davison7
Junior Member level 2
Hi~~Everybody,
Recently, I have designed a second-order Sigma-delta modulator using the CIFB architecture.
The fin = 1KHz, fs = 1MHz, and Vin = 1V.
In the first, I use the Matlab to simulate its function. The output spectrum shows that the noise floor at dc frequency is about -120dB.
But when I use the actual devices of the OPA, the Switch, and the clock generator into the Spectre for simulation, the low-frequency noise floor
rises at -90dB. It results in a worse SNR!
Could eveyone tells me that what reasons cause the low-frequency noise floor becomes worse and how to improve this problem?
Thank you very much
Recently, I have designed a second-order Sigma-delta modulator using the CIFB architecture.
The fin = 1KHz, fs = 1MHz, and Vin = 1V.
In the first, I use the Matlab to simulate its function. The output spectrum shows that the noise floor at dc frequency is about -120dB.
But when I use the actual devices of the OPA, the Switch, and the clock generator into the Spectre for simulation, the low-frequency noise floor
rises at -90dB. It results in a worse SNR!
Could eveyone tells me that what reasons cause the low-frequency noise floor becomes worse and how to improve this problem?
Thank you very much