[Pepsione1]

This isn't really audio related but I don't know where else I can find my answer from besides google. I want to build some kind of frequency selector circuit. Given a 60hz source I want to be able to break it down to 1,3,5,7,9 hz or etc. Its mainly for flashing a set of leds at different speeds. Any ideas?

 

[gsferrari]

Astable Multivibrator :-
=================

use an astable multivibrator with transistors. vary capacitor values with a switching circuit for various output frequencies.




[EDIT]

I think you want a frequency multiplier

http://216.239.39.104/search?q=cache...hl=en&ie=UTF-8

[/EDIT]

 

[Pepsione1]

Is there simple circuit I can build using basic elements. I know something can be made with flip/flops.

 

[tangent]

Instead of dividing down an existing source, setting up a square wave oscillator is simpler. As I recall, there's a way to do it with a quad inverter. The book you want is the old Forest Mims books from Radio Shack. If you have one or can dig one up (they're out of print now), they have several different square wave oscillators in them.

Be sure to keep the drive current in mind, if you want to flash more than one or two LEDs. Most of these circuits won't drive more than a few mA.

 

[guzzler]

do it with a 555 astable multivibrator; costs only a few pence for a 555 and use a pot to vary the output frequency

g

 

[gsferrari]

Yes the 555 timer circuit is a very good idea...

Or why dont you just buy a commercial function generator and live with it? Painless and more functionality than you can reasonably put in a diy function generator...more precise as well

 

[tangent]

The problem with a 555 for this purpose is that C starts to get large, and when C gets large, precision goes down -- you end up using +/-50-100% tolerance electrolytics.

Take a look at National's LM339 datasheet. There's a squarewave oscillator in there that uses smaller Cs. You can get down to a few Hz with C up to 1uF, which is still within film cap range. The datasheet doesn't give the functions, but Hz goes down as the cap value goes up.

 

[Pepsione1]

Thanks guys I think I have the problems solved by using the 555 timer. Thank you again for your time and kind suggestions.

 

[Pepsione1]

Here's what I am trying to make. If I want the led's to flash at different rates (1hz,2hz,3hz) I can just change the RC values right? What other ways can I do it?

 

[groggory]

just something to think about...

but if this is for a super high brightness bike light, try looking at the luxeon star LED's. They are REALLY REALLY bright. And being that you're thinking of 20x white LED's, the price should be similar too.

 

[Pepsione1]

Ok I got everything to work they way they should. I've decided to add switches to change the R value to control the speed of the flash. Just for fun I went and add a small speaker (like the ones inside a computer case) in parallel with the set of led's. It seems that this would turn the all the leds off that's connected to the same transistor. The biggest problem is the speaker isn't loud at all. Its makes a noise according to the freq of the timer but you have to put it close to your ears to hear it. How can it make it louder?

 

[Pepsione1]

No one wants to help? I think the reason why the leds won't turn on is because the resistance of my speaker is much lower then the resistance of the led branch so it no enough current gets to them. But my problem still remains, how can I make my speaker louder? Could it be because the freq is too low for the speaker to make a loud noise or is it because there isn't enough voltage (or current?) to drive the speaker. If power is the problem, I can just add a op-amp amplifier right?

 

[tangent]

An 8 ohm cone speaker will indeed hog current from the LED circuit, and it will indeed need a hell of a lot more power than a 555 timer will put out. If by "add an op-amp" you mean put a CMoy on it, that still won't do it. You need a small power amp to drive a low-ohm speaker. Either an LM386 or a power op-amp would be easy. Adding the amp would make let the LEDs work, since the power to drive the speaker is coming from the amp, not the 555.

If it's a piezo of some sort, you should be hearing something, but they have special drive requirements. In the case of a piezo buzzer, they want to see DC, and they oscillate at a fairly high frequency. If it's a piezo speaker element, they don't respond well at low frequencies.

Incidentally, the sound will be pretty nasty, since you're driving it with square waves. You might want to put a low-pass filter in front of it to get rid of some of the higher harmonics.

 

[gsferrari]

I dont think you should have an LED in series with a speaker. The speaker draws a lot of current and you could blow up the LED's. Also - what is your power source for this experiment? Even the most "sensitive" speakers need more than a 9V supply to be audible...

 

[Pepsione1]

I got everything working now (led with a popping noise from speaker). The speaker wasn't in series it was in parallel with the leds. Anyway given the frequency of the circuit the speaker isn't that audible. That's my theory at least. Because when I used a square wave with high enough frequency (like 500hz compared to 3hz of our circuit) the result was satisfactory. So I guess the thing I need now is a circuit that will amplify the frequency from 3hz to 300hz or more. Is there such a circuit?