Functional Description
MC9S08GB/GT Data Sheet, Rev. 2.3
Freescale Semiconductor
227
Full scale error (EFS) — This is the difference between the transition voltage to the last valid code
and the ideal transition to that code. Normally, it is dened as the difference between the actual and
ideal transition to code $3FF, but in some cases the last transition may be to a lower code. The ideal
transition to any code is:
Eqn. 14-8
Total unadjusted error (ETU) — This is the difference between the transition voltage to a given code
and the ideal straight-line transfer function. An alternate denition (with the same result) is the
difference between the actual transfer function and the ideal straight-line transfer function. This
measure of error includes inherent quantization error and all forms of circuit error (INL, DNL,
zero-scale, and full-scale) except input leakage error, which is not due to the ATD.
Input leakage error (EIL) — This is the error between the transition voltage to the current code and
the ideal transition to that code that is the result of input leakage across the real portion of the
impedance of the network that drives the analog input. This error is a system-observable error
which is not inherent to the ATD, so it is not added to total error. This error is:
EIL (in V) = input leakage * RAS
Eqn. 14-9
There are two other forms of error which are not specied which can also affect ATD accuracy. These are:
Sampling error (ES) — The error due to inadequate time to charge the ATD circuitry
Noise error (EN) — The error due to noise on VAIN,VREFH,orVREFL due to either direct coupling
(noise source capacitively coupled directly on the signal) or power supply (VDDAD,VSSAD,VDD,
and VSS) noise interfering with the ATD’s ability to resolve the input accurately. The error due to
internal sources can be reduced (and specied operation achieved) by operating the ATD
conversion in wait mode and ceasing all IO activity. Reducing the error due to external sources is
dependent on system activity and board layout.
(Current Code - 1/2)
Ideal Transition V =
2N
*(VREFH – VREFL)