• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Pi Filter Cut Off

Status
Not open for further replies.

engineergc

New Member
Dear all;
I want to use a cascaded pi filter in my design. But I cannot find any information how I can calculate cut off freqeuncy of 2nd order pi filter?
 

MikeMl

Well-Known Member
Most Helpful Member
I use this program. The author is now deceased, but the program is very good.
 

engineergc

New Member
I calculated value of inductors and capacitors by using online calculators and document that I attached to post. I want to obtain 17 kHz cut off frequency. According to document and online calculator value of capacitors and inductor are given below. Since I have restriction in component heights in my PCB, I used these values L1 and L2 4.7uH, C1 and C3 4.7uF and C2 is 22 uF. According to these values I am getting 17Khz cut off frequency in LTSpice. According to document values that I used in LTSpice is not correct, but I get correct cut offf frequency. If cut off freqeuncy of filter depends on formulas how I can get correct cut off freqeuncy in LTSpice with wrong component values? Regards.

*I am using Butterworth Filter.
http://ecee.colorado.edu/~mathys/ecen2420/pdf/UsingFilterTables.pdf
 

Attachments

Nigel Goodwin

Super Moderator
Most Helpful Member
Build it and see how it works, if you want to play computer games try an XBox :D

However, I would suggest that your chosen values probably won't make a very good filter, the inductor is too small and the capacitors too high.

What are you actually trying to do?.
 

MikeMl

Well-Known Member
Most Helpful Member
V1 needs to have a 1Ω source resistance to match the network expected by the formula.

You say you want a second order design, but you show a 5rd order.

The 1Ω input and output resistance makes no sense to me, but if that is what you really, really want, then try this:

lp.png

lp1.png
 
Last edited:

MrAl

Well-Known Member
Most Helpful Member
Hi,

Yes, for a fifth order filter as most others we need to know both the input impedance and the output impedance.

For the values given with zero input source resistance we get about 25826.55Hz for the -3db point while with an input source resistance equal to 1 ohm to match the output resistance we get about 29390.16Hz for the -3db cutoff point.
Also the zero ohm source resistance has a MUCH different response curve being partly like a bandpass filter rather than a low pass filter, while the input resistance equal to 1 ohm response is more like just a low pass.
 

Mikebits

Well-Known Member
I think I have seen 1 ohm impedance used in un-normalized filter cookbook designs. Could that be where the 1 ohm came from? Otherwise a 1 ohm impedance makes no sense.
 

MrAl

Well-Known Member
Most Helpful Member
I think I have seen 1 ohm impedance used in un-normalized filter cookbook designs. Could that be where the 1 ohm came from? Otherwise a 1 ohm impedance makes no sense.
Hi,

Yes, 1 ohm is used in filter designs that are meant to be scaled up later. It's actually the "normalized" version, normalized to 1 ohm.
For example, in that design if we have C=50uf and we want 50 ohms instead of 1 ohm then the new value of C is C/50=1uf. For an inductor, if the value is 1uH with 1 ohm then the new L with 50 ohms is 50*1uH=50uH.
There's also the assumption that Rin=Rout.
 

Mikebits

Well-Known Member
Yes, 1 ohm is used in filter designs that are meant to be scaled up later. It's actually the "normalized" version, normalized to 1 ohm.
I had a feeling I got that bass ackwards, thanks :)
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top