Understanding Delta Sigma Modulation in simulink

[jack3128]

Hi Guys, i'm new in this forum and hope to learn from u you all ^^. I'm learning 2nd order sigma delta Modulation that simulate in Matlab/simulink but i dont understand the block funtion such as why use Gain, why feed back loop , why use integral and so on. Can some one help explain the block diagram plz. :grin:

 

[jgk2004]

Hi Jack,

A sigma delta modulator consists of one feedback loop around an integrator and a quantizer. This integrator (since you have 1/S therefore continuous) producing a low pass filtering of the input signal and a high pass filtering of the quantization noise which is injected at the quantizer, hense noise shaping the noise out of band. Since you have a second order system, you have 2 integrators and two feedback paths. (You can have feedforward or feed ins but this is more advanced, please reference DSM books). Additionally, since you now have a chain of integrators, you therefore needs coefficient scaling of your integrator gains to ensure stability. That is why the gain blocks are used. If you don't have the Gain block infront, your integrators have a gain of 1.. hence 1/S. This scaling is then used, to keep your integrator swings within limits of your designs. This can be seen when looking at the scopes you have probing each node.

Hope this helps,
JGK

 

[jack3128]

Hi Jack,

A sigma delta modulator consists of one feedback loop around an integrator and a quantizer. This integrator (since you have 1/S therefore continuous) producing a low pass filtering of the input signal and a high pass filtering of the quantization noise which is injected at the quantizer, hense noise shaping the noise out of band. Since you have a second order system, you have 2 integrators and two feedback paths. (You can have feedforward or feed ins but this is more advanced, please reference DSM books). Additionally, since you now have a chain of integrators, you therefore needs coefficient scaling of your integrator gains to ensure stability. That is why the gain blocks are used. If you don't have the Gain block infront, your integrators have a gain of 1.. hence 1/S. This scaling is then used, to keep your integrator swings within limits of your designs. This can be seen when looking at the scopes you have probing each node.

Hope this helps,
JGK

Hey Thx alot for the information u really help me alot by explain these block TQ VERY MUCH ^^:lol: