# Help with a simple project for schools

Discussion in 'The Projects Forum' started by GBart, Apr 15, 2013.

1. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
In order to demonstrate the use of the theory of electromagnetism, I have built a small speaker from scratch using wire, a magnet, and a paper plate, and it works nicely, but it's very quiet. I've been trying to build a small amplifier for it, but I can't make it work. I majored in physics, so I have some experience with circuits, but not a lot.

How can I build a simple, cheap amplifying circuit with a 9-V battery and \$5 of stuff from RadioShack? I tried using a LM386 amplifier, and I thought I had it connected correctly, but it just got hot when I connected the barrery. I've tried three times now (including once using an LM741) to get it to work and nothing is happening. Help please!?

2. ### crutschow Expert

Mar 14, 2008
21,415
6,136
Do you use a capacitor to couple the output of the LM386 to the speaker?

Post the complete circuit schematic.

3. ### joeyd999 AAC Fanatic!

Jun 6, 2011
4,093
5,948
Since the project involves magnetism, and not electronics per se, why not just hook it up in place of the speaker on a cheap AM radio?

4. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
I don't know how to answer that question so I don't think so, no.

And I don't think a schematic is really necessary, it would be quicker to explain: I connected the input to the in+ and in- pins as labeled on the diagram on the TI website (and it doesn't matter if I got the polarity wrong here, because it would just invert the signal otherwise, right?), then connected the battery to Vs (+) and ground (-), then connected the speaker to output and ground.

The +'s and -'s are really confusing because I think of current moving in the direction that the actual electrons move, which is from - to +. That's the opposite of how electricians see it, right?

Last edited: Apr 15, 2013
5. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
I don't have an old radio I could use, and I want to be able to hook it up to my computer to play a specific sound.

6. ### #12 Expert

Nov 30, 2010
18,076
9,678
That pretty much covers it. The 386 chip outputs a DC level of 1/2 the power supply. That DC was running through your hand made coil and heating up the works. A fairly large electrolytic capacitor is used to block the DC and allow the sound to get to the speaker.

7. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
What does that mean? What do I need to do?

8. ### panic mode Senior Member

Oct 10, 2011
1,657
468
amplified computer speakers are \$4-5 so just buy them if you don't have them already.
use built in amplifier to drive your speaker.

Last edited: Apr 15, 2013
9. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
Yeah but I want to figure this out.

10. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
Did a post get deleted or something? What covers what?

11. ### #12 Expert

Nov 30, 2010
18,076
9,678
Your explanation that you did not use an output capacitor indicates why this failed. You must add a capacitor between the output of the chip and the speaker to make a 386 chip work correctly.

12. ### GBart Thread Starter New Member

Apr 15, 2013
7
0
Ah, ok. And keep everything else the same? What capacitance?

Thanks!

13. ### #12 Expert

Nov 30, 2010
18,076
9,678
Depends on the impedance of the speaker and the lowest frequency you want, like this:
C = 1/ 2Pi F Z
where F is your lowest frequency and Z is the impedance of the speaker.
4 instance:
C = 1/ 6.23 x 100Hz x 3.2 ohms
C = 497uf minimum.

14. ### GopherT AAC Fanatic!

Nov 23, 2012
7,983
6,775
You need to put a capacitor, in series, with your input signal to one of the two inputs on the LM386. That capacitor will convert the audio signal that is riding on a DC off-set to a simple AC signal with zero DC offset (the LM 386 internally adds 1/2 the supply voltage to your signal but you don't have to worry about that.

On the output side, a capacitor does the same thing and allows your DIY speaker to travel +x and -x from midpoint and allows more air displacement than going from mid-point to 2x.

Use a 0.1 uF capacitor on the input and something between 220 and 1000 uF cap on the output (not too critical for your demonstration.

You can also try making the paper plate more rigid and, therefore, louder and better able to reproduce the signal that is put into it. Spray starch (or concentrated liquid starch), hair spray (your grandmother's stiff type), Elmer's glue or wall paper paste all might work (most sticky things listed are interchangeable in most applications).

Last edited: Apr 15, 2013
#12 likes this.
15. ### WBahn Moderator

Mar 31, 2012
23,868
7,384
He meant that your description of your circuit pretty well covers why things are getting hot. You are running a large DC current through your speaker coil which is doing nothing but moving the plate a static amount (thus no sound for this component) but producing a lot of heat.

I'm guessing you have no idea what the impedance of your homemade speaker is? If you are using a signal from a PC or other standard source, you want to try to match your speaker's impedance to what they are intended to drive. Too much or too little and you won't get much sound out.

Now, don't just grab a VOM and measure the DC resistance of the coil, which is probably quite low (and hence the heating problem when you don't block the DC current). What matters is the AC impedance at the frequencies you are driving it at.