The AD4116 is specified for a TUE spec at ±10V full scale range. When input span is reduced the error will change as TUE is specified as a percentage of Full-scale range (%FSR).
To calculate the equivalent TUE spec for smaller ranges, we would take the assumption that gain-error and offset error both account for 50% of the total TUE. Please take note that this method provides only an estimated value, the actual performance could be better or worse as we only guaranteed the TUE spec at ±10V input range.
We can start by working through the errors in mV or V initially and then translating to a final TUE Specification
If we take an input span of ±10V and take the TUE from the datasheet specification for ±10V input at 25°C,192.5 µs settling.
For AD4116, TUE specs at 25°C,192.5 µs settling is specified as 0.1%. Thus,
Offset error = 0.05% of 20V span =>10mV
Full-Scale error at +10V (after correcting for offset error) = 0.05% of 20V span =>10mV
TUE = (10mV+10mV)/20 *100 = 0.1%
Table 2 shows the TUE estimation at 25°C across different ranges. The same approach can be applied when calculating TUE across different temperature.
Table 2. AD4116 TUE estimation at 25°C
Input span |
AD4116 |
±5V |
Offset error = 10mV; Full-Scale error = 10mV*(5V/10V) = 5mV; TUE = (10mV + 5mV)/10 *100 = 0.155% |
±2.5V |
Offset error = 10mV; Full-Scale error = 10mV*(2.5V/10V) = 2.5mV; TUE = (10mV + 2.5mV)/5*100 = 0.25% |
0 – 10 V |
Offset error = 10mV; Full-Scale error = 10mV*(10V/10V) = 10mV; TUE = (10mV + 10mV)/10 *100 = 0.2% |
0 – 5 V |
Offset error = 10mV; Full-Scale error = 10mV*(5V/10V) = 5mV; TUE = (10mV + 5mV)/5*100 = 0.3% |
0 – 2.5 V |
Offset error = 10mV; Full-Scale error = 10mV*(2.5V/10V) = 2.5mV; TUE = (10mV + 2.5mV)/2.5*100 = 0.5% |