Accuracy of AD2

[Chazzo]

I am producing some documentation for a project I have been working on and I wanted to dig into the details of the accuracy for the AD2. I do understand ADQ Resolution, but I also understand that it doesn't describe the accuracy fully

Here are the specifications that I found on the website:

• Accuracy (scale≤0.5V/div, VinCM = 0V): ±10mV±0.5%

• Accuracy (scale≥1V/div, VinCM = 0V): ±100mV±0.5%

• CMMR (typical): ±0.5%

1. What is meant by "±10mV±0.5%"? Is this showing that it can be looked at two different ways?
2. Oh, and what is "VinCM" stand for?
3. Also, for the low gain side of the AD2 ±100mV would seem like a lot of potential error? In my testing while using the API set to -20V to +20V to test voltages, it seems plenty accurate with maybe about 5-10mV of noise. Albeit, maybe I have mislead myself thinking this is the case. Deliberate testing may adjust my expectations. 
4. What is meant by "CMMR"? I have been doing some Googling but I can only come up with CMR or "Common Mode Rejection" which is related to electrical signals. 

In general I am interested in how accurate the AD2 is and the specifications aren't quite making sense to me. Thanks for any input.

Edited May 24, 2022 by Chazzo

[attila]

Hi @Chazzo

These numbers are for the worst case but the typical should be much better.

1. +/-10mV offset error +/- 0.5% of actual value, like for 1V it can be from 0.985V to 1.015V
2. common-mode input voltage
4. common-mode rejection (ratio)

https://digilent.com/reference/test-and-measurement/analog-discovery-2/hardware-design-guide