The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Mixed-Signal Design >> use chopping in a SC sigma delta modulator
https://designers-guide.org/forum/YaBB.pl?num=1479483602

Message started by xlowen on Nov 18^th, 2016, 7:40am

Title: use chopping in a SC sigma delta modulator
Post by xlowen on Nov 18^th, 2016, 7:40am

Dear All,

From the measurement results (attached Fig) of a JSSC paper "A CMOS 110-dB@40-kS/s Programmable-Gain Chopper-Stabilized Third-Order 2-1 Cascade Sigma-Delta Modulator for Low-Power High-Linearity Automotive Sensor ASICs", it seems that choppping at lower frequency (Fs/2), some kind of noise folding happens. I built a similar design (2nd FF SC SDM with chopping used in the 1st stage) and saw similar results in my transistor-level simulation (no mismatch is added into simulation!). In many papers, people just choose Fs/2 as chopping frequency but never explain why. Could someone provide some insight on this issue?

Thanks!

Title: Re: use chopping in a SC sigma delta modulator
Post by deba on Nov 18^th, 2016, 8:11am

Due to chopping at fx, the shaped quantization is 2fx will fold inband and corrupts the inband noise floor. The noise is minimum at f=0,fs,2fs etc. Thus, if chopping is done at fs/2, the minimum corruption is obtained.

Title: Re: use chopping in a SC sigma delta modulator
Post by xlowen on Nov 18^th, 2016, 8:24am

Hi deba,

This is true for a continuous time SDM, however, for a switch cap design, if you chop at proper time (eg. middle of sampling phase), I didn't see any direct correlation between chopping action and feedback signal. If you look at this paper"On Chopper Effects in Discrete-Time ΣΔ Modulators", the author concluded that noise folding happens due to unmatched parasitic caps around OTA, but in my simulation, there is no mismatch but I still see noise folding...

Thanks,
Long

Title: Re: use chopping in a SC sigma delta modulator
Post by deba on Nov 19^th, 2016, 7:38pm

Hi,

I didn't realize its a SC DSM. I just looked at the PSD plots.

Yes for a SC DSM, if the chopping is done at an appropriate time as you had mentioned one should not see this noise foldover issue. But there are voltage dependent caps on the input differential pair of the opamp. I wonder whether that mismatch can result in a noise foldover.

Thanks