The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> OTA stability analysis cadence discrepancy
https://designers-guide.org/forum/YaBB.pl?num=1378621081

Message started by RSA on Sep 7^th, 2013, 11:18pm

Title: OTA stability analysis cadence discrepancy
Post by RSA on Sep 7^th, 2013, 11:18pm

Hi,
I designed a 2 stage Miller compensated OTA for a GBW of 10MHz and PM of 63 degrees (worst case) in Cadence Spectre. In order to measure the GBW and PM, I performed an AC analysis and got them to be right. Now, I created a symbol for the OTA, and enclosed it in a unity feedback loop. And I tried to measure the stability by using cmdmprobe in spectre (I am trying the stb analysis for the first time). And when i tried to plot the loop gain from the stb analysis, i get a loop gain only 30dB against 60dB I got from the earlier measurement. The GBW also is shown to be some 3MHz and i get 88 degree phase margin. Is this discrepancy to be expected or have I done something wrong? I have attached a png file depicting the way I set the simulation for stb analysis. Please help me out.. (In the setup I have uploaded I missed to include the load capacitance which i have in my schematic)

Title: Re: OTA stability analysis cadence discrepancy
Post by raja.cedt on Sep 8^th, 2013, 5:21am

How did you simulated first time (10MHz BW and 63 PM), is it open loop? If i was in your place i would believe 2nd results from STB method...

Thanx,
Raj.

Title: Re: OTA stability analysis cadence discrepancy
Post by RSA on Sep 8^th, 2013, 5:39am

Yes. That was an open loop measurement. I did an ac analysis and got phase margin from the calculator. So I thought both the analyses should give the same result. Why is this discrepancy?

Title: Re: OTA stability analysis cadence discrepancy
Post by raja.cedt on Sep 8^th, 2013, 10:43am

Hello,
Please understand the meaning of open loop , which is physically open loop with proper loading. Lucky stb method will take care of the loading, so what ever result you got from open loop are simply useless from closed loop point of view.

Thanks,
Raj.

Title: Re: OTA stability analysis cadence discrepancy
Post by harpoon on Sep 9^th, 2013, 7:36am

Hi RSA,

You are loading your OTA with resistors ... you can see the effect when you use different resistor values in your feedback.

You will still see some discrepencies between AC open loop and closed loop for the reasons raja.cedt mentioned ...

try adding an ideal output buffer between the OTA output and the loop.

also, have a look at this post ...
http://www.designers-guide.org/Forum/YaBB.pl?num=1378453275/0#0

Title: Re: OTA stability analysis cadence discrepancy
Post by RSA on Sep 10^th, 2013, 1:09am

Thanks both of you... I got it now.. :)

Title: Re: OTA stability analysis cadence discrepancy
Post by harpoon on Sep 11^th, 2013, 6:49am

Hi RSA,

You're welcome. You may wish to participate with the following thread.

It looks like you and RaDex might be doing the same thing ... maybe we can learn from each other ...

http://www.designers-guide.org/Forum/YaBB.pl?num=1378556694