Second order passive low pass filter design

ericgibbs

Joined Jan 29, 2010
18,766
hi pp,
I am sure we would all like to help you, but how will you know when the answers that are posted will solve the problem or not, if you do not know what the problem is...??

E
corr typo!
 
Last edited:

Thread Starter

pinkyponky

Joined Nov 28, 2019
351
hi pp,
I am sure we would all like to help you, but how will you know when the answers that are posted will solve the problem or not, if you do not know what the problem is...??

E
corr typo!
Yes, I understand from your point of you, you and your team are here to help the students like me. I will update here when I understand the project.
Thank you for your support.
 

MrAl

Joined Jun 17, 2014
11,389
At least tell us what the project does - that would help!
Actually though when talking about the damping factor and response it is not really necessary to know the topology. The damping factor is related to the L and C and any R's in the circuit, but because the damping factor is a function of the L and C and all R's if we know that we know the response already so we can talk abotu the response without knowing the actual values or even how it is connected as long as it is second order.
If you look at the plots i provided you will see they vary with damping factor not with any particular value of L or C or any R's that might be present, yet we still have a clear idea what the response will be.
So really we can have say damping factor d=0.1 with L=1 and C=1 and some R's or with L=2 and C=0.01 and some other R's and the topology may even be different for both of these examples but we still know the response shape in time because the damping factor takes all these into account.
 

MrAl

Joined Jun 17, 2014
11,389
May I know why your showing the curve at -3db. Do we need to measure always at -3db?.
The -3db point on a frequency curve is the usual point to talk about because it is a defacto standard in the electrical industry. It's origin goes way back to when telegraph was still in it's infancy.
In filter theory it is most often a measure of the half amplitude frequency response of the filter but sometimes it is a measure of the half power of the response. The reference is usually the highest peak of the response.
It makes comparing filters somewhat of a standard procedure although sometimes more information is needed.
It allows us to compare bandwidths very efficiently too as well as cutoff frequency.

For example, if we have one filter that has -3db point at 100Hz and the other at 200Hz we know that one filter has lower output than the other for frequencies 100Hz and above if it is a low pass filter.
For another example if we have one filter that has two -=3bb points at 100 and 200 Hz and the otehr at 120Hz and 180Hz we know the second filter has a sharper response than the first.

So the -3db point(s) are a convenient way to compare filters and when a filter specification is given in terms of bandwidth we know they mean the -3db points. They are also sometimes called the "3db down points".


As to the request for the circuit itself, we know that when talking about damping factor we dont actually need a circuit so dont worry about that unless you have other questions that would require a circuit or a response other than 2nd order.
Of course if you want to know how the R, L and C produce a damping factor then we need to know the circuit and the values of all the components. Even so, we can look at a few different circuits if you dont have one in mind just yet. The common circuits are series and parallel with or without load. It's not too hard to talk about all of these really.
 

MrAl

Joined Jun 17, 2014
11,389
I was confused how a lowpass filter (the original subject) could have two -3dB points, then I realised.
I just thought I'd clarify it.
Oh ok sure. I was just talking about -3db points in general, how they seem to be standard for comparisons of various filter types including LP, HP, and BP, even BS.

One of my favorite filters was the Twin Tee band stop filter. I found it interesting because for one it was all passive, and two because even being passive it has a theoretically very very deep valley i think it is actually zero.
I used one once for a 60Hz filter to filter out the hum in an audio circuit. It is hard to keep it tuned perfectly though and drive impedance and load impedance have quite big effects on the efficiency of the filtering action messing up the bandwidth and the depth of the valley.
Dont know if this filter is used anymore though.
 

Thread Starter

pinkyponky

Joined Nov 28, 2019
351
The -3db point on a frequency curve is the usual point to talk about because it is a defacto standard in the electrical industry. It's origin goes way back to when telegraph was still in it's infancy.
In filter theory it is most often a measure of the half amplitude frequency response of the filter but sometimes it is a measure of the half power of the response. The reference is usually the highest peak of the response.
It makes comparing filters somewhat of a standard procedure although sometimes more information is needed.
It allows us to compare bandwidths very efficiently too as well as cutoff frequency.

For example, if we have one filter that has -3db point at 100Hz and the other at 200Hz we know that one filter has lower output than the other for frequencies 100Hz and above if it is a low pass filter.
For another example if we have one filter that has two -=3bb points at 100 and 200 Hz and the otehr at 120Hz and 180Hz we know the second filter has a sharper response than the first.

So the -3db point(s) are a convenient way to compare filters and when a filter specification is given in terms of bandwidth we know they mean the -3db points. They are also sometimes called the "3db down points".


As to the request for the circuit itself, we know that when talking about damping factor we dont actually need a circuit so dont worry about that unless you have other questions that would require a circuit or a response other than 2nd order.
Of course if you want to know how the R, L and C produce a damping factor then we need to know the circuit and the values of all the components. Even so, we can look at a few different circuits if you dont have one in mind just yet. The common circuits are series and parallel with or without load. It's not too hard to talk about all of these really.
I would like to design the filter that will cut of the frequency above 5kHz. So, I have designed RLC circuit. Please can you have a look and tell me that the circuit design is correct or not?.
 
Last edited:

Ian0

Joined Aug 7, 2020
9,668
I would like to design the filter that will cut of the frequency above 5kHz. So, I have designed RLC circuit. Please can you have a look and tell me that the circuit design is correct or not?.
View attachment 247498
still no load resistor (the load resistor goes across C1)
and there is no need for a source resistor.
Go back to Post #22, and use 5000 for the frequency.
 

Ian0

Joined Aug 7, 2020
9,668
As you demonstrated very well in post #30, if you don’t use a lower value load resistor, you get ridiculous values for the components.
 

Thread Starter

pinkyponky

Joined Nov 28, 2019
351
As you demonstrated very well in post #30, if you don’t use a lower value load resistor, you get ridiculous values for the components.
Without load resistor can't we design the filter.

Just consider that design a circuit with the RLC filter that will cut the frequency off above 5kHz.
 
Last edited:

Thread Starter

pinkyponky

Joined Nov 28, 2019
351
Absolutely true - decide on the value of the load resistor that you wish to use, then enter the figures into the equations in post #22, along with “5000” for the frequency.
Without load resistor I want to design a RLC filter. As I told you my load resistor is 30G ohm.
 
Top