Query regarding low-input amplifier for scratch sounds (urgent)

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
Hi,
I am a student bugged with a prob that's left my college project stuck halfway. My project is Scratch input.
http://courses.cit.cornell.edu/ee476/FinalProjects/s2009/gh96_jbw48/gh96_jbw48/index.html

The link contains the design and the project report of a Cornell university student who had done the same project. i'm not using microcontrollers but instead using the Java platform. However, the hardware circuit is the same. I've tried the hardware design provided in the link, but it just doesn't give the desired result. So, i changed the amplifier design i.e. instead of using the power amplifier IC (LM358) i switched over to the tranistor amplifier. The thought behind the change was that the IC may not probably detect the feeble sound and hence we decided to shift to the t-t design. BF414 amplifier gave us good results (which was salvaged from a transmitter circuit), but we do not have the design of BF414 amplifier design. The result improved,but its still not as desired. The desired result too is shown in the link (the pictures of the signals in CRO).
Please help asap. Pass on your thoughts on what can be done on the project and the BF414 design.

<snip> is my ID.
 
Last edited by a moderator:

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
Thanks for the reply. The microphone i'm using is the basic electret microphone/ condenser microphone.Few questions:Can't we do away with the pre-amp and directly amplify with a 500 gain IC?I did try a pe-amp (with BF414) and then tried apmlifying with LM358, but it didn't quite work!
 

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
Also, i found a host of circuits there. (in fact, to an extent to confuse me!). I'm in a urgent need to complete my project and have little time to fiddle around with ideas.http://www.reconnsworld.com/audio_simplepreamp.htmlThis link has no design mentioned. The problem is we find a lot of circuits on internet that works, but as a student i would really love to know and under stand the detailed design.But, please someone tell me, the output circuit shown in the link above can be an input to the amplifier LM358? Or does it demand another transistor amplifier? I need a gain of around 500.
 

beenthere

Joined Apr 20, 2004
15,819
Why does every school have a bottomless box of obsolete op amps? You can get that gain out of a 358, but I would set up one section for a gain of 10 and the other for a gain of 50. I would use the gain of 50 from the microphone and follow with the 10 stage.
 

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
That is what i'm trying to do. But, the input from microphone when directly given to op-amp LM358, didn't give an output at all!! We figured out probably, the scratch of sound is too feeble to be detected. So, we decided to stick on to the transistor transistor design. And now the prob is i get circuits on net, but none of them actually have a detailed design mentioned so that i can design for varying gains. For example, the circuit shown in the link i pasted before- http://www.reconnsworld.com/audio_simplepreamp.htmlCould anyone help me with the design. I've tried many Transistor amplifier designs with no results.
 

SgtWookie

Joined Jul 17, 2007
22,230
Why don't you combine that preamp design, and then use a unity-gain buffer stage.

Why don't you post the circuits that you have tried?

Are you using a dual rail supply, or are you trying to use a single supply?
 

Audioguru

Joined Dec 20, 2007
11,248
Your electret microphone is connected upside-down. Then its output will be full of mains hum and it might not even work as a mic.
Its metal case is the same as its ground pin and its output pin is supposed to be fed with a resistor from a positive supply voltage.

The lousy old LM358 dual opamp is not a power amp. It has a very high noise output (hiss).
There are many single audio opamps available today. I use the TL071 single audio opamp (the TL072 is a dual audio opamp and the TL074 is a quad audio opamp)to be a mic preamp but its input needs to be biased above its negative supply voltage.
 

Attachments

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
Thanks for the reply audioguru. I tried with a simple pre-amp(attachemnt) coupled to another RC coupled amplifier and got the output amplified. But, i guess that's just the voltage amplified and the power isn't amplified. True? Coz i believe i would need a power amplifer as i have to give the amplifed output to the audiojack of the PC and the software needs to analyze the signals. So, need some views on those issues. Probably TL701 op amp will make things easier. Op-amp should be better coz transistor amplifiers shun DC.True?I'm a learner. So, please bear the queries.I jus need someone to give a confirmation o nmy views :)
 

Attachments

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
TL071!There was a typo in the previous rep. And why 4.7K?Jus an arbitrary or estimated value or any particular reasoning to opt exactly 4.7k?
 

Audioguru

Joined Dec 20, 2007
11,248
An electret mic uses 0.5mA. With a 9V battery it is fed from 10k then 0.5mA in them produces 4V across the mic and 5V across the 10k resistor.
Since your supply is only 5V then I cut the resistor's value in half which is about 4.7k or 5.1k.

The transistor has not much voltage gain and has a high distortion. An opamp can have any amount of gain and has very low distortion.

A power amp drives a very low impedance speaker. You don't have a speaker. The input of a computer is a high impedance that can easily be driven by a transistor or opamp.
 

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
Hi, I 've tried the TL071 IC mentioned by Audioguru, but that doesn't trace the scratch sounds! Anyway, i used the preamp circuit shown in this figure http://www.reconnsworld.com/audio_simplepreamp.html . I achieved an amplification. SO, i just redid the same pre-amp circuit and fed its input to the output of the same first circuit. I got a lot of noise which was amplified! Any suggestions for the second stage? I'll try the op-amp, but somehow transistors seem to work better for this 'scratch input' project.Next, i worked on the simple first-order high pass filter to eliminate frequencies below 3 KHZ and clubbed it with the amplified single stage/ 2 stages output. The output surprisingly, reduced the amplification to a very very bare minimum. I need to eliminate the noises, but the filters seem to have an effect on the amplification of the signal (the scratch sound)!!!Even the envelope detector had a drastic effect on the amplification, in fact both the filter and envelope detector reverses the amplification!!! Kindly help.
 

Thread Starter

udaykrizna

Joined Mar 3, 2010
10
Here's the link again of the circuit i'm working on, in fact its the link of the entire project.http://courses.cit.cornell.edu/ee476...w48/index.html Please do keep in mind that we are dealing with the sound of a scratch, that has a time duration of not more than a second or two and has an errratic pattern. We are trying to obtain a pattern by elimnating noises, amplifying and finally just taking the envelope of the signal. Let all the circuitous brains come up with solutions :)
 
Top