SHC5320 Datasheet, PDF (5/10 Pages) Burr-Brown (TI) – High-Speed, Bipolar, Monolithic SAMPLE/HOLD AMPLIFIER

DISCUSSION OF
SPECIFICATIONS
WHAT IS A SAMPLE/HOLD AMPLIFIER?
A sample/hold amplifier (also sometimes called a track-and-
hold amplifier) is a circuit that captures and holds an analog
voltage at a specific point in time under control of an
external circuit, such as a microprocessor. This type of
circuit has many applications; however, its primary use is in
data acquisition systems which require that the voltage be
captured and held during the analog-to-digital conversion
process. Use of a sample/hold effectively increases the
bandwidth of a data acquisition system by a significant
amount. For further discussion of this capability, refer to
“Signal Digitization” in the Applications section of this data
sheet.
The ideal sample/hold amplifier in its simplest form contains
four primary components as illustrated in Figure 1, although
in actual practice they may not be internally connected
exactly as shown. Amplifier A₁, the input buffer, provides a
high impedance load to the source circuit and supplies
charging current to the holding capacitor C_H. Switch S₁
opens and closes under external control to gate the buffered
input signal to the holding circuit or to remove it so that the
most recently sampled signal will be held. Amplifier A₂
serves to present a high impedance load to the holding
capacitor and to provide a low impedance voltage source for
external loads. A minimum of three terminals are provided
for the user: input, output, and mode control (or sample/hold
control). When S₁, is closed, the output signal follows the
input signal, subject to errors imposed by amplifier band-
width and other errors as discussed below. When S , is
1
opened, the voltage stored on the holding capacitor will be
held indefinitely (in the ideal case), and will appear at the
output of the circuit until S₁, is again closed under command
of the mode control signal.
–
A₁
Input
+
Mode
Control
–
S₁
A₂
+
Output
C_H
FIGURE 1. Ideal Sample/Hold Amplifier.
The following discussion of specifications covers the critical
types of errors which may be experienced in applications of
a sample/hold amplifier. These errors are depicted graphi-
cally in Figure 2, and in the Typical Performance Curves.
Acquisition Time is the time required for the sample/hold
output to settle within a given error band of its final value
after the sample mode is initiated. Included in this time are
effects of switch delay time, slew rate of the buffer ampli-
fier, and settling time for a specified change in held voltage
value. Slew rate limitations of the buffer amplifier will cause
Settling
Time
Sample-to-Hold
Transient and
Charge Offset
Input
Feed-
though
Output
Droop
Offset
Aperture
Uncertainty
Aperture Time ^Acquisition
Hold
Time
Sample Mode Control
Slew Rate
Limited
Settling
Time
Sample
FIGURE 2. Illustration of Sample/Hold Specifications.
actual acquisition time to be highly dependent on the ampli-
tude of the voltage to be acquired, relative to the value
already held by the capacitor. Therefore, proper specifica-
tion of sample/hold amplifier performance includes defini-
tion of both output value step size and required error band
accuracy.
Aperture Time (or aperture delay time) is the time required
for switch S , to open and remove the charging signal from
1
the capacitor after the mode control signal has changed from
“sample” to “hold.” This time is measured from the 50%
point of the Hold mode transition to the time at which the
output stops tracking the input. This parameter is very
important in applications for which the input signal is
changing very rapidly when the Hold mode is initiated.
Effective Aperture Time is the difference in propagation
delay times of the analog signal and the mode control signal
from their respective input pins to switch S₁. This time may
be negative, zero, or positive. A negative value indicates that
the mode control propagation delay is shorter than the
analog propagation delay, with the result that the analog
value present on the capacitor at the time the switch opens
occurred earlier than the application of the mode control
signal by the amount of the effective aperture delay time.
Aperture Uncertainty (or aperture jitter) is the variation
observed in the aperture time over a large number of obser-
vations. This parameter is important when the analog input
is a rapidly changing signal, as aperture uncertainty contrib-
utes to lack of knowledge (at the output) about the true value
of the input at the precise time the Hold mode is initiated.
The maximum input frequency for a given acceptable error
contribution due to aperture uncertainty is
f_MAX = Maximum Fractional Error/2πt_U
where Maximum Fractional Error (MFE) is the ratio of the
maximum allowable error voltage to peak voltage, and t_U is
the aperture uncertainty time. For a bipolar ±10V signal and
a maximum uncertainty error of 1/2LSB in a 12-bit system,
the MFE is equal to 1/2LSB ÷ V_PEAK = 2.44mV ÷ 10V =
0.000244V/V, since 1/2LSB = 2.44mV for a 20V full-scale
range.
For the same system operating with a unipolar 0V to 10V
signal, MFE would be 0.000122V/V.
®
5
SHC5320