Doubt on Delta Sigma Modulators

Hi Jyothi,

Welcome to the e2e forum!  Your project sounds interesting!  Can you tell us a little more about how you are looking at the bit-stream outputs?  Assuming that both the first and second order modulators are being clocked the same way, the density of ones/zeros should be the same.  With an analog input at 50% of your full scale range, you should have an output that looks like a square wave with roughly a 50/50 duty cycle with either the 1st or 2nd order modulator.  If you are looking at the output spectrum, you should see difference.  Take a look at this application note which might help you to understand the difference.

http://www.ti.com/general/docs/lit/getliterature.tsp?literatureNumber=sbaa094&fileType=pdf

The app note above is based on the ADS1202, which is a second order D-S modulator used in current shunt applications.  The ADS1203 and ADS1204 are similar devices with different input voltage ranges.  You might take a look at the AMC1210 as well which implements various digital filter techniques for these D-S modulators allowing you to use up to four bit-stream input channels.

Hi Tom,

Thanks for the mail. Actually, we are looking at the bitstream output of the latched comparator of DS modulator. Input was a sine wave of 1KHz frequency and peak-peak amplitude of 1V centred at 2.5V for a supply of 0-5V. We have oversampled at 64KHz.  We have taken DC bias of the MOS circuit as 2.5V.

We have tried both 1st order and 2nd order and we observe different output bitstream at the latched comparator output. Are we wrong? We have not checked the output spectrum.

Can you tell us what is the kind of output you are expecting for the above specifications we mentioned. Also, can you kindly tell us how to measure the noise spectrum as shown in Fig. 5 in the app note you have shared (http://www.ti.com/general/docs/lit/getliterature.tsp?literatureNumber=sbaa094&fileType=pdf)  We are using the HSPICE tool for our experiments.

Thanks a lot.

Warm Regards,

Jyothi

Hi Jyothi,

It's difficult to say.  You're at 20% max input range so the data is going to be toggling - maybe you can send us a screen shot of your bitstream(s)? 

Hi Tom,

Sorry for this very late reply. We had vacations for 2 weeks. Thats why.  Attaching the output waveforms of both first order and second order DS modulator, that we obtained. We doubt if the problem is with the gain of the integrator. Please Help.

Output of First Order DS Modulator

Output of Second Order DS Modulator

Hi,

We were able to do the Delta Sigma modulator correctly and now we have the bitstream which has been successfully filtered using analog filter to get back the input.