[SOLVED] Integrator - Pole-at-Zero

[rightmederek]

Hi,

I have a question about integrator.

Since the pole of an ideal integrator is at 0, in a bode plot the y-axis of the pole should be at infinite left, log0 = -∞, isn't it?

Thanks,
DD

 

[LvW]

Hi,
I have a question about integrator.
Since the pole of an ideal integrator is at 0, in a bode plot the y-axis of the pole should be at infinite left, log0 = -∞, isn't it?

Thanks,
DD

I suppose, you mean that the transfer function 1/sT approaches ∞ for s=0, correct?
Yes - that`s correct from the mathematical point of you.
However, in practice there is no IDEAL integrator. You always can realize a first-order lowpass only [A/(1+sT)] which can be used as an intergator - with suffficient accuracy - for frequencies well above the corner frequency (pole) wc=1/T.
For this reason, we try to keep this lowpass pole as low as possible.

 

[crutschow]

Yes, the gain goes towards infinity as the input frequency goes towards zero. That's why a real integrator will drift towards saturation of the amp at the voltage rails, even with no input, just due to the small input offset dc voltages in the real amp.