Electronics > Beginners
Add a zero to this RC transfer function
<< < (3/6) > >>
MrAl:

--- Quote from: Wimberleytech on January 06, 2019, 08:20:18 pm ---
--- Quote ---
Hello again,

That looks like you have 3 zeros now?

However, how did you create that matrix?  That might have something to do with it.

I did this again and am now getting no extra zero as before.  What must have happened was i left that other resistor in the circuit so it already had 2 zeros before i started :-)

But anyway, i did it again and only get 1 zero.  Also, i checked it on a simulator this time and dont see a zero at the origin.

Consider this:
Make all caps the same value C, make all resistors the same value R.
Now make R=1 Ohm.
Now at low frequency the reactance of C is very large, so replace all C with 10 megohm resistors.
Now analyze the circuit for DC.  What do we  get?  We get:
Vout/Vin=1/3
So at near zero frequency we dont get zero output.
The simulator also shows this.

Maybe you could show how you got that matrix.

--- End quote ---

Actually, the numerator has two zeros, however, the zero at the origin is canceled by a pole at the origin, so there is only one zero.

You are correct--there is no additional zero added with the input capacitor...BECAUSE, there is no 'real' path to ground.  I missed that.  However, no practical circuit will have the output floating node as is the case with this circuit.

BTW, the matrix is generated with Symspice

--- End quote ---

Hi,

Yeah that reminded me i was going to ask the OP what this circuit was going to be used for.

We can connected a high resistance load to the output and re-evaluate i guess.  Might be interesting.
Wimberleytech:


--- Quote ---
Hi,

Yeah that reminded me i was going to ask the OP what this circuit was going to be used for.

We can connected a high resistance load to the output and re-evaluate i guess.  Might be interesting.

--- End quote ---

As soon as you connect it to ANY 'real' impedance (e.g. resistance), the zero at the origin will appear.  For any real circuit, there would have to be a 'real' resistance, otherwise it would not be properly biased.
bonzer:
Hello thanks a lot! I'm gonna try your answers as soon as I can as I also need some other conditions to  be verified. Anyway if you were asking yourself what I need this for it's about the compensation of a circuit. This is the transfer function corresponding to the feedback network of a op amp circuit that does the derivative. So this network is  Beta. I need that zero because in closed loop it creates a pole before the critical point of instability so this way I remove the overshoot. It's something that I still have to work on
MrAl:

--- Quote from: Wimberleytech on January 06, 2019, 10:54:46 pm ---

--- Quote ---
Hi,

Yeah that reminded me i was going to ask the OP what this circuit was going to be used for.

We can connected a high resistance load to the output and re-evaluate i guess.  Might be interesting.

--- End quote ---

As soon as you connect it to ANY 'real' impedance (e.g. resistance), the zero at the origin will appear.  For any real circuit, there would have to be a 'real' resistance, otherwise it would not be properly biased.

--- End quote ---

Hello again,

Yes and i think that is good insight there.  I would still want to look at the effectiveness of that zero however, because if we have a high impedance output then the zero is almost like it wasnt there.  I guess we'd then have to evaluate inside the application circuit.

Normally we like a zero that we have control over where we dont have to depend on the load to control.  That means instead of a numerator like:
(s+a)*s

we have:
(s+a)*(s+b)

and that gives us more control i think.  We could look at some frequency plots i guess but i have a feeling that with that one zero depending on the load value we could see a virtual zero at a frequency of 1e-12 Hz but at f=0.01 Hz it jumps right back up to 1/3 or so.



bonzer:
Meanwhile, @Wimberleytech I took a look at Symspice and it seems to be interesting for this stuff but it's not free. Anyway do you know other software like this that generate you the transfer function from the circuit? I know only how to do bode diagram. It would be really helpful to get the actual formula with components when dealing with complex rc networks to verify the result.

I'll be back with updates about my circuit in the next days.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod