The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> Other Measurements >> DFT and windowing
https://designers-guide.org/forum/YaBB.pl?num=1239186891

Message started by manodipan on Apr 8^th, 2009, 3:34am

Title: DFT and windowing
Post by manodipan on Apr 8^th, 2009, 3:34am

Hi all,
I am confused about how many bins to be taken in SNDR measurement....let's say no. of bins=1024 and OSR=64 for a delta sigma modulator....now for hanning window no. of nonzero signal bins=3 ...but what about no. of bins for offset and distortion components....normally how many bins are taken and how it's related to window type and length.....
Thanks a lot...
regards,
Manodipan

Title: Re: DFT and windowing
Post by ywguo on Apr 15^th, 2009, 8:42pm

Hi Manodipan,

The no. of bins need to be large enough in order to discern the offset, signal, and distortions. Saying no. of bins = 1024 and OSR = 64, there are 16 bins in the signal band. For hanning window no. of nonzero signal bins = 3, that setup is barely enough.

Yawei

Title: Re: DFT and windowing
Post by sheldon on Apr 18^th, 2009, 9:44pm

Manodipan,

There are some additional considerations if you want to get
a good estimate of the noise, see the Appendix A in Schrier's
book, "Understanding Delta-Sigma Data Converters". My
reading of Schrier is that you will need additional bins in the
FFT to get a good estimation of the noise. Also, if you are simulating
then the noise will only include the quantatization noise unless
you run a transient noise analysis.

Best Regards,

Sheldon

Title: Re: DFT and windowing
Post by manodipan on Apr 22^nd, 2009, 3:28am

Hi Sheldon,
yes you are right...because oversampling reduces no. of bins available in signal band.....also transient noise analysis reveals the information about noise floor...thanks for your reply.....