Band-pass filter Design Help

Thread Starter

Barryallen

Joined Apr 17, 2017
20
Hello everyone,

Could you please kindly help me design a band-pass filter for a 1V sine wave from 2000Hz to 3500Hz ?

Thank you in advance.
 

MrChips

Joined Oct 2, 2009
30,712
Thank you for the reply,
I have tried GG a lots but I dont know which one is better or easier in my case: passive or active? and I don't understand some of parameters needing to type in some online design tools.
Oh, I see.
A passive filter does not use any power or amplification. You are going to lose signal amplitude.
An active filter uses amplifier and you get set the pass-band gain.

http://www.electronics-tutorials.ws/filter/filter_4.html

http://www.electronics-tutorials.ws/filter/filter_7.html
 

MrChips

Joined Oct 2, 2009
30,712
so sorry that I don't know how to calculate Rolloff. the higher Rolloff ,the better signal? I just want the best as you can guide me to do.
The roll-off is determined by the order of the filter. This determines how sharply the filter transitions from the pass band to the stop band.
A first-order filter has a roll-off of 6dB/octave or 20dB/decade.
This is your basic filter and a good starting point.
 

Thread Starter

Barryallen

Joined Apr 17, 2017
20
The roll-off is determined by the order of the filter. This determines how sharply the filter transitions from the pass band to the stop band.
A first-order filter has a roll-off of 6dB/octave or 20dB/decade.
This is your basic filter and a good starting point.
Thank you so much for that.
It's a little difficult for me to understand some specific words of electronics no matter how hard I am trying. Could you give me a sample circuit or something?
 

Thread Starter

Barryallen

Joined Apr 17, 2017
20
Tell us the use/need of the filter and we can better guide you in what will work for you.
I want to read frequency from a Piezometer . The output signal is 0.5 mV sine wave with frequency 2000-3500Hz. I have built 2 stages amplifier to amplify it to 1.25 V (2500 gain) but the output signal seems to have so much noise, so that I want to build a filter to reduce the noise. Could you please help me with it?

Thank you in advance.
 

crutschow

Joined Mar 14, 2008
34,285
I want to read frequency from a Piezometer . The output signal is 0.5 mV sine wave with frequency 2000-3500Hz. I have built 2 stages amplifier to amplify it to 1.25 V (2500 gain) but the output signal seems to have so much noise, so that I want to build a filter to reduce the noise. Could you please help me with it?
What is the nature of the noise?
Do you have an oscilloscope to view it?

That's a very small signal so you need a low noise amp.
Post a schematic of your circuit will all part values.
 

Thread Starter

Barryallen

Joined Apr 17, 2017
20
What is the nature of the noise?
Do you have an oscilloscope to view it?

That's a very small signal so you need a low noise amp.
Post a schematic of your circuit will all part values.
Here is what I did: signal.JPG
I have an oscilloscope to watch the signal but like It has so much noise so I can't see the signal I expect.
I am waiting for TLC2272 from a member here in ThaiLan to replace my LM358. Do you have any ideas ?

Thank you
 

crutschow

Joined Mar 14, 2008
34,285
Can you post a picture of the oscilloscope noise along with the vertical and horizontal scale settings?

But it's past my bedtime now, so see you tomorrow. :)
 

Thread Starter

Barryallen

Joined Apr 17, 2017
20
Can you post a picture of the oscilloscope noise along with the vertical and horizontal scale settings?

But it's past my bedtime now, so see you tomorrow. :)
Dear Sir,
Here is my oscilloscope screen:
With the open input:
c1.jpg c2.jpg

When I pull the input to the GND:
c3.jpg c4.jpg

Please kindly give me some ideas!

Thank you in advance.
 

SLK001

Joined Nov 29, 2011
1,549
A gain of 2500 in two stages is probably too much gain. I suspect that your circuit is oscillating and a BP filter will do nothing for you.
 

crutschow

Joined Mar 14, 2008
34,285
The open input signal appears to be an oscillation.
To eliminate that you will need good decoupling (are the amps decoupled with a 100nF ceramic directly at their supply pins?) and possibly shielding between the input and the output.
The bandpass filter could also help that.

The shorted input appears to be circuit noise.
The LM358 is a noisy op amp.
The TLC2272 should help that.
 

Thread Starter

Barryallen

Joined Apr 17, 2017
20
The open input signal appears to be an oscillation.
To eliminate that you will need good decoupling (are the amps decoupled with a 100nF ceramic directly at their supply pins?) and possibly shielding between the input and the output.
The bandpass filter could also help that.

The shorted input appears to be circuit noise.
The LM358 is a noisy op amp.
The TLC2272 should help that.
I don't know why the signal is worse (more oscillating) when I add a 100nF to VCC and GND of the Op Amp.
Could you help me or guide me design a banpass filter in my case, please?
Thank you in advance.
 

MrAl

Joined Jun 17, 2014
11,389
Hello there,

As a test you can try adding a 10k resistor in series with the input cap and that will create a low pass filter effect. The output will be reduced by about 3db but at least you can see if a filter will help as the noise should be reduced more.
Wideband noise is harder fo filter though.
 
Top