The problem is this led will be used in a small waterproof box, and the LED has to have a 1kHz signal. So i cant use a signal generator with it because its too big. Is there any other way to insert signal.
I can only use small components btw,and the power supply is a 3.6 V AA battery.

 

Do you mean the LED has to be modulated (as in turned on/off at 1KHz)? This is easy enough to do, but we need a bit more information as to what you're trying to accomplish. We also need the space requirements.

 

Do you mean the LED has to be modulated (as in turned on/off at 1KHz)? This is easy enough to do, but we need a bit more information as to what you're trying to accomplish. We also need the space requirements.

Thanks for your reply.
Yes i am trying to modulate the LED. The box dimensions are 150 x 110 x 70mm. All i need to do is have an LED with 3.6V (AA battery) supply and modulate it with 1kHz signal.

 

Let's see, I translate that into 5.9" X 4.3" X 2.7", shoot that is lots of room. I'm assuming half of that is used for the battery and LED.

Be back in a bit, thinking about it.

 

OK, try this schematic...

With this IC.

It may be sensitive if the battery drops below 3V, I just don't know, but it will work. Just eliminate one of the resistors and LEDs shown in the schematic.

You don't even need to use a PCB or breadboard, use point to point wiring, then paint everything with a liquid insulator (available at most hardware stores).

 

Thanks a lot mate, i will give it a try.

 

OK, try this schematic...

With this IC.

It may be sensitive if the battery drops below 3V, I just don't know, but it will work. Just eliminate one of the resistors and LEDs shown in the schematic.

You don't even need to use a PCB or breadboard, use point to point wiring, then paint everything with a liquid insulator (available at most hardware stores).

Just tried that, but it modulate to square wave rather than sine wave.

 

Just tried that, but it modulate to square wave rather than sine wave.

Seems as if that little fact was never mentioned in the original request! See Bill'd work at http://forum.allaboutcircuits.com/showthread.php?t=12405

--Rich

 

Why do you need a sine wave?

 

That fell under the request for more information. So lets try again, does it have to be a sine wave, or can a triangle wave work? Do you need all the way off and all the way on, or would a percentage work (and what percentage).

Whilst we are asking, what current do you require through the LED? This is needed since we are now probably going to use more than one chip.

Another question, would PWM work for the sine wave, say 1Khz modulation with a base freq of 20Khz, by way of example? 555s can easily be PWM modulated using pin 5.

BTW, at this point think breadboarding before wiring a permanent circuit. Your free space just got used up.

 

That fell under the request for more information. So lets try again, does it have to be a sine wave, or can a triangle wave work? Do you need all the way off and all the way on, or would a percentage work (and what percentage).

Whilst we are asking, what current do you require through the LED? This is needed since we are now probably going to use more than one chip.

Another question, would PWM work for the sine wave, say 1Khz modulation with a base freq of 20Khz, by way of example? 555s can easily be PWM modulated using pin 5.

BTW, at this point think breadboarding before wiring a permanent circuit. Your free space just got used up.

Sorry for not mentioning it earlier Bill, but sine wave will do fine as well i guess.
I will just brefly tell you what i am doing so it will make things clear.

I have to designed a circuit and mechanics for short distance underwater communication using Green LED and a photodiode. The LED transmits a 1kHz signal, and the photodiode detects it, then amplifies it in 2 stages, and gives a output voltage. Now what i have noticed is that the signal dies away quicker with square wave when LED and photodiode move further apart (i.e smaller ouptut volatge), than sin wave.
But its not a big issue.
So many thanks for your help so far, you saved me hours.

 

That's odd, I would figure the square wave would travel further, since the RMS value is higher. So what is the max current of the LED? I'm guessing 20ma, but I have been spectacularly wrong on that value before.

I think I can boost the range a bit with just a ICM7555, assuming the receiver is powered. I'm thinking if we reduce the duty cycle of the LED to 25% and increase the max current around 300%, the brighter light might travel further.

Again, if the receiver is powered you can use active filters and op amps to really increase the range. You'll find we like technical challenges here, we just have to know what the problems are.

With just a tone this is a pretty simple link. Other than Go/No Go what are you communicating, assuming it's not proprietary.

 

BTW, have you ever seen a Type N battery? They are used for garage door openers, about half the size of a AA, but 12V. More voltage opens a lot of possibilities.

 

That's odd, I would figure the square wave would travel further, since the RMS value is higher. So what is the max current of the LED? I'm guessing 20ma, but I have been spectacularly wrong on that value before.

I think I can boost the range a bit with just a ICM7555, assuming the receiver is powered. I'm thinking if we reduce the duty cycle of the LED to 25% and increase the max current around 300%, the brighter light might travel further.

Again, if the receiver is powered you can use active filters and op amps to really increase the range. You'll find we like technical challenges here, we just have to know what the problems are.

With just a tone this is a pretty simple link. Other than Go/No Go what are you communicating, assuming it's not proprietary.

Correct!
Heres the data sheet if you want to have a look.
http://www.farnell.com/datasheets/78533.pdf

Well all i am trying to do is compare the attenuation properties of transmission through air with transmission through water. So basically plotting the output voltage in both cases against distance. So a wider range of data will be more desirable, hence increase the signal strength.
Btw the output voltage is being measured on a LED voltmeter. The receiver circuit is powered by 12V dc source.

 

Ouch, just looked at the specs. I know there is a way around it, seen it on this forum as a matter of fact, but don't remember what/where. Your forward dropping voltage of the LED is 3.2V nominal, 3.8V peak. In other words, your LED drops more voltage than your battery provides.

Any chance you could add a bit more voltage, say one more AA battery? This would give you a nominal 4.5VDC to work with. A lithium 9V battery wouldn't be much larger.

 

BTW, have you ever seen a Type N battery? They are used for garage door openers, about half the size of a AA, but 12V. More voltage opens a lot of possibilities.

i am using 12V battery for the receiver circuit already.
But you think it will be ok to use it for LED as well?

 

Yes, what kind of battery if I might ask. Be sure to compensate your LED resistor though, but calculate for 50ma since it is being pulsed.

 

Yes, what kind of battery if I might ask. Be sure to compensate your LED resistor though, but calculate for 50ma since it is being pulsed.

Its a type N battery about 28mm (half the length of AA battery)

 

Gee, I think I mentioned one of those earlier...

12V - 3.2V = 8.8V

8.8V / .05A = 180Ω @ 1/4W (1/2W is better though)

 

Gee, I think I mentioned one of those earlier...

12V - 3.2V = 8.8V

8.8V / .05A = 180Ω @ 1/4W (1/2W is better though)

True but using a high voltage battery = more heat building up in the small box with no heat sink.