The 8-bit data is received from a sensor and later on, it is either directly converted into 12-bit data or rescaled into 12-bit data (amplification). After that, it is sent to a 12-bit DAC. The analog outcome (0 to 3.3V) is then amplified to a range of 0 to 15 V. Does this process add any quantization error or any kind of noise in the signal? Or is it just better to send the 12-bit converted data to the DAC directly and only use the OpAmp to amplify the signal later on? Any suggestion?
Thank you.
Dear, I think it's better to convert 8 to 12 to use the maximal DAC range. You can send 8 bits to DAC directly, but you will get too small analog signal! ~1/16 amplitude ratio between 8 bits and 12 bits. I recommend to left shift 8 bit data before sending it to DAC to achieve 16 gain amplification: DAC_12_data = ADC_8_data
The 12 bit data can be latched in to two SFR'S depending upon rescaling.
https://books.google.co.in › books
Embedded Microcontroller Interfacing: Designing Integrated Projects
I understand from the question and explanation that the sensor gives 8 bit data of low voltage span. The output is then converted into12 bits. Please make it clear whether the actual sensor output is analog or it is the type of module which provides only digital output in 8 bits. If it's module that provides 8 bit data then obviously there will not be any effect.
In other case if the output of sensor is analog then you can modify according to your requirement by suitable signal conditioning. But in this case you need ADC and not DAC. Please clarify more for discussion.