
MB90220 Series
78
A/D Converter Glossary
Resolution
Analog changes that are identifiable with the A/D converter
When the number of bits is 10, analog voltage can be divided into 2
10
= 1024.
Total error
Difference between actual and logical values. This error is caused by a zero transition error, full-scale transition
error, linearity error, differential linearity error, or by noise.
Linearity error
The deviation of the straight line connecting the zero transition point (“00 0000 0000”
“00 0000 0001”) with
the full-scale transition point (“11 1111 1111”
“11 1111 1110”) from actual conversion characteristics
Differential linearity error
The deviation of input voltage needed to change the output code by 1 LSB from the theoretical value
Note: The values shown here are reference values.
Analog input
Comparator
C
1
C
0
R
ON2
R
ON1
R
ON1
: Approx. 1.5 k
R
ON2
: Approx. 1.5 k
C
0
: Approx. 60 pF
C
1
: Approx. 4 pF
Analog Input Circuit Mode
V
FST
V
NT
V
0T
1LSB
V
FST
– V
0T
1022
=
Linearity error
V
NT
– (N
×
1LSB + V
0T
)
1LSB
=
Differential linearity error
V
NT
– V
(N–1)T
1LSB
=
Total error
Digital output
11 1111 1111
11 1111 1110
11 1111 1101
N + 1
N
N – 1
00 0000 0010
00 0000 0001
00 0000 0000
– 1
V
(N + 1)T
N = 0 to 1022
V
NT
(N = 0) = V
0T
V
NT
(N = 1022) = V
FST
N = 1 to 1022
Linearity error
N
×
1LSB + V
0T
V
(N – 1)T
V
1T
V
2T
Theoretical value (V
NT
)
Theoretical value
Actual conversion value
AVRL
AVRH (V)
, 1LSB theoretical value
AVRH
– AVRL
1022
=
Total error
V
NT
– {(N + 0.5)
×
1LSB theoretical value}
1LSB theoretical value
=
N = 0 to 1022