In Datasheet of AD4112, 1st Page says "±10 V inputs", Page 27 Says (under VOLTAGE INPUTS)"enable an input range of ±20 V from a single 5 V power supply". Will Uni-polar setting give range of 0 to +40V & Bipolar will give +/- 20V?
AD4112 is specified with +/-10V differential inputs but could be functional up to +/-20V since the absolute input voltage pin can go from +/-20V. So the part can accept differential input voltage from +/-VREF but should meet the absolute input voltage at each VIN pin.
So in unipolar, it is 0 to +20V and not +40V. If you want +40V then you should have +/-20V at VINs.
If this is the case:
For Bi-polar mode : Code = 2(N – 1) × [(VIN × 0.1/VREF) + 1]
with Vref = 2.5V, For -10V : 223 x [ ((-10) x 0.1 / 2.5) + 1] = 223 x [0.6] = 8388608 (0.6 )= 5033165 = 0X5160AD
For +10V : 223 x [ ((10) x 0.1 / 2.5) + 1] = 223 x [-1.4] = 8388608 (1.4 )= 11744051 = 0xB33333
Difference in converted data = 6710886, (possible data values).
For Uni-polar mode :Code = 2(N – 1) × [(VIN × 0.1/VREF) ]
with Vref = 2.5V, For +10V= 16777216 (0.4 )= 6710886, (possible data values)
22 Bit resolution gives, 4194304 possible data values.
23 Bit resolution gives, 8388608 possible data values.
So, in reality, AD4112 gives, Resolution less than 23Bit, and NOT 24Bit as specified in the Data-sheet..
Am I correct ?
In unipolar mode, the ADC can only convert positive differential voltage and bipolar mode can convert both positive and negative differential voltage. I know this is something confusing with the actual analog input to consider if it is bipolar or unipolar. But the mode mentioned in the equation is more related to output coding. You can refer to this FAQ to see the difference.
In terms of resolution. There's always confusion between the data word and the actual ADC resolution. The 24 bit of any ADC in the data register indicates the number of bits it uses to digitized an input signal (data word). However, the effective number of bits is limited due to noise, this refers to the accuracy of the converter which determines how many bits in the digital output code represent useful information about the input signal. This parameter is specified as noise free bits (as shown on the RMS Noise and Resolution) of the datasheet. The value depends on the filter type, FS word or Output Data rate.
The formula in calculating the effective resolution is:
Effective resolution = log2(FS input voltage range/ADC RMS noise), thus, you'll see that the FS value really affects the actual resolution. The effective resolution is always 2.72 bits higher than the p-p resolution unless you are limited by the data output word
Thanks for the reply,
My doubt was regarding your quote "AD4112 is specified with +/-10V differential inputs but could be functional up to +/-20V".
Does this mean Conversion data is guaranteed up to "+/-10V differential inputs", it MAY work up to +/-20V ?