How to analysis poles and zeros in Analog cxt design?

[amara]

Poles and zeros

Can u explain how to analysis poles and zeros in Analog cxt design.

 

[harrytrinh]

Re: Poles and zeros

Hi,
You can make the equivalent circuit and then using current and voltage Kirchoft law to make the transfer function for a loop.
The transfer function often is writen as following:
H(s)=K*{(1-a1s1)(1-a2s2)..}/{(1-b1s1)(1-b2s2)}. where s=jw
When you have the transfer function you will get the poles and zeros.
where Pi=1/ai and Zi=1/bi
Simpler high impedance will make a pole and low impedance will make the zero.

correct me if I am wrong. Thanks inadvance.

 

[taofeng]

Re: Poles and zeros

I can not follow your description below:
" Simpler high impedance will make a pole and low impedance will make the zero. "

 

[wanily1983]

Poles and zeros

i think every node of the circuit has a pole,but in symmetrical structure the symmetrical nodes has the same pole. the zeros are always caused by feedthrough.

 

[strasno]

Re: Poles and zeros

Hi,
Multisim 9 has Pole Zero Analysis here it is some info about it

Pole Zero Analysis produces the real and imaginary coordinates of the poles and/or zeroes, depending on which analyses are enabled.

The Pole Zero Analysis provides precise results on circuits containing passive devices (resistors, capacitors and inductors). Circuits containing active devices (transistor or opamps) will not always display the expected results.

 

[ree]

Re: Poles and zeros

hello,

if you have a path between the input and the output this what'll make a zero

 

[airboss]

Re: Poles and zeros

to ree:

hi, can you explain why "if there's a path between the input and the output, that will make a zero"?

i was thinking that, a path between i/p & o/p, that means the i/p short with o/p. so, o/p voltage = i/p voltage. does that give me a zero?

can you help me on this point? thanks.

 

[harrytrinh]

Re: Poles and zeros

Yes, If there is a path from i/p to o/p then H(w)=0 that means give us a zero.

Before I said "high ipedance will make a pole and low impedance will make a zero"
just for simple way to say. Actually, which node that has infinite impedance (H(w) is infinite) then that node has a pole. And that is similar with zero. Do you agree with me?

 

[eda_freak]

Re: Poles and zeros

A fwd path contributes a zero and reverse path a pole....
Thus we can find by transfer function...dominant poles and zeros...and adapt the design accordingly...

 

[pixel]

Re: Poles and zeros

You typically have zero if there is fast path in parallel with slow path from input to output.

 

[dineshbabumm]

Re: Poles and zeros

If u use cascade,there wil b increase in the number of poles depending on how many stages you use.. If u cascade 2,then there will be an increase of 2 poles.. The transfer function analysis will make it clear.. But i am not strong in it..

 

[arebee]

Poles and zeros

This book has a new, easy method for finding poles and zeroes from various networks:
https://books.google.com/books?vid=...al+techniques&sig=UHtZtrHlLjk5l_p5hydDfu9JQPs

I highly recommend it.

 

[tomandrosea]

Re: Poles and zeros

Where can I get the book ?

Added after 8 minutes:

I think f compensation is very hard

 

[meghna]

Re: Poles and zeros

I was just reading this discussion and could not get some points:

I can understand how zero comes into picture. We say that there are two parallel paths from input to output and from these two paths, signal from input to output reaches in opposite phase, so they cancel each other and we get a zero there.

Can someone explain the effect of pole also in the similar manner. Why do we say that signal is infinite at pole frequency?

Added after 14 minutes:

I don't agree by saying that

'i was thinking that, a path between i/p & o/p, that means the i/p short with o/p. so, o/p voltage = i/p voltage. does that give me a zero? '

i/p is not shorted to o/p (in that case, transfer function is 1), o/p is shorted to ground (so the transfer function is 0)