2017-08-01 07:53 AM
Hello,
Included attachement consists screen of internal reference parameters presented in datasheet.
I think there is mistake because Vdd=31.8V seems not correct.
I try to calculate how accurately can determine Vdda voltage but it is not obvious for me.
First parameter VREFINT presents voltage variantions(1.2 to 1.25V) with conditions from -40 to 105C.
But there is third line with the same parameter that shows that reference voltage viariation over temperature range is 10mV.
(similar value is get using TCoeff parameter)
I dont understand difference between this two parameters.
There is also another aspect with adc calibration value.
VREFINT_CAL is raw value measured at 30C with VDAA=3.3V, but there is no information about 3.3V accuracy.
My interpretation of this parameters is presented following.
During production internal reference value can differ from 1.2 to 1.25V. For example, we can assume that Vref=1.24V.
After calibration value is stored in VREFINT_CAL register during manufacturing.
I assume that voltage can differ from 1.24V about 10mV with temperature.
But this is inconsistent with first parameter.
How is correct interpretation?
#adc #embedded #internal #reference2017-08-01 01:04 PM
Hi Mateusz!
The first 50mv spreading is
Guaranteed
from ST. (Fact, at all voltage range)The 10mv max spreading (over this VDD small voltage range) is guaranteed by
design
, not tested in production.During the factory calibration VREFINT_CAL calcuated and put to rom, so after 3.3 V x VREFINT_CAL / VREFINT_DATA, the result be consistent with Garanteed spreading from ST (
obviously
true, else the guarantee would not be true!)