REV. C
–5–
DEFINITION OF SPECIFICATIONS
Linearity Error
Linearity error refers to the deviation of each individual code
from a line drawn from “zero” through “full scale.” The point
used as “zero” occurs 1/2 LSB (1.22 mV for 10 V span) before
the first code transition (all zeroes to only the LSB “on”). “Full
scale” is defined as a level 1 1/2 LSB beyond the last code tran-
sition (to all ones). The deviation of a code from the true straight
line is measured from the middle of each particular code.
The K, B, and T grades are guaranteed for maximum nonlinear-
ity of ± 1/2 LSB. For these grades, this means that an analog
value that falls exactly in the center of a given code width will
result in the correct digital output code. Values nearer the upper
or lower transition of the code width may produce the next upper
or lower digital output code. The J and A grades are guaranteed
to ± 1 LSB max error. For these grades, an analog value that
falls within a given code width will result in either the correct
code for that region or either adjacent one.
Note that the linearity error is not user adjustable.
Differential Linearity Error (No Missing Codes)
A specification that guarantees no missing codes requires that
every code combination appear in a monotonic increasing sequence
as the analog input level is increased. Thus every code must have a
finite width. The AD674B and AD774B guarantee no missing codes
to 12-bit resolution, requiring that all 4096 codes must be present
over the entire operating temperature ranges.
Unipolar Offset
The first transition should occur at a level 1/2 LSB above analog
common. Unipolar offset is defined as the deviation of the actual
transition from that point. This offset can be adjusted as discussed
later. The unipolar offset temperature coefficient specifies the
maximum change of the transition point over temperature,
with or without external adjustment.
Bipolar Offset
In the bipolar mode the major carry transition (0111 1111 1111
to 1000 0000 0000) should occur for an analog value 1/2 LSB
below analog common. The bipolar offset error and temperature
coefficient specify the initial deviation and maximum change in
the error over temperature.
Quantization Uncertainty
Analog-to-digital converters exhibit an inherent quantization
uncertainty of ±1/2 LSB. This uncertainty is a fundamental
characteristic of the quantization process and cannot be reduced
for a converter of given resolution.
Left-Justified Data
The output data format is left-justified. This means that the
data represents the analog input as a fraction of full scale, rang-
ing from 0 to 4095/4096. This implies a binary point 4095 to
the left of the MSB.
Full-Scale Calibration Error
The last transition (from 1111 1111 1110 to 1111 1111 1111)
should occur for an analog value 1 1/2 LSB below the nominal
full scale (9.9963 V for 10.000 V full scale). The full-scale cali-
bration error is the deviation of the actual level at the last transi-
tion from the ideal level. This error, which is typically 0.05% to
0.1% of full scale, can be trimmed out as shown in Figures 7
and 8. The full-scale calibration error over temperature is given
with and without the initial error trimmed out. The temperature
coefficients for each grade indicate the maximum change in the
full-scale gain from the initial value using the internal 10 V
reference.
Temperature Drift
The temperature drift for full-scale calibration, unipolar offset,
and bipolar offset specifies the maximum change from the initial
(25°C) value to the value at T
MIN
or T
MAX
.
Power Supply Rejection
The standard specifications assume use of +5.00 V and ± 15.00 V
or ± 12.00 V supplies. The only effect of power supply error on
the performance of the device will be a small change in the
full-scale calibration. This will result in a linear change in all
low-order codes. The specifications show the maximum full-
scale change from the initial value with the supplies at the
various limits.
Code Width
A fundamental quantity for A/D converter specifications is the
code width. This is defined as the range of analog input values for
which a given digital output code will occur. The nominal value
of a code width is equivalent to 1 least significant bit (LSB) of the
full-scale range or 2.44 mV out of 10 V for a 12-bit ADC.
AD674B/AD774B
