Delta sigma modulation using verilogA (and Cadence)

[sammyt09]

delta sigma modulation

Hi all,

I am having some problems with the verilogA simulation of a second order delta sigma modulator. I would be very greatful for any assistance.

I took the 'sigmadelta_1storder' cell from the Cadence 'ahdlLib' and modified it to be 2nd order. The resulting verilogA code is:

-------------------------------------------------------------------------------------

//
// Ideal 2nd order delta sigma ADC
//

`include "constants.vams"
`include "disciplines.vams"

module ADC_ideal(vin, vclk, vout);
input vin, vclk;
output vout;
electrical vin, vout, vclk ;
parameter real vth=0.0 ;
parameter real vout_high=1.8 ;
parameter real vtrans_clk=0.9 ;
parameter real trise=20n from (0:inf);
parameter real tfall=20n from (0:inf);
parameter real tdel=0.0 from [0:inf);

real vsum1, vsum2;
real vd;
real vint1, vint2;
real vout_val;
real hi, lo;

analog begin

@ ( initial_step ) begin
vout_val = 1.0;
hi=1.0;
lo=-1.0;
end

@ (cross(V(vclk)-vtrans_clk, 1.0))begin

// Difference 1
vsum1 = V(vin) - vd ;

// Integrator 1
vint1 = vsum1 + vint1;

// Difference 2
vsum2 = vint1 - vd ;

// Integrator 2
vint2 = vsum2 + vint2;

// Comparator
if (vint2 > vth)
vout_val = hi ;
else
vout_val = lo ;

// D2A
vd = vout_high * vout_val ;

end
V(vout) <+ transition(vout_val, tdel, trise, tfall);
end
endmodule

-------------------------------------------------------------------------------------


I then simulated the modulator using:
Vin = +/-1V sinusoid @ Fin = 200Hz
Fclk = 1MHz
Transient Analysis of 100ms
(The OSR is 512, but I have not yet considered the filter)

However, the performance appears to be very poor. It seems to "work", just not very well. As you can see from the attached plot below, the SNR of the output signal is less than 40dB. From my Matlab simulations, I believe an ideal modulator of this spec should result in a SNR of closer to 100dB.

So, I suppose I have a couple of questions:

1) Does anyone see a problem with the verilogA code I have provided? Is this correct for an ideal second order modulator?

2) Does anyone see an error with the method I am using to measure the SNR of the modulator output? As you can see from the plot, I performed a DFT over the 100ms adc output, using N=65536.

If anyone has any useful insights, it would be greatly appreciated. If you would like some more info regarding my setup, please ask.


Thanks in advance

Sammyt09

 

[safwatonline]

verilog a

what is the snr of the input

 

[sammyt09]

best fft window delta-sigma

Hi safwatonline,

Thanks for your response. I have not strictly measured the SNR of the input as it is a perfect sine wave (200Hz) with amplitude +/-1V.

Why do you ask? Perhaps I have missed something obvious here?




Thanks again

sammyt09

 

[safwatonline]

veriloga sigma delta code

I am not sure but i thought there might be some error in the FFT as the plot of the blue spectrum shows not that pure sine-wave, specially near the signal (or maybe my eye is just fooling me)

 

[sammyt09]

what is a delta sigma modulator

Thanks safwatonline,

I think that may be the case.

However, it has made me think that perhaps I am measuring the "SNR" of the output spectrum incorrectly. I dont think it is appropriate to simply perform an FFT on the output signal.

Anyone have a suggestion as to how to measure the SNR within Cadence?



Regards

sammyt09

 

[stefannm]

delta sigma modulator

Are you taking the FFT of an integer number of periods?

Are you using a windowing function?

 

[sammyt09]

2nd order delta sigma modulator

Hi stefannm,

Thanks for your input.

With regards to your questions:

1) No, I was simply performing the FFT over the entire 100ms transient response. How will not having an integer number affect my result?

2) Yes, I was using a hamming windowing function, which was the defualt window within the Cadence DFT function.

I look forward to your thoughts.



Thanks

sammyt09

 

[swethapr565]

Re: delta sigma modulation

hi
can you share your code with me .. please it will be help full for me.. please if u dont mind... i have designed the interfacing circuitry for accelerometers using matlab.. now i have a task to built it with synopsis..and i am beginner.. can u please help mee. by sharing your code with me..
please mail to: swethapr565@gmail.com
need it urgently please help me. im in shortage of time too.. please

 

[quoctaiduong]

Re: delta sigma modulation

Hi,

I also need this code. Please send to me too ( quoctaiduong@yahoo.com )or you can post here.

Thanks so much!