sigma delta ADC中 第一级采样电容大小确定

下场噗噗雨

你的adc是几位量化呢

cfqlyx

下场噗噗雨 发表于 2024-10-18 15:29
你的adc是几位量化呢

1位

qqlpp

电容10几pF就已经很大了吧，实际做起来SNR很难达到120dB

zixin1hao

学习。。。。。

igolaps

thanks

cfqlyx

chungming 发表于 2024-10-19 09:00
as far as i know, the input signal should be 2.5V and differential. So, the input power will be Vpow ...

thanks,maybe you are right,but i still have a question,Why does the simulation result of my simulink model also need such a large capacitance, because I did not reflect the voltage amplitude when modeling?

castrader

cfqlyx 发表于 2024-10-20 10:39
thanks,maybe you are right,but i still have a question,Why does the simulation result of my simuli ...

信号幅度是一方面，公式里面的kT/OSR*C其实不够准确，应该有个系数，而这个系数很可能介于2-4之间，所以最后的电容值应该在你俩计算值之间。

sea11038

计算用的VDD有问题吧？VDD=1V和VDD=5V，信号功率差了25倍，计算的电容可是也差了25倍。

chungming

cfqlyx 发表于 2024-10-20 10:39
thanks,maybe you are right,but i still have a question,Why does the simulation result of my simuli ...

i think simply because your feedback signal voltage is 1, that's 1V, it means the VREF=1V.

might need to change the feedback voltage to be 2.5V or above, then the input signal power could be higher.
So, you can use smaller Cs for the sampling capacitor. this is my guess.

BTW, i suggest you to refer to some 24-bit sigma-delta ADC in real product.
for instance: https://www.analog.com/media/en/ ... sheets/AD7176-2.pdf

cfqlyx

sea11038 发表于 2024-10-20 12:04
计算用的VDD有问题吧？VDD=1V和VDD=5V，信号功率差了25倍，计算的电容可是也差了25倍。 ...

对的，目前看了确认是信号功率计算小了