The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Biasing of input differential pair in folded cascode OTA
https://designers-guide.org/forum/YaBB.pl?num=1298902714

Message started by Nandish Mehta on Feb 28^th, 2011, 6:18am

Title: Biasing of input differential pair in folded cascode OTA
Post by Nandish Mehta on Feb 28^th, 2011, 6:18am

Dear Friends,

The OTA in the figure below is the folded cascode. I am using it as an integrator.
On the positive terminal I have biased its input common mode by using level shifter designed from M1 & M2.

My question is how to bias the negative terminal. I tried leaving it unbiased and relying on fact that feed back will correct it. But it did not work.

Can any one please suggest me a way to bias the negative terminal?

Regards
Nandish

Title: Re: Biasing of input differential pair in folded cascode OTA
Post by philcorb on Feb 28^th, 2011, 6:58am

Hi,

The integrator is 'lossless'. If Vin and Vcm differ only slighly, the integrator will integrate to a rail. You could add a resistor in parallel with Cint, such that the dc gain (Rint/Rin) does not cause saturation. You could remove the switched-cap circuit and just use a resistor to understand the operation (use transient sims).

regards

Title: Re: Biasing of input differential pair in folded cascode OTA
Post by Nandish Mehta on Feb 28^th, 2011, 7:18am

Hi,

You are right that the integrator will integrate to rails, if there is any offset. However, this integrator will reset in fix intervals of time before it reaches rail. You are right that this integrator is lossless and that is why it is widely used in switch cap filters.

However, I am looking for some design suggestions on how to bias the input pair especially the negative terminal. Should I put a level shifter there too. I am not sure. This must be a problem with all OTA designs but I could not find any solution.

Please any one who knows point me an alternate.
Thanks
Nandish

Title: Re: Biasing of input differential pair in folded cascode OTA
Post by philcorb on Mar 1^st, 2011, 12:49am

Hi Nandish,

Sorry, I assumed you were asking about the biasing. If the input has a mid-rail common-mode, changing the connection of the switches to ground can do this. If one of the switches is connected to Vdd instead, then this will solve the problem.

Another approach is to add a second switched-cap section, that level shifts the input of the opamp. This can be done with capacitor connected to the negative input, the other end of which has two switches. One of the switches is connected to Vdd and the other to ground. The switches are on different phases.

Hope this answer is more helpful that my first!

regards

Title: Re: Biasing of input differential pair in folded cascode OTA
Post by Alexandar on Mar 1^st, 2011, 1:23am

Use a switch across Cint.

Title: Re: Biasing of input differential pair in folded cascode OTA
Post by Nandish Mehta on Mar 1^st, 2011, 5:41am

@philcorb: I will try to thing over your suggestion and try to see how to fit it in my OTA. Right now I have implemented an input level shifter similar to the one that I have for +VE terminal

@Alexandar: I can put a switch but do you think it will work if the output common mode and the input common mode are different.

Title: Re: Biasing of input differential pair in folded cascode OTA
Post by Rakesh on Mar 1^st, 2011, 1:31pm

Instead ofd coonecting the capacitors to ground connect to the bias voltage what u want