Great audio project , need your help/advice on filters

Thread Starter

simeonz11

Joined Apr 29, 2008
98
Hi all

I have this audio signal I wanna amplify . I wanna amplify this sine wave I got ( input signal ) , This signal has "HISS" in it so I pass it through a RC low-pass filter to smooth out the "HISS" ( section 1 ), this works rather well . My cutoff frequency -3dB is @ about 10 khz with a 10 nf cap and a 1.6k Ohm resistor .

This input signal stays beautiful and useable up to 20 khz or so . But its amplitude has dropped significantly . ( major bummber ) . And a higher freq lowpass-filter RC link just does not cleanse out the "HISS" as well

So basicly the reason of my post is this , I wanna "counter" this drop in amplitude with another RC link that is somewhat a reflection of my other RC link . I was thinking that as the frequency goes up the impedance of section 2 goes down and my gain goes up ... ??

thx in advance guys .

The ugly Bode plot drawing I made is a perfect example of what I want to happen , section 2 "counters" the lowered input signal amplitude with a higher gain as frequency goes up .

Is my drawing the right way to go ? Basicly my ideal end goal is to have a Hiss free signal that stays about the same amplitude from like 200 hz to 15000 hz

How do I calculate both my RC links to be "reciprocating" of eachother , tired of blidly googling for more info , need your help guys .Is there a common name for this sort of thing btw ?

thx in advance
 

Attachments

Last edited:

Thread Starter

simeonz11

Joined Apr 29, 2008
98
That drawing I made is crap .

I think I found it , thx to you post .

I will just use 2 amps , one with a unity gain low pass just like I have and another highpass filter revserse graph .

Hows that sound ?

Now I just need to find more info on this exact subject . The amplitude needs to be stable from broad range and "hiss" free..
 
Last edited:

t06afre

Joined May 11, 2009
5,934
It is up to your ear to decide what is good and bad;) If you like sound then go for it. The audio range is considered to be between 20 to 20KHz. But this is dependent of several factors like age as one example. So I would perhaps move the lowpass filter section somewhat up in the frequency range. And for the high pass section, well the most important thing is to block DC. If was you I would have used a first order filter to block the low frequency range(high pass filter). Then a second order high pass filter to block HF noise.
 

Thread Starter

simeonz11

Joined Apr 29, 2008
98
Hi t06

The noise is relatred to my fundamental signal .

I cant block the low frequency first , Im using a lowpass of 10k to block "most" of the crap like you said .

But I may need to go up to 15 without getting much lowew amplitude .

I studied this a bit at school but am still a noob ,I need an RC link somewhere on my gain side to counter-balance the input signal amplitude drop . But where ?

Does anybody have a good schematic with equations for this exact problem ?
 

Audioguru

Joined Dec 20, 2007
11,248
The hiss is at high audio frequencies. If you reduce the amplitude of the hiss frequencies then you also reduce the amplitude of the wanted audio frequencies.

A highpass filter passes hiss noise it does not block it.

Why does the input signal have hiss? An old cassette tape player?

If the signal source has hiss because it uses a noisy lousy old opamp then use a low noise audio opamp.
 

Thread Starter

simeonz11

Joined Apr 29, 2008
98
The hiss is at high audio frequencies. If you reduce the amplitude of the hiss frequencies then you also reduce the amplitude of the wanted audio frequencies.

A highpass filter passes hiss noise it does not block it.

Why does the input signal have hiss? An old cassette tape player?

If the signal source has hiss because it uses a noisy lousy old opamp then use a low noise audio opamp.
That is correct , the input signal is hissy .

So I must cleanse it , but when I cleanse it I loose amplitude @ higher than 9khz , so I need to counter this sepoeratly .
 

beenthere

Joined Apr 20, 2004
15,819
The nature of the problem is here -
I wanna amplify this sine wave I got ( input signal ) , This signal has "HISS" in it
The thing that is sensible to do is to discover the reason why the hiss is part of the signal. By eliminating the hiss at the source, the need for expensive tunable filtering is eliminated.

To expand on AG's post - what is the source of "the signal"? Can you post up the circuit?
 

Thread Starter

simeonz11

Joined Apr 29, 2008
98
Hi Beenthere , thx for your reply and your interest in my project ..

This sine wave is digitally synthetized by a switch capacitor filter and the HISS is in fact its clock noise , something totally natural and unstoppable for this component .

Post-filtering must be done , wich I have succeeded . But this filtering ahs the drawbacks of lowering the input signal amplitude past a certain frequency range .

This is my main problem , the other things are here to stay and part of the circuit .

This is an audio project in the broadest of terms , it is a bastard child improvisation project of mine , you would probably laugh if I told you.
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
I also made some sine-wave generators with switched-capacitor lowpass filters. The clock for my filters was at 100 times the sine-wave frequency and at a low amplitude so the clock "hiss" was easy to remove without affecting the sine-wave.

Your simple low-pass filter has a response that is -3dB at 10kHz and -6dB at 20kHz. You forgot to say what is the frequency of the clock you are trying to remove. Maybe a high-Q notch filter will remove it.
 
Top