Continue to Site

Welcome to

Welcome to our site! is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Filter for Sigma Delta ADC (Gm-c or Mosfet-c or Ladder)

Not open for further replies.


Newbie level 4
Jul 29, 2010
Reaction score
Trophy points
Activity points
Hi all,
I am new to IC design. I am designing a sigma delta modulator for a very low frequency application of 20 Hz(its a bio sensor so it needs to be low power also). Which is the best continuous time filter to use. ( ladder type, Gm-C or Mosfet-C ) or just a general amplifier with C in feedback.

Thanks in advance..


The filter type will be determined by your requirements of SNR. Are you implementing a continuous time or discrete time sigma delta? What is your target signal bandwidth,target SNR, desired filter order, OSR. Everything matters when trying to decide which filter is best for your application. Gm-C filters are fast but inherently non-linear since your application is low frequency and I will assume that you care about linearity, Iwon't recommend using this type.

The opamp based or active filters has good linearity and can achieve low power for low signals bandwidths, the requirement in gain or slew rate will depend if you are using continuos or discrete implementation for the sigma delta. But generally the opamp based filters are preferred.
Thanks Adrian,
The specifications are not yet decided as this is my research subject. Prof has asked me to get the best SNR and accuracy of 12 bits to start with. I am designing a continuous time sigma delta. The application supposed to be for vital signal detection (heartbeats) sensor application. The order is second order. I am looking for a filter which is most suitable for low freq as flicker noise will play a role at low frequencies.

Thanks again...

For 12 bits resolution you will need around 75dB of SNR, assuming that your signal bandwidth is around 100Hz, you can easily achieve that SNR with an OSR of 50 with a single bit quantizer and a second order loop. The main issue is your input signal level, since it is a sensor I will assume that the input level is very small, less than 10mV so your noise floor needs to be even at lower level, this means that you will need to increase your OSR or your filter order to achieve an effective 75 dB of SNR. if your input signal is at -40dB, then you will need your quantization noise, thermal and flicker noise together to lower than -115dB, then you will need an OSR around 300 to achieve that.

I will recommend you to go through the book "Understanding Delta-sigma Data Converters", Schreier and Temes. They cover pretty much all the basic stuff to start with. Then you can come with some conclusions and tradeoffs for your filter.
I am referring to David Jones Analog Design text. Is it ok to build a Gm-C filter followed by an amplifier to take care of the non linearity caused by the parasitic capacitance ( using miller architecture ) and having an OSR of say 600- 800. Meanwhile thanks for the book name. Its really helpful...


Not open for further replies.

Part and Inventory Search

Welcome to