Band-reject and Low-pass filters (Butterworth)

[Aman2301]

Hi friends,

I have just started to work on active filters and designed the following:

1. 1st Order HighPass BW (cut off @ 1kHz)
2. 1st Order LowPass BW (cut off @ 2Khz)
3. Band pass using above two (1kHz to 2kHz)
4. 2nd Order HighPass BW (cut off @ 1kHz)
5. 3rd Order HighPass BW (cut off @ 1kHz)
6. 4th Order HighPass BW (cut off @ 1kHz)

all of the above are working fine but

7. Band-reject filter using 1. and 2. does not work
8. 2nd and higher order LowPass BW (cut off @ 2Khz) are not working.
I am getting some square wave type of pattern with spikes.

I am using PSpice to stimulate the circuit. It gives the expected result but
only for ckts 7. and 8. i am unable get the results practically i.e. on PCB/breadboard.

Any help will be highly appreciated.

Thanks.

(I have attached ckt dig. Please have a look.)

 

[berkeman]

Hi friends,

I have just started to work on active filters and designed the following:

1. 1st Order HighPass BW (cut off @ 1kHz)
2. 1st Order LowPass BW (cut off @ 2Khz)
3. Band pass using above two (1kHz to 2kHz)
4. 2nd Order HighPass BW (cut off @ 1kHz)
5. 3rd Order HighPass BW (cut off @ 1kHz)
6. 4th Order HighPass BW (cut off @ 1kHz)

all of the above are working fine but

7. Band-reject filter using 1. and 2. does not work
8. 2nd and higher order LowPass BW (cut off @ 2Khz) are not working.
I am getting some square wave type of pattern with spikes.

I am using PSpice to stimulate the circuit. It gives the expected result but
only for ckts 7. and 8. i am unable get the results practically i.e. on PCB/breadboard.

Any help will be highly appreciated.

Thanks.

(I have attached ckt dig. Please have a look.)

Welcome to the PF.

How are you trying to make a band reject filter out of 1 & 2?

 

[Aman2301]

I have uploaded the circuit diagram, please have a look at 3rd pic. Ain't it correct to couple the op-amps in parallel arrangement for band reject?

 

[berkeman]

I have uploaded the circuit diagram, please have a look at 3rd pic. Ain't it correct to couple the op-amps in parallel arrangement for band reject?

Uploaded where? There were 3 pics uploaded before in your OP, no?

 

[Aman2301]

https://www.physicsforums.com/attachment.php?attachmentid=29740&d=1289304211

 

[skeptic2]

There may exist a frequency with regard to a parallel arrangement of a high pass and a low pass filter at which there is phase cancellation but I would not call that a band reject filter. For that to work the low pass filter would have to have a lower roll off frequency than the high pass filter. That is not the case with your circuit.

The simplest arrangement would be to have high pass and low pass filters in series with the low pass filter having a lower roll off frequency than the high pass filter. Note that the point at which the slopes of the low pass and high pass filters intersect will be the maximum band rejection you can get. If you need greater band rejection you will need to increase the slopes of your filters by adding more poles.

 

[Aman2301]

k...thanks, will surely try that. but i am still expecting more replies and suggestions.

 

[berkeman]

k...thanks, will surely try that. but i am still expecting more replies and suggestions.

I saw skeptic's reply, and didn't see a need to add more.

 

[Aman2301]

any references or circuits?

 

[skeptic2]

Give us some parameters, eg. center frequency, bandwidth(rejection), how many dB of rejection.

 

[Aman2301]

Band reject: 1kHz to 2kHz
It has to be Butterworth type.
AS such no restrictions on dB of rejection but has to be made by coupling Firt Order BW HP and First Order BW LP filters.
By the way i tried out constructing Higher order BW HP and BW LP filters.
I made band pass and band reject out of them as well. All of them are working fine.
I don't know what is the problem with 1O BW HP and !O BW LP coupling in band reject configuration. It is giving sort of "all-pass filter" profile on CRO.

 

[skeptic2]

I apologize for giving the wrong advice in the second part of my post #6. This turns out to be not as trivial as I had thought. This link is to a very good Texas Instruments Application Report on designing filters.

http://focus.ti.com/lit/an/sloa093/sloa093.pdf

The attached circuit is the closest I could get to meeting your parameters using 5% values.

 

[Aman2301]

thanks for the ckt but it does not involve coupling of two Butterworth filters, which i need to do. btw i saw that texas link days before...it was not of much help because they have not given specific values for R and C. one more thing (as i mentioned earlier), the MicroSim/PSpice simulations are very different from "real" circuit measurements.

It will be very convincing if you or somebody else has designed the circuit on a PCB or BB.

I forgot to mention that i have only 0.01 uF capacitors and 741 op-amp available but then i think it is just "RC" product that i have to adjust. Isn't it?

 

[vk6kro]

You asked for
Re: Band-reject and Low-pass filters (Butterworth)
but your examples are not Butterworth and they are not active filters.

Butterworth filters are L/C circuits and these are R/C circuits.

And your circuits are just passive filters with opamp buffers. This is not an active filter.

This article describes typical active filters:
http://alignment.hep.brandeis.edu/Lab/Filter/Filter.html#Active Low-Pass Filters
And there are plenty of articles describing the design of Butterworth L/C circuits.

Butterworth filters are relatively expensive due to the inductors used, so active filters are used to produce a result without using inductors.

Active filters use feedback from the output of the opamp back to the filter components, but there wouldn't be much point in using active filters with L/C Butterworth filters.

 

[skeptic2]

thanks for the ckt but it does not involve coupling of two Butterworth filters, which i need to do. btw i saw that texas link days before...it was not of much help because they have not given specific values for R and C. one more thing (as i mentioned earlier), the MicroSim/PSpice simulations are very different from "real" circuit measurements.

It will be very convincing if you or somebody else has designed the circuit on a PCB or BB.

I forgot to mention that i have only 0.01 uF capacitors and 741 op-amp available but then i think it is just "RC" product that i have to adjust. Isn't it?

I will not breadboard a circuit for you. What value is your knowledge of electronics if you need the values of the components given you?

I have used MicroSim on many projects and always found it to give reliable results. The differences between 741 and 324 opamps will not be important in this simulation.

Do you not know that with filters the values of the resistors and capacitors are not so important as the their ratios? From the values from my circuit you should be able to work out the resistor values for 0.01 uF capacitors.

It should be obvious from my circuit how to couple both filters the way you want.

 

[Aman2301]

ya skeptic2...i don't expect you to breadboard the circuit but i just asked in-case you or somebody else has already done that. i know how to adjust the R and C values after all how did i design higher order circuits. but MicroSim evaluations have proved to be a disaster...i made the same circuit at least thrice. once on PCB with proper soldering and twice on breadboard...just because my results did not match the MicroSim evals. anyways thanks for your suggestions and comments. i highly appreciate them. :)