Datasheet

""#$ %""#
""&$ %""&
SBOS077B − JUNE 1997 − REVISED MARCH 2005
www.ti.com
9
The required compensation capacitor value can be
determined from the following equation:
C
1
= 1/(2πf
C
R
1
)
Since f
C
may shift with process variations, it is
recommended that a value less than f
C
be used for
determining C
1
. With f
C
= 1MHz and R
1
= 2.5kΩ, the
compensation capacitor is about 68pF.
The selection of the compensation capacitor C
1
is
important. A proper value ensures that the closed-loop
circuit gain is greater than or equal to 5 at high frequencies.
Referring to the Open-Loop Gain vs Frequency plot in the
Typical Characteristics section, the OPA338 gain line
(dashed in the curve) has a constant slope
(−20dB/decade) up to approximately 3MHz. This
frequency is referred to as f
C
. Beyond f
C
the slope of the
curve increases, suggesting that closed-loop gains less
than 5 are not appropriate.
Figure 4 shows a compensation technique using an
inverting configuration. The low-frequency gain is set by
the resistor ratio while the high-frequency gain is set by the
capacitor ratio. As with the noninverting circuit, for
frequencies above f
C
the gain must be greater than the
recommended minimum stable gain for the op amp.
C
1
150pF
C
2
=,C
1
=(G
H
−
1)
×
C
2
1
2
π
f
C
R
2
OPA338
V
IN
V
OUT
R
1
5k
Ω
R
2
10k
Ω
Where G
H
is the high−frequency gain,
G
H
=1+C
1
/C
2
Improved slew rate versus OPA337
(see Figure 5).
C
2
15pF
Figure 4. Inverting Compensation Circuit of the
OPA338 for Low Gain
Resistors R
1
and R
2
are chosen to set the desired dc
signal gain. Then the value for C
2
is determined as follows:
C
2
= 1/(2πf
C
R
2
)
C
1
is determined from the desired high-frequency gain (G
H
):
C
1
= (G
H
− 1) × C
2
For a desired dc gain of 2 and high-frequency gain of 10,
the following resistor and capacitor values result:
R
1
= 10kΩ C
1
= 150pF
R
2
= 5kΩ C
2
= 15pF
The capacitor values shown are the nearest standard
values. Capacitor values may need to be adjusted slightly
to optimize performance. For more detailed information,
consult the section on Low Gain Compensation in the
OPA846 data sheet (SBOS250) located at www.ti.com.
Figure 5 shows the large-signal transient response using
the circuit given in Figure 4. As shown, the OPA338 is
stable in low gain applications and provides improved slew
rate performance when compared to the OPA337.
500mV/div
Time (2
µ
s/div)
OPA338
OPA337
Figure 5. G = 2, Slew-Rate Comparison of the
OPA338 and the OPA337
TYPICAL APPLICATION
See Figure 6 for the OPA2337 in a typical application. The
ADS7822 is a 12-bit, micropower, sampling analog-to-
digital converter available in the tiny MSOP-8 package. As
with the OPA2337, it operates with a supply voltage as low
as +2.7V. When used with the miniature SOT23-8 package
of the OPA2337, the circuit is ideal for space-limited and
low-power applications. In addition, the OPA2337’s high
input impedance allows large value resistors to be used
which results in small physical capacitors, further reducing
circuit size. For further information, consult the ADS7822
data sheet (SBAS062) located at www.ti.com.