M30245 Group
2.9.11 Absolute Accuracy and Differential Non-Linearity Error
• Absolute accuracy
Absolute accuracy is the difference between output code based on the theoretical A/D conversion
characteristics, and actual A/D conversion result. When measuring absolute accuracy, the voltage at
the middle point of the width of analog input voltage (1-LSB width), that can meet the expectation of
outputting an equal code based on the theoretical A/D conversion characteristics, is used as an ana-
log input voltage. For example, if 10-bit resolution is used and if V
then 1-LSB width becomes 5 mV, and 0 mV, 5 mV, 10 mV, 15 mV, 20 mV, ···· are used as analog input
www.DataSheet4U.com
voltages. If analog input voltage is 25 mV, "absolute accuracy = ± 3LSB" refers to the fact that actual
A/D conversion falls on a range from "002
pected from the theoretical A/D conversion characteristics. Zero error and full-scale error are included
in absolute accuracy.
Also, all the output codes for analog input voltage between V
Output code
(result of A/D conversion)
00B
00A
009
008
007
006
005
004
003
002
001
000
Figure 2.9.21. Absolute accuracy (10-bit resolution)
Rev.2.00 Oct 16, 2006
REJ09B0340-0200
16
16
16
16
16
16
16
16
16
16
16
16
0
5
10 15
Analog input voltage (mV)
page 235 of 354
" to "008
" though an output code, "005
16
16
+3LSB
–3LSB
20
25
30
35
40
(reference voltage) = 5.12 V,
REF
16
and AVcc becomes "3FF
REF
Theoretical A/D conversion
characteristic
45
50
55
2. A/D Converter
", can be ex-
".
16