Audio Circuit Problem

[Nirodha]

I'm building a laser microphone for my senior project. The idea is to bounce a laser off a reflective surface, capture the reflected beam with a photo-transistor, then convert the signal into sound. Vibrations in the glass from a voice will modulate the laser signal, and the voice should come out on the other side.

When I use a simple transistor amplifier circuit and connect the output to my laptop computer, it records pretty well; I can record myself talking at the piece of glass.

When I try moving the output to a 9V speaker/amp from Radioshack, the output disappears. I tried removing my amplification step entirely, I tried a LM386 low voltage audio amp, I tried an Op-Amp, and nothing is working.

My advisor didn't look at my circuit, but told me I have an impedance matching problem, so I'm not driving the speaker properly. I don't really know how to measure impedance, or where in the circuit to measure it.

Is this really my problem? If so, how can I effectively measure the appropriate values and fix them? If not, any ideas?

Any ideas or suggestions are much appreciated.

 

[neddie]

Try using an op-amp as a buffer(very high input impedance) and then your LM386/other options after that. That will rule out loading due to having too low an input impedance.

 

[Kinshoro]

sorry... posted something without knowing much....

 

[neddie]

This is not RF. You don't match speakers to the output of your amp.
The output of an lm386 is probable fractions of a ohm.
He may well be loading his phototransistor cct with to low an impedance , that is why it works with the laptop(mic input may be quite a high input impedance),
but not otherwise. Use a "buffer" between the photo cct and the power amp to stop the loading of the power amp on the photo transistor cct.

 

[keith1200rs]

I think I need to see a circuit to comment.

Keith.

 

[neddie]

That would be helpful , and limit some of the guesswork :0)

 

[stephen-7486]

try using an 8 ohm to 600 ohm audio transformer.
the audio output may be a little noisy but may work.
just an idea

steve

 

[Nirodha]

Ok, here is the circuit as it worked using the laptop as an output. The goal is to get it to work with a speaker. Right now, I have the speaker/amplifier from Radioshack, but I can buy a simple X-ohm speaker easily enough if that simplifies the problem.

 

[keith1200rs]

Is that resistor really 3.3M ohm? If so, it is surprising that it does anything at all.

Keith.

 

[Nirodha]

Yeah, it is. I increased the resistance on that resistor slowly, and each time it improved the gain of the circuit (with the laptop as output). I was surprised to find it functioned best with that large resistance, but that may mean that the amplifying transistor component isn't doing exactly what I think it's doing.

I'm sure some better circuit can be constructed, and the project isn't really any good if I can't get it to produce sound in a speaker (the laptop was just for testing).

 

[opacheco]

Yeah, it is. I increased the resistance on that resistor slowly, and each time it improved the gain of the circuit (with the laptop as output). I was surprised to find it functioned best with that large resistance, but that may mean that the amplifying transistor component isn't doing exactly what I think it's doing.

I'm sure some better circuit can be constructed, and the project isn't really any good if I can't get it to produce sound in a speaker (the laptop was just for testing).

Dear, that looks you have a serious BJT bias problems!!.......you will need to calculate the bias point and verify that amplifier is in the rigth Q point. Which one BJT are using??...What is the Base Q Current???.....How many miliamperes is flowing from the Phototransistor to the BJT Base???....

This BJT could to be in saturation all the time or no amplification at all maybe!!

Opacheco.

 

[neddie]

Apart from the obvious biasing problem , the output impedance of your cct is way to
high. Connecting even a modest load will kill any signal that may be present.
Use an opamp as a buffer. I'm going to try an attach a picture , but I'm new here so I'm not sure if it will work. Any jellybean opamp (lm324 , lm358 ) will work for your application :0) Their are obviously better opamps , but these are common as dirt :0)
Connect your lm386 audio power amp after the opamp.I used a 4n24 just to mimic your photo-transistror.

 

[Nirodha]

Well, I put together the circuit using some suggestions you guys made:

I got some static to come out of the speaker, but still no luck overall. The LM386 component is getting a little bit warm. Did I miss a component?

Thanks[/img]

 

[neddie]

If you have drawn the cct exactly as you have built it , you are going to have problems with the LM386. Check the datasheet.
First problem is the speaker is connected directly to the ic. It should have a decoupling capacitor(say 220uF).With the current configuration you ase going to have 375mA flowing through the speaker all the time(assuming 8 ohm speaker). No wonder the 386 is getting hot!!!
Once again check the datasheet.Signal into pin 3 , pin 2 grounded.
I've attached the datasheet :0)

 

[FvM]

Basically a good news: You're neighbours don't have to fear eaves dropping at the present project state.

 

[neddie]

Thats true , haa haa

 

[FvM]

The question title is misleading, the actual project challenge is optics, not a trivial audio amplifier circuit. For the vibrations
to detect, you can assume a nanometer order of magnitude. The method is known to work - with special instrumentation.

 

[neddie]

He did say that he had it working with a laptop though , so I assume he's on the right track.
The fact that he got the first part of the project (optics etc) working is surprising though :0)

 

[keith1200rs]

I don't think the LM741 will work at 6V.

Keith.

 

[neddie]

Use a single supply op-amp , like the Lm358,lm324. Not the best , but will work for this application.