August 27, 1998
GENERAL RELEASE SPECIFICATION
15.5 VOLTAGE MEASUREMENT METHODS
The various methods for obtaining a voltage measurement can use software tech-
niques to express these voltages as absolute or ratiometric readings.
NOTE
All A/D conversion methods should include a test for a maximum elapsed time in
order to detect error cases where the inputs may be outside of the design
specification.
15.5.1 Absolute Voltage Readings
The absolute value of a voltage measurement can be calculated in software by
first taking a reference reading from a fixed source and then comparing subse-
quent unknown voltages to that reading as a percentage of the reference voltage
multiplied times the known reference value.
The accuracy of absolute readings will depend on the error sources taken into
account using the features of the analog subsystem and appropriate software as
described in Table 15-6. As can be seen from this table, most of the errors can be
reduced by frequent comparisons to a known voltage, use of the inverted compar-
ator inputs, and averaging of multiple samples.
Table 15-6. Absolute Voltage Reading Errors
Accuracy Improvements Possible
Error Source
In Hardware
In Software
Calibration and storage of
reference source over temperature
and supply voltage
Change in
Reference Voltage
Provide closer tolerance reference
Change in Magnitude of
Ramp Current Source
Compare unknown with recent
measurement from reference
Not adjustable
Not adjustable
Non-Linearity of Ramp
Current Source vs. Voltage
Calibration and storage of voltages
at 1/4, 1/2, 3/4 and FS
Change in Magnitude of
Ramp Capacitor
Compare unknown with recent
measurement from reference
Provide closer tolerance ramp capacitor
Use external oscillator with crystal
Provide closer tolerance crystal
Use faster conversion times
Frequency Shift in Internal
Low-Power Oscillator
Compare unknown with recent
measurement from reference
Frequency Shift in
External Oscillator
Compare unknown with recent
measurement from reference
Sampling Capacitor
Leakage
Compare unknown with recent
measurement from reference
Compare unknown with recent
measurement from reference
OR avoid use of divided input
Internal Voltage
Divider Ratio
Not adjustable
Close decoupling at V and V pins
and reduce supply source impedance
Average multiple readings on both the
reference and the unknown voltage
DD
SS
Noise internal to MCU
MC68HC05SB7
REV 2.1
ANALOG SUBSYSTEM
MOTOROLA
15-27
FREESCALE [ Freescale ]
