The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> delta sigma modulator simulation
https://designers-guide.org/forum/YaBB.pl?num=1433874357

Message started by tul84 on Jun 9^th, 2015, 11:25am

Title: delta sigma modulator simulation
Post by tul84 on Jun 9^th, 2015, 11:25am

Hello everyone,

i am trying to simulate the accuracy of a delta sigma modulator (first order). It is a fully-differential switched capacitor integrator with a latched comparator. Transient simulation works, the question is how to evaluate the bitstream to get the accuracy for a given DC input voltage. Is it possible to perform a PSS, PAC or Pnoise simulation of the modulator?

Thank you in advance.

Title: Re: delta sigma modulator simulation
Post by RobG on Jun 9^th, 2015, 12:43pm

With a 1st order the decode is just a counter. For example, if you take 4096 samples and 100 of them are high, then your converted input is 100/4095*Vref.

Title: Re: delta sigma modulator simulation
Post by tul84 on Jun 9^th, 2015, 11:41pm

Thank you.
What have you used for counting matlab or a model directly in cadence.
Any idea concerning PSS, PAC or Pnoise simulation.

Title: Re: delta sigma modulator simulation
Post by RobG on Jun 10^th, 2015, 10:48pm

tul84 wrote on Jun 9^th, 2015, 11:41pm:

Thank you.
What have you used for counting matlab or a model directly in cadence.
Any idea concerning PSS, PAC or Pnoise simulation.

Just normalize the bitstream to 0 and 1 and take the average using your favorite method, that will give you (counts high)/(total counts).

PSS can be used to get spectrums, not sure about DC values... it is on my list of things to learn.

Title: Re: delta sigma modulator simulation
Post by tul84 on Jun 11^th, 2015, 11:07pm

I have some experience with PSS, PAC and PSTB applied to fully-differential OTA's with SC-CMFB. I have also tried it now for the DS modulator but it did not converge.
For Simulation i have now written a Verilog-A model for counting the one's the bitstream for 2^N samples. Works fine, but is not satisfying in therms of simulation time and disk space.

Title: Re: delta sigma modulator simulation
Post by sheldon on Jul 1^st, 2015, 1:33am

Tul84,

In general, PSS can't be applied to a delta sigma modulator because
their behavior is non-periodic. This is the same reason that you need
to use the Hann window function for the FFT instead of the rectangular
window, as you can do for a Nyquist rate ADC.

Sheldon

Title: Re: delta sigma modulator simulation
Post by RobG on Jul 3^rd, 2015, 8:40am

My partner uses PSS for continuous time DSMs. He models the comparator with an ideal unity gain buffer so there isn't any switching. I don't know if it would work with switched capacitor designs unless you did a continuous time equivalent. Additionally, I don't know how accurate his results are but they seem to be ballpark.