AD Converters are necessary to convert an analog signal to a digital one that our computers can understand. The output of an AD Converter is a binary number. Digital instruments most commonly use 12, 14 and 16-bit converters, although there are much higher bit converters available.
The following is a practice question from one of ISA’s classes:
1. Calculate the incremental steps for a 12-bit A/D converter
2. If the sensor connected to the A/D converter was ranged for 0-1000 F, what is the smallest increment of temperature that can be seen on the digital side of the A/D converter?
3. If I wanted to control the temperature to within 0.1 F can I do this with a 12-bit A/D converter in the sensor module?
4. What effect would a 14-bit A/D have?
5. Can you think of further improvements?
- Answer to 1.
LSB represents = 100% / 2^12 = 100%/4096 = 0.0244%
- Answer to 2.
LSB (or smallest increment of temperature on the digital side):
= 1000 F / 2^12 = 1000 F/ 4096 = 0.244 F
- A 12-bit A/D converter in the sensor module is not accurate enough.
- A 14-bit A/D converter would give a resolution or LSB of :
- = 1000 F / 2^14 = 1000 F / 16384 = 0.061 F
- 14-bit A/D converter would be better for control to within 0.1 F, however a 16-bit A/D converter would be even better because the resolution would be 0.0153 F.
Some points to remember are that although a higher bit converter allows for highly accurate instruments, we want our test equipment that we use to calibrate them to be at least 4 times more accurate. So for example if we had a transmitter with an accuracy of .02% FS, then our test equipment would have to have an accuracy of .005% FS. Also if the AD Converter on the host system’s analog input card (when the transmitter’s digital signal is being converted to 4-20) is not equal or higher in bits then the transmitter’s, the resolution that you gained is lost.