High frequency laser switching

Hello
I'm new to the forum so i hope that I'm not doing something wrong. But here's my question. I am building a laser setup with an audio control. The laser has to be turned on and off with sound coming from my audio jack on my PC( max. 20 000 Hz). I understand electronics pretty well but this project brought me to my limits. In the end i have made a Darlington transistor configurations where the first transistor is a BC107 and his base is connected via a 100 ohm resistor to the sound signal. The second transistor is a power transistor i used a 2N3055 ( I know it is too much power but doesn't matter) and everything is connected an shown on the picture. I am using a cheap 5 mW laser, something like this:
http://i.ebayimg.com/t/5mW-532nm-Powerful-Astronomy-Green-Beam-Light-Laser-Pointer-Pen-Class-3A-NEW-/00/s/NTAwWDUwMA==/$T2eC16hHJGQE9noMcT!ZBQM,UUgGY!~~60_12.JPG


(J3- signal input; J1,J2- laser connector)

And i tried another circuit that i made.

(J1- audio signal input; J2,J3- laser connector)

The same problem occurred with both circuits. When I connect them to the audio signal nothing is happening and when i connect them to 3,3 V the laser works.
What am i doing wrong? :cryyy: Please help me.
Nox.

Why do you want to turn it on/off in sync with the audio signal?! the ouput of the sound card has analog levels and depending on the amplitude of the signal, some portions of it will turn it on while others will do nothing. the darlington setup you have there would need positive swing of at least 1.8-2v to become conductive, moreover,cheap green lasers are not known for having adequate response times. in general, you have to also consider the response time of the driver preceding the laser diode.
this said, what is it that you are trying to build there? You stand better chances if you start from a scratch with a bare diode and appropriate driver circuit.

Yeah why do you need to switching to music? Are you trying to build a projector? Please give as many details about your project as possible so there isn't a ton of back and forth. Ere are many people here that can help you. Good luck.

Michael

If you're looking for an audio laser transmitter I can help with the design and implementation (see my signature, I've done it!), but at this point I don't know if that is what you're really after.

What exactly are you trying to do? Transmit audio via laser, make a laser pulse "in time" with music, send data, etc?

First of all thanks for the fast response.
The project is rather complicated and I am doing the hardware and my friend is doing the software. The project is based on transferring data between two computers with a laser.
Now let me explain all of the details. The receiver is based on an Arduino circuit and a photo diode. The diode is turned on or off with the laser. That is a logic 1 or a logic 0 if there is no laser. The transmitter however should be sending data from the first PC via the laser to the diode. I have done this because i don't have a second Arduino board and my school doesn't want to support us in our project. Because i haven't got a second D/A converter the only source of useful voltage was the stereo out from the motherboard. My friend has made a program that sends a monotone sound ( not a bunch of frequencies just one frequency)of a specified frequency to the audio port. The program takes a file that you chose and transfers it into ones and zeros, sends the 1 as a beep through the audio port and the 0 as no beep. That are the 1 and 0 that are been transferred trough my circuit (that is not working)to the laser. The laser sends them to the photo diode, the Arduino transfers them back to 1 and 0 and sends them to the COM port. The receiver program is constantly listening and when the data comes in it stores it and later he assembles the 1 and 0 to the original file that was send my the transmitter.
I hope this was helpful. :thinking:

Ahh ok, data transmission!

The problem with your topology is that the carrier audio frequency your sound card puts out will show up on your photodiode's output and the Rx pc will interpret it as a binary pulse train giving erroneous data.

Data transmission like this is either done with a true binary output signal from a com port (can be USB to serial for example) or with FM modulation of a carrier wave which is then filtered out at the receiver (ala PSK or AFSK).

The easiest one to do on a budget would be serial output based TTL modulation of the supply rail in to your laser driver. A simple MOSFET switched by the 12v Tx line from a serial port will do the job nicely. Note this doesn't work for USB-to-Serial adapters as USB is a 5V topology and MOSFETs require >10v for binary (class-D) switching. You could use a bipolar transistor instead of the MOSFET but you would have to properly bias it and ensure the maximum ratings are not exceeded, plus bipolar technology would incur a voltage drop across Vce that reduces laser power and efficiency.

Also, green lasers are a poor choice for DIY modulation as the drivers have a sluggish nonlinear response to rail modulation.

First of all, make sure you get some goggles if you use anything > 5mW. Your green laser there is probably also > 5mW, despite the label, as there are rarely checks to ensure it is eye-safe. Since you'll be trying to aim this laser at some sort of sensor, you might get some reflections that can enter your eye. These goggles are supposedly pretty decent for that wavelength.

If you're using the raw sound card output, the output is an AC signal, not DC. So you can't just send a "1" or "0" and expect the value at the output to be high/low. Rather, you need to modulate the signal onto a carrier wave to do much with it. The alternative is to do what people do for sound-card DACs: connect the electrical output before the output filter capacitors.

As for your circuit, here are some issues. You're using a 1N4004 diode. That's a very slow diode used for power applications, not signals. You should use a faster diode (e.g. a 1N4148), or just outright omit the diode altogether because you're dealing with such low-power inputs to your transistor base. Remember to connect the grounds of the circuit with the grounds of your signal output, otherwise the output will just float.

The output from a sound-card's audio output is centered around 0V. However, the output from the sound-card before the output capacitors is usually centered around some positive voltage. In both cases, the output voltage is relatively low. You should consider creating a correction amplifier for the output to ensure that your signal for your laser is within your expected range. Benm has a nice diagram you can find here.

For the lasers: green lasers are often not going to be great for data transmission. They rely on an IR laser "pumping" a laser crystal set to produce green. Depending on your modulation rate, the crystals inside the laser may not be able to recover fast enough. Still, that may not be an issue with the low powered 5mW, but given their low quality, it might just. A red laser might be a better thing to start off with. They're cheap, low-power and easy to modulate. With a red filter on the sensor you can filter out the remaining light.

If you're going to modulate something that already has a driver like that green, try just using a power MOSFET and connecting to the gate of that. That way you don't need to deal with all the resistors. If you're going to module a raw laser diode, you need to do some current regulation. Check out this MOSFET-based driver and see if you can adapt it to your needs. It works on the same principles as your circuit, but ensures a maximum current level. If you hook it up to something like a higher powered red or blue, it'll work good for that.

Benm also has a nice diagram for this here. It incorporates a correction amp as well. Note that these are generally for modulating diode-lasers, not DPSS (green) lasers. They may still work for your DPSS lasers, but they're providing current regulation that the DPSS laser driver already has.

Finally, in your diagrams, can you shift the wire connections so that they only form "T" connections, not "+" connections, or use connection dots to indicate connections. Otherwise, it's hard to determine whether you mean for the connection to cross over the wire span, or connect to the wire.

Thank you very very much.
All of those information were quite useful and i learned some things. But i am trying to send data from one computer with the laser at some speed and testing my green laser i revealed that it doesn't fully light up at frequencies higher than 200 Hz. The laser as you said is too slow, it doesn't light up fully and the input signal has already changed to going down so the laser goes down too. And so it doesn't have time to get to full power ans already has to shut down. But that doesn't matter really much I will transmit data at 100 Hz it is rather slow but I and building this thing for demonstration purposes for a competition.
By looking around and considering all the thins everyone said i did some modifications to my transistors. The first thing is that i have connected the whole thing to my mixer which has already a preamp in it so the signal coming from the audio out isn't anymore 1 V it is now 3.5 V and that is more than enough to switch my Darlington transistor. The problem is that i don' have a a serial port on my PC and my friend doesn't have one on his laptop too. But i think i will make a mosfet switch for the laser and I won't use any driver because my laser pointer already has wan built in it, it is rather poor but good enough for this job.
After doing all of this i can now turn my laser on and off up to 200 times per second (200 Hz) because above that the laser isn't powerful enough to trigger the diode.But now there is a problem with the Arduino. Because the Arduino is reading much faster than the laser is sending it is reading 11111111 instead of 1 and 00000000 instead of 0. How can i change that can i do something to the Arduino code or should I tell my friend to change something on the program? What is the easiest solution of this problem?
Thank you for your concern i already have some goggles.

If it's down to 200hz it's probably because your other components are acting like a filter on the output; it's probably somewhere on the driver for the green laser.

You'll need to do use an asynchronous serial data protocol to read in the signal properly. The receiver can't rely on the pulses being perfectly timed. A simple method you can try is when you start sending some data, you start the frame with a train of alternating 1 and 0 pulses that the receiver detects and measures the pulse length using. Then, relying on the measured time of those pulses, the receiver reads the data that follows. You should probably find and use a more sophisticated protocol for that kind of thing to eliminate errors. Maybe look at the VirtualWire library or something.

Thanks I will try to do that. :thanks:

Look up the LIN one wire interface - should work very well as the protocol described by BB. LIN can sometimes be implemented with a microcontroller's UART module.

Today I have done some serious tests with some oscilloscope and other sensitive gear. And I realized That The laser in fact is fast enough to reproduce a frequency of 1000 Hz but that frequency is creating a very low voltage ( 1-1.5 V) at the receiver diode and the Arduino does not recognize that as a logic 1. So now what I have to do is make some kind of voltage amplifier for the diode. I think That i have done this pretty good but I don't really know what circuit to use because the signal mustn't be inverted and just the signal voltage should be amplified from about 1.5 V to a minimum of 3.3 V to 5 V.
If somebody got an idea or a simple schematic would be great.Thanks.:yh:

Use a Schmitt trigger on the output of the photodiode. If you use a standard 74HC14 which inverts just feed the inverted output back in to another schmitt trigger stage (there's like 6 or 7 on the chip) so it inverts it again putting it back where it needs to be.

The 74HC version operates on 5V and is TTL compatible, and being a schmitt trigger it will impart a lot of noise immunity.

There are non-inverting amplifier topologies too, but yeah, a Schmitt trigger would probably be the best.