Simple and Cheap Laser Digital Audio Transmission

Ever since I made the laser gun, I've been thinking about modulating the laser to send over audio, either for fun (a kids intercom), or maybe to transmit data for a more sophisticated laser gun, enabling a receiver to figure out by whom he was hit. In this instructable I will focus on the audio transmission.

Many people have created analog modulated transmission systems by adding the analog audio signal to the power supply of the laser diode. This works, but it has a few serious drawbacks, mostly being the inability to amplify the signal at the receiving end without introducing a lot of noise. Also linearity is very poor.

I wanted to modulate the laser digitally using a Pulse Width Modulation (PWM) system. The cheap laser diodes used in the laser gun project can be modulated even faster than a normal LED, way into the millions of pulses per second, so this should be very doable.

Step 1: Proof of Principle (the Transmitter)

It is entirely possible to build a somewhat decent transmitter using using a triangle or sawtooth generator and comparing it's output with the signal input with an op-amp. However, it is pretty hard to get good linearity and the number of components grows out of whack pretty fast, and the usable dynamic range is often limited. Besides, I decided it was allowed to be lazy.

A bit of lateral thinking pointed me to an ultra cheap D-class audio amplifier called a PAM8403. I used it before as a real audio amplifier in the laser gun project. It does exactly what we want, pulse width modulating the audio input. Small boards with the required external components can be procured from eBay for under 1 Euro.

The PAM8404 chip is a stereo amplifier with a full H-bridge output, which means it can drive both wires to the speaker to the Vcc (plus) rail or to ground, effectively quadrupling the output power compared to just driving one wire. For this project we can simply use one one of the two output wires, of one channel only. When in complete silence the output will be driven to a square wave of approximately 230 kHz. Modulation by the audio signal changes the pulse width of the output.

Laser diodes are extremely sensitive to over-current. Even a 1 microsecond pulse can destroy it completely. The circuit shown prevents exactly that. It will drive the laser with 30 milliamp independent of VCC. However, it there is even the slightest disconnect of the diodes, normally clipping the transistor's basis voltage to 1.2 volt, the laser diode is immediately destroyed. I have blown two laser modules like this. I recommend to not build the laser driver on a breadboard, but solder it on small piece of PCB or free-form in a piece of shrink tube at the back of the laser module.

Back to the transmitter. Connect the output of the PAM8403 to the input of the laser driver circuit and the transmitter is done! When fired up, the laser is visually on and no modulation can be optically detected. This actually makes sense as the signal hovers around a 50/50 percent on/off state on a 230 kHz carrier frequency. Any visible modulation would not have been the volume of the signal, but the actual value of the signal. Only at very, very low frequencies the modulation will be noticeable.

Step 2: Proof of Principle (the Receiver, Solar Cell Version)

I investigated many principles for the receiver, such as negatively biased PIN photo diodes, non biased versions, etcetra. Different schematics had different advantages and disadvantages, such as speed versus sensitivity, but most of all things were complex.

Now I had an old IKEA Solvinden solar powered light in the garden that was destroyed by rain ingress, so I salvaged the two small (4 x 5 cm) solar cells and tried how much signal would be produced by simply pointing the modulated red laser diode on one of them. This turned out to be a surprisingly good receiver. Modestly sensitive, and good dynamic range, as in, it works with even pretty bright illumination from stray sunlight.

Of course you can search on i.e. eBay for small solar cells like this. They should retail for under 2 Euros.

I hooked up another PAM8403 D class receiver board to it (which also got rid of the DC component), and connected a simple speaker attached to it. The result was impressive. Sound was reasonably loud and distortion-free.

The downside from using a solar cell is that they are extremely slow. The digital carrier is completely wiped out and it is the actual demodulated audio frequency that is coming through as signal. The advantage is that no demodulator is needed at all: just hook up the amplifier and speaker and you're in business. The downside is that since the digital carrier is not present, and therefor cannot be restored, the performance of the receiver is completely dependent of the light intensity and audio will be distorted by all stray light sources modulated in the audio frequency range such as light bulbs, televisions and computer screens.

Step 3: Test!

I took the transmitter and receiver out at night to easily see the beam and have maximum sensitivity of the solar cell, and there was immediate success. The signal was easily picked up 200 meters down range, where the width of the beam was no more than 20 cm. Not bad for a 60 cents laser module with a non-precision collimator lens, a scavenged solar cell and two amplifier modules.

Minor disclaimer: I did not make this picture, just took it from a well known search site. As there was a little bit of moist in the air that night, the beam did indeed look like this when looking back towards the laser. Very cool, but that is beside the point.

Step 4: After Thoughts: Building a Digital Receiver

Building a Digital Receiver, PIN Diode Version

As said, without regenerating the high frequency PMW signal, stray signals are very audible. Also, without the PMW signal regenerated to a fixed amplitude, the volume, and therefor the signal-to-noise ratio of the receiver is totally dependent on how much laser light is captured by the receiver. If the PMW signal itself would be is sufficiently available at the light sensor's output, it should be very easy to filter out these stray light signals as basically everything under the modulation frequency should be considered stray. After that, simply amplifying the remaining signal should produce a fixed amplitude, regenerated PWM signal.

If have not yet build a digital receiver, but it might be very doable using a BWP34 PIN diode as detector. One would have to decide on a lens system to increase the capture area, as the BWP34 has a very small opening, about 4x4mm. Then make a sensitive detector, add a high pass filter, set to roughly 200 kHz. After filtering, the signal should be amplified, clipped to restore the original signal as good as possible. If that would all work, we have basically restored the signal as it was producted by the PAM chip and could be directly fed into a small speaker.

Maybe for a later date!

Different approach, the pro's!

There are people doing light transmissions over vastly greater distances (several tens of kilometers) than presented here. They do not use lasers because monochromatic light actually fades quicker over distance in a non-vacuum than multichromatic light. They use LED clusters, huge fresnel lenses and of course travel great distances to find clean air and long lines of sight, read: mountains.And their receivers are of very special design. Fun stuff that can be found on the internet.