Designer’s Guide Community

microe

Junior Member

Offline

Posts: 16

question on pole zero analysis
Oct 12^th, 2014, 10:22pm

Hi there,

I am trying to use pole zero analysis to find out the amplifier's open loop poles and zeros, but not sure how to set it up.

I used to use stb to check the stability of amplifier with load and feedback connected. It can gives me some idea of the pole and zeros from Bode plot, but not the exact location as pole/zero effect on mag/phase are mixed together. It is hard to identify unless they are far from each other.

That is the reason for me to try pole zero analysis. However, if I keep the feedback connected, what pole zero analysis tells me is the close loop pole/zero. What I am really interested in is open loop pole/zero.

If I break the loop to do pole/zero analysis, then the loading will be different (I understand picking a good breaking point will help to some degree).

I wonder if there is a way to set up pole/zero analysis to give me the open loop pole/zero but not have to break the loop, similar to stb analysis.

Your help is appreciated!

Back to top

IP Logged

raja.cedt

Senior Fellow

Offline

Posts: 1516
Germany

Re: question on pole zero analysis
Reply #1 - Oct 13^th, 2014, 1:54am

Hello,
Do you PZ analysis, it will give you so many pole even for a single stage opamp and search for dominate poles but still you have to think about which pole corresponds to which node.

Otherwise open the loop with large inductor(take care of the loading by adding caps on both sides of the ind) and find res and cap at each by injecting some ac current.

Thanks,
Raj.

Back to top

IP Logged

microe Junior Member Offline Posts: 16	Re: question on pole zero analysis Reply #2 - Oct 13^th, 2014, 6:17pm Hi Raj, Thanks. Is there a way to get the pole/zero without breaking the loop? Similar to stb can give us the loop mag/phase without breaking the loop Thanks.
Back to top	IP Logged

buddypoor

Community Fellow

Offline

Posts: 529
Bremen, Germany

Re: question on pole zero analysis
Reply #3 - Nov 7^th, 2014, 1:20am

microe wrote on Oct 13^th, 2014, 6:17pm:

Hi Raj,
Thanks.
Is there a way to get the pole/zero without breaking the loop? Similar to stb can give us the loop mag/phase without breaking the loop
Thanks.

Microe - if you are interested in open-loop poles/zeros why do you want to analyze the circuit under closed-loop conditions?
This I cannot understand.

Back to top

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

IP Logged

microe Junior Member Offline Posts: 16	Re: question on pole zero analysis Reply #4 - Nov 7^th, 2014, 6:39pm Thanks for the reply. It is the same reason as we connect amplifier in feedback when we do stb analysis: 1) we want it DC biased correctly 2) we want include the feedback loading
Back to top	IP Logged

buddypoor

Community Fellow

Offline

Posts: 529
Bremen, Germany

Re: question on pole zero analysis
Reply #5 - Nov 8^th, 2014, 2:28am

microe wrote on Nov 7^th, 2014, 6:39pm:

Thanks for the reply.
It is the same reason as we connect amplifier in feedback when we do stb analysis: 1) we want it DC biased correctly 2) we want include the feedback loading

OK - that`s what I have expected. However, for analyzing under open-loop conditioins you must open the loop (ac wise!) - otherwise, you have no open-loop conditions. On the other hand - I know what you means: You want to restore the dc operational point as well as the load properties.
To achive this, there are several methods - depending on the particular conditions.
For example, if you can neglect load changes (opamp output) the simplest method is to place an ac test source BETWEEN the low-resistive output and the feedback path.
As another simple method, you can use a very large inductor for ac opening the loop - and inject the test sighnal using a very large capacitor.
For a 100% correct method you must use the method as described by Middlebrook (open-loop analysis with double injection).
As far as I know, the STB analysis also makes use of Middlebrooks method.

Back to top

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

IP Logged