I have designed a voice modulated Diode laser transmitter using the Metrologic $50 pen-pointer laser. It seems to work well, although I don't claim it to be optimized. I spent only a day on this circuit, versus weeks on fine-tuning the photodiode receiver circuit presented here.
I first envisioned pulling a trick on the Diode laser. These modern Diode lasers contain not only a laser, but also a photodiode. The purpose of the photodiode is to monitor the output of the laser and provide the power supply with feedback as to how bright the laser is. This feedback circuit stabilizes the laser output over temperature changes, which they are quite sensitive to. I thought that if I shined a bright, modulated, LED into the laser's lens I could fool the feedback photodiode into modulating the laser. It didn't work at all. In fact, it worked so entirely and completely not at all that I suspect that Metrologic is really not using the feedback photodiode, even though their data sheet claims that. So much for tricks.
The trick that I thought wouldn't work, did. That is, modulating the power supply. I figured that if the Diode laser has such a critical operating point, and that it is easily damaged by static electricity, that the pen-pointer's power supply would be somewhat sophisticated and resistant to power source (battery) changes. It isn't. In fact it acts like the sophisticated laser power supply is maybe just a resistor. Varying the power supply current varies the laser's brightness. How simple! By the way, the pen-pointer in question uses two AAA batteries for power. Note that other models of pen pointers may have more sophisticated power supplies that would prevent them from working in the circuit I have designed. This circuit has been tested only on the Metrologic ML211 (Edmund G52,442) laser pen-pointer.
MIC is an electret condenser microphone element. It is Radio Shack catalog number 270-090, $1.79. The supply voltage can range from 2 to 10 VDC, but the microphone is optimized for 4.5V. (It's kind of noisy at 9V.) The two 10K resistors divide the 9V battery down to 4.5V.
IC1 is an LM324 quad op amp. Only one section is used. I'm sure that a number of other op amps would suffice. The 324 is good as it was designed for single supply operation and it draws little current.
The 1K and 100K resistors connected to op amp pin 2 set the gain. You may want to vary the 100K value for different microphones.
The 100K pot sets the voltage that the circuit "idles" at. Monitor TP with a DC voltmeter while adjusting the pot. I found 3.25 VDC +/- 0.05 VDC to be optimum. Mis-adjustment will result in a distorted voice signal.
Switch S1 has two positions. When it is connected to 9 Volts it is in the "Align" position. This will turn the laser on full brightness. When switched to the pot the switch is in the "Voice" position. Here the laser is biased for voice transmission. It will not be as bright because it is never fully on.
The transistor is needed because the op amp will not source or sink quite enough current alone to power the laser.
The string of four diodes is very important. They act like a 3 Volt Zener diode to limit the voltage to the laser. I don't know how much more than three volts the laser can take, and I don't want to spend another $50 to find out. I used 1N4007 diodes for the string. Others will work, including real Zener diodes. Make sure to check that the circuit limits the voltage to three volts before connecting the laser. You can test this by flipping S1 to the "Align" position while measuring the DC voltage where the laser would connect.
The power source for the transmitter is a 9 Volt transistor battery, but other voltages could be used too. It draws about 70 mA at 9 Volts. A 12 Volt Gel Cell or NiCad would give more amp-hours (life) than a 9 Volt battery, and would be rechargeable to boot. I wouldn't expect that any circuit changes would have to be made.
The Laser is again from a Metrologic ML211 pen-pointer, 3 Volt supply, 675 nm output wavelength, 2 milliwatt output power, 1.7 milliradian divergence. (Also known as Edmund Scientific p/n G52,442.) The Diode laser module simple pulls out of the pen-pointer housing. (There goes your 90 day warranty!) The pocket clip is positive, the coiled spring battery contact is negative.
A laser receiver consists of an antenna, a detector, an amplifier, and a speaker - the same as a simple radio receiver. In this case the antenna is a lens or mirror. The detector is a photodetector that converts light into electricity. The amplifier boosts the signal enough to drive a speaker. There may also be filters; optical and/or audio.
As with radio receivers, the antenna of a light communications receiver is the best area to make system improvements. "The bigger the better" applies here also. The purpose of the light antenna is to gather as much light from the desired source as possible, concentrate this light onto the detector, and reject light coming from unwanted sources. Unwanted sources are the sun, moon, street lights, etc. Lenses and curved mirrors are our main tools. Such elements need only be mounted in an appropriate tube or holder to make a complete antenna system. Obviously what we need is something like a telescope. Unlike a telescope, however, we needn't bother trying to obtain an actual detailed image of the light source. We need merely to gather as much light from the source as possible. Also, adjustable focus in not needed. A light communications antenna then can be much simpler and cheaper than a telescope.
The conventional lens is the most common light receiver antenna candidate. Because of weight and cost, four inches diameter is about the largest size inexpensive glass lenses appear in. Mirrors take over from there. Color corrected lenses (achromats) are not needed for our application because we are dealing with only one wavelength of light. One source of large diameter lenses are the magnifying lenses typically sold in stationary stores. These lenses may be made of plastic instead of glass and will require a little more care in handling to prevent scratches. Watch for used Luxo-type magnifying work lights that have a very large lens surrounded by a fluorescent bulb. Edmund Scientific sells a folding stand magnifier that contains a 4.3" diameter glass lens with a 8.5" focal length. It is their model G38,599 priced at $9.50. The lens can be popped out of the plastic stand and will make an excellent optical "antenna."
Fresnel lenses (pronounced fray-nel) are flat lenses, usually less that 1/8 inch thick. They come in round, square, and rectangular shapes and are usually made of plastic. One side is flat while concentric ridges are molded onto the other side. Fresnel lenses won't form much of an image, but they will gather a very large amount of light for their cost and weight. I have used a 10 1/4 inch diameter Fresnel lens mounted in a metal duct pipe to serve as an antenna for a portable 931 photomultiplier receiver. Edmund Scientific is a good source of Fresnel lenses.
Parabolic and spherical curved mirrors make excellent light antennas as witnessed by their heavy use in astronomy. Swap meets and garage sales may be sources for old reflector telescopes that use such mirrors. A small telescope with a photodetector mounted in place of the eyepiece would make a great light communications receiver, but a large Fresnel lens would probably gather much more light.
If only night time communication is anticipated, then optical
bandpass filtering can be dispensed with. Daylight
communication is a different story. The sun is quite bright,
to say the least, and can swamp the photodetector. A very
narrow bandpass filter is what we need. Luckily, Dielectric
filters, sometimes called Interference filters, fit the
bill. Edmund Scientific sells dielectric filters for several
popular laser wavelengths for about $40 each. The
specifications for their 632.8 nanometer Helium Neon laser
filter state a half power bandwidth of 10 nanometers which is
only 1.6 percent! The drawback is that the filter insertion
loss is 50%, which isn't too bad considering the bandwidth.
The filter should be placed just in front of the detector.