EFM32GG232 Datasheet, PDF (31/71 Pages) Silicon Laboratories – High Performance 32-bit processor

...the world's most energy friendly microcontrollers
Symbol
Parameter
Condition
Min
Typ
Max
Unit
GAIN_ED
Gain error drift
1.25V reference
2.5V reference
0.01²
0.01²
0.033³ %/°C
0.03³ %/°C
OFFSET_ED Offset error drift
1.25V reference
2.5V reference
0.2²
0.7³ LSB/°C
0.2²
0.62³ LSB/°C
1_{On the average every ADC will have one missing code, most likely to appear around 2048 +/- n*512 where n can be a value in}
the set {-3, -2, -1, 1, 2, 3}. There will be no missing code around 2048, and in spite of the missing code the ADC will be monotonic
at all times so that a response to a slowly increasing input will always be a slowly increasing output. Around the one code that is
missing, the neighbour codes will look wider in the DNL plot. The spectra will show spurs on the level of -78dBc for a full scale
input for chips that have the missing code issue.
2_{Typical numbers given by abs(Mean) / (85 - 25).}
3_{Max number given by (abs(Mean) + 3x stddev) / (85 - 25).}
The integral non-linearity (INL) and differential non-linearity parameters are explained in Figure 3.17 (p.
31) and Figure 3.18 (p. 32) , respectively.
Figure 3.17. Integral Non-Linearity (INL)
Digital ouput code
4095
4094
4093
4092
INL= | [(V_D- V_SS)/ V_LSBIDEAL] - D| where 0 < D < 2^N - 1
Actual ADC
tranfer function
before offset and
gain correction
Actual ADC
tranfer function
after offset and
gain correction
INL Error
(End Point INL)
3
Ideal transfer
curve
2
1
VOFFSET
0
Analog Input
2016-03-21 - EFM32GG232FXX - d0125_Rev1.40
31
www.silabs.com