27 is recommended for applications requir-
ing high accuracy over a wide gain range. This circuit provides
excellent CMR over a wide frequency range. As with the two op
amp instrumentation amplifier circuits, the tight matching of the
two op amps within the OP227 package provides a real boost in
performance. Also, the low noise, low offset, and high gain of
the individual op amps minimize errors.
A simplified schematic is shown in Figure 4. The input stage
(A1 and A2) serves to amplify the differential input V
d
without
amplifying the common-mode voltage V
CM
. The output stage
then rejects the common-mode input. With ideal op amps and
no resistor matching errors, the outputs of each amplifier will
be:
V
R
R
V
V
V
R
R
V
V
V
V
V
R
V
V
A V
O
d
CM
O
d
CM
O
O
d
O
d
1
2
2
1
1
2 1
2
1
2 1
2
1
2R
=
+
ê
ê
ˉ
ˉ
+
=
+
+
=
=
+
ê
ˉ
=
–
–
–