请问关于pole-zero doublet的问题！

pandaoyzh

wo yao kan kan

gpgpbest

kankan

741852zxt

在单端输出的差放输出点的另一端，会产生一个非主极点，但同时会产生一个非主极点频率为2倍的零点，这就是极领点对，它对相位裕度的影响很小可以忽略。

analogicer

例如，在电路中，常用的一种方法是零点和极点抵消（左半平面），当两者不能完全重合但很近是，就存在一个偶合对。有些电路结构本身就具有这种情况，如gainboost电路。

741852zxt

以前一直以为零级点相消是正好相等，谢谢analogicer的解释

kyoabc

   

原帖由 analogicer 于 2008-11-5 10:10 发表

                               
登录/注册后可看大图

例如，在电路中，常用的一种方法是零点和极点抵消（左半平面），当两者不能完全重合但很近是，就存在一个偶合对。有些电路结构本身就具有这种情况，如gainboost电路。

若零极点的位置一样的话，是否也应该被称为“pole-zero doublet“呢？？？

peterlau1984

好好看看，感谢分享！！

yuxu9710108

主要是由电容引入的零点极点，这与频率有关

prgray

零级点对是gainboosted 必须考虑的问题

prof3

1. If pole and zero are exact the same, of course they are "transparent" to the signal path.
2. The pole-zero doublet effect requires pole and zero in the same s-plane(left s-plane). If zero in right plane while pole in left plane, even they have same value, it would not be a doublet.
3. Since pole-zero are very close, normally AC simulation can show its effects in the Amplitude-frequency/phase-frequency plot, however it does impact on transient response(linear part) as follows,

V(t)=Vo(1-k1*exp(-omegabw*t)+k2*exp(-omegaz*t)
where k2=(omegaz-omegap)/omegabw,
omegabw is the unity gain-bandwidth.

When omegaz=omegap, the linear step time constant depends on GBW only

4. You will never pay attention if the pole-zero doublet is at high frequency. E.g. like ahjua compensation, there is instrinsic pole-zero doublet and most of case regarded as fully cancelled and show very minor effects on transient response of the OPA with Ahjua compensation.

Hopefully you understand.