continuous time sigma delta

[nijMcnij]

hello all,

i am trying to design a 3rd order continuous time (CT) delta sigma modulator (SDM).

first i want to get the model running in simulink.

my design methodology is as follows, design the in discrete time (DT) and map the coefficients to CT.
i started the design process by modeling a discrete time version of it in simulink, and it works perfectly.

then i applied the impulse invariant transformation to get the CT coefficients, but the damn thing does not work.....can anyone with sigma delta experience help out.

what is the best method to design a CT SDM, incase the method of transforming DT to CT is not the best one.

many thanks

 

[cgianfelici]

hello all,

Hi nijMcnij,
the impulse invariant transformation method works. Try with 2nd order modulator, first to understand it. For example, you can simulate a SD modulator with the H(z) TF:
H(z)= - (2z+1)/(z-1)²
Appling the impulse invariant transformation:
H(s)= -(1.5s+1)/(s)² to Ts=1
Then k1=-1.5 and k2=-1 for CT modulator system
We made the system with LTspice and it works perfectly.
I guess tht the systemin 3th ordeer is not more complicated.

Sorry for my english
Regards,


i am trying to design a 3rd order continuous time (CT) delta sigma modulator (SDM).

first i want to get the model running in simulink.

my design methodology is as follows, design the in discrete time (DT) and map the coefficients to CT.
i started the design process by modeling a discrete time version of it in simulink, and it works perfectly.

then i applied the impulse invariant transformation to get the CT coefficients, but the damn thing does not work.....can anyone with sigma delta experience help out.

what is the best method to design a CT SDM, incase the method of transforming DT to CT is not the best one.

many thanks

 

[baoer]

could you post your simulink file of CT SDM,
and we can find why?