AD202/AD204
AD204s are operated from a single AD246, an external bypass
capacitor should be used with a value of at least 1
µF
for every
five isolators used. Place the capacitor as close as possible to the
clock driver.
The “noninverting” circuit of Figures 5 and 6 can also be used
to advantage when a signal inversion is needed: just interchange
either the input leads or the output leads to get inversion. This
approach retains the high input resistance of the “noninverting”
circuit, and at unity gain no gain-setting resistors are needed.
When the isolator is not powered, a negative input voltage of
more than about 2 V will cause an input current to flow. If the
signal source can supply more than a few mA under such condi-
tions, the 2 kΩ resistor shown in series with IN+ should be used
to limit current to a safe value. This is particularly important
with the AD202, which may not start if a large input current is
present.
Figure 7 shows how to accommodate current inputs or sum cur-
rents or voltages. This circuit can also be used when the input
signal is larger than the
±
5 V input range of the isolator; for
example, a
±
50 V input span can be accommodated with R
F
=
20 k and R
S
= 200 k. Once again, a capacitor from FB to IN
COM is required for gains above five.
Figure 4. Clock Driver
Input Configurations.
The AD202 and AD204 have been de-
signed to be very easy to use in a wide range of applications.
The basic connection for standard unity gain applications, useful
for signals up to
±
5 V, is shown in Figure 5; some of the possible
variations are described below. When smaller signals must be
Figure 7. Connections for Summing or Current Inputs
Figure 5. Basic Unity-Gain Application
handled, Figure 6 shows how to get gain while preserving a very
high input resistance. The value of feedback resistor R
F
should
be kept above 20 kΩ for best results. Whenever a gain of more
than five is taken, a 100 pF capacitor from FB to IN COM is
required. At lower gains this capacitor is unnecessary, but it will
not adversely affect performance if used.
Adjustments.
When gain and zero adjustments are needed, the
circuit details will depend on whether adjustments are to be
made at the isolator input or output, and (for input adjust-
ments) on the input circuit used. Adjustments are usually best
done on the input side, because it is better to null the zero
ahead of the gain, and because gain adjustment is most easily
done as part of the gain-setting network. Input adjustments are
also to be preferred when the pots will be near the input end of
the isolator (to minimize common-mode strays). Adjustments
on the output side might be used if pots on the input side would
represent a hazard due to the presence of large common-mode
voltages during adjustment.
Figure 8a shows the input-side adjustment connections for use
with the “noninverting” connection of the input amplifier. The
zero adjustment circuit injects a small adjustment voltage in
series with the low side of the signal source. (This will not work
if the source has another current path to input common or if
current flows in the signal source LO lead). Since the adjust-
ment voltage is injected ahead of the gain, the values shown will
Figure 6. Input Connections for Gain > 1
(Circuit figures shown on this page are for SIP style pack-
ages. Refer to Page 3 for proper DIP package pinout.)
REV. B
–5–
AD [ ANALOG DEVICES ]
