The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Pole zero doublets and gain boosted op-amps
https://designers-guide.org/forum/YaBB.pl?num=1481765281

Message started by tenso on Dec 14^th, 2016, 5:28pm

Title: Pole zero doublets and gain boosted op-amps
Post by tenso on Dec 14^th, 2016, 5:28pm

I was reading up on the topic mentioned in the thread title.

If I understand this right pole zero doublets which arise in a gain boosted cascode should be moved to a very high frequency to not degrade the settling time response of the amplifier.

I also understand that the GBW product of the auxiliary amp. should be much larger than the GBW of the overall loop/amplifier.

my questions:

1)doesn't this mean that the UGF of the additional/aux. amp should be higher than the UGF of the overall loop/amp.?

I was reading the Bult and Geelen paper mentioned in another thread here and I was confused by figure 5 in the paper.

2) if the answer to first question is yes, shouldn't ω4 here be ideally higher than ω5 and lower than ω6?

Title: Re: Pole zero doublets and gain boosted op-amps
Post by deba on Dec 17^th, 2016, 7:04am

Hi,

In a multi-pole and zero system, the relation between GBW and settling is not very well defined. Even if the GBW is high, due to non-monotonic phase droop, the settling may take a long time.

The criteria of moving the auxiliary loop dynamics beyond the GBW of the overall loop is just a safe criteria for stability. w4 > w5,w6 will try to maintain the first order behaviour of the overall loop.

It will help to show the opamp/circuit which you are working.

Title: Re: Pole zero doublets and gain boosted op-amps
Post by nrk1 on Dec 18^th, 2016, 5:56am

> I also understand that the GBW product of the auxiliary amp. should be much larger than the GBW of the overall loop/amplifier.

The unity loop gain frequency of the auxiliary amplifier should be more than the closed loop bandwidth βω5. That way settling would correspond to the pole βω5 and not something slower than that.

Title: Re: Pole zero doublets and gain boosted op-amps
Post by tenso on Dec 20^th, 2016, 8:43am

deba wrote on Dec 17^th, 2016, 7:04am:

this is the op-amp circuit from the Bult and Geelen paper. The main stage is a folded cascode amplifier. the 2 folded cascode with pmos input transistors are the auxiliary amplifiers. The current sources are also realized as cascoded structures with additional gain.

Title: Re: Pole zero doublets and gain boosted op-amps
Post by tenso on Dec 20^th, 2016, 8:46am

nrk1 wrote on Dec 18^th, 2016, 5:56am:

if I understand you right, then the GBW of auxuliary amplifier needs to be at least larger than the 3 dB band width of the entire closed loop.

I was under the impression that the GBW of the additional amplifiers should always be larger than the GBW of the entire closed loop for stability and settling time reasons.