AD636
APPLYING THE AD636
The input and output signal ranges are a function of the supply
voltages as detailed in the specifications. The AD636 can also
be used in an unbuffered voltage output mode by disconnecting
the input to the buffer. The output then appears unbuffered
across the 10 kΩ resistor. The buffer amplifier can then be used
for other purposes. Further, the AD636 can be used in a current
output mode by disconnecting the 10 kΩ resistor from the
ground. The output current is available at Pin 8 (Pin 10 on the
“H” package) with a nominal scale of 100
µA
per volt rms input,
positive out.
OPTIONAL TRIMS FOR HIGH ACCURACY
flows into Pin 10 (Pin 2 on the “H” package). Alternately, the
COM pin of some CMOS ADCs provides a suitable artificial
ground for the AD636. AC input coupling requires only capaci-
tor C2 as shown; a dc return is not necessary as it is provided
internally. C2 is selected for the proper low frequency break
point with the input resistance of 6.7 kΩ; for a cut-off at 10 Hz,
C2 should be 3.3
µF.
The signal ranges in this connection are
slightly more restricted than in the dual supply connection. The
load resistor, R
L
, is necessary to provide current sinking capability.
C
AV
– +
C2
3.3 F
V
IN
NONPOLARIZED
If it is desired to improve the accuracy of the AD636, the exter-
nal trims shown in Figure 2 can be added. R4 is used to trim the
offset. The scale factor is trimmed by using R1 as shown. The
insertion of R2 allows R1 to either increase or decrease the scale
factor by
±
1.5%.
The trimming procedure is as follows:
1. Ground the input signal, V
IN
, and adjust R4 to give zero
volts output from Pin 6. Alternatively, R4 can be adjusted to
give the correct output with the lowest expected value of V
IN.
2. Connect the desired full-scale input level to V
IN
, either dc or
a calibrated ac signal (1 kHz is the optimum frequency);
then trim R1 to give the correct output from Pin 6, i.e.,
200 mV dc input should give 200 mV dc output. Of course,
a
±
200 mV peak-to-peak sine wave should give a 141.4 mV
dc output. The remaining errors, as given in the specifica-
tions, are due to the nonlinearity.
C
AV
– +
1
2
3
4
5
V
OUT
R
L
to 1k
ABSOLUTE
VALUE
14
13
12
+V
S
0.1 F
AD636
SQUARER
DIVIDER
20k
11
CURRENT
MIRROR
10
0.1 F
6
–
9
BUF
10k
10k
7
Figure 3. Single Supply Connection
CHOOSING THE AVERAGING TIME CONSTANT
SCALE
FACTOR
ADJUST
V
IN
1
R1
200
1.5%
–V
S
ABSOLUTE
VALUE
14
13
12
11
+V
S
2
3
4
5
The AD636 will compute the rms of both ac and dc signals. If
the input is a slowly-varying dc voltage, the output of the AD636
will track the input exactly. At higher frequencies, the average
output of the AD636 will approach the rms value of the input
signal. The actual output of the AD636 will differ from the ideal
output by a dc (or average) error and some amount of ripple, as
demonstrated in Figure 4.
E
O
IDEAL
E
O
AD636
SQUARER
DIVIDER
CURRENT
MIRROR
10
9
R2
154
V
OUT
6
–
+V
S
R3
470k
R4
500k
–V
S
OFFSET
ADJUST
DOUBLE-FREQUENCY
RIPPLE
10k
7
BUF
10k
8
Figure 4. Typical Output Waveform for Sinusoidal Input
Figure 2. Optional External Gain and Output Offset Trims
SINGLE SUPPLY CONNECTION
The applications in Figures 1 and 2 assume the use of dual
power supplies. The AD636 can also be used with only a single
positive supply down to +5 volts, as shown in Figure 3. Figure 3
is optimized for use with a 9 volt battery. The major limitation
of this connection is that only ac signals can be measured since
the input stage must be biased off ground for proper operation.
This biasing is done at Pin 10; thus it is critical that no extrane-
ous signals be coupled into this point. Biasing can be accom-
plished by using a resistive divider between +V
S
and ground.
The values of the resistors can be increased in the interest of
lowered power consumption, since only 1 microamp of current
The dc error is dependent on the input signal frequency and the
value of C
AV
. Figure 5 can be used to determine the minimum
value of C
AV
which will yield a given % dc error above a given
frequency using the standard rms connection.
The ac component of the output signal is the ripple. There are
two ways to reduce the ripple. The first method involves using
a large value of C
AV
. Since the ripple is inversely proportional
to C
AV
, a tenfold increase in this capacitance will effect a tenfold
reduction in ripple. When measuring waveforms with high crest
factors, (such as low duty cycle pulse trains), the averaging time
constant should be at least ten times the signal period. For
example, a 100 Hz pulse rate requires a 100 ms time constant,
which corresponds to a 4
µF
capacitor (time constant = 25 ms
per
µF).
–4–
+
10k
8
39k
DC ERROR = E
O
– E
O
(IDEAL)
AVERAGE E
O
= E
O
+
TIME
REV. B
AD [ ANALOG DEVICES ]
