Original question:
Hi~
1. In ADS131M08EVM, the setting value is default, and the input of the channel is connected to GND.
A voltage of 50 uV appears on three channels. What is the reason?
2. When a voltage of less than 1mV is applied to evm with cal equipment, 0V is displayed. Is it not possible to measure less than 1mV?
3. ads13108 is 24bit and the resolution is 71nV. However, uV cannot be measured. What is the reason?
4. What is the lowest measure?
Thank you.
Hello kwon,
It looks like you're just starting to dive into the world of ADCs. So first, I'd like to say welcome!
If you have not reviewed:
I highly recommend you do so. A lot of the questions you are asking are fundamental to understanding how ADCs work. As such, I think it is better to refer you resources that already exist instead of answering your questions directly. However, I'll do my best to general summarize what you need to consider.
In general, your questions deal with the concept of noise and translating ADC codes to voltage. To start, the resolution of the ADC is not 71nV, it is 114nV. The blog below can help answer why.
But your point is taken, why can the ADC measure 114nV but it seems to measure uV when 0V is on the inputs? This is because of offset error. Offset error is a DC voltage that can be calibrated out so if you know the offset, you can subtract is from your reading. Here's a TIPL video on offset:
However, ~10uV is still a lot different from 114nV so what is it so much off? Well, that because the ADC's noise is larger than the 114nV. This is outlined in the specifications of the datasheet:
As shown in Table 7-1, depending on the Data-rate (OSR) and gain, you'll get different uVRMS noise values. You probably took the default so, with 0V on the input, you should see about 5.35uVRMS on the input from noise alone. We call the difference between the 114nV and 5.35uVRMS, quantization noise and thermal noise. You can learn more about that in the TIPL video linked below:
Does this mean we can't ever measure lower than 5.35uVRMS? The answer is we can probably measure just a little bit lower that the noise listed in the datasheet. But we need to know where noise comes from and how to eliminate it. This is explained in the noise series TIPL video series linked below:
So, I'd recommend knowing what the smallest voltage you want to measure, then pick an ADC that satisfies the data-rate (or bandwidth) requirement and gain for a given noise number listed in the datasheet. Its really not a good idea to pick an ADC where the inherent noise is around the level of the smallest signal that should be measured. In the case of the ADS131M0x family of ADCs we actually have data and a reference design that can measure 5uV with 0.147% accuracy: https://www.ti.com/tool/TIDA-010036
Have fun with the videos!
Best,
-Cole
Thank you for your answer.
I have a question in ek and gui measure value.
EVM KIT AINP/AINN = connected to GND.
In the GUI, the measured value is measured to be greater than TABLE 7-1, and the larger the OSR, the larger the value is measured.
(OSR16384=>measure max95uV, OSR2048=>measure max72uV, OSR256=>measure max38uV)
Shouldn't it be measured at 0V~ noise level?
It is measured by uV when connected by GND, but why is it not measured when 500uV power is applied?
Thanks.
