Designer’s Guide Community

refugee New Member Offline Posts: 9	Help with a stability problem Jan 04^th, 2010, 5:37pm I want to ask if the buffer in the loop will affect the close loop phase margin. If yes, how does it affect the close loop phase margin. (by 90 degree?) Thanks for your kindness help
Back to top	upload.JPG IP Logged

Mayank Community Fellow Offline Posts: 334	Re: Help with a stability problem Reply #1 - Jan 4^th, 2010, 8:42pm Hi, Anything in the signal path that has poles and zeroes will affect the phase margin...Why dont you just replace the buffer with it's Transfer Function and compute Phase margin from Overall Open Loop Transfer Function...which in your case is dependent only on Opamp Gain Response. -- Mayank.
Back to top	IP Logged

buddypoor Community Fellow Offline Posts: 529 Bremen, Germany	Re: Help with a stability problem Reply #2 - Jan 5^th, 2010, 1:19am Hi, sorry, but the whole circuit looks a bit "uncommon". If you apply 100% negative feedback to a stage with a gain of "1" you get an overall gain of 0.5. Is this your intention???
Back to top	LvW (buddypoor: In memory of the great late Buddy Rich) IP Logged

rajdeep

Senior Member

Offline

Posts: 220
UK

Re: Help with a stability problem
Reply #3 - Jan 5^th, 2010, 2:48am

Hi,

If I consider an ideal unity gain buffer (having infinite BW), then it can be said that loop gain of the overall ckt is always 0 DB, and the phase margin (of the loop gain) is 180deg!!!

If I consider a real unity gain buffer then the above statements remain true upto the BW of the unity gain buffer. Isnt it??

Also, why are you concerned abt the closed loop phase margin, when in the feedback path you have a direct connection i.e. no extra poles and zeros in the feedback path?

cheers!
Rajdeep

Back to top

Design is fun, verification is a requirement.

IP Logged

refugee

New Member

Offline

Posts: 9

Re: Help with a stability problem
Reply #4 - Jan 7^th, 2010, 5:55am

Mayank wrote on Jan 4^th, 2010, 8:42pm:

Hi,
Anything in the signal path that has poles and zeroes will affect the phase margin...Why dont you just replace the buffer with it's Transfer Function and compute Phase margin from Overall Open Loop �Transfer Function...which in your case is dependent only on Opamp Gain Response.

--
Mayank.

thank you, mayank. I think I got your idea

Back to top

IP Logged

refugee

New Member

Offline

Posts: 9

Re: Help with a stability problem
Reply #5 - Jan 7^th, 2010, 6:02am

buddypoor wrote on Jan 5^th, 2010, 1:19am:

Hi,

sorry, but the whole circuit looks a bit "uncommon".
If you apply 100% negative feedback to a stage with a gain of "1" you get an overall gain of 0.5. � �Is this your intention???

sorry for the simplified diagram. �there are some extra gm block in the forward path. in fact, the output is a large power device. so i need a buffer to drive it

Back to top

IP Logged

buddypoor

Community Fellow

Offline

Posts: 529
Bremen, Germany

Re: Help with a stability problem
Reply #6 - Jan 7^th, 2010, 6:02am

refugee wrote on Jan 7^th, 2010, 5:55am:

Mayank wrote on Jan 4^th, 2010, 8:42pm:

Hi,
Anything in the signal path that has poles and zeroes will affect the phase margin...Why dont you just replace the buffer with it's Transfer Function and compute Phase margin from Overall Open Loop Transfer Function...which in your case is dependent only on Opamp Gain Response.

--
Mayank.

thank you, mayank. I think I got your idea

Yes, everything which was said about the stability of the circuit is OK.
But, REFUGEE, did you realize that the overall gain is - 6dB ??
Is this really your intention?

Back to top

LvW (buddypoor: In memory of the great late Buddy Rich)

IP Logged

loose-electron Senior Fellow Offline Best Design Tool = Capable Designers Posts: 1638 San Diego California	Re: Help with a stability problem Reply #7 - Jan 7^th, 2010, 7:58am Its a simplified conceptual drawing. Loosen up!
Back to top	Jerry Twomey www.effectiveelectrons.com Read My Electronic Design Column Here Contract IC-PCB-System Design - Analog, Mixed Signal, RF & Medical IP Logged

buddypoor Community Fellow Offline Posts: 529 Bremen, Germany	Re: Help with a stability problem Reply #8 - Jan 7^th, 2010, 8:33am loose-electron wrote on Jan 7^th, 2010, 7:58am: Its a simplified conceptual drawing. Loosen up! In this case, I would say: It is dangerous and of no value to speak about stability of a system which is given only as a "simplified conceptual" drawing. Don`t you agree?
Back to top	LvW (buddypoor: In memory of the great late Buddy Rich) IP Logged

loose-electron

Senior Fellow

Offline

Best Design Tool =
Capable Designers

Posts: 1638
San Diego California

Re: Help with a stability problem
Reply #9 - Jan 8^th, 2010, 1:28pm

How about - "Could you please provide a drawing that shows some more detail of what you are trying to do?"

buddypoor wrote on Jan 7^th, 2010, 8:33am:

loose-electron wrote on Jan 7^th, 2010, 7:58am:

Its a simplified conceptual drawing.

Loosen up!

In this case, I would say: It is dangerous and of no value to speak about stability of a system which is given only as a "simplified conceptual" drawing. Don`t you agree?

Back to top

Jerry Twomey
www.effectiveelectrons.com
Read My Electronic Design Column Here
Contract IC-PCB-System Design - Analog, Mixed Signal, RF & Medical

IP Logged