V
90=0.9(V
)
REF
REF_ADC
(5)
V
10=0.1(V
)
REF
REF_ADC
(6)
V
90=ADC
atV
90
MEAS
MEASUREMENT
REF
(7)
V
10=ADC
atV
10
MEAS
MEASUREMENT
REF
(8)
G
=
MEAS
V
90
V
10
-
MEAS
V
90
V
10
-
REF
(9)
O
=V
10
(V
10
G
)
-
MEAS
REF
MEAS
(10)
V
=AnyV ADC
AD_MEAS
IN
MEASUREMENT
(11)
V
=
ADC_CAL
V
O
-
AD_MEAS
MEAS
G
MEAS
(12)
IdealT
ransferF
unction
TransferFunction
withOffsetError+GainError
V
=+5V
REF
OffsetError=+4LSB
GainError=+6LSB
0FFFh(4.99878V)
(4.5114751443V)
(0.5056191443V)
0000h(0V)
0V
V
IN
DigitalOutput(V
)
AD
_MEAS
0.5V
(0.1
V
)
REF_ADC
4.5V
(0.9
V
)
REF_ADC
4.99878V
(V
1LSB)
REF_ADC -
www.ti.com ............................................................................................................................................ SBOS424B – MARCH 2008 – REVISED SEPTEMBER 2008
The 12-bit ADC example in
Figure 76 illustrates the
technique
for
calibrating
an
ADC
using
a
10%VREF_ADC
and
90%VREF_ADC
reading
where
VREF_ADC is the ADC reference voltage. Note that the
10%VREF reading also contains a gain error because
it is not a VIN = 0 calibration point. First, use the
2. Compute the ADC measured gain. The slope of
90%VREF and 10%VREF points to compute the
the curve connecting the measured 10%VREF and
measured gain error. The measured gain error is then
measured
90%VREF point is computed and
used to remove the gain error from the 10%VREF
compared
to
the
slope
between
the
ideal
reading, giving a measured 10%VREF number. The
10%VREF and ideal 90%VREF. This result is the
measured 10%VREF number is used to compute the
measured gain.
measured offset error.
3. Compute
the
ADC
measured
offset.
The
measured offset is computed by taking the
difference between the measured 10%VREF and
the (ideal 10%VREF) × (measured gain).
4. Compute the calibrated ADC readings.
Any ADC reading can therefore be calibrated by
removing the gain error and offset error. The
measured offset is subtracted from the ADC reading
and then divided by the measured gain to give a
Figure 76. 12-Bit Example of ADC Calibration for
corrected reading. If this calibration is performed on a
Gain and Offset Error
timed basis, relative to the specific application, gain
and offset error over temperature are also removed
from the ADC reading by calibration.
The gain error and offset error in ADC readings can
be
calibrated
by
using
10%VREF_ADC
and
For example; given:
90%VREF_ADC
calibration
points.
Because
the
12-Bit ADC
calibration is ratiometric to VREF_ADC, the exact value
ADC Gain Error = +6LSB
of VREF_ADC does not need to be known in the end
ADC Offset Error = +4LSB
application.
ADC Reference (V
REF_ADC) = +5V
Follow these steps to compute a calibrated ADC
Temperature = +25°C
reading:
1. Take the ADC reading at VIN = 90% × VREF and
Table 11 shows the resulting system accuracy.
VIN = 10% × VREF. The ADC readings for
10%VREF and 90%VREF are taken.
Table 11. Bits of System Accuracy(1) (to 0.5LSB)
ADC ACCURACY WITHOUT
ADC ACCURACY WITH PGA112
VIN
CALIBRATION
10%VREF_ADC
8.80 Bits
12.80 Bits
90%VREF_ADC
7.77 Bits
11.06 Bits
(1)
Difference in maximum input offset voltage for VIN = 10%VREF_ADC and VIN = 90%VREF_ADC is the reason for different accuracies.
Copyright 2008, Texas Instruments Incorporated
35