The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Mixed-Signal Design >> cascade2-1 Sigma Delta modulator
https://designers-guide.org/forum/YaBB.pl?num=1156421128

Message started by caosl_zju on Aug 24^th, 2006, 5:05am

Title: cascade2-1 Sigma Delta modulator
Post by caosl_zju on Aug 24^th, 2006, 5:05am

hi,all
I have made a cascade2-1 sigma delta modulator as showed in the attached picture.Where parameters are:a1=a2=0.5,b1=2,c1=0.5,a3=0.5.
The sampling frequency is 1Mhz,input signal is 0.2sin(2*pi*200).But the output bitstream amplitude is (+2,-2),This is linux matlab simulation result. Surprisingly, on my windows matlab, some of the output bitstreams has exceeded (+20,-20). If i change the parameter c1=1,thus 1/c1 is 1 too, the bit stream is (+1 -1). The PSD of it not embody the third order noise shapping effect at all,and the SNDR is very low(36dB,OSR=128).
I have readed some papers about Sigma Delta modeling, but all the examples are sigle stage,one quantizer.
Anybody can help me to solve this problem, or is there mash sigma delta modulator's simulink model? any advice is appreciative.Thanks!

Best Regards!

Title: Re: cascade2-1 Sigma Delta modulator
Post by chungmnig on Aug 27^th, 2006, 10:58pm

Hi~~~
I simulated your spec.(c1=0.5 & C1=1) using Matlab.
and i found the SNDR is more than 100dB(not include any noise)
so.......i think your matlab diagram have some wrongs.
(i means your own matlab file has some problems , not your posted mash2-1.GIF)

And should pay attention to that the output swing of integrator not saturation.

Title: Re: cascade2-1 Sigma Delta modulator
Post by caosl_zju on Sep 2^nd, 2006, 6:47am

hi,chungmnig
Thanks a lot for your reply.I have modified my matlab program,I calculate the final output's SNDR,it also reached more than 100dB.
The swing of integrator is not very large.But in logic cancellation,because not realized with digital model, it's output swing is very high,for example,through the gain of 2 after comparator model, the output swing becomes (2,-2).Have you look into the output streams, the amplitude is very high.Is this has some imfluences on system's performance?
I felt in practice, this phenomena will be avioded for the digital realizition with digital adder,shifter and so on,so the swing will be not more than (vdd,0),ie (1,-1) in simulink model.But in simulink, ........Are there any methods to solve this? Thank you very much.

Title: Re: cascade2-1 Sigma Delta modulator
Post by fangjh on Apr 8^th, 2007, 3:46am

How much is the bit for your second stage (1bit or 3bit), I have simulation the model the result is not good, it is only 63.4db? I use the 3bit quantization.