Maker Pro
Maker Pro

Simple Microphone....

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
Hey,
I was building a microphone with a piezo sensor and tested it out with my computer. I assumed(though with no real reason) that amplitude of the waves is directly proportional to the voltage, so to amplify the input (me talking), i just passed the piezo output through a transistor which was connected like this:
base- 1K resistor to piezo input
collector-+1V (from +5V using 2 resistors as a voltage divider)
emitter- microphone jack to computer

I tried increasing and decreasing the voltage input of the transistor, but the playback volume is always too low....

What am i doing wrong ?
Thanks !
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
You are making no sense.. Wave amplitude = voltage? Piezo sensor = piezo transducer/speaker/beeper disc?
Transistor as an unbiased emitter follower with no emitter resistor and with collector voltage reduced by potential divider resistors used as a volume control?
I believe a proper emitter follower would be a relatively proper "amplifier" to use for a piezo disc mike. Google for diagrams.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The other issue may be that a piezo microphone is likely to produce a relatively large signal, but one that is very high in impedance. The main issue will be to match that to a much lower impedance input of your computer.

What you need to look at is a transistor amplifier with high input impedance, and not a lot of voltage gain. If you're looking for high input impedance, a fet of some sort is going to get you there a little more easily than transistors (especially when made from discrete components.

Here is an example of a jfet amplifier for piezo pickups on an electric guitar. It may be suitable.
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
Sorry for making no sense at all, i am learning based on assumptions i make along the way...

-In a sine wave, what does amplitude correspond to? Suppose i take my sound card and input 0.5V, i would probably hear a sound. If i then input 0.8V (same waveform, just different voltage) then will i hear a louder sound or does it just not matter ?

-I meant those round circular disks with piezo crystals inside of them...
picture1.jpg


Thanks steve, i'll look into it.
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
It's starting to clear now..
A sine wave has an amplitude (peak) of 1.414 times its rms value, that goes without saying.
An increase in (either) value will of course be perceived as an increase in loudness. The smallest difference you can make out is around 3dB (= twice the voltage).
A piezo transducer has a very high voltage output but also a very high impedance. It is after all a capacitor. To use its signal one has to amplify the current, not the voltage.
A fet is better at doing this than a bipolar transistor, but depending on what you have available & what you need to achieve, a bipolar is not a bad choice either.
An emitter follower circuit (wether made with a bipolar or a fet) has 0dB voltage gain (in=out) but a very high input impedance and a very low output impedance.
A bipolar emitter follower circuit (with a transistor gain of 300) can achieve almost the same performance as the fet circuit Steve suggested.
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
from a site
An emitter follower circuit shown in the figure is widely used in AC amplification circuits
emitterfollower.gif


Is this what you are talking about ? I am unsure about what resistor values i have to use. The equations on the site have too many variables.......
Instead of solving them, is there a standard set of values that one can use for applications like this? (microphones)

Also, if this were right and i were to use it, where would i take my input for the microphone jack from? In place of RE or across RE? Or is it RL where Vout is ?
Also, in all the equations, there is no mention of the capacitors that are in the diagram.... Is this common knowledge ? (cause i have no idea) Can people here just estimate resistance values by knowing the application and looking at the circuit. If they can, then i have a lot to learn......

Thanks

EDIT: Zimbabwe ?
guyt.jpg
 
Last edited:

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
Yes. Resistor values will depend on the transistor gain. If you use 1Mohm for Rb and 10kohm for Re, that will require a transistor with a gain of 100, which is very common.
You'll want Vb & Ve to be around half of Vcc for maximum signal swing. So with the voltages being equal across each resistor you only need to select Rb 100 times Re.
(That is if the transistor has a gain of 100). You want to select the resistors as large as possible, and 10k is the most Re should be in order to easily drive the PC's ~50k.
Input (the mike) is Vin. Rs is just a representation of the mike's internal impedance, and can be disregarded. The capacitor is not needed here since the mike is a cap.
So you just connect the piezo disk straight to the base of the transistor. Your sound card is represented by Rl, and it's best to use that capacitor even if the PC has one.
C=1/(2*pi*f*Z) C= 1/(2*3.14*20*50k)=0.16uF. Use any value you have available close to that, it will only affect the low frequency. Speech only goes down to 150Hz btw.
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
Out of curiosity, how did you get those values ? A calculator of some sort, or just intuition?
All my transistors have hFE at around 300 not 100. Does this mean i can do with these values:
Re=10kOhm
Rb=300Re=3MOhm
C=0.16uF (or whatever is close enough)

Also, can i use any voltage for Vcc or does it have to be relatively high ? I heard that the sound card can't take voltages higher than 0.8V or something... So what should i use as Vcc ?
Thanks
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
Intuition, if you will, is based on knowledge one has picked up underway. In this case it's just basic electronics tuition that you'll find if you look around. There are tutorials right on this site, and this specific circuit is also called a common-collector amplifier btw. Common amplifier input impedances can be found by looking up some spec's.
Yes, those resistor values will be right for the transistor gain mentioned. Measure the actual voltages, and if they're not balanced then adjust resistor values accordingly.

Any Vcc voltage will do, and running it straight from 5V (or even 3V) should work fine. The Vcc value in itself is not what will reach (and potentially damage) the sound card.
Also, amplifiers (sound cards) have a maximum input value where they will clip, and another much higher value where they'll get damaged. I haven't checked actual values.
The piezo disks have such a high output that you might want to use the line input instead of the mike input btw.
Be aware that the disks are capable of very high voltage outputs, like if you tap or drop them they can generate 10's of Volts which could damage a sound card.

As for the map; hardly. Using that kind of projection won't get you anywhere using a straight line (as that will usually be "the long way home").
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Zimbabwe ?

No, you need to draw a straight line (where that is defined as the shortest path between 2 points on the surface of the earth).

Your line has more curves than 60's beach party :D

In many respects you could think anti-Zimbabwe.
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
Hey,
I assembled the emitter follower circuit and plugged it into the line-in of my comp. It works nicely, but i have one concern:
I had to turn up the volume a lot when using it on my comp so I'm guessing it's not such a great mic.
Suppose i take one of those headphones you get free on airlines and connect them to the amplifier i assembled, would i need some other extra amplification of some sort ?
(i would just find out myself but i can't find them at the moment)

Thanks.
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
The disc has all open edges where the sound pressure waves will pass around. Did you try to mount it by its edges in a hole (in a solid box or a baffle)?
The disc is quite stiff so it needs something to "kick" against in order to bend and do its work.
If it's still too weak for your application you can of course just try to plug it into the mic input of the sound card.
The airline headphones are of the dynamic type (meaning it's a moving coil in a magnetic field, having a low impedance (32 ohms) and hence a low signal level).
It can be plugged straight into the mic input but would probably need a common-emitter (or even better a common-base) amplifier to get a good signal level.
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
Honest to god, whenever i set up an amplifier of any sort and use it, sure it works and i need to put my comp on full volume to hear it....
But when i just connect the piezo disk straight to the jack and connect it to my craptop (one whose sound card i care not for), it is always louder !!!
I must get an amplifier working ! Or well anything that will make the odd sounds i like to make while testing a mic louder !

Please help in any way...

EDIT: and yes, i used a PVC tube and put the sensor inside it... It worked better with all the resonance and reflection and such..... but it is still pale in comparison to the disk straight to the comp...
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
Hi,
Is it silly of me to assume that the mic output happens to be a sine wave ? I know very well that it does have a positive and negative part(as in the graph) like a sine wave. I just don't know what corresponds to the crests and what to the troughs (as in, in terms of frequency or amplitude)

If i ever do manage to construct a sine wave amplification device, will it work based on my piezo crystals input or am i thinking very wrong here ?

Thanks
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Within certain constraints, it doesn't matter an awful lot what the waveform looks like.

Pure sine waves are rarely seen outside a lab, there's always a mix of signals giving you a more complex waveform (and sound).
 

eptheta

Dec 20, 2009
188
Joined
Dec 20, 2009
Messages
188
So, is it a bad idea to use transistors for amplification?
Will the circuit (emitter follower) work right or did i just build it wrong ?(twice)

Thanks...
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
No, transistors are fine for amplification. Your amplifier may need more gain, or have mismatched impedances, but the general approach is sound.

The emitter follower will tend to have low voltage gain, and you might consider a second stage of amplification (AC coupled) or to use your sound card's microphone input (as I think has been suggested)
 
Top