Datasheet

www.ti.com
CALIBRATION
DIGITAL FILTER
AdjustableDigitalFilter
DataOut
Modulator
Output
Fast-Settling
Sinc
2
Sinc
3
FILTERSETTLINGTIME
NOTE:(1)WithSynchronizedChannelChanges.
FILTER
SETTLINGTIME
(ConversionCycles)
Sinc
3
Sinc
2
Fast
3
(1)
2
(1)
1
(1)
AUTOMODEFILTERSELECTION
1 2 3 4
CONVERSIONCYCLE
Discard
Sinc
2
Sinc
3
Fast
ADS1216
SBAS171D – NOVEMBER 2000 – REVISED SEPTEMBER 2006
At the completion of calibration, the DRDY signal
goes low, which indicates the calibration is finished
The offset and gain errors in the ADS1216, or the
and valid data is available. See Application Report
complete system, can be reduced with calibration.
Calibration Routine and Register Value Generation
Internal calibration of the ADS1216 is called
for the ADS121x Series (SBAA099 ), available for
self-calibration. Self-calibration is handled with three
download at www.ti.com , for more information.
commands. One command does both offset and gain
calibration. There is also a gain calibration command
and an offset calibration command. Each calibration
process takes seven t
DATA
periods to complete. It
The Digital Filter can use either the Fast-Settling,
takes 14 t
DATA
periods to complete both an offset and
Sinc
2
, or Sinc
3
filter, as shown in Figure 27 . In
gain calibration. Self-gain calibration is optimized for
addition, the Auto mode changes the sinc filter after
PGA gains less than 8. When using higher gains,
the input channel or PGA is changed. When
system gain calibration is recommended.
switching to a new channel, it will use the
Fast-Settling filter for the next two conversions, the
For system calibration, the appropriate signal must
first of which should be discarded. It will then use the
be applied to the inputs. The system offset command
Sinc
2
followed by the Sinc
3
filter. This architecture
requires a zero differential input signal. It then
combines the low-noise advantage of the Sinc
3
filter
computes an offset that will nullify offset in the
with the quick response of the Fast-Settling time
system. The system gain command requires a
filter. See Figure 28 for the frequency response of
positive full-scale differential input signal. It then
each filter.
computes a value to nullify gain errors in the system.
Each of these calibrations will take seven t
DATA
When using the Fast-Settling filter, select a
periods to complete.
decimation value set by the DEC0 and M/DEC1
registers that is evenly divisible by four for the best
Calibration must be performed after power on, a
gain accuracy. For example, choose 260 rather than
change in decimation ratio, or a change of the PGA.
261.
For operation with a reference voltage greater than
(AV
DD
– 1.5V), the buffer must also be turned off
during calibration.
Figure 27. Filter Step Responses
17
Submit Documentation Feedback