The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Passive CT sigma delta ADC in cadence
https://designers-guide.org/forum/YaBB.pl?num=1496944412

Message started by skamlesh on Jun 8^th, 2017, 10:53am

Title: Passive CT sigma delta ADC in cadence
Post by skamlesh on Jun 8^th, 2017, 10:53am

Hi,
I have simulated passive CT sigma delta ADC in MATLAB with,
fin = 1e3Hz fs = 10e6Hz and I ran the simulink simulation for N*Ts time duration where N = 2^20 and Ts = 1/fs the data was recorded on every Ts second.

For running same simulation in cadence (IC6) what settings I should put in transient analysis specially keeping in mind that it takes alot of time to finish the simulation. Can I run the simulation for lesser duration instead of N*Ts like I did in simulink and still obtain desired results?

I have one more doubt since the input signal is periodic the output waveform is also going to be periodic, so in that case can I run the simulation for lesser time and just concatenate the array to obtain the N*Ts duration data and perform FFT.

Also I would be very grateful if you could guide me about strobe period setting and calculating FFT using spectre specifically to sigma delta ADC with very large OSR like in my case.

I have attached MATLAB simulink result file herewith: yellow curve is for fin = 1e3 and purple curve is for fin = 107/N*fs
which one should I consider for SNR and SNDR measurements?

Thank you.

Title: Re: Passive CT sigma delta ADC in cadence
Post by sheldon on Aug 3^rd, 2017, 10:43am

Skamlesh,

Use the standard setup for data converters.
fclk = 10MHz
fin = (105/2^20)*10MHz ~1001Hz
Tstop = (2^20 + a few cycles)/10MHz

Use the cosine^2 window for the FFT.

Since you have an ADC, strobing is not required. The benefit
is primarily reducing the amount of data saved.

The attached plot shows an extreme example, it is a four stage cascade,
multi-bit ADC.

Sheldon